United States
Environmental Protection
Agency
Office of Science and Technology
Off ice of Water
Washington, DC 20460
EPA-822-R-99-001
November 1999
www.epa.gov
&EPA
Legionella:
Human Health
Criteria Document
-------
ACKNOWLEDGMENTS
This document was prepared for the U. S. Environmental Protection Agency, Office of Ground Water
and Drinking Water (OGWDW) by the Office of Science and Technology (OST) under contract with ICF
Consulting Group (Contract No. 68-C6-0029). Overall planning and management for the preparation of this
document was provided byLatisha S. Parker, MS of OST.
EPA acknowledges the valuable contributions of those who reviewed this document. They include Lisa
Almodovar MPH, Robin Oshiro, MS and Stephen Schaub, Ph.D. of the U.S. EPA. EPA also recognizess the
following external peer reviewers for their excellent review and valuable comments on the draft document:
Janette E. Stout Ph.D. and Paul H. Eldelstein M.D.
-------
TABLE OF CONTENTS
I. Summary 1-1
II. General Information and Properties II-l
A. History II-l
B. Taxonomy II-2
C. Microbiology, Morphology, and Ecology II-6
D. Symbiosis in Microorganisms II-7
III. Occurrence III-l
A. Worldwide Distribution III-l
B. Occurrence in Water III-9
1. Natural Surface Water III-9
2. Groundwater 111-10
3. Man-Made Waters III-l 1
C. Occurrence in Soil 111-19
D. Occurrence in Air 111-20
E. Specific Disease Outbreaks 111-21
1. Nosocomial Outbreaks 111-22
2. Outbreaks Among Travelers 111-22
3. Community Outbreaks 111-32
F. Environmental Factors Affecting Legionella Survival 111-32
1. Symbiotic Microorganisms 111-32
2. Water Temperature 111-35
3. Other Factors 111-36
G. Summary 111-37
IV. Health Effects in Animals IV-1
A. Laboratory Studies IV-1
B. Summary IV-4
V. Health Effects in Humans V-l
A. Symptoms and Clinical Manifestations V-l
B. Clinical Laboratory Findings V-5
C. Mechanism of Action V-6
D. Immunity V-9
E. Chronic Conditions V-l 1
F. Treatment V-l 1
G. Summary V-14
VI. Risk Assessment VI-1
A. Hazard Identification VI-1
B. Dose-Response Information VT-2
C. Potential for Human Exposure to Legionella VI-3
1. Prevalence of Legionella in the Environment VI-3
2. Mode of Transmission to Humans VI-3
-------
D. Risk Factors VI-6
E. Quantification of Potential Health Effects VI-9
F. Minimizing Risk VI-10
G. Summary VI-12
VII. Analysis and Treatment VII-1
A. Analysis of Samples VII-1
1. Collection ofLegionella VII-1
2. Detection of Legionella in Environmental and Biological Samples VTI-3
B. Disinfection as a Water Treatment Practice VTI-8
C. Summary VII-13
VIII. Research Requirements VIII-1
IX. References IX-1
11
-------
LIST OF FIGURES
Figure IH-1. Summary of Reported Cases of Legionellosis in the
United States, 1984-1996 III-4
LIST OF TABLES
Table II-l. Approved Legionella Species II-3
Table III-l. Summary of Reported Cases of Legionellosis in the
United States, 1984-1996 III-3
Table III-2. Legionnaires' Disease in 24 European Countries in 1996 III-7
Table III-3. Occurrence of Legionella Bacteria in Potable Water Supplies and
Distribution Systems 111-12
Table III-4. Occurrence of Legionella Bacteria in Cooling Towers III-l 6
Table III-5. Occurrence of Legionellosis Outbreaks 111-23
Table V-l. Frequency of Symptoms of Legionnaires' Disease V-3
Table V-2. Extrapulmonary Sites of Legionella Infection V-4
Table V-3. Common Clinical Laboratory Findings in Patients with
Legionnaires' Disease V-5
Table V-4. Recommendations for Antibiotic Treatment of Legionnaires' Disease V-l 3
in
-------
(this page intentionally left blank)
IV
-------
I. Summary
This document was prepared to update information in the Environmental Protection Agency's (EPA)
Drinking Water Criteria Document on Legionella (EPA 1985) and is intended to serve as an addendum to that
report. Where appropriate, a summary of relevant information from the 1985 document is presented in each
chapter of this addendum. For a more detailed description of information published before 1986, please refer to
the 1985 document. This chapter presents a summary of the information contained in Chapters II through VII.
Chapter VIE contains a discussion of research recommendations and Chapter IX lists references.
Legionella bacteria are aerobic gram-negative rods associated with respiratory infections. Legionella
pneumophila was first recognized as a disease entity from a pneumonia outbreak at a 1976 Convention of the
American Legion in Philadelphia. Of the 42 known species of Legionella, 18 have been linked to pneumonia
infections in humans. The species L. pneumophila (particularly serogroups 1-6) has been accepted as the
principal cause of human outbreaks of legionellosis, which includes both legionnaires' disease and Pontiac
fever.
Legionella are ubiquitous in natural aquatic environments, capable of existing in waters with varied
temperatures, pH levels, and nutrient and oxygen contents. They can be found in groundwater as well as fresh
and marine surface waters. Their widespread survival in nature can be attributed to their relationships with
other microorganisms in the environment. Symbiotic existence with algae and other bacteria, particularly in
biofilms, increases the availability of nutrients. They also are able to infect protozoans and subsequently
reproduce within these organisms. These relationships provide protection against adverse environmental
conditions, including standard water disinfection techniques. Consequently, Legionella are also prevalent in
anthropogenic waters such as potable water, cooling tower reservoirs, and whirlpools.
Aerosol-generating systems such as faucets, showerheads, cooling towers, and nebulizers aid in the
transmission of Legionella from water to air. Human inhalation of contaminated aerosols leads to Legionella
infections and disease outbreaks. Historically, many of the reported outbreaks were nosocomial (i.e., hospital-
acquired), resulting from the adulteration of hospital potable water supplies, air conditioning systems, or cooling
towers. Due to increased awareness of the disease, numerous community-acquired and travel-acquired
outbreaks are now reported each year as well. However, most cases of legionnaires' disease are sporadic (i.e.,
non-outbreak related) and are acquired in the community.
1-1
-------
Collection of Legionella from natural environmental samples, anthropogenic sources such as plumbing
fixtures and potable water systems, and biological specimens is generally done by taking swab samples. These
samples are typically concentrated by filtration, treated with an acid buffer, and isolated on a BCYE agar culture
medium. An array of serological tests then are used to detect the bacteria. The most commonly used tests are
direct and indirect immunofluorescence assays; however, new techniques are consistently being developed and
improved upon as well.
The health effects of Legionella contamination have been studied in animals as well as in humans.
Experimental studies on guinea pigs and other animals have been conducted to better understand human
infection by Legionella, even though animals are not naturally infected by the bacteria. Infected animals hosting
Legionella in their lungs experience impaired respiratory system performance as a result of the disease process.
Clinical features include hypoxia, fever, seroconversion, and weight loss.
Legionella infection in humans occurs when bacteria are inhaled or aspirated into the lower respiratory
tract and subsequently engulfed by enteric pulmonary macrophages. The bacteria rapidly reproduce within the
macrophages and are eventually released when the host cell lyses. Recent research indicates that the ability of
Legionella to infect certain strains of amoeba is a factor in their infection of human lung tissue, as the amoeba
provides a habitat within the pulmonary system in which the bacteria can live and reproduce. Resistance to
Legionella infection is mainly cell-mediated, although humoral immune responses may also play a role.
Legionellosis in humans has typically been characterized as either an acute self-limiting, non-pneumonic
condition known as Pontiac fever or a potentially fatal pneumonic condition known as legionnaires' disease.
Timely treatment of legionnaires' disease is extremely important for a patient's recovery. Although
erythromycin has historically been used to treat patients with legionnaires' disease, newer macrolides and
quinolones are gaining acceptance as the first choice for treatment.
In terms of risk assessment, it is important to realize that the most prevalent source of Legionella
transmission is potable water from large buildings, particularly hospitals. Thus, although Legionella are widely
distributed in both natural and man-made water systems, transmission to humans from a water source results
mainly from inhalation or aspiration of aerosolized contaminated potable water. Potential risks caused by
Legionella in water supplies are not quantifiable by the measures of modem science. However, preventative
and corrective actions have been discovered and implemented to protect the population, especially highly
susceptible individuals (e.g., immunosuppressed people, certain hospital patients). The most effective measures
of treatment have proven to be a combination of systemic sanitization of entire water systems (e.g., thermal
1-2
-------
disinfection, hyperchlorination, copper-silver ionization) and focal disinfection of specific portions of those
systems (e.g., UV light sterilization, instantaneous heating systems, ozonation). These treatment procedures are
very useful in preventing the recolonization ofLegionella in most water distribution systems.
Fresh, innovative methods of detection and treatment ofLegionella in water supplies and sources are
consistently being uncovered and tested. In addition, new medications are being developed to treat patients
overcome with legionnaires' disease. Substantial advancements have been made since the 1985 report, and
modern science presses on with goals of further understanding Legionella and legionnaires' disease and the
eventual eradication ofLegionella colonies in water distribution systems.
(this page intentionally left blank)
1-3
-------
II. General Information and Properties
A. History
In January 1977, Joseph McDade of the U.S. Centers for Disease Control (CDC) discovered a novel
bacteria while investigating the unexplained pneumonia outbreak at the 1976 American Legion Convention in
Philadelphia (Brenner 1987). Of those attending the convention, 221 became ill with pneumonia, and 34 of
those affected died. The aerobic gram-negative bacteria isolated from infected post-mortem lung tissue and
identified as the causative agent of this pneumonia outbreak was later called Legionella pneumophila, receiving
the name Legionella to honor the stricken American legionnaires andpneumophila from the Greek word
meaning "lung-loving" (Fang et al. 1989).
The symptoms exhibited in the 1976 outbreak were termed Legionnaires' disease. Humans can be
affected by Legionella bacteria in two ways: (1) a potentially fatal multi-system disease involving pneumonia
(legionnaires' disease) and (2) a self-limited influenza-like infection (Pontiac fever) (Hoge and Brieman 1991).
Pneumonia occurs in approximately 95 percent of Legionella infections (Nguyen et al. 1991).
Subsequent to finding L. pneumophila, additional investigations ensued to determine whether prior
undetected outbreaks had occurred. Research revealed five additional outbreaks of legionellosis (i.e., diseases
caused by Legionella), which were attributed to L. pneumophila. The first occurred in 1965 at St. Elizabeth's
Hospital in Washington, D.C. Eighty-one patients became ill with pneumonia, and 14 died (Lowry et al. 1993).
The second pneumonia outbreak occurred in 1973 in Benidorm, Spain, and the third occurred in 1974 in the
same hotel as the Philadelphia outbreak of 1976. In addition, two outbreaks of Pontiac fever occurred, one in
Pontiac, Michigan, in 1968 and the other in 1973 in James River, Virginia. Aside from outbreaks, sporadic
cases of legionellosis were detected in 1943, 1947, and 1959 (Brenner 1987).
Within two years of identifying L. pneumophila, the second species of Legionella, L. micdadei, was
discovered (Dowling et al. 1992). In the following years, advances in growth and enrichment media, combined
with clinical and environmental studies, allowed for the discovery of numerous species of Legionella (Brenner
1987).
B. Taxonomy
II-1
-------
DNA-DNA hybridization studies, as well as unique cellular fatty acid composition, indicated that the
bacteria causing the pneumonia outbreak of 1976 should be classified as a new species. At the First
International Symposium on Legionnaires' Disease, held in 1978, the bacteria received the name Legionella
pneumophila and became apart of the new family Legionellaceae (Bangsborg 1997, Brenner 1986).
The 1985 Legionella Criteria Document discusses the taxonomic approaches and diagnostic techniques
used to classify Legionella species. Molecular techniques used includeDNA hybridization, genomic DNA size
comparison using (Guanine+Cytosine) content, oligonucleotide cataloguing of 16s rRNA, and plasmid analysis
(EPA 1985). Comparison of bacterial DNA and the use of antigenic analysis of proteins and peptides are the
best current methods to classify Legionella species, although some phenotypic characteristics (i.e., gram
reactivity, cell membrane fatty acid and ubiquinone content, morphology, and growth on specific media) can be
used to recognize bacteria at the genus level (Bangsborg 1997, Fang et al. 1989, Winn 1988 ).
Following the initial identification of L. pneumophila in 1977, numerous species have been discovered
within the Legionella genus. The 1985 Legionella Criteria Document listed 22 species within the genus.
Currently, the genus consists of 42 species, seven of which can be further divided into serogroups (Bangsborg
1997). The bacterial strains within a species that can be divided by serotype are genetically homologous (based
on DNA hybridization experiments), but can be differentiated by specific reactivity to antibodies (EPA 1985).
Eighteen of the 42 species of Legionella have been linked to patients with pneumonia (Bangsborg 1997). A
majority of human infections (70-90%) have been caused by L. pneumophila, especially serogroups 1 and 6 (Lo
Presti et al. 1997). Table II-l is a compilation of spedes information.
Table II-l. Approved Legionella Species
Name
L. adela idensis
L. anisa
L. birmin gham ensis
L. bozemanii * (Fluoribacter bozemanae) SG 1 -2
L. brun ensis
L. cherrii
L. cincinnatiensis
L. dumoffii* (Fluoribacter dumoffii)
Implicated in Human Disease?
No
Yes
Yes
Yes
No
Yes
Yes
Yes
II-2
-------
Name
L. erythra
L. fairfieldensis
L.feelei* SG 1-2
L. geestiana
L. gormanii* (Fluoribacter gormanii)
L. gratiana
L. hackeliae * SG 1 -2
L. israelensis
L. jamestowniensis
L. jordanis
L. lansingensis
L. londiniensis SG 1-2
L. longbeachae SG 1-2
L. lytica*
L. maceachernii (Tatlockia maceachernii)
L. micdadei* (Tatlockia micdadei)
L. moravica
L. nautarum
L. oakridgensis*
L. parisiensis
L. pneumophila* SG 1-16
L. quateirensis
L. quinlivaniiSG 1-16
L. rubrUucens
L. sainthelensi SG 1-2
L. santicrucis
L. shakespearei
L. spiritensis
L. steigerw altii
L. tucsonensis
L. wadsworthii
Implicated in Human Disease?
No
No
Yes**
No
Yes
Yes
Yes
No
No
Yes
Yes
No
Yes
Yes
Yes
Yes
No
No
Yes
Yes
(Lo Presti et al. 1997)
Yes
No
Yes
No
Yes
Yes
No
No
No
Yes
Yes
II-3
-------
Name
L. waiter sii
L. worsleiensis
Implicated in Human Disease?
No
No
Source: Bangsborg 1997, unless otherwise noted
SG = serogroup
* = species with experimentally documented ability to parasitize amoeba
** = causes Pontiac fever, but rarely pneumonia (Lo Presti et al. 1998)
An ongoing controversy about the taxonomy of the Legionellaceae family involves the single genus
designation. Because several species have a unique phenotypic characteristic (blue white fluorescence in UV
light) and very low DNA-DNA hybridization homology ioL. pneumophila, two additional genera, Tatlockia
and Fluoribacter, have been proposed by Garrity et al. and Brown et al. (see Table II-1 for the accepted and
proposed names) (Bangsborg 1997). The new genera have not been accepted by the mainstream scientific
community, but Bangsborg (1997) suggested that the classifications may be justified.
The method for determining whether two organisms are of the same genus and/or species is based on
DNA-DNA hybridization studies of Enterobacteriaceae (Bangsborg 1997) Members of the same species are
indicated by 70 percent or greater homology under optimal reaction conditions or 60 percent homology under
stringent conditions; 25-60 percent homology indicates genus member status. The Legionella species in the
proposed Tatlockia and Fluoribacter genuses share less than 25 percent DNA sequence homology withZ.
pneumophila, suggesting the need for new genera However, many argue that the use of DNA-DNA
hybridization is not an effective method to distinguish between genera, since the technology is most accurate for
organisms more closely related. Furthermore, the species in dispute exhibit phenotypic characteristics present in
the Legionella species. Finally, infection with these species results in the same human disease and is treatable
with the same antibiotics as all other Legionella species (Bangsborg 1997).
Bangsborg (1997) examined the multi-genus argument by using crossed immuno-electrophoresis of
proteins from Legionella species. Three rabbit antibody preparations, one against L. pneumophila, a second
against L. micdadei, and a third against L. bozemanii, L. dumoffii, and L. gormanii were used on the
electrophoresed proteins. Findings based on the antigenic profiles suggest that creating the Tacklockia and
Fluoribacter genera is warranted. Further taxonomic investigation is necessary to clarify this debate.
II-4
-------
Finally, identification of species isolates is another highly important taxonomic area of study, since
determining sources of outbreaks is essential to public safety. Molecular methods have been used to identify
individual isolates and will be discussed in Chapter VII, Analysis and Treatment ofLegionella.
C. Microbiology, Morphology, and Ecology
All Legionella species appear as small rods, faintly staining gram-negative. They are unencapsulated,
nonsporeforming, with physical dimensions from 0.3 to 0.9 m in width and from 2 to 20 m in length (Winn
1988). Most exhibit motility through one or more polar or lateral flagella. Legionella cell walls are unique
from other gram-negative bacteria in that they contain significant amounts of both branched-chain cellular fatty
acids and ubiquinones with side chains of more than 10 isoprene units. These bacteria are aerobic,
microaerophillic, and have a respirative metabolism that is non-fermentative and is based on the catabolism of
amino acids for energy and carbon sources (Brenner et al. 1984).
Ubiquitously found in nature, Legionella species exist primarily in aquatic environments, although some
have been isolated in potting soils and moist soil samples (Fields 1996). Legionella can survive in varied water
conditions, in temperatures of 0-63 °C, a pH range of 5.0-8.5, and a dissolved oxygen concentration in water of
0.2-15 ppm (Nguyen et al. 1991).
Even though Legionella are ubiquitous in nature, they have specific growth requirements for culturing.
The 1985 EPA Legionella Criteria document provides a detailed explanation of the process of determining
appropriate growth media to sustain Legionella bacterial growth. A typical media used to grow Legionella is
charcoal yeast extract (BCYE) agar supplemented with -ketoglutarate, L-cysteine, iron salts, and buffered to
pH 6.9 (EPA 1985). Bangsborg (1997) also provides information about Legionella growth mediums. The
BCYE agar can be further supplemented with antibacterial agents to suppress microflora(cefamandole and
vancomycin to inhibit gram-positive bacteria and polymyxin B to inhibit gram-negative bacteria), antifungal
agents (anisomycin for yeast), and inhibitors (glycine) (Nguyen et al. 1991). However, some antibiotics can be
detrimental to Legionella growth. For example, cefamandole can inhibit L. micdadei and several strains of L.
pneumophila (Winn 1993). In addition, pretreatment of respiratory tract specimens with acid before culturing
can be very useful in selecting for Legionella, since these bacteria exhibit acid resistance, unlike most other
bacteria (Nguyen et al. 1991). Optimal temperatures for culturing are 35-37°C (EPA 1985). Bacterial growth
II-5
-------
can be enhanced in a culturing environment with a CO2 concentration from 2.5- 5 percent, but not in excess of
8-10 percent, which can be inhibitory (EPA 1985, Winn 1993).
D. Symbiosis in Microorganisms
Experiments have demonstrated that Legionella in sterile tap water show long-term survival but do not
multiply, whereas Legionella in non-sterile tap water survive and multiply (Surman et al. 1994). Furthermore,
Legionella viability is maintained when they are combined with algae in culture, whereas Legionella viability
decreases once the algae are removed (Winn 1988). Legionella proliferation is dependent on their relationships
with other microorganisms.
The first evidence that Legionella share a symbiotic relationship with other microorganisms came with
the discovery of L. pneumophila 's co-existence in an algal mat from a thermally polluted lake (EPA 1985). In
contrast, Legionella survive almost entirely as parasites of single-celled protozoa (Fields 1996). This
relationship first became apparent to Rowbotham in 1980, with the demonstration of L. pneumophila's ability
to infect two types of amoeba, Acanthamoeba and Naegleria (EPA 1985). Currently, Legionella can infect a
total of 13 species of amoebae and two species of ciliated protozoa (Fields 1996). Table II-1 indicates species
of Legionella that have been shown experimentally to infect amoeba.
Legionella also can multiply intra-cellularly within protozoan hosts (Vandenesch et al. 1990).
Legionella strains that multiply in protozoa have been shown to be more virulent, possibly due to increased
bacterial numbers (Kramer and Ford 1994). The ability to infect and proliferate within hosts provides
Legionella with protection from otherwise harmful environmental conditions. Therefore, they survive in
habitats with a greater temperature range, are more resistant to water treatment with chlorine, biocides and other
disinfectants, and survive in dry conditions if encapsulated in cysts. Enhanced resistance to water treatment has
major implications for disease transmittance and water treatment procedures.
Legionella also grow symbiotically with aquatic bacteria attached to the surface of biofilms (Kramer and
Ford 1994). Biofilms provide the bacteria with protection from adverse environmental conditions (including
during water disinfection) and nutrients for growth. The concentration of Legionella in biofilms depends upon
water temperature; at higher temperatures, they can more effectively out compete other bacteria. Legionella
have been found in biofilms in the absence of amoeba (Kramer and Ford 1994). Because biofilms colonize
II-6
-------
drinking water distribution systems, they provide a habitat suitable for Legionella growth in potable water,
which can lead to human exposure.
II-7
-------
III. Occurrence
Because routine culturing for Legionella in the environment is not a common practice, the occurrence of
these bacteria is often indicated by outbreaks or sporadic cases of legionellosis (i.e., any disease caused by
Legionella). Therefore, this chapter considers the worldwide occurrence or incidence of legionellosis (Section
A) and outbreaks of legionellosis (Section E) as well as the occurrence of Legionella bacteria in water (Sections
B), soil (Section C), and air (Section D). Environmental factors influencing Legionella survival are discussed in
SectionF.
A. Worldwide Distribution
Legionellosis has been reported to occur in North and South America, Asia, Australia, New Zealand,
Europe, and Africa (Edelstein 1988). The true incidence of legionellosis is difficult to determine because
identification of cases requires adequate surveillance. Research suggests that legionnaires' disease is under
reported to national surveillance systems (Marston et al. 1994; Edelstein 1988). Its recognition depends on
physician awareness of the disease and resources available to diagnose it.
Although legionellosis is widely distributed geographically throughout the world, most cases have been
reported from the industrialized countries. The ecological niches that support Legione lla (complex recirculating
water systems and hot water at 35-55°C) are not as common in developing countries, so the incidence of
legionellosis may be comparatively low in these countries (Bhopal 1993). However, most geographical
variation in the incidence of legionellosis is probably artifact due to differences in definitions, diagnostic
methods, surveillance systems, and data presentation (Bhopal 1993).
The 1985 Legionella Criteria Document focused mainly on the distribution of legionellosis in the United
States because, at that time, national surveillance data for the United States were available from the Centers for
Disease Control (CDC), whereas surveillance programs in many other countries had not yet been developed.
Surveillance in England and Wales began in 1979, but these data were not included in the 1985 report. Since
1985, many European countries as well as Australia and New Zealand have implemented surveillance programs
to monitor the occurrence of legionellosis. Recent findings of national surveillance programs are summarized
below.
III-l
-------
United States
The CDC first began collecting data on the occurrence of legjonellosis in 1976. The 1985 Legionella
Criteria Document provides a detailed summary of the occurrence and distribution of legionellosis in the United
States through 1983. Data regarding the occurrence of legionellosis in the United States reported to CDC from
1984-1996 are summarized in Table III-l and in Figure III-l. In the United States, the number of cases per
million population rose from 3.5 in 1984 to a peak of 6.3 in 1994 and then began to decline to 4.7 in 1996.
An analysis of data reported to the CDC during the period 1980-1989 examined 3,524 confirmed cases
of legionnaires' disease in the United States (Marston et al. 1994). Disease rates did not vary by year, but rates
were higher in northern states and during the summer. L. pneumophila, serogroup 1, constituted 71.5 percent of
the fully identified isolates of Legionella. Risk factors for morbidity and/or mortality included older age, male
gender, African-American ethnicity, smoking, nosocomial acquisition of the disease, immunosuppression, end-
stage renal disease, and cancer (see Chapter VII, Section D for further discussion of risk factors).
Marston et al. (1994) also concluded that legionnaires' disease is under reported to the CDC. They cite
two studies in which diagnostic tests for legionellosis were routinely performed; Legionella infections
accounted for 3.4 and 4.6 percent of community-acquired pneumonia cases requiring hospitalization. By
projecting this proportion to the estimated total number of community-acquired pneumonia cases in the United
States annually (500,000 cases), they estimate that there would be 17,000-23,000 cases of legionnaires' disease
leading to hospitalization annually. However, fewer than 500 cases of legionnaires' disease are reported to the
CDC annually; therefore, the surveillance system detects fewer than 5 percent of Legionella pneumonia cases in
the United States.
Table III-l. Summary of Reported Cases of Legionellosis in the United States, 1984-1996
Year
1984
1985
1986
1987
Number
of Cases
750
830
980
1,038
Cases per Million
Population
3.5
3.7
4.3
4.3
III-2
-------
1988
1989
1990
1991
1992
1993
1994
1995
1996
1,085
1,190
1,370
1,317
1,339
1,280
1,615
1,241
1,198
4.4
4.8
5.5
5.3
5.3
5.0
6.3
4.8
4.7
Sources: CDC 1994, CDC 1996, CDC 1997b
Figure III-l. Summary of Reported Cases of Legionellosis in the United States, 1984-1996
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
3.5
3.7
4.3
4.3
4.4
4.8
5.5
5.3
5.3
5
6.3
4.8
4.7
Sources: CDC 1994, CDC 1996, CDC 1997b
United Kingdom
III-3
-------
Legionnaires' disease is not a statutorily notifiable disease in England and Wales; therefore, cases are
reported on a voluntary basis. The National Surveillance Scheme for Legionnaires' Disease for residents of
England and Wales was set up in 1979 by the Public Health Laboratory Service (PHLS) Communicable Disease
Surveillance Centre (CDSC), and data have been collected each year since. In addition, the PHLS CDSC
obtains information about cases of legionnaires' disease in residents of England and Wales that are the result of
travel, either abroad or in the United Kingdom, from the European Surveillance Scheme for Travel-Associated
Legionnaires' Disease, which was established in 1987. Data on the occurrence of legionnaires' disease in
residents of England and Wales in 1996 were reported in the Communicable Disease Report (Joseph et al. 1997)
and are summarized below.
In 1996, 201 cases of legionnaires' disease were reported to the PHLS CDSC (Joseph etal. 1997). The
number of cases associated with various sources of infection were: 101 (50%) cases resulting from travel, either
abroad or in the United Kingdom; two (1%) hospital-acquired cases; and 98 (49%) community-acquired cases.
The number of cases linked to outbreaks or clusters was 55 (27%), and the remaining 146 cases (73%) were
reported as single cases. Six outbreaks were associated with industrial sites, and nine outbreaks or clusters were
associated with travel.
A total of 3,005 cases of legionnaires' disease in residents of England and Wales were reported during
the period 1980-1996 (Joseph et al. 1997). Overall, travel and community cases each accounted for 46 percent,
and hospital-acquired infections accounted for the remaining 8 percent. The annual totals of reported cases
fell between 1989 and 1991, following a peak of 279 cases reported in 1988. Since 1993, the annual totals have
been increasing; there was a sharp increase in the number of cases of legionnaires' disease reported in 1996 (201
cases) compared to 160 in 1995. The 201 cases reported in 1996 was the highest total recorded since 1989.
Cases associated with travel abroad accounted for the second highest number of travel cases reported since
1980, and community-acquired cases the largest since 1989. In contrast, the number of hospital-acquired cases
was lower in 1996 than in any of the previous years.
Legionnaires' disease has been a notifiable disease in Scotland since 1988 (Joseph et al. 1997); data are
reported to the Scottish Centre for Infection and Environmental Health (SCIEH). The most recent data available
are for 1996, which were summarized in the Communicable Disease Report (Christie 1997).
In 1996, 24 cases of legionnaires' disease in residents of Scotland were reported to SCIEH (Christie
1997). A total of 15 cases resulted from travel, 13 from travel outside the UK and two from travel within the
III-4
-------
UK. The travel cases were associated with three clusters and two linked groups (cases linked to the same
accommodation but who became ill more than six months apart). There were no cases of hospital-acquired
infection in 1996. Two cases may have been associated with workplace exposure (article does not specify
occupation). The remaining seven cases were presumed by the author to have been acquired in the community.
The total number of cases reported in Scotland in 1996 (24) is 14 fewer than in 1995.
Europe
Since 1993, 24 collaborating European countries have been submitting information on cases of
legionnaires' disease in Europe through completion of the annual reporting forms prepared by the PHLS CDSC
in London. The annual results for 1996 were reported in the Weekly Epidemiological Record (Anonymous
1997b) and are summarized below.
In 1996, 1,566 cases of legionnaires' disease were reported in 24 European countries including England,
Wales, and Scotland (Anonymous 1997b). The number of cases as well as the rate of infection for each of the
24 countries is shown in Table III-2. Four countries reported more than 100 cases each: Spain, 430; France,
294; England and Wales, 200; and Germany (North and South-East), 181. The highest rates of infection (per
million) occurred in Germany (30.17), Croatia (16.00), Denmark (14.40), Spain (11.03), Greece (7.00), and
France (5.25). In all other countries, the rate of infection was less than 5.00 per million population.
In 1996, there were nearly 3 00 more cases than in 1995 and nearly 400 more cases than in 1994
(Anonymous 1997b). The increase was attributed mainly to a large community outbreak in Spain in 1996. In
addition, the average European rate of 4.45 cases per million population in 1996 reflected an increase of almost
1 case per million population from 1995.
Table III-2. Legionnaires' Disease in 24 European Countries in 1996
Country
Austria
Belgium
Croatia
Czech Republic
Cases
20
16
24
12
Population Rate per Million
(millions) Population
8
10
1.5
10.5
2.50
1.60
16.00
1.14
m-5
-------
Country
Denmark
England and Wales
Finland
France
Germany (North and South-east)
Greece
Ireland
Italy
Malta
Netherlands
Northern Ireland
Norway
Portugal
Russian Federation (Moscow)
Scotland
Slovakia
Spain
Sweden
Switzerland
Turkey
Total
Cases
72
200
18
294
181
7
0
84
0
40
0
1
16
45
24
o
3
430
40
26
13
1,566
Population Rate per Million
(millions) Population
5
52
5
56
6
1
3.5
57
0.4
15.5
1.6
4.3
10
10
5
5
39
9
7
30
352.3
14.40
3.85
3.60
5.25
30.17
7.00
0.00
1.47
0.00
2.58
0.00
0.23
1.60
4.50
4.80
0.60
11.03
4.44
3.71
0.43
4.45
Source: Anonymous 1997b
The distribution of cases between various sources of infection were: 16 percent of cases resulting from
travel; 6 percent hospital-acquired cases; 40 percent community-acquired cases; and 38 percent of unknown
origin (Anonymous 1997b). The proportion of community-acquired cases rose from 16 percent in 1994 and 21
percent in 1995 to 40 percent in 1996 largely due to a decline in the proportion of cases from unknown origin,
which represented 55 percent in 1994, 50 percent in 1995, and 38 percent in 1996.
In 1996, individual European countries detected 22 outbreaks: two linked to hospitals, eight to the
community, and 12 to travel (Anonymous 1997b). This distribution represents a decline in nosocomial
outbreaks and a rise in community outbreaks in comparison to 1995 data. The number of outbreaks and the
III-6
-------
number of cases linked to outbreaks maybe largely under reported because many countries are still unable to
provide any epidemiological data associated with the cases of legionnaires' disease they diagnose. For example,
the European Surveillance Scheme for Travel-Associated Legionnaires' Disease detected around 20 travel-
related outbreaks, whereas individual countries detected only 12 travel-related outbreaks. It also is likely that
many industrial-related community outbreaks remain undetected in countries without enhanced surveillance.
The majority of European cases (75%) reported in 1996 were caused by L. pneumophila, serogroup 1
(Anonymous 1997b). L pneumophila of other or undetermined serogroups accounted for 18 percent, and the
remaining 7 percent were attributed to other or unknown Legionella species.
Australia
The Communicable Diseases Network Australia New Zealand collects data on cases of legionellosis in
Australia and New Zealand as part of the National Notifiable Diseases Surveillance System. There have been
1,041 notifications of legionellosis in Australia since 1991, with similar numbers of cases reported each year
(Anonymous 1997a). Since 1995, 255 notifications provided species identification. The majority of cases
were caused by L. pneumophila (41%); however, at least 22 percent of legionellosis cases were attributed toL.
longbeachae (2 percent of cases attributed to other species, and 35 percent of cases not speciated). The report
suggests that these data indicate a microbiological difference in the incidence of legionellosis in Australia
because L. pneumophila has been reported as responsible for at least 90 percent of legionellosis infections in
other countries (Anonymous 1997a).
B. Occurrence in Water
The 1985 Legionella Criteria Document states that Legionella are widely distributed in the aqueous
environment in the United States and, apparently, wherever they are sought (EPA 1985). Since 1985, research
has revealed that Legionella thrive in biofilms, and interaction with other organisms in biofilms is essential for
their survival and proliferation in aquatic environments (Kramer and Ford 1994, Yu 1997, Lin et al. 1998a).
Legionella survival is enhanced by symbiotic relationships with other microorganisms; sediment within biofilms
stimulates the growth of these commensal microflora, which stimulate the growth of Legionella (see Section F
in this chapter for further discussion of symbiotic microorganisms). This section considers the specific
occurrence of Legionella in natural water bodies (surface water and groundwater) as well as man-made waters
(e.g., potable water, cooling towers, whirlpools, etc.).
III-7
-------
1. Natural Surface Water
Legionella are considered to be ubiquitous in the aqueous environment, although few studies examine
natural nonepidemic surface waters for their presence. The 1985 Legionella Criteria Document cited several
studies that clearly demonstrate the widespread occurrence of Legionella from natural surface freshwater
sources (e.g, lakes and streams) in the United States. At the time of the 1985 report, there was little evidence
that the marine environment is a normal habitat for Legionella although they had been isolated from estuarine
waters in Puerto Rico (EPA 1985). More recent studies indicate that Legionella are fairly common in marine
waters (Ortiz-Roque and Hazen 1987, Palmer et al. 1993).
Ortiz-Roque and Hazen (1987) investigated the occurrence of Legionella at twenty-six sampling sites in
Puerto Rico (16 marine, 8 freshwater, and 2 estuarine). L. pneumophilawas the most abundant species at all
sites, with highest densities reported for sewage-contaminated coastal waters. L. pneumophilawas found in
densities several orders of magnitude higher than those in corresponding natural aquatic habitats in the United
States, which the researchers attributed to the presence of higher concentrations of organic matter in the water.
Several other species were widely distributed at all sites, including L. bozemanii, L. dumqffii, L. gormanii, L.
longbeachae, and L. micdadei. The study notes the occurrence of Legionella in water samples taken from
epiphytic rain forest plants, which further demonstrates the ubiquitous nature of these organisms in natural
surface water.
Palmer et al. (1993) studied the occurrence of Legionella in ocean water as part of an investigation of
their presence in raw and treated sewage and nearby receiving waters in California. Ocean-receiving water
located five miles offshore from where treated sewage was discharged contained Legionella; however, ocean
water between the discharge site and coastal bathing beaches was negative. The presence of Legionella at a
nearby beach swimming area was attributed to surface runoff from a flood control channel and river, which
tested positive for Legionella.
2. Groundwater
The 1985 Legionella Criteria Document reported that no studies had documented the occurrence of
Legionella in groundwater (EPA 1985). Recognizing the need for data on the occurrence of Legionella in
groundwater, the U.S. EPA and the American Water Works Association Research Foundation (AWW ARF)
sponsored a study in which untreated groundwater samples from 29 public water supply system wells were
m-8
-------
analyzed for the presence ofL. pneumophila (Lieberman et al. 1994). A variety of hydrogeologic settings were
represented by the wells selected. Samples positive for L. pneumophila were collected from six (21%) of the
sampling sites. In contrast, Campo and Apraiz (1988) sampled water coming from wells in Spain that were not
subject to disinfection; of the 29 samples from eight wells, none were positive for Legionella.
3. Man-Made Waters
As noted previously, Legionella thrives in biofilms. Because bacteria in biofilms are relatively resistant
to standard water disinfection procedures., Legionella are able to enter and colonize potable water supplies
(Kramer and Ford 1994, Lin et al. 1998a). Artificial aquatic habitats (e.g., components of water distribution
systems and cooling towers) are believed to function as amplifiers or disseminators of Legionella present in
potable water (EPA 1985). The 1985 Legionella Criteria Document clearly establishes that these bacteria occur
in a variety of man-made water sources, including components of internal plumbing systems (e.g., faucets and
showerheads), cooling towers, respiratory-therapy equipment, humidifiers, and whirlpools.
Potable Water Supplies and Distribution Systems
In 1980, British investigators first demonstrated that plumbing fixtures in potable water systems
contained Legionella (EPA 1985). The 1985 Legionella Criteria Document provides extensive evidence of
Legionella occurrence in a variety of plumbing equipment, including faucets, shower heads, hot water tanks,
and water storage tanks. Since that time, numerous studies have continued to document the occurrence of
Legionella in components of potable water distribution systems; these studies are summarized in Table ni-3.
As awareness of the ecology and epidemiology of Legionella has increased, attention has shifted from
heat-exchange units, such as cooling towers, to potable water distribution systems as sources of human exposure
and infection. The 1985 Legionella Criteria Document notes the
III-9
-------
Table III-3. Occurrence of Leg/owe//a Bacteria in Potable Water Supplies and Distribution Systems
Setting
community, hospitals,
hotels, residential
community
community
community
community
community
community
community
community
hospitals
hospital
hospital
hospital
hospital
hospital
hospital and hotel
hospital
hospital
hospital
Year
1987
1986-1987
1985-1987
NS
NS
NS
NS
NS
NS
1994-1995
1993-1994
1990-1992
1990
1986-1990
1989
1985-1987
1984-1986
1985
1984-1985
Location
Alicante, Spain
North West England
England
NS
Columbus, Ohio
Columbus, Ohio
Adelaide, Australia
Pittsburgh, Pennsylvania
U.S. Virgin Islands
Allegheny County, Pennsylvania
Taiwan
England and Scotland
Halifax, Nova Scotia, Canada
Halifax, Nova Scotia, Canada
Stanford University Medical
Center, California
Lower Saxony, Germany
Brussels, Belgium
London, England
Dublin, Ireland
Species (Serogroup)
L. pneumophila (serogroups 1,8,6)
L. pneumophila
L. pneumophila (serogroup 1)
NS
L. pneumophila (serogroup 1)
L. pneumophila (serogroup 1)
proposed name: L. w alter sii
L. pneumophila (serogroups 1,3,4-6,12)
L. pneumophila (serogroups 1-6)
L. micdadei
L. gorm anii
L. pneumophila (serogroups 1,3,5)
L. pneumophila (serogroup 1)
L. pneumophila (serogroups 1,4 6)
L. pneumophila (serogroups 1,5)
L. pneumophila (serogroup 1)
L. dum offii
L. pneumophila (serogroups 1-6,9,10)
L. dum offii
L. anisa
L. pneumophila (serogroups 6,10)
L. pneumophila (serogroups 1,4)
L. pneumophila (serogroups 3,5,6)
References
Campo and Apraiz 1988
Jones andAshcroft 1988
Colbourneet al. 1988,
Colbourne and Dennis 1989
Hsu 1986
Voss et al. 1985
Voss et al. 1986
Benson etal. 1996
Stout et al. 1992a
Broadhead etal. 1988
Goetz et al. 1998
Pan et al. 1996
Liu etal. 1993
Bezanson etal. 1992
Marrie etal. 1992
Lowry etal. 1991
Habicht and Mullerl988
Ezzeddineet al. 1989
Oppenheim etal. 1987
Haugh et al. 1990
III-10
-------
Setting
hospital
hospitals
hospital
hospital
hospital
hospitals
hospital
hospital
hospital
hospital
hospitals
hotel, residential, and
industrial
hotel
laboratory
residential
residential
residential
residential
residential and
institutional
Year
1984-1985
1983
1982-1983
1982
1981
1980-1981
NS
NS
NS
NS
NS
NS
1986
1989
1989-1991
1982-1983
NS
NS
NS
Location
Torino, Italy
Canada
NS
France
Pittsburgh, Pennsylvania
Chicago, Illinois
Los Angeles, California
Germany
England
Duesseldorf, Germany
Quebec, Canada
Sao Paulo, Brazil
Bangladesh
Greece
Detroit, Michigan
Finland
Chicago, Illinois
Germany
The Netherlands
Austria
South-eastern Germany
South Africa
Species (Serogroup)
L. pneumophila (serogroup 1)
L. pneu moph ila (serogroups 1,3)
L. dumoffii
L. pneu moph ila
L. anisa
L. pneumophila (serogroup 1)
L. anisa
L. pneu moph ila
L. pneumophila (serogroup 1)
L pneumophila (serogroups 1,6)
L. pneumophila (serogroups 1-6,8)
L. longbeachae (serogroups 1,2)
L. micdadei
L. pneumophila (serogroups 1,6)
L. pneumophila
L. pneumophila (serogroups 1,8)
NS
L. pneu moph ila
L. pneumophila (serogroups 1-6)
L. pneu moph ila
L. pneumophila (serogroups 1,3,6,10)
NS
References
Moiraghi Ruggenini et al. 1989
Tobin et al. 1986
Stout et al. 1985b
Bornstein et al. 1985
Stout et al. 1982
Gorman et al. 1985
Botzenhart et al. 1986
Ribeiroet al. 1987
Hell 1989
Alary and Joly 1992
Pellizari and Martins 1995
Hossain and Hoque 1994
Alexiouet al. 1989
Paszko-Kolva et al. 1991
Zacheus and Martikainen 1994
Arnow et al. 1985
Tiefenbrunner etal. 1993
Luck et al. 1993
Augoustinos et al. 1995
III-l 1
-------
Setting
residential
residential
residential
NS
Year
NS
NS
NS
1987-1988
Location
Pittsburgh, Pennsylvania
Pittsburgh, Pennsylvania
Vermont
New S outh W ales, Austra lia
Species (Serogroup)
L. pneu moph ila
L. pneu moph ila
L. pneu moph ila
NS
References
Stout et al. 1992b
Lee et al. 1988
Witherell et al. 1988
Hedges and Roser 1991
NS = not specified
m-12
-------
presence ofLegionella in water distribution systems of hospitals, hotels, clubs, public buildings, homes,
and factories; recent studies confirm that these systems continue to be a major source ofLegionella
exposure (see Table III-3 for examples).
The 1985 Legionella Criteria Document stated that no isolations of Legionella had been reported
from the extramural components of community water distribution systems. Based on indirect evidence,
water in distribution systems was believed to be contaminated with Legionella infrequently and with low
numbers of organisms (EPA 1985). At that time, Legionella were thought to be introduced into the
distribution systems Ihrough cross connections with equipment such as cooling towers, evaporative
condensers, lawn sprinkling equipment, and hoses (EPA 1985). Since 1985, studies have shown that
Legionella are present in all segments of community water supplies, including treatment facilities
(Campo and Apraiz 1988, Colbourne and Dennis 1989, Colbourne et al. 1988, Voss et al. 1986).
Cooling Towers
The first outbreak of Pontiac fever in 1968 was later linked to the presence of Legionella in a
defective evaporative condenser in a county health department building (EPA 1985). The 1985
Legionella Criteria Document notes numerous outbreaks of legionellosis that have been linked to
cooling towers and evaporative condensers in hospitals, hotels, and public buildings, clearly establishing
these water sources as habitats for Legionella. Table III-4 summarizes more recent studies that
document the continued presence of Legionella in cooling towers and evaporative condensers.
Whirlpools and Spas
Whirlpools and spas serve as an ideal habitat for Legionella because they are maintained at
temperatures ideal for their growth (Hedges and Roser 1991). In addition, organic nutrients suitable for
bacterial growth often accumulate in these waters. Whirlpools and spas can produce
111-13
-------
Table III-4. Occurrence ofLegionella Bacteria in Cooling Towers
Setting
commercial
commercial
commercial
commercial
hospital
hospital
hotels, universities,
hospitals
industrial
industrial
sewage treatment plant
NS
NS
NS
NS
NS
NS
NS
Year
NS
NS
NS
NS
1993-1994
1985
1983
1980-1981
NS
NS
1993
1988-1991
1988-1991
1987-1988
1987
1983-1987
NS
Location
NS
Sao Paulo, Brazil
NS
San Juan, Puerto Rico
Taiwan
Singapore
Canada
Jamestown, New York
Bangladesh
Adelaide, Australia
Fall River, Massachusetts
Adelaide, Australia
United States
New S outh W ales, Austra lia
Singapore
Israel
South Africa
Species (Serogroup)
L. pneumophila (serogroups 1,3,5,6,10)
L. pneumophila (serogroups 1,6)
L. bozem anii
NS
L. pneumophila (serogroups 1-6)
L. bozem anii
L. micdadei
L. gorm anii
L. dumoffii
L. pneumophila (serogroup 1)
L. pneumophila (serogroups 1,4)
L. pneumophila (serogroups 1,4,6)
L. anisa
L. pneumophila
informal name: L. genomospecies 1
L. pneumophila (serogroup 1)
L. pneumophila (serogroups 1-14)
L. anisa
L. rubrUucens
NS
NS
L. pneumophila (serogroups 1,5,7,8)
L. dumoffii
NS
NS
References
Kusnetsov et al. 1993
Pellizari and Martins 1995
Cappabianca et al. 1994
Negron-Alvira etal. 1988
Pan et al. 1996
Nadarajah and Goh 1986
Tobin et al. 1986
Gorman et al. 1985
Hossain and Hoque 1994
Benson etal. 1996
Keller et al. 1996
Bentham 1993
Shelton et al. 1994
Hedges and Roser 1991
Meers et al. 1989
Shuval et al. 1988
Grabow et al. 1991
m-14
-------
Setting
NS
Year
NS
Location
Japan
Species (Serogroup)
L. pneumophila (serogroups 1,3,6)
References
Ikedo and Yabuuchi 1986
NS = not specified
m-15
-------
water droplets of respirable size that have the potential to transmit Legionella to humans (Jernigan
1996). The 1985 Legionella Criteria Document notes two outbreaks resulting from the presence of
Legionella in whirlpools, one involving a therapeutic whirlpool and another involving a recreational
whirlpool (EPA 1985). Recent studies document the continued presence of Legionella in whirlpools.
Hsu et al. (1986) detected Legionella in 5 of 140 (13%) spa whirlpool samples. Hedges and Roser
(1991) tested spas in New South Wales, Australia and found that 11 of 43 (26%) contained Legionella.
In addition, several spa filters were found to have higher Legionella counts than the water contained in
the pool, suggesting that spa filters can act as protective reservoirs or niches for Legionella. Fallen and
Rowbotham (1990) also isolated Legionella from whirlpool water and filters while investigating a large
outbreak of legionellosis at a leisure complex in Scotland. Jernigan etal. (1996) isolated Legionella
from the sand filter in a cruise ship whirlpool spa following an outbreak of legionnaires' disease among
cruise ship passengers.
Other related sources of Legionella include spring water spas and saunas. Spring water therapy
is medicinally accepted in many European countries and often involves aerosol exposure or bathing in
certain spring waters (thermal or non-thermal). During an epidemiologic survey of spa waters in France,
15 different Legionella species were isolated, including a species that had never before been identified,
L. gratiana (Bornstein et al. 1989a, 1989b). Den Boer et al. (1998) reported a case of legionnaires'
disease linked to an air-perfused footbath at a sauna in The Netherlands that was found to be
contaminated with L. pneumophila.
Wastewater
The 1985 Legionella Criteria Document reports only a few instances of Legionella isolation from
wastewater (EPA 1985). The 1985 document notes the difficulty of isolating Legionella from
wastewater because it contains so many other microorganisms. Palmer et al. (1993) conducted an
extensive study to determine whether Legionella were present in the influent of a major metropolitan
sewage treatment plant and to determine how well the bacteria could survive the different stages of
sewage treatment. They found that Legionella were always present in all phases of the sewage treatment
process, including the secondary effluent that was discharged through an ocean outfall. They also noted
that population numbers did not significantly decline in different stages of the treatment process.
111-16
-------
In a later study, Palmer et al. (1995) examined tertiary treated (including chlorination) sewage
effluents that are used as reclaimed water and aerosols obtained from above a secondary sewage
treatment basin for the presence ofLegionella. The bacteria were detected in samples of reclaimed
water at all four sites tested using two detection methods: polymerase chain reaction and direct
fluorescent antibody (see Chapter 7, Section A for explanation of detection methods). The researchers
noted that they were not able to culture Legionella obtained from any of the reclaimed water samples,
suggesting that chlorine may injure Legionella and cause them to enter a viable but nonculturable state.
Legionella were detected in the air obtained from above secondary treatment (activated sludge) aeration
tanks at one site using polymerase chain reaction, direct fluorescent antibody, and plate culture.
C. Occurrence in Soil
The 1985 Legionella Criteria Document reported that Legionella had been isolated from mud and
sandy, moist soil on the edge of streams containing the bacteria. The 1985 document noted a lack of
data indicating soil is involved in the transmission of Legionella to humans although excavations and
other soil disturbances had been associated with some Legionella epidemics. At that time, Legionella
had only been from mud or moist soil (EPA 1985). More recently, one species, L. longbeachae, was
shown to inhabit and thrive in soil (Steele et al. 1990). Following an outbreak of legionellosis due toL.
longbeachae in South Australia in 1988 and 1989, Steele et al. (1990) analyzed a number of water and
soil samples to find the source of the organism. L. longbeachae was not isolated from any of the water
samples or natural soil samples; however, the bacteria was isolated from three samples of potting soil
mixes and from soil surrounding two potted plants. L. longbeachae was able to persist for seven months
in two potting mixes stored at room temperature. The researchers concluded that the isolation and
prolonged survival of L. longbeachae in potting mixes suggest that soil rather than water is the natural
habitat of this species and may be a source of human exposure.
D. Occurrence in Air
As discussed in Sections B and C of this chapter, the natural habitat for Legionella appears to be
aquatic bodies and perhaps, for L. longbeachae, soil. However, Legionella can be found in air as part of
aerosols. The 1985 Legionella Criteria Document establishes aerosolization as an important component
of Legionella transmission from the aquatic environment to the human respiratory system (see Chapter
VI, Section C.2 for further discussion of transmission to humans). At the time of the 1985 report,
111-17
-------
aerosol-generating systems that had been linked to disease transmission included cooling towers,
evaporative condensers, plumbing equipment (e.g., faucets, showerheads, hot water tanks), humidifiers,
respiratory-therapy equipment (e.g., nebulizers), and whirlpool baths (EPA 1985). Studies published
after the 1985 report have confirmed the presence ofLegionella in aerosols from several of these
systems (Bollin et al. 1985, Seidel et al. 1987).
In most cases, disease outbreaks resulting from Legionella aerosolization have involved indoor
exposure and outdoor exposure to within 200 meters. However, Addiss et al. (1989) describe an
outbreak that occurred in Wisconsin in which aerosolized L. pneumophila from an industrial cooling
tower was disseminated at least one mile (1.6 km) and perhaps up to two miles (3.2 km).
Meteorological conditions that suppress vertical mixing and favor horizontal transport of aerosols (e.g.,
fog, high humidity, and cloud cover) occurred before and intermittently during the outbreak and
presumably contributed to the lengthy transport.
E. Specific Disease Outbreaks
Legionellosis can occur as sporadic cases or as outbreaks. The majority of cases of legionnaires'
disease are sporadic rather than outbreak related (Stout et al. 1992a). The study of outbreaks caused by
Legionella has yielded essential information about these bacteria and the illnesses they cause. Early
outbreaks illustrated the clinical course of legionnaires' disease and Pontiac fever. Subsequently,
epidemics provided information regarding the sources of human exposure, risk factors for the
development of disease, and the efficacy of treatment options.
Legionellosis outbreaks have been attributed most frequently to exposure to contaminated
cooling towers, potable water, or components of water distribution systems. Outbreaks of legionellosis
caused by contaminated cooling towers can be explosive with numerous cases over a short period of
time (e.g., Addiss etal. 1989, Fiore et al. 1998, Gecewicz et al. 1994, O'Mahoney et al. 1990).
Legionellosis outbreaks due to contaminated water or water distribution systems tend to be more
insidious and may only be revealed after active surveillance is introduced (e.g., Brady 1989, Colville et
al. 1993, Goetz et al. 1998, Guiget et al. 1989, Hanrahan et al. 1987, Helms et al. 1988, Le Saux et al.
1989, Meenhorst et al. 1985, Schlech et al. 1985, Struelens et al. 1992).
m-18
-------
Establishing the source ofLegionella bacteria causing a legionellosis outbreak can be
problematic due to their ubiquitous nature in the environment. Epidemiologic investigations of
outbreaks often rely on multiple molecular subtyping techniques to match clinical isolates ofLegionella
with isolates from environmental samples (Johnston et al. 1987, Mamolen et al. 1993, Struelens et al.
1992, Whitney et al. 1997). Detection ofLegionella in environmental and biological samples is
discussed further in Chapter VJJ.
Outbreaks of legionellosis typically are categorized as nosocomial (i.e., hospital-acquired),
travel-acquired, or community-acquired. Table III-5 summarizes outbreaks of legionellosis that have
been reported since 1985, including the type of outbreak, the setting in which the outbreak occurred, the
source of the outbreak, the number of individuals affected, the species implicated, and the location and
time of the outbreak. In addition, specific characteristics and features of the various types of outbreaks
are described below.
1. Nosocomial Outbreaks
Studies have linked nosocomial legionellosis to air conditioning systems and cooling towers;
however, numerous studies demonstrate the importance of hospital potable water supplies as a source of
nosocomial infections (see Table III-5 for examples). L. pneumophila has most commonly been
implicated as the causative agent in hospital-acquired legionellosis outbreaks (see Table III-5 for
examples).
2. Outbreaks Among Travelers
Travelers are usually exposed toLegionellavia contaminated hotel potable water or
contaminated whirlpool spas at hotels, resorts and cruise ships (see Table III-5 for examples). Two
reported outbreaks resulted from exposure to Legionella-contaminated water in decorative fountains
(Fensterheib et al. 1988, Hlady et al. 1993). As with nosocomial legionellosis outbreaks, the most
commonly implicated species is L. pneumophila (see table III-5 for examples).
Among U.S. residents, travel-associated legionellosis outbreaks are extremely difficult to detect,
and extensive case investigations often are required. The European Surveillance Scheme for Travel
Associated Legionnaires' Disease has greatly enhanced detection of travel-associated outbreaks in
111-19
-------
European cities because individual cases are entered into a centralized database, which is then searched
for other cases linked to the same place of accommodation (Joseph et al. 1997).
111-20
-------
Table III-5. Occurrence ofLegionellosis Outbreaks
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
Nosocomial
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital and
community
1996
January- June, 1996
July 2-12, 1995
January, 1985-
April,1993
March, 1992
February-March,
1992
March, 1983-
September, 1991
December, 1990-
February, 1991
June-October,
1990
1988-1990
Arizona
Ohio
Franklin County,
Pennsylvania
Innsbruck
Univers ity
Hospital, Austria
Albany Medical
Center, New York
Providence, Rhode
Island
Ontario, Canada
Varnamo, Sweden
Glasgow Royal
Infirmary, Scotland
Sao Paulo, Brazil
hot water
distribution
system
hot water
distribution
system
cooling towers
and rooftop air
samples
hot water system
potable water
system used in
nasogastric tubes
potable water
tap water, shock
absorbers within
water pipes
hot water supply
fire hydrants
connected to
main water
supply
NS
8
2
22
14
2
2
13
31
3
5
L. pneu moph ila
(serogroups 6,10)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 6)
L. pneu moph ila
L. pneu moph ila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroups 1)
Kioski et al. 1997
Kioski et al. 1997
Fiore etal. 1998
Prodingeret al. 1994
Venezia etal. 1994
Mermel et al. 1995
Memish et al. 1992
Darelidet al. 1994
Patterson et al. 1994
Levin etal. 1993
m-21
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
June-August, 1989
May, 1989
July, 1988-April,
1989
1984-1988
1977-1988
October, 1985-
September, 1987
September, 1985-
February, 1986
October-
December, 1985
December, 1984-
December, 1985
August, 1982-
December, 1985
January, 1983-
December, 1985
April 16-May 16,
1985
May-September,
1984
NS
Stanford
University
Medical Center
Nottingham,
England
Atlanta, Georgia
Charlotte sville
Virginia
Brussels, Belgium
Paris, France
Glasgow Royal
Infirmary, Scotland
Manitoba, Canada
Columbus, Ohio
Berlin, Germany
District General
Hospital, Stafford,
England
University of Utah
School of
Medicine, Utah
cooling towers
bath water
domestic hot
water system
nebulizer and
water system
study suggests
potable water
water system
shower supply
and water tank
cooling tower
water system,
renal transplant
unit sink
potable water,
showers
water supply
system
air conditioning
unit
cooling tower
3
3
12
13
16
32
4
16
6
7
35
68 confirmed
35 suspected
4
L. pneu moph ila
(serogroup 1)
L. dumoffii
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 3)
L. micdadei
L. pneumophila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneumophila
L. pneu moph ila
L. pneumophila
L. pneu moph ila
(serogroup 1)
Shelton et al. 1994
Lowry etal. 1991
Colville etal. 1993
Mastro etal. 1991
Doebbeling et al. 1989
Struelens etal. 1992
Meletiset al. 1987
Winter etal. 1987
Le Saux etal. 1989
Brady 1989
Ruf etal. 1988
Anonymous 1985
Dennis 1991
O'Mahony etal. 1990
Johnston et al. 1987
111-22
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
hospital
rehabilitation
center
renal transplant
unit
June-July, 1984
1983
August, 1978-
November, 1983
June 27 -August
25, 1983
November, 1982-
March, 1983
February-
September, 1982
1981
1981
NS
NS
NS
June, 1989-March,
1990
Halifax, Nova
Scotia
NS
Leiden University
Hospital, The
Netherlands
Rhode Island
Paris, France
Upstate New Yo rk
Iowa City, Iowa
Paris, France
Quebec City
NS
Germany
Sao Paul, Brazil
shower heads,
faucets, ac filter
hot water
hot potable
water system
water in cooling
tower
water supply
potable water,
showers, and
water system
hot and cold
water systems
hot water system
distilled water
hot water supply
system
potable water
potable water
system
8
NS
21
15
47
7 confirmed
4 suspected
16
6
5
19
11
8
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
L. pneu moph ila
(serogroups 1,10)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. dum offii
L. pneu moph ila
(serogroup 1)
L. anisa
L. pneumophila
(serogroup 1 )
L. pneu moph ila
(serogroup 1)
Martin etal. 1988
Palmer 1986
Meenhorstet al. 1985
Garbe 1985
Guiguetet al. 1987
Hanrahan et al. 1987
Helms etal. 1988
Neill etal. 1985
Joly etal. 1986
Bornstein et al. 1986
Nechwatalet al. 1993
Levin etal. 1991
Travel- Acquired
cruise ship
July-August, 1994
Cruise ship to
Bermuda
whirlpool
14
L. pneu moph ila
(serogroup 1)
Guerrero et al. 1994
Guerrero and Filippone
1996
m-23
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
cruise ships
hotel
hotel
hotel
hotel
hotel
hotel
hotel
hotels
hotel
leisure complex
leisure complex
ski lodges
ski resort
April, 1994
September-
October, 1996
May 1996
May-August, 1995
January 6-February
2, 1992
1986-1990
1988
August- September,
1987
1973-1987
1979-1982
1988
January-March,
1995
October, 1987
January, 1991
New York City to
Bermuda
Mamora Bay,
Antiqua
Minorca, Spain
Kusadasi, Turkey
Orlando, Florida
Ischia Island,
Naples, Italy
Santa Clara
County, California
Yugoslavia
Northern Italy
U.S. Virgin Islands
Lochgoilhead,
Scotland
Northwest England
Vermont
Vermont
whirlpool spas
and aerosols
solar powered
hot water system
hot water system
water supply
decorative
fountain
hot- water supply
fountain in lobby
NS
potable water
potable water
system
whirlpool and
filter
whirlpool
water sources,
whirlpool spa
hot tub
16 confirmed
34 probable
3
4-5
7
5
6
34
15
117
27
NS
8 confirmed
32 possible
17
6
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
NS
L. anisa
NS
L. pneu moph ila
(serogroups 1,3,4)
L. pneumophila
(serogroup 1)
L. micdadei
L. micdadei
L. pneumophila
(serogroup 1)
L. pneu moph ila
Jernigan et al. 1996
Joseph etal. 1997
Joseph etal. 1996,1997
Anonymous 1995a, 1995b
Hlady etal. 1993
Castellani Pastoris et al.
1992
Fenstersheib et al. 1990
Anonymous 1988
Passi et al. 1990
Schlechet al. 1985
Fallon and Rowbotham
1990
Newton etal. 1996
Mamolen et al. 1993
Thomas et al. 1993
111-24
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
Community- Acquired
artesian well
construction site
business district
coal mine
commercial
building (BBC)
community
community
community
community
community
(Piccadilly Circus)
community
community
grocery store
October, 1990
April 11-20, 1992
1979-1982
April, 1988
September 1 1-
October 18, 1996
September-
November, 1991
1988-1990
May, 1987-June,
1989
January-F ebruary,
1989
May 30, 1986 and
August27-
October 27, 1986
August 10-29,
1986
October 10-
November 13,
1989
Apulia, Italy
Fairfield, Sydney,
Australia
South Wales, UK
London, England
Alcala de Henares,
Spain
Chorley, United
Kingdom
Sao Paulo, Brazil
South Australia
London, England
Gloucester,
England
Sheboygan,
Wisconsin
Bogalusa,
Louisiana
groundwater
not determined
open pit pond
cooling systems
cooling towers,
water storage
tanks
cooling tower
NS
potting soils,
mixes
cooling towers
wet cooling
towers
industrial
cooling tower
mist machine,
aerosols
2
26
3
NS
49 confirmed
197 possible
11
-3
3
30
33 confirmed
10 suspected
15
29
33
L. pneu moph ila
L. pneumophila
(serogroup 1)
L. pneu moph ila
L. pneumophila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneu moph ila
L. pneu moph ila
(serogroups 1,5)
L. longbeachae
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
Miragliotta etal. 1992
Levy etal. 1994
Davies etal. 1985
Dennis 1991
Anonymous 1996
Peiriset al. 1992
Levin etal. 1993
Steeleet al. 1990
Watson et al. 1994
Hunt etal. 1991
Addiss etal. 1989
Mahoney etal. 1992
m-25
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
home and butcher
shop
hospital
hot spring
hotel
industrial
industrial estate
industrial foundries
industrial plant
industrial plants
locker room
nursing home
nursing home
nursing home
September 1986
July 2-12, 1995
1986
April 22-27, 1993
October, 1988
1996-1997
October-
November, 1996
June-August, 1994
July, 1987
May 15-17, 1982
1994
1994
December, 1990
Italy
Franklin County,
Pennsylvania
France
Sydney, Australia
Lostock, England
Northamptonshire
Enlgand
West Midlands,
England
Birmingham,
England
NS
Michigan
Ontario, Canada
Ontario, Canada
Nagasaki, Japan
shower and
condensation
water
cooling towers
and rooftop air
samples
spring water
system
cooling towers
water cooling
system
cooling towers
cooling tower
cooling towers
potable water
whirlpool aerator
water system
water system
not determined
3
22
5
4
57
20
7
8
3
14
10
9
2
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroups 1,3)
L. pneu moph ila
(serogroup 1)
NS
L. pneumophila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 6)
L. sainthelensi
(serogroup 1)
L. sainthelensi
(serogroup 1)
L. pneu moph ila
(serogroup 1)
Castellani Pastoris et al.
1988
Fiore etal. 1998
Bornstein et al. 1989a
Bell etal. 1996
Anonymous 1989
Joseph etal. 1997
Joseph etal. 1997
Joseph etal. 1995
Muraca et al. 1988
Mangione etal. 1985
Tang et al. 1995
Tang et al. 1995
Maesaki etal. 1992
111-26
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
office building
office building
plastics factory
plastics factory
plastics factory
police HQ building
power station
prison
recycling plant
retail store
retail store
retirement home
January-February,
1990
April, 1984
October-
November, 1996
1996
August, 1996
October, 1985
September-
October, 1981
August- September,
1993
June, 1994
September 29-
October 22,1996
May-June, 1986
June 10- July 22,
1988
Christchurch, New
Zealand
New York City,
New York
Trent, England
Wales
Yorkshire, England
United Kingdom
United Kingdom
Michigan
South England
Southwestern
Virginia
Maryland
Los Angeles,
California
cooling tower
cooling tower
unregistered
cooling tower
cooling towers
water from an
uncovered
outdoor tank
air conditioning
system
small capacity
cooling towers
cooling towers
cooling tower
whirlpool spa
display
not determined
evaporative
condenser and
potable water
4 confirmed
3 suspected
86
2
4
2
6
3 confirmed
2 suspected
17
5
23
27
6
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
NS
L. pneu moph ila
(serogroup 1)
L. pneumophila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
L. pneu moph ila
(serogroup 1)
L. pneu moph ila
(serogroup 1)
Mitchell et al. 1991
Friedman et al. 1987
Joseph etal. 1997
Joseph etal. 1997
Joseph eral. 1997
O'Mahony etal. 1989
Morton et al. 1986
Gecewicz et al. 1994
Joseph etal. 1995
Hersheyet al. 1997
Redd 1990
Breiman etal. 1990
m-27
-------
Setting
Dates
Location
Source
# Affected
Species
(Serogroup)
References
sauna
town building
town building
December, 1992-
January, 1996
July August, 1993
August -October,
1993
The Netherlands
Fall River,
Massachusetts
Rhode Island
hot water system
cooling towers
cooling towers
6
11
17
L. pneu moph ila
L. pneu moph ila
(serogroup 1)
L. pneumophila
(serogroup 1)
Den Boer et al. 1998
Gecewicz et al. 1994
Keller et al. 1996
Gecewicz et al. 1994
Whitney et al. 1997
Unknown
NS
NS
March-April, 1993
1986-1996
Georgia and
Florida
Singapore
NS
cooing towers,
fountains, spa
pools
1 confirmed
24 suspected
22 confirmed
236 presumed
NS
L. pneu moph ila
Anonymous 1993
Heng etal. 1997
NS = not specified
m-28
-------
3. Community Outbreaks
Cooling towers and potable water are the most common causes of community outbreaks of
legionellosis (see Table III-5 for examples). Other less common sources reported include a whirlpool
spa display at aretail store (Hershey et al. 1997) and a grocery store mist machine (Mahoney et al. 1992).
Some outbreaks involve residential exposure (e.g., Breiman et al. 1990, Tang et al. 1995), whereas
others involve exposure at the workplace (e.g, Anonymous 1989, Dennis 1991, Joseph et al. 1997,
Joseph et al. 1995, Muracaet al. 1988). Community-acquired outbreaks have often been associated with
urban rather than rural areas (Joseph et al. 1997), which is not surprising given the increased availability
of artificial water bodies in urban areas. As with nosocomial and travel outbreaks, L. pneumophila is the
species most commonly implicated in community-acquired outbreaks.
As noted previously, the vast majority of cases of legionnaires' disease are community-acquired
sporadic (i.e., non-outbreak related) (Stout et al. 1992a). Straus et al. (1996) studied 146 adults
diagnosed with having nonepidemic, community-acquired legionnaires disease and the possible link to
residential potable water. Legionella was isolated from water in six percent of case patients homes (1-8
sites per home) compared to three percent of control patients homes. The researchers suggest that
transmission of Legionella from domestic water may have occurred in more instances than the study
results indicate, since sampling occurred as much as six weeks after a patient's illness.
F. Environmental Factors Affecting Legionella Survival
1. Symbiotic Microorganisms
Legionella can only exist on artificial cultured media in very specific conditions and under
particular temperature, pH, and nutritional requirements. Nevertheless, they survive in an extremely
wide range of conditions in natural and man-made aquatic habitats. Their survival is enhanced by
symbiotic relationships with other microorganisms such as protozoa, algae, and other bacteria, which
provide them with advantages in the natural environment as well as in anthropogenic potable water
distribution systems. Legionella have the unique ability to multiply within protozoan cells, which helps
111-29
-------
them survive over a wide temperature range and resist the effects of chlorine, biocides, and other
disinfectants.
Amoebae and Other Protozoa
Many species ofLegionella can infect amoebae and other protozoa and subsequently reproduce
within these protozoans. Legionella have been found to infect and incorporate themselves into at least
13 species of amoebae including Acanthamoeba, Hartmanella, Valkampfia, and Naegleria, and two
strains of ciliates, Tetrahymena and Cyclidium (Lee and West 1991, Paszko-Kolva et al. 1993, States et
al. 1989, Kramer and Ford 1994, Henke and Seidel 1986, Fields 1996, Vandenesch et al. 1990). Further,
a study by Vandenesch et al. (1990) illustrated thatZ. pneumophila can infect and reproduce within the
amoeba Acanthamoeba, even when the ratio of Legionella cells to amoebae is low. Various species of
Legionella have been detected recently that are able to grow intracellularly in protozoan cells even
though they have never been capable of growth on standard Legionella media. These organisms have
been called LLAP (Legionella-\ike amoebal pathogens) organisms, and they have the ability to infect
and propagate in many mammalian and protozoan cells (Fields 1996). After the bacteria are
phagocytosed by amoebae, they multiply within their vesicles and remain encapsulated in the cysts until
the vesicles and/or amoeba rupture (States et al. 1989). Because Legionella replicate rapidly
intracellularly within protozoan hosts for prolonged periods of time, amoebic vesicles can contain
hundreds ofLegionella cells at once (Berk et al. 1998, Lee and West 1991). In addition, replication
within protozoa can contribute to enhanced virulence of Legionella (Kramer and Ford 1994).
The fact that Legionella have the ability to infect and grow in protozoa is extremely critical to
their maintenance and survival. Not only can they multiply quickly within protozoan cells, but they also
obtain protection from disinfectants and other adverse environmental conditions. For example,
Legionella caught in encysted protozoa have demonstrated better resistance to chlorine thanE. coli, a
common indicator of water quality (Paszko-Kolva et al. 1993, States et al. 1989, Kramer and Ford 1994).
Legionella trapped in the amoeba A. polyphaga have been shielded from the effects of exposure to 50
mg/L of free chlorine (Paszko-Kolva et al. 1993, Fields 1996). Intracellularly grown Legionella are also
more resistant to biocides, chemical disinfection, and other physical stresses than Legionella grown on
cultured media. Because protozoa ingest virulent strains of L. pneumophila, they also augment growth
111-30
-------
of the bacteria in cooling towers and other epidemic sources (Barbaree et al. 1986). In addition,
encapsulation in cysts allows Legionella to survive in the dry conditions of an aerosol for extended time
periods, thus allowing the bacteria to persist, disperse, and infect human hosts (Fields 1996).
Algae and Other Bacteria
Certain algae such as the cyanobacterium Fischerella and the green algae Scenedesmus,
Chlorella, and Gleocystis have fostered the growth of Legionella, but only in the presence of light (Lee
and West 1991, Kramer and Ford 1994, States et al. 1987, Henke and Seidel 1986, Paszko-Kolva et al.
1993). States et al. (1987) found that the highest incidence of Legionella multiplication came from
samples gathered from zones affected by the accumulation of algal materials and leaf litter. Legionella
growth is further supplemented by their utilization of the nutrients supplied by the decomposition and
excretion of algae, as well as decaying organic matter from leaf litter (States et al. 1987).
Legionella also have formed colonies in media deficient in cysteine or iron salts, which they
require for growth. The colonies have been found around strains of common aquatic bacteria such as
Flavobacterium, Pseudomonas, Alcaligenes, and Acinetobacter, which are presumed to provide these
nutrients (Lee and West 1991, Paszko-Kolva et al. 1993, Kramer and Ford 1994, Stout et al. 1985a).
Legionella have also been found attached to the surface of biofilms in water systems (Kramer and Ford
1994). Biofilms are encased microcolonies made up of bacterial cells and attached to a conglomerate of
polysaccharides. They trap nutrients for growth and provide a protective layer for many microbes.
Legionella survive in these biofilms via nutritional symbiosis with other inhabiting organisms (Kramer
and Ford 1994).
2. Water Temperature
Legionella exhibit the ability to survive in an incredibly wide range of temperatures. As a lower
limit, Bentham (1993) observed growth at a water temperature of 16.5°C. The highest water
temperature of a sample cultivated byBotzenhart et al. (1986) was 64°C, while Henke and Seidel (1986)
claimed Legionella to be a "thermoresistant" organism, exhibiting survival in natural warm waters of up
III-3 1
-------
to 60°C and artificially heated waters of 66.3°C. Optimum temperatures for Legionella reproduction
range from 32 to 45°C (Vickers 1987, Kramer and Ford 1994).
Nevertheless, temperature has a formidable effect on the persistence and dissemination of
Legionella in aquatic habitats. While Legionella populations seem to be controlled by extremely low
temperatures, they are enhanced by heat and elevated temperatures found in areas like whirlpools, hot
springs, and blast zones (Henke and Seidel 1986, Lee and West 1991, Verissimo et al. 1991). Colbourne
and Dennis (1989) contend that although Legionella are not thermophilic, they exhibit thermo-tolerance
at temperatures between 40 and 60°C, which gives them a survival advantage over other organisms
competing in man-made warm water systems. Although temperatures between 45 and 55°C are not
optimal for Legionella, these temperatures enable them to reach higher concentrations than other
bacteria commonly found in drinking water, thus providing Legionella with a selective advantage over
other microbes (Kramer and Ford 1994). Legionella were found in natural surface waters of Puerto Rico
in densities several orders of magnitude higher than those in corresponding natural habitats in the United
States (Ortiz-Roque and Hazen 1987) although these differences may be due to factors other than
temperature (e.g., increased nutrient availability). In contrast, the distribution and abundance of
Legionella in south-eastern Australia is comparable to the United States and Europe (Hedges and Roser
1991).
3. Other Factors
Although interaction with other microorganisms and water temperature are the most significant
and evident factors affecting Legionella growth and survival, there are a few other factors, such as
sediment and metals content, that are notable influences as well. These factors are usually amplified by
ideal water temperature or coexisting environmental microflora.
Stout et al. (1985a) tested different external influences of Legionella growth and sustenance. The
results indicated that growth of L. pneumophila declined in the absence of environmental microflora
such as algae and amoebae. The results also showed that as the amount of sediment increased, so did the
population of L. pneumophila. This was largely attributed to the fact that the scale, or mineral deposits,
and detritus, or decaying plant matter, that make up sediment, are used by Legionella organisms as a
111-32
-------
major source of nutrients. In this study, the greatest effect onLegionella growth and survival was caused
by the presence of both sediment and other microbes. The researchers theorized that the sediment
stimulates the growth of environmental microorganisms, which prompts the growth of Legionella that
rely on their environmental by-products and availability as hosts (Stout et al. 1985a).
States et al. (1987) noted that Legionella growth was more evident at the corners and bottoms of
tanks, sedimentation basins, and reservoirs than anywhere else due to the excess sediment and scale in
those areas. It follows that total organic carbon and turbidity are also factors that motivate Legionella
growth since these influences are found in water zones rich in sediment. Vickers et al. (1987) studied
the design of water distribution systems and concluded that vertical tanks were more prone to Legionella
growth due to thicker accumulation of sediment at the bottom of the tank Also, greater amounts of
scale and sediment in older tanks may contribute to increased growth of Legionella. Sediment is
important to Legionella growth because it provides essential nutrients, aids in the growth of other
coexisting microflora, and shelters the organism as well (Vickers et al. 1987).
Changes in water pressure and flow rates of water distribution systems may cause disruption of
the biofilm, resulting in increased concentrations of Legionella in water supplies (Kramer and Ford
1994). Mermel et al. (1995) remarked that repressurization of potable water upon completion of a
construction project may lead to increased concentration of Legionella in the water. They noted that this
phenomenon could occur in the absence of construction (i.e., any situation in which the water pressure is
changed). Straus et al. (1996) reported that recent residential plumbing repair is an independent risk
factor for community-acquired legionnaires' disease.
Water hardness is determined primarily by the amount of calcium and magnesium in scale
deposits. Legionella have been found to flourish in areas where these metallic cations are present
(Vickers et al. 1987). Low levels of iron, zinc, and vanadium also may stimulate the growth of
Legionella (Kusnetsov 1993, States et al. 1987, Stout et al. 1992b), while higher concentrations of
metals like copper, iron, manganese, and zinc may actually be toxic (Kusnetsov 1993).
G. Summary
m-33
-------
Cases of legionellosis have been documented throughout the world; however, the true incidence
of the disease is unknown due to inadequate surveillance. Geographical variation in the incidence of
legionellosis has been attributed to differences in definitions, diagnostic methods, surveillance systems,
and data presentation. National surveillance programs currently are conducted in the United States, 24
European countries (including England), and Australia and New Zealand. In the United States, the
number of cases per million population rose from 3.5 in 1984 to a peak of 6.3 in 1994 and then began to
decline to 4.7 in 1996. In England and Wales, annual totals declined briefly after apeak in 1988 but
have been increasing since 1993.
Legionella are widely distributed in the aqueous environment, including both natural water
bodies (surface water and groundwater) and man-made waters (e.g., potable water, cooling towers,
whirlpools, etc.). The presence of Legionella has been documented in fresh surface water sources (e.g,
lakes and streams), estuarine and marine surface water sources, and groundwater. Legionella thrive in
biofilms, and interact!on with other organisms inbiofilms is essential for their survival and proliferation
in aquatic environments. Bacteria in biofilms are relatively resistant to standard water disinfection
procedures, and therefore, Legionella are able to enter potable water supplies. Legionella find niches
suitable for survival and growth in artificial aquatic habitats (e.g., internal plumbing systems, cooling
towers, respiratory-therapy equipment, humidifiers, and whirlpools), which function as amplifiers or
disseminators of these bacteria.
Although water has been the most documented source of Legionella in the environment, these
bacteria have been isolated from mud, moist soil, and potting soil. Legionella can be transmitted from
water to air by aerosol-generating systems such as cooling towers, evaporative condensers, plumbing
equipment (e.g., faucets, showerheads, hot water tanks), humidifiers, respiratory-therapy equipment
(e.g., nebulizers), and whirlpool baths. Inhalation of Legione I la-contaminated aerosols is an important
source of human exposure and infection.
Human exposure to Legionella-contaminated sources can result in outbreaks of legionellosis.
Outbreaks can be categorized as nosocomial (i.e., hospital-acquired), travel-acquired, or community-
acquired. Nosocomial outbreaks have been linked to hospital potable water supplies as well as air
conditioning systems and cooling towers. Travelers are usually exposed to Legionella in contaminated
111-34
-------
hotel potable water or contaminated whirlpool spas. Community outbreaks are caused by exposure to
the widest variety of sources, but potable water and cooling towers are the most common. L.
pneumophila has most frequently been implicated as the causative agent for all three types of outbreaks.
The majority of cases of legionnaires' disease, however, are community-acquired sporadic (i.e., non-
outbreak related).
The growth and survival of Legionella in the environment is enhanced by their ability to form
symbiotic relationships with other microorganisms. Legionella are able to infect and multiply
intracellularly within at least 13 species of amoebae, allowing them to survive over a wider range of
environmental conditions and resist the effects of chlorine, biocides, and other disinfectants. Because
Legionella replicate rapidly intracellularly within these protozoan hosts, often for prolonged periods of
time, a single amoebic vesicle can contain hundreds ofLegionella. Relationships with certain algae and
bacteria in biofilms also foster the growth ofLegionella., presumably due to the increased availability of
nutrients and resistence to disinfection. Other factors influencing the survival of Legionella in the
environment include water temperature, presence of sediment, and metal content.
111-35
-------
(this page intentionally left blank)
m-36
-------
IV. Health Effects in Animals
A. Laboratory Studies
Although Legionella are widely distributed in the environment, there are no reports of their
isolation from naturally infected animals, and they are considered to be strictly human pathogens. As
discussed in detail in the 1985 Legionella Criteria Document, there is considerable serological evidence
that exists to support exposure or possible subclinical infection in animals, such as horses, cattle, sheep,
swine, nonhuman primates, goats, dogs, and protozoa It should be noted, however, that controversy
exists in the applicability of the utilization of titer criteria in animals, an evaluation method which is
established for measurement of antibody levels in humans.
Animals have been primarily used as hosts for the isolation of the Legionella, models for the
study of the disease process in human legionellosis, models for the study of the virulence of various
Legionella species, as well as for the testing of new diagnostic techniques, immunological responses,
and possible therapeutic approaches. Guinea pigs have been studied extensively due to similarities
between the natural legionnaires' disease in humans, and the experimental disease in guinea pigs. Other
species including rats, gerbils, mice, hamsters, rabbits, nonhuman primates and embryonated hens' eggs
have also been utilized for study of infection by Legionella.
Experimental routes of exposures have been primarily respiratory, including small particle
aerosols, intranasal instillation, nose-only inhalation and intratracheal injections. Infections have also
been induced by ingestion (drinking water and gastric intubation) and intraperitoneal injection routes.
Clinical Features and Symptomatology in Guinea Pigs
The disease process, following an inhalation exposure of L. pneumophila in guinea pigs, has
been characterized by investigators as fever for several days, bacteremia, and fibrinopuiulent pneumonia
with congestion and eventually consolidation (Baskerville 1984, Davis et al. 1983c). The most striking
clinical symptoms are fever and weight loss (Twisk-Meijssen et al. 1987). In fact, weight loss, fever and
seroconversion are considered to be the only dependable clinical criteria of aerosol infection (Berendt et
IV-1
-------
al. 1980). Clinical symptoms and mortality are dose-dependent in nature, and are discussed at length in
the 1985 Legionella Criteria Document. The defense mechanism initially involves resident alveolar
macrophages and polymorphonuclear cells (PMNs), followed by the presence of immunospecific
antibodies (Davis et al. 1983b). The multiplication of the Legionella in recruited macrophages results in
destruction of the macrophages and the release of toxic products in the lungs of susceptible animals.
The alveolar membrane integrity is destroyed, and serum proteins, together with PMNs, macrophages,
bacteria, and debris, fill the air sacs producing hypoxia and impairing respiratory function.
In contrast, intraperitoneal exposure in guinea pigs to L. pneumophila causes a diffuse
fibrinopurulent peritonitis involving the liver and spleen (Chandler et al. 1979c, Hambleton et al. 1982).
In addition, foci of inflammation and necrosis may also be found in the lungs, lymph nodes, pancreas,
and heart. The histological features of thepneumonitis induced by intraperitoneal inoculation are quite
different from those observed in animals infected by the aerosol route; lesions are more focal, the
interstitium is more strongly involved, and necrosis and fibrin in the alveolar exudate is minimal. It was
also noted by Hambleton et al. (1982) that extrapulmonary symptoms, such as diarrhea, kidney or liver
failure, or neurologic disturbances, that are observed with intraperitoneal infections are seldom observed
in guinea pigs infected by the aerosol route. Biochemical changes observed in guinea pigs infected by
the intraperitoneal route include hyponatremia, striking changes in serum trace metals, amino acids and
proteins, changes in liver enzymes indicating hepatic necrosis, and evidence of leukocytosis followed by
leukopenia (Hambleton et al. 1982).
Guinea pigs infected with L. pneumophila by an oral route of exposure demonstrate a febrile
disease with mild pneumonitis and splenitis (Katz and Matus 1984). In one study, subacute exposure to
L. pneumophila in drinking water over a period of 17 days did not cause clinical illness, and none of the
guinea pigs seroconverted (Conner and Gilbert 1979).
Other Animal Models
Rats have also been used as models for L. pneumophila infection. Winn et al. (1982) found that
acute pneumonia occurred in both rats and guinea pigs; however, the rats appeared to be more resistant
to lethal infection and extrapulmonary inflammatory lesions. Davis et al. (1983a) also demonstrated a
IV-2
-------
milder illness in rats similarly exposed. Exposure of marmosets to small-particle aerosols of Z.
pneumophila produces acute fibrinopurulent pneumonia like that observed in guinea pigs (Baskerville et
al. 1983b). Rhesus monkeys are less susceptible than the marmosets or guinea pigs, and the pulmonary
lesions are less severe.
Mice have also been used as models for Legionella infection. Fitzgeorge et al. (1983) found that
Porton mice were highly resistant to aerosol infection withZ. pneumophila, mice remained healthy and
did not develop antibodies. ICR mice infected by intraperitoneal injection had a moderate to low
susceptibility to infection by Legionella, and Mongolian gerbils were found to be highly susceptible
(Patton et al. 1979). Atabular summary of the dose responses of various animals to experimental L.
pneumophila infection provided in the 1985 Legionella Criteria Document aids in emphasizing that there
is great species variation in susceptibility to Legionella infection.
LD50Data
LD50 data and median 50% infection doses (ID50) have been documented in guinea pigs exposed
to L. pneumophila by the aerosol route:
ID50 of <129 bacteria with an LD50 of 1.4xl05 organisms (Berendt et al. 1980)
• LD50 in the range of 500-5000 retained CPU and a fever production ID50 dose of 20 CPU
(Huebneretal. 1984)
retained LD50 of 104 (Baskerville 1984)
Long-Term Effects
The long-term effects of LegioneMa-induced pneumonia are pulmonary fibrosis and functional
impairment of the lung. Studies in surviving guinea pigs, Rhesus monkeys and marmosets exposed to
aerosol infections of L. pneumophila have shown that alveolar fibrosis, cellular infiltration of alveolar
walls, and blockage of some terminal airways are common features 10 days after exposure, and were still
present in guinea pigs after one month (Baskerville et al. 1983a). The infecting organism did not persist
in the lungs, and pulmonary abscesses did not develop. In Syrian hamsters intratracheally infected with
IV-3
-------
L. pneumophila, the alvedar response to the infection was still prominent after 90 and 180 days in some
lungs, and the severity of the inflammation was correlated with a persistent restrictive defect in lung
elasticity (Parent! et al. 1989).
B. Summary
Although animals are not naturally infected by Legionella, their use as models for the study of
human legionellosis is beneficial in understanding the etiology of its clinical manifestations.
Experimental studies of legionellosis in animals, particularly guinea pigs exposed by the respiratory
route of infection, provide useful information on human legionellosis due to the close similarities of
these diseases. These similarities are discussed in detail in the 1985 Legionella Criteria Document.
The disease process in the lungs of susceptible animals is characterized by multiplication of the
Legionella in recruited macrophages; destruction of the macrophages eventually results inhypoxia and
impaired respiratory function. Clinical features include weight loss, fever and seroconversion. The LD50
for guinea pigs exposed to L. pneumophila by the aerosol route is somewhat less than 105 cells. The
long-term effects of Legione//a-induced pneumonia are pulmonary fibrosisand functional impairment of
the lung.
There are varying degrees of susceptibility to Legionella infection among animal species. In
comparison to guinea pigs, which have been studied extensively, rats, monkeys, marmosets and mice are
more resistant to infection by Legionella aerosols. Gerbils are highly susceptible to infection by the
intraperitoneal route.
IV-4
-------
(this page intentionally left blank)
IV-5
-------
V. Health Effects in Humans
Legionellosis in humans has typically been characterized as either anon-pneumonic condition
known as Pontiac fever or a pneumonic condition known as legionnaires' disease. This chapter
summarizes new information available since publication of the 1985 Legionella Criteria Document on
legionellosis in humans, specifically symptoms and clinical manifestations, clinical laboratory findings,
mechanism of action, immunity, chronic conditions, and treatment.
A. Symptoms and Clinical Manifestations
Pontiac fever is described as an acute, self-limiting illness with "flu-like" symptoms. The illness
is characterized by an attack rate of greater than 90 percent of exposed persons and an incubation period
ranging from 24 to 48 hours (Nguyen and Yu 1991, Roig et al. 1994). The symptoms include fever,
chills, headache, myalgia, and malaise (Muder et al. 1989, Nguyen and Yu 1991). The illness typically
resolves without complications within two to five days (Muder et al. 1989). Upper or lower respiratory
tract symptoms have not been associated with this illness. No additional information on Pontiac fever
was located.
Since publication of the 1985 Legionella Criteria Document, the course of legionnaires' disease
has been more precisely defined (Davis and Winn 1987, Ampel and Wing 1990; Nguyen et al. 1991,
Stout and Yu 1997). The incubation period for legionnaires' disease is two to ten days, although
incubation periods exceeding ten days have been reported (WHO 1990). Malaise, myalgia, anorexia,
and headache typically occur within 48 hours. These symptoms are usually accompanied by a rapidly
rising fever that frequently reaches 39°C or 40°C. Chills may also occur with the fever. A dry cough is
typically present in the early stages of the illness. Although the cough may become productive with
minimally or moderately purulent sputum within several days, hemoptysis is rarely observed. Other
common early features of the illness include neurologic abnormalities (e.g., confusion, disorientation,
lethargy) and gastrointestinal symptoms (e.g., nausea, vomiting, watery diarrhea). As the illness
progresses, chest pain (often pleuritic), dyspnea, and respiratory distress may be observed.
V-l
-------
Frequencies of these common symptoms vary. Table V-l summarizes frequencies for these
symptoms based on estimates provided in two review articles. An important point to note is that the
clinical features described for legionnaires' disease do not distinguish it from other bacterial pneumonias
(Roig et al. 1994). Recent studies have shown that symptoms initially thought to occur with greater
frequencies in patients with legionnaires' disease are actually not distinctive. For example, Edelstein
(1993) reported that diarrhea, which has historically been considered a distinctive feature of
legionnaires' disease, occurred with similar frequency in patients with legionnaires' disease (0-25%)
compared to patients with pneumonias caused by other agents (3-36%). Similarly, bradycardia and
neurologic abnormalities have been "discredited" as distinctive features (Roig et al. 1994, Stout and Yu
1997). Edelstein (1993) concluded that prospective comparative studies have demonstrated that no one
clinical feature can be used to distinguish legionnaires' disease from pneumonia caused by other agents.
Extrapulmonary diseases resulting from legionnaires' disease are rare, but have been reported
with increasing frequency since publication of the 1985 Legionella Criteria Document. Infections in
which Legionella species have been implicated include sinusitis (Lowry andTompkins 1993), cellulitis
(Waldor et al. 1993, Kilborn et al. 1992), pancreatitis (Kesavan et al. 1993, Eitrem et al. 1987),
peritonitis (Lowry and Tompkins 1993), brain abscess (Andersen and Sogaard 1987), perirectal abscess
(Lowry and Tompkins 1993), acalculous cholecystitis (Earle and Hoffbrand 1990), transient aplastic
anemia (Martinez et al. 1991), myositis (Warner et al. 1991), and various wound infections (Lowry and
Tompkins 1993). Stout and Yu (1997) stated that the heart is the most common extrapulmonary site.
This assertion is supported by numerous reports of myocarditis (De Lassence et al. 1994, Armengol et al.
1992, Devriendt et al. 1990), pericarditis (Lowry and Tompkins, 1993, Domingo et al. 1989), and
endocarditis (Berbari et al. 1997, Chen et al. 1996).
Table V-l. Frequency of Symptoms of Legionnaires' Disease
Symptoms
Fever
>38.2°C
Frequency (% of Patients)
A1
—
—
B2
99
71
V-2
-------
>39°C
>39.4°C
Cough
Chills
Headache
Myalgias
Dyspnea
Neurological/Confusion
Diarrhea/Nausea
Chest Pain
70-95
—
75-95
59-73
32-75
38-75
—
25-50
13-54
30-42 5
—
79 (65)
89
78
50
—
48
37 3
45 4
45
3
4
5
The source of information is Davis and Winn 1987. The authors did not provide any indication
regarding the number of patients evaluated, but noted that the frequency was a "composite
estimate from published series."
The source of information is Ampel and Wing 1990. The authors indicated the frequency was
based on 231 patients. Mimbers indicated in parentheses are exceptions.
Symptom was listed as "neurologic abnormalities."
Symptom was listed simply as "diarrhea."
Symptom was listed as "pleuritic pain."
The kidney is also a common extrapulmonary site. In the Philadelphia epidemic of 1976,14 of
the 123 cases of legionnaires' disease developed acute renal failure (Shah et al. 1992). Since 1976, at
least 53 additional cases of legionnaires' disease complicated with acute renal failure have been reported
(Lin et al. 1995). Based on the limited histopathology that has been conducted, the acute renal failure
appears to be a result of either acute tubulointerstitial nephritis (Shah et al. 1992, Haines and Calhoon
1987) or acute tubular nephritis (Shah et al. 1992, Fenves 1985), although acute pyelonephritis (Shah et
al. 1992) and glomerulonephritis (Pai et al. 1996, Wegmuller et al. 1985) have been reported.
Typically, extrapulmonary infections occur concurrently with pneumonia and are believed to
result from bacteremia(Stout and Yu 1997, Edelstein 1993). Where extrapulmonary infections develop
prior to the onset of pneumonia, identifying the primary site of infection may be difficult. Several cases
of infections attributed to Legionella species in the absence of pneumonia have been reported (Edelstein
1993). These infections may be the result of direct inoculation of a site with water contaminated with
V-3
-------
Legionella bacteria. Table V-2 summarizes those extrapulmonary sites in whichLegionella infections
have been implicated in the presence and absence of pneumonia. As a final note, extrapulmonary
infections tend to occur with greater frequency in immunocompromised patients or in patients with
severe cases of legionnaires' disease (Edelstein 1993).
Table V-2. Extrapulmonary Sites of Legionella Infection (Source: Edelstein 1993)
Presence of Pneumonia
Blood
Brain
Bowel
Kidney
Liver
Spleen
Hemodialysis shunt
Peritoneum
Prostate
Pericardium
Bone marrow
Skin and fascia
Rectum
Myocardium
Thyroid
Pancreas
Testes
Muscle
Peripheral lymph nodes
Absence of Pneumonia
Blood
Surgical wounds
Bowel
Respiratory sinus
Endocardium
Peritoneum
Pericardium
Skin and fascia
B. Clinical Laboratory Findings
Many abnormalities in standard clinical laboratory tests have been noted in patients with
legionnaires' disease. Some of the more common findings are summarized in Table V-3. The clinical
laboratory findings that are most frequently associated with legionnaires' disease are hyponatremia
(Stout and Yu 1997, Roig et al. 1994, EPA 1985) and elevated levels of serum transaminase or
transpeptidase enzymes (Edelstein 1993, EPA 1985). Edelstein (1993), however, reported that in only
one of four prospective studies, patients with legionnaires' disease showed an increased incidence of
hyponatremia compared to patients with pneumonia caused by some other agent. Furthermore,
hyponatremia was observed in only about 20 percent of the patients with legionnaires' disease in these
studies. Edelstein also reported that only one of four prospective studies showed an increased incidence
of elevated serum enzyme levels in patients with legionnaires' disease compared to patients with
V-4
-------
pneumonia caused by some other agent. One study indicated the converse, and two indicated no
difference. Edelstein noted that "the most reasonable conclusion is that nonspecific test results cannot be
used to clearly distinguish between those with or without legionnaires' disease."
Table V-3. Common Clinical Laboratory Findings in Patients with Legionnaires' Disease :
Leukocytosis Serum glutamic oxaloacetic transaminase
Hyponatremia Serum glutamic pyruvic transaminase
Hypophosphatemia Serum lactic dehydrogenase
Proteinuria Alkaline phosphatase
Hematuria Creatine phosphokinase
Liver function abnormalities
1 Information was compiled from the following sources: Ampel and Wing 1990, Muder et al. 1989, Strampferand Tu
1988, Ching and Meyer 1987.
Although investigators appear to agree that no one clinical feature or laboratory finding
distinguishes legionnaires' disease, some have recently reported that a diagnosis of legionnaires' disease
may be made using a multifactorial clinical model (Breiman andButler 1998, Cunha 1998). Cunha
(1998) noted that one problem is that the "literature does not address the diagnostic significance of
characteristic signs and symptoms in concert." He recently reported a weighted point evaluation system
to aid physicians in the diagnosis of legionnaires' disease.
The majority of patients with legionnaires' disease exhibit abnormalities in the chest radiograph
(Muder et al. 1989). Although "all types of roentgenographic patterns are seen in cases of legionnaires'
disease" (Edelstein 1993), unilateral alveolar infiltrates, which may be segmental, lobar, or diffuse, are
typically observed in the early stages of the disease (Muder et al. 1989). These infiltrates may enlarge
and consolidate as the disease progresses (Ampel and Wing 1990). Pleural effusion is typically observed
in one-third of patients with legionnaires' disease (Ampel and Wing 1990, Stout and Yu 1997). The
frequency, however, ranges from 6 to 63 percent and, therefore, is not a distinguishing feature (Edelstein
1993). Nodular opacities and cavitation are uncommon, except in immunocompromised patients
(Strampfer and Tu 1988, Muder et al. 1989, Stout and Yu 1997). Radiographic progression can occur
even with appropriate antibiotic therapy, and resolution is typically slow (i.e., may require one to four
V-5
-------
months) (Muder et al. 1989, Stout and Yu 1997). Similar to clinical laboratory findings, clinicians have
concluded that "no characteristic radiographic pattern helps to distinguish any one type of pneumonia"
(Colette and Fein 1998).
C. Mechanism of Action
The typical progression of a Legionella infection can be characterized by the following steps
(Cianciotto et al. 1989). First, Legionella is inhaled or instilled into the lower airways of the lungs. The
mechanism for evasion of the body's non-specific defenses has not been established. Second, alveolar
macrophages phagocytize the bacteria by either a conventional or coiling mechanism. The resulting
phagosome becomes studded with ribosomes within four to six hours. Intracellular survival of the
bacteria maybe attributed to one or more of the following factors: reduced oxidative burst, failure of
phagosome to acidify, failure of phagosome to fuse with lysosome, and/or bacterial resistance to
lysosomal contents. Third, the bacteria undergo rapid intracellular growth. Si fact, the bacterial growth
within infected macrophages has been estimated at 100- to 1000-fold wilhin 48 to 72 hours of infection,
which is considered remarkable compared to that of other intracellular opportunistic bacteria (e.g.,
Salmonella., Mycobacterium, Listeria) (Friedman et al. 1998). Finally, the host cell dies and releases the
bacteria. Intracellular infection and bacterial growth is then escalated. Atthis stage, tissue damage and
induction of an inflammatory response may occur as a result of exposure to bacterial cellular
components and/or extracellular products from the bacteria.
Significant effort has been invested into elucidating the factors responsible for the pathogenesis
of Legionella. The 1985 Legionella Criteria Document described three "toxic" bacterial components: a
lipopolysaccharide (LPS) located in the outer membrane of Legionella bacteria, an extracellular acid-
soluble toxin isolated from several Legionella species, and an extracellular cytotoxin isolated fromZ.
pneumophila. A variety of proteolytic enzymes were also recognized as potentially important factors in
the pathogenesis of Legionella. Since completion of the 1985 Legionella Criteria Document, a variety of
cellular components and extracellular products have been identified (Rechnitzer 1994). Their
involvement in the pathogenesis of Legionella, however, has not been established.
V-6
-------
One biologically important extracellular bacterial component that has been isolated is azinc
metalloprotease. This enzyme has exhibited proteolytic, hemolytic, and cytotoxic activity and has
received a variety of names (e.g., tissue-destructive protein (TDP), major secretory protein, cytolysin)
(Rechnitzer 1994). This enzyme has been shown to elicit the same type of pulmonary lesions that
develop in legionnaires' disease and has been found in lungs of guinea pigs infected with/,, pneumophila
at a level equal to the dose of the purified protease known to cause death in experimental animals
(Conlan et al. 1988). This enzyme has also been shown to degrade two phosphorylated proteins
generated by a phosphokinase system isolated from the pulmonary cells of rabbits (Belyi 1990).
Although the significance of this specific system is unknown, phosphokinase systems generally are
involved in controlling intracellular metabolic processes. Therefore, this enzyme may disrupt metabolic
processes of the host cell in addition to causing necrosis.
One additional factor recently recognized that may contribute to the pathogenesis ofLegionella is
the symbiotic relationship of the bacteria with amoebae. Brieland et al. (1996) investigated the effect of
intratracheal coinoculation of L. pneumophila and Hartmannella vermiformis into A/J mice. A/J mice
are recognized as an animal model for human legionnaires' disease and have been used extensively to
investigate many aspects ofLegionella infections. H. vermiformis is "the most prevalent species of
amoebae in potable water supplies in the United States and has been epidemiologically linked to
outbreaks of legionnaires' disease." They found that coinoculation resulted in significantly increased
intrapulmonary growth of L. pneumophila, an increased severity of infection, and significant mortality
when compared to inoculation with only L. pneumophila. Furthermore, they found that coinoculation
with L. pneumophila and H. vermiformis into a resistant host (i.e., BALB/c mice) resulted in an eight-
fold increase in intrapulmonary bacterial growth when compared to inoculation with only L.
pneumophila.
To confirm that the amoebae were providing a niche for bacterial replication, Brieland et al.
(1997a) investigated the effect of coinoculation of A/J mice with//, vermiformis and mutant strains of L.
pneumophila that had reduced virulence for H. vermiformis. They found that the intrapulmonary
bacterial growth was not significantly increased in mice coinoculated with//, vermiformis and the
mutant strains. The authors concluded that virulence for the amoebae is necessary for increased bacterial
V-7
-------
growth and, therefore, "inhaled amoebae may potentiate intrapulmonary growth of L. pneumophila by
providing a niche for bacterial replication."
As a final note, Brieland et al. (1997b) investigated the effect of inoculating A/J mice with//.
vermiformis infected withZ. pneumophila. They found that the infected amoebae were more pathogenic
than an equal number of L. pneumophila or a mixture of L. pneumophila and uninfected amoebae. The
authors concluded that amoebae infected withZ. pneumophila may be the infectious particles in
Legionella infections.
D. Immunity
Both humoral and cell-mediated immune responses to Legionella infections have been
documented (EPA 1985, Friedman et al. 1998). Based on the results of serological tests, antibodies are
produced in response ioLegionella infections that interact with specific bacterial components.
Furthermore, bacterial infection results in the activation of complement, which has been shown to occur
through both the classical and alternative pathway (Friedman et al. 1998, Mintz et al. 1992).
Opsonization of bacteria (i.e., binding of antibodies and/or complement to the bacteria) has been shown
to increase phagocytosis by human peripheral blood monocytes and animal macrophages; however, the
ability of the bacteria to replicate within these cells does not appear to be diminished (Friedman et al.
1988). Therefore, the protection provided by specific antibodies in vivo is not currently known.
Cell-mediated immunity is recognized as the primary defense to Legionella infection (Susa et al.
1998). Research indicates that cytokines secreted by TH1 helper cells or macrophages play a primary
role in limiting bacterial replication (Friedman et al. 1998). For example, interleukin-2, which is
secreted by TH1 helper cells, appears to activate natural killer cells to lyse cells infected with Legionella.,
thus limiting bacterial growth by killing the host cell (Friedman et al. 1998).
Interferon- is also an essential component in host resistance to Legionella infection. This
cytokine, which is also secreted by TH1 helper cells, appears to activate macrophages and monocytes to
inhibit bacterial growth. In fact, bacterial growth has been shown to decrease 100-fold in activated
macrophages compared to non-activated infected macrophages (Friedman et al. 1998). The limited
V-8
-------
growth may be the result of down-regulation of the transferrin receptors, which results in a decreased
availability of intracellular iron, an essential component m Legionella growth (Skerrett and Martin 1991,
Susaetal. 1998).
Many studies have confirmed the important role of interferon- in the resistance to Legionella
infection. For example, based on a comparison of aged mice to young mice, Fujio et al. (1995) proposed
that the susceptibility of the elderly to Legionella infection may be the result of a decreased capacity to
produce interferon- . Finally, Heath et al. (1996b) investigated the effect of Legionella infection in
BALB/c mice (i.e., a resistant species) and in BALB/c mice in which the interferon- gene was
disrupted. Mice were inoculated intratracheally withZ. pneumophila. Bacterial growth was not
observed in the BALB/c mice; however, the mutant BALB/c mice developed "persistent, replicative
intrapulmonary L. pneumophila infections with extrapulmonary dissemination of the bacteria to the
spleen." Intratracheal administration of interferon- to the mutant BALB/c mice increased clearance of
the bacteria from the lungs. The authors concluded that these results confirm the importance of
interferon- in the resistance to Legionella infection.
One additional factor that appears to play an important role in resistance to Legionella infection
is tumor necrosis factor- , which is a cytokine secreted by macrophages. Blanchard et al. (1988)
reported that polymorphonuclear leukocytes treated with tumor necrosis factor- exhibited increased
bactericidal activity against L. pneumophila. Furthermore, mice treated with tumor necrosis factor-
prior to infection exhibited reduced mortality, which correlated with increased clearance of bacteria from
the lungs. Matsiota-Bernard et al. (1993) reported that treatment of human peripheral monocytes with
tumor necrosis factor- significantly inhibited the growth of L. pneumophila. Inhibition was not
observed when an inhibitor of tumor necrosis factor production or anti-tumor necrosis factor antibodies
were added to the culture medium. The mechanism by which tumor necrosis factor- inhibits bacterial
growth has not been established, although one proposal is that this cytokine may potentiate nitric oxide
release (Susa et al. 1998, Skerrett and Martin 1996).
E. Chronic Conditions
V-9
-------
As discussed in the 1985 Legionella Criteria Document, most patients with legionnaires' disease
recover without chronic manifestations. Ching and Meyer (1987), however, reported that fatigue and
weakness may persist for several months following treatment. Furthermore, as noted above, resolution
of infiltrates on chest radiographs is slow and may take from one to four months (Stout and Yu 1997).
Respiratory abnormalities resulting from legionnaires' disease occasionally occur. Gea et al.
(1988) reported the outcome of 11 patients with legionnaires' disease followed for 53 months. Mild to
moderate ventilatory and/or gas exchange abnormalities were observed several months following
discharge from the hospital. At study termination, the majority of patients (8/11) exhibited one or more
of the following respiratory abnormalities: a restrictive ventilatory defect, a low transfer factor, and/or
hypoxemia. Because the majority of patients were smokers, some with chronic bronchitis, the authors
could not dismiss the possibility that these manifestations were the result of pre-existing conditions.
More serious respiratory abnormalities are rare. Pulmonary pathology that has been reported
includes pulmonary fibrosis, bronchiolitis obliterans, chronic vasculitis, and chronic organizing pleuritis
(Ching and Meyer 1987, EPA 1985).
F. Treatment
Early initiation of appropriate treatment is recognized today as crucial for a successful outcome
of legionnaires' disease. Heath et al. (1996a) conducted a retrospective analysis of serologically
confirmed cases of legionnaires' disease to determine factors associated with increased mortality. After
multiple logistic regression analysis, the only factor associated with increased mortality was a delay in
initiation of appropriate therapy.
Retrospective analyses of early epidemics of legionnaires' disease have helped define
"appropriate" therapy. Early studies indicated that patients treated with erythromycin had a lower
mortality rate than patients treated with aminoglycosides, -lactam antibiotics, orchloramphenicol (6%
versus 30-40%) (Roig et al. 1993). The poor response to these antibiotics has been related to their
inability to penetrate phagocytic cells. In fact, studies have indicated that clinically effective antibiotics
V-10
-------
must have the following features: (1) superior in vitro activity against Legionella species, (2) "the ability
to enter and concentrate within phagocytic cells," and (3) "the ability to achieve high concentrations in
lung tissue and alveolar exudate" (Roig et al. 1993).
Current recommendations for antibiotic treatment of legionnaires' disease are provided in Table
V-4. Erythromycin has historically been considered the first choice in treatment of legionnaires' disease
(Stout and Yu 1997). Treatment with this antibiotic, however, is associated with several adverse side
effects, including transient hearing loss, phlebitis, gastrointestinal intolerance, and, more rarely,
ventricular arrhythmia (Roig et al. 1993). Newer macrolides, such as azithromycin and clarithromycin,
are attractive because they have exhibited superior activity against Legionella species and greater
intracellular penetration with potentially fewer adverse effects (Klein and Cunha 1998, Stout and Yu
1997, Roig et al. 1993). With development of intravenous formulations, these newer macrolides (e.g.,
azithromycin) may replace erythromycin as the treatment of choice (Stout and Yu 1997).
Quinolones have shown greater activity against Legionella species and higher intracellular
penetration than the macrolides (Klein and Cunha 1998, Stout and Yu 1997, Edelstein et al. 1996).
These antibiotics have been recommended for transplant recipients with legionnaires' disease because,
unlike the macrolides and rifampicin, they do not interfere with metabolism of immunosuppressive
medications (Stout and Yu 1997). Although successful outcomes have been reported using these
antibiotics, Baty et al. (1997) reported a case of pneumoniain an immunocompetent patient resulting
from L. jordanis that was unresponsive to ciprofloxacin and ofloxacin. These antibiotics, however, were
administered at dose levels lower than those suggested in Table V-4.
Other antibiotics that have shown variable success in treatment of legionnaires' disease include
the tetracyclines (e.g., doxycycline, minocycline, and tetracycline) and the combination of trimethoprim-
sulfamethoxazole (Stout and Yu 1997, Roig et al. 1993). Rifampicin is an antibiotic
Table V-4. Recommendations for Antibiotic Treatment of Legionnaires' Disease1
Antibiotic
Dose
Route
Frequency
Macrolides
Azithromycin
500 mg
oral or intravenous
every 24 hours
V-ll
-------
Antibiotic
Clarithromycin
Erythromycin
Dose
500 mg
1000 mg
500 mg
Route
oral or intravenous3
intravenous
oral
Frequency
every 12 hours
every 6 hours
every 6 hours
Quinolones
Levofloxacin
Ciprofloxacin
Ofloxacin
500 mg2
400 mg
750 mg
400 mg
oral or intravenous
intravenous
oral
oral or intravenous
every 24 hours
every 8 hours
every 12 hours
every 12 hours
Tetracyclines
Doxycycline
Minocycline
Tetracycline
Trimethoprim-
sulfametho xazole
100 mg2
100 mg2
500 mg
160 mg/800 mg
160 mg/800 mg
oral or intravenous
oral or intravenous
oral or intravenous
intravenous
oral
every 12 hours
every 12 hours
every 6 hours
every 8 hours
every 12 hours
1 Source of information is Stout and Yu 1997. Recommendations are based on clinical experience rather than controlled trials.
Specific recommendations may vary slightly depending on the source of information.
2 Doubling of the first dose was recommended by Stout and Yu (1997). Edelstein (1998), however, does not recommend this
suggested practice forazithromycin and levofloxacin.
3 Intravenous route is under investigation in United States.
that is used in combination therapy for severely ill patients and is typically administered either orally or
intravenously at dose levels of 300-600 mg every 12 hours (Stout and Yu 1997). Although rifampicin
has shown excellent in vitro and in vivo activity against Legionella species, it is not administered as a
monotherapy due to the potential of developing rifampicin-resistant strains of Legionella (Roig et al.
1993). Rifampin has been combined with many antibiotics, but some uncertainty exists regarding the
clinical efficacy of rifampicin and quinolone combinations (Roig et al. 1993, Edelstein et al. 1993).
Furthermore, the clinical efficacy of the conventional combination of rifampicin and erythromycin has
been questioned. Hubbard et al. (1993) conducted a retrospective analysis of patients with legionnaires'
disease requiring intermittent positive pressure ventilation. Those patients receiving rifampicin in
combination with erythromycin had a significantly increased incidence of jaundice, had significantly
higher levels of bilirubin, and did not have decreased mortality compared to those patients that did not
receive rifampicin.
For the treatment of legionnaires' disease, the preferred route of administration of any antibiotic
therapy is intravenous (Stout and Yu 1997). This route ensures the greatest potential concentration of
V-12
-------
antibiotic in the lung tissue. Intravenous treatment should continue until the patient becomes afebrile.
At this point, intravenous treatment can be replaced by oral therapy. The total duration of therapy
depends on the patient history. For a patient with a mild illness exhibiting significant improvement,
therapy should continue for a period of approximately two weeks. For the severely ill or
immunocompromised patient, therapy should continue for three weeks. Newer macrolides (e.g.,
azithromycin) may allow for a shorter course of treatment.
G. Summary
Since publication of the 1985 Legionella Criteria Document, much has been learned regarding
legionnaires' disease in humans. Although its progression has been more precisely defined, no one
symptom has been recognized that can distinguish legionnaires' disease from other bacterial
pneumonias. Similarly, abnormalities in standard clinical laboratory tests and chest radiographs cannot
be used to distinguish legionnaires' disease from other pneumonias. Some investigators, however, have
recently reported that a multifactorial clinical approach may be helpful in the diagnosis of legionnaires'
disease.
Although extrapulmonary diseases resulting from legionnaires' disease are still relatively rare,
they have been reported with increasing frequency since publication of the 1985 Legionella Criteria
Document. The kidney is a common site of extrapulmonary infection; however, the heart is now
recognized as the most common site of extrapulmonary infection. These extrapulmonary infections can
occur in the absence of pneumonia. No significant new information has been located on chronic
conditions resulting from legionnaires' disease.
Since 1985, the mechanism of bacterial replication has been more precisely defined. Briefly,
bacteria are inhaled or instilled in the lower airways of the lung and are phagocytized by alveolar
macrophages. Bacteria undergo rapid intracellular growth within the phagosome. The host cell lyses
and releases the bacteria, which escalates the bacterial infection.
Significant effort has been invested into the elucidation of factors responsible for the
pathogenesis of Legionella. One important discovery was the isolation of a zinc metalloprotease, an
V-13
-------
enzyme that elicits pulmonary lesions similar to those that develop in legionnaires' disease. Although
not a bacterial component or product, one factor that may affect the pathogenesis oi Legionella is their
ability to infect amoebae. Recent research suggests that Hartmannella vermiformis may provide a niche
for bacterial replication in the lungs. In fact, one study suggests that amoebae infected withZ.
pneumophila may be responsible for bacterial infection.
Recent research has continued to document that both humoral and cell-mediated immune
responses to Legionella infection occur. Although specific antibodies are produced, the protection that
these antibodies provide in vivo is still unknown. Cell-mediated immunity is currently recognized as the
primary defense against Legionella infection. Research also has emphasized the importance of specific
cytokines (e.g., interferon- , tumor necrosis factor- ) in host resistance to Legionella infection. Much
more research is needed to understand the host's mechanisms of resistance to these bacteria.
Since publication of the 1985 Legionella Criteria Document, many advancements in the
treatment of legionnaires' disease have been made. Although erythromycin has historically been
considered the first choice for the treatment of legionnaires' disease, newer macrolides
(e.g., azithromycin) are available that exhibit superior activity to Legionella and greater intracellular
penetration with potentially fewer adverse effects. Furthermore, quinolones show promising activity
against Legione lla infections and are recommended for patients on immunosuppressive medication.
Early initiation of appropriate therapy is crucial for a successful outcome to legionnaires' disease.
V-14
-------
VI. Risk Assessment
Risk assessment is a tool for the synthesis of available scientific information, in both a qualitative
and quantitative manner, in order to characterize the probability of potential public health hazards
resulting from exposure to a toxic or infectious agent. The results of such an assessment can then be
employed in making informed risk management decisions. Over the past 25 years, scientists have
developed methodologies to assess risks to human health from exposure to chemicals in the
environment, foods, or drugs. The application of this methodology to the assessment of risks from
microbial pathogens is a much newer field, however. This chapter presents the relevant information,
where available, for a risk assessment oi Legionella in water supplies.
A. Hazard Identification
As discussed in the preceding chapter as well as in the 1985 Legionella Criteria Document,
Legionella are opportunistic pathogens that cause a pneumonic condition known as legionnaires' disease
in some individuals. Outbreaks and sporadic cases have been reported following exposure in the general
community and among hospitalized persons (i.e., nosocomial cases). Legionella are considered
opportunistic pathogens because, although they are highly prevalent in the environment, relatively few
people develop a clinical infection. Yu and colleagues (1993) characterized the attack rate for
Legionella as "strikingly low."
Knowledge gained from advances in laboratory identification techniques and more rigorous
epidemiological studies suggests that Legionella are responsible fora growing percentage of both
community- and hospital-acquired pneumonias. These advances have allowed a better understanding of
the relative impact of Legionella-caused pneumonia in the U.S. From a review of pneumonia patients in
Ohio, Marston and colleagues (1997) estimated that between 8,000 and 18,000 (2-4 %) of the total
485,000 community-acquired cases of pneumonia requiring hospitalization annually in the U.S. are due
to Legionella. This estimate is associated with significant uncertainty, however, because the causative
organism is identified in only 50 percent of pneumonia cases (Reynolds 1996, Marrie et al. 1996).
VI-1
-------
Legionnaires' disease is the most serious illness caused by Legionella organisms. The clinical
course of this disease, which was described in detail in Chapter V, is quite similar for community- or
hospital-acquired infections (Petro-Botet et al. 1995). Other infections caused by Legionella are self-
limiting (e.g., Pontiac fever) or are much more rare (e.g., infection of surgical incisions or other wounds)
(Lowry and Tompkins 1993). Therefore, risk assessment of this organism is focused on legionnaires'
disease as the endpoint of concern.
B. Dose-Response Information
The 1985 Legionella Criteria Document noted that quantitative data on the infectivity of
Legionella in humans had not been reported. Unfortunately, sufficient information is still not available
to support a quantitative characterization of the threshold infective dose (i.e., the dose required to
produce infection) of Legionella. Animal models show a great interspecies variation in susceptibility to
infection with Legionella, as described in Chapter IV. Due to the potentially serious health effects,
experiments to identify the infective dose in humans are not possible. Legionella are opportunistic
pathogens that replicate within host cells, reach target tissue via several routes (primarily inhalation or
aspiration), and exhibit a very low attack rate or virulence in the general population; therefore, it is not
surprising that definitive dose-response information continues to be elusive.
VI-2
-------
C. Potential for Human Exposure to Legionella
1. Prevalence of Legionella in the Environment
As discussed in Chapter HI and in the 1985 Legionella Criteria Document, there is clear
consensus that Legionella bacteria are widely distributed in the environment, especially in treated or
potable water supplies. Important niches or reservoirs for Legionella can occur within treated water
supply systems due to their ability to form symbiotic relationships with other microorganisms (including
biofilm formation) and their subsequent resistance to standard disinfection techniques (e.g.,
chlorination). Since 1985, there have been numerous studies documenting the presence of Legionella in
potable water and in water distribution systems of all types of large buildings including hospitals, office
buildings and hotels, and also smaller buildings and family residences (see Chapter HI). In some cases,
Legionella occur in the absence of any reported cases of legionnaires' disease (Oppenheim et al. 1987,
Stout et al. 1992b). Through the combination of environmental sampling studies with laboratory and
epidemiological findings, a better understanding has been gained for the relative importance of various
reservoirs for transmission of Legionella to humans.
2. Mode of Transmission to Humans
Given the widespread prevalence of Legionella in the environment and their niches within
reservoirs of water supply systems, it is important to have an understanding of the circumstances under
which Legionella bacteria can reach the lower respiratory tract of humans and potentially cause serious
disease. Based on such knowledge of the important reservoirs and routes of transmission of Legionella
to humans, the most appropriate preventive measures can be selected.
Legionella are transmitted directly from the environment to humans. There is very little, if any,
evidence of human-to-human transmission, and there is no evidence of any animal reservoirs with
public health relevance for this organism. In the past decade, considerable interest and controversy have
been focused on the mechanisms by which Legionella bacteria reach the lower respiratory tract, where
they are engulfed by alveolar macrophages and commence the pathological process of infection. One
route is the inhalation of an aerosol containing respirable droplets of water (or other liquids)
VI-3
-------
contaminated with Legionella. Alternatively, Legionella may be deposited in upper airways and
subsequently aspirated into the deeper portions of the lung. As mentioned in Chapter IV and in the 1985
Legionella Criteria Document, infection in animals following ingestion of Legionella has also been
demonstrated experimentally.
At the time of the 1985 Legionella Criteria Document, scientists believed transmission of
Legionella in the community or hospital setting occurred primarily via inhalation of infectious aerosols;
however, this assumption was based, for the most part, on circumstantial evidence. Now, with advances
in laboratory identification techniques and the availability of more rigorous epidemiological and
experimental data, there is increasing emphasis on the role of aspiration as a route of transmission.
Thus, it follows that there is an increased focus on potable water as a primary source of infection.
Environmental sampling from outbreaks (in communities or in hospital s) most frequently has implicated
potable water as the source (Stout and Yu 1997).
The 1985 Legionella Criteria Document reported that the most common reservoirs of
transmission for community-acquired Legionella infection are aerosols from: heat-rejection equipment
(cooling towers, evaporative condensers, steam turbine cleaning), components of plumbing systems
(showers, faucets, hot water tanks), nebulizers, humidifiers, whirlpool spas, or public fountains. Less
common sources include: ingestion of potable water, immersion in raw water, inhalation of
contaminated oil/water mixtures, and excavations (dust or soil) (EPA 1985). Additional types of aerosol
generators (e.g., grocery store mist machines) have been linked to outbreaks of legionnaires' disease
(Mahoney et al. 1992). No additional categories of sources have been identified during the period since
the 1985 EPA report, but the relative importance of contributions from some of these sources has
shifted. Although cooling towers are still a source of some community-acquired cases (eg., Castellani
Pastoris et al. 1997, Bhopal and Fallen 1988), potable water (with subsequent inhalation or aspiration of
aerosols) is acknowledged as a much more important source (Stout and Yu 1997, Neill et al. 1985).
There are still insufficient data to support quantification of the relative contributions from these various
sources (Bhopal 1995).
One of the most interesting and important advances made recently in the study of Legionella
transmission concerns the role of amoebae and other larger protozoa in enabling or enhancing the
VI-4
-------
transmission of Legionella. Working with Acanthamoeba in culture with L. pneumophila (isolated from
a cooling tower), Berk et al. (1998) examined the Legionella-fi\\ed vesicles formed and expelled by the
amoebae. The vesicles were 2.1-6.4 m in diameter (i.e., respirable size), and the study authors
calculated that, based on volumes, each vesicle could contain as many as 200 bacteria. Other
investigators have estimated even higher numbers of bacteria per vesicle (Rowbotham et al. 1986 as
cited in Berk et al. 1998). Berk and colleagues also demonstrated that vesicles free in the medium were
resistant to biocide and that the biocide treatment facilitated the release of large numbers of vesicles as it
induced encystment of the amoebae. Such infectious vesicles may represent a very important vehicle of
transmission for Legionella, by protecting the bacteria from dessication while in the atmosphere and
delivering a possibly infective dose to the respiratory tract. Thus, these preliminary findings contribute
to the complexity of modeling exposure and dose-response relationships for Legionella infections in
humans.
For nosocomial cases of legionnaires' disease, there is a growing body of evidence from case
observation and experimental data that points to aerosolization of potable water (tap water) as the most
important source of transmission of Legionella. Blatt et al. (1993) analyzed 14 nosocomial cases
that occurred in a military hospital and compared them with controls. Environmental sampling for
Legionella showed colonization of 15% of potable water sites, one hot water tank, and the groundwater
supply to the hospital, while no Legionella were isolated from the hospital cooling towers, building air
intakes or other hospital air and oxygen supplies. This case-control comparison showed a negative
association between showering and acquiring legionnaires' disease, although earlier studies have
sometimes reported a positive association with showering (EPA 1985, Breiman et al. 1990, Hanrahan et
al. 1987 as cited in Blatt et al. 1993).
Potable water is now consistently identified as the most common source of Legionella in
hospitals (Yu 1993, Blatt et al. 1993, Woo et al. 1992). Observation of hospital cases indicates a high
risk of infection for patients who have received ventilation support or have been exposed to respiratory
equipment (e.g., nebulizers), suggesting a major role for aspiration as a route of transmission for
hospital-acquired legionnaires' disease. The relative significance of aspiration is also supported by the
very low infection rates (or antibody liters) among hospital personnel where nosocomial Legionella
outbreaks have occurred (Marrie et al. 1986).
VI-5
-------
The sources of Legionella that play an important role in transmitting the bacteria to humans have
been fairly well characterized by now. Knowledge gaps exist, however, regarding the relationship
between environmental concentrations of Legionella and the ultimate risk of infection in exposed
individuals. It is, therefore, useful to review the factors that may place an individual at increased risk for
developing legionnaires' disease.
D. Risk Factors
For opportunistic pathogens such as Legionella bacteria, identification of risk factors in
susceptible individuals is an essential element forthe selection of the most appropriate control and
prevention measures. Based on the very low attack rates associated with this organism, it is clear that
the general U. S. population is quite resistant to infection by Legionella.
Certain patient populations are clearly at increased risk for contracting nosocomial legionnaires'
disease. These populations include patients who require intubation, patients who have received
ventilation assistance (including patients who have undergone surgery), and patients receiving
respiratory therapy with potentially contaminated medical equipment or whose care includes the use of
aerosol generators such as humidifiers ornebulizers (England etal. 1981, Marston et al. 1994, Stout and
Yu 1997).
Certain demographic factors are associated with an increased susceptibility to legionnaires'
disease following exposure. Subpopulations at increased risk include men over the age of 50, heavy
smokers, and heavy drinkers (Bhopal 1995, Marston et al. 1994, England et al. 1981). These findings
are based on analyses of very large series of legionnaires' disease cases. For example, Marston and
colleagues (1994) reviewed data for 3,254 patients whose cases were reported to the CDC between 1980
and 1989. The findings reported by England et al. (1981) represent the first 1,000 confirmed cases of
legionnaires' disease reported to the CDC (through September 1979).
People with certain underlying health conditions also have a significantly increased risk of
contracting legionnaires' disease. Such medical conditions include: chronic obstructive pulmonary
disease, diabetes, head or neck cancer, other malignancy, or end-stage renal disease. In addition, any
VI-6
-------
disease state (e.g., AIDS) or medical treatment (e.g., drugs such as corticosteroids or cancer
chemotherapy, or procedures such as hemodialysis) that suppresses or depletes a patient's immune
system can lead to an increased susceptibility to opportunistic infections such as legionnaires' disease.
Several patient populations (e.g., renal transplant patients, especially those requiring hemodialysis) are at
an extremely high risk for legionnaires' disease, as they have both an increased risk of exposure (via their
surgery and other ventilation needs), and an increased susceptibility (due to corticosteroid therapy and
dialysis) (Woo et al. 1986, LeSaux et al. 1989).
Many of these risk factors contribute not only to increased incidence of legionnaires' disease
among these groups, but also increased severity of the disease and increased mortality. Marston and
colleagues (1994) found that, among 3,254 legionnaires' disease cases reported to the CDC Legionella
surveillance system between 1980 and 1989, the following factors were significantly associated with
increased mortality attributed to legionnaires' disease: the use of steroids or other immunosuppressive
drugs; the presence of cancer, diabetes, or renal disease requiring dialysis; hospital-acquired infection;
older age; male gender; isolation of L. pneumophila subgroup 6 (Lp6); or isolation of more than one
Legionella species or L. pneumophila serogroup. More severe legionnaires' disease has also been
documented in smaller series of cases among bone marrow transplant patients (Harrington et al. 1996)
and patients receiving immunosuppressive drugs (with or without chronic disease) (Pedro-Botet et al.
1998).
People immunocompromised due to HIV infection are also at risk of developing more severe
legionnaires' disease, but Legionella infections (in the absence of other pneumonia-causing pathogens) in
this population are relatively rare. This may be due, in part, to exposure to other more common (and
more virulent) pathogens in the environment and, in part, to increased infection control vigilance
(including concern for waterborne pathogens) when patients with AIDS are hospitalized. Marston and
colleagues at the CDC (1994) reported an increased prevalence of legionnaires' disease among AIDS
patients compared to the general U.S. population (8 people with AIDS among 2,575 legionnaires' disease
cases; 0.19 expected). Bangsborg et al. (1990) reported that among 180 AIDS patients with pneumonia,
only six had Legionella infection, but four of these six patients were also infected with the fungus
Pneumocystis carinii. A high rate of coexistent pulmonary infection (again, with P. carinii) was also
reported by Blatt et al. (1994): among seven HIV-infected individuals who had legionnaires' disease, six
VI-7
-------
were also infected with P. carinii. Of bacterial pneumonias reported in persons with AIDS, other
species that are more pathogenic and hardier than Legionella are reported most frequently, including
Haemophilus influenzae, various Streptococci, andBranhamella catarrhalis (Chaisson 1998).
Another population that may be at increased risk of contracting Legionella infection is neonates,
due to their underdeveloped immune systems, intensive ventilation procedures, and corticosteroid
therapy. Nosocomial cases of legionnaires' disease have been reported, albeit infrequently, in this
population (Holmberg etal. 1993, Horie et al. 1992). Older infants and children who have the risk
factors identified for adult populations (e.g., are receiving corticosteroid therapy or are undergoing
mechanical ventilation) are also at increased risk of contracting legionnaires' disease (Carlson et al.
1990). But even though pneumonia (all types/sources) is common in the general pediatric population,
reports of legionnaires' disease in otherwise healthy children is extremely rare (Abernathy-Carver et al.
1994, Carlson et al. 1990, Famiglietti et al. 1997).
E. Quantification of Potential Health Effects
The 1985 Legionella Criteria Document reported that our understanding of the mechanisms of
transmission of and infection by Legionella was inadequate for quantification of the potential health
effects or the development of specific recommendations for control (EPA 1985). Although
improvements in laboratory isolation and identification techniques for Legionella, along with important
experimental, epidemiological, and ecological study results, have greatly expanded our understanding of
Legionella infections in humans, the current state of the science still does not allow estimation of the
probability of the potential adverse health effects caused by Legionella. Estimation of the infective dose
(i.e., the dose required to produce infection) is necessary for completing a risk assessment of a microbial
pathogen. Legionella are opportunistic pathogens with widespread environmental occurrence and a very
low attack rate in the general population. Legionella survive and thrive inside vesicles after being
ingested by amoebae in water reservoirs, but much more information is needed on the implications of
this potential vehicle for enhanced transmission and infectivity. More complete information is also
needed concerning the conditions under which a population is likely to be exposed to the infective dose,
with models that accommodate aspiration and inhalation routes, as well as the variability introduced by
the bacteria's potential replication within host cells.
VI-8
-------
Despite deficiencies in understanding several of the factors necessary to determine the risk of
infection by Legionella, the current state of knowledge is sufficient to support specific recommendations
to control and prevent legionnaires' disease.
F. Minimizing Risk
In the 22-year period since the sentinel outbreak in Philadelphia of what is now known as
legionnaires' disease, a great deal of knowledge has been gained on the behavior and occurrence of
Legionella. Based on this knowledge, efforts to minimize the risks of Legionella infection have been
instituted, especially for the protection of susceptible or high-risk individuals.
Because there is little if any person-to-person transmission of Legionella and no vaccine is
available to prevent infection, risk minimization efforts are focused on breaking the chain of
transmission between environmental sources of Legionella and human hosts. Approaches used for
controlling the growth of Legionella in treated water, frequently used in combination, include heat,
chlorination, ultraviolet light, copper-silver ionization, and ozone treatment. These various treatment
options are detailed in Chapter VII of this report. For hospitals and other health care settings, regular
environmental surveys of both hot water systems and distal sites should be conducted; some health
departments have issued mandates for such testing (Allegheny County Health Department 1997). In
health care institutions, these environmental surveys can also serve to raise awareness and the index of
suspicion of health practitioners for consideration of Legionella as the causative agent in nosocomial
pneumonia cases (Yu 1997).
Active surveillance for Legionella infection, especially among hospital patients at highest risk of
acquiring nosocomial infection (i.e., transplant patients, immunocompromised patients, or patients with
certain chronic underlying health conditions) is also an important tool for minimizing risk of
legionnaires' disease because it allows for prompt remedial actions and rapid diagnosis and treatment of
confirmed cases. As discussed in Chapter V of this document, earlier treatment is associated with
reduced severity of disease and reduced mortality. Both the control measures and active surveillance for
cases can be expensive, however, and ultimately require cost-benefit decisions. Several recent
publications have outlined some of the important considerations in making such cost-benefit decisions.
VI-9
-------
Shelton and colleagues (1993) developed a system for determining when Legionella detection
may justify preventive or remedial actions. Analysis of samples from buildings with reported
legionnaires' disease outbreaks and with no reported cases (a total of 900 samples) revealed a strong
association between amplified Legionella levels and legionnaires' disease outbreaks. Their system
matches action levels (scale of 1 to 5) with specific environmental concentrations of viable Legionella
detected in three types of sources: cooling towers and evaporative condensers; potable water; and
humidifiers/foggers. The concentrations corresponding to the action levels vary depending on the
environmental source. For example, the highest level of concern (Hazard Level 5) corresponds to
Legionella concentrations above 1,000 organisms/mL in cooling towers, above 100/mL in potable water,
or above 10/mL in a humidifier/fogger. For Hazard Level 5, the authors recommend immediate
disinfection of equipment. Hazard Level 3 represents "a low but increased level of concern" and
corresponds to Legionella concentrations above 10/mL in cooling towers and above 1/mL in potable
water or a humidifier/fogger. The study authors noted, however, that among the 900 samples, some of
the samples rated Hazard Level 5 were obtained in buildings without any reported legionnaires' disease
cases. Such findings illustrate the lack of a direct relation between detected environmental levels of
Legionella and risk of disease. Nevertheless, preventive or remedial actions may be warranted when
Legionella concentrations exceed certain limits.
In 1997, the Centers for Disease Control (CDC) for the first time included information on
Legionella infections in their revised "Guidelines forNosocomial Pneumonia" (CDC 1997a). For
hospitals without any identified cases, the CDC outlined two primary prevention measures: (a) routine
culturing of the potable water system, with initiation of active surveillance (i.e., increasing the use and
availability of diagnostic laboratory tests for Legionella) when 30 percent or more of environmental
samples are positive for Legionella; or (b) utilizing diagnostic laboratory tests for high risk patients with
nosocomial pneumonia, with routine maintenance of potable water supplies (i.e., with sufficient heat and
chlorination), and initiation of an environmental investigation once one definite or two possible cases of
legionnaires' disease have been identified.
The CDC Guidelines also outline secondary prevention measures for hospitals where nosocomial
legionnaires' disease cases have been identified, with a caveat that full-scale environmental
investigations and decontamination measures may not always be indicated in all hospitals. The decision
VI-10
-------
to initiate such measures depends on the level of risk for infection and mortality from Legionella in the
given patient population. Nonetheless, the CDC also cautions that a low threshold for initiating an
environmental investigation may be appropriate because nosocomial cases have typically been
underdiagnosed, and additional recent or ongoing nosocomial cases typically are identified once several
cases have been confirmed. The Guidelines (CDC 1997) describe five important steps in conducting
such an environmental investigation: (1) review of medical records; (2) active surveillance to identify
recent or ongoing nosocomial cases; (3) identification of risk factors for infection and comparison of
cases and controls, through the collection of information on environmental exposures (e.g., showering or
the use of respiratory therapy equipment); (4) collection of water samples from the implicated sources
and other potential aerosol sources; and (5) subtype matching of patient isolates and the environmental
samples. Decontamination or replacement of the identified environmental sources must also take place.
Clearly, these secondary prevention measures can require extensive resources.
G. Summary
Given that legionnaires' disease is the most serious infection caused by Legionella, risk
assessment of these organisms should be focused on legionnaires' disease as the endpoint of concern.
Legionella are opportunistic pathogens with widespread distribution in the environment but a very low
rate of infection in the general population. The sources of transmission of Legionella to humans have
been well characterized, and almost all of these sources (with the exception of contaminated medical
equipment) involve the aerosolization of water contaminated with Legionella and subsequent inhalation
or aspiration. Potable water, especially in hospitals and other buildings with complex hot water systems,
is considered to be the most important source of Legionella transmission.
Despite many advances in laboratory isolation and identification techniques and the availability
of findings from recent epidemiological and experimental studies, the current state of the science does
not allow for quantification of the potential risks caused by Legionella in water supplies. Nevertheless,
important preventive (or remedial) actions have been identified that can minimize the risks of Legionella
infection, especially for the protection of high-risk or susceptible individuals.
VI-11
-------
(this page intentionally left blank)
VI-12
-------
VII. Analysis and Treatment
A. Analysis of Samples
Legionella can survive in a wide range of conditions including variable temperatures, pH-levels,
and dissolved oxygen concentrations. In addition, algae and other water bacteria can promote their
growth (Nguyen et al. 1991). Detection of Legionella contamination in potable water and plumbing
fixtures, as well as in biological samples, is a major concern, particularly of hospitals experiencing cases
of legionnaire's disease. The 1985 Legionella Criteria Document discusses collection (disassembly,
swabbing and scraping, centrifugation, filtration), isolation (culturing), and detection techniques (Direct
Fluorescent Antibody (DFA), Indirect Fluorescent Antibody (IFA), monoclonal antibodies, and
radioimmunoassay). However, more recent studies and additional data on the collection, isolation, and
detection of Legionella in both water and biological samples have been published and are described
below.
1. Collection of Legionella
Most outbreaks of legionellosis come from warm waters, as higher temperatures generally
stimulate the growth of these organisms. It is difficult to culture Legionella in waters below 20°C
(Colbourne et al. 1988). Test samples for Legionella typically come from anthropogenic sources such as
faucets, sink outlets, taps, filters, and showerheads, which are usually sampled by disassembling,
swabbing, and scraping to obtain Legionella-bearing debris or scale (Stout et al. 1992b, Helms et al.
1988, Stout and Yu 1997, Barbaree et al. 1987). As reported in the 1985 Legionella Criteria Document,
the most effective manner of obtaining the sample is by insertion of sterile cotton swabs into the interior
surface of the water source. Ta et al. (1995) found that swabbing recovered greater concentrations of
Legionella organisms than two other methods (water sampling before swabbing and water sampling after
swabbing), exposing an average of 30.2 CFU in each swab sample while the concentration of Legionella
in the water samples averaged 4.7 CFU. Swab sampling is also the preferred sampling method because
the swab is easier to transport and requires less processing time than straight water samples (Ta et al.
1995).
VI-13
-------
The study by Ta et al. (1995) also concluded that concentration of the water sample, either by
filtration or centrifugation, greatly improved the ability to detect Legionella in the samples. Filtration
was proven more effective than centrifugation, recovering 77 percent of the expected organism count
while centrifugation recovered only approximately 34 percent (Ta et al. 1995).
The 1985 Legionella Criteria Document provided summaries of several studies that used
filtration or centrifugation to concentrate Legionella, and recommended heat and acid wash treatment to
isolate Legionella from environmental specimens. Because Legionella can survive at high temperatures,
heating (at 60°C for 1-2 minutes) was found to reduce the strains of other bacteria contaminants (e.g.,
Pseudomonas aeruginosa) by 98 percent while leaving the Legionella unaffected.
Acid wash treatment is used to isolate Legionella because unlike most bacteria, Legionella
strains are acid resistant (Nguyen et al. 1991). Ta et al. (1995) showed that although acid buffer
treatment did not enhance the recovery of L. pneumophila bacteria, it was in fact required for an optimal
yield of other strains. A detailed procedure for isolation of Legionella from environmental water
samples by acid treatment was described by Bopp et al. (1981) and summarized in the 1985 Legionella
Criteria Document. Water samples were pretreated, either concentrated by centrifugation or not
concentrated, with anHCl-KCl buffer mixture at pH2.2 for periods of 5-60 minutes. The greatest
quantity of isolations were obtained by acid treatment of centrifuged samples for 5 minutes (Bopp et al.
1981). More current studies have shown that samples treated with acid for three minutes can minimize
the development of competing bacteria (Taet al. 1995).
Following the collection and pretreatment steps, the samples are plated onto appropriate media.
Legionella do not grow on standard culture media. They have complex nutritional requirements,
featuring an unusually high iron requirement. The 1985 Legionella Criteria Document described various
media that can be used for culturing Legionella including a charcoal yeast extract (CYE) medium, which
was improved to the buffered charcoal yeast extract (BCYE) that is presently used to successfully isolate
these organisms. This medium is ACES buffered charcoal yeast extract (BCYE) agar supplemented with
-ketoglutarate (BCYE ), a Krebs-cycle intermediate that is readily catabolized by these bacteria
(Edelstein 1987). An incubation period of two to six days ensues when Legionella are cultured on this
medium (Grimont 1986). The buffer maintains the pH within a range that is critical for Legionella
VI-14
-------
(around pH 6.9) while the -ketoglutarate stimulates growth. Growth is further enhanced by the
addition of L-cysteine, keto acids, and ferric ions. Antimicrobials such as glycine (inhibitor),
cefamandole, polymyxin B, vancomycin (antibacterials), and anisomycin and cyclohaxamide
(antifungals) are added to inhibit or prevent the overgrowth of contaminants (Nguyen et al. 1991).
Selective media containing dyes, glycine, vancomysin, and polymyxin (DGVP) is used for
environmental sampling (Lin et al. 1998).
2. Detection ofLegionella in Environmental and Biological Samples
An array of serological tests have been used for detecting Legionella in water, sputum, blood,
serum, and urine samples. Kohler (1986) reports that antigens can be detected in the urine of
approximately 80 percent of patients withZ. pneumophila serogroup 1 pneumonia, and that the
specificity of these assays is greater than 99 percent. Most tests are used on lower respiratory tract
secretions, specifically tissue specimens, bronchial and tracheal secretions, and sputum. Sputum
specimens are pretreated with acid and cultured on selective media, similar to the pretreatment of
environmental samples. Tracheal aspirate specimens on culture media can provide a yield of 90 percent
sensitivity (Nguyen et al. 1991).
The two main serologic tests performed on bacteria are direct and indirect fluorescent assays,
which are applicable to both environmental and clinical specimens. Fluorescent organic compounds are
attached to antibody molecules that are bound to a cell or tissue's surface antigens, and then these tags
are detected by a fluorescent microscope. In the direct method, the antibody against the organism is
fluorescent, while the indirect method has the fluorescent antibody detected against a nonfluorescent
antibody on the surface of the cell. The Direct Immunofluorescence Assay (DFA) and the Indirect
Immunofluorescence Assay (IFA), which were examined in the 1985 Legionella Criteria Document, are
described further below. Other serologic tests described in the 1985 Legionella Criteria Document and
discussed below are enzyme-linked immunosorbent assays, monoclonal antibodies, and
radioimmunoassay. Serologic tests that are currently being used for detection ofLegionella antigens or
antibodies, including Polymerase Chain Reaction, and nucleic acid and DNA probes, are also discussed
below. The serologic tests differ primarily in sensitivity, specificity, predictive value, and complexity.
VI-15
-------
Direct Immunofluorescence Assay (DFA)
The most common and rapid test for Legionella is the DFA. According to Nguyen et al. (1991),
the exhibited sensitivity of DFA tests ranges anywhere from 25 to 85 percent. Sputum, lung specimens,
and bronchial and tracheal secretions are excellent samples to test by the DFA method (Grimont 1986).
Kohler (1986) reports that as long as proper quality control exists, the specificity of DFA testing in
respiratory specimens is greater than 90 percent. However, he cites examples of DFA accuracy results
for serogroup 1 infections of 50 percent, 47 percent, 68 percent, and 33-47 percent, where sensitivity and
specificity were not ideal in testing respiratory specimens. There are also DFA tests that use species-
specific monoclonal antibodies that are particularly useful for lower respiratory tract samples and tissues.
The monoclonal reagent is optimal for detection of L. pneumophila in respiratory specimens due to its
ability to detect multiple serogroups of L. pneumophila and decreased non-specific fluorescence of the
specimen and other bacteria. It may not be useful in the detection of environmental specimens (Grimont
1986).
Indirect Immunofluorescence Assay (IFA)
Legionella bacteria, and antibodies in patient sera, are detected through IF A. Heat-fixed antigens
are commonly used in the United States, but formolized antigens, used primarily in Europe, are said to
actually be more specific than heat-fixed antigens. Because seroconversion only occurs after a rather
long time period in humans, the IF A test is often used in conjunction with other tests (Kohler 1985). A
series of serological tests are typically conducted to test for antibodies, and they are most often run in
conjunction with the IFA (Colbourne et al. 1988, Grimont 1986, Edelstein 1987, Kohler 1985, Ehret et
al. 1986, Kashuba and Ballow 1996).
Enzyme-Linked Immunoabsorbent Assays (ELISA)
Enzyme-Linked Immunosorbent Assays (ELISA), radioimmunoassay (RLA), and agglutination
assays have also been used to detect Legionella antibodies. These methods employ enzymes and
radioisotopes to detect antibody molecules. The ELISA method is used to detect Legionella antibodies
in patient sera, but it has also been used to detect Legionella antigens in urine. The RIA method has also
VI-16
-------
been used for the detection oi Legionella antigens in urine, but is no longer commercially available. The
agglutination method has been used to detect antibodies in serum and antigens in urine. These tests are
all extremely sensitive because radioactivity and enzyme reaction products can be measured in very
small amounts. Enzyme-linked methods are preferred over radioactively tagged methods of discovery to
eliminate the problem of radioactive material disposal even though a longer incubation period is required
for these types of tests. With ELKA, preheating of the specimen is required to avoid false positives.
Agglutination assays, which clump organisms to other particles, are simpler and faster assays that are
generally easier to perform than the others (Kohler 1986).
Monoclonal Antibody
Monoclonal antibody tests, tests with antibodies formed from a single clone of cells, have been
found to be more accurate than polyclonal tests due to the suppression of background fluorescence.
Also, false positive results from cross-reactivity with non-Legionella organisms are eliminated (Stout
and Yu 1997). Monoclonal antibody tests are effective due to their high specificity for a single antigenic
determinant. Monoclonal antibodies can be produced to react only with a particular species, or even
strain, of bacteria. According to Kohler (1986), numerous laboratories have asserted that antibodies for
L. pneumophila have been developed to be species-specific and even serogroup-specific. These
monoclonal antibodies can be aimed against subsets of a specific serogroup and then used for antigen
detection by ELISA (Kohler 1986).
Polymerase Chain Reaction (PCR)
The Polymerase Chain Reaction (PCR) test uses two disparate primers. One is specific for
Legionella species, and the other is specific for L. pneumophila only. The primers are specific for the
gene sequences of the 5S ribosomal RNA gene. The PCR was converted into a kit called the
EnviroAmp® Legionella Kit; however, the kit is no longer commercially available. PCR is a relatively
new method designed to rapidly multiply DNA target genes in a laboratory setting to yield detectable
quantities for testing. A study done by Fricker and Fricker (1995) compares this technique to standard
culture methods for water samples. The positive results of the PCR matched those of the culture in all
but 4 of 87 cases, where the PCR reaction was apparently inhibited. Generally, the PCR was found to be
VI-17
-------
a very useful screening test because it is both fast and accurate. The main problem with the PCR method
is that it identified several negative cultures as positive. This issue is being investigated, but it is either
due to false positive results, discovery of dead Legionella, or detection of viable but non-culturable
Legionella (Flicker and Flicker 1995).
Murdoch et al. (1996) studied the ability of PCR tests to detect Legionella DNA in urine and
serum samples of pneumonia patients. There was a 64 percent detection rate in the urine and/or serum
samples, with this figure rising to 73 percent if testing was done within four days of the onset of
symptoms.
Nucleic Acid and DNA Probes
Nucleic acid and DNA probes can also be used to detect Legionella. With these methods, probes
are marked with RNA or DNA sequences that are specific to a particular species or strain of bacteria.
Nucleic acid probes require that the nucleic acids of the bacteria become accessible and prepared to react
with the tagged probe. Detection using 1his method has been reported to be anywhere from 5 to 100
percent, withZ. pneumophila giving the highest values (Grimont 1986). According to Edelstein (1987),
a probe kit generally has a sensitivity of 75 percent and a specificity of 100 percent "if certain samples
are excluded from the analysis." Although it is more sensitive toL. pneumophila detection, it will still
quickly recognize all Legionella species (Edelstein 1987). This new detection method, however, as
reported by Nguyen et al. (1991), has yet to be clinically validated and is rather insensitive and costly.
Urinary Antigen
There is a commercially available enzyme immunoassay (EIA) test for the Legionella antigen in
urine from the company Binax, Inc. in Portland, Maine. Nguyen et al. (1991) reports that the test
exhibits 99 percent specificity and greater than 90 percent sensitivity and that it is relatively inexpensive.
The main drawback to this urinary antigen test is that it only detects antigens of L. pneumophila
serogroup 1. However, since this species accounts for upwards of 80 percent of all legionellosis
infections, this weakness is rather slight (Nguyen et al. 1991).
VI-18
-------
B. Disinfection as a Water Treatment Practice
Legionella are found in natural aquatic environments, artificial aquatic environments such as heat
rejection devices (cooling towers and evaporative condensers), and water distribution systems (Muraca
et al. 1990). Water distribution systems in hospitals, hotels, institutional buildings, and domestic homes,
as well as personal respiratory therapy equipment, freestanding room humidifiers in hospitals, industrial
cutting oil/water emulsions, and communally used whirlpools and spas have been shown to be reservoirs
for Legionella (World Health Organization 1990, Moreno 1997). Legionella colonization is promoted
by temperatures below 50°C (122°F), scale and sediment accumulation, stagnation (which prevents
disinfectant from reaching the bacteria), and design of the hot water tank (see Chapter III, Section F for
further discussion of factors affecting Legionella survival) (Muraca et al. 1990). The 1985 Legionella
Criteria Document indicated that Legionella surviving initial water treatment may colonize pipe joints,
cul-de-sacs, and corroded areas or adhere to the surface or sediment of storage tanks, especially those
constructed of wood (EPA 1985). New distribution systems may also be a source of Legionella
contamination; the 1985 document cited cases in which Legionella outbreaks have occurred in new
distribution systems (EPA 1985).
There are several control methods available for disinfection of water distribution systems. These
include thermal (superheat and flush), hyperchlorination, copper-silver ionization, ultraviolet light
sterilization, ozonation, and instantaneous steam heating systems. These disinfection methods are
discussed below. The use of heat, chlorine, ultraviolet sterilization, and ozone were discussed in the
1985 Legionella Criteria Document, however, recent studies have been conducted that provide updated
information. Because one methodology may not be sufficient, a combination of these techniques may be
more effective in eradicating Legionella from the system and preventing recolonization (Yu et al. 1993).
Thermal Disinfection
Thermal disinfection is a common practice for water distribution systems in hospitals, hotels, and
other institutional buildings. The hot water temperature is elevated to above 70°C (158° F), and distal
sites, such as faucets and showerheads, are flushed for thirty minutes (Nguyen et al. 1991, Miuetzner et
al. 1997, Stout and Yu 1997). L. pneumophila is killed at temperatures above 60°C (140°F). At 70°C
VI-19
-------
(158°F), it takes ten minutes to eliminate L. pneumophila from water, and at 60°C (140°F) L.
pneumophila are eradicated in 25 minutes (Muraca et al. 1990). In cases of outbreak, thermal
disinfection can be quickly implemented. No special equipment is needed, and it is relatively
inexpensive (Stout and Yu 1997, Muraca et al. 1990, Nguyen et al. 1991). The disadvantages to this
method are the potential for scalding and the fact that many personnel are required to monitor distal
sites, tank water temperatures, and flushingtimes (Nguyen et al. 1991, Muraca et al. 1990). In addition,
recolonization will occur within months because disinfection using this method is only temporary (Lin et
al. 1998).
In state development centers for mentally and physically handicapped people, hot water tanks
positive for Legionella were heated to 71°C for 72 hours followed by flushing for 15 minutes. In one
center, Legionella reoccurred after three months. Consequently, a quarterly heating schedule was
established in both centers (Beam et al. 1984).
Hyper chlor matron
Hyperchlorination of water distribution systems requires the installation of a chlorinator. Shock
hyperchlorination involves the addition of chlorine to a water system, raising chlorine throughout the
system to a concentration of 20 to 50 mg/L. The chlorine levels of the system are returned to 0.5 to 1
mg/L after one to two hours (Lin etal. 1998). Continuous hyperchlorination entails the addition of
chlorinated salts (e.g., calcium hypochlorite (solid) or sodium hypochlorite (aqueous)) to the water at
concentrations ranging from 2 to 6 mg/L(ppm) (Stout and Yu 1997, Muraca et al. 1990). Domestic
residual levels are typically 1 mg/L (ppm) (Muraca et al. 1990). The 1985 Legionella Criteria
Document suggests using chlorine levels of 1-2 mg/L (ppm), however, recent studies have shown that
using chlorine levels of 3-5 mg/L is more effective (Helmes et al. 1988). The chlorinator will maintain a
set level of chlorine throughout the system, which should completely eliminate Legionella.
Unfortunately, this method is relatively expensive, and it does have some drawbacks. This method leads
to corrosion of the pipes of the system after five to six years of operation, and eventually parts of the
system may be destroyed. Corrosion can be reduced by the use of a silicate coating on the water pipes
(Nguyen et al. 1991). In addition, mechanical failure of the chlorinator, if not detected, could result in
Legionella recolonizing the system (Nguyen et al. 1991). Human health problems are another result of
VI-20
-------
hyperchlorination. High levels of trihalomethanes develop in the hot water of the system when chlorine
levels exceed 4 mg/L (Helmes et al. 1988, Muraca et al. 1990). Trihalomethanes are potentially
carcinogenic, andean be reduced by maintaining the concentration of the chlorine below 4 mg/L
(Muraca et al. 1990).
Ezzeddine et al. (1989) describes disinfection in ahospital where 6 ppm of free residual chlorine
was used in a heating tank during a 6-hour period of time. Legionellawas eliminated from the tank;
however, chlorination of the mixer tank, where the temperature was 45°C, was not successful even when
chlorine levels were raised to 6 ppm over 48 hours.
Helmes et al. (1988) combined the hyperchlorination method with an elevated water temperature
at a University of Iowa hospital after a 1981 outbreak of nosocomial legionellosis. Chlorine levels were
set at 3-5 mg/L, while temperatures were raised to 60-70°C. After six months of hyperchlorination,
Legionella was no longer detected in samples.
Copper-Silver lonization
Copper-Silver lonization distorts the permeability of the Legionella cell, denatures proteins, and
leads to lysis and cell death. A commercial system can be easily installed to perform this ionization.
This system sends an electrical current to copper/silver electrodes, which generate positively charged
ions. These positively charged ions electrostatically bond to the negatively hypercharged sites on the
cell walls of the microorganisms (Nguyen et al. 1991, Miuetzner et al. 1997, Muraca et al. 1990). The
Legionella are then killed, making it unlikely that recolonization will occur. Copper-silver ionization is
less expensive than hyperchlorination and provides residual protection throughout the water distribution
system (Nguyen et al. 1991, Muraca et al. 1990). A disadvantage of this approach is that the system's
performance will suffer unless scale is removed regularly from the electrodes and the pH of the system is
maintained below 8. Also, extremely high concentrations of copper and silver ions will turn the water a
blackish color, which can stain porcelain (Lin et al. 1998). Another disadvantage is that over an
extended period of time, human consumption of the water from this system may result in accumulation
of copper and silver and toxic effects (Muraca et al. 1990). However, because copper and silver ions are
typically only added to hot water recirculating lines, human exposure would be minimal since
VI-21
-------
consumption of large amounts of water is unlikely (Lin et al. 1998). In addition, the levels of ions in hot
water are maintained below the EPA recommended levels for cold drinking water which are 1.3 ppm
copper and 100 ppb for silver (a secondary minimum contaminant level).
Miuetzner et al. (1997) used a flow-through cell containing two sets of four copper-silver
electrodes. A single cell was installed in each of three hot water circuits of a hospital. The copper-silver
ionization system significantly reduced the amount of L. pneumophila recovered from the faucets from
72 percent to 2 percent within one month. Control of Legionella was maintained for at least 22 months
after the ionization treatment.
Ultraviolet Light Sterilization
Ultraviolet light kills Legionella by disrupting cellular DNA synthesis (Muraca et al. 1990). An
ultraviolet light sterilization system can be installed easily. It can be positioned to disinfect the incoming
water, or it can be installed at a specific place in the pipe system that services a designated area. The UV
system consists of low-pressure mercury lamps in quartz sleeves. Sterilization is most effective at UV
energy wavelengths of 254 nm and temperatures of 40°C (104°F) (Muraca et al. 1990). A filter should
be used to remove particulates from the water to keep UV light transmission optimal (World Health
Organization 1990). No chemical by-products are produced, and the taste and odor of water from a
water distribution system containing a UV sterilizer are not affected (Muraca et al. 1990). The UV
sterilization system requires continuous maintenance in order to prevent scale from coating the UV
lamps. The system does not provide residual protection, so distal areas must be disinfected (Nguyen et
al. 1991, Muraca et al. 1990). Operational problems, such as electrical malfunction and water leaks, are
possible, in which case experienced technicians are needed (Muraca et al. 1990).
Ozonation
Ozone can be used to kill L. pneumophila. It can be created using ozonators, which electrically
excite oxygen (O2) to ozone (O3). Ozone instantaneously inactivates Legionella, however, it has a short
half-life and decomposes quickly back to oxygen. A second form of disinfection may be required in the
distribution system for residual protection. Also, ozonation is more expensive than hyperchlorination,
VI-22
-------
and a large amount of space is required for the air preparation equipment or oxygen tanks and contacting
tank (Muraca et al. 1990). Ozonation was described in the 1985 Legionella Criteria Document as a
possible method of eliminating Legionella from a water distribution system. At the time, few studies
had been conducted and the results were inconclusive.
Muraca et al. (1987) recommend using a 1-2 mg/L ozone residual for treatment of domestic
water. They demonstrated that a 1-2 mg/L ozone residual caused a 5 log decrease in aL. pneumophila
population of 107 CFU/mL over five hours within a model plumbing system.
Instantaneous Steam Heating
Instantaneous Steam Heating systems entail flash heating water to temperatures greater than
88°C (190°F) and then blending the hot water with cold water to attain a designated water temperature
(Nguyen et al. 1991, Muraca et al. 1990). These systems are often cost-effective because specialized
personnel are not needed to operate them; maintenance can be performed by regular building staff. The
maintenance is, however, more complex than the maintenance of a conventional hot water tank.
Instantaneous Steam Heating systems work best when installed as the original system of a building
rather than when the building has already been contaminated by Legionella. Another drawback to this
system is that it can only be used to control Legionella in the hot water supply system. The cold water
portion of the distribution system i s not disinfected (Muraca et al. 1990). Any Legionella that may have
colonized the system downstream of the heater will be unaffected. In order for disinfection to be
complete, the hot water temperature at outlet sites must exceed 60°C. These heaters may not have the
ability to flush large amounts of outlets with superheated water for thirty minutes (Lin et al. 1998).
C. Summary
The examination of water for the presence of Legionella is best done by taking swab samples of
the medium over which the water flows. The specimen should then be concentrated by filtration, treated
with an acid buffer to enhance Legionella recovery, and cultured on a BCYE agar medium. Legionella
can be detected in environmental and biological samples by a number of tests, the most common of
which are direct and indirect immunofluorescence assays.
VI-23
-------
Contamination by Legionella has occurred in the water distribution systems of many hospitals,
hotels, and other buildings. Various means of disinfection have been established and utilized. Some
methods have not always proven completely successful or have not provided permanent protection from
recolonization. A combination of these methods may be the most effective way of managing water
systems and preventing future outbreaks. Yu et al. (1993) defines two categories of disinfection, focal
and systemic. Focal disinfection is directed at a specific portion of the system and would include
ultraviolet light sterilization, instantaneous heating systems, and ozonation. Systemic methods, such as
thermal, hyperchlorination, copper-silver ionization, disinfect the entire system. Selecting a combination
of focal and systemic disinfection techniques would ensure eradication of present Legionella colonies
and prevent recolonization of the water distribution system.
VI-24
-------
VIII. Research Requirements
From all of the information presented in the previous chapters, it is clear that Legionella bacteria
are an important cause of community- and hospital-acquired pneumonia, and they can be associated with
serious morbidity and mortality, especially when the infection is not rapidly diagnosed and treated.
Legionella are widely distributed in the environment, including treated water supplies. In the past 13
years (i.e., during the time since publication of the 1985 EPA Criteria Document on Legionella).,
dramatic advances have been achieved in our understanding of the behavior and transmission of
Legionella., including information on: special ecological niches occupied by these organisms, including
their presence in biofilms and their symbiotic relationships with larger microbes such as amoebae;
improved techniques for the clinical isolation (e.g., culture techniques) and characterization (e.g., PCR
technology) of these organisms; improved methods for identifying patients recently or currently infected
with these organisms (e.g., urinary antigen assay); factors important for understanding the epidemiology
of legionellosis infection; and effective measures for eradicating these organisms from treated water
supplies.
Despite the important advances in the 13 years since the previous EPA Legionella Criteria
Document, additional information is needed to institute optimal prevention and control measures and to
minimize the morbidity and mortality associated with Legionella. Specific information gaps include the
following:
• The relative influence of the symbiotic relationship between Legionella organisms and larger
microbes on Legionella survival, transmission, virulence, and susceptibility to disinfection.
More information is also needed on the implications of the intracellular replication of Legionella
inside host microbes.
• Key environmental factors promoting the growth of Legionella in biofilms. Additional
information is needed about the structure and physiology of biofilms, and in particular, the
effects of changing environmental conditions on their ecology.
VIII-1
-------
• More comprehensive data on the occurrence ofLegionella in groundwater, especially as it relates
to potable water supplies.
• Further information on the relative importance of various reservoirs of the organism (and thus the
allocation of expenditures for disinfection); in particular, the diminishing role of cooling towers
and the increasing prominence of potable water distribution systems as reservoirs for Legionella.
• The nature of the dose-response relationship for this organism, including the development of
models, particularly for exposures from potable water. An effort should be made to determine
the predictive value ofLegionella concentrations found in a given reservoir. Research is also
needed to establish the minimal infectious dose for high-risk populations.
• A clearer definition of the important factors involved in transmission of this infectious agent
from a specific source, which would be facilitated by more accurate identification of legionellosis
cases, especially of sporadic cases, and the corresponding improved epidemiological and
environmental analyses.
• The further characterization of risk factors for acquiring legionellosis, particularly for
community-acquired, sporadic cases. Many cases of legionellosis undoubtedly still go
unrecognized. Information indicating patients at greatest risk ofLegionella infection should also
be disseminated more widely to clinicians, with the hope of more accurately and rapidly
identifying (and treating for) Legionella as the causative agent, thus reducing morbidity and
mortality associated with these organisms.
• The risk for development of legionnaires' disease from Legionella present in residential water
systems (single family or multi-family dwellings).
• Identification of the most effective (and most cost-effective) biocidal treatments for a given
source ofLegionella.
VIII-2
-------
• Delineation and development of specific design and operational/physicochemical modifications
for building water supply systems, in order to minimize colonization by Legionella and symbiont
hosts, including biofilm eradication.
Given the potentially high costs of surveillance for, and eradication of, Legionella from treated
water supplies, new information that fills some of these gaps will be of great value in identification and
institution of the best strategies for prevention of legionellosis.
VIII-3
-------
(this page intentionally left blank)
VIII-4
-------
IX. References
Abernathy-Carver KJ, Fan LL, Boguniewicz M, Larsen GL, Leung DY. 1994. Legionella and
Pneumocystis pneumonias in asthmatic children on high doses of systemic steroids. PediatrPulmonol.
18(3):135-138.
Addi ss DG, Davis JP, La Venture M, Wand PJ, Hutchinson MA, McKinney RM. 1989. Community-
acquired Legionnaires' Disease Associated with a Cooling Tower: Evidence for Longer-Distance
Transport of Legionella Pneumophila. Am J Epidemiol. 130(3):557-568.
Alary M, Joly JR. 1992. Factors contributing to the contamination of hospital water distribution systems
by Legionellae. J Infect Dis. 165(3):565-569.
Alexiou SD, Antoniadis A, Papapaganagiotou J, StefanouTH. 1989. Isolation of Legionella
pneumophila from hotels in Greece. European Journal of Epidemiology. 5(1):47-50.
Allegheny County Health Department. 1997. Approaches to prevention and control of Legionella
infection in Allegheny County health care facilities. Pittsburgh, PA: Allegheny County Health Dept.
Ampel NM, Wing EJ. 1990. Legionella infection in transplant patients. Semin Respir Infect. 5(1):30-
37.
Andersen BB, Sogaard I. 1987. Legionnaires'disease and brain abscess. Neurology. 37(2): 333-334.
Anonymous. 1997a. Communicable diseases surveillance: Legionellosis. Commun Dis Intell.
21(10):137-143.
Anonymous. 1997b. Legionnaires'disease in Europe, 1996. Wkly Epidemiol Rec. 72(34):253-257.
Anonymous. 1996. An outbreak of Legionnaires'disease in Spain. CDR Weekly. Novembers.
6(45).
Anonymous. 1995a. Cluster of cases of Legionnaires'disease associated with travel to Turkey. CDR
Weekly. September 15. 5(37).
Anonymous. 1995b. Legionnaires'disease associated with Turkey: update. CDR Weekly. 5(40): 193.
Anonymous. 1993. Legionnaires' disease associated with travel in the USA. CDR Weekly. April 30.
Anonymous. 1989. Communicable Disease Report, October to December 1988. Community Med.
11(2):168-172.
IX-1
-------
Anonymous. 1988. Report from the PHLS Communicable Disease Surveillance Centre. Br Med J
(Clin Res Ed). 296(6624):778-779.
Anonymous. 1985. Communicable Disease Report: April to June 1985. Community Med (Great
Britain). 7(4):304-308.
Armengol S, Domingo C, Mesalles E. 1992. Myocarditis: a rare complication during Legionella
infection. IntJCardiol. 37(3):418-420.
Arnow PM, Weil D, Para MF. 1985. Prevalence and significance of Legionellapneumophila
contamination of residential hot-tap water systems. J Infect Dis. 152(1):145-151.
Augoustinos MT, Venter SN, KfirR. 1995. Assessment of water quality problems due to microbial
growth in drinking water distribution systems. Environmental Toxicology and Water Quality.
10(4):295-299.
Bangsborg JM. 1997. Antigenic and genetic characterization of Legionella proteins: contributions to
taxonomy, diagnosis and pathogenesis. APMK Supplementum: 70(105):l-53.
Bangsborg JM, Jensen BN, Friis-Moller A, BruunB. 1990. Legionellosis in patients with HIV
infection. Infection. 18(6):342-346.
BarbareeJM, Fields BS, FeeleyJC, Gorman GW, Martin WT. 1986. Isolation of protozoa from
water associated with a legionellosis outbreak and demonstration of intracellular multiplication of
Legionella pneumophila. Appl Environ Microbiol. 51(2): 422-424.
Barbaree JM, Gorman GW, Martin W, Fields BS, Morrill WE. 1987. Protocol for sampling
environmental sites for Legionellae. Appl Environ Microbiol. 53(7): 1454-1458.
Baskerville A. 1984. Pathology and pathophysiology. In: Legionella. Proc. 2nd Int. Symp., June 19-23,
1983; Atlanta, GA. Thornsberry C, Balows A, FeeleyJC, Jakubowski W.(Eds.). American Society
for Microbiology, Washington, DC. p. 136-140. (As cited in EPA 1985)
Baskerville A, DowsettAB, FitzgeorgeP, HambletonP, BrosterM. 1983a. Ultrastructure of
pulmonary alveoli and macrophages in experimental leionnaires'disease. JPathol. 140:77-90. (As
cited in EPA 1985)
Baskerville A, Fitzgeorge RB, BrosterM, HambletonP. 1983b. Histopathology of experimental
legionnaires'disease in guinea pigs, rhesus monkeys and marmosets. JPathol. 139:349-362. (As cited
in EPA 1985)
BatyV, HoenB, Schuhmacher H, Amiel C, Reyrolle M, Garin H, Canton P. 1997. Legionella
jordanis pneumonia unresponsive to fluoroquinolones in a non-immunocompromised host. Scand J
Infect Dis. 29(3)319-320.
IX-2
-------
BeamTRJr, Moreton D, Raab TA, Heaslip, Yu VL 1984. Epidemiology and Control of
Legionellaceae in State Developmental Centers. In: Legionella. Proc. 2nd International Symposium.
June 19-23, 1983. Atlanta, GA. Thornsberry C, Balows A, Feeley JC, Jakubowski W. (Eds.).
American Society for Microbiology, Washington, DC. 236 -237.
BellJC, JormLR, Williamson M, ShawNH, Kazandjian DL, ChiewR, Capon AG. 1996.
Legionellosis linked with a hotel car park—how many were infected? Epidemiol Infect. 116(2): 185-192.
Belyi YuF. 1990. Action of Legionella cytolysin on components of the phosphokinase system of
eukaryotic cells. Biomed Sci. 1(5): 494-498.
Benson RJ, Thacker WL, Daneshvar MI, Brenner DJ. 1996. Legionella waltersii sp. nov. and an
unnamed Legionella genomospecies isolated from water in Australia. International Journal of
Systematic Bacteriology. 46(3):631-634.
Bentham RH. 1993. Environmental factors affecting the colonization of cooling towers by
Legionella spp. in South Australia. Int BiodeteriorBiodegrad. 31(l):55-63.
BerbariE, Cockerill FR 3rd, Steckelberg JM. 1997. Infective endocarditis due to unusual or fastidious
microorganisms. Mayo Clin Proc. 72(6):532-542.
Berendt RF, Young HW, Allen RG, Knutsen GL. 1980. Dose-response of guinea pigs exerimentally
infected with aerosol of Legionella pneumophila. J Infect Dis. 141(2):186-192. (As cited in EPA 1985)
Berk SG, TingRS, Turner GW, AshburnRJ. 1998. Production of respirable vesicles containing live
Legionella pneumophila cells by two Acanthamoeba spp. Appl Environ Microbiol. 64(l):279-286.
Bezanson G, Burbridge S, Haldane D, Yoell C, Marrie T. 1992. Diverse populations of Legionella
pneumophila present in the water of geographically clustered institutions served by the same water
reservoir. J Clin Microbiol. 30(3):570-576.
Bhopal R. 1995. Sourceof Infection for Sporadic Legionnaires'Disease: A Review. J Infect. 30(1):9-
12.
Bhopal RS. 1993. Geographical Variation of Legionnaires' Disease: a Critique and Guide to Future
Research. Int J Epidemiol. 22(6): 1127-1136.
Bhopal RS, WagstaffR. 1993. Prospects for the elimination of Legionnaires'disease. J Infect.
26(3):239-243.
BlanchardDK, DjeuJY, Klein TW, Friedman H, Stewart WE II. 1988. Protective effects of tumor
necrosis factor in experimental Legionella pneumophila infections of mice via activation of PMN
function. JLeukocBiol. 43(5): 429-435.
IX-3
-------
Blatt SP, DolanMJ, Hendrix CW, Melcher GP. 1994. Legionnaires'disease in human
immunodeficiency virus-infected patients: Eight cases and review. Clinical Infectious Diseases.
18(2):227-232.
Blatt SP, Parkinson MD, Pace E, Hoffman P, DolanD, LauderdaleP, ZajacRA, Melcher GP. 1995.
AmJMed. 195(1): 16-22.
Bollin GE, Plouffe JF, ParaMF, Hackman B. 1985. Aerosols containing Legionellapneumophila
generated by shower heads and hot-water faucets. Appl. Environ. Microbiol. 50(5): 1128-1131.
Bopp CA, Sumner JW, Morris GK, Wells JG. 1981. Isolation of Legionella from environmental water
samples by low-pH treatment and use of a selective medium. J Clin Microbiol. 13(4):714-719. (As
cited in EPA 1985)
Bornstein N, Marmet D, Surgot M, Nowicki M, Arslan A, Esteve J, Fleurette J. 1989a. Exposure to
Legionellaceae at a hot spring spa a prospective clinical and serological study. Epidemiol Infect.
102(l):31-36.
Bornstein N, Marmet D, Surgot M, Nowicki M, Meugnier H, Fleurette J, Ageron E, Grimont F, Grimont
PD. 1989b. Legionella GratianaNew-species Isolated from French Spa Res Microbiol. 140 (8):541-
552.
Bornstein N, Vieilly C, Marmet D, Surgot M, Fleurette J. 1985. Isolation of Legionella anisa from a
hospital hot water system. European Journal of Clinical Microbiology. 4(3):327-330.
Bornstein N, Vieilly C, Nowicki M, Paucod JC, Fleurette J. 1986. Epidemiological evidence of
legionellosis transmission through domestic hot water supply systems and possibilities of control. Isr J
MedSci. 22(9):655-661.
BotzenhartK, Heizmann W, Sedaghat S, Heeg P, Hahn T. 1986. Bacterial Colonization and
Occurrence of Legionella-pneumophilain Warm and Cold Water in Faucet Aerators and in Drains of
Hospitals. Zentralbl Bakteriol Mikrobiol Hyg Ser B Umwelthyg Krunkenhaushyg Arbeitshyg Praev
Med. 183(l):79-85.
Brady MT. 1989. Nosocomial Legionnaires' Disease in a children's hospital sect. J Pediatr. 115(1):46-
50.
Breiman RF, Butler JC. 1998. Legionnaires' disease: Clinical, epidemiological, and public health
perspectives. Semin. Respir. Infect. 13(2): 84-89.
Breiman RF, Cozen W, Fields BS, Mastro TD, CarrSJ, Spika JS, MascolaL. 1990. Role of air
sampling in investigation of an outbreak of legionnaires' disease associated with exposure to aerosols
from an evaporative condenser. J Infect Dis. 161(6): 1257-1261.
Brenner DJ. 1987. Classification of Legionellae. Seminars in Respiratory Infections. 2(4): 190-205.
IX-4
-------
Brenner DJ. 1986. Classification of Legionellaceae. Current Status and Remaining Questions. Israel
Journal of Medical Sciences. 22:620-632.
Brenner DJ, FeeleyJC, Weaver RE. 1984. Family VSlLegionellaceae. In Bergey's Manual of
Systematic Bacteriology. Krieg NR, Holt JG ( Eds). Williams and Wilkins, Baltimore, MD. (1):279.
(As sited in EPA 1985)
Brieland J, Fantone JC, Remick DG, LeGendre M, McClain M, Engleberg NC. 1997b. The Role of
Legionella pneumophila-lnfectedHartmannella vermiformis as an infectious particle in a murine model
of Legionnaire's disease. Infect Immun. 65(12): 5330-5333.
Brieland J, McClain M, Heath L, Chrisp C, Huffnagle G, LeGendre M, Hurley M, Fantone J, Engleberg
C. 1996. Coinoculation withHartmannella vermiformis enhances replicative Legionellapneumophila
lung infection in a murine model of Legionnaires' disease. Infect Immun. 64(7): 2449-2456.
Brieland J, McClain M, LeGendre M, Engleberg C. 1997a. Intrapulmonary Hartmannella vermiformis:
a potential niche for Legionella pneumophila replication in a murine model of legionellosis. Infect
Immun. 65(ll):4892-4896.
Broadhead AN, Negron-alvira A, BaezLA, Hazen TC, CanoyMJ. 1988. Occurrence of Legionella
species in tropical rain water cisterns. Caribb J Sci. 24(l-2):71-73.
Campo AM, Apraiz D. 1988. Epidemiological study of the Legionella pneumophila presence in
potable water in Alicante municipal waters of Alicante, Spain. Aqua (The Journal of the International
Water Supply Association). 3:116-119.
Cappabianca RM, Jurinski NB, Jurinski JB. 1994. A comparison of Legionella and other bacteria
concentrations in cooling tower water. Appl Occup Environ Hyg 9(5):358-361.
Carlson NC, Kuskie MR, Dobyns EL, Wheeler MC, Roe MH, AbzugMJ. 1990. Legionellosis in
children: an expanding spectrum. Pediatr Infect Dis J. 9(2): 133-137.
CastellaniPastorisM, Benedetti P, Greco D, Volpi E, Billo N, Fehrenbach FJ, Hohl P, Horbach I,
Wewalka G. 1992. Six cases of travel-associated Legionnaires' disease in Ischia involving four
countries. Infection. 20(2):73-77.
CastellaniPastorisM, Ciceroni L, Lo Monaco R, Goldoni P, MentoreB, Flego G, Cattani L,
Ciarrochi S, Pinto A, ViscaP. 1997. Molecular epidemiology of an outbreak of Legionnaires'disease
associated with a cooling tower in Genova-Sestri Ponente, Italy. Eur J Clin Microbiol Infect Dis.
16(12):883-892.
CastellaniPastorisM, ViganoEF, Passi C. 1988. A family cluster of Legionella pneumophila
infections. Scand J Infect Dis. 20(5):489-93.
CDC. 1997a. Guidelines for Prevention of Nosocomial Pneumonia. MMWR. 46(Rr-l):l-79.
IX-5
-------
CDC. 1997b. Summary of Notifiable Diseases, United States, 1996. MMWR. 45(53): 1-103.
CDC. 1996. Summary of Notifiable Diseases, United States, 1995. MMWR. 44(53): 1-87.
CDC. 1994. Summary of Notifiable Diseases, United States, 1993. MMWR. 42(53): 1-87.
Chaisson RE. 1998. Bacterial pneumonia in patients with human immunodeficiency virus infection.
Semin Resp Infect. 4(2):133-138.
Chandler FW, McDade JE, FficklinMD, Blackmon JA, Thomason BM, Ewing EP. 1979c.
Pathological findings in guinea pigs innoculated intraperitoneally with the legionnaires' disease
bacterium. AnnlntMed. 90(4):671-675. (As cited in EPA 1985)
Chen TT, Schapiro JM, LoutitJ. 1996. Prosthetic valve endocarditis due toLegionellapneumophila.
J Cardiovasc Surg. 37(6):631-633.
ChingWT, Meyer RD. 1987. Legionella infections. Infect Dis Clin North Am. 1(3)595-614.
Christie P. 1997. Legionnaires' disease in residents of Scotland: 1996. Commun Dis Rep CDR Rev.
7(11):R159.
Cianciotto N, Eisenstein BI, EnglebergNC, Shuman H. 1989. Genetics and molecular pathogenesis of
Legionella pneumophila., an intracellular parasite of macrophages. Mol Biol Med. 6(5):409-424
Colbourne JS, Dennis PJ. 1989. The ecology and survival of Legionella Pneumophila. Thames Water
Authority Journal of the Institution of Water and Environmental Management. 3(4):345-350.
Colbourne JS, Dennis PJ, Trew RM, Berry C, Vesey G. 1988. Legionella and Public Water Supplies.
Water Science and Technology. 20(11-12):5-10.
Coletta FS, Fein AM. 1998. Radiological manifestations of Legionella/Legionella-\ike organisms.
Semin. Respir. Infect. 13(2): 109-115.
Colville A, Crowley J, DeardenD, Slack RGB, Lee JV. 1993. Outbreak of Legionnaires' disease at
University Hospital, Nottingham. Epidemiol Infect. 110(1): 105-116.
Conlan JW, Williams A, Ashworth LA. 1988. In vivo production of a tissue-destructive protease by
Legionella pneumophila in the lungs of experimentally infected guinea-pigs. J Gen Microbiol. 134(Pt
1):143-149.
Conner RW, Gilbert DN. 1979. Legionella pneumophila in drinking water of guinea pigs. Ann Intern
Med. 91:323. (As cited in EPA 1985)
CunhaBA. 1998. Clinical features of legionnaires'disease. Semin. Respir. Infect. 13(2):116-127.
IX-6
-------
Darelid J, Bengtsson L, Gastrin B, Hallander H, Lofgren S, Malmvall BE, Olinder-Nielsen AM,
Thelin AC. 1994. An outbreak of Legionnaires'Disease in a Swedish hospital. Scandinavian Journal
of Infectious Diseases. 26(4):417-425.
Davies DH, Hill EC, Howells CHL, Ribeiro CD. 1985. Legionellapneumophila in coal miners. BrJ
IndMed. 42:421-425.
Davis GS, Winn WC Jr. 1987. Legionnaires' disease: respiratory infections caused by Legionella
bacteria. Clin Chest Med. 8(3):419-439.
Davis GS, Winn WC Jr., Gump DW, Craighead JM, BeatyHN. 1983a. Legionnaires'pneumonia
after aerosol exposure in guinea pigs and rats. Am Rev Respir Dis. 126:1050-1057. (As cited in EPA,
1985)
Davis GS, Winn WC Jr., Gump DW, Craighead JM, BeatyHN. 1983b. The kinetics of early
inflammatory events during experimental pneumonia due to Legionella pneumophila in guinea pigs. J
Infect Dis. 148(5):823-835. (As cited in EPA 1985)
Davis GS, Winn WC Jr., Gump DW, Craighead JM, BeatyHN. 1983c. Legionnaires'pneumonia in
guinea pigs and rats produced by aerosol exposure. Chest. 83S:15S-16S. (As cited in EPA 1985)
De Lassence A, Matsiota-Bernard P, Valtier B, Franc B, JardinF, Nauciel C. 1994. A case of
myocarditis assodated with Legionnaires' disease. Clin Infect Dis. 18(1):120-121.
Den Boer JW, Yzerman E, Van Belkum A, Vlaspolder F, Van Breukelen FJ. 1998. Legionnaire's
disease and saunas. Lancet. 351(9096):114.
Dennis PJ. 1991. Legionella in the United Kingdom and water quality in buildings. ASHRAE Winter
Meeting, Conference No. 15219. ASHRAE Transactions, p.271-274.
DevriendtJ, Staroukine M, Schils E, Sivaciyan B, Van Beers D. 1990. Legionellosis and "torsades
depointes". ActaCardiol. 45(4):329-33.
Doebbeling BN, Ishak MA, Wade BH, Pasquale MA, Gerszten RE, Groschel DHM, Kadner J,
Wenzel RP. 1989. Nosocomial Legionella micdadei pneumonia: 10 years experience and a case-
control Study. JHosp Infect. 13(3):289-298.
Domingo C, Roig J, Seres J. 1989. Pericardial effusion as a clinical sign of Legionnaires'disease. Int
J Cardiol. 23(3):407-409.
DowlingJN, SahaAK, GlewRH. 1992. Vir ulence Factors of the Fa mil y Legionellaceae. Microbiol
Rev. 56(1):32-60.
EarleKA, HoffbrandBI. 1990. Acalculous cholecystitis complicating Legionnaires' disease. BrJ Clin
Pract. 44(11):783.
IX-7
-------
EdelsteinPH. 1998. Legionnaires' disease. NEnglJMed. 338(3 ):200.
EdelsteinPH. 1993. Legionnaires' disease. Clin Infect Dis. 16(6):741-747.
Edelstein PH. 1988. Nosocomial Legionnaires' disease: a global perspective. J Hosp Infect. Suppl
A:182-188.
EdelsteinPH. 1987. Laboratory Diagnosis of Infections Caused by Legionellae. Eur J Clin Microbiol.
EdelsteinPH, Edelstein MA, LehrKH, Ren J. 1996. In-vitro activity of levofloxacin against clinical
isolates of Legionella spp, its pharmacokinetics in guinea pigs, and use in experimental Legionella
pneumophila pneumonia. J Antimicrob Chemother. 37(1): 1 17-126.
EdelsteinPH, SnitzerJB, Bridge JA. 1982. Enhancement of recovery of Legionella pneumophila from
contaminated respiratory tract specimens by heat. J Clin Microbiol. 16(6): 1061-1065.
EhretW, vonSpechtBU, Ruckdeschel G. 1986. Discrimination between clinical and environmental
strains of Legionella pneumophilaby a monoclonal antibody. Isr J Med Sci. 22(10):715-723.
Eitrem R, Forsgren A, Nilsson C. 1987. Pneumonia and acute pancreatitis most probably caused by a
Legionella longbeachae infection. Scand J Infect Dis. 19(3):381-382.
England AC m., FraserDW, Plikaytis BD, Tsai TF, Storch G, Broome CV. 1981. Sporadic
legionellosis in the United States: The first thousand cases. Ann. Intern Med. 94(2): 164-170. (As sited
in EPA 1985)
EPA. 1985. Legionella Criteria Document. United States Environmental Protection Agency, Office of
Water. Washington, DC.
Ezzeddine H, Van Ossel C, Delmee M, Wauters G. 1989. Legionella-spp in a Hospital Hot Water
System: Effect of Control Measures. J Hosp Infect. 13(2)121-132.
Fallen RJ, Rowbotham TJ. 1990. Microbiological investigations into an outbreak of Pontiac Fever
due to Legionella micdadei associated with use of a whirlpool. J Clin Pathol. 43(6):479-483.
Famiglietti RF, Bakerman PR, Saubolle MA, RudinskyM. 1997. Cavitary legionellosis in two
immunocompetent infants. Pediatrics. 99(6):899-903.
Fang GD, Yu VL, Vickers RM. 1989. Disease due to the Legionellaceae (other than Legionella
pneumophila): Historical, microbiological, clinical, and epidemiological review. Medicine (Baltimore).
68(2):116-132.
Fenstersheib M, Miller M, Diggins C, Li ska S, DetwilerL, Werner SB, LindquistD, ThackerWL,
Benson RR. 1 990. Outbreak of Pontiac Fever due to Legionella anisa. Lancet (North American
Edition). 336(8706):35-37.
IX-8
-------
Fenves AZ. 1985. Legionnaires' disease associated with acute renal failure: a report of two cases and
review of the literature. Clin Nephrol. 23(2)96-100.
Fields BS. 1996. The molecular ecology of Legionellae. Trends Microbiol. 4(7):286-90.
Finch R. 1988. Minimi sing the risk of Legionnaires' disease. Br Med J (Clin Res Ed).
296(6633):1343-1344.
FioreAE, Nuorti JP, Levine OS, Marx A, Weltman AC, Yeager S, Benson RF, Pruckler J, Edelstein
PH, GreerP, Zaki SR, Fields BS, Butler JC. 1998. Epidemic Legionnaires'Disease Two Decades
Later: Old Sources, New Diagnostic Methods. Clinical Infectious Diseases. 26(2): 426-433.
Fitzgeorge RB, Baskerville A, BrosterM, Hambleton P, DennisPJ. 1983. Aerosol infection of
animals with strains of Legionellapneumophila of different virulence: Comparison with intraperitoneal
and intranasal routes of infection. JHyg. 90:81-89. (As cited in EPA 1985)
Fricker EJ, Fricker CR. 1995. Detection of Legionella spp.using a commercially available polymerase
chain reaction test. Water Science and Technology. 31(5-6): 407-408.
Friedman S, Spitalny K, Barbaree J, Faur Y, McKinney R. 1987. Pontiac Fever Outbreak Associated
with a Cooling Tower. American Journal of Public Health. 77(5):568-572.
Friedman H, Yamamoto Y, Newton C, Klein T. 1998. Immunologic Response and Pathophysiology of
Legionella Infection. Seminars in Respiratory Infections. 13(2): 100-108.
Fujio H, Kawamura I, Miyamoto H, Mitsuyama M, Yoshida S. 1995. Decreased capacity of aged mice
to produce interferon-gamma in Legionella pneumophila infection. Mech Ageing Dev. 81(2-3): 97-106.
GarbePI, Davis BJ, Weisfeld JS, MarkowitzL, Miner P, GarrityF, Barbaree JM, Reingold AL.
1985. Nosocomial Legionnaires' Disease: Epidemiologic Demonstration of Cooling Towers as a
Source. JAMA. 254(4):521-524.
GeaJ, Rodriguez-Roisin R, Torres A, Roca J, Agusti-Vidal A. 1988. Lung function changes
following Legionnaires'disease. EurRespirJ. 1(2): 109-114.
GecewiczTE, SaravoL, Lett SM, LkudtPE, DeMariaAJr., Stobierski MG, Johnson D, Hall W,
Dietrich S, Stiefel H, Robinson-Dunn S, Shah S, Hutchinson C, Mermel LA, Giorgio OH, Agostino
LD, Rittmman M, Stoeckel M. 1994. Legionnaires' disease associated with cooling towers -
Massachusetts, Michigan, and Rhode Island, 1993. MMWR. 43(27):491-499.
GoetzAM, Stout JE, Jacobs SL, Fisher MA, PonzerRE, Drenning S, Yu VL. 1998. Nosocomial
legionnaires' disease discovered in community hospitals following cultures of the water system: seek and
ye shall find. American Journal of Infection Control. 26(1):8-11.
IX-9
-------
Gorman GW, FeeleyJC, Steigerwalt A, EdelsteinPH, Moss CW, Brenner DJ. 1985. Legionella
anisa: a new species of Legionella isolated from potable waters and a cooling tower. Appl Environ
Microbiol. 49(2):305-309.
Grabow NA, Pienaar EJ, Kfir R. 1 99 1 . The occurrence of Legionella bacteria in cooling towers in
South Africa. Water Science and Technology. 24(2): 149-152.
GrimontPA. 1986. Rapid methods for identification of Legionella— a review. Isr J Med Sci.
22(10):697-702.
Guerrero I, Genese C, HungMJ, Ragazzoni H, Brook J, Finelli L, Spitalny KC, MojicaBA,
Mahoney KJ, Heffernan RT, Kondracki SF, Morse DL, Cartter ML, Hadler J, Rankin JT Jr, Groves
C. 1994. Update: Outbreak of Legionnaires' Disease Associated with a cruise ship, 1994. MMWR.
43(31):574-575.
Guerrero 1C, Filippone C. 1996. A cluster of Legionnaires' disease in a community hospital—a clue to a
larger epidemic. Infect Control Hosp Epidemiol. 17(3): 177-178.
GuiguetM, Pierre J, Brun P, Berthelot G, Gottot S, Gibert C, Valleron A. 1987. Epidemiological
survey of a major outbreak of nosocomial Legionellosis. Int J Epidemiol. 16(3): 466-471.
HabichtW, Mueller FIE. 1988. Occurrence and parameters of frequency of Legionella in warm water
systems of hospitals and hotels in Lower Saxony. Zentralbl Bakteriol Mikriobiol Hyg. 186(l):79-88.
Haines JD Jr., Calhoon H. 1987. Interstitial nephritis in a patient with Legionnaires' disease. Postgrad
Med. 81(3):77-79.
HambletonP, Baskerville A, Fitzgeorge RB, Bailey NE. 1982. Pathological and biochemical features
of Legionella pneumophila infection in guinea pigs. J Med Microbiol. 15:317-326. (As cited in EPA
1985)
HanrahanJP, Morse DL, ScharfVB, Debbie JG, Schmid GP, MckinneyRM, ShayeganiMA. 1987.
Community hospital outbreak of Legionellosis: transmission by potable hot water. Am J Epidemiol.
125(4):639-649.
Harrington RD, Woolfrey AE, BowdenR, McDowell MG, HackmanRC. 1996. Legionellosis in a
bone marrow transplant center. Bone Marrow Transplant. 18(2):361-368.
Haugh C, HoneR, Smyth CJ. 1990. Legionella in Dublin hospital water supplies. J Med Sci.
Heath CH, Grove DI, Looke DF. 1996a. Delay in appropriate therapy of Legionella pneumonia
associated with increased mortality. Eur J Clin Microbiol Infect Dis. 15(4):286-290.
Heath L, Chrisp C, Huffnagle G, LeGendre M, Osawa Y, Hurley M, Engleberg C, Fantone J, Brieland
J. 1996b. Effector mechanisms responsible for gamma interferon-mediated host resistance to
IX- 10
-------
Legionella pneumophila lung infection: the role of endogenous nitric oxide differs in susceptible and
resistant murine hosts. Infect Immun. 64(12):5151-5160.
Hedges LJ, RoserDJ. 1991. Incidence of Legionella in the urban environment in Australia. Water
Research. 25(4):393-399.
Hell W. 1990. Legionellapneumophila in Copper Water-pipe-systems. Zentralbl Hyg Umweltmed.
189(4): 3 72.
Helms CM, Massanari RM, WenzelRP, PfallerMA, MoyerNP, Hall N. 1988. Legionnaires'
Disease Associated with a Hospital Water System: A five-year progress report on continuous
hyperchlorination. JAMA. 259(16):2423-2427.
Heng BH, Goh KT, Ng DLK, Ling AE. 1997. Surveillance of Legionelosis and Legionella Bacteria in
the Built Environment in Singapore. Ann Acad Med Singapore. 26(5):557-565.
Henke M, Seidel KM. 1986. Association between Legionella pneumophila and amoebae in water. Isr
JMedSci. 22(9):690-695.
Hershey J, Burrus B, Marcussen V, Notter J, Watson K, Wolford R, Shaffner RE m, Barrett E,
WoolardD, Branch L, HacklerR, Rouse B, Gibson L, Jenkins S, Rullan J, Miller GJr, Curran S.
1997. Legionnaires disease associated with a whirlpool spa display-Virginia. September-October,
1996. MMWR. 46(4):83-86.
Hlady WG, Mullen RC, Mintz CS, Shelton BG, Hopkins RS, Daikos GL. 1993. Outbreak of
Legionnaires' disease linked to a decorative fountain by molecular epidemiology. American Journal of
Epidemiology. 138(8):555-562.
Hoge CW, Brieman RF. 1991. Advances in the epidemiology and control of Legionella infections.
Epidemiol Rev. 13:329-40.
Holmberg RE Jr., Pavia AT, Montgomery D, Clark JM, EggertLD. 1993. NosocomialLegionella
Pneumonia in the Neonate. Pediatrics. 92(3):450-453.
HorieH, Kawakami H, MinoshimaK, Kamohara T, Nakamura T, KurokiH, Nakamura A. 1992.
Neonatal Legionnaires' disease. Histopathological findings in an autopsied neonate. Acta Pathol Jpn.
42(6):427-431.
Hossain MS, Hoque MM. 1994. Isolation of Legionella pneumophila from chlorinated water and water
from industrial cooling tower. Bangladesh Journal of Microbiology. 11(2):111-114.
Hsu SS. 1986. Isolation of Legionella species from chlorinated tap water and whirlpool baths.
Advances in Water Analysis and Treatment. 79-86.
Hubbard RB, MathurRM, MacFarlane JT. 1993. Severe community-acquired Legionella pneumonia:
treatment, complications and outcome. Q J Med. 86(5):327-332.
IX-11
-------
HuebnerRE, ReeserPW, SmithDW. 1984. Comparison of the virulence of the Philadelphia and
Pontiac isolates of Legionellapneumophila. In: Legionella. Proc. 2nd Int. Symp., June 19-23, 1983;
Atlanta, GA. Thornsberry C, Balows A, Feeley JC, Jakubowski W.(Eds.). American Society for
Microbiology, Washington, DC. p. 123-124. (As cited in EPA 1985)
Hunt DA, Cartwright KV, Smith MC, Middleton J, Bartlett CLR, Lee JV, Dennis PJ, Harper D.
1991. An Outbreak of Legionnaires' Disease in Gloucester England UK. Epidemiol Infect. 107(1): 133-
142.
Ikedo M, Yabuuchi E. 1986. Ecological studies of Legionella Species: viable counts of Legionella
pneumophila in cooling towers. Microbiol Immunol. 30(5):413-424.
JerniganDB, HofmannJ, Cetron MS, Genese CA, Nuorti JP, Fields BS, Benson RF, Carter RJ,
EdelsteinPH, Guerrero 1C, Paul SM, LipmanHB, BreimanRF. 1996. Outbreak of Legionnaires'
disease among cruise ship passengers exposed to a contaminated whirlpool spa. Lancet (North
American Edition). 347(9000):494-499.
Johnston JM, Latham RH, Meier FA, Green JA, Boshard R, Mooney BR, EdelsteinPH. 1987.
Nosocomial outbreak of Legionnaires' disease. Infect Control (Thorofare). 8(2):53-58.
JolyJR, DeryP, GauvreauL, CoteL, Trepanier C. 1986. Legionnaires'disease caused by Legionella
dumoffii in distilled water. CanMed Assoc J. 135 (11): 1274-1277.
Jones F, Ashcroft C. 1988. Survey to detect Legionella pneumophila in potable waters in North West
England. North West Water Journal of the Institution of Water and Environmental Management.
2(5):460-464.
Joseph C, Morgan D, Birtles R, Pelaz C, Martin-Bourgon C, Black M, Garcia-Sanchez \ Griffin M,
BornsteinN, Bartlett C. 1996. An international investigation of an outbreak of legionnaires'disease
among UK and French tourists. European Journal of Epidemiology. 12(3):215-219.
Joseph CA, Harrison TG, flijic-CarD, Bartlett CL. 1997. Legionnaires'disease in residents of
England and Wales: 1996. CommunDis RepCDRRev. 7(11):R153-159.
Joseph C A, Hutchinson EJ, DedmanD, Birtles RJ, Watson JM, Bartlett CL. 1995. Legionnaires'
disease surveillance: England and Wales 1994. Commun Dis Rep CDR Rev. 5(12):R180-R183.
KashubaAD, Ballow CH. 1996. Legionella urinary antigen testing: potential impact on diagnosis and
antibiotic therapy. Diagn Microbiol Infect Dis. 24(3): 129-139.
Katz SM, Matus JP. 1984. Febrile Illness Produced in Guinea Pigs by Oral Inoculation of Legionella
pneumophila. In: Legionella: Proc. 2nd Int. Symp., June 19-23,1983; Atlanta, GA. Thornsberry C,
Balows A, Feeley JC, Jakubowski W. (Eds.) American Society for Microbiology, Washington, DC.
p. 124-127. (As cited inEPA 1985)
IX-12
-------
Keller DW, Hajjeh R, DeMaria A Jr, Fields BS, Pruckle JM, Benson RS, KludtPE, Lett SM,
Mermel LA, Giorgio C, BreimanRF. 1996. Community outbreak of Legionnaires'disease: An
investigation confirming the potential for cooling towers to transmit Legionella species. Clinical
Infectious Diseases. 22(2):257-261.
Kesavan CR, Pitchumoni CS, Marino WD. 1993. Acute painless pancreatitis as a rare complication in
Legionnaires disease. Am J Gastroenterol. 88(3):468-469.
KilbornJA, Manz LA, OBrien M, Douglass MC, HorstHM, Kupin W, Fisher EJ. 1992.
Necrotizing cellulitis caused by Legionella micdadei. Am J Med. 92(1): 104-106.
Kioski C, Cage G, Johnson B, Resales C, England B, Halpin TJ. 1997. Sustained transmission of
nosocomial Legionnaires disease—Arizona and Ohio. MMWR. 46(19):416-421.
Klein NC, Cunha BA. 1998. Treatment of legionnaires'disease. Semin. Respir. Infect. 13(2):140-146
Kohler RB. 1986. Antigen detection for the rapid diagnosis of mycoplasma and Legionella pneumonia.
Diagn Microbiol Infect Dis. 4(3 Suppl):47S-59S.
Kramer MH, Ford TE. 1994. Legionellosis: ecological factors of an environmentally'new'disease.
Zentralbl HygUmweltmed. 195(5-6):470482.
Kusnetsov JM, Martikainen PJ, Jousimies-Somer HR, Vaisanen M, Tulkki AI, Ahonen HE, Nevalainen
AI. 1993. Physical, chemical and microbiological water characteristics associated with the occurrence
of Legionella in cooling tower systems. Water Research. 27(1): 85.
LeSauxNM, SeklaL, Mcleod J, Parker S, RushD, Jeffrey JR, BrunhamR.C. 1989. Epidemic of
Nosocomial Legionnaires' Disease in Renal Transplant Recipients a Case-control and Environmental
Study. CanMedAssocJ. 140(9): 1047-1053.
Lee JV, West AA. 1991. Survival and growth of Legionella species in the environment. Soc Appl
Bacteriol Symp Ser. 20:1218-1298.
Lee TC, Stout JE, Yu VL. 1988. Factors predisposing to Legionellapneumophila colonization in
residential water systems. Archives of Environmental Health. 43(l):59-62.
Levin ASS, Caiaffa Filho HH, Sinto SI, SabbagaE, Barone AA. 1991. An outbreak of nosocomial
Legionnaires' disease in a renal transplant unit in Sao Paulo, Brazil. Legionellosis J Hosp Infect.
18(3):243-248.
Levin ASS, Mazieri AO, Carvalho NB, MeirelesLP, de Andrade DR, Barone AA. 1993. Five Cases of
Nosocomial and Community-Acquired Legionnaires' Disease in Sao Paulo, Brazil. Rev Inst Med Trop
Sao Paulo. 35(1):103-106.
Levy M, Westley-Wise V, BlumerC, FrommerM, Rubin G, Lyle D. 1994. Legionnaires'disease
outbreak, Fairfield 1992: public health aspects. Aust J Public Health. 18(2):137-143.
IX-13
-------
Lieberman RJ, Shadix LC, Newport BS, Crout SR, Buescher SE, Safferman RS, Stetler RE, Lye D,
ShayFoutG, Dahling DR. 1994. Source water microbial quality of some vulnerable public ground
water supplies. Proceedings 1994 Water Quality Technology Conference, Part n p. 1425-1436.
Lin SL, Chen HS, Yu CJ, Yen TS. 1995. Legionnaires'disease with acute renal failure: report of two
cases. J Formos Med Assoc. 94(3): 123-126.
Lin YE, Stout JE, Yu YL, Vidic RD. 1998a. Disinfection of water distribution systems for Legionella.
Seminars in Respiratory Infections. 13(2): 147-159.
Lin YE, Vidic RD, Stout JE, Yu VL. 1998b. Legionella in Water Distribution Systems: Regular
culturing of distribution system samples is the key to successful disinfection. J American Water Works
Assoc. 90:112-121.
Lo Presti F, Riffard S, Neyret C, Celard M, VandeneschF, Etienne J. 1998. First isolation in Europe
of Legionella feeleii from two cases of pneumonia. Eur J Clin Microbiol Infect Dis. 17(l):64-66.
Lo Presti F, Riffard S, VandeneschF, Reyrolle M, Ronco E, Ichai P, Etienne J. 1997. The First
Clinical Isolate of Legionella Parisiensis, from a Liver Transplant Patient with Pneumonia J Clin.
Microbiol. 35(7): 1706-1709.
Lowry PW, Blankenship RJ, Gridley W, Troup NJ, TompkinsLS. 1991. A cluster of Legionella
sternal-wound infections due to postoperative topical exposure to contaminated tap water. N Engl J
Med. 324(2): 109-13.
Lowry PW, TompkinsLS. 1993. Nosocomial legionellosis: a review of pulmonary and
extrapulmonary syndromes. Am J Infect Control. 21(l):21-27.
Luck PC, DingerE, Helbig JH, Thurm V, Keuchel H, Presch C, OttM. 1994. Analysis of
Legionella pneumophila strains associated with nosocomialpneumonia in a neonatal intensive care unit.
European Journal of Clinical Microbiology & Infectious Diseases. 13(7):565-571.
Maesaki S, Kohno S, KogaH, Kaku M, Yoshitomi Y, Yamada H, Matsuda H, Higashiyama Y, Kara K,
Seto M, Makaguchi S. 1992. An Outbreak of Legionnaires'Pneumonia in a Nursing Home. Internal
Medicine. 31(4):508-512.
MahoneyFJ, Hoge CW, Farley TA, Barbaree JM, Breiman RF, Benson RF, McFarlandLM. 1992.
Communitywide outbreak of Legionnaires' disease associated with a grocery store mist machine. J
Infect Dis. 165(4):736-739.
MamolenM, Breiman RF, Barbaree JM, Gunn RA, Stone KM, Spika JS, Dennis DT, Mao SH, Vogt
RL. 1993. Use of multiple molecular subtyping techniques to investigate a Legionnaires' disease:
outbreak due to identical strains at two tourist lodges. J Clin Microbiol. 31(10):2584-2588.
IX-14
-------
Mangione EJ, Remis RS, Tait KA, Mcgee HB, Gorman GW, Wentworth BB, Baron PA, Hightower
AW, Barbaree JM, Broome CV. 1985. An outbreak of Pontiac Fever Related to Whirlpool Use.
JAMA. 253 (4):535-539.
Marrie TJ, George J, Macdonald S, HaaseD. 1986. Are Health Care Workers at Risk for Infection
During an Outbreak of Nosocomial Legionnaires' Disease? Am J Infect Control. 14(5):209-213.
Marrie TJ, Haldane D, Bezanson G, PeppardR. 1992. Each water outlet is a unique ecological niche
for Legionellapneumophila. Epidemiol Infect. 108(2):261-270.
Marrie TJ, Peeling RW, Fine MJ, Singer DE, ColeyCM, KapoorWN. 1996. Ambulatory patients
with community-acquired pneumonia: The frequency of atypical agents and clinical course. Am J Med.
101(5):508-515.
Marston BJ, LipmanFffi, BreimanRF. 1994. Surveillance for Legionnaires'disease. Risk factors for
morbidity and mortality. Arch Intern Med. 154(21):2417-2422.
Marston BJ, Plouffe JF, File TM, Hackban BA, Salstrom SJ, Lipman HB, Kolczak MS, Breiman RF.
1997. Incidence of community-acquired pneumonia requiring hospitalization. Results of a population-
based active surveillance study in Ohio. The Community-Based Pneumonia Incidence Study Group.
Arch Intern Med. 157(15): 1709-1718.
Martin RS, Marrie TJ, Haase D, Sumarah RK. 1988. An environmental study of a nosocomial
outbreak of Legionellosis in a city hospital. Can J Public Health. 79(6):440-442.
Martinez E, Domingo P, Ruiz D. 1991. Transient aplastic anaemia associated with Legionnaires'
disease. Lancet. 338(8761):264.
MastroTD, Fields BS, BreimanRF, Campbell J, Plikaytis BD, SpikaJS. 1991. Nosocomial
Legionnaires' disease and use of medication nebulizers. J Infect Dis. 163(3):667-671.
Matsiota-Bernard P, Lefebre C, Sedqui M, Cornillet P, Guenounou M. 1993. Involvement of tumor
necrosis factor alpha in intracellular multiplication of Legionella pneumophila in human monocytes.
Infect Immun. 61(12): 4980-4983.
Meenhorst PL, Reingold AL, Groothuis DG, Gorman GW, Wilkinson HW, Mckinney RM, Feeley JC,
Brenner DJ, VanFurthR. 1985. Water-related nosocomial Pneumonia caused by Legionella
pneumophila serogroups 1 and 10. J Infect Dis. 152(2):356-364.
Meers PD, Goh KT, Lim EW. 1989. Legionella species, serogroups, and subgroups found in the
environment in Singapore. Ann Acad Med Singapore. 18(4):375-378.
Meletis J, Arlet G, DournonE, Pol S, Devergie A, Sportes C, Peraldi MN, Mayaud C, Perol Y,
Gluckman E. 1987. Legionnaires' disease after bone marrow transplantation bone marrow transplant
unit. Bone Marrow Transplant. 2(3):307-314.
IX-15
-------
Memish ZA. Oxley C, Contant J, Garber GE. 1994. Plumbing System Shock Absorbers as a Source of
Legionellapneumophila. Amer Journal of Infection Control. 20(6):305-309.
Mermel LA, Josephson SL, Giorgio CH, DempseyJ, Parenteau S. 1995. Association of
Legionnaires' disease with construction: contamination of potable water? Infect Control Hosp
Epidemiol. 16(2): 76-81.
Mintz CS, Schultz DR, Arnold PI, Johnson W. 1992. Legionella pneumophila \ipopo\ysaccharide
activates the classical complement pathway. Infect. Immun. 60(7):2769-2776.
Miragliotta G, Del Prete R, Sabato R, Cassano A, Carnimeo N. 1992. Legionellosis Associated with
Artesion Well Excavation. European Journal of Epidemiology. 8(5):748-749.
Mitchell P, ChereshskyA, Haskell AJ, Brieseman MA. 1991. Legionellosis in New Zealand: First
Recorded Outbreak. New Zealand Medical Journal. 104(915):275-276.
Miuetzner S, Schwille RC, Farley A, Wald ER, Ge JH, States SJ, Libert T, and R.M. Wadowsky.
1997. Efficacy of thermal treatment and copper-silver ionizati on for controlling Legionella
pneumophila in high-volume hot water plumbing systems in hospitals. American Journal of Infection
Control. 25(6):452-457.
Moiraghi Ruggenini A, Catellani Pastoris M, Dennis PJ, Barral C, Sciacovelli A, Carle F, Bolgiani
M, Passarino G, MingroneMG, Passi C, Lombardo M. 1989. Legionella pneumophila in a hospital
in Torino, Italy: A retorspective one-year study. Epidem Inf. 102(l):21-29.
Moreno C, DeBlasI, Miralles F, ApraizD, Catalan V. 1997. A simple method for the eradication of
Legionella pneumophila from potable water systems. Canadian Journal of Microbiology. 43 (12): 1189-
1196.
MortonS, DyerJV, Bartlett CLR, Bibby LF, Hutchinson DN, Dennis PJ. 1986. Outbreak of
Legionnaires' disease from a cooling water system in a power station (Heysham). Br J Ind Med. (United
Kingdom). 43(9):630-635.
Muder RR, Yu VL, Fang GD. 1989. Community-Acquired Legionnaires'Disease. SeminRespir
Infect. 4(l):32-39.
MuracaPW, Stout JE, Yu VL, Yee YC. 1988. Legionnaires'disease in the work environment:
implications for environmental health. Am Ind Hyg Assoc J. 49(11):584-590.
Muraca PW, Stout JE, Yu VL. 1987. Comparative assessment of chlorine, heat, ozone, and UV light
for killing Legionella pneumophila within a model plumbing system. Appl Environ Microbiol. 53:447-
453. (As cited in Muraca etal. 1990).
Muraca PW, Yu VL, Goetz A. 1990. Disinfection of water distribution systems for Legionella: A
Review of Application Procedures and Methodologies. Infect Control Hosp Epidemiol. 11(2):79-88.
IX-16
-------
Murdoch DR, Walford EJ, Jennings LC, Light GL, Schousboe MI, Chereshsky AY, Chambers ST,
Town GL 1996. Use of the polymerase chain reaction to detect Legionella DNA in urine and serum
samples from patients with pneumonia. Clin Infect Dis. 23(3):475-80.
NadarajahM, GohKT. 1986. Isolation of Legionella pneumophila from hospital cooling towers. Ann
Acad Med Singapore. 15(l):6-8.
NechwatalR, EhretW, Hatte OJ, ZeisslerHJ, Prull A, LutzH. 1993. Nosocomial Outbreak of
Legionellosis in a Rehabilitation Center Demonstration of Potable Water as a Source. Infection.
21(4):235-240.
Negron-Alvira A, Perez-Suarez I, Hazen TC. 1988. Legionella spp. in Puerto Rico cooling towers.
Applied and Environmental Microbiology. 54(10):2331-2334.
NeillMA, Gorman GW, Gibert C, Roussel A, Hightower AW, Mckinney RM, Broome CV. 1985.
Nosocomial Legionellosis, Paris, France: Evidence for Transmission by Potable Water. Am J Med.
78(4):581-588.
Newton LH, Joseph CA, Hutchinson EJ, Harrison TG, Watson JM, Bartlett CL. 1996. Legionnaires'
disease surveillance: England and Wales, 1995. Commun Dis Rep CDR Rev. 6(11)R151-R155.
Nguyen MH, Stout JE, Yu VL. 1991. Legionellosis. Infectious Disease Clinics of North America.
5(3):561-584.
Nguyen MLT, Yu VL. 1991. Legionella infection. Clin Chest Med. 12(2):257-268.
O'Mahony M, Lakhani A, Stephens A, Wallace JG, Young ER, Harper D. 1989. Legionnaires' disease
and the sick-building syndrome. Epidemiological Infections. 103(2):285-292.
O'Mahony MC, Stanwell-Smith RE, Tillett HE, Harper D, Hutchinson JGP, Farrell ID, Hutchinson
DN, LeeJV, Dennis PJ, Duggal HV, Scully JA, Denne C. 1990. The Stafford England UK Outbreak
of Legionnaires'Disease. Epidemiol Infect. 104(3 ):361-380.
OppenheimBA, Sefton AM, Gill ON, Tyler JE, O'mahony MC, Richards JM, Dennis PJL, Harrison
TG. 1987. Widespread Legionella-pneumophila Contamination of Dental Stations in a Dental School
Without Apparent Human Infection. Epidemiol Infect. 99(1): 159-166.
Ortiz-Roque CM, Hazen TC. 1987. Abundance and distribution of Legionellaceae in Puerto Rican
waters. Appl Environ Microbiol. 53(9):2231-2236.
Pai P, Kumar S, Bell GM, Ahmad R. 1996. Rapidly progressive crescentic glomerulonephritis and
Legionnaires'disease. ClinNephrol. 45(3):209-210.
Palmer CJ, Bonilla GF, Roll B, Paszko-Kolva C, Sangemano LR, Fujioka RS. 1995. Detection of
Legionella species in reclaimed water and air with the EnviroAmp Legionella PCR Kit and direct
fluorescent antibody staining. Applied and Environmental Microbiology. 61(2):407-412.
IX-17
-------
Palmer CJ, Tsai Y-L, Paszko-Kolva C, Mayer C, Sangermano LR. 1993. Detection of Legionella
species in sewage and ocean water by polymerase chain reaction, direct fluorescent-antibody, and plate
culture methods. Applied and Environmental Microbiology. 59(11):3618-3624.
Palmer SR, Zamiri I, Ribeiro CD, Gajewska A. 1986. Legionnaires' disease cluster and reduction in
hospital hot water temperatures. Br Med J (Clin Res Ed). 292(6534): 1494-1495.
Pan TM, Yea HL, Huang HC, Lee CL, Horng CB. 1996. Legionellapneumophila infection in Taiwan:
A preliminary report. Journal of the Formosan Medical Assoc. 95(7):536-539.
Parent! CM, Richards SW, Hoidal JR, Niewoehner DE. 1989. Long-term Pulmonary Sequelae after
Legionella pneumophila Infection in the Hamster. 139(4):988-995.
Passi CR, Maddaluno R, Castellani Pastoris M. 1990. Incidence of Legionellapneumophila infection
in tourists: Italy. Public Health. 104(3)183-188.
Paszko-Kolva C, ShahamatM, Col well RR. 1993. Effect of temperature on survival of Legionella
pneumophila in the aquatic environment. Microb Releases. 2(2):73-79.
Paszko-Kolva C, YamamotoH, ShahamatM, Sawyer TK, Morris G, Col well RR. 1991. Isolation of
Amoebae and Pseudomonas-spp and Legionella-spp from Eyewash Stations. Appl Environ Microbiol.
57(1):163-167.
Patterson WJ, Seal DV, CurranE, Sinclair TM, McLuckie JC. 1994. Fatal nosocomial Legionnaires'
disease: relevance of contamination of hospital water supply by temperature-dependent buoyancy-driven
flow from spur pipes. Epidem Infect. 112(3):513-525.
Patton CM, Johnson SR, Kaufmann AF. 1979. Susceptibility of laboratory rodents, birds, and the
rabbit to Legionella pneumophila (Legionnaires' Disease bacterium). Presented at the Annual Meeting
of the American Society of Microbiology, 1979. Paper number 879. (As cited in EPA 1985).
Pedro-Botet ML, Sabria-Leal M, Haro M, Rubio C, Gimenez G, Sopena N, Tor J. 1995. Nosocomial
and community-acquired Legionella pneumonia, clinical comparative analysis. Eur Respir J.
8(11):1929-1933.
Pedro-Botet ML, Sabria-Leal M, SopenaN, Manterola JM, Morera J, Blavia R, Padilla E, Matas L,
Gimeno JM. 1998. Role of immunosuppression in the evolution of Legionnaires'disease. Clin Infect
Dis. 26(1):14-19.
Peiris V, Prasad MKD, Bradley D, Zaqistowicz W, Sivayoham S, Naqvi SNH, Hutchinson DN. 1992.
Legionnaires' Disease in Elderly People: The First Sign of an Outbreak in the Community? Age and
Ageing. 21(6):451455.
Pellizari VH, Martins MT. 1995. Occurrence of Legionella sp in water samples from man-made
systems of Sao Paulo. Revista de Microbiologia. 26(3):186-191.
IX-18
-------
Prodinger WM , BonattiH, Allerberger F, Wewalka G, Harrison TG, Aichberger C, Dierich MP,
Margreiter R, Tiefenbrunner F. 1994. Legionellapneumonia in transplant recipients: a cluster of cases
of eight years'duration. JHosp Infect. 26(3): 191-202.
Rechnitzer C. 1994. Pathogenetic aspects of Legionnaires' Disease: Interaction of Legionella
pneumophila with cellular host defences. APMIS Suppl. 43:1-43.
ReddSC, LinFYC, Fields BS, BiscoeJ, Plikaytis BB, Powers P, Patel J, Lim BP, Joseph JM,
Devadason C, Israel E, Cohen ML. 1990. A Rural Outbreak of Legionnaires'Disease Linked to
Visiting a Retail Store. Amer Journ of Public Health. 80(4):431-434.
Reynolds HY. 1996. Respiratory infections: Community-acquired pneumonia and newer microbes.
Lung. 174(4):207-224.
Ribeiro CD, Burge SH, Palmer SR, Tobin JO, Watkins ID. 1987. Legionella-pneumophila'ma
Hospital Water System Following a Nosocomial Outbreak Prevalence Monoclonal Antibody
Subgrouping and Effect Of Control Measures. Epidemiol Infect. 98(3):253-262.
Roig J, Carreres A, Domingo C. 1993. Treatment of Legionnaires'disease. Current recommendations.
Drugs. 46(l):63-79.
RoigJ, Domingo C, MoreraJ. 1994. Legionnaires'disease. Chest. 105(6):1817-1825.
RufB, Schuermann D, Horbach I, Seidel K, Pohle HD. 1988. Nosocomial Pneumonia
Demonstration of Potable Water as the Source of Infection. Epidemiol Infect. 101(3):647-654.
Schlech WF III, Gorman GW, Payne MC, Broome CV. 1985. Legionnaires' Disease in the Caribbean
an Outbreak Associated with a Resort Hotel. Arch Intern Med. 145(11):2076-2079.
Seidel K, Baez G, Boernert W, Seeber E, Seifert B, Esdorn H, Fischer M, Rueden H, Wegner J(Eds.).
1987. Legionellae in aerosols and splashwaters in different habitats. Conference Title: INDOOR AIR
'87: 4th international conference on indoor air quality and climate. Berlin, F.R. Germany. (l):690-693.
Shah A, Check F, Baskin S, Reyman T, MenardR. 1992. Legionnaires'disease and acute renal
failure: case report and review. Clin Infect Dis. 14(1):204-207.
Shelton BG, Flanders WD, Morris GK. 1994. Legionnaires'disease outbreaks and cooling towers with
amplified Legionella concentrations. Current Microbiology. 28(6):359-363.
Shelton BG, Morris GK, Gorman GW. 1993. Reducing Risks Associated with Legionella Bacteria in
Building Water Systems. Legionella Current Status and Emerging Perspectives, 4th International
Symposium on Legionella, Orlando, Florida. Jan. 26-29, 1992. American Society for Microbiology.
0(0):279-281.
Shuval HI, Fattal B, BercovierH. 1988. Legionnaires Diseases and the Water Environment in Israel
Water Science and Technology. 20(11-12):33-38.
IX-19
-------
Skerrett SJ, Martin TR. 1996. Roles for tumor necrosis factor alpha and nitric oxide in resistance of rat
alveolar macrophages to Legionellapneumophila. Infect Immun. 64(8):3236-3243.
Skerrett SJ, Martin TR. 1991. Alveolar macrophage activation in experimental legionellosis. J
Immunol. 147(l):337-345.
States SJ, ConleyLF, Knezevich CR, Keleti G, Sykora JL, Wadowsky RM, Yee RB. 1989. Free-
Living Amoebae in PublicWater Supplies: Implications for Legionella, Giardia, and Cryptosporidia.
Proceedings Water Quality Technology Conference Advances in Water Analysis and Treatment. St.
Louis, Missouri, November 13-17, 1988. p 109-125.
States SJ, ConleyLF, KuchtaJM, OleckBM, Lipovich MJ, Wolford RS, Wadowsky RM,
McNamara AM, Sykora JL, Keleti G, Yee RB. 1987. Survival and Multiplication of Legionella
pneumophila in Municipal Drinking Water Systems. Appl Environ Microbiol. 53(5): 979-986.
Steel e, Trevor W, LanserJ, SangsterN. 1990. Isolation of Legionella longbeachae Serogroup 1 from
Potting Mixes Appl Environ Microbiol. 56(1): p49(5).
Stout JE, Yu VL. 1997. Current Concepts (Review Article): Legionellosis. NEngl JMed. 337:682-
687.
Stout JE, Yu VL, Best MG. 1985a. Ecology of Legionella pneumophila within Water Distribution
Systems. Appl. Environ. Microbiol. 49(l):221-228.
Stout JE, Yu VL, Muraca P. 1985b. Isolation of Legionella pneumophila from the Cold Water of
Hospital Ice Machines: Implications for Origin and Transmission of the Organism. Infection Control.
6(4):141-146.
Stout JE, Yu VL, MuracaP, Joly J, TroupN, TompkinsLS. 1992a. Potable water as a cause of
sporadic cases of community-acquired legionnaires' disease. NEngl JMed. 326(3):151-155.
Stout JE, Yu VL, VickersRM, Zuravleff J, Best M, Brown A, Yee RB, Wadowsky R. 1982.
Ubiquitousness of Legionella pneumophila In The Water Supply Of A Hospital With Endemic
Legionnaires'Disease. NEngl JMed. 306:466-468.
Stout JE, Yu VL, Yee YC, Vaccarello S, Diven W, Lee TC. 1992b. Legionella pneumophila in
residential water supplies: environmental surveillance with clinical assessment for Legionnaires' disease.
Epidemiol Infect. 109(l):49-57.
StrampferMJ, Tu RP. 1988. Nosocomial Legionnaires'disease. Heart Lung. 17(6 Pt 1):601-604.
Straus WL, Plouffe JF, File TM, Lipman HB, Hackman BH, Salstrom S, Benson RF, Breiman RF.
1996. Risk factors for domestic acquisition of Legionnaires Disease. Arch Intern Med. 156:1685-1692.
Struelens MJ, Maes N, RostF, Deplano A, Jacobs F, Liesnard C, Bornstein N, Grimont F, Lauwers S,
Mclntyre MP, Serruys E. 1992. Genotypic and Phenotypic Methods for the Investigation of a
IX-20
-------
Nosocomial Legionella-pneumophila Outbreak and Efficacy of Control Measures. J Infect Dis.
166(1):22-30.
Surman SB, Morton LHG, Keevil CW. 1994. The dependence of Legionellapneumophila on other
aquatic bacteria for survival on R2A medium. International Biodeterioration & Biodegradation.
33(3):223-236.
Susa M, Ticac B, Rukavina T, Doric M, Marre R. 1998. Legionella pneumophila Infection in
Intratracheally Inoculated T Cell-depleted or-Nondepleted A/j Mice. J Immunol. 160(1): 316-321.
Ta AC, Stout JE, Yu VL, Wagener MM. 1995. Comparison of Culture Methods for Monitoring
Legionella Species in Hospital Potable Water Systems and Recommendations for Standardization of
SuchMethods. Journal of Clinical Microbiology. 33(8):2118-2123.
Tang P, Boleszczuk P, Brodsky M H, Krishnan C. 1995. Outbreaks of legionellosis associated with
Legionella sainthelensi serogroup 1 in Ontario, Canada. Abstracts of the General Meeting of the
American Society for Microbiology. 95(0):56.
Thomas DL, MundyLM, Tucker PC. 1993. Hot Tub Legionellosis. Legionnaires'Disease and
Pontiac Fever after a Point-Source Exposure toL. Pneumophila. Arch InternMed. 153(22):2597-2599.
Tiefenbrunner F, Arnold A, DierichMP, Emde K. 1993. Occurrence and Distribution of Legionella-
pneumophila in Water Systems of Central European Private Homes. Barbaree JM, Breiman RF,
Dufour AP (Ed.). Legionella. Current Status and Emerging Perspectives, 4th International Symposium
on Legionella, Orlando, Florida. January 26-29, 1992. American Society for Microbiology.
Washington, DC. 0(0):235-238.
Tobin RS, Ewan P, Walsh K, Dutka B. 1986. A Survey of Legionella pneumophila in Water in 12
Canadian Cities. Water Research. 20(4):495-501.
Twisk-Meijssen MJM, Meenhorst PL, van Cronenburg BJ, Mulder JD, Scheffer E, Van Furth R. 1987.
The Course of Legionella pneumophila in Guinea Pigs after Inhalation of Various Quantities of
L.pneumophila. Immunobiol. 176:108-124.
Vandenesch F, Surgot M, Bornstein N, Paucod JC, Marmet D, Isoard P, Fleurette J. 1990. Relationship
between free amoeba and Legionella: studies in vitro and in vivo. Zentralbl Bakteriol. 272(3):265-275.
Venezia RA, Agresta MD, Hanley EM, Urquhart K, Schoonmaker D. 1994. Nosocomial Legionellosis
Associated with Aspiration of Nasogastric Feedings Diluted with Tap Water. Infect Control Hosp
Epidemil. 15(8):529-533.
Verissimo A, Marrao G, Gomes da Silva F, da CostaMS. 1991. Distribution of Legionella spp. in
hydrothermal areas in continental Portugal and the island of Sao Miguel, Azores. Applied and
Environmental Microbiology. 57(10):2921-2927.
IX-21
-------
Vickers RM, Yu VL, Hanna SS, MuracaP, Diven W, Carmen N, Taylor FB. 1987. Determinants of
Legionella pneumophila contamination of water distribution systems: 15-hospital prospective study.
Infect Control . 8(9):357-363.
Voss L, Button KS, Lorenz RC, Tuovinen OH. 1986. Legionella Contamination of a Preoperational
Treatment Plant. Journal of the American Water Works Assoc. 78(l):70-75.
Voss L, Button KS, Tuovinen, OH. 1985. Legionella pneumophila in a metropolitan water
distribution system. Environ Tech Letter. (10):429-438.
WaldorM, WilsonB, Swartz M. 1993. Cellulitis caused by Legionella pneumophila. Clin Infect Dis.
Warner CL, Fayad PB, HeffnerRRJr. 1991. Legionella myositis. Neurology. 41(5):750-752.
Watson JM, Mitchell E, GabbayJ, MaguireH, Boyle M, Bruce J, Tomlinson M, Lee J, Harrison TG,
Uttley A, O'Mahony M, Cunningham D. 1994. Piccadilly Circus Legionnaires' Disease Outbreak.
Journal of Public Health Medicine. 16(3): 34 1-347.
WegmullerE, Weidmann P, Hess T, Reubi FC. 1985. Rapidly progressive glomerulonephritis
accompanying Legionnaires' disease. Arch Intern Med. 145(9):171 1-1713.
Whitney CG, HofmannJ, Pruckler JM, Benson RF, Fields BS, Bandyopadhyay U, Donnally EF,
Giorgio-Almonte C, Mermel LA, Boland S, Malyas BT, BreimanRF. 1997. The role of arbitrarily
primed PCR in identifying the source of an outbreak of Legionnaires' disease. Journal of Clinical
Microbiology. 35(7): 1800-1804.
WHO. 1990. Epidemiology, Prevention, and Control of Legionellosis: Memorandum From a WHO
Meeting. Bulletin of the World Health Organization. 68(2): 155-164.
WinnWCJr. 1993. Legionella and the clinical microbiologist. Infect Dis Clin North Am. 7(2):377-
92.
Winn WC Jr. 1988. Legionnaires disease: Historical Perspective. Clin Microbiol Rev. 1(1):60-81.
WinnWCJr., Davis GS, Gump DW, Craighead JE, Beaty HN. 1982. Legionairres' pneumonia after
intratracheal inoculation of guinea pigs and rats. Lab Invest. 47(6): 568-578.
Winter JH, McCartney AC, Fallen RJ, Telfer ABM, Drury JK, Reece IJ, TimburyMC. 1987.
Rapid Diagnosis of an Outbreak of Legionnaires' Disease at Glasgow Royal Infirmary. Thorax.
42(8):596-599.
WitherellLE, Duncan RW, Stone KM, Stratton LJ, Orciari L, Kappel S, Jillson DA. 1988.
Investigation of Legionella pneumophila in Drinking Water. JournaL AWWA. 80(2): 87-93.
IX-22
-------
Woo AH, Goetz A, Yu VL. 1992. Transmission oi Legionella by Respiratory Equipment and Aerosol
Generating Devices. Chest. 102:1586-1590.
Woo AH, Yu VL, Goetz A. 1986. Potential in-hospital modes of transmission of Legionella
pneumophila. Demonstration experiments for dissemination by showers, humidifiers, and rinsing of
ventilation bag apparatus. AmJMed. 0(4): 567-573.
Yu VL. 1997. Prevention and control of Legionella: An idea whose time has come [editorial]. Infect.
Dis. Clin. Pract. 6(7):42(M21.
Yu VL. 1993. Could aspiration be the major mode of transmission for Legione Hal AmJMed.
YuVL, LiuZ, Stout JE, Goetz A. 1993. Legionella disinfection of water distribution systems:
principles, problems, and practice. Infection Control and Hospital Epidemiology. 14(10):567-570.
Zacheus OM, Martikainen PJ. 1994. Occurrence of legionellae in hot water distribution systems of
Finnish apartment buildings. Can J Microbiol/Rev Can Microbiol. 40(12): 993-999.
IX-23
-------
(this page intentionally left blank)
IX-24
------- |