United States
       Environmental Protection
       Agency
 Off ice of Water
EPA 815-D-15-001
  October 2015
Draft - Technologies for Legionella Control:
         Scientific Literature Review

-------
Office of Water (4607M)
EPA815-D-15-001
October 2015

-------
This page intentionally left blank.

-------
                                     Disclaimer

This draft document was prepared by the U.S. Environmental Protection Agency (EPA) as a
technical resource for primacy agencies, building water system operators, and building owners to
consider as they evaluate technologies to respond to the risks associated with Legionella
colonization of premise plumbing. This draft document is a summary of publicly available, peer-
reviewed, technical literature that evaluates the effectiveness of six technologies used for
Legionella control. The draft document also discusses water quality issues that could result when
using the various approaches, and summarizes operational conditions for each technology. It also
discusses critical multiple-barrier approaches to address microbial (including Legionella)^
physical and chemical risks in various parts of the building water system, such as water
management programs (WMPs), hazard analysis and critical control point (HACCP), and water
safety plans (WSPs). This document also provides an overview of other strategies that primacy
agencies, building water system operators, and building owners could consider when addressing
threatening public health risks associated with a legionellosis outbreak.

This draft document is not a regulation; it is not legally enforceable; and it does not confer legal
rights or impose legal obligations on any party, including EPA, states or the regulated
community. While EPA has made every effort to ensure the accuracy of any references to
statutory or regulatory requirements, the obligations of the interested stakeholders are
determined by statutes, regulations or other legally binding requirements, not this document. In
the event of a conflict between the information in this document and any statute or regulation,
this document would not be controlling.

Although this document describes technologies for controlling Legionella in finished water, the
information presented may not be appropriate for all situations and alternative approaches may
be applicable.

Mention of trade names or commercial products does not constitute an EPA endorsement or
recommendation for use.

-------
                                  Table of Contents
Table of Contents	i
Table of Exhibits	Hi
Abbreviations and Acronyms	iv
Acknowledgements	vi
Preface	vii
Executive Summary	1
1   Background	2
1.1   Purpose and Scope	2
1.2   Legionella: Overview	2
  1.2.1    General Information	2
  1.2.2    Epidemiology and Pathogenesis	3
  1.2.3    Ecology and Physiology	5
1.3   Legionella Occurrence and Risk from the Distribution System and Building Water
      System	6
1.4   Regulatory Context	7
2   Multi-Barrier Approaches and Technologies to Control Legionella	8
2.1   Overview of Current State of Knowledge	8
2.2   Multi-Barrier Approaches	10
  2.2.1    Background	10
  2.2.2    Applications of Multi-Barrier Approaches	11
  2.2.3    Environmental Testing	12
2.3   Technologies	13
  2.3.1    Chlorine	13
      2.3.1.1   Background	13
      2.3.1.2   Characterization of Effectiveness against Legionella	14
      2.3.1.3   Potential Water  Quality Issues	19
      2.3.1.4   Operational Conditions	20
  2.3.2    Monochloramine	21
      2.3.2.1   Background	21
      2.3.2.2   Characterization of Effectiveness against Legionella	22
      2.3.2.3   Potential Water  Quality Issues	27
      2.3.2.4   Operational Conditions	28
  2.3.3    Chlorine Dioxide	30
      2.3.3.1   Characterization of Effectiveness against Legionella	30
      2.3.3.2   Potential Water  Quality Issues	33
      2.3.3.3   Operational Conditions	34
  2.3.4    Copper-Silver lonizati on	36
      2.3.4.1   Background	36
      2.3.4.2   Characterization of Treatment Technology Effectiveness against Legionella.... 36
      2.3.4.3   Potential Water  Quality Issues	38
      2.3.4.4   Operational Conditions	39

Draft - Technologies for Legionella Control:
Scientific Literature Review                       i                                 October 2015

-------
 2.3.5    Ultraviolet Light Disinfection	41
      2.3.5.1   Background	41
      2.3.5.2   Characterization of Effectiveness against Legionella	42
      2.3.5.3   Potential Water Quality Issues	43
      2.3.5.4   Operational Conditions	44
 2.3.6    Ozone	45
      2.3.6.1   Background	45
      2.3.6.2   Characterization of Effectiveness against Legionella	45
      2.3.6.3   Potential Water Quality Issues	47
      2.3.6.4   Operational Conditions	47
3   Other Strategies Used to Control for Legionella	49
3.1   Emergency Remediation	49
 3.1.1    Superheat-and-Flush Disinfection	49
      3.1.1.1   Background	49
      3.1.1.2   Characterization of Effectiveness against Legionella	49
      3.1.1.3   Potential Water Quality Issues	51
      3.1.1.4   Operational Conditions	52
 3.1.2    Shock Hyperchlorinati on	53
      3.1.2.1   Background	53
      3.1.2.2   Characterization of Effectiveness Against Legionella	53
      3.1.2.3   Potential Water Quality Issues	54
      3.1.2.4   Operational Conditions	55
3.2   Point-of-Use Filtration	55
 3.2.1    Background	55
 3.2.2    Characterization of Effectiveness against Legionella	57
 3.2.3    Potential Water Quality Issues	59
 3.2.4    Operational Conditions	59
4   Questions and Answers on Legionella Control in Building Water Systems	60
4.1   Public Health Concerns	60
4.2   Potential Regulatory Requirements	60
4.3   Control Measures	62
4.4   New Technology Approval	64
4.5   Permitting	64
4.6   Sampling and Monitoring	65
4.7   Operator Certification	66
4.8   Unintended Consequences	67
4.9   Additional Sources of Information	67
5   References	68
A   Appendix	93
A. 1   Elements of Hazard Analysis and Critical Control Points (HACCP)	93
A.2   Elements of a Water Management Program	96
A.3   Elements of Water Safety Plan	97
Draft - Technologies for Legionella Control:
Scientific Literature Review                       ii                                  October 2015

-------
                                  Table of Exhibits

Exhibit 1-1: Legionella transmission	4
Exhibit 2-1: Kuchta et al. (1983) findings on CT values (min-mg/L) for 2-log (99 percent)
            reduction of L. pneumophila using chlorine	15
Exhibit 2-2: Jacangelo et al. (2002) findings on CT values (min-mg/L) for 2-log (99 percent)
            reduction of L. pneumophila using chlorine	15
Exhibit 2-3: UV doses (ml/cm2) for inactivation of L. pneumophila	43
Exhibit 3-1: Membrane filtration guide for removal of microbial contaminants	56
Draft - Technologies for Legionella Control:
Scientific Literature Review                      Hi                                 October 2015

-------
                          Abbreviations and Acronyms
ANSI
AOC
ASHRAE
ATP
AWWA
C
CDC
CFR
CPU
CSI
CT

DBF
DBPR
DPD
EPA
F
FIFRA
GWR
HAA
HAAS
HACCP
HPC
ICP/MS
ICU
km
MCL
MF
mg/L
ml/cm2
mM
MRDL
NDMA
NF
OSHA
POE
POU
ppm
PWS
qPCR
RO
SDWA
SMCL
American National Standards Institute
Assimilable organic carbon
American Society of Heating, Refrigerating and Air-Conditioning Engineers
Adenosine triphosphate
American Water Works Association
Celsius
Centers for Disease Control and Prevention
Code of Federal Regulations
Colony-forming units
Copper/silver ionization
The product of disinfectant residual concentration "C" and contact time "T"
(CxT)
Disinfection byproduct
Disinfectants and Disinfection Byproducts Rule
N,N-diethyl-p-phenylenediamine
United States Environmental Protection Agency
Fahrenheit
Federal Insecticide, Fungicide and Rodenticide Act
Ground Water Rule
Haloacetic acid
Sum of the  mass concentrations of five haloacetic acid species
Hazard analysis and critical control points
Heterotrophic plate count
Inductively coupled plasma/mass spectrometry
Intensive care unit
Kilometers
Maximum contaminant level
Microfiltration
Milligrams  per liter
Millijoule per square centimeter
Millimolar
Maximum residual disinfectant level
jV-nitrosodimethylamine
Nanofiltration
Occupational Safety and Health Administration
Point-of-entry
Point-of-use
Parts per million
Public water system
Quantitative polymerase chain reaction
Reverse osmosis
Safe Drinking Water Act
Secondary maximum contaminant level
Draft - Technologies for Legionella Control:
Scientific Literature Review
                                                          October 2015

-------
spp.              All species within a genus
SWTR            Surface Water Treatment Rul e
THM             Trihalomethane
TTHM            Total trihalomethane
|i                 Micron (millionth of a weight, distance and/or volume unit)
Hg/L              Micrograms per liter
|im               Micrometer
UF               Ultrafiltration
U.S.              United States
UV               Ultraviolet
UVT              UV transmittance
VBNC            Viable but non-culturable
WHO             World Health Organization
WMP             Water management programs
WSG             Water supply guidance
WSP              Water safety plan
Draft - Technologies for Legionella Control:
Scientific Literature Review                       v                                  October 2015

-------
                               Acknowledgements
EPA's Office of Ground Water and Drinking Water would like to thank the Association of State
Drinking Water Administrators; the States of Maryland, Minnesota, Nevada, New York, Ohio,
Pennsylvania, Washington; and the U.S. Centers for Disease Control for their contributions to this
document. EPA thanks these participants for their meaningful input which was critical for the
development of this draft document.

The following individuals helped to develop and/or review this draft document:
       Cesar Cordero (U.S. EPA)1
       Mike Finn (U.S. EPA)1
       TomGrubbs(U.S. EPA)1
       Hannah Holsinger (U.S. EPA)1
       Crystal Rodgers-Jenkins (U.S. EPA)1
       Lili Wang (U.S. EPA)1
       Stacy Pfaller (U.S. EPA)1
       Darren Lytle (U.S. EPA)1
       Mark Rodgers (U.S. EPA)1
       Jennifer Carr (Nevada Department of
       Environmental Protection)1
       Ross Cooper (Nevada Department of
       Environmental Protection)1
       David  Dziewulski (New York State
       Department of Health)1
       Samuel Perry (Washington State
       Department of Health)1
       Darrell Osterhoudt, (Association of
       State Drinking Water
       Administrators)
       Laurel Garrison (CDC)
       Natalia A. Kozak-Muiznieks (CDC)
       Claressa Lucas (CDC)
       Leslie  Darman (U.S. EPA)

       lLead author
Phil Berger (U.S. EPA)
StigRegli(U.S. EPA)
KenRotert(U.S. EPA)
John Hebert (U.S. EPA)
JingrangLu(U.S. EPA)
Emily Mitchell (U.S. EPA)
Mark Perry (U.S. EPA)
Jonathan Pressman (U.S. EPA)
Nicole Shao (U.S. EPA)
AlysaSuero(U.S. EPA)
Elizabeth Messer (Ohio EPA)
Jerry Smith (Minnesota Department
of Health)
Lloyd Wilson (New York State
Department of Health)
Nick Codru (New York State
Department of Health)
Lisa Daniels (Pennsylvania
Department of Environmental
Protection)
Saeid Kasraei (Maryland Department
of the Environment)
Michael Elovitz (U.S. EPA)
Draft - Technologies for Legionella Control:
Scientific Literature Review
                      October 2015

-------
                                        Preface
This draft document is a compilation of publicly available, peer-reviewed, technical literature
that evaluates the effectiveness of technologies to control for Legionella. The United States
Environmental Protection Agency (EPA) developed this document because the agency
recognizes that Legionella is a public health threat. EPA also recognizes that many facility
managers are choosing to install treatment systems to prevent or mitigate Legionella in their
building's plumbing systems. The agency expects this document will help to further improve
public health by helping the targeted audience make science-based, risk management decisions
regarding treatment and  control of Legionella in buildings. The EPA is not promoting or
endorsing treatment technologies as a preferred means of Legionella control. Rather the agency
seeks to provide decision makers with scientific information on the effectiveness of these
technologies and the operational requirements. The target audience for this document includes,
but is not limited to, primacy agencies, building water system operators, building owners and
technology developers and vendors.

The intent of this document is to provide a summary of scientific information on Legionella
control technologies that may be considered when assessing a particular building water
system(s). The EPA did  not evaluate individual study quality with the goal of making
recommendations for or against the use of any of the technologies discussed in the document.

The scientific information presented in this draft document comes from published literature
related to six technologies used for Legionella control (chlorine, monochloramine, chlorine
dioxide, copper-silver ionization (CSI), ultraviolet (UV) disinfection and ozone). The draft
document also discusses water quality issues that could result when using the various
approaches, and it summarizes operational conditions for each technology. It also discusses
critical multi-barrier approaches for addressing microbial, physical and chemical risks in various
parts of the building water system, such as water management programs (WMPs), hazard
analysis and critical control point (HACCP), and water safety plans (WSPs). This document also
provides an overview of other strategies that primacy agencies, building water system operators
and building owners could consider when addressing a public health threat such as a legionellosis
outbreak.

The EPA developed this draft document in collaboration with state co-regulators. Legionella
subject matter experts at the Centers for Disease Control and Prevention (CDC) reviewed and
provided feedback on portions of the draft document. All parties were invited to compile the
peer-reviewed literature  referenced in this document. The scientific information in this document
spans from the 1970s to  2014 and is limited to peer-reviewed literature. Information published in
trade journals or popular magazines is not included in this document.

There is not a single one-size-fits-all approach to addressing Legionella concerns in all building
water systems. A determination of which strategy is best suited for a particular building water
system is case-specific due in part to the complex and diverse nature of building water systems.

This document does not  recommend the addition of treatment nor the installation of any of the
technologies discussed herein, but rather provides technical information, based on the publicly
Draft - Technologies for Legionella Control:
Scientific Literature Review                      vii                                 October 2015

-------
available, peer-reviewed literature, about technologies and other approaches for controlling
Legionella and other microbial contaminants. In some facilities, risks associated with the
building water system (including Legionella) may be addressed without the addition of
treatment.

Stakeholders (e.g., primacy agencies, technology developers and vendors) who are interested in
information about the approval process for a new or alternative drinking water treatment
technology are advised to consult EPA's Water Supply Guidance (WSG) 90, "State Alternative
Technology Approval Protocol" (USEPA, 1996). The goal of WSG 90 is to provide a
streamlined and consistent protocol to facilitate state approval of new drinking water treatment
technologies. WSG 90 is not meant to replace current state plan review and approval processes.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       viii                                 October 2015

-------
                                Executive Summary
[This is a placeholder. The Executive Summary will be drafted after comments from the public
and expert peer review have been incorporated.]
Draft - Technologies for Legionella Control:
Scientific Literature Review                       1                                 October 2015

-------
1   Background

    1.1   Purpose and Scope

The purpose of this document is to characterize the current body of knowledge regarding the
effectiveness of available technologies for the control ofLegionella in building finished drinking
water systems (building water systems).l Throughout this document the term building water
system refers to the pipe infrastructure inside a building used to deliver finished drinking water
intended for human consumption. The U.S. Environmental Protection Agency (EPA) defines
water "intended for human consumption" as water used for drinking, bathing, showering, hand
washing, teeth brushing, food preparation, dishwashing, and maintaining oral hygiene (40 CFR
141.801). Discussions ofLegionella control issues related to cooling towers are not within the
scope of this document. The EPA developed this document in collaboration with state co-
regulators. Legionella subject matter experts at the Centers for Disease Control (CDC) reviewed
and provided feedback on portions of the draft document. All parties were invited to compile the
peer-reviewed literature that is summarized and referenced in this document.

The agency expects this document will help to further improve public health by helping the
primacy agencies,2 building water system operators, building owners, technology developers and
vendors make science-based risk management decisions regarding treatment and control of
Legionella in buildings. The EPA is not promoting or endorsing treatment technologies  as a
preferred means ofLegionella control in buildings. Rather, the agency seeks to provide decision
makers with scientific information on the effectiveness of these technologies and the operational
requirements. The EPA did not evaluate individual study  quality or the body of evidence from
available studies with the goal of making recommendations for or against the use of any of the
technologies discussed in the document.

    1.2  Legionella: Overview

    1.2.1  General Information

The genus Legionella currently includes more than 50 bacterial species and approximately 70
distinct serogroups, many of which are considered pathogenic (DSMZ, 2014; LPSN, 2014;
Pearce et al., 2012; Bartram et al., 2007; Fields et al., 2002).  Legionellapneumophila was the
first species to be  described following an outbreak of pneumonia in 1976 among members of the
American Legion, who were attending a convention in Philadelphia, Pennsylvania (Fields et al.,
2002; McDade et  al., 1979). Approximately half of the Legionella species described to date have
been associated with clinical cases of legionellosis (any disease caused by Legionella), but it is
likely that most Legionellae can cause human disease under the appropriate conditions (Borella
1 For the purposes of this document, the term "Legionella" refers to the genus Legionella (any species). The plural
form Legionellae and Legionella spp. are also used to denote the genus Legionella.
2 Primacy - States and Indian Tribes are given primary enforcement responsibility (e.g., primacy) for public water
systems in their State if they meet certain requirements.

Draft - Technologies for Legionella Control:
Scientific Literature Review                       2                                 October 2015

-------
et al., 2005; Fields, 1996; Fang et al., 1989). There are several EPA regulations that provide
some degree of protection against Legionella (see Section 1.4 for additional information).

Legionellae are gram-negative, rod-shaped bacteria. Legionellosis is acquired by inhaling or
aspirating aerosolized water or soil (potting soil, compost soil) contaminated with Legionella
(Travis et al., 2012), as opposed to person-to-person contact, animal-to-person transmission,
consumption of contaminated food, or ingestion of contaminated water. Though animals can be
infected by Legionella and develop disease, they have not been identified as carriers of
Legionella, nor has transmission from animals to humans been documented (Cunha, 2006;
USEPA, 1999a).

     1.2.2  Epidemiology and Pathogenesis

Legionellosis includes Legionnaires' disease, characterized by pneumonia (Fraser et al., 1977),
and Pontiac fever, a milder flu-like illness without pneumonia (Kaufmann et al., 1981; Glick et
al., 1978). Hospitalization and intensive care is common among Legionnaires' disease patients;
inpatient costs are estimated at $433 million per year, with a case fatality rate of 5-30 percent
(Collier et al., 2012). The economic costs associated with loss of productivity and death are not
included in these estimates and are likely to be significant.

Legionellosis is a nationally notifiable disease, which means that any case that is confirmed by a
laboratory is reported to CDC by state health departments (CDC, 2005). However, many cases of
pneumonia that could be Legionnaires' disease are empirically treated with antibiotics and never
tested for Legionella, so the incidence could be much higher than reported (CDC, 2011; Marston
et al., 1997). Between 3,000 and 4,000 cases of legionellosis are reported to CDC each year;
however, the actual number of hospitalized cases is estimated to be between 8,000 and 18,000
(CDC, 2013a; CDC, 2012; Marston et al., 1997).

In the United States, waterborne disease outbreaks associated with Legionella have been tracked
through the Waterborne Disease and Outbreak Surveillance System since 2001  (Craun et al.,
2010). Between 2009 and 2010, CDC reported that Legionella accounted for  19 of the 33
drinking water-related waterborne disease outbreaks in the United States, causing 72 illnesses
and 8 deaths. Environmental conditions within building water systems were identified as the
cause of 11 of the 19 Legionella outbreaks (CDC, 2013b).

Strains of L. pneumophila belonging to serogroup 1 are responsible for most legionellosis cases
in the United States and Europe (Borella et al., 2005; Yu et al., 2002; Fields et al., 2002; Marston
et al., 1994). L. pneumophila serogroup 6 may be the second most common serogroup, based on
the frequency with which it is isolated from clinical samples (Marston et al., 1994).

Although L. pneumophila causes most cases of Legionnaires' disease, other species can also
cause the disease, particularly in hospital-acquired cases.  Of the reported non-L. pneumophila
infections, the most common causes of infection are L. micdadei, L. bozemanii, L. dumoffii and
L. longbeachae (Fang et al., 1989; Reingold et al.,  1984).

While anyone can develop Legionnaires' disease, factors associated with an increased risk of
developing infection  include age (>50 years), gender (male), smoking and drinking habits,
existing lung conditions (e.g., asthma, chronic obstructive pulmonary disease),
Draft - Technologies for Legionella Control:
Scientific Literature Review                       3                                 October 2015

-------
immunosuppressed or immunocompromised status (e.g., persons receiving transplants or
chemotherapy, those with kidney disease, diabetes or AIDS), and recent surgery or intubation
(Health Canada, 2013; Newton et al., 2010; Bartram et al., 2007; Stout and Yu, 1997).

Exhibit 1-1 shows different factors and events that could affect the transmission oi Legionella in
environmental and clinical settings. Legionellosis outbreaks can occur when Legionellae
multiply under particular conditions in water systems and the water is then aerosolized and
subsequently inhaled or aspirated by susceptible persons (Donohue et al., 2014; Fields et al.,
2002; Blatt et al., 1993; Stout et al., 1985; Fliermans et al., 1981). For these reasons, the presence
ofLegionella is a particular concern in large buildings that house susceptible populations, such
as facilities in the healthcare and hospitality industries (Health Canada, 2013; Williams et al.,
2013; Buse and Ashbolt, 2012; CDC, 2008; Rusin et al., 1997; Colbourne and Dennis, 1989).
However, recent outbreaks have demonstrated that legionellosis infections are not limited to
those environments.

                          Exhibit 1-1: Legionella transmission
              Environmental
                                                               Clinical
     Temp., pH,
      Nutrients
 Miorobial Associations
    Events

       1
   Survival in
Reservoir (Nature)
                                                      Events
              Diagnosis of
             Legionnaire's
               Disease
                     Factors

                    Symptoms
                    Lab Tests
                   Surveillance
 MieroWal Associations
      Nutrients
      Biocides
  System Cleanliness
   Temp., Humidity,
   Droplet Production
  Amplification
  Dissemination
 (Aerosolization)
    Risk
Minimization
(Prevention!
    6
 Multiply in
  Human
Phagocytes
                                                  Susceptible Host
                                                     Exposure
                                                   Virulence
                                     Age
                                   Disease
                               Immunodeficiency
Source: ASHRAE, 2000
                                      Transmission
                                        Humidity
                                       Droplet Size
                                        Distance
Building water systems can be colonized with Legionella and transmit the bacteria through
showerheads, faucets, whirlpool spas, respiratory therapy devices, ultrasonic mist machines,
humidifiers, cooling towers, decorative fountains and industrial-use water (Haupt et al., 2012;
Wallensten et al., 2010; Carducci et al., 2010; Edelstein, 2007; Stout and Yu, 1997; CDC, 1997;
Blatt et al., 1993; Addiss et al., 1989; Muder et al., 1986; Bollin et al., 1985; Dondero et al.,
1979; Glick et al., 1978). Cases have also been linked to ice machines, windshield washer fluid
Draft - Technologies for Legionella Control:
Scientific Literature Review
                                                           October 2015

-------
and birthing pools (Public Health England, 2014; Wallensten et al., 2010; Nagai et al., 2003;
Franzin et al., 2001; Graman et al., 1997). In addition, several infections have been linked to
exposures to potting soil (Whiley and Bentham, 2011; CDC, 2000).

    1.2.3  Ecology and Physiology

Fresh water is the major natural reservoir for Legionellae. The bacteria are found worldwide in
many different natural aquatic environments (e.g., lakes, rivers and groundwater); however,
exposure to these sources typically does not result in legionellosis. Legionella species have also
been found to occur in natural soil, potting soil and compost samples (van Heijnsbergen et al.,
2014; Travis et al., 2012; CDC,  2000).

Legionella exhibit several properties that allow them to persist in extreme environmental
conditions such as low and high temperatures, presence of disinfectants, low pH, low nutrients
and high salinity (Health Canada, 2013; Borella et al., 2005; Kuchta et al., 1983; Fliermans et al.,
1981). Ideal growth conditions are in warm water between 35 and 46 degrees Celsius (C) (95-
114.8 degrees Fahrenheit (F)) (Buse and Ashbolt, 2011; Katz and Hammel,  1987; Wadowsky et
al., 1985; Yee and Wadowsky, 1982; Dondero et al., 1979; Glick et al., 1978). High relative
humidity increases the viability  of Legionella species in contaminated aerosols  (Heng et al.,
1995). Legionellae are considered thermotolerant bacteria, able to withstand temperatures of 50
degrees C (122 degrees F) for several hours (Bartram et al., 2007). This characteristic allows
Legionella species to occur frequently in heated water systems (Taylor et al., 2009). Legionella
species can also survive at temperatures below 20 degrees C (68 degrees F)  and even below
freezing (Borella et al., 2005).

Legionella species are often found to be protected from adverse environmental  conditions as a
result of their association with biofilms, as well as their symbiotic and parasitic interactions with
other microorganisms. The association of L. pneumophila with many different microorganisms
in aqueous environments has been widely demonstrated. Studies have shown the ability of
Legionella to parasitize and multiply in several species of protozoa including amoebae, ciliated
protozoa and slime mold (Cervero-Arago et al., 2014; Escoll et al., 2013; Buse  et al., 2013; Buse
and Ashbolt, 2011; Taylor et al., 2009; Fields, 1996) as well as establish symbiotic interactions
with other bacteria (Taylor et al., 2009; Rowbotham, 1986; Wadowsky et al., 1985; Bohach and
Snyder, 1983; Wadowsky and Yee, 1983; Fliermans et al., 1981). The ability of Legionella
species to parasitize certain protozoa that are commonly found to graze on biofilms in
distribution systems is considered particularly important in their ability to survive and grow
under adverse environmental conditions (Hoffman et al., 2014; Escoll et al., 2013; Richards et
al., 2013; Bartram et al., 2007; Hwang et al., 2007; Molmeret et al., 2004; Storey et al., 2004a;
Storey et al., 2004b; Thomas et al., 2004; Fields  et al., 1984). Legionella can parasitize alveolar
macrophages (white blood cells that are part of the immune system) in human lungs the same
way it parasitizes protozoa (Hoffman et al., 2014).

Multiple studies suggest that protozoa play a major role in the transmission of L. pneumophila
and subsequently, legionellosis. Some of the research indicates that infectivity may be
substantially increased if amoebae infected by Legionella are inhaled, as opposed to individual
free-living Legionella cells (Richards et al., 2013; Newton et al., 2010; Borella  et al., 2005;
Cirillo et al., 1999; Brieland et al., 1996). Infected amoebae may contain hundreds of Legionella
cells, which, when released from the amoeba, could allow a large number of bacteria to reach the

Draft - Technologies for Legionella Control:
Scientific Literature Review                       5                                 October 2015

-------
lungs (Buse and Ashbolt, 2012; Ohno et al., 2008; Berk et al., 1998; Kwaik et al., 1998; O'Brien
and Bhopal, 1993). A study by Berk et al. (1998) also showed that protozoa can release vesicles
(membrane-bound, sack-like structures within a cell) of respirable size containing live L.
pneumophila. The vesicles are resistant to freeze-thawing and sonication (a procedure that uses
sound waves to break cells), and the bacteria within the vesicles are highly resistant to biocides.

Another survival mechanism ofLegionella spp. is their ability to enter a viable but not culturable
(VBNC) state. Bacteria in a VBNC state fail to grow on culture media, where they would
normally grow, yet are still alive and could cause disease (Oliver, 2010). Numerous chemical
and environmental factors have been reported to induce a VBNC state, including nutrient
starvation, extreme temperatures, high salt concentrations, low oxygen concentration, heavy
metals and chemical treatment (including water disinfection) (Ducret et al., 2014; Aileron et al.,
2013; Oliver, 2010; Kana et al., 2008; Colbourne and Dennis, 1989). Studies suggest that
bacteria in the VBNC state can maintain their infectivity, multiply in their hosts and recover their
ability to grow on solid media (Ducret et al., 2014; Aileron et al., 2013; Oliver, 2010; Steinert et
al., 1997).

     1.3  Legionella Occurrence and Risk from the Distribution System and Building

          Water System

Building water systems have been identified as sources ofLegionella infection (Stout et al.,
1992; Muder et al., 1986), particularly through exposure via showers and hot water systems.
Within healthcare facilities such as hospitals and nursing homes drinking water is the most
common source  of exposure (Lin et al., 201 la). Exposure to Legionella has also been associated
with other types of building water systems (e.g., hotels and other buildings with complex water
distribution systems) (Silk et al., 2012; Hung et al., 1993; Tobin et al.,  1981a and 1981b).

L. pneumophila  has been found in the biofilms of water mains in distribution systems, although
proliferation has not been shown (Armon et al., 1997; States et al., 1990). Legionella spp. are
known to occur in finished water from water treatment plants and therefore, drinking water is
known to be a source of Legionella found in building water systems (Donohue et al., 2014;
Schaechter et al., 1998). Section 1.2.3 discusses optimal conditions for Legionella growth in the
distribution system.

Several surveys  have found Legionella in building water systems, including buildings that had
not been linked to recognized outbreaks:

    •  Donohue et al. (2014) used two quantitative polymerase chain reaction (qPCR)3 assays to
        evaluate  incidence of L. pneumophila serogroup 1 in 272 water samples collected in 2009
        and 2010 from 68 public and private cold drinking water taps across the United States. L.
       pneumophila serogroup 1 was detected in 47 percent of the taps.
3 A quantitative polymerase chain reaction assay detects a specific gene target known to be associated with a
specific genus/species/serogroup but it cannot distinguish between viable and nonviable cells (Donohue et al., 2014).

Draft - Technologies for Legionella Control:
Scientific Literature Review                        6                                 October 2015

-------
    •   Wadowsky et al. (1985) found that naturally occurring L. pneumophila multiplied at a
       temperature between 25 and 37 degrees C, at pH levels of 5.5 to 9.2, and at
       concentrations of dissolved oxygen of 6.0 to 6.7 mg/L.

    •   Wadowsky et al. (1982) sampled showerheads, shower pipes, and water and sediment
       collected from the bottom of hot water tanks in 11 buildings, including five homes and
       three hospitals. L. pneumophila serogroups 1, 5 and 6 were isolated from the drinking
       water fixtures in seven buildings including 1 of the 5 homes. Legionellae were also
       present in water and sediment in hot water tanks maintained at temperatures from 39 to
       54 degrees C (102.2 to 129.2 degrees F), but not found in tanks maintained between 71
       and 77 degrees C (between 159.8 and 170.6 degrees F). The authors hypothesized that hot
       water tanks are the major source and seed of L. pneumophila in building water systems.

    •   Tobin et al. (1981b) conducted a survey of 31 building water systems in hospitals and
       hotels, 6 of which were associated with sporadic cases or outbreaks of Legionnaires'
       disease. For the 6 buildings (hospitals  and hotels) associated with cases of Legionnaires'
       disease, the study found L. pneumophila in all of the building water systems and in the
       cooling water for each of the 3 buildings with cooling towers. For buildings that had not
       previously experienced an outbreak, the study found L. pneumophila in 4 out of 24 taps
       or showers, 3 out of 9 cooling towers,  and  1 out of 15 storage tanks.

    1.4   Regulatory Context

EPA regulates Legionella under the Surface Water Treatment Rule (SWTR). The SWTR has
treatment technique requirements to control for Giardia and viruses. The SWTR's treatment
technique requirements presume that if sufficient treatment is provided to control for Giardia
and viruses (i.e., 3-log inactivation of Giardia and 4-log inactivation  of viruses), then Legionella
risks will also be controlled. In addition, the Revised Total Coliform Rule and the Ground Water
Rule have treatment technique requirements that address bacteria, which provide  some control of
Legionella. All of these rules apply to public water systems (PWSs).

Building water systems may or may not be subject to federal drinking water regulations under 40
CFR Part 141. States and/or local governments may have drinking water standards for such
systems even if federal regulations do not apply. To ensure adequate public health protection,
federal and state oversight may be needed  since adding certain technologies in a building water
system could impact the chemical and microbial quality of the water within the system. The EPA
issued guidance that primacy agencies may use as they make regulatory implication decisions
OJSEPA.  1976. USEPA. 1990).

A determination of which technology is best suited for a particular building water system is case-
specific in part due to the complex and diverse nature of building water systems. This document
does not recommend the addition of treatment nor the installation of any of the technologies
discussed herein; however, it does provide information regarding the operational requirements
that regulated PWSs must comply with. This information is included only to provide the reader
with a comprehensive understanding of the technologies.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       7                                 October 2015

-------
Building owners who are considering adding treatment to their building water systems may wish
to consult with their water supplier (i.e., PWS) to better understand any potential water quality
issues before making treatment related decisions. If a decision to add treatment to the building
water system seems likely, EPA advises building owners to consult with their primacy agency
for any specific requirements that may apply before they add any treatment.

2   Multi-Barrier Approaches and Technologies to Control Legionella

    2.1   Overview of Current State of Knowledge

The following sections  of this document describe multi-barrier approaches  and technologies for
controlling Legionella in building water systems. The information presented is based on the
references reviewed during the preparation of this document. Section 2.2 introduces multi-barrier
approaches as a framework for identifying and prioritizing hazards within a particular building
water system and determining the specific control measures for each priority hazard. Section 2.3
introduces several commercially available technologies that show some effectiveness in
mitigating potential exposure to Legionella in building water systems, including chlorine,
monochloramine, chlorine dioxide, copper/silver ionization (CSI), ultraviolet (UV) light
disinfection and ozone. For each technology, the document provides background information,
general characterization of its effectiveness against Legionella, potential water quality issues,
and operational conditions (including monitoring frequency and location).  This document does
not rank or recommend any one technology over another. The information in Section 2.3 is
presented in the context of national drinking water requirements. Applicability of such
requirements to a building water system would need to be determined by the building water
system operator in  consultation with the primacy agency and/or water supplier.

In Section 3, other  strategies (i.e., remediation methods) are discussed, including emergency
superheat-and-flush disinfection, shock hyperchlorination and point-of-use (POU) filtration. This
section summarizes what is currently known about the performance of these individual
technologies for controlling the occurrence of Legionella bacteria and other waterborne
pathogens in buildings.

In general, all of the technologies discussed in this document have been shown to offer some
degree of effectiveness  against Legionella. However, the long-term eradication of Legionella
from a building water system has not been demonstrated consistently with any of these
technologies. Complex  plumbing systems, such as those found in a multi-story building, may
have areas where there is less exposure to disinfectants, which could provide opportunities for
bacteria to grow. Legionella bacteria can be found in biofilms or in stagnant sections of the
plumbing system. The effectiveness  of a technology against Legionella in biofilm or Legionella
ingested by amoebae is often cited as a concern.

The maintenance of a disinfectant residual throughout the system is critical for the effectiveness
of chlorine, monochloramine, chlorine dioxide and CSI treatments. Maintaining a disinfectant
residual provides increased protection in the event Legionella is released into the building water
system (e.g., sloughing off of biofilm material containing Legionella) or enters a building water
system through the PWS distribution system. Ozone and UV disinfection do not produce a
disinfectant  residual. Therefore, water treated with these methods, in some  cases, may be

Draft - Technologies for Legionella Control:
Scientific Literature Review                       8                                 October 2015

-------
susceptible to subsequent contamination. For these reasons, more than one type of treatment or
control measure may be necessary to inhibit Legionella growth in a building water system
(Department of Veterans Affairs, 2014). The use of multi-barrier approaches is further discussed
in Section 2.2.

The effectiveness of a particular technology is dependent upon building-specific characteristics
such as pipe material, age and condition; water usage rates and water age; and water quality
parameters (e.g., pH, hardness, organic contaminants, inorganic contaminants, types of
waterborne pathogens). Therefore, decision makers may want to consider the specific conditions
of each building water system before making a decision and ensure that the conditions are
adequate for the selected approach.

The physical and chemical characteristics of the finished water have an impact on the
effectiveness of all the treatment technologies discussed, albeit not to the same degree. For
example, chlorine and chlorine dioxide disinfectant residuals may be difficult to maintain as the
water temperature increases due to faster reaction with organic materials or pipe surfaces. In
contrast, temperature has little impact on the effectiveness of copper or silver ions. The pH of the
finished water will significantly impact the effectiveness of chlorine, monochloramine and
copper and silver ions, but it will have less of an impact on the effectiveness of chlorine dioxide.
Other physical parameters (such as turbidity) and chemical constituents (such as chlorides and
dissolved organic carbon) can also affect the performance of specific technologies. These issues
are covered in more detail in Section 2.3.

Ensuring proper maintenance is a priority for all of the technologies discussed. Failures of
technologies put in place to protect building inhabitants from exposure to Legionella have
resulted in outbreaks (CDC, 2013b). Safety concerns also exist for most of the technologies
(USEPA, 1999b, 1999c). The use of strong oxidants such as chlorine requires proper handling to
avoid adverse health risks. The Stage 1 Disinfection Byproduct Rule (DBPR) requires PWSs
using chlorine, monochloramine and chlorine dioxide to maintain disinfection byproduct (DBF)
concentrations below levels that can cause negative human health implications (USEPA, 1998;
Rohr et al., 1999; States et al., 1998). These water quality issues are discussed in Section 2.3.

Unless a legionellosis outbreak occurs, the decision to employ additional treatment is often
difficult for building owners. Some building owners choose to install supplemental disinfection
treatment systems as a preventative measure based on economic, insurance or marketing reasons.
The detection of Legionella bacteria in finished water samples from a building is likely the most
common reason some facilities may choose to add treatment.

The CDC does not recognize a safe level of Legionella and recommends certain preventative and
corrective actions in health facilities that care for patients who are at higher risk for Legionella
infection (CDC, 2003). However, because of the ubiquitous nature of environmental Legionella,
a robust response to every positive test result is likely unnecessary and could be both costly and
damaging to the building infrastructure. Outbreaks of legionellosis have occurred when this
metric was applied. In some instances, risks associated with the building water system (including
Legionella) can be addressed by taking measures other than the addition of treatment after
careful analysis of the particular conditions of the building water system. Thus, decision makers
may want to consider the specific conditions of a building water system to better inform
decisions made in response to detection of Legionella. Multi-barrier approaches  commonly

Draft - Technologies for Legionella Control:
Scientific Literature Review                       9                                  October 2015

-------
include consideration of parameters such as the building water system layout, water processes
such as heating and water softening, methods of transmission such as showers and spas, and the
risk level of the potentially exposed population based on age and immune status.

    2.2  Multi-Barrier Approaches

    2.2.1   Background

Multi-barrier approaches refer to programs that systematically apply risk management principles
to reduce biological (including Legionella), chemical and physical risks associated with building
water systems. Different names  are used throughout the literature to describe multi-barrier
approaches. Some examples of multi-barrier approaches include water management programs
(WMPs), hazard analysis and critical control point programs (HACCP), and water safety plans
(WSPs).

The HACCP concept was established in the early 1960s by The Pillsbury Company in
coordination with the National Aeronautic and Space Administration and the U.S. Army
Laboratories. It was created to ensure the safety of food from microbiological hazards for
astronauts working in space (Mortimore and Wallace, 2001). Beginning in the mid-1970s,
HACCP principles were applied to the food industry as a preventative approach for  addressing
biological, chemical and physical hazards. This approach to food safety was recognized by WHO
as being essential for controlling foodborne disease. In 1993, the Codex Alimentarius
Commission food  code, established by the Food and Agriculture Organization of the United
Nations and the World Health Organization (WHO), adopted the HACCP approach (FAO,
1998). After seeing the success of HACCP in the food industry, water utilities began to
implement the HACCP approach. The process for using HACCP in a water system was
originally described within a food journal, Food Control, in 1994 (Havelaar, 1994).

HACCP can be viewed as a continuous multi-barrier approach for protecting finished water and
building water systems from hazards that may occur. While many water systems use some
aspects of the HACCP approach, implementing a full HACCP program that evaluates an entire
system in detail ensures the highest level of public health protection (Deere and Davison,  1998).

WSPs are also considered a comprehensive risk-management approach; WSPs use multiple
barriers to ensure public health protection from the source to the tap (WHO, 2011). The use of
HACCP in the water industry was the basis for the development of WSPs by the WHO (Figueras
and Borrego, 2010; WHO, 2005).

The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE)
Standard 188 describes a multi-barrier approach that establishes minimum legionellosis risk
management requirements for building water systems. ASHRAE uses the term water
management programs to describe the multi-barrier approach (ASHRAE, 2015).

The application of any multi-barrier approaches,  such as WMP, HACCP or WSPs can be
beneficial for water systems and building water systems in protecting water quality and public
health in general. For more information on WMPs please refer to the ASHRAE Standard  188
(ASHRAE, 2015). For more information on WSPs, the reader is referred to WHO documents
(WHO 2011; WHO 2005). For information about the HACCP, WMP and WSP elements

Draft - Technologies for Legionella Control:
Scientific Literature Review                      10                                October 2015

-------
elements, see the Appendix of this document. Slight variations can be observed in the elements
or steps described by each approach. The EPA does not make any specific recommendation
regarding the use of any particular approach. The EPA advises building water system operators
and owners to determine which approaches may be more suitable to their specific needs or
whether a combination of approaches is appropriate.

    2.2.2   Applications of Multi-Barrier Approaches

Water system managers have found success in implementing multi-barrier approaches such as
WMP, HACCP and WSP, similar to the successes seen in the food industry for many years. In
1999, Brisbane, Australia, employed HACCP as a means of protecting the water treatment
process, storage and distribution system (Gray and Morain, 2000).  A system in Melbourne,
Australia, implemented HACCP in 1999 and improved outcomes including streamlined work
procedures, a net decrease in customer complaints, and a better understanding of the quality
issues (Mullenger et al., 2002). Five full-scale HACCP applications in Australian water
distribution systems resulted in reductions in customer complaints  and water quality incidents.
Other improvements attributed to HACCP were noted in work processes, documentation and
recordkeeping, and the system's capability to demonstrate due diligence. Pilot-scale applications
in U.S. distribution systems showed that HACCP was feasible and practical, but the time and
resource requirements for implementing the plans were greater than expected (Martel et al.,
2006). Researchers in Japan have concluded that HACCP ensures safe and high quality drinking
water; they also have proven success with safe water through the previous uses of HACCP for
bottled water and ice production (Yokoi at al., 2006). In Iceland, an estimated 68 percent of the
population consumes drinking water from systems with WSPs. In a 2008 evaluation of water
systems, the authors noted that compliance with drinking water standards improved considerably
upon implementation of HACCP (Gunnarsdottir and Gissurarson, 2008).

Implementing multi-barrier risk management concepts for building water systems has also been
shown to be successful. The Occupational Safety and Health Administration (OSHA) recognizes
the importance of having controls for building water systems in place, as under the right
conditions any water source can be a source of disease and illness (OSHA, 1999). ASHRAE also
recognizes the importance of risk management for building water systems in its Standard  188,
Legionellosis: Risk Management for Building Water Systems (ASHRAE, 2015). NSF
International is developing a draft standard based on application of HACCP to building water
systems for Legionella control. In addition to applying multi-barrier risk management concepts
to existing building water systems,  building designers can also use these concepts in the design
phase for new building water systems to help reduce and control hazards (Krageschmidt et al.,
2014).

Multi-barrier approaches have proven to be  effective for controlling the growth of significant
pathogens in building water systems (Gillilland et al., 2014), as documented in the following
case studies:

   •   In 2004, a university clinic in Germany adopted the WSP concept based on HACCP. One
       immediate success this clinic noted was the correction of an infrastructural failure that
       was identified during the process. Three years after implementation, two additional
       improvements were noted: a lowered rate of sepsis in very low birth weight neonates and


Draft - Technologies for Legionella Control:
Scientific Literature Review                       11                                October 2015

-------
       no cases of nosocomial (hospital-acquired) Legionnaires' disease since implementation
       (Dyck et al., 2007).

    •   In Minnesota, the Mayo Clinic used HACCP principles to build a water management
       program for its multi-campus healthcare facilities. During implementation of HACCP the
       clinic found distribution piping design issues and determined that additional hazard
       controls were needed. The water management program improved the awareness of water
       quality issues (Krageschmidt et al., 2014).

    •   Evaluations of outbreaks of Legionnaires' disease have shown system deficiencies to be
       contributing factors to outbreaks (CDC, 2013b). The implementation of HACCP plans or
       WSPs may identify and help to correct these deficiencies.

In addition to applying multi-barrier risk management concepts to existing building water
systems, building designers can also use these concepts in the design phase for new building
water systems to help reduce and control hazards (Krageschmidt et al., 2014). For example,
designing a system to minimize water age and dead-end pipelines may limit the occurrence of
waterborne pathogens. Another example is to exclude  the use of decorative fountains which can
be a source of Legionella; the Veterans Administration has concluded that they should not be
included in healthcare interior designs (Department of Veterans Affairs, 2012).

Addition of treatment as part of a multi-barrier approach into a building water system's operation
and maintenance program could have regulatory implications. The EPA advises building owners
who are considering adding treatment to consult with their water supplier and primacy agency
for any specific considerations or requirements that may apply.

    2.2.3  Environmental Testing

Environmental testing involves collecting water  samples from the building water system and
analyzing for L. pneumophila or other hazards of concern, as well as for water quality parameters
(pH, temperature, disinfectant residual) that may indicate efficacy of treatment performance and
overall water quality. Environmental testing may be performed in the context of an outbreak
investigation in order to determine the source and stop transmission or as part of a multi-barrier
Legionella prevention plan such as a WMP, HACCP or WSP (ASHRAE, 2015; Sidari et al.,
2014; Kozaketal., 2013).

Using Legionella test results as a measure of risk for disease transmission is problematic due to
many knowledge  gaps, including but not limited to, infectious dose, efficiency of aerosol-
generating devices, susceptibility of potential hosts, and virulence of the strain. While detection
of Legionella in a building water system may indicate  conditions conducive to Legionella
persistence, it is also clear from worldwide studies that the strains of Legionella most often
detected during routine environmental testing are rarely the strains that cause disease (Kozak-
Muiznieks et al., 2014; Euser et al., 2013;  Harrison et  al., 2009, Kozak et al., 2009; Doleans et
al., 2004). The lack of reliable and definitive human infectious dose information for Legionella
makes environmental monitoring results difficult to translate into action levels that can  directly
reduce human health risks (Buse et al., 2012;  Schoen and Ashbolt, 2011; Storey et al., 2004a;
Storey et al., 2004b; O'Brien and Bhopal,  1993;  Fitzgeorge et al., 1983).

Draft - Technologies for Legionella Control:
Scientific Literature Review                      12                                 October 2015

-------
Guidelines on routine environmental testing for Legionella vary among different agencies,
including the Veterans Health Administration (VHA), CDC, and WHO (Barker et al., 2015).
VHA recommends routine environmental testing for Legionella in VHA facilities (VHA, 2014).
CDC and WHO recognize that environmental Legionella counts alone cannot predict the
probability of human infection from a water system because other factors, such as the exposure
dose and level of host susceptibility, contribute to the likelihood of infection (Bartram et al.,
2007; Sehulster and Chinn, 2003). Despite the limitations of environmental monitoring, WHO
continues to recommend using Legionella testing as one way to validate a WSP (Bartram et al.,
2007). Similarly, a WMP or HACCP approach for protecting a building water system might
involve routinely monitoring water temperature, disinfectant residual levels, and the functioning
of any other treatment system,  leaving the more complicated testing for Legionella for validation
of the plan being used. Current challenges to environmental testing for Legionella include the
following:

    •   Despite a number of published procedures for the detection of Legionella in water
       samples, standard culture methods remain limited by their sensitivity and unreliability in
       detecting a wide range of Legionella spp. on a consistent basis (Buse et al., 2012) and
       detecting VBNC Legionella (Oliver, 2010). Further, the time it takes to receive results
       limits the utility of testing.

    •   Wide fluctuations occur in Legionella testing results from the same tap on a daily basis
       and from the same water sample between laboratories (Lucas et al., 2011).

    •   There is a lack of standardized protocols for the selection of sampling sites and the
       frequency of sampling (Lucas et al., 2011; Bartram et al., 2007).

If a decision is made to conduct routine environmental testing for Legionella as part of a multi-
barrier approach, a building-specific sampling plan should be developed that specifies the
location of sampling sites, the type of samples, the frequency of sampling, the sample collection
method and the sample analysis method (Krageschmidt et al., 2014). However, there is no
consensus on how many and which types of samples to take (e.g., bulk water or biofilm), nor
how often to perform the  sampling in order to accurately assess the risk from Legionella.

    2.3   Technologies

    2.3.1  Chlorine

2.3.1.1  Background

Chlorine and chlorine-based compounds are disinfectants that can serve the dual role of
efficiently inactivating microorganisms during water treatment, as well as maintaining the
quality of the water as it flows  from the treatment plant to the consumer's tap (Calomoris and
Christman, 1998). Chlorine is a powerful oxidant that effectively inactivates a large variety of
microbial waterborne pathogens,  including those that can cause typhoid fever, dysentery, cholera
and Legionnaires' disease.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       13                                 October 2015

-------
Chlorine is added to drinking water as elemental chlorine (chlorine gas), sodium hypochlorite
solution or dry calcium hypochlorite. Chlorine as sodium hypochlorite is the form of disinfectant
most often applied in buildings (Rosenblatt and McCoy, 2014). Chlorine can be applied by
facilities for routine treatment of both hot and cold domestic water; it can be applied to the cold
and hot water tanks or to the entire distribution system. Chlorine can also be used at high doses
for emergency disinfection of potable water systems through shock chlorination (also called
shock hyperchlorination). Shock chlorination is covered in more detail in Section 3.1.2.

For chlorine to be effective against microorganisms, it must be present in sufficient
concentration, and it must have adequate time to react (Calomoris and Christman, 1998).  This
combination of concentration and reaction time is expressed as C (mg/L) x T (min), or CT. For
continued protection against potentially harmful organisms in distribution systems or building
water systems, some level of chlorine needs to be maintained after the initial application.  The
remaining chlorine is known as residual chlorine.

The addition of chlorine to water creates two chemical species that together make up "free
chlorine." These species, hypochlorous acid (HOC1, electrically neutral) and hypochlorite ion
(OC1", electrically negative), behave very differently. Hypochlorous acid is more reactive than
the hypochlorite  ion and is also the stronger disinfectant and oxidant.  The ratio of hypochlorous
acid to hypochlorite ion in water is determined by pH. At low pH (6-7), hypochlorous acid
dominates, while at high pH (>8.5) the hypochlorite ion dominates. Thus, the pH of the incoming
water may be a factor when deciding upon the use of chlorine as a disinfectant, or in the
engineering design when addressing issues such as CT for the target organism(s).

Chlorine was first used as a primary disinfectant of drinking water in  Jersey City, New Jersey,  in
1908. Chlorine is widely credited with virtually eliminating outbreaks of waterborne disease in
the United States and other developed countries (Calomoris and Christman, 1998). The use of
chlorine to control microbes has the  lowest production and operating  costs  of any disinfectant,  as
well as the longest history for large continuous disinfection operations. Among PWSs that
disinfect, chlorine is the most commonly used disinfectant (AWWA Disinfection Systems
Committee, 2008).

2.3.1.2  Characterization of Effectiveness against Legionella

Both laboratory and full-scale studies have been conducted to assess the effectiveness of chlorine
against Legionella. These studies included a range of physical and chemical water conditions
such as chlorine  dose and residual levels, temperature and pH. Lin et  al. (2002) reviewed
available literature on the efficacy of various disinfectants against Legionella; findings related to
chlorine disinfection include the following:

           o  Relatively high doses of chlorine (2-6 mg/L) were  needed for continuous control
             of Legionella in water systems.
           o  The effectiveness of chlorine increased with temperature, although chlorine
             residual decay also increased.
           o  The association of Legionella with protozoa required much higher doses of
             chlorine for inactivation. Lin et al. (2002) noted that this association with
             protozoa may explain why chlorine can suppress Legionella in water systems but
             cannot usually prevent regrowth of Legionella.

Draft - Technologies for Legionella Control:
Scientific Literature Review                       14                                 October 2015

-------
The laboratory studies that follow examined the effectiveness of chlorine in inactivating
Legionella under a range of pH, temperature and chlorine residual levels. Results showed a wide
range of CT values needed for all inactivation levels.

    •   Kuchta et al. (1983) studied the effects of various chlorine concentrations, temperatures
       and pH levels on Legionella in tap water. The chlorine residuals used (0.1 and 0.5 mg/L)
       were consistent with residual levels that would be expected in PWSs. The ranges of pH
       and temperature conditions evaluated are shown in Exhibit 2-1. Results show that lower
       CT is required for higher temperature and lower pH. The authors noted that contact times
       for the clinical and other environmental sources of Legionella were as long as, or longer,
       than those required for river samples, although long contact times were needed regardless
       of serogroup or origin. The authors concluded that low chlorine concentrations (0.1
       mg/L) allowed Legionella to survive for relatively long periods of time. Increasing the
       total chlorine concentration predictably enhanced the bactericidal effect, resulting in a 99
       percent (2-log) kill within the first 5 minutes at a concentration of 0.5 mg/L.

   Exhibit 2-1: Kuchta et al. (1983) findings on CT values (min-mg/L) for 2-log (99
                percent) reduction of L. pneumophila using chlorine
Temperature in degrees C
(in degrees F)
4 (39.2)
21 (69.8)
32 (89.6)
pH6.0

0.5

pH7.0

1-6
3.2
pH7.6
6-9
4
<3
Source: Kuchta et al. (1983)

    •   Jacangelo et al. (2002) conducted laboratory studies to examine the efficacy of current
       disinfection practices (e.g., chlorine dioxide, free chlorine and monochloramine) for
       inactivation of waterborne emerging pathogens including Legionella. Chlorine doses of
       1.0 to 4.0 mg/L were used. Three different temperatures (5, 15 and 25 degrees C, or 41,
       59 and 77 degrees F, respectively) and three different pH (6.0, 7.0 and 8.0) values were
       examined. Results are presented as CT (min-mg/L) values. The observed CT values for
       2-log (99 percent) reduction of L. pneumophila are shown in Exhibit 2-2. These CT
       values were at least an order of magnitude higher than those reported by Kuchta et al.
       (1983). The wide range of CT values reported in the literature could be due to different
       water quality conditions and test protocols used for inactivating Legionella.

 Exhibit 2-2: Jacangelo et al. (2002) findings on CT values (min-mg/L) for 2-log (99
                percent) reduction of L. pneumophila using chlorine
Temperature in degrees C
(in degrees F)
5(41)
15(59)
25 (77)
pH6.0
>50 to >320
100to>320
40 to 500
pH7.0
50 to 250
60 to >320
100 to 160
pHS.O
250 to >1, 000
25to>710
130 to 250
Source: Jacangelo etal. (2002)
Draft - Technologies for Legionella Control:
Scientific Literature Review
15
October 2015

-------
The following pilot studies evaluated the efficacy of chlorine disinfection for inactivating
Legionella without co-occurring microbial organisms. Both studies were completed using warm
water conditions.

    •   Muraca et al. (1987) compared chlorine, heat, ozone and UV for inactivating Legionella
       in a model building water system. A suspension of Legionella was added to the system
       and allowed to circulate. Chlorine disinfection consisted of maintaining a residual
       concentration between 4 and 6 mg/L by multiple additions of chlorine. Chlorine
       experiments were conducted at 25, 43 and 45 degrees C (77, 109.4 and 113 degrees F,
       respectively). Continuous chlorination at a dose of 4 to 6 mg/L resulted in  a 5- to 6-log
       decrease of L. pneumophila in six hours. Chlorine disinfection at 43 degrees C (109.4
       degrees F) inactivated L. pneumophila more reliably and completely than disinfection at
       23 degrees C (73.4 degrees F). Due to thermal decomposition of chlorine residual,
       approximately 120 percent more chlorine was needed to maintain a residual of 4-6 mg/L
       at 43 degrees C (109.4 degrees F). The authors noted that in  addition to the higher doses
       required to overcome residual decomposition, a drop in chlorine levels or failure of
       chlorination equipment could allow Legionella to survive. As a  result, the authors
       concluded that chlorination of hot water systems is more difficult to regulate than that of
       cold water systems.

    •   Saby et al. (2005) tested the efficiency of several disinfectants in a hot water system pilot
       unit. The pilot unit was supplied by tap water pre-heated to 30 degrees C (86 degrees F).
       Legionella-contaminated water was mixed with the tap water before heating.
       Colonization of the biofilm by Legionella was found after seven weeks. After
       colonization of pipes in the pilot unit, various treatments were tested. Shock
       hyperchlorination at 50 mg/L of free chlorine residual for 12 hours was found to be very
       effective in reducing Legionella in the water; however, the pipe networks were
       recolonized in three to four weeks. The authors stated this could be explained by the
       inefficiency of shock hyperchlorination treatment on bacteria in biofilms. Continuous
       chlorine at a dose of 3 mg/L for two periods of four weeks was  also examined. The
       results showed that this treatment was very effective at maintaining viable  bacteria,
       including Legionella, at low levels. However, a malfunction of the chlorination system
       resulted in a positive result for Legionella within 28 hours. The  authors concluded that
       continuous chlorination allows only for containment of Legionella and that technical
       problems with treatment could result in rapid recolonization. Temperature  control at 40
       degrees C (104  degrees F) and 55  degrees C (131 degrees F) was also evaluated as part  of
       this study. While temperature control at 55 degrees C was the best technical and
       economic solution to Legionella control, continuous chlorination was also  a good
       solution.

The interaction of Legionella with co-occurring organisms can affect the efficacy of chlorine in
the inactivation of Legionella. The following laboratory studies evaluated the effects of co-
occurring amoebae on Legionella inactivation by chlorine disinfection:

    •   In a study of the interaction of thermotolerant amoebae and Legionella, Storey et al.
       (2004a) evaluated the efficacy of heat and chlorine as disinfectants. The study found that
       a 2-1 og (99 percent) reduction of free-living (planktonic) L. pneumophila was achieved  at

Draft - Technologies for Legionella Control:
Scientific Literature Review                       16                                 October 2015

-------
       30 minutes with free chlorine concentrations of 1 mg/L and 2 mg/L (at 37 degrees C, or
       98.6 degrees F). A 3-log (99.9 percent) reduction of L. pneumophila was achieved after
       10 minutes with a free chlorine concentration of 10 mg/L (at 37 degrees C, or 98.6
       degrees F). The efficacy of free chlorine in the reduction of Acanthamoeba castellanii (an
       amoeba)-bound L. erythra was also evaluated. A free chlorine dose of 1 mg/L achieved
       less than 0.5-log reduction at contact times of 60 minutes or less, whereas a 2 mg/L dose
       resulted in a 3-log reduction (99.9 percent) at contact times of >30 minutes (at 37 degrees
       C or 98.6 degrees F). A free chlorine dose of 10 mg/L and  contact time of 10 minutes
       achieved a 3.2-log reduction. The study found that the interaction of Legionellae and
       Acanthamoebae increased the resistance of Legionellae to thermal treatment and
       increased their sensitivity to chlorine. The authors also noted the tolerance of
       Acanthamoebae to high chlorine doses and thermal treatment. Both cysts retained their
       viability at free chlorine levels of 100 mg/L after 10 minutes and at free chlorine levels of
       less than 10 mg/L after 30 minutes. The authors cited a prior study by Kilvington and
       Price (1990) that found that cysts were able to maintain their viability at free chlorine
       concentrations of 50 mg/L or less.

    •   Dupuy et al. (2011) also investigated the interaction of amoebae and Legionella. The
       authors compared the efficiency of three oxidizing disinfectants (chlorine,
       monochloramine and chlorine dioxide). These disinfectants were used on three
       Acanthamoeba strains, L. pneumophila alone, and Acanthamoeba and L. pneumophila in
       co-culture. Chlorine efficiency was  evaluated at 30 degrees C (86 degrees F) and at 50
       degrees C (122 degrees F). An initial dose between 2 mg/L and 3  mg/L was applied with
       a residual free chlorine residual of 1 mg/L at the end of the treatment. Results were
       presented as CT (min-mg/L) values. Chlorine was found to inactivate all three strains of
       Acanthamoeba studied, both infected with L. pneumophila and not infected. At least a 3-
       log inactivation (99.9 percent) was obtained for all strains at a CT of approximately 60
       min-mg/L. There was a significant difference in inactivation between the strains of
       Acanthamoeba studied, with more than 3-log inactivation found at a CT of less than 10
       min-mg/L for one strain. Inactivation efficiency was slightly higher at 50 degrees C (122
       degrees F).

The following laboratory studies evaluated the effectiveness of chlorine when biofilm is present:

    •   In  a 1994 paper, de Beer et al. studied the degree to which  chlorine penetrates a biofilm
       based on bulk concentration.  For this study, biofilms consisting of Pseudomonas
       aeruginosa and Klebsiella pneumoniae were grown for one week, with a maximal
       thickness of 150-200 micrometers (|im). Transient chlorine concentration profiles were
       measured in biofilms with a microelectrode that was developed for the investigation and
       was sensitive to concentrations of chlorine in the micromolar range. The transient
       chlorine microprofiles showed slow chlorine penetration into the biofilm, with the rate
       dependent on the bulk concentration of chlorine. The penetration time exceeded 60
       minutes even at the highest concentration tested (0.36 millimolar  (mM)). The biofilm
       matrix, consisting of cells and extracellular polymeric substances, was determined to be a
       substrate for the chemical reduction of chlorine. Chlorine concentrations measured in
       biofilms were typically only 20 percent or less of the concentration of the bulk liquid.
       The microprofiles showed that following exposure to 2.5 mg/L chlorine for one hour,

Draft - Technologies for Legionella Control:
Scientific Literature Review                      17                                 October 2015

-------
       only the upper 100 jim of the cell clusters was penetrated by chlorine. Findings showed
       that the limited penetration of chlorine into the biofilm (as determined by penetration
       depth and rate of penetration) is likely a key factor influencing the reduced efficacy of
       chlorine against biofilms compared to its effectiveness against planktonic cells. Rapid
       regrowth after chlorine treatment may have originated from areas within biofilms that are
       highly resistant to chlorine.

    •   Loret et al.  (2005) expanded on the de Beer et al. (1994) study by using a simulated
       distribution system consisting of pipe loops to compare disinfectants for Legionella
       control in biofilms in building water systems. The pilot unit also included piping off of
       the main pipe loop to simulate areas at the ends of a water system (dead ends) with
       stagnant flow conditions. Tap water and injection of cultured natural Legionella strains
       were used to establish biofilms. Low temperature (35 degrees C, or 95 degrees F) relative
       to hot water systems and low water velocity, as well as high retention times, were
       maintained to favor the growth of Legionella and biofilms. Each pipe loop was treated
       with one of the studied disinfectants for three months. The loop receiving chlorine was
       maintained with  a residual dose of 2 mg/L. Each type of disinfectant used in the study
       displayed rapid initial results in the treated loops, with Legionella populations decreasing
       to undetected levels (less than 500 CFU/L, or colony-forming units per liter) within three
       days of treatment, in all cases. However, Legionella remained undetected over the whole
       study period only with sodium hypochlorite, electro-chlorination, chlorine dioxide and
       monochloramine. (Ozone and copper/silver allowed occasional re-emergence of
       detectable Legionella.) Ozone, electro-chlorination and chlorine treatments resulted in a
       reduction of biofilm thickness to below detection limits (<5 jim) after one week. A
       chlorine dosage rate of 2.5 mg/L removed biofilm better than a chlorine dioxide dosage
       rate of 0.5 mg/L. Flushing of the dead ends at a rate of 20 percent of the volume per day
       did not result in a significant reduction in  Legionella. After a single complete flushing, all
       so-called dead end sections of piping returned to their initial  contamination level within
       24 hours. The study concluded that chlorine and chlorine dioxide were the most effective
       treatment methods in this study (as compared to ozone, monochloramine and
       copper/silver). Ozone was found to be effective for controlling the planktonic and biofilm
       populations within the pipe loops but was ineffective within dead end sections.
       Monochloramine was found  to be ineffective for the amoebae and in the biofilm. The
       authors suggest that the experimental protocol did not allow for maintenance of a stable
       product and resulted in insufficient dosing in the pipe loops.

    •   Using copper and stainless steel coupons,  Cooper and Hanlon (2009) found that mature
       L. pneumophila biofilms (one and two months old) survived  a one-hour treatment with 50
       mg/L chlorine and continued to grow after treatment, reaching a population of 106 CFU
       per coupon (20 mm diameter disc). The authors also found that planktonic Legionellae
       were able to survive and persist at free chlorine concentrations of 0.5 mg/L.

Additional studies  that compare the effectiveness of other disinfectants to chlorine to control for
Legionella are cited in subsequent sections for various technologies.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       18                                 October 2015

-------
    •   In a study oi Legionella control in full-scale water systems of older hospital buildings in
       Rome, Italy, Orsi et al. (2014) evaluated the effectiveness of shock hyperchlorination and
       continuous chlorination over a five-year period. Thirty-eight buildings were studied and
       1,308 samples were analyzed for the presence ofLegionella. Samples were collected
       before and/or after several chlorination treatment scenarios (before and after shock
       hyperchlorination,  shock hyperchlorination followed by continuous chlorination) from
       cold water piping, mixed cold and hot piping, and hot water piping. Shock
       hyperchlorination was described as an applied concentration of 20-50 parts per million
       (ppm), and continuous chlorination was described as a continuously applied
       concentration of 0.5-1.0 ppm. The study found a significant association between the
       presence ofLegionella in the building drinking water systems and the lack of continuous
       chlorination following shock hyperchlorination. Isolation ofLegionella was more
       frequent in mixed water samples (20-40 degrees C (68-113 degrees F)) than in cold or
       hot water samples.  The authors concluded that continuous free chorine levels of 0.5 to  1.0
       mg/L resulted in significant reductions in Legionella counts in the old hospital water
       systems. However, this treatment did not completely eradicate Legionella.

    •   Lin et al. (1998a) reported that some hospitals that initially adopted chlorination
       converted to other methods of disinfection because of failure to control Legionella and
       corrosion of the building water system. Also, Casini et al. (2014) isolated Legionella
       strains more tolerant of free chlorine from a water system after years of chlorine
       treatment.

2.3.1.3  Potential Water  Quality Issues

Chlorine can react with organic material in the water to form DBFs in systems  with areas of low
flow or stagnant waters. Some DBFs have been shown to cause cancer and reproductive effects
in lab animals and may cause bladder cancer and reproductive effects in humans (USEPA, 2010).
In a simulated distribution system of pipe loops, Loret et al. (2005) found trihalomethane (THM)
levels >100 micrograms per liter (|ig/L), bromate levels >10 |ig/L, and chlorite levels >0.2 mg/L
with an applied chlorine dose of 2 mg/L. Orsi et al.  (2014) noted that special equipment was
needed in certain health care settings (e.g., dialysis, neonatal care) to reduce free  chlorine and
THM levels.

Continuous chlorination at high levels in building water systems can result in objectionable
tastes and odors along with irritation of skin, eyes and mucous membranes.

Continuous chlorination can contribute to corrosion, with associated leaks, in plumbing systems
and may require the simultaneous use of corrosion-inhibiting chemicals. Sarver et al. (2011)
reported that continuous hyperchlorination increased leaks by up to 30-fold,  consistent with
extensive laboratory work in soft higher-pH waters  (Sarver et al., 2011). In a study by
Grosserode et al. (1993), leaks first appeared in the  copper pipes of a water distribution system
about two years after installation of the chlorine injectors. Significant deterioration was noted
only in the hot water system. The addition of silicate corrosion inhibitors reduced the total
number of leaks per year by >80 percent.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       19                                 October 2015

-------
2.3.1.4  Operational Conditions

Parameter Conditions Indicating Operational Effectiveness

The efficacy of chlorination is affected by many factors, including chlorine concentration,
contact time, pH, temperature, buffering capacity of the water, concentration of organic matter,
and the number and types of microorganisms in the water system (in biofilms and free-living).
Sidari et al. (2014) reported that the typical concentration of supplemental chlorine is 2-4 mg/L
as free residual chlorine. Lin et al. (2002) reported that 2-6 mg/L of chlorine was needed for
continuous control ofLegionella in water systems.  The bactericidal action of the chlorine is
enhanced at higher temperatures and at lower pH levels.  The anti-microbial efficacy of chlorine
declines as pH increases >7, with significant loss of efficacy at pH >8. However, free chlorine is
degraded rapidly at elevated water temperatures, which is a concern for hot water chlorination
(Health Protection Surveillance Centre,  2009).

Operational Considerations

To help ensure its effectiveness, standard industry practices recommend how sodium
hypochlorite should be stored and handled to minimize decomposition of the product
(GLUMRBSPPHEM 2012). It should be stored in the original shipping containers or compatible
containers and sited away from direct sunlight in a  cool area. Feed rates should be regularly
adjusted to account for any losses in chlorine content during storage or handling.

NSF/ANSI Standard 60 certification can help ensure that the quality and effectiveness of water
treatment chemicals have been reviewed and found to be acceptable for potable water
applications. A facility considering application of chlorine gas as the form of chlorine to be used
for disinfection would also need to consider potential safety and security concerns. The Chemical
Facility Anti-Terrorism Standards include information on standard practices for storage and
handling of chlorine gas (Department of Homeland Security, 2015). Additional safety procedures
will likely be required for personnel training and equipment. Existing OSHA, state or local fire
authority regulations may apply and may need to be consulted.

Monitoring Frequency and Location

The Surface Water Treatment Rule (SWTR) (USEPA, 1989a) requires that all PWSs using
chlorine and using surface water or ground water under the direct influence of surface water
monitor for the presence of the residual  disinfectant in the distribution system or at the point-of-
entry (POE) to the distribution  system. The disinfectant level must be >0.2 mg/L at the POE and
detectable within the distribution system.

The Stage 1 Disinfectants and Disinfection Byproducts Rule (Stage 1  DBPR) requires PWSs that
use chlorine to maintain a maximum residual disinfectant level (MRDL) as running annual
average less than 4.0 mg/L (USEPA, 1998).

As stated in the SWTR, PWSs that use chlorine are required to monitor for combined or total
chlorine residual or heterotrophic plate count (HPC) bacteria in the distribution system at
locations that have been approved by the primacy agency. All these parameters could provide
Draft - Technologies for Legionella Control:
Scientific Literature Review                      20                                 October 2015

-------
operational information to indicate the need for chlorine dose adjustments, system flushing and
managing water age within finished water storage facilities.

Maintenance Needs

Operating and maintenance practices for chlorine disinfection systems include maintenance of an
appropriate disinfectant residual, regular system cleaning and flushing, inspections, and water
quality monitoring. Newly constructed or rehabilitated piping systems are cleaned and flushed
prior to initial disinfection. Routine flushing and water quality monitoring are recommended to
assure that adequate disinfectant levels are maintained throughout the building water system
(HSE, 2014; Rosenblatt and McCoy, 2014).

Since chlorine is recognized as being less effective than other disinfectants at penetrating and
controlling established biofilms, chlorination may not be effective if large amounts of scale and
sediment are present in the system. These solids are prone to biofilm formation and may need to
be removed by cleaning before effective disinfection can be achieved (HSE, 2014). Loret et al.
(2005) recommended flushing dead ends daily with disinfected water and removing building
finished water fixtures and pipes that are rarely used.

     2.3.2  Monochloramine

2.3.2.1  Background

The primary use of monochloramine (NIHbCl) in water systems is for residual disinfection to
maintain a disinfectant residual in the distribution system. Monochloramine has a more persistent
and stable disinfectant residual than chlorine (USEPA, 1994). It causes fewer unpleasant tastes
and odors in drinking water than other disinfectants (USEPA, 1994). Monochloramine has a
much lower disinfection efficacy than free chlorine (Symons, 1978) and if used as a primary
disinfectant it requires a much longer contact time. Often ammonia is added after chlorine has
acted as a primary disinfectant for a period of time, and the resulting monochloramine is used as
a residual disinfectant (USEPA, 1999b; USEPA, 1999c).

Monochloramine is effective for controlling bacterial regrowth and controlling biofilms due to its
ability to penetrate the biofilm, although excess ammonia can cause biofilm growth (USEPA,
1999c; LeChevallier et al., 1988a). Monochloramine and chlorine have different  mechanisms of
action; monochloramine is more specific, and chlorine reacts with a wider array of compounds.
When inactivating bacteria in the biofilm, monochloramine is able to penetrate, whereas chlorine
may get consumed through reactions that do not occur with monochloramine (Lee et al., 2011;
LeChevallier,  1988b). For equivalent chlorine concentrations, monochloramine was shown to
initially penetrate biofilm 170 times faster than free chlorine, and even after subsequent
application to a monochloramine-penetrated biofilm, free chlorine penetration was limited (Lee
et al., 2011). The mechanism of inactivation for chloramine is thought to involve inhibition of
proteins or protein-mediated processes such as respiration (USEPA, 1999c).

Monochloramine can be formed by first adding chlorine then ammonia or vice versa. Although
monochloramine is the dominant form produced under conditions typically found in a drinking
water system, two other forms of chloramines (dichloramine and trichloramine (nitrogen
trichloride)) can also be produced when excessive levels of hypochlorite are present or at low pH

Draft - Technologies for Legionella Control:
Scientific Literature Review                      21                                 October 2015

-------
levels (USEPA, 1994). Monochloramine is the preferred form of chloramine for use in drinking
water treatment due to fewer taste and odor issues and its disinfection efficacy. Monochloramine
is a colorless water-soluble liquid (Kirk-Othmer,  1979) with a freezing point at -66 degrees C (-
86.8 degrees F).

Monochloramine has been used in the treatment of drinking water for nearly 100 years (USEPA,
2009). It was first used in water treatment in the mid-1910s; the City of Ottawa first used
chloramines in 1915 due to the rising costs of bleach. Denver, Colorado, started using
monochloramine around the same time as a way to control organisms in the distribution system
(Symons, 1978). Its use gained popularity in the 1930s and 1940s but soon declined due to the
shortage of ammonia during World War II. The use of monochloramine has been increasing in
the past couple of decades due to concerns over DBFs associated with chlorine use (USEPA,
1999c).  As of 2009, 1 in 5 Americans were using drinking water treated with chloramines
(USEPA, 2009) and this usage rate is projected to increase due to implementation of the Stage 2
DBPR (Seidel et al., 2005; USEPA, 2005a).

2.3.2.2  Characterization of Effectiveness against Legionella

Laboratory studies  have used a wide range of CT  values under different water quality test
conditions for inactivating Legionella.

   •   Jakubek et al. (2013) evaluated inactivation of L. pneumophila in nuclear power plant
       cooling circuits with monochloramine formed by combining sodium hypochlorite and
       ammonia solution with a chlorine-to-ammonia mass ratio of 4.8 (at pH 7.5-8.5 and 25-35
       degrees C (or 77-95 degrees F)). The results showed 99.9 percent (3-log) inactivation of
      L. pneumophila with a CT range between  16.14ฑ3.07 min-mg/L and 64.88ฑ19.07 min-
       mg/L. The study also found that temperature, pH and initial bacterial concentration
       affected the ability of monochloramine to  inactivate Legionella. Increasing the
       temperature had a positive effect on monochloramine activity but a negative effect on the
       contact time required to inactivate 99.9 percent of the Legionella. Increasing the pH had a
       negative effect on monochloramine activity but a positive effect on the contact time
       required to inactivate 99.9 percent of the Legionella (Jakubek et al., 2013).

   •   Jacangelo et al. (2002) examined inactivation of waterborne emerging pathogens such as
      Legionella by selected disinfectants, including monochloramine. Pre-formed
       monochloramine was used at a target pH of 7.0. Two different temperatures (5 degrees C
       (41 degrees F) and 25 degrees C (77 degrees F)) and two different mass ratios of chlorine
       to ammonia (3:1 and 7:1) were examined.  The observed CT values for 99 percent
       inactivation (2-log reduction) of L. pneumophila ranged from >320 to >1,000 min-mg/L.
       At a water temperature of 5 degrees C (41 degrees F), the CT value at a 3:1 ratio was
       >1,000 min-mg/L and was >320 to >1,000 min-mg/L at a 7:1 ratio.  At a temperature of
       25 degrees C (77 degrees F), the CT was >630 to >1,000 min-mg/L at a 3:1 ratio and was
       >320 to >1,000 min-mg/L at a 7:1 ratio. These CT values were similar to CT values for
       Giardia inactivation under the same conditions (Jacangelo et al., 2002).

   •   Donlan et al. (2002) conducted a study with three different monochloramine
       concentrations (0.2 mg/L, 0.5 mg/L and 1.5 mg/L) and three different contact periods (15,

Draft - Technologies for Legionella Control:
Scientific Literature Review                     22                                October 2015

-------
       60 and 180 minutes). All scenarios involved a temperature of 30 degrees C (86 degrees F)
       and a pH of 7. A monochloramine concentration of 0.2 mg/L was ineffective for all
       contact periods. At the 0.5 mg/L concentration and 180 minute contact time, 99 percent
       ofLegionella was inactivated (i.e., 2-log removal). Using the 1.5 mg/L concentration of
       monochloramine,  99.9 percent ofLegionella was inactivated (i.e., 3-log removal) at 60
       and 180 minutes contact time (Donlan et al., 2002).

    •   A study conducted by Cunliffe (1990) evaluated Legionella contact time in a lab
       simulated model experiment. This study used a 2.5:1 chlorine-to-ammonia mass ratio
       prepared by mixing ammonium chloride with sodium hypochlorite at 30 degrees C (86
       degrees F) and pH 8.4-8.6. The average CT level for 99 percent inactivation was 15 min-
       mg/L. The results showed that Legionella was more sensitive to monochloramine than E.
       coli (Cunliffe, 1990).

The wide range of CT values reported in the literature  could be due to different water quality
conditions and different methodologies used for inactivating Legionella.

Several studies reported on the efficacy of monochloramine in controlling Legionella when
biofilm is present on pipe surfaces.

    •   Wang et al. (2012) evaluated the effects of disinfectant (chlorine and chloramine), water
       age (1 to 5.7 days), and pipe material (polyvinyl chloride, iron and cement) on multiple
       pathogens, including Legionella, using simulated distribution systems. Two sampling
       events occurred after six and 14 months. The results showed systems treated with
       chloramines  had higher levels of bacteria and protozoa at shorter water ages than systems
       treated with chlorine. Chloramine concentrations were depleted faster than chlorine due
       to nitrification of the chloramine. The effects of pipe type on pathogen growth mainly
       became evident after water age reached 5.7 days, after the majority of the disinfectant
       residual was depleted. Legionella was only detected during the 14-month sampling event
       in bulk water and at lower water ages for chloraminated systems.

    •   Dupuy et al.  (2011) evaluated the inactivation of both free and intracellular L.
       pneumophila (co-occurring with Acanthamoebd) using different disinfectants. The results
       showed no difference between the inactivation of both forms ofLegionella by
       monochloramine, while the other disinfectants  (chlorine and chlorine dioxide) were not
       as efficient in inactivating the intracellular Legionella.

    •   Loret et al. (2005) evaluated disinfectants and their effects on biofilm. They studied
       Legionella control in a pipe loop receiving continuously treated water. Monochloramine
       treatment was evaluated for one month. The ratio of chlorine to ammonia for
       monochloramine was 2:1, and an average dose of 0.5 mg/L was used. Planktonic
       Legionella decreased to undetectable levels after three days  and stayed undetectable for
       the remainder of the month. There were no viable Legionella in the biofilm after six days
       of treatment. Biofilm thickness increased with monochloramine treatment after one
       month of treatment, unlike with the other disinfectants (e.g., chlorine, chlorine dioxide).
       The study results showed that monochloramine was effective against Legionella, but it
       was not effective in removing the biofilm completely (Loret et al., 2005).

Draft - Technologies for Legionella Control:
Scientific Literature Review                      23                                 October 2015

-------
    •   Tiiretgen (2004) conducted a study of the efficacy of monochloramine and chlorine
       against biofilms using different contact times and different concentrations on a full-scale
       and model system cooling tower. Monochloramine was found to be significantly more
       effective against cooling tower biofilms than free chlorine. In both systems, a 3-log (99.9
       percent) reduction of heterotrophic biofilm bacteria was achieved using a
       monochloramine concentration of 1.5 ppm for a contact time of about 35 minutes.
       Monochloramine is unaffected by the elevated pH levels within the cooling towers,
       unlike chlorine. To completely remove biofilms from a cooling tower additional
       treatment would be needed, such as physical cleaning (Tiiretgen, 2004).

    •   Lee et al. (2011) and Pressman et al.  (2012) used microelectrodes to investigate the
       penetration of chlorine, monochloramine, oxygen and free ammonia in nitrifying biofilm.
       While this research clearly demonstrated that monochloramine had a greater penetration,
       the authors found this penetration did not necessarily translate to immediate viability loss.
       Even though free chlorine's penetration was limited compared to that of
       monochloramine, it more effectively (on a cell membrane integrity basis) inactivated
       microorganisms near the biofilm surface.  The authors also found that the presence of
       higher free ammonia concentrations allowed a larger biomass to remain active during
       monochloramine application, particularly the organisms deeper within the biofilm,
       leading to faster recovery in oxygen utilization when monochloramine was removed. The
       authors suggested that limiting the free ammonia concentration during monochloramine
       application would slow the onset of nitrification episodes by maintaining the biofilm
       biomass at a state of lower activity.

    •   Donlan et al. (2002) evaluated Legionella levels within a biofilm reactor. They found
       monochloramine to be more effective than chlorine in identical conditions for Legionella
       inactivation, leading the authors to conclude that monochloramine may be more effective
       for the inactivation of Legionella in drinking water distribution systems.

Several studies evaluated Legionella control in building water systems receiving water from a
distribution system where the treatment plant converted from chlorine to monochloramine.
Several studies evaluated the addition of monochloramine for the treatment of building water
systems. Other studies compared Legionella control in water systems using different disinfection
methods.

    •   Baron et al. (2014) studied the microbial ecology of a hot water system within a hospital
       following the introduction of monochloramine. Samples were taken three months before
       and immediately prior to the addition of an on-site monochloramine generation system
       and then every month for six months after the addition. Monochloramine levels were
       targeted at 1.5-3.0 mg/L as chlorine. Samples were taken at multiple sites within the
       hospital's hot water system and analyzed by three methods. The authors observed a shift
       in microbial ecology immediately after the addition of the disinfectant, and the number of
       operational taxonomic units significantly increased. Microbial ecology variation based on
       sampling location within the hospital's hot water system (including automatic and
       standard faucets) increased after the addition of monochloramine. There was a
       statistically significant increase in the relative abundance of genera associated with

Draft - Technologies for Legionella Control:
Scientific Literature Review                      24                                 October 2015

-------
       denitrification after the addition of monochloramine. Waterborne pathogen-containing
       genera were also examined. After the addition of monochloramine, an increase in counts
       of Acinetobacter, Mycobacterium, Pseudomonas and Sphingomonas were observed.
       Trends for Legionella counts varied but did not show an increase. The addition of
       monochloramine to the hospital's water system had an impact on the types and amounts
       of microorganisms found in the hot water system.

   •   Whiley et al. (2014) measured Legionella spp., L. pneumophila and mycobacterium
       avium complex in two drinking water distribution systems: DS1, using chlorine
       disinfection, and DS2, using chloramine disinfection. Samples were collected and
       disinfectant residual was measured four times throughout the year and at different
       distances. In DS1, the five sampling sites were located between 5 and 22 kilometers (km)
       from the treatment plant and had free chlorine residuals in the range of 0.2 to 1.3 mg/L.
       In DS2, the five sampling sites were located between 1 and 137 km from the treatment
       plant and had monochloramine residuals in the range of <0.05 (at a dead-end location) to
       3.9 mg/L. All three microbes were detected throughout the distribution system and at
       different points throughout the year. The only recurring trend was an increase in
       microorganisms when the disinfectant residual decreased (for both chlorine and
       chloramine), especially at dead ends in the system (<0.05 mg/L of monochloramine).

   •   Duda et al. (2014) observed a significant reduction in Legionella at distal sites after a
       monochloramine generation system was installed in a hospital hot water system.  The
       observations were based on 29 months of monitoring data including a five-month
       baseline period and 24 months' data following installation.

   •   A hospital in Italy added monochloramine treatment into a hot water network within the
       building using  a device to continuously distribute monochloramine (Marches! et  al.,
       2013; Marches! et al., 2012). The disinfectant levels were maintained between 1.5 and 3.0
       mg/L. Hot water samples were analyzed for Legionella spp. and Pseudomonas spp. over
       a one-year period.  Both organisms decreased in terms of the number of positive samples.
       Before the addition of continuous treatment, 97 percent of samples were positive for
       Legionella. After treatment,  13.3 percent of samples were positive for Legionella. The
       authors concluded that based on this full-scale study, continuous injection of
       monochloramine in a building hot water system has potential for controlling Legionella
       (Marches! et al., 2012). Marches! et al. (2013) continued the study for a total of 36
       experimental months with the same parameters for monochloramine and confirmed that
       Legionella control with monochloramine was rapid, as 7 out of the 8 positive samples
       occurred within the first eight months of the total 36-month experimental period. The
       eighth positive sample occurred at 15 months, when the monochloramine dosage rate
       decreased below 1 mg/L. Use of monochloramine did not increase chlorite levels and
       nitrification did not occur. The authors suggested that a monochloramine concentration
       between 2 and  3 mg/L should be maintained to assure a Legionella concentration below
       102 CFU/L.

   •   Weintraub et al. (2008) evaluated water and biofilm samples from 53 buildings in San
       Francisco before and after conversion to monochloramine for residual disinfection in
Draft - Technologies for Legionella Control:
Scientific Literature Review                      25                                October 2015

-------
       February 2004. Chlorine was used for primary disinfection throughout the study period.
       The total chlorine level in finished water was 0.6 mg/L on average prior to conversion
       and 1.97 mg/L on average in 2004 following conversion. Samples were collected from
       each building six times during the two-year study period—three samples before and after
       the conversion to monochloramine. Sampling results showed that 60 percent of hot water
       systems and 72 percent of buildings contained Legionella before conversion to
       monochloramine compared to 4 percent of hot water systems and 9 percent of buildings
       after the conversion. After the conversion to monochloramine, there was an approximate
       10-fold increase in the concentration of total chlorine in the buildings' hot water systems.
       Also, prevalence of Legionella decreased by 96 percent in POU outlets when controlling
       for building and water characteristics (Weintraub  et al., 2008).

    •   Flannery et al. (2006) compared Legionella colonization of hot water systems for two
       years to determine if a conversion from chlorine to monochloramine in the drinking water
       system would reduce Legionella levels in the building hot water system. The results
       showed 60 percent colonization of the hot water system before conversion and 4 percent
       colonization after the conversion. After switching to a disinfectant with a more stable
       residual, higher concentrations of total chlorine were measured within building hot water
       systems. The authors concluded that increasing the amount of water supplies disinfecting
       with monochloramine might reduce the incidence of Legionnaires'  disease.

    •   Moore et al. (2006) evaluated Legionella colonization within building water systems after
       the wholesale PWS had converted from chlorine to monochloramine for residual
       disinfection treatment. Legionella colonization of building water systems decreased from
       19.8 percent (19 of 96 buildings) to 6.2 percent (6 of 96 buildings). The samples in this
       study were taken a few months before and a few months after the conversion to
       monochloramine (Moore et al., 2006).

    •   Heffelfinger et al. (2003) concluded that hospital water systems using a monochloramine
       disinfectant residual were at a lower risk of Legionnaires' disease cases than systems
       using a chlorine residual, based on survey data. Out of 459 surveys sent, 166 hospitals
       responded (36 percent response rate). Of the 166 survey respondents, 38 (25 percent of
       survey respondents) were selected as case studies because they had reported definite
       cases of Legionnaires' disease in the period 1994 to 1998 or outbreaks of hospital-
       acquired Legionnaires' disease in the period 1989 to 1998, and they had not changed
       their water disinfection practices during the study period. Six of the 38 case study
       hospitals (16 percent) used monochloramine for disinfection of the  municipal water
       supply. Of the 128 survey respondents that reported no cases of Legionnaires' disease
       during the study period, 59 (46 percent) used monochloramine disinfection. The hospitals
       supplied by drinking water with a monochloramine disinfectant residual were less likely
       to have definite cases or outbreaks than hospitals with chlorine disinfectant residuals
       (adjusted odds ratio: 0.20; confidence interval (95 percent): 0.07-0.56).

    •   Kool et al. (2000) conducted a case control study comparing disinfection methods in
       water supplied to hospitals with reported Legionnaires' disease (32 hospitals) with the
       disinfection methods used in water  supplied to control hospitals (48 hospitals) with no


Draft - Technologies for Legionella Control:
Scientific Literature Review                      26                                  October 2015

-------
       reported disease. They found that hospital water systems supplied with water treated by
       chlorine were more likely to have reported an outbreak of Legionnaires' disease than
       hospitals supplied with water treated by monochloramine (odds ratio 10:2 and 95 percent
       confidence interval: 1.4-460). The authors infer that 90 percent of the outbreaks might
       have been prevented had the residual used in the case hospitals contained
       monochloramine (Kool et al., 2000; Kool et al., 1999).  The cases in this study were based
       on previous records of infections and not on Legionella measurements in the water
       supply (Kim at al., 2002; Lin et al., 2000a).

2.3.2.3  Potential Water Quality Issues

Potential water quality issues with monochloramine include corrosion, formation of DBFs, and
nitrification. Monochloramine can impact kidney dialysis and should be removed from the
dialysate water. Monochloramine should be removed from water used for fish tanks due to
detrimental effects.

An unintended consequence of using monochloramine is corrosion of the pipes and materials
used in water systems. Corrosion can occur in two forms, including pitting and a more uniform
thinning of pipe surfaces. Kirmeyer et  al. (2004) reported that chloramine can attack rubber and
plastic components in a water system and that 23 percent of utilities surveyed experienced an
increase in degradation of rubber materials after chloramine disinfection was implemented.
Water temperature, pH and disinfectant concentration also affect corrosion rates.
Monochloramine can react with pipe scale differently than other disinfectants, resulting in lead
leaching in system materials containing lead (Edwards and Dudi, 2004). Corrosion control and
maintenance of building water systems will be important to consider before adding disinfectants.
Further  research is needed to evaluate the interactions of disinfectants with water chemistry and
piping materials in a building water system and to better understand the effects of these
interactions on the efficacy of pathogen inactivation (Rhoads et al., 2014).

Another unintended consequence of monochloramine disinfection is its ability to react with
organics in the water to form DBFs. Although chloramination significantly reduces some DBFs,
such as  THM and haloacetic acids (HAA), its usage can contribute to the formation of other
DBFs such as nitrosamines. For more information regarding nitrosamines please see the N-
nitrosodimethylamine (NDMA) fact sheet (USEPA, 2014b) on EPA's website.

Nitrification is a potential  problem for utilities that utilize chloramines as a disinfectant and may
occur when finished water contains excess ammonia and low chloramine  residual (Kirmeyer et
al., 2004). Areas of the distribution system with higher water age and warmer temperatures are
more susceptible to nitrification. Nitrification is a microbiological process that oxidizes ammonia
to form  nitrite and nitrate.  Increased nitrate levels provide nutrients for the growth of nitrifying
bacteria. Nitrification  can also degrade the aesthetic quality of the water resulting in taste and
odor issues as  well as  particles in the water (AWWA, 2013). Breakpoint chlorination can occur
due to imbalances in chlorine and ammonia concentrations, resulting in the formation of nitrate,
nitrogen chloride  and nitrogen gas. Once nitrification occurs, maintaining monochloramine
disinfectant residual becomes very difficult within the nitrified areas of the distribution  system,
allowing pathogenic organisms that may be present in biofilm or pipe scale to proliferate. The
Draft - Technologies for Legionella Control:
Scientific Literature Review                       27                                 October 2015

-------
American Water Works Association's (AWWA) Manual M56 recommends that any utility using
chloramines develop and implement a nitrification control plan (AWWA, 2013).

Monochloramine can inhibit biological growth on filters, which could be positive in that it helps
keep the filters clean, but this inhibition can also reduce biodegradable dissolved organic carbon
removal, a problem if the filters were put in place for that purpose.

Converting disinfection to monochloramine can have an impact on organisms other than
Legionella. A study by Moore et al. (2006) found that, in addition to Legionella, building water
systems were colonized with mycobacteria before and after a conversion from chlorine to
monochloramine in the PWS. The proportion of buildings colonized with mycobacteria increased
from 19.1 percent during the chlorine phase to 42.2 percent after the conversion to
monochloramine. The number of samples within the distribution system containing detectable
levels of coliform increased from two samples during the chlorine phase to twenty  samples after
the conversion (Moore et al., 2006).

Pry or et al. (2004) saw similar results in a study conducted in Florida. After conversion to
monochloramine, mycobacteria increased, total  coliforms and heterotrophic bacteria levels
increased, and nitrification occurred  in the storage tanks (Pryor et al., 2004). Building water
system operators who consider treating water with monochloramine to control for Legionella
should be cognizant of potential unintended consequences such as increases in mycobacteria and
other waterborne pathogens and take the necessary protective measures to protect public health.

2.3.2.4  Operational Conditions

Parameter Conditions Indicating Operational Effectiveness

The normal dosage rate for monochloramine is between 1.0 and 4.0 mg/L.

The case studies cited above generally support maintaining a chloramine residual in the building
water system in the range of 1 to 2 mg/L as an effective means for containing biofilm growth,
minimizing Legionella colonization, and preventing outbreaks.  As such, building water system
maintenance such as  appropriate pH, chlorine-to-ammonia ratios, flushing, and  frequent
monitoring to demonstrate residual maintenance on an ongoing basis are essential.  The current
practice is to use a chlorine-to-ammonia ratio of 3:1  to 5:1 to produce monochloramine. The
amount of organic nitrogen in the water prior to addition of ammonia will  also affect how much
ammonia is needed to reach the desired ratio (USEPA, 1999c).

The rate of reaction for the conversion of chlorine to monochloramine is sensitive to pH and can
also be affected by contact time and  temperature. The optimum pH range for formation of
monochloramine is 7.5 to 9 (WHO, 2004). Monochloramine is relatively stable  under varying
temperatures  once formed. Cunliffe (1990) evaluated monochloramine decay at two different
temperatures. Water incubated  at 55  degrees C (131  degrees F) showed a loss of residual after 50
hours, from 1.3  to 0.35 mg/L. After 5 days at 30 degrees C (86 degrees F), the concentration
dropped from 1.3 to 0.8 mg/L.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      28                                 October 2015

-------
Installation Considerations

Guidelines for design and implementation of chloramination systems include the following:

    •   AWWA M56 Manual, Nitrification Prevention and Control in Drinking Water. Second
       Edition. (AWWA, 2013).
    •   Simultaneous Compliance Guidance Manual for the Long Term 2 and Stage 2 DBF Rules
       (USEPA, 2007).
    •   The Water Research Foundation manual Optimizing Chloramine Treatment (Kirmeyer et
       al., 2004).
    •   Alternative Disinfectants and Oxidants Guidance Manual (EPA 815-R-99-014) (USEPA,
       1999c).

Monitoring Frequency and Location

The SWTR (USEPA, 1989a) requires that all PWSs using monochloramine and using surface
water or ground water under the direct influence of surface water monitor for the presence of a
disinfectant residual in the distribution system and at the POE to the distribution system. The
disinfectant level must be at least 0.2 mg/L at POE and detectable in at least 95 percent of
samples collected within the distribution system.

Stage 1 DBPR also requires PWSs that use monochloramine to maintain an MRDL running
annual average of less than 4.0 mg/L (USEPA, 1998).

PWSs that use chloramines are required to monitor for combined or total chlorine residual or
HPC in the distribution system at locations that have been approved by the primacy agency. All
of these parameters could provide operational information to indicate the need for chloramine
dose adjustments, system flushing and water age management within finished water storage
facilities.

Monochloramine can be  measured by amperometric titration (Symons, 1978), N,N-diethyl-p-
phenylenediamine (DPD) ferrous titrimetric, DPD colorimetric methods (USEPA, 1999c), and
commercially available adapted indophenol methods (Hach MonochlorF) (Lee et al., 2007). The
EPA has approved multiple methods for measuring combined chlorine as well  as total chlorine.
A list of approved methods is available through EPA's website (USEPA, 2014a).

Other monitoring should be conducted to  identify the onset of nitrification, which is common in
systems that use chloramination. Kirmeyer et al. (2004) recommended monitoring HPC,
chloramine residual, ammonia, nitrate and nitrite to detect nitrification in the distribution system.
A system-specific monitoring plan should be developed to identify sampling locations,
parameters and sampling frequency.

Maintenance Needs

Operating and maintenance practices for chloramine disinfection systems include maintenance of
an appropriate disinfectant residual, regular system cleaning and flushing, inspections, and water
quality monitoring. Newly constructed or rehabilitated piping systems are cleaned and flushed
prior to initial disinfection. Routine flushing and water quality monitoring are recommended to

Draft - Technologies for Legionella Control:
Scientific Literature Review                      29                                 October 2015

-------
assure that adequate disinfectant levels are maintained throughout the building water system
(HSE, 2014; Rosenblatt and McCoy, 2014).

Systems using monochloramine as a residual disinfectant periodically use free chlorine to
eliminate biological growth that may have occurred in the distribution system or on equipment
(AWWA, 2013; Lin et al., 2000b).

Approaches for preventing nitrite and nitrate formation within the distribution system include
decreasing water age through flushing or operational changes, increasing the pH, decreasing
temperature, decreasing total organic carbon concentration, increasing monochloramine
residuals, increasing the chlorine-to-ammonia ratio, and decreasing the excess ammonia
concentration (USEPA, 1999c).

    2.3.3  Chlorine Dioxide

Chlorine dioxide is a water-soluble gas that can easily diffuse through cell membranes of
microorganisms. It has been found to be superior in penetrating biofilms as compared to chlorine
(Lin et al., 201 Ib). It is a very effective disinfectant (when used correctly) at inactivating
bacterial, viral and protozoan pathogens and has a high oxidation potential (USEPA, 1999b). Its
use as a biocide can be maintained over a wider pH range than chlorine or CSI (Lin et al.,
201 Ib).

Chlorine dioxide was first used as a disinfectant in the early 1900s at a spa in Belgium, and its
use in drinking water disinfection became more common in the 1950s (USEPA, 1999b). In the
1970s, more than 100 U.S. water treatment facilities used chlorine dioxide for taste and odor
control, iron and manganese oxidation, or final disinfection, while in Europe, it was being used
at several thousand water treatment facilities, primarily for final disinfection (Symons et al.,
1977). In the 1980s, use of chlorine dioxide as an alternative primary disinfectant to chlorine
increased in the United States after EPA promulgated a regulation for total trihalomethanes
(TTHMs) (Aieta and Berg, 1986). Since the late 1980s, chlorine dioxide has been evaluated
(Dupuy et al., 2011; Loret et al., 2005; Jacangelo et al., 2002; Berg et al.,  1988) and later
implemented as an effective disinfectant to control Legionella and biofilm in hot and cold
building water systems (Casini et al., 2014; Marches! et al., 2013; Cristino et al., 2012; Marches!
et al., 2011; Zhang et al., 2009; Sidari et al., 2004).

Use of chlorine dioxide in PWSs is regulated by the DBPR. Chlorine dioxide itself can cause
acute health effects and has an MRDL of 0.8 mg/L. Chlorite, a DBF of chlorine dioxide
disinfection, is also regulated by EPA due to potential health concerns. The  Stage 1 DBPR sets a
maximum contaminant level (MCL) of 1.0 mg/L for chlorite.

2.3.3.1  Characterization of Effectiveness against Legionella

Chlorine dioxide is usually generated on site from sodium chlorite solutions and one or more
other chemical precursors  (e.g., sodium hypochlorite, hydrochloric acid, sulfuric acid) or by an
electrochemical oxidation  process. Stock solutions produced on-site typically have a
concentration of 500 mg/L. Chlorine dioxide gas cannot be compressed or stored commercially
because it is explosive under pressure. Therefore, it is never shipped (USEPA, 1999b). Water
Draft - Technologies for Legionella Control:
Scientific Literature Review                       30                                 October 2015

-------
treatment chemicals must meet the appropriate ANSI/AWWA standards or NSF/ANSI Standard
60 (GLUMRBSPPHEM 2012).

Laboratory and pilot-scale testing have generally shown that chlorine dioxide disinfection can be
effective in controlling Legionella:

   •   Dupuy et al. (2011) compared chlorine dioxide, chlorine and monochloramine in treating
       L. pneumophila and Acanthamoeba strains alone or in co-cultures (i.e., L. pneumophila
       grown within amoebae). Dosage rates were 0.4 mg/L for chlorine dioxide, 2-3 mg/L for
       chlorine (for a residual free chlorine concentration of ~1 mg/L), and 0.8 mg/L for
       monochloramine. All samples were treated with disinfectant for one hour and then the
       disinfectant residual was measured. Chlorine and chlorine dioxide were more efficient at
       reducing (i.e., providing at least a 3-log (99.9 percent) reduction of the bacterial
       population at study conditions)  free L. pneumophila than co-cultured L. pneumophila.
       Chlorine dioxide was found to be highly efficient in inactivation of Acanthamoeba M3
       amoebae only,  less so in inactivating the other two Acanthamoeba strains discussed in the
       paper.

   •   Loret et al. (2005) compared the performance of several alternative disinfectants under a
       controlled pilot-scale simulation of a typical building water system. Tap water and
       injection of cultured natural Legionella strains were used to establish biofilms in each
       pipe loop. Low temperature (35 degrees C, or 95 degrees F) and low water velocity were
       maintained to favor the growth  of Legionella and biofilms. Each pipe loop was treated
       with one of the studied disinfectants for three months. The target dosage rate for chlorine
       dioxide was 0.5 mg/L. The authors determined that chlorine dioxide and chlorine were
       the most effective in controlling Legionella, biofilm and protozoa. Chlorine  dioxide had
       longer residual activity in the system than did chlorine. Chlorite levels were measured in
       the chlorine dioxide pipe loop at levels >0.2 mg/L. Legionella populations decreased to
       undetected levels (<500 CFU/L) within the first three days of treatment for all
       disinfectants. Biofilm reduction started one week after treatment was initiated, and
       biofilm thickness was reduced to <5 jim with chlorine dioxide and several other
       disinfectants, as compared to a measured biofilm thickness of 13-35 jim in the untreated
       pipe loop.

   •   Jacangelo et al. (2002) conducted laboratory studies to evaluate chlorine dioxide, free
       chlorine and monochloramine for inactivation of waterborne emerging pathogens,
       including Legionella. The chlorine dioxide dose rate was 1.0 mg/L. Two different
       temperatures (5 and 25 degrees  C, or 41 and 77 degrees F) and two different pH values
       (6.0 and 8.0) were examined. The observed CT values for 2-log reduction of Legionella
       were reported. At 5 degrees C, the observed CT values ranged from >320 to >1,000 min-
       mg/L at pH 6.0 and from >250 to 630 min-mg/L at pH 8.0. At 25 degrees C, the observed
       CT values ranged from 50 to 200 at pH 6.0 min-mg/L and from 50 to 130 min-mg/L at
       pH 8.0.

Chlorine dioxide disinfection systems have been installed  in many hospitals to control Legionella
and biofilm in hot and cold water systems (Casini et al., 2014;  Marches!  et al., 2013; Cristino et
al., 2012; Marches! et al., 2011; Zhang  et al., 2009; Sidari  et al., 2004).

Draft - Technologies for Legionella Control:
Scientific Literature Review                      31                                October 2015

-------
    •   Casini et al. (2014) reported the application of a WSP approach using multiple
       disinfectants and filtration to control Legionella in a hospital's hot water system. The
       multi-barrier strategy was developed and refined over a 9-year period. Continuous
       disinfection with chlorine dioxide (0.4-0.6 mg/L in recirculation loops) was provided.
       Additional treatment included endpoint filtration and a shift to monochloramine
       disinfection (2-3 mg/L) after chlorine-tolerant Legionella spp. were identified. After nine
       years, the number of sampling sites that were positive for Legionella decreased by 51
       percent, from 66.7 percent to 32.9 percent, and the mean Legionella count decreased by
       78 percent.

    •   Marches! et al. (2013) reported a strong reduction in Legionella contamination in three
       hospital hot water systems over a three-year period compared to the untreated systems. A
       dosage rate of 0.50-0.70 mg/L chlorine dioxide was applied to the hot water systems,
       with the goal of maintaining a minimum concentration of 0.30 mg/L at distal sites (i.e.,
       sink taps, tubs and showers located at distant points in the building water system. On
       average, the three systems reduced Legionella occurrence from 96 percent of sampling
       sites to 46 percent.

    •   Cristino et al. (2012) described use of chlorine  dioxide for chemical shock treatment and
       continuous treatment after the hot water system in a long-term care facility was found to
       be colonized with L. pneumophila. In addition to thermal shock and chemical shock with
       peracetic acid, chlorine dioxide was applied at a dose  sufficient to obtain a 5-mg/L
       residual throughout the building water systems for a one-hour contact time.  The water
       was then drained and fresh water introduced to the systems until the chlorine dioxide
       residual was <0.3 mg/L. A continuous chlorine dioxide system was installed in the hot
       water supply at a dose sufficient to maintain a minimum  0.3-mg/L residual at distal taps.
       The shock treatment reduced Legionella counts from 104-105 CFU/L to zero to 102
       CFU/L. Environmental monitoring conducted during the continuous chlorine dioxide
       treatment period showed that Legionella counts remained at stable levels (zero to 103
       CFU/L). No cases of hospital-acquired legionellosis occurred during the study period.

    •   Marches! et al. (2011) compared the performance of treatment alternatives for controlling
       Legionella contamination in hospital hot water systems, including two hot water plants
       that installed chlorine dioxide treatment systems in 2005. Chlorine dioxide successfully
       maintained Legionella levels at <100 CFU/L. Electric boilers and POU filters had better
       performance than chlorine dioxide.4 The authors suggested implementing chlorine
       dioxide and electric boilers in parallel to control Legionella.

    •   Zhang et al. (2009) evaluated using chlorine dioxide to treat for L. pneumophila in two
       hospitals reporting cases of hospital-acquired legionellosis. Water quality parameters
       were very similar between the two systems, except pH was 7.70 for Hospital A and 8.57
       for Hospital B. Hospital B had previously tried a superheat-and-flush of the hot water
       system and replacing a storage tank that was colonized with Legionella, but those
 The electric boilers were installed on cold water lines in high risk areas; each boiler served one to two patient
rooms.

Draft - Technologies for Legionella Control:
Scientific Literature Review                       32                                 October 2015

-------
       measures were unsuccessful. Chlorine dioxide was injected into the cold water main at a
       dosage rate of 0.5-0.7 mg/L to control Legionella and maintain chlorite levels below the
       MCL at both hot and cold water taps. The residual concentration for chlorine dioxide at
       the distal sites varied from zero to 0.11 mg/L depending on the building, date and type of
       tap (hot or cold). The occurrence of Legionella at hot water taps decreased from 60
       percent of sampling sites before chlorine dioxide addition to <10 percent of sampling
       sites after treatment; however, a significant decrease in the concentration of Legionella in
       positive samples was not observed. No cases of hospital-acquired Legionnaires' disease
       were detected after installation of the chlorine dioxide system.

    •   Sidari et al. (2004) reported on the first controlled field evaluation in the United States of
       a chlorine dioxide treatment system installed in June 2000, at a hospital linked to three
       cases of hospital-acquired Legionnaires'  disease. Legionella was eradicated (i.e., the
       authors had no positive results) from the  hot water system, but only after 20 months of
       treatment. Rates of positive Legionella detections decreased significantly, from 23
       percent to 12 percent for hot water taps, and to almost zero percent for the cold water
       reservoir. Average chlorine dioxide residuals were 88 percent lower for the hot water tap
       than the cold, or 0.08 mg/L and 0.68 mg/L, respectively. No cases of Legionnaires'
       disease occurred after the chlorine dioxide treatment system was installed.

2.3.3.2   Potential Water Quality Issues

Chlorine dioxide does not form the high levels of chlorinated DBFs that chlorination does (Gates
et al., 2009). Chlorite and chlorate are the most prominent byproducts of chlorine dioxide (Gates
et al., 2009). Chlorite could cause anemia in some people and affect the nervous systems of some
infants, young children and fetuses of pregnant women. Ongoing exposure to chlorate ion can
lead to an enlarged thyroid (USEPA, 2012). In an Italian hospital where  chlorine dioxide was
used to control Legionella in the hot water supply, chlorite levels higher than 0.7 mg/L were
measured when the chlorine dioxide residual was >0.3 mg/L (Marches! et al., 2013). In another
study (Zhang et al., 2009), chlorine dioxide concentrations among different sampling locations
over two years ranged from zero to 0.70 mg/L, whereas chlorite concentrations ranged from zero
to 0.82 mg/L. The average chlorate concentrations in hot and cold water were below the
detection limit of 0.10 mg/L.

Water treatment utilities use chlorine dioxide for taste and odor control;  however, various odors
were reported when chloride dioxide residuals exceeded 0.05 mg/L in the winter or 0.15 mg/L in
the summer (Gates et al., 2009). Kerosene-like and cat-urine-like odors were reported in some
homes with new carpets when volatizing chlorine dioxide reacted with airborne volatiles
(Dietrich and Hoehn, 1991). The taste and odor threshold for chlorine dioxide in water has been
reported  to be as low as 0.2 mg/L (Roche and Benanou, 2007).

Chlorine dioxide is considered less corrosive than chlorine (Lin et al., 201 Ib). Some reports
suggest that chlorine dioxide can cause damages to polyethylene pipes (Chord et al., 2011; Yu et
al., 2011); however, information on other types of pipes is sparse (Gates et al., 2009).
Draft - Technologies for Legionella Control:
Scientific Literature Review                       33                                 October 2015

-------
2.3.3.3  Operational Conditions

Parameter Conditions Indicating Operational Effectiveness

Dosage rate is an important design criterion for chlorine dioxide disinfection systems. Chlorine
dioxide dosage rates of 0.4 to 0.7 mg/L were reported by systems experiencing successful
treatment performance (HSE, 2014; Casini et al., 2014; Marches! et al., 2013; Zhang et al.,
2009). Zhang et al. (2009) reported that the dosage rate was between 0.5 and 0.7 mg/L depending
on the flow rate of cold water entering the buildings.

The required disinfectant dosage rate is dependent on system-specific conditions including pipe
material and condition, the water's disinfectant demand, the extent of biofilm on pipe surfaces,
pipe diameter and length, complexity of the building water system, treatment goals (e.g.,
Legionella control), and the water turnover rate. For example, corrosion of galvanized piping can
increase the water's disinfectant demand and reduce the residual chlorine dioxide level (Lin et
al., 201 Ib). One  study (Zhang et al., 2009) reported the chlorine dioxide demand of the building
water was determined to be 0.20 mg/L after six hours of contact time at 23 degrees C (73.4
degrees F) and pH 7.8.

Maintaining a total chlorine dioxide residual of 0.1-0.5 mg/L at the tap is usually sufficient to
control Legionella, although higher residuals may be necessary in a heavily colonized system
(HSE, 2014). Some systems have established a treatment goal of maintaining a minimum
chlorine dioxide residual of 0.3 mg/L at distal taps (Marches! et al., 2013; Cristino et al., 2012;
Sidari et al., 2004). If treatment with a residual higher than 0.8 mg/L is determined to be
necessary, the facility may want to ensure that emergency disinfection procedures  are developed
and followed so that human consumption of a concentration of chlorine dioxide greater than the
MRDL  does not occur.

Several studies identified factors that limited the effective performance of chlorine dioxide
disinfection for Legionella control:

    •   Casini et al. (2014) reported that performance of a hospital water system appeared to be
       affected by an accidental event in a water tank of the municipal water system that caused
       sediment buildup and increased water contamination in the hospital water system. After
       the event in December 2006, no further reduction in Legionella colonization was
       observed with only disinfection treatment. As a result, the hospital modified its WSP to
       include POU filtration with 0.2-|im sterile filters in critical areas of the system (e.g.,
       intensive care units (ICUs), transplant wards).

    •   Marches! et al. (2013) reported that some samples associated with the chlorine dioxide
       treatment systems had chlorite levels higher than the Italian regulatory limit of 0.7 mg/L.
       Such exceedances would occur when the chlorine dioxide residual was >0.30 mg/L at
       distal taps. Because successful treatment performance (stated above) was based on
       maintaining a chlorine dioxide residual >0.30 mg/L at distal taps, this case highlights the
       importance of balancing disinfection needs with minimizing DBF formation.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       34                                 October 2015

-------
    •   The lack of proper monitoring for a chlorine dioxide treatment system for drinking water
       was noted at a New York heath care facility after it experienced a Legionella outbreak in
       2010 (CDC, 2013b). Two hospitalizations for acute respiratory illness were reported and
       no deaths occurred.

    •   Zhang et al. (2009) noted the importance of achieving an adequate disinfectant residual
       and its effect on the amount of time needed to control Legionella. Disinfection systems
       installed at Hospital A (January 2003) and Hospital  B  (April 2004) injected chlorine
       dioxide  at the cold water service line to each building using a dosage rate of 0.5-0.7
       mg/L. Although both hospitals achieved significant  reductions in Legionella occurrence
       in the hot water system, Hospital B achieved control after six to 10 months of treatment,
       whereas Hospital A required 18 months. The longer treatment period for Hospital A was
       attributed to the longer time needed to achieve a chlorine dioxide residual >0.10 mg/L.

    •   Sidari et al. (2004) reported differences in the time required to eradicate Legionella in a
       hospital's hot and cold water systems after a chlorine dioxide treatment system was
       installed in June 2000. The water treatment goal was to achieve  a minimum chlorine
       dioxide  residual of 0.3 mg/L at distal sites. Chlorine dioxide residuals in the hot water
       loop averaged 0.08 mg/L (88 percent lower) compared to an average of 0.33 mg/L at cold
       water taps, which may explain why 20 months of treatment was required to eradicate
       (i.e., achieve zero occurrence) Legionella from the hot water system, whereas only 15
       months  of treatment was needed for the cold water system.

Installation Considerations

The location of disinfectant application point(s) is a critical design decision. The location may
affect the required dosage rate and the time needed to inactivate Legionella. For example, if
chlorine dioxide is added at the cold water service entry point to the building, the dosage rate
should be sufficient to achieve an adequate disinfectant residual at hot water taps at distant points
in the building.  However, the need to comply with drinking water standards may drive a design
decision to install multiple treatment units in the building system.

Monitoring Frequency and Location

The SWTR (USEPA, 1989a) requires that all PWSs using chlorine dioxide monitor the level of
residual disinfectant present in the water supply. The Stage 1 DBPR also requires that these
PWSs monitor  daily at each entry point to the distribution system to ensure it is not exceeding
the MRDL (USEPA, 1998). Chlorine dioxide is a contaminant with acute health effects. Chlorine
dioxide has a short sample hold time and should be measured immediately after sample
collection; therefore, under the Stage  1 DBPR, on-site analysis at the water system is required.

If the daily chlorine dioxide measurement exceeds 0.8 mg/L, three follow-up distribution system
chlorine dioxide samples must be measured the following day, as required by the Stage 1 DBPR.

The Stage 1 DBPR and Stage 2 DBPR require that all PWSs using chlorine dioxide monitor
chlorite for compliance with the MCL (USEPA, 2006c). Chlorite must be monitored daily at the
entry point to the distribution system, in addition to being measured in a three-sample set each

Draft - Technologies for Legionella Control:
Scientific Literature Review                      35                                  October 2015

-------
month in the distribution system as detailed in the Code of Federal Regulations (CFR) at 40 CFR
141.132(b)(2). Daily monitoring can be conducted on-site; monthly monitoring should be
conducted at a certified laboratory (see Table Footnote 8 in 40 CFR 141.131(b)).

Maintenance Needs

Operating and maintenance practices for chlorine dioxide  disinfection systems include
maintenance of a disinfectant residual, regular system cleaning and flushing, inspections, and
water quality monitoring. A newly constructed or rehabilitated piping system should be  cleaned
and flushed prior to initial disinfection (GLUMRBSPPHEM, 2012). Routine flushing and water
quality monitoring are recommended to assure that adequate disinfectant levels are maintained
throughout the building water system (HSE, 2014; Rosenblatt and McCoy, 2014).

    2.3.4   Copper-Silver lonization

2.3.4.1 Background

Commercially available CSI systems typically consist of flow cells that contain metal bars or
anodes (containing copper and silver metals) surrounding  a central chamber, through which
piped water flows. A direct electric current is passed between these anodes, releasing the copper
and silver ions into the water stream. The amount of ions released depends on the composition of
the anode and is controlled by the electrical current applied to the bars and the water flow rate.

The earliest reports of CSFs antibacterial properties were published in the mid-1970s (Spadaro et
al., 1974; Berger et al., 1976). The earliest combined use of copper and  silver ions as water
treatment focused on the disinfection of swimming pools (Yahya et al.,  1989) as an alternative to
using high levels of chlorine. A 1994 report on the use of CSI treatment was the first to address
the efficacy of this treatment for controlling Legionella in  hospital water systems (Liu et al.,
1994). CSI systems are currently used in buildings with complex water  systems to control the
growth and occurrence of Legionella bacteria. Lin et al. (201 Ib) documented CSI applications
eradicating Legionella from hospitals worldwide.

2.3.4.2 Characterization of Treatment Technology Effectiveness against Legionella

Case studies constitute the majority of the published reports on the efficacy of CSI in controlling
Legionella in building water systems (Chen, 2008; Modol  et al., 2007; Blanc et al., 2005; Stout
and Yu, 2003; Kusnetsov et al., 2001; Rohr et al., 1999; Liu et al., 1998; States et al., 1998; Liu
et al., 1994). The studies generally describe situations where Legionella bacteria were found in a
building water system and CSI was initiated in an attempt  at Legionella control. Many of the
reviewed laboratory studies indicate that copper and silver ions can reduce the cultivability of
Legionella and the incidence of legionellosis. However, as with other technologies, other studies
showed that Legionella can be protected from copper and  silver ions when it is associated with
biofilms or amoebae. The potential for Legionella to develop resistance to copper and silver ions
has also been suggested by several studies.

   •   Lin et al. (1996)  and Landeen et al.  (1989) indicate that copper ions (at 0.4 mg/L) and
       silver ions (at 0.04 mg/L) can effectively reduce the cultivability of Legionella bacteria.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       36                                October 2015

-------
       A 3- to 4-log (99.9 to 99.99 percent) reduction of culturable Legionella bacteria at these
       ion concentrations is reported to occur at pH 7.0-7.3, with exposure times reported to be
       <1 hour to 24 hours (Lin et al., 2002; Lin et al., 1996 and Landeen et al., 1989).

    •   Hwang et al. (2007) showed that Legionella bacteria that are ingested by amoebae (in
       biofilms along piping walls) are protected from inactivation by copper and silver ions; a
       concentration of copper and silver that would normally result in a 7-log reduction of
       planktonic Legionella in 30 minutes would not similarly affect Legionella bacteria inside
       an amoeba, where they can survive for many days.

    •   States et al. (1998) reported that CSI treatment was successful in reducing the percentage
       of samples testing positive for Legionella from 100 percent to less than 17 percent on
       average over a two-year period.

    •   Kusnetsov et al. (2001) reported a 100-fold decrease in culturable Legionella from
       biofilm  samples after CSI treatment was employed.

    •   Rohr (1999) indicated an initial impact on Legionella occurrence, where 100 percent of
       sampling sites were positive for Legionella before treatment and 55 percent of sampling
       sites had positive  results one year after treatment was initiated. Over the next three years,
       75-78 percent of samples were positive for Legionella.

    •   Modol et al. (2007) described how a large hospital experienced success in decreasing the
       number of positive Legionella samples after initiating CSI, only to see the number of
       positive samples increase from 20 percent to 65 percent during two months when the
       treatment system was under repair.

    •   Blanc et al. (2005) reported that the addition of copper and silver ions alone had no
       impact on the number of Legionella-positive water and biofilm samples in a large
       hospital.

    •   Survey results by  Stout and Yu (2003) showed that of 13 hospitals reporting at least 30
       percent Legionella-positive samples before CSI treatment began, nine hospitals reported
       a sustainable (over a period of 6-9 years) decrease in the number of Legionella-positive
       samples; five hospitals reported no positive samples after treatment. This survey also
       showed that all of the hospitals reported cases of hospital-associated Legionnaires'
       disease before CSI treatment, and  all but one reported no cases after treatment.

Other studies have reported the potential occurrence of Legionella strains that appear to be less
sensitive to the  toxic effects of certain chemicals such as copper and silver. Microorganisms are
highly adaptive, and it is well documented that within the bacterial world there are cellular
mechanisms which allow bacteria to survive hostile environments.

There is an understanding of how bacterial gene systems can confer resistance to copper and
silver (Nies, 1999). Some of these gene systems are found in Legionella (Bondarczuk and
Piotrowska-Seget, 2013). One common resistance mechanism in gram-negative bacteria (such as
Legionella) requires an energy-dependent protein that protects the cell by acting as a pump to

Draft - Technologies for Legionella Control:
Scientific Literature Review                      37                                October 2015

-------
export copper ions out of the cell (Bondarczuk and Piotrowska-Seget, 2013). The occurrence of
Legionella strains tolerant of copper and silver at the levels employed in CSI treatment has been
noted (Rohr et al.,  1999). Hypochlorous acid, the active disinfecting chlorine species, is in part
toxic to bacterial cells by virtue of interfering with the production of energy (in the form of
adenosine triphosphate (ATP)) that is needed for many cellular processes including heavy metal
resistance enzymes (Barrette et al.,  1989). The synergy between free chlorine and heavy metal
ions on Legionella copper resistance mechanisms and Legionella susceptibility is generally
unstudied. However, Landeen et al. (1989) showed increased (although not statistically
significant) inactivation rates of Legionella with copper and silver ions in the presence of 0.4
mg/L free chlorine.

2.3.4.3  Potential Water Quality Issues

Use of CSI may result in corrosion. Investigations have shown that test systems that exposed
galvanized and mild steel (i.e., carbon steel) coupons to CSI resulted in corrosion (Loret et al.,
2005). The authors noted that copper deposition resulted in extensive pitting in ferrous materials
via localized galvanic corrosion events. Copper coupons in the same system were covered with
copper deposits at pH 7.6. Some evaluation of deposition, pitting and corrosion of copper piping
in a system using CSI was presented by Boffardi and Hannigan (2013). Both the presence of
additional copper ions and the pre-existing water conditions appeared to contribute to Type III
pitting corrosion. This type of pitting usually occurs in soft water with  alkaline pH >8.0
(Edwards et al., 1994),  at distal or stagnant locations, and at moderately warm temperatures
(Boffardi and Hannigan, 2013; Edwards et al., 1994). Lytle and Schock (2008) found that waters
with high pH (pH 9 and possibly  as low as 8), low dissolved inorganic carbon  (<10 mg/L and
possibly as high as 25 mg/L) and chloride levels of 14-38 mg/L promoted pitting corrosion.

Materials compatibility and water quality will dictate the severity of corrosion, and awareness of
the types of materials and water chemistry in a building water system is critical to maintaining
system integrity.

High concentrations of both copper and silver have been reported in systems employing CSI, to
levels approaching the maximum contaminant level goal and action level for copper (1.3 mg/L)
and the secondary maximum contaminant level (SMCL) for silver (0.1 mg/L). (States at al.,
1998; Rohr et al., 1999). As copper levels in copper piping can rise during periods of stagnation,
high levels of copper can occur in early morning first-draw water samples (Araya et al., 2004;
Araya et al., 2003a; Araya et al., 2003b; Araya et al., 2003c; Araya et al., 2001; Knobeloch et al.,
1994). Copper, but not  silver, was found to concentrate on biofilm material in building water
systems employing CSI treatment (Liu et al., 1998; Zevenhuizen et al., 1979).

Copper toxicity from ingestion of drinking water has been reported even without the contribution
of copper from CSI systems (Araya et al., 2004; Araya et al., 2003a; Araya et al., 2003b; Araya
et al., 2001; Knobeloch et al., 1998; Knobeloch et al., 1994).  Symptoms of copper toxicity
include nausea, abdominal cramps,  vomiting and diarrhea.

Both copper and silver  can have negative aesthetic effects on water: color, taste and odor, and
staining issues (Hong et al., 2010; Dietrich, 2009; Stout and Yu, 2003; Edwards et al., 2000;
Knobeloch et al., 1998; Knobeloch et al., 1994). Edwards et al. (2000)  attributed the rare
occurrence of blue water to corrosion of copper plumbing; no other causative factors have been

Draft - Technologies for Legionella Control:
Scientific Literature Review                       38                                 October 2015

-------
identified. Ingesting high levels of silver can also lead to a skin discoloration condition called
"argyria" (Drake and Hazelwood, 2005; WHO, 1996; USEPA, 1989b). According to WHO
(2003), the lowest dose of silver that may lead to occurrence of argyria has not been determined,
but, in general, silver levels up to 0.1 mg/L can be tolerated without risk to health. Silver levels
approaching the SMCL of 0.1 mg/L have been reported in building water systems using CSI
treatment (Rohr et al., 1999; States et al., 1998).

Laboratory studies have been conducted on the effectiveness of CSI in reducing levels of other
bacterial species commonly found in built environments, including Pseudomonas,
Stenotrophomonas, Acinetobacter (all gram-negative bacteria like Legionella) and
Mycobacterium.  These studies showed that Legionella are as much as 10-fold more sensitive to
copper than Pseudomonas, Stenotrophomonas and Acinetobacter, but that Legionella are less
sensitive to silver than Pseudomonas and Stenotrophomonas (Huang et al., 2008;  Hwang et al.,
2007). Copper and silver  ions appear to act synergistically (the total effect is greater than the sum
of the individual effects) toward Legionella (Lin et al.,  1996), Pseudomonas and Acinetobacter
(Huang et al., 2008), while the ions act antagonistically (the interaction of the two metals lessens
the effect of each metal acting individually) toward Stenotrophomonas (Huang et al., 2008).
Mycobacterium was shown to be 100-fold less sensitive to copper and silver ions  than Legionella
(Lin et al., 1998b), and copper and silver levels that controlled Legionella were unable to control
the occurrence of Mycobacterium in a hospital building water system (Kusnetsov et al., 2001).

Field studies of CSI have reported some effectiveness in reducing fungi in hospital water
systems, especially Fusarium spp. (Chen et al., 2013). A report on healthcare facilities in Spain
with (n=9) and without (n=7) ionization treatment systems cited a fungal isolation rate of 28
percent versus 77 percent, respectively (Pedro-Botet et al., 2007). CSI has not been reported to
reduce levels of heterotrophic bacteria or amoebae in either a controlled laboratory study (Rohr
et al., 2000) or a case study (States et al., 1998).

2.3.4.4 Operational Conditions

Parameter Conditions Indicating Operational Effectiveness

Several physicochemical  parameters that could impact treatment effectiveness are discussed in
more detail below.

Maintaining copper and silver at the levels recommended by the manufacturer is a best practice
in achieving operation effectiveness. Note that monitoring typically includes measurement of the
total metal concentration, which includes copper and silver that are bound up as complexes, as
well as copper and silver  ions.  The presence of copper and silver ions is thought to be critical for
treatment effectiveness, so maintaining proper pH and avoiding interfering materials  (e.g.,
phosphates, chlorides) is also important (Zevenhuizen et al.,  1979). Phosphates, such as those
added  for corrosion control, can bind to copper ions as well as silver ions, reducing their
treatment effectiveness (Zevenhuizen et al.,  1979). It has also been reported that in the presence
of 20-40 mg/L of chloride ions, silver ion levels are significantly (60 percent) decreased by
complexing with chloride (and are presumably less microbiocidal) (Lin et al., 2002).
Draft - Technologies for Legionella Control:
Scientific Literature Review                       39                                 October 2015

-------
The presence of dissolved organic carbon at 2 mg/L, calcium at 100 mg/L, magnesium at 80
mg/L and bicarbonate at 150 mg/L did not appear to decrease the treatment efficacy of copper
and silver ions against Legionella in a laboratory study (Lin et al., 2002).

The impact of pH on the ionic nature (and thus the microbiocidal action) of copper in solution is
also important. At a pH of 7, exposure to 0.4 mg/L of copper resulted in a 4-log (99.99 percent)
reduction of culturable Legionella in one hour in a controlled laboratory study (Lin et al., 2002);
however, at a pH of 9, there was no appreciable  decrease in culturable Legionella over the same
period of time with the same copper exposure. At pH levels >6.0, copper forms insoluble
complexes with a number of compounds. While in the pH range typical of potable waters (pH 6
- 9), silver ions are not diminished.

With regard to the effects of temperature,  one study (Landeen et al., 1989) found no significant
difference in L. pneumophila inactivation rates in experiments conducted at room temperature
(21-23 degrees C, or 69.8-73.4 degrees F) and elevated temperature (39-40 degrees C, or
102.2-104 degrees F) using water with 0.2 mg/L free chlorine, with or without 400 |ig/L of
copper and 40 |ig/L of silver.

Installation Considerations

CSI systems can be plumbed into either the cold water entry pipe or plumbed into the hot water
line. Care should be taken to install devices downstream of any process that will remove or
exchange copper and silver ions. Note that construction that includes new copper pipe can add
copper to water for a time via leaching.

Newly installed CSI systems generally require a period of time to adjust system output in order
to achieve the desired level of metal ions.  Representatives from the manufacturer are typically
involved in on-site start-up and balancing of the system.

Monitoring Frequency and Location

Initial monitoring during start-up is critical to ensure the copper action level in the Lead and
Copper Rule is not exceeded. A facility that is considering installation of CSI should consult
with their primacy agency to determine a protocol for initial monitoring. During the initiation of
CSI, weekly monitoring with inductively coupled plasma/mass spectrometry (ICP/MS) (e.g.,
EPA Method 200.8) or atomic absorption spectroscopy (e.g., Standard Methods 311 IB) can be
conducted to determine accurate levels of copper and silver.5 As treatment proceeds, the
frequency of analysis may be reduced, but these methods remain the only reliable and accurate
means to determine copper and silver concentrations.

Operational monitoring of copper is generally conducted weekly at a variety of locations
throughout the building water system to monitor for process changes in copper concentration,
high copper concentrations that may be indicative of improper application, and no detectable
copper. Handheld colorimeters and reagents are  available for monitoring copper in the field but
5 The National Environmental Methods Index (https://www.nemi.gov/home/') "is a searchable database that allows
scientists and managers to find and compare analytical and field methods for all phases of environmental
monitoring."

Draft - Technologies for Legionella Control:
Scientific Literature Review                       40                                  October 2015

-------
laboratory analysis is needed to confirm field measurements (Sidari et al., 2014). These wet
chemical tests have practical limits on detection when near the lower limit of detection (0.04
mg/L). These field methods are not approved for compliance monitoring.

According to the United Kingdom Health and Safety Executive (2014), both copper and silver
levels should be monitored monthly, or no less than quarterly, at the same locations within the
building, using appropriate sampling procedures and submitted for analysis by ICP/MS or atomic
absorption (HSE, 2014). Sampling locations will vary in specific buildings and should include
both taps that are frequently  and infrequently used. When appropriate, the facility operator
should work with the primacy agency to determine a site sampling plan for the water system.
First-draw and flushed samples will often yield different results (Liu et al., 1994). First-draw
sample testing can indicate how periods of low water flow may affect metal levels given the
water quality conditions found in a specific building, while flush samples will measure the metal
levels in the main cold or hot water lines feeding individual taps. Knowledge of how water flows
in any particular building is essential in determining the best monitoring frequency and locations.

Maintenance Needs

The copper- and silver-containing anodes are sacrificial and should be rehabilitated periodically
as they become smaller, according to the recommendations of the manufacturer. Anodes can also
wear down due to high shear velocities (Chen et al., 2008). The anodes typically will develop
scale from calcium in all but the softest waters and should be cleaned by scraping/acid treatment
on a regular basis. Scale build-up reduces the surface area from which ions can be released,
lowering the ion output. Any time a component of a water system is opened to the environment
for maintenance, such as scraping, procedures should ensure that the system components are re-
installed in a sanitary condition (i.e., disinfected).

Regular flushing of water lines (either through the frequent use of taps or routine weekly
flushing) was cited as a critical factor in maintaining the effectiveness of CSI systems
(Kusnetsov et al., 2001; Liu et al., 1994).

    2.3.5  Ultraviolet Light Disinfection

2.3.5.1  Background

UV disinfection is a well-established treatment technology for inactivating pathogens present in
the environment.

In the drinking water context, UV disinfection was initially most widely used in Europe, with
hundreds of installations in place by 1985 (USEPA, 2006e). In North America, UV disinfection
has been more widely employed in drinking water applications  since 2000 to address health
concerns associated with Cryptosporidium. As of the spring of 2008, there were at least 300
water systems in the United States and Canada with UV installations treating flows >350 gallons
per minute (Wright et al., 2012).

UV reactor validation helps to define the operational conditions under which the pathogens of
concern are inactivated. Validation is a method of determining the operating conditions under
which a UV reactor delivers  a specified dose. This generally involves initial tests using a

Draft - Technologies for Legionella Control:
Scientific Literature Review                       41                                  October 2015

-------
surrogate organism (e.g., bacteriophage MS2) rather than the target pathogen (e.g.,
Cryptosporidium) to establish the dose relationship between the two organisms. This bioassay
helps determine the dose required for full-scale testing using the surrogate. The conditions that
are examined for full-scale testing to establish dose are flow rate, UV transmittance (UVT) (a
measure of the fraction of incident light transmitted through a material) and lamp output. The
EPA has developed guidance for validation of UV reactors (USEPA, 2006e). Independent
organizations have used this EPA guidance to develop their own validation approaches. There
are also several validation standards for UV reactors from organizations based in Austria
(Osterreichisches Normungsinstitut - ONORM) and Germany (Deutsche Vereinigung des Gas-
und Wasserfaches - DVGW) that are widely accepted and use a benchmark UV dose of 40
millijoules per square centimeter (mJ/cm2) (GLUMRBSPPHEM, 2012).

There are two methods for validating UV units. In short, the setpoint approach establishes a
measured UV intensity that corresponds to a specific dose and flow rate. The dose control
method (also referred to as the calculated dose approach) provides  a means of determining the
required intensity that corresponds to a specific flow rate, UVT and dose. For applications
related to Legionella control in buildings and other moderately sized facilities, reactors using the
setpoint approach will likely be installed due to the ease of operations and control.

2.3.5.2  Characterization of Effectiveness  against Legionella

There are several important lessons from installations of UV disinfection in hospital settings and
UV installations in general:

   •   UV disinfection has been shown to be effective at decreasing and, in some cases,
       eliminating Legionella  from facility piping.
   •   UV is only  effective at  inactivating Legionella in the water that flows through the UV
       reactor.  For existing facilities with Legionella  that is present in the piping systems
       downstream of a UV reactor, supplemental controls such as thermal treatment or
       chemical disinfection will be necessary.
   •   UV reactors need to be maintained to remain effective. The quartz sleeves that house the
       reactors can be fouled by iron, manganese, calcium carbonate or other deposits that
       decrease UV output. Lamps and other reactor components also need to be replaced
       periodically in order to maintain treatment effectiveness.

Relatively low UV doses appear to inactivate L. pneumophila (Exhibit 2-3). A dose of 1 mJ/cm2
was found adequate to achieve 99 percent (2-log) reduction in six different Legionella spp. using
irradiation times from 33 to 63 minutes depending on the species (Gilpin et al., 1985). A dose of
30 mJ/cm2 achieved 99.999 percent (5-log) reduction  in 20 minutes in a model building water
system (Muraca et  al., 1987).

Usually, there are limited opportunities for exposure to light for water treated and held in
building water systems. However, if there is a significant opportunity for light repair (repair of
UV-induced DNA  damage using photoreactivating light), such as in water used in tubs, pools
and baths, a higher UV dose should be considered. At a UV dose adequate to achieve 99.9
percent (3-log)  reduction of Legionella, subsequent exposure to fluorescent light for one hour
resulted in only a 68 percent (0.5-log) reduction following initial inactivation by low pressure
UV lamps and only 60 percent (0.4-log) reduction following inactivation by medium pressure

Draft - Technologies for Legionella Control:
Scientific Literature Review                       42                                 October 2015

-------
UV lamps (Oguma et al., 2004). Similar significant light-repair ofLegionella has been observed
by others (Knudson,  1985).

         Exhibit 2-3: UV doses (mJ/cm2) for inactivation of L. pneumophila
L. pneumophila
strain
Philadelphia Type 2
Philadelphia 1
(no light repair)
Philadelphia 1
(with light repair)
Philadelphia 1
ATCC33152
Philadelphia 1
ATCC33152
ATCC43660
Lamp
Type
LP
LP
LP
LP
MP
LP
1-log
0.92
0.5
2.3
1.6
1.9
3.1
2-log
1.84
1.0
3.5
3.2
3.8
5.0
3-log
2.76
1.6
4.6
4.8
5.8
6.9
4-log
No data
No data
No data
6.5
7.7
9.4
Reference
Antopol and Ellner,
1979
Knudson, 1985
Knudson, 1985
Oguma etal., 2004
Oguma etal., 2004
Wilson etal., 19921
Notes:
LP = Low pressure lamps, which have a single output of UV peaking around a wavelength of 254 nanometers.
MP = Medium pressure lamps, which have polychromatic (or broad spectrum) output of UV at multiple wavelengths.
1 As reported by Wright et al. 2012.

Given the sensitivity ofLegionella to UV disinfection, a well-maintained UV disinfection system
can be effective at significantly decreasing the presence ofLegionella in building water systems
to less than 10 CFU/mL (Franzin et al., 2002; Liu et al., 1995). In the case of one new facility, a
UV disinfection unit was  installed on the incoming water supply, and none of the 930 cultures of
hospital water were positive. In addition, there were no confirmed hospital-acquired Legionella
infections over a 13-year  study period (Hall et al., 2003). UV disinfection may not be effective in
pipe systems that have been colonized  by bacteria. One study found no statistical difference in
showers treated with UV  and those not treated. However, Liu et al. (1995) showed that when the
lines were disinfected, the UV treated showers remained Legionella-free for a month, while the
untreated showers were recolonized by Legionella within a week.

Fouling of the UV lamps  was found to decrease effectiveness of the UV treatment. Legionella
recolonized showers  treated with UV after one month of operation, but when filters were added
to remove particles that foul the UV lamps, the showers remained Legionella-free for a period of
three months (Liu et  al., 1995).

2.3.5.3  Potential Water Quality Issues

Water quality data is needed to adequately characterize the water to be treated by the UV reactor
and identify any pre-treatment that may be required. Manufacturers may have their own  data
requests, though the following list will cover most water quality information needed (Alaska
DEC, 2014; AWWA, 2012; GLUMRBSPPHEM, 2012; WSDOH, 2009; USEPA, 2006e):

    •   Temperature - Some reactor components may not be tolerant of water >35 degrees C (95
       degrees F). For this reason, UV should be installed on the cold water supply upstream of
       water heaters and  continuous hot water recirculation loops.
Draft - Technologies for Legionella Control:
Scientific Literature Review
43
October 2015

-------
    •   Disinfectant type and residual - Some reactor components may not be tolerant of certain
       disinfectants or high doses, so UV equipment manufacturers should be consulted about
       exposure of UV reactors to chemical disinfectants.
    •   UVT - Components in the water can absorb UV light and reduce the dose delivered to the
       microorganisms from the UV reactor. UVT (also measured as UV absorbance) is a key
       parameter in making sure that the UV reactor is properly sized for the facility.
    •   Iron and manganese - These constituents can foul quartz sleeves, leading to decreased
       UV output. Iron concentrations >0.1 mg/L may cause operational issues.
    •   Hardness - Calcium and magnesium salts may precipitate on quartz sleeves leading to
       decreased UV output. A hardness of >120 mg/L is a threshold of concern
       (GLUMRBSPPHEM, 2012).

2.3.5.4  Operational Conditions

Parameter Conditions Indicating Operational Effectiveness

The operation of a small UV reactor is  typically governed by two key parameters: the flow
through the reactor and UV sensor reading(s). Overtime, UV sensors will drift out of calibration.
For this reason, the readings from a UV duty sensor installed in the reactor should be compared
against a reference sensor temporarily inserted in the reactor. PWSs typically make these sensor
checks on a monthly basis. If the calibration ratio between the duty and reference sensor readings
is >1.2, then follow-up actions such as  recalibration or replacement of the UV sensor should be
taken (USEPA 2006e).

Installation Considerations

There are several sources of design guidance for the application of UV disinfection on potable
water supplies (Alaska DEC, 2014; AWWA, 2012; GLUMRBSPPHEM, 2012; WSDOH, 2009;
USEPA, 2006e). These references cover a range of applications from those producing only a few
gallons per day to millions of gallons per day. The following checklist is tailored to institutional
settings for Legionella control:

    •   Hydraulics should allow for even flow through the reactor. Control valves and reducers
       should be avoided within five pipe diameters upstream of the UV reactor to avoid jetting
       and swirling flow through the UV reactor.
    •   Redundancy includes providing more than one UV reactor to allow for chemical cleaning
       and equipment maintenance.
    •   Valves to isolate UV reactors are necessary. In some cases, such as when UV reactors are
       flooded with cleaning chemicals, special valve arrangements, such as double-block-and-
       bleed valves, may be required on the outlet and inlet piping.
    •   Power quality analysis includes review of sub-second power interruptions and voltage
       sags at the location of a proposed UV installation. An uninterruptible power supply or
       power conditioning equipment may need to be considered.
    •   Alarm and reactor shutdown  conditions should be clearly identified.
    •   A lamp breakage response plan should be developed that defines emergency response
       actions that will be taken if a lamp breaks. Low velocity traps or other piping
Draft - Technologies for Legionella Control:
Scientific Literature Review                      44                                 October 2015

-------
       configurations to collect broken lamp components should be considered. The potential for
       hydraulic transients should be evaluated because they may cause the quartz sleeves that
       house UV lamps to fail.

Maintenance Needs

While UV reactors are relatively simple to maintain, compared to more complex treatment
equipment, they do require routine maintenance to ensure that the UV dose remains adequate for
inactivation of pathogens. Some of the basic maintenance items include cleaning the quartz
sleeves housing the lamps and periodically replacing the lamps, as their output decreases with
time. In addition,  some reactor components can be affected by disinfectants, including chlorine,
added prior to the reactor, and additional maintenance may be required. Most UV lamps installed
in smaller reactors will typically be rated for 10,000-12,000 hours of operation (one year of
continuous operation equals 8,736 hours). For a detailed list of recommended maintenance
activities for a UV reactor, please see EPA's Ultraviolet Disinfection Guidance Manual for the
Final Long Term 2 Enhanced Surface Water Treatment Rule (USEPA, 2006e).

     2.3.6  Ozone

2.3.6.1  Background

Ozone is used in drinking water treatment for disinfection and oxidation (USEPA, 1999d, 2007).
It is generated on-site as a gas using either air or liquid oxygen and is then transferred (dissolved)
into the water phase.  When dissolved in water, molecular ozone (Os) is unstable and decomposes
to hydroxyl radical, which is a stronger and typically more reactive oxidizing agent than
molecular ozone.  Ozone decomposes quickly during water treatment (USEPA, 1999d, 2007).
Therefore, during a typical ozonation process, both molecular ozone and the hydroxyl radical
may contribute  to the oxidation of contaminants of concern. The relative importance of these two
oxidants depends on the concentrations of the oxidants and the reactivity of the contaminant with
each oxidant.

The use of ozone in U.S. water treatment facilities has been increasing over the last four decades
(Thompson et al., 2013). As an oxidant, ozone can be used to oxidize iron, manganese, taste and
odor compounds, and DBF precursors. It can oxidize  organic matter into smaller molecules that
are more easily biodegradable. As a primary disinfectant, ozone is more effective than chlorine,
chloramines and chlorine dioxide for inactivation  of Cryptosporidium, Giardia and viruses
(USEPA, 1999d, 2007). However, ozone cannot be used as a secondary disinfectant because it
decays very rapidly and cannot maintain a residual in the distribution system (USEPA, 1999d,
2007).

2.3.6.2  Characterization of Effectiveness against Legionella

Several researchers have reported rapid and effective  inactivation of Legionella, mostly in
laboratory studies (Edelstein et al., 1982; Muraca  et al., 1987; Domingue et al., 1988; Jacangelo
et al., 2002). They found little to no effect of pH (from 7.2-8.9), turbidity and temperature (from
25-43 degrees C, or 77-109.4 degrees  F) on ozone inactivation of L. pneumophila. However,
information on  ozone systems installed in hospitals and other types of buildings for L.

Draft - Technologies for Legionella Control:
Scientific Literature Review                      45                                 October 2015

-------
pneumophila control is lacking. Only one paper, by Edelstein et al. (1982), evaluated the efficacy
of ozone for eradicating L. pneumophila in hospital plumbing fixtures, but the result of the study
was inconclusive.
    •  Edelstein et al (1982) applied continuous ozonation to the water supply in an unoccupied
       hospital building to evaluate whether it would eradicate L. pneumophila from the
       plumbing fixtures with positive cultures. The water supply was split into two wings: one
       treated with ozone, while the other was untreated. In their laboratory study using distilled
       water, more than 3-log reduction in L. pneumophila was achieved by exposure to 0.32
       mg/L of ozone for 20 minutes. However, results from ozonation of the water supply
       system (average ozone levels of 0.79 and 0.58 mg/L in two study phases, respectively)
       were difficult to interpret because the non-ozonated water in the control wing also
       showed inactivation of L. pneumophila due to a higher water usage rate and an
       unexpected rise in the chlorine residual in the control wing (average chlorine residual
       levels of 0.24 and 0.12 mg/L in two study phases, respectively). Although the treatment
       wing had a smaller number of positive cultures (3 of 12) than the control wing (8 of 12),
       the researchers could not reach a conclusion on the role of ozone in the inactivation of L.
       pneumophila. The study indicated that when ozonation was stopped, L. pneumophila
       regrew and reached levels close to the pre-test conditions at the end of the stagnation
       phase. Moreover, the authors pointed out one important factor for continual dosing of
       ozone, namely that residual ozone at the faucet or shower head led to the release of
       gaseous ozone into the air (an issue discussed in Section 2.3.6.4).

    •  Muraca et al.  (1987) compared the efficacy of chlorine, heat, ozone and UV light for
       inactivating L. pneumophila in a bench-scale model plumbing system. Legionella was
       added to the system and allowed to circulate. Continuous ozonation for five hours at a
       concentration of 1 to 2 mg/L achieved a 5-log inactivation of L. pneumophila at 25 and
       43 degrees C  (77 and 109.4 degree F, respectively). Neither turbidity nor the higher
       temperature (43 degrees C, or 109.4 degrees F) was reported to affect the efficacy of
       ozone. However, the conclusion regarding no effect of turbidity was drawn from a
       comparison between non-turbid water (tap water) and turbid water (containing 4 to 5
       mg/L of suspended solids, prepared by making 1:10  dilution of a concentrated hot-water
       tank effluent sample). The turbidity of neither water was measured or reported in the
       paper. The report of the insignificant effect of temperature on inactivation in this study
       may have more to do with the experimental design (in which log-inactivation was not
       actually measured as a function of CT) than the inherent temperature dependence of the
       ozone- L. pneumophila reaction.

    •  Domingue et  al. (1988) conducted laboratory experiments to compare the bactericidal
       effects of ozone, hydrogen peroxide and free chlorine on "free" L. pneumophila cells.
       Ozone was the most potent of the three disinfectants, with a greater than 2-1 og kill of L.
       pneumophila  occurring during a 5-minute exposure to 0.10-0.3 mg/L ozone. They also
       reported little to no effect of pH on ozone inactivation of L. pneumophila., with pH
       ranging from  7.2 to 8.9. Experiments were conducted at 25, 35 and 45 degree C, and
       slightly lower ozone doses were required at 35 and 45 degree C than at 25 degree C. At  a
Draft - Technologies for Legionella Control:
Scientific Literature Review                      46                                 October 2015

-------
       higher temperature, an enhanced rate of disinfection could have been offset by a higher
       rate of ozone decomposition. Overall, the effect of temperature in this work is not clear.

    •   Jacangelo et al. (2002) conducted laboratory studies to evaluate multiple disinfectants,
       including ozone, for inactivation of waterborne emerging pathogens including
       Legionella. The ozone dose rate was 1.0 mg/L. The model-predicted CT values for a 2-
       log inactivation of Legionella at pH 7 were 2.5, 0.16 and 1.1 min-mg/L at 5, 15 and 25
       degrees C (41, 59  and 77 degrees F), respectively.

    •   Ruiz et al. (2007)  designed and constructed a pilot plant to simulate the growth of
       Legionella in biofilms and to evaluate the use of ozone to reduce both Legionella and
       biofilms. They discussed technical issues with the application of ozone for control of
       Legionella in water systems, including injection methods, automation systems, liquid and
       gas phase online measurements, and interaction of ozone with the structure of the facility.
       However, no test results were reported in this paper.
2.3.6.3  Potential Water Quality Issues

Ozone itself does not form halogenated DBFs such as THMs and HAAs. However, ozonation of
water containing inorganic bromide can produce bromate, a regulated DBF with an MCL of 10
|ig/L. The disinfection process of a PWS will likely have transformed any bromide in water to
organically bound bromine or inorganic bromamines. In either case, these forms of bromine are
less likely to contribute to bromate formation via an ozonation process in a building water
system. As such, bromate formation may not be as relevant as in the water treatment plant. Other
ozonation byproducts include aldehydes and organic acids that are more readily biodegradable
and that may contribute to assimilable organic carbon (AOC) and hence biological growth in the
distribution system. In addition, these ozonation byproducts are more likely to form some types
of DBFs upon chlorination or chloramination (Carlson and Amy, 2001; Shah and Mitch, 2012).
However, these general concepts regarding ozonation pertain to treatment of water at the plant.
Ozonation of water that has already undergone treatment, including exposure to a  chlorine or
chloramine residual in the distribution system en route to the building (e.g., hospital) has not
been studied to a great extent. Therefore, impacts of ozonation on AOC or DBF formation in a
building water system are still unclear.

2.3.6.4  Operational Conditions

Ozone disinfection is not typically impacted by pH in the range of 6 to 9. As water temperature
increases, ozone disinfection efficiency increases (USEPA, 1999d). However, because ozone
decomposes quickly in hot water, it is difficult to maintain an effective concentration throughout
the system to control Legionella. Therefore, there is a need to balance the tradeoffs between
potentially higher inactivation rates and lower CT with increased water temperature. Due to the
faster decomposition of ozone in warm water, water leaving the ozone contactor with a
concentration of 1 to 2 mg/L may not have a concentration high enough to inactivate Legionella
when it reaches distal parts of the system.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      47                                 October 2015

-------
One important aspect of ozone-based treatment in a building is the potential for ozone residual
that reaches the tap to degas from the water and expose building occupants to ozone gas. Ozone
is a toxic gas (e.g., being a principal component of smog). Ozone is corrosive and can corrode
steel pipes and fittings, concrete, rubber gaskets and other material it comes into contact with
(USEPA, 2007). Due to safety concerns and the corrosiveness of ozone, on-site generation of
ozone gas requires containment or a separate structure. Ambient air monitoring may also be
required for compliance with local regulations.

Ozone disinfection is a relatively complex process. Operational and maintenance demands are
significantly greater than those for chlorine and chloramines (USEPA, 2007).
Draft - Technologies for Legionella Control:
Scientific Literature Review                       48                                  October 2015

-------
3   Other Strategies Used to Control for Legionella

    3.1   Emergency Remediation

Emergency remediation of a building water system is triggered by an outbreak of legionellosis
associated with a potable water system, identification of suspected cases of the disease associated
with a potable water system, or identification of Legionella-poshive water results during routine
environmental testing (ASHRAE, 2015; Department of Veterans Affairs, 2014).  Several
agencies and organizations have published standards or guidance documents on when and how to
conduct emergency remediation (HSE, 2014; Department of Veterans Affairs, 2014; ASHRAE,
2015; HSE, 2009; CDC, 2003; ASHRAE, 2000).  Some of these documents apply to not only
building water systems but also cooling towers and evaporative condensers; whirlpool spas;
decorative fountains; and other aerosol-generating air coolers, humidifiers, and air washers. This
chapter provides an overview of commonly used emergency remediation methods, including
superheat-and-flush disinfection, shock hyperchlorination, POU filtration, and any combination
of these methods.

    3.1.1   Superheat-and-Flush Disinfection

3.1.1.1  Background

The superheat-and-flush disinfection method involves raising the water temperature in the hot
water heater sufficiently high to ensure hot water is delivered to outlets; circulating the hot water
through all water outlets, faucets and showerheads; and then flushing with the hot water for a
suitable period. Because Legionella can easily be killed at temperatures >60 degrees C (140
degrees F), raising the temperature of hot water tanks to 71-77 degrees C (160-170 degrees F)
and keeping the water temperature at outlets >65 degrees C (149 degrees F) during flushing are
recommended (Department of Veterans Affairs, 2014;  Sehulster and Chinn, 2003; ASHRAE,
2000). The optimal flush time reported varies from 10 to 30 minutes depending on the
characteristics of the building water system. A 30-minute flush, first adopted by Best et al.
(1983), is recommended as a good practice (Department of Veterans Affairs, 2014).

3.1.1.2  Characterization of Effectiveness against Legionella

The superheat-and-flush method can be effective  as an emergency disinfection procedure for
building hot water systems, particularly in hospital outbreak scenarios.

   •   Best et al.  (1983) first reported the use of superheat-and-flush to control Legionella from
       a hospital water supply by raising the temperature of hot water tanks as high as 77
       degrees C  (170.6 degrees F) for 72 hours and flushing the water outlets for 30 minutes
       with hot water. After flushing,  the number of samples testing positive for Legionella was
       reduced, followed by a decline in the incidence of legionellosis. The temperature of the
       hot water storage tanks was intermittently increased on eight occasions to 60-77 degrees
       C (140-170.6 degrees F), resulting in a decrease in the number of months in which  cases
       of Legionnaires' disease occurred and the proportion of nosocomial pneumonias caused
       by L. pneumophila and Pittsburgh pneumonia agent.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      49                                 October 2015

-------
    •   Darelid et al. (2002) reported the successful application of thermal shock disinfection
       after a  1991 nosocomial outbreak of Legionnaires' disease in a Swedish hospital. The hot
       water temperature was raised from 45 degrees C to 65 degrees C (113 degrees F to 149
       degrees F) to maintain the circulating hot water temperature above 55 degrees C (131
       degrees F) to control the bacteria. Environmental monitoring was conducted over a 10-
       year period to confirm whether this thermal shock treatment was sufficient or if chemical
       disinfection was required. The monitoring results showed that complete eradication of
       Legionella was not possible, but the occurrence of nosocomial Legionnaires' disease was
       controlled by maintaining the circulating hot water temperature above 55 degrees C (131
       degrees F).

However, an inadequate temperature for the superheat (below 65 degrees C, or 149 degrees F) or
a short flush time (such as five minutes) is ineffective for the control of Legionella, as
experienced at some hospitals (Chen et al., 2005). The shock treatment may not provide long-
term control of Legionella if the building water system does not maintain a proper temperature or
a residual chlorine level.

    •   Chen et al. (2005) conducted superheat-and-flush treatment on the water supply for a
       1,070-bed medical center in southern Taiwan. The treatment procedure involved
       removing faucet aerators and showerheads at distal sites; flushing distal sites with cold
       water for two minutes; and flushing distal sites with hot water at 60 degrees C for five
       minutes. The procedure was conducted once a day for five consecutive  days on each
       portion of the water system. Water samples were collected before treatment and 10 days
       after treatment. The first heat and flush treatment, performed over an eight-week period,
       eliminated Legionella from patient wards and reduced the colonization  rate in ICUs from
       80 percent to 25 percent. But two months later, the colonization rate had increased from
       zero to 15 percent in patient wards, and from 25 percent to 93 percent in the ICUs. The
       second superheat-and-flush treatment, performed over a 2-day period, resulted in much
       smaller reductions in the colonization rate.

    •   Mietzner et al. (1997) conducted thermal treatment of a hot water circuit in a hospital by
       flushing hot water (>60 degrees C, or >140 degrees F) through distal fixtures for 10
       minutes. Sampling of the faucets showed that positive samples decreased from
       approximately 80  percent to 1 or 2 percent of samples immediately following the initial
       treatment, then increased to 36 percent within 61 days of the treatment. Three additional
       heat-flush treatments resulted in zero detection of Legionella. But recolonization
       occurred within 29 days of the last treatment. The heat-flush treatment failed to provide
       long-term control  of Legionella.

Combining the superheat-and-flush method with supplemental continuous chlorination (Cristino
et al., 2012; Heimberger et al., 1991; Snyder et al., 1990) or UV light irradiation (Liu et al.,
1995) has achieved some  success in decontaminating hospital water systems.

    •   Snyder et al. (1990) reported a successful application of heat flushing followed by
       continuous supplemental chlorination to reduce L. pneumophila in a hospital hot water
       system. Twelve of 74 sampling sites in the hot water system were culture-positive for L.
       pneumophila. Heat flushing (>60  degrees C,  or > 140 degrees F) at hot water system

Draft - Technologies for Legionella Control:
Scientific Literature Review                      50                                 October 2015

-------
       outlets for 30 minutes alone reduced the number ofLegionella positive samples by 66
       percent, but within four months, the number of positive samples had increased.
       Continuous supplemental chlorination was added to the hot water system at a dosage rate
       of 2 mg/L. After six weeks, the number of Legionella-positive samples decreased from
       37 percent (43 of 115 samples) to 7 percent (8 of 115 samples). After 17 months of
       continuous supplemental chlorination, no new cases of legionellosis had occurred.

    •   Heimberger et al. (1991) reported the successful application of hot water flushing and
       supplemental chlorination to control Legionella at a tertiary care hospital in Syracuse,
       New York. L. pneumophila was found in 6 of 32 water samples including samples from
       one of two hot water tanks. Initial treatment of the hot water system included tank
       cleaning, hot water flushing and shock chlorination but did not include continuous
       supplemental chlorination. One month after initial treatment, L. pneumophila was again
       detected from a hot water tank and several taps, and another case of legionellosis
       occurred. In response, hot water flushing, shock chlorination and continuous
       supplemental chlorination were conducted. On a monthly basis, each hot water tank is
       taken offline, cleaned and treated with hot water. In the 7.5 months after these practices
       were employed, all samples were negative for Legionella and  no new cases of
       legionellosis had occurred.

    •   Cristino et al. (2012) reported the successful application of various  shock disinfection
       methods (e.g., heat shock, chemical shock with peracetic acid  and chlorine dioxide)
       followed by continuous chlorination for long-term care facilities, including three hot
       water systems that were colonized by L. pneumophila and one hot water system
       colonized by L. londiniensis. No cases of hospital-acquired legionellosis occurred during
       the study period. Although three of four systems reported that 100 percent of samples
       were positive for Legionella before and after shock treatment,  the mean Legionella count
       was reduced by up to 69 percent as a result of shock disinfection. Two years of
       environmental monitoring after shock disinfection showed that Legionella counts either
       continued to decrease or remained at post-treatment levels.

    •   Liu et al. (1995) conducted superheat-and-flush and shock chlorination treatment prior to
       UV treatment of a hospital's hot and cold water systems. Five years of surveillance data
       at untreated control sites (three showers and 20 other water outlets) showed that 30-80
       percent of sites were persistently colonized with L. pneumophila (i.e., 1-300 CPU per
       swab). The UV treatment units were located near points  of use such as showers. Filters
       were added to prevent scale accumulation on the UV lamps. The study showed that UV
       plus pre-filtration could prevent Legionella recolonization for  three months after shock
       treatment.
3.1.1.3  Potential Water Quality Issues

Regrowth ofLegionella following superheat-and-flush has been identified as an issue (Stout and
Yu, 2003). Recolonization could be caused by the survival properties ofLegionella spp. (i.e., the
ability to colonize biofilms, ability to parasitize and multiply within protozoa, and ability to enter
Draft - Technologies for Legionella Control:
Scientific Literature Review                      51                                 October 2015

-------
a VBNC state, as discussed in Section 1.2.3), or failure to properly address the conditions that
caused the problem (such as dead ends, stagnation and low flow). Researchers have revealed that
Legionella can rapidly proliferate after temperatures are lowered, presumably via microbial
response to the nutrients released by the newly killed biofilm (necrotrophy) (Temmerman et al.,
2006). This finding indicates that disturbing the microbial ecology on a short-term basis may
exacerbate pathogen regrowth in the long-term (Pruden et al., 2013).

The superheat-and-flush method generally does not require special equipment; however,  it is
labor-intensive and time-consuming due to the need to monitor hot water temperature and
flushing time. Several limitations of the superheat-and-flush method need to be recognized:

    •   Superheat-and-flush is only applicable to hot water systems with sufficient heat capacity.
       It may be ineffective when parts of the boiler/water heater or the water system fail to
       reach the required temperature (Lin et al., 2000a) or in systems that do not have hot water
       lines to every distal site.
    •   Superheat-and-flush requires considerable energy and manpower resources.
    •   Thermal disinfection will not disinfect downstream of thermostatic mixer valves and so is
       of limited value where such valves are installed (HSE, 2009).
    •   Scalding is a significant hazard (Rosenblatt and McCoy, 2014). Caution must be taken
       during emergency disinfection to avoid potential  scalding.
    •   The high temperature employed in this method can damage pipes and may cause failure
       of elastomeric seals, resulting in pump failure and leakage across valves (Rosenblatt and
       McCoy, 2014). Pipe material should be assessed before considering this approach.
3.1.1.4  Operational Conditions

Recommendations for conducting an effective superheat-and-flush, based on the published
standards and guidelines (HSE, 2014; Department of Veterans Affairs, 2014; ASHRAE, 2015;
HSE, 2009; CDC, 2003; ASHRAE, 2000), are summarized as follows:

    •   When possible, perform flushing when the fewest building occupants are present (e.g.,
       nights and weekends).
    •   Post signage and warning notices at all areas of the building to alert occupants of the
       potential scalding hazard.
    •   Maintain water heater temperatures at 71-77 degrees C (160-170 degrees F) while
       progressively flushing each outlet in the system for up to 30 minutes at 65 degrees C (149
       degrees F).
    •   Flushing multiple outlets simultaneously can save time, but should not exceed the
       capacity of the water heater and the flow capacity of the system.
    •   Perform flushing in a manner that reduces the risk of scalding and aerosolization of
       potable water in patient-care areas.
    •   Following superheat-and-flush treatment, maintain hot water  system temperature >60
       degrees C  (140 degrees F) in all hot water lines.
    •   At the end of the procedure, collect samples of water at distal outlets of the water system.
       After the water temperature has returned to normal, Legionella culture should be done to
       determine  efficacy of the treatment. Culture should be repeated within two weeks of

Draft - Technologies for Legionella Control:
Scientific Literature Review                      52                                 October 2015

-------
       treatment to determine if there is any short-term control. Repeat the procedure until
       decontamination is achieved. Following decontamination, microbiological checks must
       be repeated periodically.

    3.1.2  Shock Hyperchlorination

3.1.2.1  Background

Hyperchlorination involves injecting chlorine at an elevated concentration into the building
water system in one of two modes: shock or continuous hyperchlorination. Shock
hyperchlorination, often used for emergency disinfection, is the injection of chlorine to achieve a
level of 20-50 mg/L of free chlorine (as chlorine) (HSE, 2014; Department of Veterans Affairs,
2014). After a sufficient contact time, the water is flushed and the residual chlorine is returned to
its normal level. Continuous hyperchlorination is accomplished by continuous injection of
chlorine to achieve at least 0.5-1.0 mg/L (as chlorine) free chlorine (HSE, 2014). It is often
performed as a post-emergency disinfection procedure to aid the control ofLegionella and
biofilms. Continuous hyperchlorination in building water systems is discussed in Section 2.3.1.

3.1.2.2  Characterization of Effectiveness \gainstLegionella

Hyperchlorination can be applied to the cold- and hot-water tanks and to the entire building
water system. It may be the only option in some healthcare facilities where superheat-and-flush
cannot be used because hot water lines are not available at every distal site or they cannot reach
the required high temperature.

The success of hyperchlorination in the control ofLegionella has been mixed, as shown in the
case studies below and other studies in Section 2.3.1:

   •   Grosserode et al. (1993) reported a 10-year follow-up study of the efficacy and
       environmental effects of hyperchlorination for control of nosocomial legionellosis at a
       university hospital. In the 10 years following an outbreak in 1981, the incidence fell
       dramatically from 35 to less than 1 per 1,000 admissions; the frequency of cases of
       legionellosis also declined significantly from 16 cases among 21 tested to 5 cases among
       294 tested. No Legionella spp. were isolated from the more than 500 water samples
       collected during the 10-year period.

   •   Garcia et al. (2008) conducted long-term surveillance and studied the persistence of
      Legionella in finished water systems at a hospital  and a hotel before and after multiple
       hyperchlorination treatments. Each facility had been associated with cases of
       Legionnaires' disease. Prior to May 1998, the hotel's finished water system was
       interconnected with the industrial water system. Over the period August 1992 to April
       2001, at least seven hyperchlorination treatments were applied using a dosage rate of 10
       ppm and contact time of five hours, or a dosage rate of 20 ppm and contact time of eight
       hours. Between 1984 and 1995, the hospital's water system was treated with
       hyperchlorination four times (dosage rate and contact time not stated by authors).
       Environmental monitoring after each treatment showed that Legionella was absent for a
       period of a few months. New cases of Legionnaires' disease also occurred after

Draft - Technologies for Legionella Control:
Scientific Literature Review                       53                                  October 2015

-------
       hyperchlorination. The results of Legionella sampling also demonstrated that successive
       hyperchlorination treatments did not modify the susceptibility of bacteria to new
       treatments with chlorine or other disinfectants. The authors noted that interaction with
       other microorganisms, such as amoebae, could favor the persistence ofLegionella, as
       noted in a previous investigation by Kilvington and Price (1990).

    •   Despite shock hyperchlorination (with 50 mg/L) being applied to the cold- and hot-water
       tanks and to the whole building water system, Legionella colonization persisted in a 250-
       bed hospital (Biurrun et al., 1999). A continuous chlorine system was installed in the
       cold-water tanks to achieve approximately 0.8 mg/L  of free residual chlorine at the cold-
       water outlets (higher levels of chlorine and thermal treatment were not desired due to the
       poor condition of the piping). This, along with elimination of dead ends, replacement of
       contaminated fixtures, and other corrective measures, reduced the number of positive
       sample sites from 88 percent to  17 percent. But one month later, colonization was
       detected at a positive rate of 58 percent.

    •   A study of 62 hotels in Spain evaluated the use of continuous chlorination at 1-2 mg/L of
       free residual chlorine in the cold water, combined with intermittent thermal treatment in
       the hot water. Samples positive for Legionella dropped from approximately 30 percent
       after the first year of application, to 20 percent after three years, and to 6 percent after
       five years (Crespi and Ferra, 1998).

In general, the efficacy of shock hyperchlorination is affected by the same factors as continuous
hyperchlorination, as described in Section 2.3.1. Shock hyperchlorination, if conducted alone,
would not achieve long-term control ofLegionella. Researchers reported that Legionella could
be protected within  free-living protozoan cysts ofAcanthamoebae, which can survive free
chlorine concentrations up to 50 mg/L, the same concentration used in shock hyperchlorination
(Storey et al., 2004a; Kilvington and Price, 1990).

3.1.2.3  Potential Water Quality Issues

Regrowth ofLegionella may occur within days or weeks after shock hyperchlorination is
discontinued (Cooper and Hanlon, 2009), just as with the superheat-and-flush method. Multiple
shock hyperchlorination treatments may be needed in response to positive potable water cultures,
followed by continuous hyperchlorination or other treatment measures to achieve long-term
results.

Caution must be taken during shock hyperchlorination to avoid exposures to high disinfectant
levels. Signs and warning labels should be posted at  sinks and other outlets to warn building
occupants not to use the water (HSE, 2014). When possible, shock hyperchlorination should be
performed when the fewest building occupants are present (e.g., nights and weekends).
Draft - Technologies for Legionella Control:
Scientific Literature Review                       54                                 October 2015

-------
3.1.2.4  Operational Conditions

Recommendations for conducting effective shock hyperchlorination, based on published
guidelines (HSE, 2014; HSE, 2009; ASHRAE, 2000; Grosserode et al., 1993), are summarized
below:

Shock hyperchlorination of hot and cold water systems can significantly impact the physical
integrity of the piping system and water fixtures if applied incorrectly or too often. Corrosion of
metal pipes and appurtenances may occur from exposure to high levels of free chlorine (CDC,
2003). Therefore, routinely performing these procedures is not recommended (HSE, 2014).

Other recommendations include the following:

    •   Post signage and warning notices at all areas of the building to alert occupants of the
       potential chemical hazard.
    •   When possible, shut off and bypass any existing water treatment equipment (e.g., water
       softeners, carbon filters).
    •   Clean the tanks and associated fittings. Remove  sediment, sludge and stagnant water.
       Correct other problems that may harbor Legionella. Rubbers containing thiuram disulfide
       do not enhance the growth of Legionella., and some have suggested the use of such rubber
       in water systems (Niedeveld et al., 1986).
    •   To prevent colonization from recurring after emergency disinfection is discontinued, the
       initial conditions that caused the problem (such as stagnation and low flow) need to be
       identified and corrected; the water temperature needs to be maintained at a proper level
       (>60 degrees C, or 140 degrees F); or a residual  disinfection treatment needs to be
       installed for long-term routine operation.
    •   Consider the use of continuous hyperchlorination, or other form of long-term treatment, if
       cases continue to be identified or if a Legionella strain isolated from patients persists in
       the building water system. If Legionella isolates are limited to the hot water system,
       continuous hyperchlorination should be initiated for the hot water system alone. Chlorine
       levels need to be adjusted as required to keep DBFs at acceptable levels.
    •   Monitor the hyperchlorinated building water system for pipe damage. Assays for levels
       of copper, lead and iron, along with use of corrosion and water stability indexes, may
       permit early detection and control of corrosion problems.

    3.2   Point-of-Use Filtration

    3.2.1   Background

POU filtration is defined as the use of a device applied to a single tap for the purpose of reducing
contaminants in drinking water at that one tap. POU filtration can be used at specific taps,
faucets and showerheads as a temporary measure to provide a physical barrier against
Legionella. Hospitals have used this technology to try to reduce disease transmission (Ortolano
et al., 2005). POU water filtration may be an effective measure for remediation situations if a
limited patient area can be targeted. Filters can be installed immediately and are a better
alternative than restricting showering and providing bottled water.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       55                                  October 2015

-------
Advances in membrane filter technology have resulted in POU filtration systems capable of
removing microorganisms (USEPA, 2005b; USEPA, 2001). These treatment systems include
microfiltration (MF), ultrafiltration (UF), nanofiltration (NF) and reverse osmosis (RO)
processes. MF and UF use hollow-fiber membrane material contained in cartridges which
separate particles using a sieving mechanism based on the pore size and particle size. NF and RO
use spiral-wound (consisting of a flat sheet of membrane material wrapped around a central
collection tube) filter elements or cartridges. They are semi-permeable membranes without
definable pores. NF and RO are both pressure-driven systems that function similarly in their
removal mechanisms for microbial contaminants. Exhibit 3-1 describes the average pore size and
molecular weight cut-off requirements for different membrane filtration devices. For
comparison, Legionella cells are typically 0.3-0.9 micrometers (|im) wide and 2-20 jim long
when grown in laboratory culture (Bartram et al., 2007).

   Exhibit 3-1:  Membrane filtration guide for removal of microbial contaminants
 Nominal Pore Size (microns)
                 0.0001   0.001
          0.01
0.1
1.0
10
100
 Molecular Weight
      (daltons)
200      20,000    200,000
     Microbial
   Contaminants
                                      Cryptosporidium

                                         Viruse
     Membrane
  Filtration Process
Source: USEPA, 2005b.

EPA defines two criteria for membrane filtration technology for pathogen removal under the
Safe Drinking Water Act's (SOWA) Long Term 2 Enhanced Surface Water Treatment Rule (40
CFR141.2):

   •   The filtration system must be a pressure- or vacuum-driven process and remove
       particulate matter larger than 1  jim (for Cryptosporidium, specifically) using an
       engineered barrier, primarily via a size exclusion mechanism.
   •   The process must have a measurable removal efficiency for a target organism that can be
       verified through the application of a direct integrity test.
Draft - Technologies for Legionella Control:
Scientific Literature Review
                56
                            October 2015

-------
Many homeowners; building owners; and operators of hospitals, nursing homes, and hotels
utilize POU membrane filtration devices, often in a proactive manner but also in response to
emergencies (USEPA 2006d). Some hospitals use POU membrane filtration treatment in areas
populated with high-risk patients (e.g., in oncology, bone marrow and solid organ transplant, and
ICUs).

Two ANSI standards exist for certification of POU devices used for removal of microbial
contaminants: Standard 53 (Drinking Water Treatment Units - Health Effects) and Standard 58
(Reverse Osmosis Drinking Water Treatment Systems). POU filtration devices have been
certified by NSF International for removal of protozoa, bacteria and viruses in general, using
surrogate microorganisms as challenge organisms during testing and evaluation. Lists of POU
devices certified by independent, accredited laboratories to meet these standards are available
from Underwriters Laboratories (www.ul.com), NSF International (www.nsf.org/certified/dwtu),
and the Water Quality Association (www.wqa.org). Note that although some POU filtration
devices have been certified to meet bacterial removal standards, they have not been certified
specifically for removal of Legionella.

    3.2.2  Characterization of Effectiveness against Legionella

Several case studies describe the effectiveness of POU membrane filtration devices for removal
of Legionella.

   •   Casini et al. (2014) reported the efficacy of POU filtration installed in selected wards of
       an Italian hospital to further reduce Legionella growth within the building hot water
       system after chloride dioxide disinfection. POU filters used in this study had a 0.2-|im
       nominal pore size and 30-day replacement rate. This  integrated disinfection-filtration
       strategy, although expensive, significantly reduced Legionella counts to less than 103
       CFU/L and achieved a positive sample rate of less than 30 percent.

   •   Baron et al. (2014) evaluated a new faucet filter at five sinks in a cancer center and found
       that Legionella was removed from all filtered samples for  12 weeks, exceeding the
       manufacturer's recommended maximum duration  of use of 62 days. The filters contain a
       30-|im pre-filtration layer, a l-|im membrane and a 0.2-|im membrane.

   •   Marches! et al. (2011) performed a 10-year review of multiple treatment methods to
       control for Legionella at a hospital in Italy, including POU filtration, though information
       on the characteristics of the filters was not supplied. Filters were placed in high-risk units
       of the hospital only, where high levels of Legionella  contamination were identified, and
       were replaced every 30 days. No Legionella were detected at taps containing POU filters.

   •   Daeschlein et al. (2007) evaluated a reusable POU filter for removing waterborne
       pathogens, including L. pneumophila, in a hospital's  transplant unit for eight weeks.
       Filters had three configurations: (1) hollow fiber of polyethersulfone with pore size 0.2
       |im and surface area of 800  cm2; (2) hollow fiber of polyethersulfone with pore size 0.2
       jim, surface area of 1100 cm2, and inner encasement  coated with nanosilver; and (3) same
       as (2) with metallic silver outlet.

Draft - Technologies for Legionella Control:
Scientific Literature Review                      57                                 October 2015

-------
       Filters were placed on 18 taps (12 taps, six showers) in the hospital's transplant unit and
       each filter was monitored for pathogens at one, four and eight weeks, reprocessed and re-
       used in three additional trials. Over the test period, no Legionella or other pathogens were
       detected in any filter effluent. Because bacterial counts in filtered water exceeded the
       limit of > 100 CFU/mL eight times, the following criteria were developed to prevent
       carry-over contamination from re-use of the filters:  filters were cleaned with a strong
       chemical followed by flushing and thermal disinfection in a quality control-compliant
       washer-disinfector once a week, in addition to alcohol disinfection of the filter
       encasement. With this reprocessing, the authors determined that filters should be changed
       after four weeks in high-risk areas and after eight weeks in moderate-risk areas.

    •   A newer version of the filters  described in the example by Daeschlein et al. (2007) was
       evaluated by Vonberg et al. (2008) at a hospital in Germany. The new version had a
       membrane surface coated with nanosilver. Fifteen taps in a thoracic  surgery department
       were selected and sampled before adding filters. Filters were placed on those taps and
       sampled after one, two, three and four weeks of usage. Samples were analyzed for the
       pathogens Legionella and Pseudomonas, in addition to the indicators enterococci and
       heterotrophic bacteria. Legionella were detected in nearly half (48.3 percent) of taps
       before filters were added and only one sample (week 1) after filters were added (L.
       pneumophila serogroup  1, 4 CFU/mL); no Pseudomonas were detected.  The authors did
       not attempt to reprocess the filters as in the Daeschlein study and did see heterotrophs
       increase to > 100 CFU/mL in some filters after one week of use. The authors concluded
       that incorporation of nanosilver in the filter's membrane surface  coating may prevent
       biofilm growth in this POU device and that use of these POU filters  with weekly
       replacement in high-risk patient wards may be effective at preventing nosocomial
       legionellosis.

    •   Sheffer et al. (2005) evaluated POU filtration devices containing positively charged nylon
       membranes with a 0.2-|im nominal pore size. Filters were placed on four taps in the
       administration building at a hospital and monitored for Legionella, heterotrophic bacteria,
       and mycobacteria, along with three taps without filters, every 2-3 days for 13 days,
       before and after a one-minute flush. Samples from taps with filters before flush were
       negative for Legionella during the 13-day period, while mean concentration in taps
       without filters was 104.5 CFU/mL. Mycobacterium gordonae was isolated from 10.3
       percent of taps without filters  before flushing, but no mycobacteria were isolated from
       taps with filters before flushing. Heterotrophs were significantly reduced at taps with
       filters. One post-flush sample from a tap with a filter was positive for Legionella on day
       10 with a concentration of 5 CFU/mL. No post-flush samples from taps with or without
       filters were positive for mycobacteria. The authors concluded that the POU filters used in
       this study effectively eliminated Legionella and mycobacteria through seven days of use
       and yielded a >99 percent reduction in heterotrophic bacteria.

    •   Molloy et al. (2008) evaluated three types of POU solid block activated carbon filters for
       removal of L. pneumophila in a laboratory-simulated domestic water system: (1) carbon
       containing copper, (2) carbon containing copper and silver, and (3) carbon without


Draft - Technologies for Legionella Control:
Scientific Literature Review                      58                                 October 2015

-------
       metals. Filters were challenged with tap water seeded with L. pneumophila multiple times
       and water was monitored under simulated domestic use for six weeks. Levels of
       Legionella were reduced by all three filters by nearly 8 log, but they were detected in all
       filter effluents for the length of the study. The authors concluded that the organisms
       attached to the carbon blocks and sloughed off over time.
    3.2.3  Potential Water Quality Issues

POU filters have the potential to concentrate bacteria and foster growth of pathogens. Failure of
filters could lead to the release of high levels of pathogens. Membranes may foul, clog with scale
(from salts in feed water), or be degraded by microorganisms.


    3.2.4  Operational Conditions

In general, most POU devices include pre-filtration (usually granular activated carbon) to treat
inlet water and prevent clogging of the central membrane, the central filtration membrane, and
post-filtration, in a module configuration. Design guidance for POU filtration devices can be
found in the EPA's Membrane Filtration Guidance Manual (USEPA, 2005b). Operators are
advised to follow the manufacturer's operational guidance for the POU system being employed.
There is a variety of commercially available systems with unique design features and operational
conditions. Additional guidance on operation and maintenance for POU treatment devices,
including examples of maintenance logs, can be found in EPA's POU or POE Treatment
Options for Small Drinking Water Systems (USEPA, 2006d). A detailed maintenance log should
be kept for each system, based on the state's requirements, if any. Maintenance typically
includes the following:

   •   Tracking flows - Flow meters are used to measure the total flow treated, as flow values
       may be used to determine filter membrane or other component replacement parameters.
   •   Replacement parts - Components should be replaced as required by the manufacturer or
       monitoring data, to ensure water free of microbial contaminants. Minimal components
       needing regular replacement include exhausted membranes and  pre- and post-filters. A
       30-day replacement rate was reported in the studies using POU filters for Legionella
       control in hospitals (Casini et al., 2014; Marches! et al., 2011).
   •   Visual check of mechanical conditions - All components, including the mechanical
       warning device, should be inspected visually on a regular basis and parts
       replaced/repaired if necessary, in addition to being replaced as specified by the routine
       replacement schedule.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      59                                October 2015

-------
4   Questions and Answers on Legionella Control in Building Water Systems

    4.1   Public Health Concerns

Ql. What are the threats from Legionella in a building water system?

Legionella is a naturally occurring bacterial pathogen that can be present in municipal and other
water supplies. Building water systems may provide conditions (e.g., low flow/long retention
times, optimal temperature, and low disinfectant residual levels) that favor its growth to levels
that may result in increased risk to public health (i.e., developing legionellosis). For more details
please see Section 1.2 of this document.

Q2: Do all species of Legionella cause disease?

Although most of the diagnosed cases of legionellosis (Legionnaires' disease and Pontiac fever)
are associated with Legionellapneumophila (serogroup 1), approximately half of all the species
of Legionella have been associated with clinical cases of legionellosis. However, it is likely that
most Legionellae can cause human disease under the appropriate conditions (e.g., in individuals
in higher risk groups) (Borella et al., 2005; Fields, 1996; Fang et al., 1989). For additional
information, please see Section 1.2 of this document.

Q3. Do you need to eliminate all Legionellae in order to have a safe building environment?

Not necessarily. Due to the highly variable and inconclusive information that is available, it is
not feasible to establish a definitive action level below which the risk from disease is eliminated.

Building water system operators may choose to assess the population they serve for individual
factors that may increase the risk of disease (e.g., age, immunosuppression) to reduce the  risks
from Legionellae. See Section 1.2.2 for additional information on risk factors. The building
water operator and/or manager may want to evaluate the building water system processes  that
could contribute to Legionella growth (e.g., long hot water holding times). This assessment
should allow the building water system manager to determine the necessary stringency of the
control plan and measures (see Section 2.2 for additional information).

    4.2   Potential Regulatory Requirements

Q4. What constitutes being a regulated public water system?

The criteria for being a regulated public water system (PWS) are codified at 40 CFR 141.3.
Where there are questions about the application of these criteria, the primacy agency (typically
the state) will make the determination based on  these criteria and any relevant site-specific
considerations.

Q5. Will a building that installs a treatment specifically designed for Legionella and  serves
a population above the threshold of a PWS definition be subject to SDWA requirements?

See response to Question 4.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      60                                 October 2015

-------
Q6. Why do I need to comply with drinking water standards if I'm only treating the hot
water (not for drinking purposes)?

EPA considers water for human consumption to include water for drinking and food preparation
as well as water for brushing teeth, showering and hand washing (See 63 FR41940; August 5,
1998).

Q7. If I comply with the Federal Insecticide,  Fungicide and Rodenticide Act (FIFRA)
pesticide registration requirements, am I in compliance with the Safe Drinking Water Act
(SDWA) requirements?

The pesticide registration requirements under FIFRA are independent of the SDWA
requirements, and each mandate targets complementary, yet different, environmental and public
health protection objectives. Registration of a pesticide product under FIFRA does not mean that
it meets the requirements of other environmental and public health protection statutes, including
the SDWA or vice versa.

The objective of FIFRA is to protect human health and the environment through federal control
of pesticide distribution, sale and use. All pesticides distributed or sold in the United States must
be registered (licensed) by EPA. Registration assures that pesticides are properly labeled and
that, if used in accordance with their approved labeling, they will not cause unreasonable adverse
effects on the environment or human health.

Registration is required for pesticide products that are sold or distributed in the United States for
antimicrobial applications. However, this requirement does not necessarily preclude the use of
other disinfectants recognized under the Surface Water Treatment Rules such as chlorine,
chlorine  dioxide and chloramines.

The SDWA is the main federal law that ensures the quality of Americans' drinking water. Under
SDWA, EPA sets standards for drinking water quality and oversees the states, localities and
water suppliers who implement those standards.

While there are no requirements under SDWA that prohibit the installation of a given
technology, the primacy agency is responsible for accepting the installation or usage of new
technologies  in PWSs. Both SDWA and FIFRA allow states  to have stricter standards than those
prescribed in federal regulations. This includes  the authority  to request additional data or
information before approving a drinking water treatment technology or pesticides (see response
to Question 8 for additional information) to be used within the state. In the  case of a pesticide, a
state can require compliance with a state-specific  pesticide registration process in addition to the
EPA registration. With regard to technologies for drinking water treatment, primacy agencies
and technology manufacturers can refer to EPA's Water Supply Guidance (WSG) 90 for
guidance on some of the types of data or information that may be requested as part of the
primacy  agency's evaluation and approval of alternative drinking water treatment technologies.

 Q8. What are the pesticide registration requirements related to pesticide  products and
devices for the control of Legionella (and other microbial contaminants)?

Pesticide products and devices that make antimicrobial claims of efficacy against L.
pneumophila are subject to certain EPA regulatory requirements. FIFRA defines a pesticide as

Draft - Technologies for Legionella Control:
Scientific Literature Review                       61                                October 2015

-------
any substance or mixture of substances intended for preventing, destroying, repelling or
mitigating any pest. The term pesticide includes antimicrobials (e.g., sanitizers and disinfectants)
in addition to various other substances used to control pests. Products that contain a substance or
mixture of substances and that make a pesticidal claim must be registered by EPA prior to sale or
distribution. For products that claim efficacy against a public health pest, an applicant must
submit data demonstrating that efficacy to obtain a registration.

While devices are subject to certain EPA regulatory requirements, they do not require
registration as pesticide products. A pesticide device is defined in FIFRA as an instrument or
contrivance (without a chemical substance) that is used to destroy, repel, trap  or mitigate any
pest such as insects, weeds, rodents, animals, birds, mold/mildew, bacteria and viruses. A device
is subject to the FIFRA prohibition against misbranding and must be produced in an EPA-
registered establishment. To be considered a device, the item must not be sold or distributed with
a substance or mixture of substances to perform its intended pesticidal purpose. For example,
drinking water treatment technologies that incorporate a substance (e.g., chlorine, chloramine,
silver or copper in the form of an electrode) for pesticidal purposes are not considered to be
pesticide devices and must be registered. Additional information on pesticide  devices and the
associated FIFRA requirements is available on EPA's website and includes fact sheets and a
registration manual. For additional information regarding obtaining installation or operating
permits see response to Question 20.

    4.3   Control Measures

Q9. What measures can a building operator take to control the colonization and
amplification ofLegionella in a building water system?

Buildings can vary in their characteristics (e.g., dimensions, location with respect to the servicing
PWS) as well as their purposes. The appropriate treatment depends on those characteristics  and
purposes. A multi-barrier approach can be used to ensure a comprehensive preventative approach
is taken to address potential health risks related to the building water system.  See Section 2.2 of
this document for information on multi-barrier approaches.

Q10. Does EPA regulate Legionella?

EPA regulates Legionella under the Surface Water Treatment Rule (SWTR).  The SWTR has
treatment technique requirements to control for Giardia and viruses. The SWTR assumes that if
sufficient treatment is provided to control for Giardia and viruses (i.e., 3-log inactivation of
Giardia and 4-1 og inactivation of viruses), then Legionella should be addressed as well. In
addition, the Revised Total Coliform Rule and the Ground Water Rule have treatment technique
requirements that address bacteria, which provide some control ofLegionella. All of these rules
apply to PWSs. They would not apply to building water systems unless the facility is a regulated
PWS. See response to Question 4.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       62                                 October 2015

-------
Qll. What treatment technologies does EPA approve for control ofLegionella in drinking
water?

EPA does not approve any treatment technologies specifically for control ofLegionella in
drinking water. See response to Question 17.

Q12. What is supplemental disinfection?

For the purposes of this document, supplemental disinfection refers to any additional treatment
added, such as that added to reduce Legionella, to supplement or boost the treatment provided by
the distributor of the water being received. To address water quality and pathogen control needs,
building operators and owners, after a careful review of building and water system conditions,
may wish to implement a supplemental application of a disinfectant specific to and within the
building water system. Supplemental treatment in this case does not refer to emergency
disinfection by the building owner or operator (i.e., water disinfected as a result of the
emergency that is not intended for human consumption).

Q13. What happens if I add supplemental disinfection in my building?

You will need to determine whether the application of that treatment triggers additional SDWA
requirements, for example, requiring the building water system to be fully regulated as a PWS.
You may wish to consult with your water supplier (i.e., PWS) to better understand any potential
water quality issues before making treatment-related decisions. If a decision to add treatment to
the building water system seems likely, EPA advises building owners to consult with their
primacy agency to determine if any SDWA requirements apply; in addition, there may be state or
local requirements that apply to the treatment or the water system.  See Section 1.4 for additional
information.

Q14. What should I do before I consider supplemental water treatment?

Building owners and operators considering the addition of a supplemental water treatment
system are encouraged to contact their primacy agency and other state and local authorities and
familiarize themselves with applicable federal, state and local regulations (e.g., building codes,
local health codes). Building owners and operators should also become very familiar with the
characteristics and needs of their system to help determine the most appropriate supplemental
water treatment. Please see Section 2.3 of this document for more information on treatment
technologies that could be used as supplemental treatment.

Q15. Are there any advantages to supplemental disinfection?

Facilities that design, operate, control  and monitor supplemental treatment systems ensure that a
high level of water quality is maintained, improving public health protection and reducing
liabilities. Providing supplemental disinfection can help maintain the high level of water quality
throughout the building water system.

Q16. Are there any disadvantages to supplemental disinfection?

Operating supplemental water treatment requires the commitment of financial, physical and staff
resources. An additional disadvantage is that installation of supplemental  treatment could lead to
Draft - Technologies for Legionella Control:
Scientific Literature Review                      63                                October 2015

-------
a false sense of security. Installation of supplemental disinfection does not negate the need for
building owners/customers to respond to water supply emergencies (i.e., boil water advisories,
"do not consume" notices, "do not use" notices) issued by the selling system.

    4.4   New Technology Approval

Q17. What is EPA's process for approval of new treatment technologies to control
Legionella in drinking water?

Rather than approve new treatment technologies, EPA recognizes them for their capacity to
achieve treatment technique requirements for control of pathogens (under rules such as the
SWTR. GWR and LT2ESWTR). The EPA defers to the primacy agency for the approval of
technologies that PWSs can use to comply with treatment technique regulations. The EPA, in
cooperation with the Association of State Drinking Water Administrators, state drinking water
program personnel, industry representatives, and other stakeholders, developed a Water Supply
Guidance document (WSG 90) that provides a streamlined protocol to facilitate consistent state
approvals of new drinking water treatment technologies. WSG 90 is not meant to replace current
state plan review and approval processes. Note that new technologies may need to comply with
registration or other requirements for pesticide products and devices under FIFRA (See response
to Question 8).

Q18. How do states approve new treatment technologies for Legionella control?

Many states utilize a plan approval or permitting process to approve the installation of treatment
at PWSs. For new treatment technologies, many states (47) require conformance to ANSI/NSF
Standard 60 and/or 61. In addition, states may require third-party validation of efficacy. States
may also use  the protocol described in WSG 90 to facilitate consistent  state approvals of new
drinking water treatment technologies. WSG 90 is not meant to replace current state plan review
and approval  processes.

If you are planning to install additional treatment in your building/facility, consult with your
primacy agency regarding any additional specific requirements. If you  require further assistance,
contact the appropriate EPA regional office for additional information. For additional
information, please refer to Question 30 (Additional Sources of Information).

    4.5   Permitting

Q19. What is the procedure for plan review and permitting to operate a Legionella
treatment system?

The procedure for plan review and approval or permitting varies from state to state. Some states
require a permit to construct/install a treatment system and a separate permit to operate the
system. Water system owners should consult with their water provider and primacy agency to
find out specific procedures and requirements. Alternatively, contact the appropriate EPA
regional office for additional information.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      64                                 October 2015

-------
    4.6   Sampling and Monitoring

Q20. If I am only treating the hot water, where should I take compliance samples?

SDWA requirements for PWSs apply to any water for human consumption regardless of the
temperature of the water. Consult with your primacy agency regarding site-specific
requirements. Alternatively, contact the appropriate EPA regional office for additional
information.

Q21. What type of sampling (based on the selected treatment) will I be required to do?

Consult with your primacy agency regarding applicable sampling requirements. Alternatively,
contact the appropriate EPA regional office for additional information.

Q22. What residual disinfectants and DBFs do I need to monitor?

The Stage 1 DBPR and Stage 2 DBPR established MCLs for DBFs and MRDLs for disinfectant
residuals. It also specified the monitoring requirements that PWSs must perform for residual
disinfectants and DBFs (type, frequency and location), depending on the type of systems,
population served and type of disinfectants being used. For example, a regulated PWS using
chlorine or chloramine must monitor for TTHM, HAAS and residual chlorine. A regulated PWS
using chlorine dioxide must monitor for TTHM, HAAS, chlorite and chlorine dioxide. See the
Stage 2 DBPR Quick Reference Guide for more information.

Q23. How do I monitor for chlorine dioxide?

For regulated PWSs, chlorine dioxide is a contaminant with acute health effects and must be
monitored daily at the entry point to the distribution system and at each treatment unit location to
ensure it is not exceeding the MRDL of 0.8  mg/L. Chlorine dioxide has a short sample hold time
and must be measured immediately after collection; therefore, on-site analysis at the water
system is required.

The approved analytical methods for measuring residual disinfectant concentration for chlorine
dioxide include Standard Method 4500-C1O2 D and E (40CFR 141.131(cY). These methods
indicate that systems may also measure residual disinfectant concentrations for chlorine dioxide
by using DPD colorimetric test kits, and the person conducting the measurements must be
approved by EPA or the state.

Facilities considering installation of this technology should coordinate with the primacy agency
to determine requirements for approval. Some primacy agencies are requiring the water systems'
staff members who are conducting on-site analysis of chlorine dioxide and chlorite to prove
analytical competency by performing an initial demonstration of capability as well as an ongoing
demonstration of capability.

Q24. Can I send daily  chlorite and chlorine dioxide samples to a lab?

If you are subject to Stage 2 requirements, you are required to analyze daily chlorite samples on-
site and send monthly chlorite samples to a certified laboratory (see table footnote 8 in 40 CFR
141.131(b)). You cannot send daily chlorine dioxide samples to a laboratory for analysis because

Draft - Technologies for Legionella Control:
Scientific Literature Review                       65                                October 2015

-------
chlorine dioxide has a short sample hold time and must be measured on-site immediately after
collection.

Q25. Does EPA require Legionella monitoring if treatment for its control is installed? If so,
what are the targets for meeting control?

No; EPA does not have requirements for Legionella monitoring. However, state or local agencies
may specify such requirements in the permit conditions for the permit to operate issued to the
facility. In addition, there may be requirements for monitoring of water quality parameters or
treatment process parameters on a routine basis.

Q26. If a facility has treatment and has either an outbreak or has Legionella test results
showing detections, are they required to report to the primacy agency?

If the facility is a regulated PWS, there will be reporting requirements defined by the state in
response to a waterborne disease outbreak, Legionella detection, or even water quality conditions
that may contribute to an outbreak. Facilities that are not PWSs might have to share information
about an outbreak or Legionella detection with local health authorities. These agencies will be
able to assist with response actions.

     4.7   Operator Certification

Q27. What is operator certification?

Operator certification is a program that establishes minimum professional standards for the
operation and maintenance of PWSs to help protect human health and the environment. In 1999,
EPA issued operator certification program guidelines specifying minimum standards for
certification and recertification of the operators of community and non-transient non-community
PWSs. These guidelines are currently being implemented through state operator certification
programs. While the specific requirements vary from state to state, the goal of all operator
certification programs is to ensure that skilled professionals are overseeing the treatment and
distribution of safe drinking water. Operator certification is an important step in promoting
compliance with SDWA. More information on operator certification is available on EPA's
website.

Q28. Do I need to have a certified operator for this treatment system?

If your facility is a regulated PWS, (see questions in Section 4.2 to determine if you are
regulated), you may be required to have a certified operator for your treatment system. There are
two general types of operator certification — one for treatment systems and one for distribution
systems. Within each type, there  are different levels of certification. Consult with your primacy
agency to determine the type and level of certified  operator required for your specific system.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       66                                 October 2015

-------
    4.8  Unintended Consequences

Q29. What are some of the unintended consequences of installing additional treatment for
Legionellal

For unintended consequences related to specific treatment technique requirements, please see the
specific treatment sections in this document.

    4.9  Additional Sources of Information

Q30. How can I obtain additional information on each treatment method?

For additional information on any remaining general questions you can contact:
    •   Safe Drinking Water Hotline by phone or email at:
          -  (800)426-4791
          -  hotline-sdwa@epa.gov
Draft - Technologies for Legionella Control:
Scientific Literature Review                      67                                October 2015

-------
5   References

Addiss, D.G., J.P. Davis, M. La Venture, PJ. Wand, M.A. Hutchinson, and R.M. McKinney.
1989. Community-acquired Legionnaires' disease associated with a cooling tower: evidence for
longer-distance transport ofLegionellapneumophila. American Journal of Epidemiology,
130(3): 557-568.

Aieta, E.M. and J.D. Berg. 1986. A review of chlorine dioxide in drinking water treatment.
JournalAWWA, 78(6): 62-72.

Alaska DEC. 2014. Treatment - ultraviolet (UV) disinfection system checklist. Alaska
Department of Environmental Conservation. Available online at:
http://dec.alaska.gov/eh/dw/Engineering/Plan rev checklist.htm.

Aileron, L., A. Khemiri, M. Koubar, C. Lacombe, L. Coquet, P. Cosette, T. Jouenne, and J.
Frere. 2013. VBNC Legionellapneumophila cells are still able to produce virulence proteins.
Water Research, 47(17): 6606-6617.

American Society for Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE). 2000.
Guideline 12-2000. Minimizing the risk of legionellosis associated with building water systems.

American Society for Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE). 2015.
ANSI/ASHRAE Standard 188-2015 Legionellosis: Risk management for building water systems.

American Water Works Association (AWWA). 2012. ANSI-AWWA Standard Fl 10 - ultraviolet
disinfection systems for drinking water. American Water Works Association, Denver, CO.

American Water Works Association (AWWA). 2013. ManualM56.  Nitrification Prevention and
Control in Drinking Water. Second Edition. American Water Works Association, Denver CO.

American Water Works Association (AWWA) Disinfection Systems Committee. 2008.
Committee Report: Disinfection Survey, Part 1 - Recent Changes, Current Practices and Water
Quality. Journal AWWA, 100(10):  76-90.

Antopol,  S.C. and P.D. Ellner. 1979. Susceptibility of Legionella pneumophila to ultraviolet
radiation. Applied and Environmental Microbiology, 38(2): 347-348.

Araya, M., B. Chen, L.M. Klevay,  JJ. Strain, L. Johnson, P. Robson, W. Shi, F.  Nielsen, H. Zhu,
M. Olivares, F. Pizzarro, and L.T. Haber. 2003a. Confirmation of an acute no-observed-adverse-
effect and low-observed-adverse-effect level for copper in bottled drinking water in a multi-site
international study. Regulatory Toxicology and Pharmacology, 38(3):389-399.

Araya, M., M.C. McGoldrick, L.M. Klevay, JJ.  Strain, P. Robson, F. Nielsen, M. Olivares, F.
Pizarro, L. Johnson, and K.A. Poirier. 2001. Determination of an acute no-observed-adverse-
effect level (NOAEL) for copper in water. Regulatory Toxicology and Pharmacology,
34(2): 137-145.


Draft - Technologies for Legionella Control:
Scientific Literature Review                      68                                 October 2015

-------
Araya, M., M. Olivares, F. Pizarro, M. Gonzalez, H. Speisky, and R. Uauy. 2003b.
Gastrointestinal symptoms and blood indicators of copper load in apparently healthy adults
undergoing controlled copper exposure. American Journal of Clinical Nutrition, 77(3):646-650.

Araya, M., M. Olivares, F. Pizarro, A. Llanos, G. Figueroa, and R. Uauy. 2004. Community-
based randomized double-blind study of gastrointestinal effects and copper exposure in drinking
water. Environmental Health Perspectives, 112(10): 1068-1073.

Araya, M., C. Pefia, F. Pizarro, and M. Olivares. 2003c. Gastric response to acute copper
exposure. The Science of the Total Environment, 303(3):253-257.

Armon, R., J. Starosvetzky, T. Arbel, andM. Green. 1997. Survival of Legionellapneumophila
and Salmonella typhimurium in biofilm systems. Water Science & Technology, 35(11-12): 293-
300.

Baron, J.L., T. Peters, R. Shafer, B. MacMurray, and I.E. Stout. 2014. Field evaluation of a new
point-of-use faucet filter for preventing exposure to Legionella and other waterborne pathogens
in health care facilities. Amer. Journal of Infection Control., 42(2014): 1193-1196.

Baron, J. L., A. Vikram, S. Duda,  J. E. Stout, and K. Bibby. 2014. Shift in the microbial  ecology
of a hospital hot water system following the introduction of an  on-site monochloramine
disinfection system. PLOS ONE. 9(7).

Barrette, W.C., D.M. Hannum, W.D. Wheeler, and J.K. Hurst.  1989. General mechanism for the
bacterial toxicity of hypochlorous  acid: Abolition of ATP production. Biochemistry, 28:  9172-
9178.

Bartram, J., Y. Chartier, J.V. Lee,  K. Pond, and S. Surman-Lee. 2007. Legionella and the
prevention of legionellosis.  1st edition, World Health Organization, Geneva, Switzerland.

Berg, J.D., J.C. Hoff, P.V. Roberts, and A. Matin. 1988. Resistance of bacterial subpopulations
to disinfection by chlorine dioxide. JournalAWWA, 80(9): 115-119.

Berger, T.J., J.A.  Spadaro, S.E. Chapin, andR.O. Becker. 1976. Electrically generated silver
ions: quantitative effects on bacterial and mammalian cells. Antimicrobial Agents and
Chemotherapy, 9(2): 357-358.

Berk, S.G., R.S. Ting, G.W. Turner, and RJ. Ashburn.  1998. Production of respirable vesicles
containing live Legionella pneumophila cells by two Acanthamoeba spp. Applied and
Environmental Microbiology. 64:  279-286.

Best, M., V.L. Yu, I.E. Stout, R.R. Muder, A. Goetz, and F. Taylor. 1983. Legionella in  the
hospital water supply: Epidemiologic link with disease and evaluation of a method of control of
nosocomial Legionnaires' disease  and Pittsburgh pneumonia. Lancet, 2: 307-310.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      69                                 October 2015

-------
Biurrun, A., L. Caballero, C. Pelaz, E. Leon, A. Gago, 1999. Treatment of a Legionella
pneumophila-colonized water distribution system using copper-silver ionization and continuous
chlorination. Infection Control and Hospital Epidemiology.  20(6): 426-428.

Blanc, D.S., P. Carrara, G. Zanetti, and P. Francioli. 2005. Water disinfection with ozone, copper
and silver ions, and temperature increase to control Legionella: seven years of experience in a
university teaching hospital. Hospital Infection, 60: 69-72.

Blatt, S.P., M. D. Parkinson, E. Pace, P. Hoffman, D. Dolan, P. Lauderdale, R. Zajac, and G.P.
Melcher. 1993. Nosocomial Legionnaires' disease: aspiration as a primary mode of disease
acquisition.  The American Journal of Medicine, 95: 16-22.

Boffardi, B.P. and  J. Hannigan. 2013. A limited evaluation of pitting corrosion of copper piping
in a hospital domestic hot water system using copper-silver  ionization for Legionella control. The
Analyst Technology Supplement, 4: 38-42.

Bohach, G.A. and I. S. Snyder. 1983. Cyanobacterial stimulation of growth and oxygen uptake
by Legionella pneumophila. Applied and Environmental Microbiology. 46(2): 528-531.

Bollin, G.E., J.F. Plouffe, M.F. Para, and B. Hackman. 1985. Aerosols containing Legionella
pneumophila generated by shower heads and hot-water faucets. Applied and Environmental
Microbiology, 50(5): 1128-1131.

Bondarczuk, K. and Z. Piotrowska-Seget. 2013. Molecular basis of active copper resistance
mechanisms in Gram-negative bacteria.  Cell Biology Toxicology, 29: 397-405.

Borella, P., E. Guerrieri, I. Marches!, M. Bondi, and P. Messi. 2005. Water ecology of
Legionella and protozoan: environmental and public health  perspectives. Biotechnology Annual
Review, 11:  355-380.

Brieland, J., M. McClain, L. Heath, C. Chrisp, G. Huffnagle, M. LeGendre, M. Hurley, J.
Fantone, and C. Engleberg. 1996. Coinoculation with Hartmannella vermiformis enhances
replicative Legionella pneumophila lung infection in a murine model of Legionnaires' disease.
Infection and Immunity, 64: 2449-2456.

Buse, H.Y. andNJ. Ashbolt. 2011. Differential growth of Legionella pneumophila strains within
a range of amoebae at various temperatures associated with  in-premise plumbing. Letters in
Applied Microbiology, 53:217-224.

Buse, H.Y. andNJ. Ashbolt. 2012. Counting Legionella cells within single amoeba host cells.
Applied and Environmental Microbiology, 78(6): 2070-2072.

Buse, H. Y., J. Lu,  I. T. Struewing, N. J. Ashbolt. 2013. Preferential colonization and release of
Legionella pneumophila from mature drinking water biofilms grown on copper versus
unplasticized polyvinylchloride coupons. Int. J. Hyg. Environ. Health, doi
10.1016/j.ijheh.2013.04.005


Draft - Technologies for Legionella Control:
Scientific Literature Review                      70                                 October 2015

-------
Buse, H.Y., M.E. Schoen and NJ. Ashbolt. 2012. Legionellae in engineered systems and use of
quantitative microbial risk assessment to predict exposure. Water Research, 46: 921-933.

Calomoris, JJ. and K. Christman. 1998. How does chlorine added to drinking water kill bacteria
and other harmful organisms? Why doesn't it harm us? Scientific American. Available online at:
http://www.scientificamerican.com/article/how-does-chlorine-added-t-1998-05-04/.

Carducci, A., M. Verani, and R. Battistini. 2010. Legionella in industrial cooling towers:
monitoring and control strategies. Letters in Applied Microbiology, 50(1): 24-29.

Carlson, B.H. and G.L. Amy, 2001. Ozone and Biofiltration Optimization for Multiple
Obj ectives. Journal A WWA ,93:1:88.

Casini, B., A. Buzzigoli, M.L. Cristina, A.M. Spagnolo, P. DelGiudice, S. Brusaferro, A. Poscia,
U. Moscato, P. Valentini, A. Baggiani, and G. Privitera. 2014. Long-term effects of hospital
water network disinfection on Legionella and other waterborne bacteria in an Italian University
Hospital. Infection Control and Hospital Epidemiology., 35(3): 293-299.

Centers for Disease Control and Prevention (CDC). 1997. Sustained transmission of nosocomial
Legionnaires' disease - Arizona and Ohio. Morbidity and Mortality Weekly Report (MMWR),
46(19): 416-421.

Centers for Disease Control and Prevention (CDC). 2000. Legionnaires' disease associated with
potting soil - California, Oregon and Washington, May-June 2000. Morbidity and Mortality
Weekly Report (MMWR). 49: 777-778.

Centers for Disease Control and Prevention (CDC). 2003. Guidelines for environmental infection
control in health-care facilities. Recommendations of CDC and the Healthcare Infection Control
Practices Advisory Committee  (HICPAC). Morbidity and Mortality Weekly Report (MMWR), 52
(RR10): 1-42.

Centers for Disease Control and Prevention (CDC). 2005. Legionellosis  (Legionnaire's disease or
Pontiac fever) case definition. Available online at:
http://wwwn.cdc.gov/NNDSS/script/casedef.aspx?CondYrID=741&DatePub=l/l/2005%2012:0
0:00%20AM.

Centers for Disease Control and Prevention (CDC). 2008. Surveillance for waterborne disease
and outbreaks associated with drinking water and water not intended for drinking - United
States, 2005-2006. Morbidity andMortality Weekly Report Surveillance  Summaries,  57: 39-62.

Centers for Disease Control and Prevention (CDC). 2011. Legionellosis - United States, 2000-
2009. Morbidity andMortality  Weekly Report (MMWR), 60: 1083-1086.

Centers for Disease Control and Prevention (CDC). 2012. Summary of Notifiable Diseases -
United States, 2012. Morbidity andMortality Weekly Report (MMWR), 61(53): 1-121.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      71                                October 2015

-------
Centers for Disease Control and Prevention (CDC). 2013a. Legionella (Legionnaires' disease and
Pontiac fever). Available online at: http://www.cdc.gOv/Zeg-/owe//a/index.htmL.

Centers for Disease Control and Prevention (CDC). 2013b. Surveillance for waterborne disease
outbreaks associated with drinking water and other non-recreational water — United States,
2009-20 \Q. Morbidity and Mortality Weekly Report Surveillance Summaries, 62(35): 714-720.

Cervero-Arago, S., R. Sommer, and R.M Araujo. 2014. Effect of UV irradiation (253.7 nm) on
free Legionella and Legionella associated with its amoebae hosts. Water Research, 67 (2014):
299-309.

Chen, C.H., L.C. Lin, Y. J. Chang, C.E. Liu, M.S.  Soon, and C.S. Huang. 2013. Efficacy of
copper-silver ionization for controlling fungal colonization in water distribution  systems.  Water
and Health, 11(2): 277-280.

Chen, Y.S., Y.E. Lin,  Y.C. Liu, W.K. Huang, H.Y. Shih, S.R. Wann, S.S. Lee, H.C. Tsai, C.H.
Li, H.L. Chao, C.M. Ke, H.H. Lu, and C.L. Chang. 2008. Efficacy of point-of-entry copper-
silver ionization system in eradicating Legionellapneumophila in a tropical tertiary care hospital:
implications for hospitals contaminated with Legionella in both hot and cold water. Hospital
Infection,  68: 152-158.

Chen, Y.S., Y.C. Liu, S.S. Lee, H.C. Tsai, S.R. Wann, C.H. Kao, C.L.  Chang, W.K. Huang, T.S.
Huang, H.L. Chao, C.H. Li, C.M. Ke, and Y.S. Lin. 2005. Abbreviated duration  of superheat-
and-rush and  disinfection of taps for Legionella disinfection: lessons learned from failure.
American Journal of Infection  Control.  33(10): 606-610.

Chord, F., P. Fascia, F. Mallaval, J. Cornillon, L. Roesch, B. Pozzetto, F. Grattard, P. Berthelot.
2011. Chlorine dioxide for Legionella spp. disinfection: a danger for cross-linked polyethylene
pipes? Hospital Infection, 78(3): 242-243.

Cirillo, J.D., S.L.  Cirillo, L. Yan, L.E. Bermudez, S. Falkow, and L.S.  Tompkins. 1999.
Intracellular growth in Acanthamoeba castellanii affects monocyte entry mechanisms and
enhances virulence of Legionella pneumophila. Infection and Immunity, 67(9): 4427-4434.

Colbourne, J.S. andPJ. Dennis. 1989. The ecology and survival of Legionella pneumophila.
Water and Environment. 3: 345-350.

Collier, S.A., LJ. Stockman, L.A. Hicks, L.E. Garrison, FJ. Zhou,  andMJ. Beach. 2012. Direct
healthcare costs of selected diseases primarily or partially transmitted by water. Epidemiology
and Infection,  140: 2003-2013.

Cooper, I.R. and G.W. Hanlon. 2009. Resistance of Legionella pneumophila serotype 1 biofilms
to chlorine-based disinfection. Journal of Hospital Infection., 74 (2010): 152-159.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       72                                 October 2015

-------
Craun, G.F., J.M. Brunkard, IS. Yoder, V.A. Roberts, J. Carpenter, T. Wade, R.L. Calderon,
J.M. Roberts, MJ. Beach, and S.L. Roy. 2010. Causes of outbreaks associated with drinking
water in the United States from 1971 to 2006. Clinical Microbiology Reviews, 23(3): 507-528.

Crespi, S., and J. Ferra. 1998. Chlorination and thermal treatment in the control ofLegionella in
hotels: 5 years follow-up. 13th Meeting of the European Working Group on Legionella
Infections, Abstract 3.

Cristino, S., P.P. Legnani, and E. Leoni. 2012. Plan for the control ofLegionella infections in
long-term care facilities: Role of environmental monitoring. InternationalJournal of Hygiene
and Environmental Health, 215(2012): 279-285.

Cunha, B.A. 2006. The atypical pneumonias: clinical diagnosis and importance. Clinical
Microbiology and Infection, 12(Suppl. 3): 12-24.

Cunliffe, D.A. 1990.  Inactivation of Legionella pneumophila  by monochloramine. Applied
Bacteriology, 68: 453-459.

Daeschlein, G., W.H. Kriiger, C.  Selepko, M. Rochow, G. Dolken, and A. Kramer. 2007.
Hygienic safety of reusable tap water filters (GermLyserฎ) with an operating time of 4 or 8
weeks in a haematological oncology transplant unit. BMC Infectious Diseases, 7:45.

Darelid, J., S. Lofgrenm,  and B.E. Malmvall. 2002. Control of nosocomial Legionnaires' disease
by keeping the circulating hot water temperature above 55  degrees C: experience from a 10-year
surveillance programme in a district general hospital. Hospital Infection, 50(3): 213-219.

de Beer, D., R. Srinivasan, and P.S. Stewart. 1994. Direct measurement of chlorine penetration
into biofilms during disinfection. Applied and Environmental Microbiology.  60(12): 4339-4344.

Deere, D., and A. Davison. 1998. Safe drinking water. Are food guidelines the answer? Water,
Nov/Dec.

Department of Homeland Security. 2015. Chemical Facility Anti-Terrorism Standards (CFATS).
Available online at http://www.dhs.gov/chemical-facility-anti-terrorism-standards.

Department of Veterans Affairs. 2012. Indoor water features, decorative fountains: recommend
non-use. 003C2B-DA-138.

Department of Veterans Affairs. 2014. VHA Directive 1061. Prevention of healthcare-associated
Legionella disease and scald injury from potable water distribution systems. Washington, DC:
Veterans Health Administration.

Dietrich, A.M. 2009. The sense of smell: contributions of orthonasal and retronasal perception
applied to metallic flavor of drinking water. Journal of Water Supply: Research and Technology
-AQUA, 58(8): 562-570.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      73                                 October 2015

-------
Dietrich, A.M. and R. Hoehn, 1991. Taste and odor problems associated with chlorine dioxide.
Denver, Colo.: Water Research Foundation and AWWA.

Doleans A., H. Aurell, M. Reyrolle, G. Lena, J. Freney, F. Vandenesch, J. Etienne, and S.
Jarraud. 2004. Clinical and environmental distributions of Legionella strains in France are
different. J Clin Microbiol, 42(1):458-60.

Domingue, E.L., R.L. Tyndall, W.R. Mayberry, and O.C. Pancorbo. 1988. Effects of Three
Oxidizing Biocides on Legionella pneumophila Serogroup 1. Applied and Environmental
Microbiology, 54(3): 741-747.

Dondero, T.J., H.W. Clegg, T.F. Tsai, R.M. Weeks, E. Duncan, J. Strickler, C. Chapman, G.F.
Mallison, B. Politi, M.E. Potter, and W. Schaffner. 1979. Legionnaires' disease in Kingsport,
Tennessee. Annals of Internal Medicine, 90: 569-573.

Donlan, R., R. Murga, J. Carpenter, E. Brown, R. Besser, and B. Fields. 2002. Monochloramine
disinfection of biofilm-associated Legionella pneumophila in a potable water model system. In:
Legionella, R. Marre, Y.A. Kwaik, and C. Bartlett, (eds.). 406-410. Washington, DC: American
Society for Microbiology.

Donohue, M.J., K.O. O'Connell, S.J. Vesper, J.H. Mistry, D. King, M. Kostich, and S. Pfaller.
2014. Widespread molecular detection of Legionella pneumophila serogroup 1 in cold water taps
across the United States. Environmental Science & Technology, 48: 3145-3152.

Drake, P.L. and KJ. Hazelwood. 2005. Exposure-Related Health Effects of Silver and Silver
Compounds: A Review. Annals of Occupational Hygiene, 49:575-585.

DSMZ. 2014. List of prokaryotic names validly published. Braunschweig, Germany. Available
online at:
http://www.dsmz.de/fileadmin/Bereiche/ChiefEditors/BacterialNomenclature/DSMZ Bactnames
.pdf

Ducret, A., M. Chabalier, and S. Dukan. 2014. Characterization and resuscitation of'non-
culturable' cells of Legionella pneumophila. BMC Microbiology, 14:  3-10.

Duda, S.,S. Kandiah, I.E. Stout, J.L. Baron, M. Yassin, M. Fabrizio, J. Ferrelli, R. Hariri, M.M.
Wagener, J. Goepfert, J. Bond, J. Hannigan, and D. Rogers. 2014. Evaluation of a New
Monochloramine Generation System for Controlling Legionella in Building Hot Water Systems.
Infection Control and Hospital Epidemiology, 35(11):  1356-1363.

Dupuy, M., S. Mazoua, F. Berne, C. Bodet, N. Garrec, P. Herbelin, F. Menard-Szczebara, S.
Oberti, M.H. Rodier, S. Soreau, F. Wallet, and Y. Hechard. 2011. Efficiency of water
disinfectants against Legionella pneumophila and Acanthamoe ba. Water Research, 45: 1087-
1094.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      74                                October 2015

-------
Dyck, A., M. Exner, and A. Kramer. 2007. Experimental based experiences with the introduction
of a water safety plan for a multi-located university clinic and its efficacy according to WHO
recommendations. BMC Public Health, 7(34).

Edelstein, P.H. 2007. Legionella*. In Manual of Clinical Microbiology, 9th ed., P.R. Murray
(ed.). p. 835-849. Washington, DC: ASM Press.

Edelstein, P.H., R.E. Whittaker, R.L. Kreiling, and C.L. Howell. 1982. Efficacy of Ozone in
Eradication of Legionella pneumophila from Hospital Plumbing Fixtures. Applied and
Environmental Microbiology, 44(6): 1330-1334.

Edwards, M. and A. Dudi. 2004. Role of chlorine and chloramines in corrosion of lead-bearing
plumbing materials. JournalAWWA, 96(10): 69-81.

Edwards, M., J.F. Ferguson, and S.H. Reiber. 1994. The pitting corrosion of copper. Journal
AWWA, 86: 74-90.

Edwards, M., S.  Jacobs, and RJ. Taylor. 2000. The blue water phenomenon. Journal AWWA,
92(7):72-82.

Escoll, P., M. Rolando, L. Gomez-Valero and C. Buchrieser. 2013. From amoeba to
macrophages: exploring the molecular mechanisms of Legionella pneumophila infection in both
hosts. Current Topics in Microbiology and Immunology, 376: 1-34.

Euser S.M., J.P.  Bruin, P. Brandsema, L. Reijnen, S.A. Boers, and J.W. Den Boer. 2013.
Legionella prevention in the Netherlands: an evaluation using genotype distribution. Eur J Clin
Microbiol Infect Dis, 32(8): 1017-22.

Fang, G.D., V.L. Yu, and R.M. Vickers. 1989.  Disease due to the Legionellaceae (other than
Legionella pneumophila). Historical, microbiological, clinical and epidemiological review.
Medicine Baltimore, 68(2):  116-132.

Fields, B. S. 1996. The molecular ecology ofLegionellae. Trends in Microbiology, 4(7): 286-
290.

Fields, B.S., R.F. Benson, and R.E. Besser, 2002. Legionella and Legionnaires' disease:  25 years
of investigation.  Clinical Microbiology Reviews, 15: 506-26.

Fields B.S., E.B. Shotts, J.C. Feeley, G.W. Gorman, and W.T. Martin. 1984. Proliferation of
Legionella pneumophila as an intracellular parasite of the ciliated protozoan Tetrahymena
pyriformis. Applied and Environmental Microbiology, 47: 467-471.

Figueras, M.J., and JJ. Borrego. 2010. New perspectives in monitoring drinking water microbial
quality. International Journal of Environmental Research and Public Health, 7: 4179-4202.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      75                                 October 2015

-------
Fitzgeorge, R.B., A. Baskerville, M. Broster, P. Hambleton, and PJ. Dennis. 1983. Aerosol
infection of animals with strains of Legionellapneumophila of different virulence: comparison
with intraperitoneal and intranasal routes of infection. Journal of Hygiene, 90: 81 - 89.

Flannery, B., L.B. Gelling, DJ. Vugia, J.M. Weintraub, JJ. Salerno, MJ. Conroy, V.A. Stevens,
C.E. Rose, M.R. Moore, B.S. Fields, and R.E. Besser. 2006. Reducing Legionella colonization of
water systems with monochloramine. Emerging Infectious Diseases, 12(4): 588-596.

Fliermans, C.B., W.B. Cherry, L.H. Orrison, SJ. Smith, D.L. Tison, andD.H. Pope. 1981.
Ecological distribution of Legionella pneumophila. Applied and Environmental Microbiology,
41: 9-16.

Food and Agriculture Organization (FAO). 1998. Food quality and safety systems - a training
manual on food hygiene and the hazard analysis and critical control point (HACCP) system.
Food and Agriculture Organization of the United Nations Publishing Management Group, FAO
Information Division, Rome. Available online at:
http://www.fao.org/docrepAV8088E/w8088e05.htmtfsection.

Franzin, L., D. Cabodi, and C. Fantino. 2002. Evaluation of the efficacy of ultraviolet irradiation
for disinfection of hospital water contaminated by Legionella. Hospital Infection, 51(4): 269.

Franzin, L, C. Scolfaro, D. Cabodi, M. Valera, and P. A. Tovo. 2001. Legionella pneumophila
pneumonia in a newborn after water birth: a new mode of transmission. Clinical Infectious
Diseases- Brief Report, 33: el 03.

Fraser, D.W., T.R.  Tsai, W. Orenstein, W.E. Parkin, HJ. Beecham, R.G. Sharrar, J. Harris, G.F.
Mallison, S.M. Martin, I.E. McDade, C.C. Shepard, andP.S. Brachman. 1977. Legionnaires'
disease: description of an epidemic of pneumonia. New England Journal of Medicine. 297:
nl!86-1196.

Garcia, M.T., B. Baladron, V. Gil, M.L. Tarancon, A. Vilasau, A. Ibafiez, C. Elola, and C. Pelaz.
2008. Persistence of chlorine-sensitive Legionella pneumophila in hyperchlorinated installations.
Applied Microbiology,  105(3): 837-847.

Gates, D., G. Ziglio, and K. Ozekin. 2009. State of the Science of Chlorine Dioxide in Drinking
Water. Water Research Foundation and Fondazione AMGA.

Gillilland, C., A.A. Rosenblatt, and W.F. McCoy. 2014. HACCP for building water systems.
Water Conditioning and Purification International. Jan.

Gilpin, R. W., S.B. Dillon, P. Keyser, A. Androkites, M. Berube, N. Carpendale, J. Skorina, J.
Hurley, and A.M. Kaplan.  1985. Disinfection of circulating water systems by ultraviolet light
and halogenation. Water Research, 19(7): 839-848.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      76                                 October 2015

-------
Glick, T.H., M.B. Gregg, B. Berman, G. Mallison, W.W. Rhodes Jr., and I. Kassanoff. 1978.
Pontiac fever. An epidemic of unknown etiology in a health department: I. Clinical and
epidemiological aspects. American Journal of'Epidemiology, 107(2): 149-160.

Graman, P.S., G.A. Quinlan, and J.A. Rank. 1997. Nosocomial legionellosis traced to a
contaminated ice machine. Infection Control and Hospital Epidemiology. 18(9): 637-640.

Gray, R., and M. Morain. 2000. HACCP application to Brisbane water. Water, Jan/Feb.

Great Lakes - Upper Mississippi River Board of State and Provincial Public Health and
Environmental Managers (GLUMRBSPPHEM). 2012. Recommended Standards for Water Works.
Health Education Service, Albany, NY.

Grosserode, M., R. Wenzel, M. Pfaller, and C. Helms.  1993. Continuous hyperchlorination for
control of nosocomial Legionellapneumophila pneumonia:  a 10-year follow-up of efficacy,
environmental effects, and costs. In: Legionella:  current status and emerging perspectives, J.M.
Barbaree, R.F. Breiman, A.P. Dufour, (eds.). 226-229. Washington, DC: American Society of
Microbiology.

Gunnarsdottir, M.J., and L.R. Gissurarson. 2008. HACCP and water safety plans in Icelandic
water supply: Preliminary evaluation of experience. Water and Health, 6(3): 377-382.

Hall, K.K., E.T. Gianetta, S. I. Getchell-White, L. J. Durbin, and B. M. Farr. 2003. Ultraviolet
light disinfection of hospital water for preventing nosocomial Legionella infection: a 13-year
follow-up. Infection Control and Hospital Epidemiology, 24(8): 580-583.

Harrison T.G., B. Afshar, N. Doshi, N.K. Fry,  and J.V. Lee. 2009. Distribution of Legionella
pneumophila serogroups, monoclonal antibody subgroups and DNA sequence types in recent
clinical  and environmental isolates from England and Wales (2000-2008). Eur J Clin Microbiol
Infect Dis, 28(7): 781-91.

Haupt, T.E., R.T. Heffernan, JJ. Kazmierczak, H. Nehls-Lowe, B. Rheineck, C. Powell, K.K.
Leonhardt, A.S. Chitnis, and J.P Davis. 2012. An outbreak of Legionnaires disease associated
with a decorative water wall fountain in a hospital. Infection Control and Hospital Epidemiology,
33(2): 185-191.

Havelaar, A.H. 1994. Application of HACCP to drinking water supply. Food Control,  5(3): 145-
152.

Health and Safety Executive (HSE). 2009. National Guidelines for the Control of Legionellosis
in Ireland.

Health and Safety Executive (HSE). 2014. Legionnaires' disease part 2: The control of Legionella
bacteria in hot and cold water systems. HSG274 Part 2.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      77                                October 2015

-------
Health Canada. 2013. Guidance on waterborne bacterial pathogens. Water, Air and Climate
Change Bureau, Health Environments and Consumer Safety Branch, Health Canada, Ottawa,
Ontario (Catalogue No. H129-25/1-2014E-PDF).

Health Protection Surveillance Centre. 2009. National Guidelines for the Control of
Legionellosis in Ireland, 2009. Report of Legionnaires' Disease Subcommittee of the Scientific
Advisory Committee. Dublin, Ireland: Health Protection Surveillance Centre.

Heffelfinger, J.D., J.L. Kool, S. Fridkin, VJ. Fraser, J. Hageman, J. Carpenter, and C.G.
Whitney. 2003. Risk of hospital-acquired Legionnaires' disease in cities using monochloramine
versus other water disinfectants. Infection Control and Hospital Epidemiology, 24(8): 569-574.

Heimberger, T., G. Birkhead, D. Bornstein, K. Same, and D. Morse. 1991. Control of
nosocomial Legionnaires' disease through hot water flushing and supplemental chlorination of
potable water. Infectious Diseases, 163(2): 413.

Heng, B.H., K.T. Goh, and L.K. Ng. 1995. Surveillance and control ofLegionella bacteria in the
built environment. In Health and the Built Environment.  Singapore: Ministry of the
Environment.

Hoffman, C., C.F. Harrison, and H. Hilbi. 2014. The natural alternative: protozoa as  cellular
models for Legionella infection. Cellular Microbiology, 16(1): 15-26.

Hong, J-H; S.E. Duncan, A.M. Dietrich. 2010. Effect of copper speciation at different pH on
temporal sensory attributes of copper. Food Quality and Preference, 21(1): 132-139.

Huang, H.I., H.Y. Shih, C.M. Lee, T.C. Yang, J.J. Lay, and Y.E. Lin. 2008. In vitro efficacy of
copper and silver ions in eradication Pseudomonas aeruginosa, Stenotrophomonas maltophilia
and Acinetobacter baumannii: implications for on-site disinfection for hospital infection control.
Water Research, 42(1-2): 73-80.

Hung, L., D.C.  Copperthite, C.S. Yang, F.A. Lewis, andF.A. Zampiello. 1993. Environmental
Legionella assessment in office buildings of continental United States. Indoor Air, 3: 349-353.

Hwang, M.G., H. Katayama, and S. Ohgaki. 2007. Inactivation of Legionella pneumophila and
Pseudomonas aeruginosa: evaluation of the bactericidal ability of silver cations.  Water
Research,  41:4097-4104.

Jacangelo, J.G., N.L. Patania, R.R. Trussel, C.N. Haas, and C. Gerba. 2002. Inactivation of
Waterborne Emerging Pathogens by Selected Disinfectants. Denver, Colo.: AWWA Research
Foundation and AWWA.

Jakubek, D., C. Guillaume, M. Binet, G. Leblon, M. Dubow, and M. Le Brun. 2013.
Susceptibility ofLegionella strains to the chlorinated biocide, monochloramine. Microbes and
Environments, 28(3): 336-345.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      78                                 October 2015

-------
Kana, B.D., E.G. Gordhan, K. J. Downing, N. Sung, G.Vostroktunova, E. E. Machowski,
L.Tsenova, M. Young, A. Kaprelyants, G. Kaplan, and V. Mizrahi. 2008. The resuscitation-
promoting factors of Mycobacterium tuberculosis are required for virulence and resuscitation
from dormancy but are collectively dispensable for growth in vitro. Molecular Microbiology,
67(3): 672-684.

Katz, S.M., and J.M. Hammel. 1987. The effect of drying, heat and pH on the survival of
Legionella pneumophila. Annals of Clinical and Laboratory Science, 17(3): 150-156.

Kaufmann A.F., I.E. McDade, C.M. Patton, J.V. Bennett, P. Skaliy, J.C. Feeley, D.C. Anderson,
M.E. Potter, V.F. Newhouse, M.B. Gregg, and P.S. Brachman. 1981. Pontiac fever: isolation of
the etiologic agent (Legionellapneumophila) and demonstration of its mode of transmission.
American Journal of Epidemiologyl 14(3): 337-47.

Kilvington, S. and J. Price, 1990. Survival of Legionella pneumophila within  cysts of
Acanthamoeba polyphaga following chlorine exposure. Applied Bacteriology, 68: 519-525.

Kim, B.R.,  I.E. Anderson, S.A. Mueller, W.A. Gaines, and A.M.  Kendall, 2002. Literature
review - efficacy of various disinfectants against Legionella in water systems. Water Research,
36(18): 4433-4444.

Kirk-Othmer. 1979. Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed.

Kirmeyer, G., K. Martel, G. Thompson, L. Radder, W. Klement, M. LeChevallier, H. Baribeau,
and A. Flores. 2004. Optimizing Chloramine Treatment. Denver,  Colo. Water Research
Foundation and AWWA.

Knobeloch  L., C. Schubert, J. Hayes, J. Clark, C. Fitzgerald, and  A. Fraundorff. 1998.
Gastrointestinal Upsets and New Copper Plumbing - Is There a Connection? Wisconsin Medical
Journal, 49-53.

Knobeloch, L., M. Ziarnik, J. Howard, B. Theis, D. Farmer, H. Anderson, and M. Proctor. 1994.
Gastrointestinal upsets associated with ingestion of copper-contaminated water. Environmental
Health Perspectives, 102(11): 958-961.

Knudson, G.B. 1985. Photoreactivation of UV-inadiated Legionella pneumophila and other
Legionella species. Applied and Environmental Microbiology, 49(4): 975-980.

Kool, J.L., J.C. Carpenter, and B.S. Fields. 1999.  Effect of monochloramine disinfection of
municipal drinking water on risk of nosocomial Legionnaires' disease. The Lancet, 353: 272-
277.

Kool, J.L., J.C. Carpenter, and B.S. Fields. 2000.  Monochloramine and Legionnaires' disease.
Journal AWWA, 92(9): 88-96
Draft - Technologies for Legionella Control:
Scientific Literature Review                      79                                 October 2015

-------
KozakN.A., R.F. Benson, E. Brown, N.T. Alexander, T.H. Taylor Jr., E.G. Shelton, and B.S.
Fields. 2009. Distribution of lag-1 alleles and sequence-based types among Legionella
pneumophila serogroup 1 clinical and environmental isolates in the United States. JClin
Microbiol, 47(8): 2525-35.

 Kozak, N.A., C.E. Lucas, and J.M. Winchell. 2013. Identification of Legionella in the
environment. In: Legionella: Methods and Protocols, Methods in Molecular Biology Vol. 954,
Carmen Buchrieser and Hubert Hilbi (eds). p 3-25. New York: Springer + Business Media.

Kozak-Muiznieks N.A., Lucas C.E., Brown E., T. Pondo, T.H. Taylor Jr., M. Frace, D.
Miskowski, and J. M. Winchell. 2014. Prevalence of sequence types among clinical and
environmental isolates of Legionella pneumophila serogroup 1 in the United States  from 1982 to
2012. JClin Microbiol, 52(1): 201-211.

Krageschmidt, D.A., A.F. Kubly, M.S. Browning, A.J. Wright, J.D. Lonneman, M.J. Detmer,
and W.F. McCoy. 2014. A comprehensive water management program for multicampus
healthcare facilities. Infection Control and Hospital Epidemiology, 35(5): 556-563.

Kuchta, J.M., S.J. States, A.M. McNamara, R.M. Wadowsky, andR.B. Yee. 1983. Susceptibility
of Legionella pneumophila to Chlorine in Tap Water. Applied and Environmental Microbiology,
46(5): 1134-1139.

Kusnetsov, J., E. livanainen, N. Elomaa, O. Zacheus, and P.J. Martikainen. 2001. Copper and
silver ions more effective against Legionellae than against mycobacteria in a hospital warm
water system. Water Research, 35(17): 4217-25.

Kwaik, Y.A., L.Y.  Gao, B.J. Stone, C. Venkataraman, and O.S. Harb.  1998. Invasion of protozoa
by Legionella pneumophila  and its role in bacterial ecology and pathogenesis. Applied and
Environmental Microbiology. 64(9): 3127-3133.

Landeen, L.K., M.T. Yahya, and C.P. Gerba. 1989. Efficacy of copper and silver ions and
reduced levels of free chlorine in inactivation of Legionella pneumophila. Applied and
Environmental Microbiology, 55(12): 3045-50.

LeChevallier, M.W., C.D. Cawthon, and R.G. Lee. 1988a. Inactivation of biofilm bacteria.
Applied and Environmental Microbiology, 54(10): 2492-2499.

LeChevallier, M.W., C.D. Cawthon, and R.G. Lee. 1988b. Factors promoting  survival of bacteria
in chlorinated water supplies. Applied and Environmental Microbiology, 54(3): 649-654.

Lee, W.H., D.G. Wahman, P.L. Bishop, and J.G. Pressman. 2011. Free chlorine and
monochloramine application to nitrifying biofilm: Comparison of biofilm penetration, activity
and viability. Environmental Science and Technology, 45, 1412-1419.

Lee, W., P. Westerhoff, X. Yang, and C. Shang. 2007. Comparison of colorimetric and
membrane introduction mass spectrometry techniques for chloramine analysis. Water Res.,
41(14): 3097-3102.

Draft - Technologies for Legionella Control:
Scientific Literature Review                      80                                October 2015

-------
Lin, Y.E., I.E. Stout, and V.L. Yu. 201 la. Prevention of hospital-acquired legionellosis. Current
Opinion in Infectious Diseases, 24: 350-356.

Lin, Y.E., I.E. Stout, and V.L. Yu. 201 Ib. Controlling Legionella in hospital drinking water: An
evidence-based review of disinfection methods. Infection Control and Hospital Epidemiology,
32(2): 166-173.

Lin, Y.E., I.E. Stout, V.L. Yu, R.D. Vidic, and SJ. States. 2000b. Legionnaires disease in an
apartment building: disinfection methods and recommendations. Abstract for Health-related
Water Microbiology Symposium. Submitted for oral presentation at 1st International Water
Congress, International Water Association. Paris, France. July 3-7, 2000.

Lin, Y.E., R.D. Vidic, I.E. Stout, C.A. McCartney, and V.L. Yu. 1998b. Inactivation of
Mycobacterium avium by copper and silver ions. Water Research. 32(7): 1997-2000.

Lin, Y.E., R.D. Vidic, I.E. Stout, and V.L. Yu. 1996. Individual and combined effects of copper
and silver ions on inactivation of Legionella pneumophila. Water Research, 30(8): 1905-13.

Lin, Y.E., R.D. Vidic, I.E. Stout, and V.L. Yu. 1998a. Legionella in water distribution systems.
JournalAWWA, 90(9), 112-122.

Lin, Y.E., R.D. Vidic, J. E.  Stout, and V.L. Yu. 2002. Negative effect of high pH on biocidal
efficacy of copper and silver ions in controlling Legionella pneumophila. Applied and
Environmental Microbiology, 68(6): 2711-15.

Lin, Y.E., V.L. Yu, R.D. Vidic, and SJ. States. 2000a. Discussion of "monochloramine and
Legionnaires' disease." American Water Works Association. 92(10): 88-90.

Liu, Z., I.E.  Stout, M. Boldin, J. Rugh, W. Diven, and V.L. Yu. 1998. Intermittent use of copper-
silver ionization for Legionella control in water distribution systems: a potential option in
buildings housing individuals at low risk of infection. Clinical Infectious Diseases, 26: 138-140.

Liu, Z., I.E.  Stout, L. Tedesco, M. Boldin, C. Hwang, W.F. Diven, and V.L. Yu. 1994.
Controlled evaluation of copper-silver ionization in eradicating Legionella pneumophila from a
hospital water distribution system. Infectious Diseases, 169: 919-922.

Liu, Z., I.E.  Stout, L. Tedesco, M. Boldin, C. Hwang, and V.L. Yu. 1995. Efficacy of UV light in
preventing Legionella colonization of a hospital water distribution system. Water Research,
29(10): 2275-2280.

Loret, J.F., S. Robert, V. Thomas, A.J. Cooper, W.F. McCoy, and Y. Levi. 2005. Comparison of
disinfectants for biofilm, protozoa and Legionella control. IWA Journal of Water and Health.,
3(4):  423-433.

LPSN. 2014. (List of prokaryotic names with standing in nomenclature ().) Available online at:
http://www.bacterio.net/Legionella.htmL.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       81                                  October 2015

-------
Lucas, C.E., T.H. Taylor Jr., and B.S. Fields. 2011. Accuracy and precision oi Legionella
isolation by US laboratories in the ELITE program pilot study. Water Research, 45: 4428-4436.

Lytle, D.A., and M.R. Schock. 2007. Pitting Corrosion of Copper in Waters with High pH and
Low Alkalinity. JAWWA, 100(3):115-129.

Marches!, I, S. Cencetti, P. Marchegiano, G. Frezza, P. Borella, and A. Bargellini. 2012. Control
ofLegionella contamination in a hospital water distribution system by monochloramine.
American Journal of Infection Control, 40: 279-281.

Marches!, I, G. Ferranti, A. Bargellini, P. Marchegiano, G. Predieri, I.E. Stout, and P. Borella.
2013. Monochloramine and chlorine dioxide for controlling Legionellapneumophila
contamination: biocide levels and disinfection byproduct formation in hospital networks. IWA
Journal of Water and Health, 11(4): 738-747.

Marches!, I, P. Marchegiano, A. Bargellini, S. Cencetti, G. Frezza, M. Miselli, and P. Borella.
2011. Effectiveness of different methods to control Legionella in the water supply: ten-year
experience in an Italian university hospital. Hospital Infection. 77(1): 47-51.

Marston, B.J., H.B. Lipman, and R.F. Breiman. 1994. Surveillance for Legionnaires' disease.
Risk factors for morbidity and mortality. Archives of Internal Medicine,  154(21): 2417-2422.

Marston, B.J., J.F. Plouffe, T.M. File Jr., B.A. Hackman, S.J. Salstrom, H.B. Lipman, M.S.
Kolczak, and R.F. Breiman.  1997. Incidence of community-acquired pneumonia requiring
hospitalization: Results of a population-based active surveillance study in Ohio. Archives of
Internal Medicine, 157(15): 1709-1718.

Martel, K., G. Kirmeyer, A. Hanson, M. Stevens, J. Mullenger, and D. Deere. 2006. Application
ofHACCPfor distribution system protection. Denver, Colo.: Water Research Foundation.

McDade, I.E., DJ. Brenner,  and F.M. Bozeman. 1979. Legionnaires' disease bacterium isolated
in 1947. Annals ofInternal Medicine, 90: 659-661.

Miami-Dade County Health Department (MDCHD). 2010. Outbreak of Legionnaire's disease:
Miami-Dade County Health Department Final Report. Miami, FL.

Mietzner, S., R.C. Schwille, A. Farley, E.R. Wald, J.H. Ge, S.J.  States, T. Libert, R.M.
Wadowsky. 1997. Efficacy of thermal treatment and copper-silver ionization for controlling
Legionella pneumophila in high-volume hot water plumbing systems in hospitals. American
Journal of Infection Control, 25(6): 452-457.

Modol, J., M. Sabria, E.  Reynaga, M.L. Pedro-Botet, N. Sopena, P. Tudela, I. Casas, and C. Rey-
Joly. 2007. Hospital-acquired Legionnaires disease in a university hospital: impact of the copper-
silver ionization system. Clinical Infectious Diseases,  44: 263-265.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      82                                 October 2015

-------
Molloy, S.L., R. Ives, A. Hoyt, R. Taylor, and J.B. Rose. 2008. The use of copper and silver in
carbon point-of-use filters for the suppression ofLegionella throughput in domestic water
systems. Applied Microbiology, 104: 998-1007.

Molmeret, M., D.M. Bitar, L. Han, and Y.A. Kwaik. 2004. Cell biology of the intracellular
infection by Legionellapneumophila. Microbes and Infection. 6(1): 129-139.

Moore, M.R., M. Pryor, B. Fields, C. Lucas, M. Phelan, and R.E. Besser. 2006. Introduction of
monochloramine into a municipal water system: Impact on colonization of buildings by
Legionella spp. Applied and Environmental Microbiology, 72: 378-383.

Mortimore, S.E. and C.A. Wallace. 2QQ1.HACCP, Blackwell Science, UK.

Muder, R.R., V.L. Yu, and A.H. Woo. 1986. Mode of transmission of Legionellapneumophila: a
critical review. Archives of Internal Medicine, 146: 1607-1612.

Mullenger, J., G. Ryan, and J. Hearn. 2002. A water authority's experience with HACCP. Water
Science and Technology, 2(5-6): 149-155.

Muraca, P., I.E. Stout, and V.L.Yu.  1987. Comparative assessment of chorine, heat, ozone and
UV light for killing Legionellapneumophila within a model plumbing system. Applied and
Environmental Microbiology,  53(2): 447-453.

Nagai, T., H. Sobajima, M. Iwasa, T. Tsuzuki, F. Kura, J. Amemura-Maekawa, and H.
Watanabe. 2003. Neonatal sudden death due to Legionella pneumonia associated with water
birth in a domestic spa bath. Clinical Microbiology, 41(5): 2227-2229.

National Environmental Methods Index. Available online at: http://www.nemi.gov/home/.
Accessed October 1, 2014.

Newton, H.J., D.K. Ang, I.R. van Driel, and E.L. Hartland. 2010. Molecular pathogenesis of
infections caused by Legionellapneumophila. Clinical Microbiology Reviews, 23(2):  274-298.

Niedeveld, C.J., F.M. Pet, and P.L. Meenhorst. 1986. Effect of rubbers and their constituents on
proliferation of Legionellapneumophila in naturally contaminated hot water. Lancet.,  II: 180-
184.

Nies, D. H. 1999. Microbial heavy-metal resistance. Applied Microbiology Biotechnology,
51:730-50.

O'Brien, SJ. and R.S. Bhopal. 1993. Legionnaires' disease: the infective dose paradox. Lancet,
342:5.

Occupational Safety and Health Administration (OSHA). 1999. OSHA Technical Manual. TED
01-00-015.  Section III, Chapter 7. Available online at:
http://www.osha.gov/dts/osta/otm/otm_iii/otm_iii_7.htmL.


Draft - Technologies for Legionella Control:
Scientific Literature Review                      83                                 October 2015

-------
Oguma, K., H. Katayama and S. Ohgaki. 2004. Photoreactivation of Legionellapneumophila
after inactivation by low- or medium-pressure ultraviolet lamp. Water Research, 38(11): 2757-
2763

Ohno, A., N. Kato, R. Sakamoto, S. Kimura, and K. Yamaguchi. 2008. Temperature-dependent
parasitic relationship between Legionella pneumophila and a free-living amoeba (Acanthamoeba
castellanii). Applied and Environmental Microbiology, 74(14): 4585-4588.

Oliver, J.D. 2010. Recent findings on the viable but nonculturable state in pathogenic bacteria.
FEMSMicrobiology Reviews, 34: 415-425.

Orsi, G.B., M. Vitali, L. Marinelli, V. Ciorba, D. Tufi, A. Del Cimmuto, P. Ursillo, M. Fabiani,
S. DeSantis, C. Protano, C. Marzuillo, and M. DeGiusti. 2014. Legionella control in the water
system of antiquated hospital buildings by shock and continuous hyperchlorination: 5 years'
experience. BMC Infectious Diseases, 14: 394. Available online at:
http://www.biomedcentral.com/content/pdf/1471-2334-14-394.pdf

Ortolano, G.A., M.B. McAlister, J.A. Angelbeck, J. Schaffer, R.L. Russell, E. Maynard, and B.
Wenz. 2005. Hospital water point-of-use filtration: A complementary strategy to reduce the risk
of nosocomial infection. American Journal of Infection Control, 33(5): S1-S19.

Pearce, M.M., N. Theodoropoulos, MJ. Mandel, E. Brown, K.D.  Reed, and N.P. Cianciotto.
2012. Legionella cardiaca sp. nov., isolated from a case of native valve endocarditis in a human
heart. InternationalJournal ofSystematic and Evolutionary Microbiology, 62:  2946-2954.

Pedro-Botet, M.L., I. Sanchez, M. Sabria, N. Sopena, L. Mateu, M. Garcia-Nunez, and C. Rey-
Joly. 2007. Impact of copper and silver ionization on fungal colonization of the water supply in
health care centers: implications for immunocompromised patients. Clinical Infectious Diseases.
45:  84-86.

Pressman, J.G., W.H. Lee, P.L. Bishop, and D.G. Wahman. 2012. Effect of free ammonia
concentration on monochloramine penetration within a nitrifying biofilm and its effect on
activity, viability and recovery, Water Research, 46(3), 882-894.

Pruden, A., M.A. Edwards, and J.O. Falkinham, III. 2013. State of the science and research
needs for opportunistic pathogens in premise plumbing. Water Research Foundation.

Pryor, M., S. Springthorpe, S. Riffard, T. Brooks, Y. Huo, G. Davis, and S.A. Sattar. 2004.
Investigation of opportunistic pathogens in municipal drinking water under different supply and
treatment regimes. Water Science and Technology, 50(1): 83-90.

Public Health England. 2014. Public Health England advice on home birthing pools.  Available
online at: http://www.gov.uk/government/news/public-health-england-advice-on-home-birthing-
pools.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      84                                 October 2015

-------
Reingold, A.L., B.M. Thomason, BJ. Brake, L. Thacker, H.W. Wilkinson, and IN. Kuritsky.
1984. Legionella pneumonia in the United States: the distribution of serogroups and species
causing human illness. Infectious Diseases, 149(5): 819.

Rhoads, W.J., A. Pruden, and M.A. Edwards. 2014. Anticipating challenges with in-building
disinfection for control of opportunistic pathogens. Water Environment Research. 86(6): 540-
549.

Richards, A.M., I.E. Von Dwingelo, C.T. Price and Y.A. Kwaik. 2013. Cellular microbiology
and molecular ecology of Legionella-amooba interaction. Virulence, 4(4): 307-314.

Roche, P. and D. Benanou, 2007. Impact of chlorination on the formation of odor compounds
and their precursors in treatment of drinking water. Available online at:
http://www.techneau.Org/fileadmin/files/Publications/Publications/Deliverables/D5.3.8.pdf.

Rohr, U., M. Senger, F. Selenka, R. Turley, and M. Wilhelm. 1999. Four years of experience
with silver-copper ionization for control of Legionella in a German university hospital hot water
plumbing system. Clinical Infectious Diseases, 29: 1507-11.

Rohr, U., S. Weber, F. Selenka, and M. Wilhelm. 2000. Impact of silver and copper on the
survival of amoebae and ciliated protozoa in vitro. InternationalJournal of Hygiene and
Environmental Health, 203: 87-89.

Rosenblatt, A.A. and W.F. McCoy. 2014. HACCPfor Building Water Systems. Participants'
Handbook. Version 1.0. NSF International.

Rowbotham, TJ. 1986. Current views on the relationships between amoebae, Legionellae and
man. IsraelJournal of Medical Sciences. 22(9): 678-689.

Ruiz, B., J. Bauza,  J. Benito, and A. Pascual. 2007. Use of ozone for Legionella reduction in
water systems. IOA Conference and Exhibition, Spain. October 29-31, 2007.

Rusin, P. A., J.B. Rose, C.N. Haas,  and C.P. Gerba. 1997. Risk assessment of opportunistic
bacterial  pathogens in drinking water. Reviews of Environmental Contamination and Toxicology.
152: 57-83.

Saby, S., A. Vidal and H. Suty. 2005. Resistance of Legionella to disinfection in hot water
systems.  Water Science and Technology. 52(8): 15-26.

Sarver, E., K. Dodson, R.P.  Scardina, R. Lattyak-Slabaugh, M. Edwards, and C. Nguyen. 2011.
Copper pitting in chlorinated, high-pH potable water. JournalAWWA,  103(3):  86-97.

Schaechter, M., N.C. Engelberg, B.I.  Eisenstein, and G. Medoff. 1998. Mechanisms of microbial
disease. Third Edition. Williams and Wilkins. Baltimore, MD.

Schoen, M.E.,  and NJ. Ashbolt. 2011. An in-premise model for Legionella exposure during
showering events. Water Research. 45: 5826 - 5836.

Draft - Technologies for Legionella Control:
Scientific Literature Review                       85                                 October 2015

-------
Sehulster, L., and R.Y.W. Chinn. 2003. Guidelines for Environmental Infection Control in
Health-Care Facilities, 2003:  Recommendations of CDC and the Healthcare Infection Control
Practices Advisory Committee. Morbidity and Mortality Weekly Report. 52:1-44. Available
online at: http://www.cdc.gov/hicpac/pdf/guidelines/eic_in_HCF_03.pdf

Seidel, C.J., MJ. McGuire, R.S. Summers, and S. Via. 2005. Have utilities switched to
chloramines? JournalAWWA. 97(10): 87-97.

Shah, A. and W.A. Mitch. 2012. Halonitroalkanes, Halonitriles, Haloamides and N-nitrosamines;
A Critical Review of Nitrogenous Disinfection Byproduct Formation Pathways. Environmental
Science and Technology, 46:1:119.

Sheffer, P.J., I.E. Stout, M.M. Wagener, and R.R. Muder.  2005. Efficacy of new point-of-use
water filter for preventing exposure to Legionella and waterborne bacteria. Infection Control, 33:
S20-S25.

Sidari, F.P., I.E. Stout, S. Duda, D. Grubb, and A. Neuner. 2014. Maintaining Legionella control
in building water systems. Journal AWWA, 106(10): 24-32.

Sidari, F.P., I.E. Stout, J.M. Vanbriesen, A.M. Bowman, D. Grubb, A. Neuner, M.M. Wagener,
and V.L. Yu. 2004. Keeping Legionella out of water systems. Journal AWWA, 96(1): 111-119.

Silk, B.J., M.R. Moore, M. Bergtholdt, RJ. Gorwitz, N.A. Kozak, M.M. Tha, E.W. Brown, J.L.
Winchester, BJ. Labus, P. Rowley, J.P. Middaugh, B.S. Fields, and L.A. Hicks. 2012. Eight
years of Legionnaires' disease transmission in travellers to a condominium complex in Las
Vegas, Nevada. Epidemiology and Infection, 140(11): 1993-2002.

Snyder, M.B., M. Siwicki, J.  Wireman, D. Pohlod, M. Grimes, S. Bowman-Riney, and L.D.
Saravolatz. 1990. Reduction in Legionellapneumophila through heat flushing followed by
continuous supplemental chlorination of hospital hot water. Infectious Diseases,  162: 127-132.

Spadaro, J.A., TJ. Berger, S.D. Barranco, S.E. Chapin, and R.O. Becker. 1974. Antibacterial
effects of silver electrodes with weak direct current. Antimicrobial Agents and Chemotherapy,
6(5): 637-642.

States, S., J. Kuchta, W. Young, L. Conley, J. Ge, M. Costeloa, J. Dowling, and R. Wadowsky.
1998. Controlling Legionella using copper-silver ionization. American Water Works Association.
90(9): 122-129.

States, S.J., R.M. Wadowsky, J.M. Kuchta, R.S. Wolford,  L.F. Conley,  andR.B. Yee. 1990.
Legionella in drinking water. In: Drinking Water Microbiology. G.A. McFeters, (ed.). p. 340-
367. New York, NY: Springer-Verlag.

Steinert, M., L. Emody, R. Amann and J. Hacker. 1997. Resuscitation of viable but
nonculturable Legionella pneumophila Philadelphia JR32 by Acanthamoeba castellanii. Applied
and Environmental Microbiology, 63(5): 2047-2053.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      86                                October 2015

-------
Storey, M.V., NJ. Ashbolt, and T.A. Stenstrom. 2004b. Biofilms, thermophilic amoeba and
Legionellapneumophila - a quantitative risk assessment for distributed water. Water Science
and Technology. 50(1): 77-82.

Storey, M.V., J. Winiecka-Krusnell, NJ. Ashbolt, and T.A. Stenstrom. 2004a. The efficacy of
heat and chlorine treatment against thermotolerant Acanthamoebae and Legionellae.
Scandinavian Journal of Infectious Diseases. 36(9): 656-662.

Stout, I.E. and V.L. Yu. 1997. Legionellosis. New England Journal of Medicine, 337: 682-687.

Stout, I.E. and V.L. Yu. 2003. Experiences of the first 16 hospitals using copper-silver ionization
for Legionella control: implications for the evaluation of other disinfection modalities. Infection
Control and Hospital Epidemiology. 24: 563-568.

Stout, I.E., V.L. Yu, and M.G. Best. 1985. Ecology of Legionella pneumophila within water
distribution systems. Applied and Environmental Microbiology. 49(1): 221-228.

Stout, I.E., V.L. Yu, P. Muraca, J. Joly, N. Troup, and L.S. Tompkins. 1992. Potable water as a
cause of sporadic cases of community-acquired Legionnaires' disease. New England Journal of
Medicine. 326: 151-155.

Symons, J. M. 1978. Ozone, chlorine dioxide and chloramines as alternatives to chlorine for
disinfection of drinking water. U.S. Environmental Protection Agency, Cincinnati, Ohio.

Symons, J.M., J.K. Carswell, R.M. Clark, P. Dorsey, E.E.  Geldreich, W.P. Heffernan, J.C. Hoff,
O.T. Love, LJ. McCabe, and A.A. Stevens. 1977. Ozone,  Chlorine Dioxide and Chloramines as
Alternatives to Chlorine for Disinfection of Drinking  Water. State-of-the-Art. Water Supply
Research, Office of Research and Development. Cincinnati, Ohio. November 1977.

Taylor, M., K. Ross, and R. Bentham. 2009. Legionella, protozoa, and biofilms: interactions
within complex microbial systems. MicrobialEcology. 58: 538-547.

Temmerman, R., H. Vervaeren, B. Noseda, N. Boon,  and W. Verstraete. 2006. Necrotrophic
growth of Legionella pneumophila. Applied and Environmental Microbiology, 72(6), 4323-4328.

Thomas, V., T. Bouchez, V. Nicolas, S. Robert, J.F. Loret and Y. Levi. 2004. Amoebae in
domestic water systems: resistance to disinfection treatments and implication in Legionella
persistence. Applied Microbiology, 97: 950-963.

Thompson, C., J. Drago, B. Loeb, and G. Hunter. 2013. Forty Years of Ozone Experience
Treating Municipal Water Supplies in the United States. Proc. 2013 Joint IOA/IUVA World
Congress.

Tobin, J.O., C.L. Bartlett, S.A. Waitkins, G.I. Barrow, A.D. Macrae, A.G. Taylor, R.J. Fallen,
and F.R. Lynch.  198la. Legionnaires' disease: further evidence to implicate water storage and
distribution systems as sources. British Medical Journal, 282: 573.


Draft - Technologies for Legionella Control:
Scientific Literature Review                      87                                October 2015

-------
Tobin, J.O., R.A. Swann, and C.L. Bartlett. 1981b. Isolation of Legionellapneumophila from
water systems: methods and preliminary results. British MedicalJournal, 282: 515-517.

Travis, T.C., E.W. Brown, L.F. Peruski, D. Siludjai, P. Jorakate, P. Salika, G. Yang, N.A.
Kozak,, M. Kodani, A.K. Warner, C.E. Lucas, K.A. Thurman, J.M. Winchell, S. Thamthitiwat,
and B.S. Fields. 2012. Survey of Legionella species found in Thai soil. InternationalJournal of
Microbiology, 2012: 1-4.

Tiiretgen, I. 2004. Comparison of the efficacy of free residual chlorine and monochloramine
against biofilms in model and full scale cooling towers. Biofouling. 20(2): 81-85.

United States Environmental Protection Agency (USEPA). 1989a. Federal Register Notice
National Primary Drinking Water Regulations: Filtration, Disinfection, Turbidity, Giardia
lamblia, Viruses, Legionella and Heterotrophic Bacteria; Final Rule. 54 FR 27486. (June 29,
1989). Available online at: http://water.epa.gov/lawsregs/rulesregs/sdwa/swtr/upload/SWTR.pdf.

United States Environmental Protection Agency (USEPA). 1989b. Integrated Risk Information
System: Silver (CASRN 7440-22-4). Available online at:
http ://www. epa.gov/iris/sub st/0099. htm.

United States Environmental Protection Agency (USEPA). 1994. Drinking water criteria
document for chloramines. ECAO-CIN-D002.

United States Environmental Protection Agency (USEPA). 1996. Water Supply Guidance-90:
State Alternative Technology Approval Protocol. Available online at:
http://water.epa.gov/lawsregs/guidance/sdwa/upload/wsg 90.pdf

United States Environmental Protection Agency (USEPA). 1998. Federal Register Notice
National Primary Drinking Water Regulations; Disinfectants and Disinfection Byproducts; Final
Rule. 63 FR 69390. (December 16, 1998). Available online at:
http://www.gpo.gov/fdsvs/pkg/FR-1998-12-16/pdf/98-32887.pdf.

United States Environmental Protection Agency (USEPA). 1999a. Legionella: Human health
criteria document. EPA-822-R-99-001.

United States Environmental Protection Agency (USEPA). 1999b. 25  years of the Safe Drinking
Water Act:  history and trends. EPA 816-R-99-007.

United States Environmental Protection Agency (USEPA). 1999c. Alternative disinfectants and
oxidants guidance manual. EPA 815-R-99-014.

United States Environmental Protection Agency (USEPA). 1999d. Microbial and Disinfection
Byproducts Rules simultaneous compliance guidance manual. EPA 815-R-99-015.

United States Environmental Protection Agency (USEPA). 2001. Low pressure membrane
filtration for pathogen removal: application, implementation and regulatory issues. EPA 815-C-
01-001.

Draft - Technologies for Legionella Control:
Scientific Literature Review                      88                                 October 2015

-------
United States Environmental Protection Agency (USEPA). 2005a. Economic analysis for the
Final Stage 2 Disinfectants and Disinfection Byproducts Rule. EPA 815-R-05-010.

United States Environmental Protection Agency (USEPA). 2005b. Membrane filtration guidance
manual. EPA 815-R-06-009.

United States Environmental Protection Agency (USEPA). 2006a. Hazard Analysis Critical
Control Point (HACCP) strategies for distribution system monitoring, hazard assessment
and control.

United States Environmental Protection Agency (USEPA). 2006c. National Primary Drinking
Water Regulations: Stage 2 Disinfectants and Disinfection Byproducts Rule; Final Rule. 71 FR
388. (January 4, 2006). Available online at: http://www.gpo.gov/fdsvs/pkg/FR-2006-01 -
04/pdf/06-3.pdf.

United States Environmental Protection Agency (USEPA). 2006d. Point-of-use or point-of-entry
treatment options for small drinking water systems. EPA-815-06-010.

United States Environmental Protection Agency (USEPA). 2006e. Ultraviolet disinfection
guidance manual for the Final Long Term 2 Enhanced Surface Water Treatment Rule. EPA 815-
R-06-007.

United States Environmental Protection Agency (USEPA). 2007. Simultaneous compliance
guidance manual for the Long Term 2 and Stage 2 DBF Rules. EPA 815-R-07-017.

United States Environmental Protection Agency (USEPA). 2009. Chloramines Q &A's. EPA
815-B-09-001.

United States Environmental Protection Agency (USEPA). 2010. Comprehensive Disinfectants
and Disinfection Byproducts Rules (Stage 1 and Stage 2): quick reference guide.  EPA 816-F-10-
080. Available online at: http://nepis.epa.gov/Exe/ZyPDF.cgi?Dockev=P 100C8XW.txt.

United States Environmental Protection Agency (USEPA). 2012. The Third Unregulated
Contaminants Monitoring Rule (UCMR3): Fact Sheet for Assessment Monitoring of List 1
Contaminants. EPA 815-F-12-003. Available online at:
http://water.epa.gov/lawsregs/rulesregs/sdwa/ucmr/ucmr3/upload/UCMR3 FactSheet__Listl.pdf.

United States Environmental Protection Agency (USEPA). 2014a. Analytical methods approved
for drinking water compliance monitoring under the Disinfection Byproduct Rules. EPA 815-B-
14-004. Available online at:
http://nepis.epa.gov/Exe/ZvPlJRL.cgi?Dockev=P100J7AR.txt.

United States Environmental Protection Agency (USEPA). 2014b. NDMA Fact Sheet. Available
online at: http://www2.epa.gov/sites/production/files/2014-
03/documents/ffrrofactsheet  contaminant ndma  ianuarv2014  final.pdf.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      89                               October 2015

-------
van Heijnsbergen, E., A.M. de Roda Husman, W. J. Lodder, M. Bouwknegt, A.E. Docters van
Leeuwen, J.P. Bruin, S.M. Euser, J.W. den Boer, and J.A.C. Schalk. 2014. Viable Legionella
pneumophila bacteria in natural soil and rainwater puddles. Applied Microbiology. 117(3): 882-
890.

Vonberg, R.P., D. Sohr, J. Bruderek, and P. Gastmeier. 2008. Impact of a silver layer on the
membrane of tap water filters on the microbiological quality of filtered water. BMC Infectious
Diseases. 8: 133.

Wadowsky R.M., R. Wolford, A.M. McNamara, and R.B. Yee. 1985. Effect of temperature, pH
and oxygen level on the multiplication of naturally occurring Legionella pneumophila in  potable
water. Applied and Environmental Microbiology.  49(5): 1197-1205.

Wadowsky, R.M. and R.B. Yee. 1983. Satellite growth of Legionella pneumophila with an
environmental isolate of Flavobacterium breve. Applied and Environmental Microbiology.
46(6): 1447-1449.

Wadowsky, R. M., R. B. Yee, L. Mezmar, E. P. Wing, and J. N. Dowling. 1982. Hot water
systems as sources of Legionella pneumophila in hospital and non-hospital plumbing fixtures.
Applied and Environmental Microbiology.  43: 1104-1110.

Wallensten, A., I. Oliver, K. Ricketts, G. Kafatos, J. M. Stuart,  and C. Joseph. 2010. Windscreen
wiper fluid without added screenwash in motor vehicles: a newly identified risk factor for
Legionnaires' disease.  European Journal of Epidemiology. 25(9): 661-665.

Wang, H., S. Masters,  Y. Hong, J. Stallings, J.O. Falkinham III, M.A. Edwards, and A. Pruden.
2012. Effect of disinfectant, water age and  pipe material on occurrence and persistence of
Legionella, mycobacteria, Pseudomonas aeruginosa and two amoebas. Environmental Science
and Technology. 46: 11566-11574.

Washington State Department of Health (WSDOH). 2009. Water system design manual.  DOH
331-123.

Weintraub, J.M., B. Flannery, DJ. Vugia, L.B. Gelling, JJ. Salerno, V.A. Stevens, C.E. Rose,
R.E. Besser, B.S. Fields, M.R. Moore, and MJ. Conroy. 2008.  Legionella reduction after
conversion to monochloramine for residual disinfection. JournalAWWA.  100(4): 129-139.

Whiley, H. and R. Bentham. 2011. Legionella longbeachae and Legionellosis. Emerging
Infectious Diseases.  17(4): 579-583.

Whiley, H., A. Keegan, H. Fallowfield, and R. Bentham. 2014. Detection of Legionella, L.
pneumophila and mycobacterium avium complex (MAC) along potable water distribution
pipelines. InternationalJournal of Environmental Research and Public Health. 11: 7393-7405.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      90                                 October 2015

-------
Williams, M.M., C.R. Armbruster and MJ. Arduino. 2013. Plumbing of hospital premises is a
reservoir for opportunistically pathogenic microorganisms: a review. Biofouling: The Journal of
Bioadhesion andBiofilm Research. 29(2): 147-162.

Wilson, B.R., P.P. Roessler, E. Van Dellen, M. Abbaszadegan, and C.P. Gerba. 1992. Coliphage
MS-2 as a UV disinfection efficacy test surrogate for bacterial and viral pathogens. In
Proceedings of Water Quality Technology Conference. Denver, Colo., American Water Works
Association.

World Health Organization (WHO). 1996. Silver in Drinking Water: Background document for
development of WHO Guidelines for Drinking-water Quality. WHO/SDE/WSH/03.04/14.
Geneva, Switzerland.

World Health Organization (WHO). 2004. Monochloramine in drinking-water; Background
document for development of WHO Guidelines for Drinking-water Quality.
WHO/SDE/WSH/03.04/83. Geneva, Switzerland.

World Health Organization (WHO). 2005. Water Safety Plans. Managing drinking-water quality
from catchment to consumer. WHO/SDE/WSH/05.06. Geneva,  Switzerland.

World Health Organization (WHO). 2011. Water safety in buildings. Geneva, Switzerland.

Wright, H., D. Gaithuma, M. Heath, C. Schulz, T. Bogan, A. Cabaj, A. Schmalweiser, M.
Schmelzer, and J. Finegan-Kelly. 2012. UV disinfection knowledge base. Denver, Colo., Water
Research Foundation.

Yahya, M.T., S.M. Kutz, L.K. Landeen, and C.P. Gerba. 1989. Swimming pool disinfection: an
evaluation of the efficacy of copper/silver ions. Environmental Health. 51(5): 282-285.

Yee, R.B. and Wadowsky, R.M., 1982. Multiplication of Legionellapneumophila in Unsterilized
Tap Water. Applied and Environmental Microbiology, 43(6): 1330.

Yokoi, H., M.Y. Embutsu, and K. Waseda. 2006. Study on the introduction of hazard analysis
and critical control point (HACCP) concept of the water quality management in water supply
systems. Water Science and Technology. 53(4-5): 483-492.

Yu, W., B. Azhdar, D. Anderson, T. Reitberger. J. Hassinen, T. Hjertberg, U.W. Gedde. 2011.
Deterioration of polyethylene pipes exposed to water containing chlorine dioxide. Polymer
Degradation and Stability. 96:  790-797.

Yu, V.L., J.F. Plouffe, M.C. Pastoris, I.E. Stout, M. Schousboe, A. Widmer, J. Summersgill, T.
File, C.M. Heath, D.L. Paterson, and A. Chereshsky. 2002. Distribution of Legionella species
and serogroups isolated by culture in patients with sporadic community-acquired legionellosis:
an international collaborative survey. Infectious Diseases. 186: 127-128.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      91                                October 2015

-------
Zevenhuizen, L.P., J. Dolfmg, EJ. Eshuis, and IJ. Scholten-Koerselman. 1979. Inhibitory effects
of copper on bacteria related to free ion concentration. MicrobialEcology. 5: 139-146.

Zhang, Z., C. McCann, J. Hanrahan, A. Jencson, D. Joyce, S. Fyffe, S. Piesczynski, R. Hawks, J.
E. Stout, V.L. Yu, and R.D. Vidic. 2009. Legionella control by chlorine dioxide in hospital water
systems. JournalAWWA, 101(5): 117-127.
Draft - Technologies for Legionella Control:
Scientific Literature Review                       92                                  October 2015

-------
A  Appendix

    A.I   Elements of Hazard Analysis and Critical Control Points (HACCP)

HACCP is based on an engineering concept of failure, mode and effects analysis. It has five
initial steps followed by seven main principles (Mortimore and Wallace, 2001).

The five initial steps are below (FAO, 1998):

    •   Step 1: Assemble HACCP team
    •   Step 2: Describe the product
    •   Step 3: Identify intended use
    •   Step 4: Construct a process flow diagram
    •   Step 5: Verification of the process flow diagram

The seven principles include the following:

    •   Principle 1: Conduct a hazard analysis
    •   Principle 2: Determine the critical control points
    •   Principle 3: Establish critical limit(s)
    •   Principle 4: Establish a system to monitor control of critical control points
    •   Principle 5: Establish corrective actions
    •   Principle 6: Validate/verify HACCP plan
    •   Principle 7: Establish documentation and recordkeeping procedures

The first step, assembling the HACCP team, is best completed by bringing together a multi-
disciplinary team. The team may consist of a lead coordinator along with team members
representing different areas of expertise or responsibilities for managing the building water
system. This can include but is not limited to building managers, building water system
operators, engineers, maintenance staff, laboratory managers, plumbing experts, public health
risk assessors, financial experts and environmental health specialists. In some cases, the HACCP
team may include external experts (WHO, 2011; Mortimore and Wallace, 2001). The inclusion
of staff with a broad range of expertise helps to ensure that all priority risks will be identified and
decisions on control measures will be practical to implement (Mattel et al., 2006). The roles
(e.g., task manager, trainer, lead reviewer), responsibilities (e.g., collecting samples) and
expertise (e.g., backflow prevention, water treatment) of each team member are usually
summarized in the HACCP plan.

The second step involves describing the product, in this case the building water. The description
usually includes water characteristics such as pH, storage conditions, temperatures, current
treatment, and source water information (e.g., the type of treatment that was applied to the water
prior to it reaching the building service connection, how/where the water was stored and
distributed throughout the PWS, and standards the water must meet before it reaches the service
connection) (USEPA, 2006a).

The third step is to identify the intended use(s) of the water. Examples can include drinking,
bathing, swimming, cleaning, laundry, flushing toilets,  building heating, cooling and fire

Draft - Technologies for Legionella Control:
Scientific Literature Review                       93                                 October 2015

-------
protection. These intended uses can help indicate the different routes of exposures such as
consumption, inhalation and dermal absorption (WHO, 2005). The intended uses may vary
depending on the type of facility containing the building water system. For example, a hospital
might have additional water uses that a hotel or condominium would not have, and vice versa.
The population using the water will also vary among different types of facilities (e.g.,  a nursing
home will most likely have a larger elderly and immunocompromised population than a hotel).

The fourth step is to construct a process flow diagram. This diagram describes in detail the
building water system from the point where the water is first received (the building service
connection) to where the water meets its intended uses, such as at the tap for consumption,
showering, flushing toilets, or a hot tub. This diagram will include all aspects of the building
water system, such as hot water networks, cold water networks, equipment installed at point-of-
use (e.g., filters), backflow prevention assemblies, and cross connections (direct and indirect).
Descriptions of the water at different points in the system are also usually included, such as
temperatures in the hot and cold water networks, pressures at backflow prevention devices, and
water age throughout the system. Water uses and patterns are also important to include within the
process flow diagram, including intended and unintended uses (WHO, 2011).

The fifth step is to verify the process flow diagram. This refers to an on-site visual inspection of
the entire building water system during initial development of the HACCP plan and periodic
review after construction or maintenance work. The periodic review occurs during various times
to ensure accuracy during all operational processes. Not only does this step ensure all of the
different operational aspects of the building water system are accounted for, it also confirms the
different uses and use patterns in the process flow diagram (FAO, 1998). During the on-site
inspection, the HACCP team confirms that the building water system  meets applicable codes
(Rosenblatt and McCoy, 2014).

After the five initial steps have been completed, the seven principles of HACCP are initiated.
The first principle of HACCP involves conducting a hazard analysis to identify and prioritize
possible hazards in a particular building water system and to identify appropriate control
measures. Hazards within a building water system can be biological, chemical, radiological or
physical. These can occur at many points throughout the entire building water system  (WHO,
2011; FAO, 1998). An  example of a physical hazard in a building water system is scalding due
to hot water or steam (Rosenblatt and McCoy, 2014). An example of chemical exposure is lead
leaching from pipework, and an example of microbial exposure is Legionella growing in
biofilms. Along with identifying hazards, it is important to specify how likely the contaminant or
hazard is to occur, the different control measures put in place for the identified hazard, and the
severity of consequences of the hazard occurring (Martel  et al., 2006;  USEPA, 2006a).

The second principle of HACCP is identification of critical control points (FAO, 1998). Critical
control points are specific points in the building water system that are essential to preventing and
eliminating hazards and where controls can be applied (new or controls currently in place) (FAO,
1998). Decision trees can be useful tools in determining which control points are critical
(Mortimore and Wallace, 2001). Examples of critical control points in a building water system
can include hot water heaters, the location where disinfection is applied, locations where
backflow prevention assemblies are installed or should  be installed, the location of POU
controls, and the points at which routine flushing is conducted.
Draft - Technologies for Legionella Control:
Scientific Literature Review                      94                                 October 2015

-------
The third HACCP principle is establishment of critical limits for the control measures employed
at each critical control point. These critical limits are measurable criteria that separate safe from
potentially unsafe conditions (WHO, 2011; Mortimore and Wallace, 2001). Examples of critical
limits in a building water system can include temperature ranges for a hot water heater,
temperature ranges for hot and cold water lines, disinfectant levels at the point of application or
at distal taps, and the frequency for changing POU filters.

The fourth HACCP principle is identification of monitoring procedures for each critical limit set
for each control point (Mortimore, 2001; WHO, 2011). The monitoring procedures describe what
parameter is being monitored, the monitoring frequency and location, monitoring methods, and
reporting and recordkeeping procedures. According to FAO (1998) the procedures should also
identify staff responsible for each aspect of the monitoring procedures, such as which staff are
responsible for sample collection and analysis.

The fifth HACCP principle is the establishment of corrective actions. When a critical limit is
exceeded, the corrective action procedure will be carried out to restore values to the expected
range. The procedure will also include processes for determining the cause of the exceedance to
ensure it does not recur (FAO, 1998). For example, if the disinfectant residual at distal taps is too
low, the corrective action may include increasing the chemical dosage rate or boosting it at
another location in the system. Other corrective action procedures for managing critical  limits
within building water systems could include maintenance on equipment, inspecting chemical
feed pumps, inspecting boiler controls, preventing the use of contaminated water (e.g., providing
bottled water, shutting off the tap, using an alternative water source), superheating the system,
and flushing the piping system (WHO,  2011). Similar to monitoring procedures described under
the fourth principle, corrective action procedures identify the staff responsible for each task.

The sixth HACCP principle is establishment of verification and validation procedures. The
validation step ensures the system is safe from hazards and that the HACCP plan is effectively
controlling for hazards. Validation may include monitoring physical, chemical or
microbiological parameters, as well as review of clinical surveillance data. Validation can also
help identify unnecessary and ineffective control measures (Rosenblatt and McCoy, 2014;
Mortimore, 2001; FAO, 1998). The verification step ensures the HACCP plan is working
correctly and provides evidence that the HACCP plan is being implemented as intended.
Verification activities may involve a supervisor checking that staff members have performed
monitoring, maintenance and recordkeeping tasks  outlined in the HACCP plan.

The seventh HACCP principle is the establishment of documentation and recordkeeping. It is
imperative to the HACCP process that complete and accurate records  are maintained (FAO,
1998). Recordkeeping allows the HACCP team to track the system's performance of the HACCP
plan as well as the performance of control measures  (USEPA, 2006a). In the event that a water
system experiences a waterborne disease outbreak or another water quality event that could
affect public health, historical water quality data and system records can be helpful to prove due
diligence (i.e., that appropriate steps were taken to control known hazards).
Draft - Technologies for Legionella Control:
Scientific Literature Review                      95                                 October 2015

-------
    A.2  Elements of a Water Management Program
  PROGRAM TEAM—Identify persons responsible for Program
  development and implementation.
  DESCRIBE WATER SYSTEMS/FLOW DIAGRAMS—Describe
  the potable and nonpotable water systems within the building and
  on the building site and develop water-system schematics.
  ANALYSIS OF BUILDING WATER SYSTEMS—Evaluate where
  hazardous conditions may occur in the water systems and
  determine where control measures can be applied.
  CONTROL MEASURES—Determine locations where control
  measures must be applied and maintained in order to stay within
  established control limits.
  M 0 NI TO Ri N G/CORRECTIVE ACTIONS—Establish procedures
  for monitoring whether control measures are operating within
  established limits and, if not, take corrective actions.
  CONHRMATION—Establish procedures to confirm that
    "  the Program is being implemented as designed (verification),
       and
    •  the Program effectively controls the hazardous conditions
       throughout the building water systems (validation).
  DOCUMEN1AI ION—Establish documentation and communication
  procedures for all activities of the Program.
Source: ASHRAE, 2015.
Draft - Technologies for Legionella Control:
Scientific Literature Review
96
October 2015

-------
     A.3  Elements of Water Safety Plan
                     Ik sen I* vtubcr supply
                                  4
                   L'mduLt huyiinl ssiiiih sis
                   IX' I inc 
-------