PreprintPDF Available

Occupational Immunity and Natural Vaccination

Preprints and early-stage research may not have been peer reviewed yet.


People who work with animals are frequently exposed to dangerous pathogens. Disease and subsequent immunity may result. Alternatively, occupational exposure to animals may lead to natural vaccination: the acquisition of immunity in the absence of overt disease. We use anthrax, Q fever, Campylobacter and influenza to illustrate aspects of dose, route and frequency of exposure that may be particularly favorable to natural vaccination. We then explore how exposure and immunity in those who work with animals provide clues about the epidemiology of emerging infectious diseases.
Electronic copy available at:
Occupational Immunity and Natural Vaccination
Robin M. Bush* & Steven A. Frank*
*Department of Ecology and Evolutionary Biology, University of California, Irvine, CA
92697-2525 USA
People who work with animals are frequently exposed to dangerous pathogens.
Disease and subsequent immunity may result. Alternatively, occupational exposure to
animals may lead to natural vaccination: the acquisition of immunity in the absence of
overt disease. We use anthrax, Q fever, Campylobacter and influenza to illustrate
aspects of dose, route and frequency of exposure that may be particularly favorable to
natural vaccination. We then explore how exposure and immunity in those who work
with animals provide clues about the epidemiology of emerging infectious diseases.
Emerging infectious diseases arise from zoonotic pathogens that transmit from animals to
humans. The most obvious recent zoonotic threat to human health comes from avian
influenza. But the potential for zoonotic assault spans a wide array of pathogens including
SARS, Ebola, anthrax, HIV, monkeypox, and the diverse bacteria of common farm animals
that sometimes cause severe enteric or neural damage in humans. Zoonotic pathogens are
also among the most commonly listed agents for use as bioterror weapons.
Yet, for all the threat that zoonotic pathogens pose to humans, there are entire
working classes of people who are frequently and in some cases continually exposed to
zoonotic agents, including veterinarians, farmers, ranchers, tanners, and food processors.
These people usually make it through the working day without incident, or so it seems.
Perhaps frequent zoonotic exposure and relatively rare disease per exposure occur because
successful infection is rare. But in some cases, natural vaccination may arise by infection
Electronic copy available at:
and subsequent acquisition of immunity in the absence of overt disease 1. Such immunity
would alter the dynamics of infection and the spread of disease in the first line of contact
with zoonotic pathogens.
We searched the literature on occupational exposure to zoonotic pathogens. We found
surprisingly few reports about the patterns and mechanisms of exposure and the
consequences for immunity. Because occupational exposure may be the primary source of
many emerging infectious diseases, there is great need for more information about this
subject. In this paper, we develop a conceptual framework to clarify what we need to know
about the sporadic exposures at the source of emerging infectious disease. We use three key
questions to organize the observations and concepts.
First, do occupational exposures to zoonotic pathogens actually lead to higher levels
of immunity than observed in the rest of the population? For example, veterinarians
encounter more zoonotic pathogens than the average individual. But do they show
serological evidence that they have been infected by, and developed immunity to, those
Second, if occupationally exposed individuals show higher levels of immunity to
zoonotic pathogens, was that immunity more likely to have been acquired by illness or by
subclinical infection? If occupational exposure leads to illness, the resulting immunity
would not be considered, by our definition, natural vaccination.
Third, are particular aspects of exposure, such as dosage, route of inoculation, or
frequency of exposure, more likely to cause in subclinical (asymptomatic) infection with
resulting immunity (natural vaccination) as opposed to disease 1? If the answer to the third
Electronic copy available at:
question is yes, it might be possible to utilize these mechanisms in the prevention of
zoonotic disease.
We develop four case studies. Anthrax is an occupational hazard historically confined
to ranchers and tannery workers, but has recently posed challenges to postal workers and
hazardous material crews. Q fever is a bacterial disease transmitted from cattle, sheep and
goats. Campylobacter jejuni is a zoonotic bacterial pathogen that causes a significant
fraction of infections leading to severe gastroenteritis. Influenza A transmits from wild
birds to domestic animals, presenting an occupational threat to the farmers, veterinarians
and others who work with these animals.
Each case illustrates some of the key attributes of occupational immunity and natural
vaccination. Infection occurs by various routes, and the route of infection usually influences
the severity of disease. Dosage varies and may be related to the route of infection,
influencing whether subclinical or full-blown disease results. The frequency of exposure
differs by occupation and may cause variation in the boosting of immunity. From these
varied observations, we paint a picture of occupational exposure, illness and immunity.
Overall, we conclude that, for some diseases, natural vaccination probably occurs
frequently, but that zoonotic pathogens differ in the amount of immunity they induce and in
the pathways by which such immunity develops. We believe that the consequences of
different routes of infection have been particularly neglected 1. Further study of this topic
will provide insight into the frequency of natural vaccination in those subpopulations most
at risk for zoonotic infection, who form the front line in the spread of emerging infectious
Electronic copy available at:
Anthrax is caused by Bacillus anthracis, a bacterium that typically infects grazing animals
such as cattle, sheep and goats. Infection by B. anthracis can occur through inhalation of
spores from infected animals or animal products such as hides, by cutaneous infection from
handling these products, or by ingestion of undercooked meat. Prompt treatment with
antibiotics generally cures anthrax. Untreated inhalational anthrax has a mortality rate
above 50%; gastrointestinal and cutaneous cases cause much lower mortality rates. Thus
the route by which infection occurs greatly affects disease outcome.
There are few infected animals today in the developed industrial nations. Most recent
exposure to anthrax in the US resulted from bioterrorism by spore-laden letters that infected
mail recipients and postal workers. Recent outbreaks in bison and cattle in Canada arose
from anthrax spores, which can live in the soil for many years; such outbreaks pose risk to
ranchers and disposal crews. Rare sporadic inhalational cases develop from a variety of
sources, such as exposure to spore-laden cow hides imported from Africa for constructing
drums 2.
In developing countries, most inhalational cases are similar to those reported recently
in cattle processors in Kazakhstan, which arose from exposure to cattle infected by soil-
borne spores from old outbreaks 3. Cutaneous and gastrointestinal cases are more common
and occur sporadically in many developing countries (CDC web site).
Most information about occupational exposure to anthrax comes from studies in the
US during the years 1900-1960. In several studies of animal hair and hide workers, the rate
of disease was low in spite of continuous exposure to air-borne spores 4-6.
Electronic copy available at:
One study suggested that workers may have inhaled up to 510 spores per working day
7. Another study reported that a significant fraction of workers had spores in anterior nasal
swabs and pharyngeal washings 8. In three mills, up to 66% of the surfaces of the animal
material handled by the workers had spores, suggesting common skin exposure 9. This
series of papers supports the idea that exposure among animal hide and hair workers was
common, but that symptomatic inhalational disease was rare.
We now turn to our three questions about occupational exposure and natural
vaccination. First, can occupational exposure to anthrax actually lead to immunity? The
answer is yes; we discuss the evidence in conjunction with the next question.
Second, is there is any evidence of subclinical infection by anthrax? 10 made the
strongest case for this point of view, based on their study of a goat hair processing mill in
New Hampshire, USA, following an outbreak of inhalational anthrax in 1957. They
measured antibody titers against the protective antigen of anthrax in unaffected workers
during the three months following the outbreak.
10 divided workers into three classes: prior anthrax victims, vaccinated individuals,
and unaffected individuals. Among eleven individuals who had symptomatic anthrax more
than two years before testing, only two tested positive for antibody titers. In serial studies
of cutaneous anthrax, four of five individuals reverted from detectable to undetectable
antibody titers after three months. Only 15 of 33 vaccinated individuals had detectable
antibodies. The vaccinated individuals were sampled just before a six-month booster shot
was due. The rapid waning of antibody titers in prior cases and in vaccinated individuals
supports the observation that protective immunity decays over the year following
inoculation 11.
Electronic copy available at:
10’s most interesting result concerns unaffected workers. Among those without
symptoms, 11 of 72 had detectable antibodies. Those individuals with raised titers worked
mainly in the regions of the factory with the highest air-borne concentrations of spores.
Given that detectable antibodies appear to wane rapidly after infection, the 11 positive tests
suggest a high frequency of subclinical infection or boosting of immunity by continual
exposure. In addition, among 56 unvaccinated and unaffected individuals from three other
processing plants, 8 had positive titers, among which 4 worked in the most intensively
exposed part of the plant. To test for the possibility of false-positive reactions, samples of
242 unexposed individuals from a military base were included in the samples examined in
the laboratory, without any labeling to distinguish exposed from unexposed individuals.
None of the 242 unexposed samples yielded positive titers, suggesting that false-positive
results must be very rare by the methods used, supporting the conclusion of subclinical
From these results, 10 conclude: “Spontaneous recovery from inhalation anthrax has
been reported 12, 13 and is common in cutaneous anthrax so that it seems possible that the
disease may be manifested by lesions so minor as to go unnoticed but sufficient to cause a
serologic response.”
Although 10 provides a convincing and well designed analysis of subclinical infection,
no other study has ever turned up such clear evidence. Subsequent authors who focused on
inhalational anthrax in industrialized countries tend to invoke or reject the importance of
subclinical infection in a haphazard way, without significantly advancing the subject.
Third, do particular mechanisms of infection, such as route of inoculation, bias
outcome toward subclinical infection or overt disease? Most studies of anthrax focus on
Electronic copy available at:
inhalational inoculation, because that route has the most severe symptoms. However,
cutaneous anthrax is much more common than inhalational anthrax 14, and gastrointestinal
anthrax also occurs frequently in developing countries 15, 16. The relatively common routes
of cutaneous and gastrointestinal inoculation may cause seroconversion and less severe
symptoms—a potential form of natural vaccination.
In support of 10’s suggestion that cutaneous inoculation and subclinical infection may
have played a role in the resistance of certain workers to the outbreak of inhalational
anthrax in the goat hair processing mill, 17 found direct evidence that cutaneous anthrax
caused seroconversion for the protective antigen. Cutaneous symptoms can often be rather
mild, with small skin lesions that heal without significant clinical consequences. It would
be interesting to know how often animal workers get inoculations into cuts that induce or
boost immunity without noticeable symptoms. Vaccine research supports the idea that
cutaneous exposure can be protective: an epidermal patch protects laboratory animals
against subsequent challenge by inhalational exposure 18.
Among gastrointestinal anthrax exposures, subclinical or mild and misdiagnosed
cases likely occur, but few studies have focused on this problem 15, 19. Seroconversion may
occur in subclinical cases 19. Oral vaccines are considered a promising approach to inducing
protective immunity to anthrax 20, which suggests that a better understanding of the effects
of naturally occurring gastrointestinal exposure would be interesting.
In summary, some evidence supports mild or subclinical cases of anthrax by routes of
infection that differ from the most severe inhalational form of the disease. Occupation or
lifestyle likely influences the patterns of exposure and dosage by different routes. The main
limitation with regard to occupational immunity and natural vaccination probably arises
Electronic copy available at:
from the short-lived course of protective immunity following exposure. Whether repeated
boosting by continual exposure can maintain protective immunity remains an open
question. More generally, we discussed anthrax because of the available evidence on
alternative routes of infection with different consequences for disease. We suspect that
other pathogens share anthrax’s variable consequences among routes of infection but also
have longer-lasting immunity. If so, then natural vaccination may be a significant factor in
the dynamics of some zoonotic infections. Our next case continues to build the
circumstantial evidence in favor of this view.
Q fever
Immunity to anthrax wanes over time, whereas immunity after a bout of Q fever appears to
be lifelong 14, 20. Effective natural vaccination against anthrax may require repeated
exposure; for Q fever, a single exposure is probably sufficient for natural vaccination to
Q fever is a disease of humans caused by the zoonotic bacteria Coxiella burnetii. The
primary reservoirs of C. burnetii are cattle, sheep and goats. Humans at risk of occupational
exposure include veterinarians, meat processors, dairy workers and livestock farmers.
Human infections vary in severity; about half of cases are subclinical. When symptoms do
occur, infected individuals suffer fever, sore throat, chills, coughing, nausea, vomiting,
diarrhea or chest pain. Mortality ranges from 1%-2%. Recovery results in immunity that is
thought to be lifelong (CDC web site).
We now turn to our three questions. First, does occupational exposure to C. burnetii
lead to more frequent infection than occurs in the rest of the population? Two studies
suggest this is so. In a random sample of 583 people across geographic regions of Cyprus,
Electronic copy available at:
53% of individuals were reported as seropositive. Rural, semi-urban, and urban areas
differed, with rates of 61%, 48%, and 34%, respectively, presumably because of greater
exposure to animals in the rural regions. In particular, contact with sheep or goats increased
risk by 80% 21. In the second example, 22 found seropositive reactions in 14% of 267
veterinarians in Japan, compared with seropositivity in 4% of 2003 blood donors and 5% of
352 medical workers. The methods to determine positive reactions varied between the
above studies; the important results concern differences between geographic location or
occupation within studies, rather than the absolute levels of seroprevalence. Both studies
support the idea that people who are exposed to zoonotic pathogens are also being infected
by them.
Second, are individuals occupationally exposed to C. burnetii more likely to be
infected subclinically than those who became infected by a chance exposure? And third, do
the mechanisms of infection, such as route and dose, differ between those who become
infected subclinically and those who become clinically ill?
We found relatively little direct information about our second and third questions.
Thus, even though Q fever poses a significant occupational hazard to a variety of human
workers, we lack basic information about dose, route of infection, and severity of
symptoms. There is, however, a substantial literature on various aspects of transmission and
dosage that do help us to access the probability of natural vaccination in indirect ways. We
first review the most common mode of Q fever transmission, inhalation, then follow with
the less common routes of ingestion and tick-borne transmission.
It is widely believed that nearly all symptomatic human cases of Q fever arise from
inhalation of bacteria 23-25. If occupational exposure involves frequent inhalation of small
Electronic copy available at:
doses of bacteria in the workplace, then natural vaccination instead of disease could result
if disease symptoms occur more frequently at higher doses. A single inhaled bacterium is
sufficient to initiate infection in both humans and guinea pigs 26. However, we found no
studies that relate the number of inhaled bacteria to the severity of symptoms in humans.
Dose-response studies have been done in guinea pigs, which appear to be a good model for
the pathology of human infections 24, 27, 28. Humans and guinea pigs have similar dose-
response curves for the time between inhalation and the onset of fever 26.
28 infected guinea pigs by inhalation with six dose levels, increasing by factors of 10
from 2 X 101 to 2 X 106. All animals seroconverted, indicating infection with an immune
response. The intensity of fever and pathology of the lungs, liver, and spleen, increased
steadily with dose. At the lowest dose, almost no pathology was detected; at the highest
dose, moderate pathology occurred in all tested tissues. These results suggest that low doses
may often lead to subclinical or mild disease, whereas high doses may lead to severe cases.
Further clues about subclinical infection and the potential for natural vaccination
come from studies of non-inhalational infections of Q fever. Human infections by C.
burnetii can occur by ingestion of unpasteurized milk 29. Most evidence suggests that
ingestion often leads to seroconversion but rarely to symptomatic disease, although the
occurrence and severity of clinical symptoms by this route of infection remain open
problems 30, 31. Longitudinal serological studies of farm families or others who routinely
drink unpasturized milk could help shed light on whether natural vaccination is occurring in
these populations.
Ticks comprise the final route of infection discussed in the literature. Field surveys
show that ticks often carry C. burnetii 24. In experimental settings, tick bites successfully
Electronic copy available at:
infect guinea pigs 24. Ticks could, therefore, be a source of human infection, but almost all
discussion in the literature concludes that tick-borne transmission of Q fever to humans is
rare. 32 did find in two patients simultaneous bacteremia of C. burnetii and the tick-borne
pathogen Rickettsia conorii, suggesting that tick-borne transmission of C. burnetii may
occur; similarly, 33 found three patients simultaneously infected by these two bacteria.
Thus, although tick-borne transmission is widely discounted in the literature and may
indeed be rare, the actual evidence on this topic is rather limited.
We find these studies of tick-borne transmission intriguing because they suggest how
occupational exposure via cutaneous inoculation might lead to natural vaccination against
Q fever. Experimental transmission of C. burnetii by ticks into guinea pigs shows that
subcutaneous inoculation can be highly effective in causing infection. Various
subcutaneous vaccination strategies provide good protection against subsequent challenge
23, 24, 28. In animals, killed bacteria provide effective vaccines 23, suggesting that cutaneous
exposure to dead bacteria may produce or boost immunity.
Animal workers must often be exposed to bacteria through cuts in their skin. Among
veterinarians and their assistants, scratches and bites from handling animals commonly
occur 34. How often does accidental inoculation through skin cuts lead to infection and
seroconversion? What is the dose-response relationship between cutaneous infection and
the severity of symptoms? Can accidental cutaneous inoculation of dead bacteria lead to
natural vaccination or the boosting of immunity induced by prior infection?
In summary, the necessary conditions for natural vaccination against Q fever may
occur. Occupational exposure, infection and immunity occur frequently. Many infections
are subclinical. But the relative extent to which exposure by different routes or doses leads
Electronic copy available at:
to illness versus subclinical infection remains unknown. Immunity without overt disease
may be obtained occupationally by exposure via skin cuts and scratches. Cutaneous
inoculation is in fact a common method of vaccination against many pathogenic organisms.
Campylobacter jejuni
Our third example is campylobacteriosis, an acute gastroenteritis caused primarily by
Campylobacter jejuni 35. Campylobacteriosis is one of the most common causes of human
diarrhea. Most infections result from handling or eating raw or undercooked poultry.
Campylobacteriosis is an occupational hazard to meat processors. Non-occupational
exposure occurs in the home during food preparation and in petting zoos. The risk of
exposure is heightened by the fact that infected animals often exhibit no signs of illness
(CDC web site).
This example particularly highlights the ways in which different routes of inoculation
may influence the probability of subclinical infection following occupational exposure. We
now turn to our three questions.
First, does occupational exposure lead to a higher level of immunity against
Campylobacter than is observed in the rest of the population? Occupational exposure
among food processing workers is associated with high antibody titers against
Campylobacter jejuni 36. Only 2% of prenatal patients from Manchester, England and 5%
of similar patients from more rural areas had detectable antibodies. By contrast, sampling
detected antibodies in 27-68% of poultry workers from five different sites, 36% of workers
exposed to cattle, and 18% of veterinary assistants. In a poultry abattoir, short-term workers
employed less than one month had significantly lower levels of IgG antibodies against
Campylobacter jejuni than did long-term workers 37.
Electronic copy available at:
It is not known whether detectable antibodies against Campylobacter indicate
protective immunity. Various lines of circumstantial evidence suggest that infection yields
protection against subsequent exposure. Experimental oral infections in animals 38 and in
human volunteers 39 lead to increased antibody titers. Naturally occurring infections cause
increased levels of specific IgG antibodies at one year post-infection 40. In developing
countries, young children frequently harbor the bacteria, but disease is rare among those
over two years of age, and antibody levels increase with age 41, 42. 37 mention that “It is
anecdotal among poultry abattoir workers that during the first period of employment they
suffer from episodes of diarrhoea. However, over time the number of diarrhoeal episodes
apparently decreases, suggesting acquired immune protection.”
U.S. military personnel were screened for antibodies against Campylobacter before
and after travel to Thailand 43. Those with higher titers before travel had significantly lower
incidence of diarrhea during their time in Thailand. Symptomatic seroconversion during
travel occurred four times more frequently among those with low initial titers.
Campylobacter were the most commonly identified enteropathogens in stool samples.
Thus, high antibody levels before travel appeared to provide protection against
Campylobacter enteritis. On the whole, it appears that the tendency for increased antibody
titer with greater exposure correlates with better protection against disease—in other words,
occupational immunity appears to be common.
Second, are occupationally exposed individuals more likely to acquire immunity by
illness or by subclinical infection? We were unable to find any studies of Campylobacter
that directly address this question. Subclinical infection does occur (CDC web site), but we
did not find reports of the frequency of subclinical infection in particular groups.
Electronic copy available at:
Third, do aspects of exposure, such as route and dose, differ between those who
become infected subclinically and those who become clinically ill? The primary route of
symptomatic infection is by ingestion. The CDC web site reports that it takes fewer than
500 Campylobacter cells in an ingested inoculum to initiate infection. In industrial abattoirs
of developed countries, special worker clothing and training probably minimize cutaneous
exposure via cuts. But in less regulated food processing environments, infection through
cuts would provide another route of infection, with yet another relationship between
dosage, symptoms, and protective immunity. No studies have focused on all of these issues
in natural cutaneous exposure.
Some studies suggest significant airborne concentrations of Campylobacter in
industrial poultry abattoirs, with worker exposure from airborne droplets 44, 45. Such
droplets may be ingested by the typical oral route of infection.
If ingestion of airborne droplets is indeed a significant source of infection in abattoirs,
then two differences likely occur between occupational exposure and sporadic exposure of
typical members of the population. First, the distribution of dosages likely differs—perhaps
airborne droplets in the moist workplace more often inoculate workers with subclinical
doses compared with sporadic ingestion via food. Second, previously exposed workers
likely get a continual boost of immunity by repeated exposure, whereas sporadically
infected individuals would not receive boosting inoculations so often.
Pulmonary inoculation apparently rarely leads to symptomatic infections, but it
would be interesting to study how repeated inhalation of subclinical doses affects protective
immunity. Wilson suggests that UV treatment of air in abattoirs may improve the health of
workers 44, but if repeated inhalation actually boosts protective immunity for long-term
Electronic copy available at:
workers, then cleaning the air might alter the dosage and infection profiles in ways that
actually increase symptomatic disease. That is, of course, a speculative idea—the point is
that some fascinating and important questions remain open with regard to how occupational
conditions affect exposure and natural vaccination.
In summary, long-term immunity following infection with Campylobacter probably
occurs, although boosting by repeat infection may play a role. There is no direct evidence
with regard to the frequency or causes of subclinical infection. Among animal workers, it
would be particularly interesting to know whether cutaneous or pulmonary exposure could
cause subclinical infection and protective immunity, and whether frequent small doses by
ingestion of airborne droplets affect the long-term maintenance of protective immunity.
Avian and swine influenza
Influenza A viruses derive from wild birds. Much has been written about the pathways of
transition between wild birds and humans 46, Suarez, 2000 #55, Van Reeth, 2007 #56. One recurring
theme concerns infections of humans through contact with domestic animals that harbor
avian-derived viruses. This theme leads to our topic of occupational exposure.
Several recent sporadic infections of humans have derived from contact with
domestic animals. In these cases, a small number of people became infected, but the virus
did not spread widely in human populations. We use those sporadic cases to address the
same three questions we applied to anthrax, Q fever and Campylobacter.
First, does occupational exposure to zoonotic influenza in domestic animals lead to a
higher level of immunity against those foreign viruses than is observed in the rest of the
human population? Several recent reviews demonstrate increased antibody titers among
Electronic copy available at:
individuals occupationally exposed to swine 47 48. For example, 49 found higher
seropositivity to swine-adapted influenza viruses in swine farmers, their families and
employees than controls not having contact with swine. 50 found that, for swine isolates of
H1N1 and H1N2, farm workers, veterinarians, and meat-processing workers all had greatly
elevated seroprevalence compared with controls.
Serological studies of poultry workers present mixed results, partly because of the
technical complexity of such studies 48. However, several studies do clearly show strongly
elevated antibody titers among occupationally exposed poultry workers. For example, in
the H7N7 poultry outbreak in the Netherlands during 2003, raised titers were observed in
49% of 508 poultry cullers and 64% of 63 people exposed to humans infected with H7N7
51. In a comparison of 42 veterinarians with 66 controls for antibody titers against nine
different avian influenza strains, the veterinarians had significantly elevated titers against
three of the nine strains 52.
The recent human infections by the H5N1 avian influenza virus also present mixed
serological results. Studies of poultry workers, cullers and health care workers involved in
the initial outbreaks in 1997 showed elevated seropositivity, suggesting that there had been
some degree of subclinical infection by H5N1 53, Bridges, 2002 #59, Katz, 1999 #61 . However, studies
of later outbreaks failed to show similar results 54 (need more citations). The current view is
that H5N1 has difficulty starting an infection in humans, which would explain the low
seropositivity and the relatively few cases per contact with infected animals (see review by
55). However, sampling has been limited and the serology seems to be particularly difficult
to interpret for this virus.
Electronic copy available at:
In summary, several lines of evidence suggest that occupational exposure to avian
and swine influenza viruses may lead to a higher level of seropositivity than observed in the
general population. Differences in seropositivity occur between various avian-derived
strains and between the type of animal exposure. Swine exposure presents clearly elevated
antibody levels among animal workers; poultry exposure presents a more complex picture.
Second, if occupationally exposed individuals show higher levels of immunity to
zoonotic influenza viruses, was that immunity more likely to have been acquired via illness
or by subclinical infection? The high levels of seropositivity reported above from swine and
poultry exposures were not associated with widespread clinical symptoms. Thus,
subclinical infections can lead to seropositivity. However, positive tests for serum
antibodies does not necessarily mean protective immunity. The relation between serology
and immunity remains a key issue in interpreting the current literature.
Third, do particular aspects of infection, such as dosage, route of inoculation, or
frequency of exposure, influence the probability that exposure results subclinical infection
with resulting immunity (natural vaccination) as opposed to disease? No direct evidence
addresses this question in humans. 56 reported a laboratory study of mice that contrasted the
intensity of symptoms between two different routes of inoculation. They found that
pulmonary inoculation of mice with H1N1 influenza viruses led to lethal infections at
moderate doses, whereas nasal inoculations caused death only at very high doses. This
result on the less severe consequences of nasal inoculation leads to an interesting general
problem with regard to occupational exposure. Could animal workers be exposed
frequently to viral particles in inhaled dust? Would such exposures sometimes act as nasal
vaccines? Similarly, how would workers be affected by cutaneous exposures through
Electronic copy available at:
scratches and cuts? Apart from the specific issues of natural vaccination, we simply do not
have much information about routes of exposure, symptoms, and immunity.
In recent years, emerging infectious diseases have grown in importance and attracted
increased research attention. Because occupationally exposed individuals often provide the
first line of zoonotic transmission into human populations, it will be particularly important
to learn more about infection and immunity in animal workers. To set the background, we
developed the conceptual framework of occupational immunity and natural vaccination.
That framework provided the basis on which to organize existing information about the key
problems for future study.
In particular, we emphasized three aspects of exposure and immunity for which it
would be important to know more. First, to what extent do occupationally exposed
individuals actually develop infections and immunity? Second, how often do occupational
infections go undetected because they cause relatively mild symptoms? Third, how do
unusual aspects of dosage and route of inoculation among animal workers cause those
individuals to develop infection, symptoms, and immunity?
Each of our four pathogen examples provides some information about these
questions. From our review, it appears that occupational exposure does typically cause
increased infection and immunity among animal workers. Several studies suggest that
occupational infections are sometimes, perhaps often, subclinical. Interestingly, the routes
of inoculation that may be particularly prevalent among workers, such as cutaneous
exposure through cuts and scratches or nasal inhalation, may be particularly likely to cause
subclinical infection and natural vaccination. However, the consequences for different
Electronic copy available at:
routes of exposure and different dose levels have rarely been studied. On the whole, our
concepts and review of the literature show the importance of the subject and what we need
to learn in order to move ahead.
More references needed in text??
Annotate bibliography
1. Frank, S.A. & Jeffrey, J.S. The probability of severe disease in zoonotic and
commensal infections. Proc Biol Sci 268, 53-60 (2001).
2. Anonymous. Inhalation anthrax associated with dried animal hides--Pennsylvania
and New York City, 2006. MMWR Morb Mortal Wkly Rep 55, 280-2 (2006).
3. Woods, C.W. et al. Risk factors for human anthrax among contacts of anthrax-
infected livestock in Kazakhstan. Am J Trop Med Hyg 71, 48-52 (2004).
4. Plotkin, S.A., Brachman, P.S., Utell, M., Bumford, F.H. & Atchison, M.M. An
epidemic of inhalation anthrax, the first in the twentieth century. I. Clinical features. Am J
Med 29, 992-1001 (1960).
5. Brachman, P.S., Plotkin, S.A., Bumford, F.H. & Atchison, M.M. An epidemic of
inhalation anthrax: the first in the twentieth century. II. Epidemiology. Am J Hyg 72, 6-23
6. Glassman, H.N. Discussion. Bacteriol Rev 30, 657-659 (1966).
Last updated in 2008
Electronic copy available at:
7. Dahlgren, C.M. et al. Bacillus anthracis aerosols in goat hair processing mills. Am J
Hyg 72, 24-31 (1960).
8. Carr, E.A., Jr. & Rew, R.R. Recovery of Bacillus anthracis from the nose and throat
of apparently healthy workers. J Infect Dis 100, 169-71 (1957).
9. Brachman, P.S. & Fekety, F.R. Industrial anthrax. Ann N Y Acad Sci 70, 574-84
10. Norman, P.S., Ray, J.G., Jr., Brachman, P.S., Plotkin, S.A. & Pagano, J.S. Serologic
testing for anthrax antibodies in workers in a goat hair processing mill. Am J Hyg 72, 32-7
11. Dinh, P.N. Risk Factors for Human Infection with Avian Influenza A H5N1,
Vietnam, 2004. Emerg Infect Dis 12, 1841-7 (2006).
12. Lodge, S., Jr. La maladie des trieurs de laine (Charbon bronco-pulmonaire). Arch.
Med. Exper. et d'Anat. Pathol. 2, 759-771 (1890).
13. Kronberger, H. Ein Fall von Lungenmilzbrand mit güstigen Ausgang. Klin.
Tubercular Berl. 38, 135-144 (1917).
14. Dixon, T.C., Meselson, M., Guillemin, J. & Hanna, P.C. Anthrax. N Engl J Med
341, 815-26 (1999).
15. Sirisanthana, T. & Brown, A.E. Anthrax of the gastrointestinal tract. Emerg Infect
Dis 8, 649-51 (2002).
16. Beatty, M.E., Ashford, D.A., Griffin, P.M., Tauxe, R.V. & Sobel, J. Gastrointestinal
anthrax: review of the literature. Arch Intern Med 163, 2527-31 (2003).
Electronic copy available at:
17. Harrison, L.H., Ezzell, J.W., Abshire, T.G., Kidd, S. & Kaufmann, A.F. Evaluation
of serologic tests for diagnosis of anthrax after an outbreak of cutaneous anthrax in
Paraguay. J Infect Dis 160, 706-10 (1989).
18. Kenney, R.T. et al. Induction of protective immunity against lethal anthrax
challenge with a patch. J Infect Dis 190, 774-82 (2004).
19. Turnbull, P.C. et al. Anthrax in wildlife in the Luangwa Valley, Zambia. Vet Rec
128, 399-403 (1991).
20. Baillie, L.W. Past, imminent and future human medical countermeasures for
anthrax. J Appl Microbiol 101, 594-606 (2006).
21. Psaroulaki, A. et al. Epidemiological study of Q fever in humans, ruminant animals,
and ticks in Cyprus using a geographical information system. Eur J Clin Microbiol Infect
Dis 25, 576-86 (2006).
22. Abe, T. et al. A seroepidemiological study of the risks of Q fever infection in
Japanese veterinarians. Eur J Epidemiol 17, 1029-32 (2001).
23. Arricau-Bouvery, N. & Rodolakis, A. Is Q fever an emerging or re-emerging
zoonosis? Vet Res 36, 327-49 (2005).
24. Maurin, M. & Raoult, D. Q fever. Clin Microbiol Rev 12, 518-53 (1999).
25. Raoult, D., Marrie, T. & Mege, J. Natural history and pathophysiology of Q fever.
Lancet Infect Dis 5, 219-26 (2005).
26. Tigertt, W.D., Benenson, A.S. & Gochenour, W.S. Airborne Q fever. Bacteriol Rev
25, 285-93 (1961).
Electronic copy available at:
27. La Scola, B., Lepidi, H. & Raoult, D. Pathologic changes during acute Q fever:
influence of the route of infection and inoculum size in infected guinea pigs. Infect Immun
65, 2443-7 (1997).
28. Russell-Lodrigue, K.E., Zhang, G.Q., McMurray, D.N. & Samuel, J.E. Clinical and
pathologic changes in a guinea pig aerosol challenge model of acute Q fever. Infect Immun
74, 6085-91 (2006).
29. Benson, W.W., Brock, D.W. & Mather, J. Serologic Analysis of a Penitentiary
Group Using Raw Milk from a Q Fever Infected Herd. Public Health Rep 78, 707-10
30. Hatchette, T.F. et al. Goat-associated Q fever: a new disease in Newfoundland.
Emerg Infect Dis 7, 413-9 (2001).
31. Cerf, O. & Condron, R. Coxiella burnetii and milk pasteurization: an early
application of the precautionary principle? Epidemiol Infect 134, 946-51 (2006).
32. Janbon, F., Raoult, D., Reynes, J. & Bertrand, A. Concomitant human infection due
to Rickettsia conorii and Coxiella burnetii. J Infect Dis 160, 354-5 (1989).
33. Rolain, J.M. et al. Concomitant or consecutive infection with Coxiella burnetii and
tickborne diseases. Clin Infect Dis 40, 82-8 (2005).
34. Nienhaus, A., Skudlik, C. & Seidler, A. Work-related accidents and occupational
diseases in veterinarians and their staff. Int Arch Occup Environ Health 78, 230-8 (2005).
35. Nachamkin, I., Blaser, M.J. & Tompkins, L.S. (eds.) Campylobacter jejuni: Current
status and future trends (American Society of Microbiology, Washington, D.C., 1992).
36. Jones, D.M. & Robinson, D.A. Occupatonal exposure to Campylobacter jejuni
infection. Lancet 1, 440-1 (1981).
Electronic copy available at:
37. Cawthraw, S.A., Lind, L., Kaijser, B. & Newell, D.G. Antibodies, directed towards
Campylobacter jejuni antigens, in sera from poultry abattoir workers. Clin Exp Immunol
122, 55-60 (2000).
38. Russell, R.G. in Campylobacter jejuni: current status and future trends (eds.
Nachamkin, I., Blaser, M.J. & Tompkins, L.S.) (American Society of Microbiology,
Washington, DC, 1992).
39. Black, R.E., Perlman, D., Clements, M.L., Levine, M.M. & Blaser, M.J. in
Campylobacter jejuni: current status and future trends (eds. Nachamkin, I., Blaser, M.J. &
Tompkins, L.S.) (American Society of Microbiology, Washington, DC, 1992).
40. Cawthraw, S.A., Feldman, R.A., Sayers, A.R. & Newell, D.G. Long-term antibody
responses following human infection with Campylobacter jejuni. Clin Exp Immunol 130,
101-6 (2002).
41. Blaser, M.J., Black, R.E., Duncan, D.J. & Amer, J. Campylobacter jejuni-specific
serum antibodies are elevated in healthy Bangladeshi children. J Clin Microbiol 21, 164-7
42. Tauxe, R.V. in Campylobacter jejuni: current status and future trends (eds.
Nachamkin, I., Blaser, M.J. & Tompkins, L.S.) 9-19 (American Society of Microbiology,
Washington, D.C., 1992).
43. Walz, S.E. et al. Pre-exposure anti-Campylobacter jejuni immunoglobulin a levels
associated with reduced risk of Campylobacter diarrhea in adults traveling to Thailand. Am
J Trop Med Hyg 65, 652-6 (2001).
44. Wilson, I.G. Airborne Campylobacter infection in a poultry worker: case report and
review of the literature. Commun Dis Public Health 7, 349-53 (2004).
Electronic copy available at:
45. Posch, J. et al. Transmission of Campylobacter spp. in a poultry slaughterhouse and
genetic characterisation of the isolates by pulsed-field gel electrophoresis. Br Poult Sci 47,
286-93 (2006).
46. Ito, T. & Kawaoka, Y. Host-range barrier of influenza A viruses. Veterinary
Microbiology 74, 71-5 (2000).
47. Myers, K.P., Olsen, C.W. & Gray, G.C. Cases of swine influenza in humans: a
review of the literature. Clin Infect Dis 44, 1084-8 (2007).
48. Gray, G.C., Trampel, D.W. & Roth, J.A. Pandemic influenza planning: shouldn't
swine and poultry workers be included? Vaccine 25, 4376-81 (2007).
49. Olsen, C.W. et al. Serologic evidence of H1 swine influenza virus infection in swine
farm residents and employees. Emerging Infectious Diseases 8, 814-819 (2002).
50. Myers, K.P. et al. Are swine workers in the United States at increased risk of
infection with zoonotic influenza virus? Clin Infect Dis 42, 14-20 (2006).
51. Meijer, A. et al. Measurement of antibodies to avian influenza virus A(H7N7) in
humans by hemagglutination inhibition test. J Virol Methods 132, 113-20 (2006).
52. Myers, K.P., Setterquist, S.F., Capuano, A.W. & Gray, G.C. Infection due to 3
avian influenza subtypes in United States veterinarians. Clin Infect Dis 45, 4-9 (2007).
53. Bridges, C.B. et al. Risk of influenza A (H5N1) infection among health care
workers exposed to patients with influenza A (H5N1), Hong Kong. Journal of Infectious
Diseases 181, 344-8 (2000).
54. Vong, S. et al. Low frequency of poultry-to-human H5NI virus transmission,
southern Cambodia, 2005. Emerg Infect Dis 12, 1542-7 (2006).
Electronic copy available at:
55. Horimoto, T. & Kawaoka, Y. Influenza: lessons from past pandemics, warnings
from current incidents. Nat Rev Microbiol 3, 591-600 (2005).
56. Yetter, R.A., Lehrer, S., Ramphal, R. & Small, P.A., Jr. Outcome of influenza
infection: effect of site of initial infection and heterotypic immunity. Infection and
Immunity 29, 654-62 (1980).
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
The first outbreak of avian influenza A (H5N1) occurred among humans in Hong Kong in 1997. To estimate the risk of person-to-person transmission, a retrospective cohort study was conducted to compare the prevalence of H5N1 antibody among health care workers (HCWs) exposed to H5N1 case-patients with the prevalence among nonexposed HCWs. Information on H5N1 case-patient and poultry exposures and blood samples for H5N1-speciflc antibody testing were collected. Eight (3.7%) of 217 exposed and 2 (0.7%) of 309 nonexposed HCWs were H5N1 seropositive (P = .01). The difference remained significant after controlling for poultry exposure (P = .01). This study presents the first epidemiologic evidence that H5N1 viruses were transmitted from patients to HCWs. Human-to-human transmission of avian influenza may increase the chances for the emergence of a novel influenza virus with pandemic potential.
Full-text available
An abnormally high mortality among hippos (Hippopotamus amphibius) in the Luangwa River valley between June and November 1987 and estimated to number more than 4000 deaths was attributed to anthrax. Several other species, particularly Cape buffalo (Syncerus caffer) and elephant (Loxodonta africana), appear to have been affected. A smaller outbreak of anthrax in hippos occurred between August and September 1988, approximately 100 km up-river. A field study was arranged in August 1989 to assess the extent of environmental contamination by Bacillus anthracis and the risks to people in the area, to study possible methods of control and to equip local laboratory staff for continued monitoring of the disease. The study confirmed the enzootic status of the region. The characteristics of the outbreaks of anthrax in 1987 and 1988, and the results of the field study are described.
Full-text available
In Bangladesh and other developing countries, isolation of Campylobacter jejuni is common in healthy children, and the illness/infection ratio falls with age. To determine whether specific serum antibodies correlate with this phenomenon, using an enzyme-linked immunosorbent assay, we studied sera from 93 healthy Bangladeshi children and 121 healthy U.S. children under 15 years of age. For each age group (less than 1, 2 to 4, and 5 to 14 years) studied, specific serum antibody levels were significantly higher in the Bangladeshi children. Among Bangladeshi children, for each of the three immunoglobulin subclasses, the change in antibody levels with age was different. Specific immunoglobulin A antibody levels rose linearly with age, immunoglobulin G levels peaked in the 2- to 4-year age group and then fell, and immunoglobulin M levels peaked in the 2- to 4-year age group and then plateaued. Elevated serum antibody levels to C. jejuni in Bangladeshi children may be protective in themselves or may reflect other protective phenomena.
Full-text available
An infection established throughout the total respiratory tract of mice with a highly lung adapted influenza virus (H0N1) led to death from viral pneumonia. The 50% lethal dose (LD(50)) was approximately the same as the 50% infectious dose (ID(50)). An infection with the same virus initiated in the nasal mucosa spread to the trachea and lungs over a 3- to 5-day period but was not lethal except at very high infecting doses. The LD(50) was 30,000 times the ID(50). Mice that had recovered from a prior infection with A/PC/73(H3N2) demonstrated enhanced recovery (heterotypic immunity) when challenged with A/PR/8/34(H0N1). Heterotypically immune mice infected while anesthetized with this potentially lethal virus stopped shedding virus from the nose, trachea, and lungs by day 7 and recovered. Heterotypically immune mice, infected awake, stopped shedding virus from the nose by day 5, and, in fact, the virus did not spread to the trachea or lungs. Thus, some of the variation in the severity of influenza infections may be explained by two factors: the site of initial infection and previous infection with heterotypic influenza virus.
When swallowed, anthrax spores may cause lesions from the oral cavity to the cecum. Gastrointestinal anthrax is greatly underreported in rural disease-endemic areas of the world. The apparent paucity of this form of anthrax reflects the lack of facilities able to make the diagnosis in these areas. The spectrum of disease, ranging from subclinical infection to death, has not been fully recognized. In some community-based studies, cases of gastrointestinal anthrax outnumbered those of cutaneous anthrax. The oropharyngeal variant, in particular, is unfamiliar to most physicians. The clinical features of oropharyngeal anthrax include fever and toxemia, inflammatory lesion(s) in the oral cavity or oropharynx, enlargement of cervical lymph nodes associated with edema of the soft tissue of the cervical area, and a high case-fatality rate. Awareness of gastrointestinal anthrax in a differential diagnosis remains important in anthrax-endemic areas but now also in settings of possible bioterrorism.
An outbreak of at least 21 cases of cutaneous anthrax occurred in rural Paraguay. A case-control study revealed that disease was associated with touching the raw meat of an ill cow (odds ration = 16.5, P = .02). Serum drawn from 12 cases and 16 colony and 2 noncolony controls 6 w after the outbreak were analyzed by electrophoretic-immunotransblots (EITB) to detect serum antibodies to the protective antigen (PA) and lethal factor components of anthrax toxin. Serum was also tested by enzyme-linked immunosorbent assay (ELISA) for the presence of antibodies to poly-D-glutamic acid capsule. Of 12 cases, 11 had a positive PA screen, for a sensitivity of 91.7% (76.1%–100%, 95% confidence interval [CI]) whereas none of the 18 controls was positive for a specificity of 100% (84.8%, one-sided binomial 95% CI). Only 6 (50%) of 12 cases (21.7%–78.3%, 95% CI) had positive lethal factor titers; all controls were negative. At a cutoff of ⩾1:32 for antibodies to capsule, 11 (91.7%) of 12 (76.1%–100%, 95% CI) were positive; 16 (88.9%) of 18 controls (74.5%–100%, 95% CI) were negative. These data suggest that the EITB for detection of antibody to PA, and ELISA for detection of anticapsule antibodies are both sensitive for the retrospective diagnosis of anthrax. Both tests were specific, but EITB may be more so than ELISA.