United States
Environmental Protection
Agency
Health Effects Research
Laaboratory
Cincinnati OH 45268
Research and Development
EPA-600/S1-81-032 May 1981
Project Summary
Health Effects of Sewage
Aerosols: Additional
Serological Surveys and
Search for Legionella
pneumophila In Sewage
R. Northrop, C. Becker, R. Cordell, M. Sulita, N. Altman, R. Anderson, and
J. Kusek
Antibody levels to Legionella pneu-
mophila. serogroup 1 and Hepatitis A
virus (HAV) were determined for 433
persons living within a 1.6 km radius
of an activated sludge plant. Sera of
children 6 to 13 years of age were also
tested for antibody to Norwalk virus.
The antibody prevalence rates for L.
pneumophila. HAV and Norwalk virus
were 23.2%, 31.5% and 7.1%, respec-
tively. The prevalence of L. pneumo-
phila antibody was higher than ex-
pected. Seroconversions were not
observed except for a 4-fold rise in
HAV antibody in one pair of sera,
suggesting new infections did not
occur in the 8-month study period by
any of these agents.
Using an index of exposure for the
number of viable organisms generated
at the plant to which participants were
exposed, the average indexfor persons
with antibody to L. pneumophila, or
HAV, was not statistically different
from the averages of those without
antibody.
In addition, there was no association
between how close people lived to the
plant, or how long they lived in the
study area and being seropositive.
These findings augment but still did
not prove the previous findings that
sewage aerosols emitted from this
plant had no overt health effects.
In search of L. pneumophila in sew-
age and aerosols, guinea pigs were
inoculated with the particulates of
sewage and air samples collected at
the plant. The inoculum was adjusted
to contain not more than 106 organisms
by standard plate count. In one experi-
ment 4 animals were inoculated with
undiluted sewage. L. pneumophila
was identified in spleen cells by direct
fluorescent antibody (DFA) 6-7 days
later. After 10-fold dilution of the
sample none of the 4 animals had
evidence of infection. In a second
experiment using undiluted inocula, 2
of 4 guinea pigs developed infection
as determined by OFA. Infections
were not detected in animals inoculated
with the respective aerosol samples.
These findings suggested that viable
L. pneumophila or a related agent
occurs in sewage.
This Project Summary was developed
by EPA's Health Effects Research
Laboratory, Cincinnati, OH, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back.).
-------
Introduction
Objectives of Study
This study is an addendum to a previ-
ous study on the health effects of
sewage aerosols carried out in the
vicinity of the North Side Sewage Treat-
ment Works (NSSTW) in Skokie, Illinois.
At the time of the original study, serum
samples were collected from 433 per-
sons in the study population. The sera
were initially tested for antibody to
several enteroviruses in an attempt to
discover whether there was a relation-
ship between the level of exposure to
sewage aerosols and infection with
these enteroviruses. In addition, attempts
were made to detect Legionnaires'
Disease Bacteria (LBD) in sewage and
sewage aerosols.
The purpose of this addendum was to
determine: 1) whether there was a
relationship between exposure to sew-
age aerosols and prevalence of antibodies
to Legionella pneumophila, Hepatitis A
virus and Norwalk virus; 2) to determine
if there were bactericidal substances in
sewage which would have precluded
the occurrence of LDB; 3) to evaluate if
the reported methodologies for recover-
ing LDB from environmental samples
could be applied to sewage; and 4) to
attempt to isolate and identify LDB in
sewage and aerosols from the NSSTW.
Study Design
An area within a 1.6-kilometer radius
of the NSSTW was designated as the
study area, whwch was estimated to
consist of approximately 16,000 persons.
The design and conduct of this main
study has been described in detail (EPA-
600/1-79-019). Of the 16,000 possible
residents in the study area a stratified
random sample of 2,378 (approximately
1 5%) were recruited for a questionnaire
survey. Of these, 318 persons voluntarily
gave two blood samples, one at the
beginning and a second at the end of the
eight month study period. An additional
115 people gave one blood sample either
at the beginning or at the end of the
study. Sera of these 751 bloods were
tested for presence of antibody to
Legionella pneumophila or Hepatitis A
virus. Blood was not collected from
children under six years of age and only
blood samples from children six to
twelve years of age were tested for
antibody to Norwalk virus.
The presence of antibody in the first or
only blood collected was used to calcu-
late prevalence rates of immune indi-
viduals. A four-fold rise in antibody
between paired sera was taken as
evidence of recent infection and was
used to calculate incidence rates of
infection. Rates were then evaluated
with respect to the demographic char-
acteristics, income, residence, occupa-
tion, underlying health conditions, and
exposure index to viable aerosol particles
documented from the questionnaire,
health or environmental surveys previ-
ously described. From these data sets a
dose-response relationship between
different levels of exposure to sewage
plant emissions and illness and infection
rates was investigated. Standard tech-
niques such as regression and analysis
of variance were used to relate the
health and environmental data.
Isolation of Legionella pneumophila
was also attemped two years after the
sera were collected. Grab samples of
sewage influent in the aeration basin of
the NSSTW were obtained. Upwind and
downwind air samples were collected
during the time the influent traveled
through the basin. Following suitable
preparation in the laboratory, the par-
ticulates of sewage and air samples
were inoculated intra-pentoneally into
separate groups of guinea pigs. It was
necessary to determine whether or not
guinea pigs would survive the insult of
resident organisms in these samples
and subsequently display evidence of
infection by the slow growing Legionella
pneumophila. Evidence of infection
would be based on a rise in rectal
temperature and/or death, and dem-
onstration of Legionella pneumophila
antigen in peritoneal cells and splenic
tissue by direct fluorescent antibody
techniques. Preliminary and parallel
experiments were conducted to address
the other objectives described above
and to facilitate interpretation of the
results.
Background and Literature
Review
Legionella pneumophila
An epidemic of pneumonia at the
Legionnaries'convention in Philadelphia
in the summer of 1976 led to the
discovery of the Legionnaires' Disease
Bacterium now called Legionella pneu-
mophila. Investigations of this and other
epidemics caused by this bacterium
have produced evidence implicating an
environmental source. Isolation of the
bacterium from the environment during
or following the epidemic was possible
in a few cases. Some serological studies
of populations exposed to Legionella
pneumophila in these epidemics were
done with varying results. The bacterium
has been found in nonepidemic-related
habitats such as lakes, streams and
soils. It has also been isolated from
residual water in cooling towers of air
conditioning units in association with
disease outbreaks, suggesting that
disease in man results from exposure to
aerosols containing the agent. It was
deemed useful to study a population of
healthy individuals living around an
activated sludge plant with a defined
environmental exposure to sewage
aerosol in a non-epidemic situation.
Hepatitis A
Although the Hepatitis A virus has not
yet been officially classified, evidence
indicates that it is most likely an entero-
virus of the picorna- or parvovirus type.
While enteroviruses have been infre-
quently identified in sewage aerosols to
date, they have been found in abundance
in sewage. Primary and trickling filter
treatments have only a modest effect on
virus levels in sewage. While activated
sludge treatment removes 90 to 98% of
the enteroviruses there is concern that
a large number of potentially infectious
virus particles may be aerosolized in the t
process, creating a health hazard for
sewage treatment plant workers and
persons residing nearby. The respirable
particles may become trapped in the
nasopharynx of exposed persons, an
appropriate site for an enterovirus to
initiate an infection. Thus, the danger of
Hepatitis A virus infection occurring
through inhalation of the virus into the
nose or mouth is a possibility.
Little, if any, work has been done
concerning the occurrence of Hepatitis
A virus in sewage aerosols. Although a
study was done in Israel which suggested
a population living near a wastewater
irrigation system experienced two to
four times the incidence of infectious
hepatitis compared to a control popula-
tion, the interpretation of these findings
awaits reanalysis. It was thus felt that
since sera and environmental data were
available from the original NSSTW
study, a serological screening for anti-
body prevalence in the study population,
as related to exposure indices would be
important. Serological analysis would
be an appropriate indication of experi-
ence with the Hepatitis A virus since
most infections are subclinical and are '
not manifested as overt disease. _
-------
Norwalk Virus
Norwalk virus is one of the viral
agents recently shown to be associated
with epidemics of gastroenteritis in
older children and adults. Infections,
either clinical or subclinical, are rela-
tively infrequent (<5%) in children
before age 10 years while 50-60% of
adults have serological evidence of such
infections. The virus isa small, parvovirus-
hke particle, apparently spread by the
fecal-oral route and shed to maximal
amounts in feces during episodes of
diarrhea.
Since the Norwalk virus occurs in
stools, it may be present in sewage and
possibly aerosols for spread to exposed
persons around sewage treatment
works. Sera of children 1 2 years of age
and less in the study population were
tested to determine if the prevalence of
a 5-10% antibody would be found as
expected from previous studies.
Rotaviruses, which are Reovirus-like
viruses, have also been associated with
epidemics of diarrhea in infants and
children less than 3 years of age. Sero-
epidemiological studies have shown that
over 70% of children in the 3-5 year old
age group have Rotavirus antibody.
There was initial interest in determining
Rotavirus antibody prevalence in the
study population aged 12 years and
less, but since blood was not collected
from children under 6 years of age, it
was decided to withhold those deter-
minations until the Norwalk virus anti-
body data were evaluated.
Conclusions
1. The prevalence of antibody of Le-
gionella pneumophila, serogroup 1,
in the study population was higher
than expected (23.2%), but there was
no association between total viable
particle exposure index or length of
residency in the study area to impli-
cate aerosols from the sewage treat-
ment plant as a source of exposure.
2. New cases (incidence) of Legionella
pneumophila, based on a 4-fold
antibody rise over the 8-month
observation period, were not detected.
3. Prevalence of antibody to Hepatitis A
virus was found to be 31.5% which
was an expected rate for that study
population. Personal exposure
indices or length of residency m the
study area were not associated with
having antibody.
4. One new infection with Hepatitis A
virus was detected serologically
' US GOVERNMENT PRINTING OFFICE 1961-757-012/7102
suggesting a two-to-three fold higher
incidence rate than expected, but the
:ase was considered a spu-
single
rious fihding
5. Norwalk
year
old
virus antibody in the 6-12
age group occurred at an
expected prevalence rate (7.1 %)sug-
gestinc no unusual occurrence of
infection by this gastroenteritis virus
in the population.
6. In two experiments, pathogenic
Legionella pneumophila was tenta-
tively recovered from grab samples
of sewage entering the aeration
basin of an activated sludge plant.
7. A liqu d medium containing the
particu ates captured from approxi-
mately 200 M3 of air collected 50
meters upwind or downwind from
the sewage aeration basins did not
contain viable, pathogenic/lep/oAJe//a
pneumophila when inoculated into
guinea
The parent study, entitled Health
Effects of Aerosols Emitted from an
Activated Sludge Plant, Grant No
805003, is available as EPA-600/1-79-
019 from the National Technical Infor-
mation Service, Order No. PB 299583.
pigs
Recommendations
1. Prospective studies of populations
newly exposed to sewage aerosols
would jrovide more definitive infor-
mation on the risk of infection by
Legionella pneumophila, Hepatitis A
or any of the gastroenteritis viruses.
2. Since the epidemiology of Legion-
naires' Disease is poorly understood,
it wou d contribute to this under-
Standing to determine the occurrence
of Legicnellapneumophila in sewage.
Further studies are needed to deter-
mine tr e source of the organism, i.e.,
man 01 the environment; seasonal
variation in its occurrence; and
conditions that enhance its patho-
genicity in man.
R Nortnrop, C. Becker, R. Cordell, M. Sulita, N. Altman, R. Anderson, and J.
Kusei;, are with the School of Public Health, University of Illinois at the Medical
Center, Chicago, IL 60608.
Walter Jakubowski is the EPA Project Officer (see below).
The conplete report, entitled "Health Effects of Sewage Aerosols: Additional
Seroligical Surveys and Search for Legionella pneumophila m Sewage,"
(Orde<- No. PB 81-179 830; Cost. $6 50, subject to change) will be available
only Irom:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone- 703-487-4650
The EP/fi Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
HElilUN 5
230 b DEAPbUH".
GO IL
------- |