é :iirnéri .0

 

 

.53....3355 £153..

 

 

 

 

 

 

 

 

 

.53..

+

 

‘ .6535“ In

.11.: 3:5
:1...!....:.2 :3" V E.
Afinvfikafl 5n 1.
«i a,

A..."

a" In. .fivs. ;

. 2i... 1.
.32!
$9.43..

(3.. lit...“
3.

.713:
3!: iii.
a! .

 

TENS

70°?

 

 

LIBRARY

Michigan State

University

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

OCT 1 7 2005

 

DC] 16 2006

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRCIDataDue.p65-p.15

 

A RETROSPECTIVE STUDY OF
CAMPYLOBACTER JEJUNI ENTERITIS AND
GUILLAIN-BARRE SYNDROME IN MICHIGAN: 1992 — 1999
By

Rachel Church Potter, DVM

A THESIS

Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Epidemiology

2002

ABSTRACT

A RETROSPECTIVE STUDY OF CAMPYLOBACTER JEJUNI ENTERITIS AND
GUILLAlN-BARRE SYNDROME IN MICHIGAN: 1992-1999

By

Rachel Church Potter, DVM

The association between Campylobacterjejuni and Guillain-Barré Syndrome (GBS)
has been extensively researched in recent years. Because C. jejuni enteritis is common
and GBS is severe, understanding the distribution and risk factors for C. jejuni enteritis
and how this infection leads to GBS has become a priority.

In this work, three major objectives were addressed in three studies. The first study
was designed to determine if increased rate of C. jejuni infection was evident in areas of
high poultry density. The second study assessed if GBS rates were also higher in these
areas. The third study was designed to determine what animal species were associated
with increased odds of infection in rural counties at the individual level.

In the first study, Michigan was divided into two regions of high and low poultry
density and C. jejuni incidence rates were compared between the regions. The incidence
of C. jejuni was 1.3 times higher in the high poultry density region. In the second study,
the GBS rate was compared between the high and low C. jejuni enteritis regions
determined by the first study. No significant differences were found. In the third study,
poultry husbandry was found to significantly increase the odds for C. jejuni enteritis in

rural areas.

ACKNOWLEDGEMENTS

I would first like to thank the Population Medicine Center and the National Food
Safety and Toxicology Center for their support of this project.

Thanks also to the people in rural Michigan who took the time to fill out our
questionnaire and so helped us learn about Campylobacterjejuni. I hope that our
findings will benefit them in the future.

It has been a pleasure to work with this committee. Without Bill Hall’s enthusiastic
support this project would not have been possible, and he has my most sincere gratitude
for facilitating this study all along the way. Thanks to Joseph Gardiner for his expertise
and patience. He has been a great teacher and has held up well under a steady barrage of
emails. Claudia Holzman has helped to steer my course through this graduate program
and, as a private practitioner moving into epidemiology, her career has been an
inspiration.

Special thanks go to my advisor, John Kaneene. I have learned a great many things
from him and by his example. I feel very lucky to have had the opportunity to work with
him. It has been a great experience and I look forward to working with him again in the

future.

TABLE OF CONTENTS

LIST OF TABLES ..................................................................... vi
LIST OF FIGURES ..................................................................... vii
INTRODUCTION ....................................................................... 1
Purpose ................................................................................ 1
Objectives ............................................................................. 1
Hypotheses ............................................................................ 1
Overview .............................................................................. 2
CHAPTER 1
RISK FACTORS FOR SPORADIC C. JEJUNI ENTERITIS IN DEVELOPED
COUNTRIES: A REVIEW .............................................................. 3
Introduction ........................................................................... 3
Descriptive Epidemiology ........................................................... 4
Analytical Epidemiology ............................................................ 6
Conclusion ............................................................................. 15
CHAPTER 2
CAMPYLOBACTER JEJUNI, GUILLAIN-BARRE SYNDROME, AND THE
CAUSAL CRITERIA .................................................................... 24
Introduction ........................................................................... 24
Strength of the Association ......................................................... 25
Consistency ........................................................................... 27
Specificity ............................................................................. 30
Temporality ........................................................................... 33
Biologic Gradient ..................................................................... 34
Plausibility ............................................................................. 35
Analogy ................................................................................. 39
Conclusion .............................................................................. 40
CHAPTER 3

CAMPYLOBACTER JEJUNI ENTERITIS IN MICHIGAN 1992-1999:
A COMPARISON OF INCIDENCE RATES IN AREAS OF HIGH AND LOW

POULTRY DENSITY ................................................................... 46
Abstract ................................................................................ 46
Introduction ........................................................................... 46
Methods ................................................................................ 47
Results .................................................................................. 49
Discussion .............................................................................. 50

iv

CHAPTER 4
DESCRIPTIVE EPIDEMIOLOGY OF GUILLAIN—BARRE SYNDROME IN

MICHIGAN: 1992 — 1999 ............................................................ 60
Introduction .......................................................................... 60
Methods .............................................................................. 60
Results ................................................................................ 61
Discussion ........................................................................... 63

CHAPTER 5

RISK FACTORS FOR CAMPYLOBACTER JEJUNI INFECTIONS IN RURAL

MICHIGAN: A PROSPECTIVE CASE-CONTROL STUDY .................. 67
Abstract ............................................................................... 67
Introduction .......................................................................... 67
Methods .............................................................................. 68
Results ................................................................................ 71
Discussion ........................................................................... 75

SUMMARY AND CONCLUSIONS ................................................ 84

APPENDIX
Consent Form ....................................................................... 87
Questionnaire ....................................................................... 89

REFERENCES ......................................................................... 102

Table 1-1

Table 1-2

Table 1-3

Table 3-1

Table 3-2

Table 4-1

Table 5-1

Table 5—2

Table 5-3

LIST OF TABLES

Odds ratios for C. jejuni enteritis associated with undercooked
chicken consumption

Odds ratios for C. jejuni enteritis associated with any chicken
consumption

Odds ratios for C. jejuni enteritis associated with animal contact
High and low density counties by year 1992 — 1999

Distribution of cases between areas of high and low poultry
density

Age-specific incidence rates for GBS in Michigan 1992-1999

Univariable analysis of covariates included in the summary
husbandry variable

Matched odds ratios for animal contact, demographic
characteristics, and food consumption habits for cases of
C. jejuni enteritis: exposures significant in univariable
analysis (p < 0.15)

Final multivariable model

vi

12

54

54

62

78

79

80

Figure 3—1

Figure 3-2

Figure 3-3

Figure 3-4

Figure 4-1

Figure 4-2

Figure 5-1

LIST OF FIGURES

Distribution of cases between areas of high and low poultry
density regions: Michigan 1992-1999

Gender-specific incidence rates in high and low poultry
density regions: Michigan 1992-1999

Age-specific incidence rates in high and low poultry density
regions: Michigan 1992-1 999

Seasonal incidence rates in high and low poultry density
regions: Michigan 1992-1999

Crude annual incidence rates for GBS in Michigan 1992-1999

Age-specific incidence rates for GBS in high and low C. jejuni
enteritis regions in Michigan 1992-1999

Demonstration of a linear dose-response in number of types
of contact with poultry or bovines and odds ratios

vii

55

55

56

57

65

65

81

INTRODUCTION

Purpose

C ampylobacter jcj zmi is a gram-negative bacterium that can cause severe
gastroenteritis in humans, but is a commensal in many warm-blooded animals. These
animals are the reservoirs for the organism. Campylobacterjejuni infection is thought to
be the most common cause of Guillain-Barré Syndrome (GBS), a post-infectious flaccid
paralysis. Previous ecological studies have revealed that C. jejuni incidence rates are
highest in farming areas, followed by rural areas, with urban areas having the lowest
rates, but the risk factors underlying these rates have not been elucidated. There is a need
to identify the risk factors in rural and farming communities so that control and
prevention measures can be focused appropriately.
Objectives

1. Describe the epidemiology of GBS and C. jejuni enteritis in Michigan.

2. Identify the risk factors for C. jejuni enteritis in rural counties.

3. Compare the distribution of GBS cases to that of C. jejuni cases.

4. Assess the usefulness of the Michigan Department of Community Health

reporting system for future studies of GBS in Michigan.

Hypotheses Tested

1. The odds of developing C. jejuni enteritis are higher in counties with high poultry

density than those with low density (hypothesis tested in Chapter 3).
2. Campylobacterjejuni is strongly associated with Guillain-Barre’ Syndrome

(Hypothesis tested in Chapter 4).

3. Exposure to food animals is a major risk factor for C. jejuni enteritis and the odds
of infection changes with exposure to different food animal species (Hypothesis
tested in Chapter 5).
Overview
Chapter 1 is a literature review of the risk factors for C. jejuni enteritis. Particular
attention is paid to urban and rural disease determinants. Chapter 2 is a review of the
association between C. jejuni and GBS. Gaps in the current understanding of this
association are identified and areas where future research is needed are described.
Chapter 3 is an ecological study of C. jejuni enteritis in Michigan in the eight-year period
between 1992 and 1999. Chapter 4 describes the epidemiology of GBS in Michigan
between 1992 and 1999. Chapter 5 is a case-control study of C. jejuni enteritis in rural
Michigan. Each chapter is in a format for independent publication. The overall
contribution of this work to the understanding of both diseases is presented in the

conclusion.

CHAPTER 1

RISK FACTORS FOR SPORADIC C. JEJUNI ENTERITIS IN DEVELOPED
COUNTRIES: A REVIEW
Introduction

Campylobacterjq/zmi is the leading cause of food-borne disease outbreaks in the
United States. The yearly incidence rate was 15.7 / 100,000 person-years in 2000,
according to the CDC FoodNet active surveillance system (CDC, 2001). Campylobacter
jejuni is a gram-negative, microaerophilic, thermophilic organism. Morphologically, it is
a spiral or curved rod that is highly motile due to a single polar flagellum. Illness occurs
three to five days after ingestion of a dose as low as 500 organisms (Robinson, 1981).
Clinical signs and symptoms include fever, profuse, often bloody, diarrhea, abdominal
pain, which may be severe enough to mimic appendicitis, and nausea. Clinical signs may
remain severe for 24 to 48 hours before resolving gradually over a period of one week or
longer. Rarely, infection occurs at other sites causing cholecystitis, pancreatitis, cystitis,
or septic abortion. Fatal bacteremia may occur in immunocompromised hosts (Manfredi
et al., 1999). Sequelae to the infection include Guillain-Barre syndrome (GBS),
rheumatic or reactive arthritis (Blaser, 1997; Bremell et al., 1991), and Hemolytic-uremic
syndrome (HUS) (Sillero and Almirall, 1999).

GBS is a subacute polyneuropathy affecting motor, sensory, and autonomic nerves
that supply the limbs and respiratory muscles. Cranial nerves may also become involved.
The GBS mortality rate is approximately ten percent and recovery is often incomplete
and / or delayed (Hughes and Rees, 1997). An annual incidence of 0.20 to 1.46 cases of

GBS associated with Campylobacrer per 100,000 population has been estimated (Buzby

et al., 1997). A retrospective cohort study in Sweden in revealed that C. jejuni cases
experienced an incidence of GBS that was 100 times higher what was expected in the
general population (McCarthy and Giesecke, 2001). Reactive arthritis may cause pain
and incapacitation for several weeks to months in approximately one percent of C. jejuni
patients (Skirrow and Blaser, 2000). HUS is characterized by microangiopathic anemia,
thrombocytopenic purpura, and acute oliguric renal failure. It is only rarely reported in
association with Campj'lobacterjejuni (Sillero and Almirall, 1999).

The main focus of this review will be the epidemiology of sporadic cases, which cause
the burden of disease, in developed countries. Sources of infection will be discussed and
areas where additional research is needed will be highlighted.

Descriptive Epidemiology

The epidemiologic features of C. jejuni enteritis are unique. Incidence rates vary by
season, gender, age, and place.
Season

Sporadic carnpylobacteriosis in the US. overall shows a seasonal distribution
characterized by a low number of reported cases from January through April, an increase
in May and June, and a peak in July and August. After August, the rate falls off (CDC,
2000). Investigators in Scotland showed that the seasonal distribution of rural and urban
cases differed, however. Isolations were more frequent in the third quarter of the year in
cities, whereas in rural populations they were more predominant in the first quarter of the
year. The authors noted that the first quarter peak in infection rates in rural areas

coincided with spring calving (Sibbald and Sharp, 1985). This is plausible because it has

been found that calves have a higher prevalence of C. jejuni carriage before they are
weaned than after (Busato et al., 1999).
Gender and Age

For the whole United States, age-specific incidence rates show a bimodal distribution.
The highest isolation rate occurs in infancy (<1 year of age) followed by a later peak in
infection among young adults. Males are affected more commonly than females. The
reasons for the age and gender distribution are not currently known. The high rates in
infants may be due to increased care seeking by parents, because they may rapidly
become dehydrated, or because of the immaturity of their immune systems. The second
peak occurring in young adults may be due to the so-called second weaning effect, in
which self-reliance in food preparation is first learned.

The ratio of childhood to adult infections was significantly higher in rural areas than
urban areas in one study (Sibbald and Sharp, 1985). A hypothesis generated from this
finding was that children might have early exposure to C. jejuni, via raw milk
consumption, and then go on to develop and maintain immunity that protects them from
clinical infection later in life (Sibbald and Sharp, 1985). A case-control study of endemic
carnpylobacteriosis in Dubuque, Iowa, a small city in a rural area had findings that
supported the raw milk hypothesis generated by Sibbald and Sharp in 1985. Of 15 cases
who drank raw milk, 12 were less than 10 years old (80%) and nine (60%) lived in rural
areas, both significant differences from those that did not drink raw milk. Even more
interestingly, the six cases who drank raw milk and were from urban areas had visited

rural areas and consumed raw milk there (Schmid et al., 1987). It has also been shown

that regular raw milk consumption is protective against clinical signs of disease (Blaser et
al., 1987; Jones et al., 1981).

The incidence of campylobacteriosis in HIV-infected patients is almost 39 times
higher than the rate in the general population. In Los Angeles County between 1983 and
1987, the reported incidence of carnpylobacteriosis in AIDS patients was 519 cases per
100,000 population (Sorvillo, 1991).

Place

A few ecological studies have made comparisons between urban and rural disease
rates in developed countries with mixed results. A study in Norway found a higher urban
rate, which was explained by a higher proportion of imported cases in urban areas.
Imported cases are those that are acquired during foreign travel. The isolation rates were
almost the same when these imported cases were excluded (Kapperud and Aasen, 1992).
In Yugoslavia, higher Campylobacter spp. isolation rates were found in rural areas, but
this was due to the high proportion of C. coli isolates in rural areas; C. jejuni isolation
rates were actually lower in rural areas than urban areas (Popovic-Uroic, 1989). A study
in New Zealand (Briesman, 1990) found elevated rates in the rural population and a study
in Canada (Thompson et al., 1986) found higher rates in the rural farm population.
Analytical Epidemiology

The risk factors for sporadic infections and outbreaks are different. Outbreaks are
most commonly linked to the consumption of untreated water (Alary and Nadeau, 1990;
CDC, 1999; Millson et a1. 1991) or unpasteurized milk (Evans et al. 1996; Harris et al.
1987; Komblatt et al., 1985; Orr et al., 1995; Potter et al., 1983; Wood et al., 1992). A

continuous source outbreak of carnpylobacteriosis has been traced to one supplier of

poultry, however (Pearson et al., 2000). The majority ofcases of C. jejuni are sporadic,
i.e., not associated with an outbreak (Friedman et al., 2000a). The epidemiological
features of outbreaks are beyond the scope of this review. The following discussion is
limited to sporadic cases.
Person-to-Person Transmission

Asymptomatic carriage of C. jejuni does not typically occur in industrialized nations
(Blaser et al., 1980). Therefore, person to person transmission is uncommon but there are
a few reports in the literature. In an outbreak involving 35 nursery school students a
father of one of the children developed enteritis (Itoh et al., 1980). In another study,
spread to household contacts was documented in six of 24 families with children testing
positive for C. jejuni (Pai et al., 1979). In another outbreak due to contaminated milk in
school age children, 21% of the cases apparently became infected due to sibling contact
(Jones et al., 1981). It is noteworthy that person—to-person transmission often involves
children as index cases and may reflect poor hygiene in this age group
F oodborne Transmission

Case-control studies have shown repeatedly that the majority of sporadic infections
are food-bome, usually associated with the consumption of undercooked or raw chicken
(see Table l-l). Acquiring campylobacteriosis from undercooked chicken is biologically
plausible because it is known that chickens harbor the organism in their intestinal tract
(Achen ct al., 1998), that their carcasses can become infected during processing (Atabay
and Corry, 1997, Rivoal, et al., 1999), and that the carcass is still contaminated when on
sale in the retail market (Uyttendaele et al., 1999). Thorough cooking of chicken is

required to kill the organism. The consumption of undercooked poultry may contribute

to the age-specific and seasonal features of disease. The age peak in young adults is
thought to correspond with self-reliance in food preparation and probably undercooking

chicken. This has been described as a “second weaning” effect (Altekruse et al., 1994).

Additionally, fried chicken and barbecued chicken, which tend to be undercooked, are

consumed more commonly in the summer months and this may explain the increased

incidence in warmer months (Ikram et al., 1994). Other studies have shown an

association with eating any chicken, undercooked or not.

Table 1-1: Odds ratios for C. jejuni enteritis associated with undercooked chicken

 

 

consumption

Author Study Design Sample Size OR* 95% CIT

Hopkins, et al. 1984 Case-control 40 cases 2.77 1.10, 12.7
71 controls

Harris, et al. 1986 Case-control 218 cases 7.6 2.1, 27.6
526 controls

Deming, et al. 1987 Case-control 45 cases 9.0 1.1, 71.0
45 controls

Ikram, et al. 1994 Case-control 100 cases 4.94 1.03, 23.62
100 controls

Eberhart-Philips, et al. 1997 Case-control 621 cases 3.71 2.24, 6.13
621 controls

Neal and Slack 1997 Case-control 531 cases 2.7 1.1, 7.2
512 controls

Friedman, et al. 2000 Case-control 1463 cases 1.9 1.3, 2.9

1317 controls

 

OR“ = Odds Ratio CIT = Confidence Interval

The finding that consuming any chicken at all would increase the odds of infection

(Table 1-2) may reflect some unidentified undercooked chicken consumption or cross-

contamination in the kitchen. Foods that were cross-contaminated with chicken juices

have been linked to cases of C. jejuni enteritis (CDC. 1998). Thus, improper handling of

raw chicken may confound this exposure.

Table 1-2: Odds ratios for C. jejuni enteritis associated with any chicken

 

 

consumption

Author Design Sample Size OR* 95% Cl‘j'

Harris, et a]. 1986 Case-control 218 cases 2.4 1.6, 3.6
526 controls

Deming, et al. 1987 Case-control 45 cases 4.7 1.3, 16.2
45 controls

Kapperud, et al. 1992 Case-control 52 cases 2.69 1.23, 5.87
103 controls

Neal and Slack. 1997 Case-control 531 cases 1.3 1.1, 1.7
512 controls

Studahl and Andersson. C ase-control 101 cases 2.29 1.29, 4.23

2000 198 controls

Effler, et al. 2001 Case-control 21 1 cases 1.8 1.1, 2.9

21 1 controls

 

OR*= Odds Ratio; 95% C It = 95% Confidence Interval

Other foods associated with carnpylobacteriosis include poultry liver (Schorr et al.,

1994), barbecued sausages (Kapperud et al., 1992), pork chops or loin (Studahl and

Andersson, 2000), raw seafood (Friedman et al., 2000b), and raw milk (Friedman et al.,

2000b; Hopkins et al., 1984; Schmid et al., 1987; Eberhart-Philips et al., 1997). Any raw

or undercooked meat, fish, or shellfish may also increase the odds for sporadic infection

(Eberhart-Philips et al.. 1997; Abeyta et al., 1993). The consumption of milk from
bottles with bird-damaged tops is also associated with illness (Neal and Slack, 2000;
Lighton et al., 1991). These risk factors are also biologically plausible.

Pigs were once thought to carry primarily C. coli, but they are now known to carry C.
jejuni as well. In the Netherlands, 79% of pigs were intestinal carriers of C. jejuni
(Oosterom et al., 1985). and the cecal contents of 3 1% of Texas pigs tested positive in
another study (Harvey et al., 1999). Campylobacter is not commonly isolated from pork
for retail sale, however (Oosterom et al., 1985; Duffy et al., 2001).

Milk may become contaminated with Campylobacter either by direct excretion of the
organism from a mastitic udder or, more commonly, by fecal contamination of milk.
Campylobacterjejuni mastitis has been induced experimentally and C. jejuni has been
reisolated from the milk (Lander and Gill, 1979). Contamination of milk due to C. jejuni
mastitis is thought to be rare, but direct milk excretion leading to human cases has been
reported (Orr et al., 1995). Surveys of raw milk from bulk tanks have revealed variable
rates of infection. Only 1.4% of raw milk samples from 27 farms were positive for C.
jejuni, in one survey. but these farms were chosen specifically because they were known
to produce high quality milk (Desmasures et al., 1997). Another survey of 256 raw milk
samples in Manitoba found only 1.2% of the samples were contaminated (Davidson et al.,
1989) while 12.3% of the milk shipped to an east Tennessee processing plant tested
positive (Rohrbach et al., 1992). Only 500 organisms in milk are needed to induce
enteritis in humans (Robinson, 1981). Campylobacter enters milk through improper
hygiene when milking (Dilworth et al., 1988) or when instruments come in contact with

the milking parlor floor (Humphrey and Beckett, 1986). Though Campylobacter cannot

10

multiply in milk, it can survive for up to 21 days at refrigerator temperatures (Doyle and
Roman, 1982). Pasteurized milk can become contaminated after processing when bottle
tops are peeked by birds (Lighton et al., 1991; Southern et al.. 1990).
Animal Contact

Campylobacterje/‘uni is known to be a normal enteric commensal in the digestive tract
of many warm-blooded animals. Zoo animals, including the llama, cape hydrax,
chimpanzee, flamingo, common peafowl, and fantail pigeon have been found to carry the
organism without exhibiting symptoms of infection (Misawa et al., 2000). Dogs and cats
harbor the organism (Hold and Madsen, 1997; Baker et al., 1999; Fox et al., 1985), as do
many bird species. Wild birds such as crows, gulls, and domestic pigeons can also harbor
and transmit the organism (Kapperud and Rosef, 1983, Riordan et al., 1993).
Investigators in Georgia collected fecal droppings and intestinal and cloacal samples of
wild birds within a 200 foot radius of broiler chicken houses. Ten percent of these total
samples were positive for C. jejuni, indicating that they carry the organism and could
transmit it (Craven et al., 2000). Campylobacterjejuni has also been isolated from the
bill and cloaca of birds (Hudson et al., 1991). Food animals, including chickens
(Uyttendaele et al., 1999; Atabay and Corry, 1997; Achen et al., 1998; Rice et al., 1997),
turkeys (Wallace et al.. 1998), goats (Harris et al., 1987), sheep (Jones et al., 1999), cows
(Busato et al., 1999), and swine (Harvey et al., 1999) carry C. jejuni in their digestive
tracts. It is not surprising, then, that direct contact with animals is associated with C.
jejuni enteritis. Examples of the kind of odds associated with animal contact are

presented in Table 1-3.

Table 1-3: Odds ratios for C. jejuni enteritis associated with animal contact

 

 

Author Study Sample Size Animal OR* 95% CIT
Design
Hopkins ct Case-control 40 cases Cat 3.21 1.25, 8.3
al.1984 71 controls
Deming et al. C ase-control 45 cases Cat 9.0 1.1, 71.0
1987 45 controls
Kapperud et al. C ase-control 52 cases Dog 5.04 1.81, 14.02
1992 103 controls
Saeed et al. 1993 Case-control 218 cases Adult Dog: 11.5 2.0, 65.9
526 controls PuppyI 8.6 1.3, 57
Adak et al. 1995 C ase-control 598 cases Household pet: 2.39 1.09, 5.25
738 controls Livestock and 0.44 0.21, 0.92
feces
(occupational)
Eberhart — C ase-control 621 cases Puppy 3.94 1.57, 9.88
Philips et al. 621 controls Calf feces 4.40 1.34, 14.39
1997
Neal and Slack. Case-control 313 cases Puppy 11.3 1.2, 105
1997 512 controls
Friedman et al. Case-control 1463 cases Farm animals 2.2 1.5, 3.1
2000 1317 Puppy 2.0 1.5, 2.8
controls
Studahl and Case-control 101 cases Chicken 11.83 3.41, 62.03
Andersson 2000 198 controls

 

OR” = Odds Ratio; 95% C IT = 95% Confidence Interval; 1= diarrheic

Companion Animals. Some studies have investigated the association between pets

and C. jejuni infection, but have not found an association (Engleberg et al., 1984; Lighton

et al., 1991). In at least one of these studies (Engleberg et al., 1984), however, the sample

size was too small to detect a risk to pet ownership.

The age and health status of companion animals in the aforementioned studies is

worth noting. In a study of household pets and kennel dogs and cats, the highest isolation

rates were in puppies and kittens (Blaser et al., 1980). Campylobacter jejuni has been
isolated from cats and dogs with and without gastroenteritis (Gondrosen et al., 1985), but
dogs that are less than 12 months old are almost three times as likely to be shedding C.
jejuni if they have diarrhea, compared to those that do not (Burnens et al., 1992). A
randomly selected sample of 11-17 week old puppies found that of 72 puppies, 29% were
positive by culture for Campylopbacter spp. and 76% of those isolates were C. jejuni
(Hald and Madsen, 1997). Other studies have also shown an inverse relationship
between shedding of the organism and age of the dog, with the highest rates in dogs less
than six months of age (Torre and Tello, 1993). The same study also showed that
shedding was heaviest during summer and autumn, compared to winter. This increased
shedding, in addition to the habit of acquiring new puppies in the summer months, may
be another factor contributing to the seasonal distribution of disease in humans. It has
been shown at the ecological level that the months of peak canine births correspond with
the peak months of campylobacteriosis in humans (Evans, 1993). Similarly, the age
distribution may be explained by puppy exposure, since children often have more
intimate contact with puppies than adults do. A small study found a significant
association between Campylobacter infection and presence of puppy in the household in
the zero to five-year old age group (Salfield and Pugh, 1987).

Pets have been directly linked to individual cases. A case report of C. jejuni enteritis
traced the source of infection back to a pet cat. The cat was shown to be excreting C.
jejuni and the patient’s serum titer to the isolate from the cat was high (Blaser et al.,
1982). In a case series. five patients are described; all of which had contact with diarrheic

puppies. Campylobacrerjefuni was isolated from these dogs or their littermates (Blaser

et al., 1978). Another pet that is gaining in popularity and is known to harbor the
organism is the ferret (Taylor et al., 1989). To date, there is no epidemiological evidence
that the ferret is associated with sporadic cases of campylobacteriosis. Future case-control
studies should include an analysis of this pet exposure.

Food Animals. A study in New Zealand revealed the increased odds with several food
animal contacts. These included handling of calf or bovine feces, any contact with cattle,
with new or aborted calves, or with cattle or calf carcasses in the ten days prior to illness
or at work (Eberhart-Philips et al., 1997). Another study of Campylobacter enteritis on
Hopi and Navajo reservations found that case households were more likely than control
households to own farm animals, including poultry, cattle, sheep. goats, horses, or
rabbits. Direct contact with poultry is associated with carnpylobacteriosis.

While case-control studies have shown that daily contact with poultry increases the
odds of infection (Studahl and Andersson, 2000) there is evidence that long-term
occupational contact with chickens is protective. That poultry workers are exposed to C.
jejuni is clear. Campylobacterjejuni organisms have been isolated from the hands of
processing-line poultry workers (Oosterom et al., 1983) and antibody response has been
demonstrated (Jones and Robinson, 1981). A study of chicken abattoir workers in
Sweden found evidence of higher mean lgG antibody levels in long term workers than
short term workers and blood donors and, of eight workers with positive fecal cultures
only one, a new employee, had enteritis (Cawthraw et al., 2000). This has been supported
by decreased risk for occupational contact with livestock in a case-control study (Adak et
al., 1995). Future studies on farm animal contact should assess the specific type of

contact, if the contact is occupational, how frequently the contact occurs, and its duration.

14

Waterborne
Contaminated treated or well water and untreated surface water are both associated

with C. jejuni enteritis. In a case-control study in Colorado, drinking untreated water
was strongly associated with Campylobacterjejuni, increasing the odds for infection
more than 10-fold (Hopkins et al., 1984). Investigators in England found that drinking
untreated water from lakes, rivers, or streams also increased the odds of infection
significantly (Adak et al.. 1995). Non-urban water supplies were associated with illness
in New Zealand (Ikram et al., 1994). Isolation rates for C. jejuni was significantly higher
in rural areas than urban areas in a study carried out in Finland (Martikainen et al., 1990).
The source of this contamination was thought to be grazing animals and pig or cow
manure applied as fertilizer. Campylobacter contamination of a river system was found
to increase after heavy rains increased surface run-off in rural areas (Bolton et al., 1987).
C ampylobacter jejuni was isolated from groundwater after a leaking slurry pit on a dairy
farm contaminated it (Stanley et al., 1998). These findings may explain the increased
incidence noted in rural areas.
Conclusion

Though much is known about how Campylobacter/eiuni infection is acquired, much
remains to be learned. The reasons for the unique age, gender, and seasonal distribution
are not yet fully understood, and the reasons may be different in urban and rural
populations. Future research should be geared towards understanding the risk factors for

disease in these groups because prevention should be targeted to be most effective.

15

References

Abeyta C, Deeter FG, Kaysner CA, Stott RF, Wekell MM. Campylobacter jejuni in a
Washington State Shellfish Growing Bed Associated With Illness. J Food Prot 1993; 56:
323-325.

Achen M, Morishita TY, Ley EC. Shedding and Colonization of Campylobacter jej uni in
broilers from day-of-hatch to slaughter age. Avian Dis 1998; 42: 732-7.

Adak GK, Cowden JM. Nicholas S, Evans HS. The Public Health Laboratory Service
national case-control study or primary indigenous sporadic cases of campylobacter
infection. Epidemiol Infect 1995; 115: 15-22.

Alary M and Nadeau D. An Outbreak of Campylobacter Enteritis Associated with a
Community Water Supply. Can J Public Health, 1990; 81: 268-271.

Altekruse SF, Hunt JM. Tollefson LK, Madden JM. Food and animal sources of human
Campylobacterjejuni infection. J Am Vet Med Assoc 1994; 204(1); 57-61.

Atabay HI and Corry JEL. The prevalence of campylobacters and arcobacters in broiler
chickens. J Appl Microbiol 1997; 83: 619-26.

Baker J, Barton MD, Lanser J. Campylobacter spp. in Cats and Dogs in South Australia.
Aust VetJ 1999; 77: 662-666.

Blaser M, Cravens J , Powers BW, Wang WL. Campylobacter enteritis associated with
canine infection. Lancet 1978; 2(8097): 979-981.

Blaser MJ, Laforce M, Wilson NA, Wang WLL. Reservoirs for Human
Campylobacteriosis. J Infect Dis 1980; 141: 665-669.

Blaser MJ, Weiss SH, Barrett TJ. Campylobacter enteritis associated with a healthy cat.
JAMA, 1982: 247; 816.

Blaser, M.J., Sazie E., Williams P. The Influence of Immunity on Raw Milk-Associated
Campylobacter Infection. JAMA, 1987; 257: 43-46.

Blaser, Martin J. Epidemiologic and Clinical Features of Campylobacter jejuni
infections. J Infect Dis 1997; 176 (Suppl 2): 8103-5.

Bolton F J , Coates D, Hutchinson DN, Godfree AF. A study of thermophilic
campylobacters in a river system. Journal of Applied Bacteriology, 1987; 62: 167-176.

Bremell T, Bjelle A, Svedhem A. Rheumatic symptoms following an outbreak of
campylobacter enteritis: a five year follow up. Ann Rheum Dis 1991; 50: 934-938.

Brieseman MA. A further study of the epidemiology of Campylobacter jejuni infections.
NZ Med J 1990; 103: 207-209.

Bumens AP, Angeloz-Wick B, Nicolet J. Comparison of Campylobacter carriage rates in
diarrheic and healthy pet animals. J. Vet. Med. B, 1992; 39: 175-80.

Busato A, Hofer D, Lentze T, Gaillard C, Bumens A. Prevalence and infection risks of
zoonotic enteropathogenic bacteria in Swiss cow-calf farms. Vet Microbiol 1999; 69:
251-263.

Buzby JC, Allos BM, Roberts T. The Economic Burden of Campylobacter-associated
Guillain—Barre Syndrome. J Infect Dis 1997; 176 (suppl 2): S192-7.

Casemore D.P., E.G. Jessop, D. Douce, and F .B. Jackson. Cryptosporidium plus
campylobacter: an outbreak in a semi-rural population. Journal of Hygiene, 1986; 96:
95-105.

Cawthraw SA, Lind L. Kaijser B, Newell DG. Antibodies, directed towards
Campylobacterjejuni antigens, in sera from poultry abattoir workers. Clin Exp Immunol,
2000; 122: 55-60.

CDC. Outbreak of Campylobacter Enteritis Associated with Cross-Contamination of
Food— Oklahoma, 1996. MMWR 1998; 47(07): 129-131.

CDC. Outbreak of Escherichia coli 01572H7 and Campylobacter among attendees of the
Washington County Fair — New York, 1999. MMWR 1999; 48(36); 803-4.

CDC. Preliminary F oodNet Data on the Incidence of Foodbome Illnesses --- Selected
Sites, United States, 2000. MMWR 2001: 50(13): 241-6.

Craven SE, Stern NJ, Line E, Bailey J S, Cox NA, F edorka-Cray P. Determination of the
Incidence of Salmonella spp., Campylobacter jejuni, and Clostridium perfringens in Wild

Birds near Broiler Chicken Houses by Sampling Intestinal Droppings. Avian Dis 2000;
44: 715-20.

Davidson RJ, Sprung DW, Park CE, Rayman MK. Occurrence ofListeria
monocytogenes, Campylobacter spp. and Yersinia enterocolitica in Manitoba Raw Milk.
Canadian Institute of Food Science and Technology Journal, 1989; 22: 70-74.

Deming MS, Tauxe RV, Blake PA, Dixon SE, Fowler BS, Jones TS, et al.
Campylobacter enteritis at a university: transmission from eating chicken and from cats.
Am J Epidemiol 1987; 126(3): 526-534.

Desmasures, N., F. Bazin, and M. Gueguen. Microbiological composition of raw milk
from selected farms in the Camembert region of Normandy. J Appl Microbiol 1997; 83:
53-58.

Dilworth CR, Lior H, Belliveau MA. Campylobacter Enteritis Acquired from Cattle.
Can J Public Health, 1988; 79: 60-62.

Doyle MP and Roman DJ. Prevalence and Survival of Campylobacterjejuni in
Unpasteruized Milk. Appl Environ Microbiol 1982; 44: 1154-1158.

Duffy EA, Belk KE. Sofos JN, Bellinger GR, Pape A, Smith GC. Extent of Microbial
Contamination in United States Pork Retail Products. J Food Prat 2001; 64: 172-178.

Eberhart-Philips J, Walker N, Garrett N, Bell D, Sinclair D, Rainger W, et al.
Campylobacteriosis in New Zealand: results of a case-control study. J Epidemiol
Community Health, 1997; 51: 686-691.

Effler P, Ieong M-C, Kimura A, Nakata M, Burr R, Cremer E, et al. Sporadic
Campylobacter jejuni Infections in Hawaii: Associations with Prior Antibiotic Use and
Commercially Prepared Chicken. J Infect Dis 2001; 183: 1152-5.

Engleberg NC, Correa-Villasenor A, North CQ, Crow T, Wells JG, Blake PA.
Campylobacter enteritis on HOpi and Navajo Indian Reservations. West J Med 1984;
141(1): 53-56.

Evans SJ. The seasonality of canine births and human carnpylobacteriosis: a hypothesis.
Epidemiol Infect 1993; 110: 267-272.

Evans MR, Roberts RJ, Ribeiro CD, Gardner D, Kembrey D. A milk-bome
campylobacter outbreak following an educational farm visit. Epidemiol Infect 1996; 117:
457-462.

Fox JG, Ackerman JA, Newcomer CE. The Prevalence of Campylobacterjejuni in
Random Source Cats Used in Biomedical Research. J Infect Dis 1985; 151(4): 743.

Friedman C, Reddy S, Samual M, Marcus R, Bender J, Desai S, et al. Risk Factors for
Sporadic Campylobacter Infections in the United States: A Case-Control Study on

FoodNet Sites. 2nd International Conference on Emerging Infectious Diseases. Atlanta,
GA, July 2000.

Friedman CR, Neimann J, Wegener HC, Tauxe RV. Epidemiology of Campylobacter
jejuni Infections in the United States and Other Industrialized Nations. In: Nachamkin I,
Blaser MJ, editors. Campylobacter 2nd ed. Washington, DC, ASM Press; 2000. p. 127.

Gondrosen B, Knaevelsrud T, Dommarsnes K. Isolation of Thermophilic
Campylobacters from Norweigan Dogs and Cats. Acta vet Scand, 1985; 26: 81-90.

18

Hald B and Madsen M. Healthy Puppies and Kittens as Carriers of Campylobacter spp.,
with Special Reference to Campylobacter upsaliensis. J Clin Microbiol 1997; 35(12):
3351-2.

Harris NV, Weiss NS. Nolan CM. The Role of Poultry and Meats in the Etiology of
Campylobacter jejuni/coli enteritis. Am J Public Health, 1986; 76: 407-11.

Harris NV, Kimball TJ, Bennett P, Johnson Y, Wakely D, Nolan CM. Campylobacter
jejuni enteritis associated with raw goat’s milk. Am J Epidemiol 1987; 126(2): 179-186.

Harvey RB, Young CR. Ziprin RL, Hume ME, Genovese KJ, Anderson RC, Droleskey
RE, Stanker LH, Nisbet DJ. Prevalence of Campylobacter spp isolated from the

intestinal tract of pigs raised in an integrated swine production system. J Am Vet Med
Assoc 1999; 215: 1601-1603.

Hopkins RS, Olmsted R, Istre GR. Endemic Campylobacter jej uni infection in Colorado:
Identified Risk Factors. Am J Public Health, 1984; 74: 249-250.

Hudson SJ, Lightfoot NF, Coulson JC, Russell K, Sisson PR, Sobo AO. Jackdaws and
magpies as vectors of milkbome human campylobacter infection. Epidemiol Infect 1991;
107: 363-372

Hughes and Rees. Clinical and Epidemiologic Features of Guillain—Barré Syndrome. J
Infect Dis 1997: 176(suppl 2): 892-8.

Humphrey TJ and Beckett P. Campylobacterjejuni in dairy cows and raw milk.
Epidemiol Infect 1987; 98: 263-269.

Ikram R, Mitchess P, Brieseman MA, Ikram OH. A case control study to determine risk
factors for campylobacter infection in Christchurch in the summer of 1992-3. N Z Med J
1994; 107: 430-32

Itoh T, Saito K, Maruyama T, Sakai S, Ohashi M, Oka A. An Outbreak of Acute
Enteritis Due to Campylobacter fetus Subspecies jejuni at a Nursery School in Tokyo.
Microbiol Immunol, 1980; 24(5): 371-9.

Jones DM, Robinson DA. Occupational Exposure to Campylobacter jejuni infection.
Lancet ,1981; i : 440-441.

Jones EM, Robinson DA, and Eldridge J. Serological studies in two outbreaks of
Campylobacter jejuni infection. Journal of Hygiene, 1981; 87: 163.

Jones K, Howard S, Wallace J S. Intermittent shedding of thermophilic campylobacters
by sheep at pasture. J Appl Microbiol 1999; 86: 531-536.

19

Kapperud G, Rosef O. Avian Wildlife Reservoir of C ampylabacter fetus subsp. jejuni,
Yersinia spp., and Salmonella spp. in Norway. Appl Environ Microbiol 1983; 45: 375-
380.

Kapperud G, Skjerve E, Bean NH, Ostroff SM, Lassen J. Risk Factors for Sporadic
Campylobacter Infections: Results of a Case-control Study in Southeastern Norway, J
Clin Microbiol 1992; 30(12): 3117-3121.

Kapperud G, Aasen S. Descriptive epidemiology of infections due to therrnotoleant
Camppylobacter spp. in Norway, 1979-1988. APMIS 1992; 100: 883-90.

Komblatt AN, Barrett T. Morris GK, Tosh FE. Epidemiologic and laboratory
investigation of an outbreak of Campylobacter enteritis associated with raw milk. Am J
Epidemiol 1985; 122(5): 884-889.

Lander KP, Gill KPW. Experimmental infection of the bovine udder with Campylobacter
coli/jejuni. Journal of Hygiene, 1980; 84: 421-428.

Lighton LL, Kaczmarski EB, Jones DM. A study of Risk Factors for Campylobacter
Infection in Late Spring. Public Health, 1991; 105: 199-203.

Manfredi R, Nanetti A. F erri M, Chiodo F. Fatal Campylobacter jejuni bacteraemia in
patients with AIDS. J Med Microbial, 1999; 48: 601-3.

McCarthy N, Giesecke J. Incidence of Guillain-Barré Syndrome following infection with
Campylobacterjejuni. Am J Epidemiol 2001; 153(6): 610-614.

Millson M, Bokhout M. Carlson J, Spielberg L, Aldis R, Borczyk A, et al. An Outbreak
of Campylobacter jej uni gastroenteritis linked to meltwater contamination of a municipal
well. Can J Public Health, 1991; 82: 27-31.

Misawa N, Shinohara S, Sato H, Itoh H, Shinohara K, Shimomura K, et al. Isolation of

Campylobacter species from zoo animals and polymerase chain reaction-based random
amplified polymorphism DNA analysis. Vet Microbial 2000; 71: 59-68.

Neal KR, Slack RCB. Diabetes mellitus, anti-secretory drugs and other risk factors for
campylobacter gastro-enteritis in adults: a case-control study. Epidemiol Infect 1997;

119: 307-11.

Oosterom J, Notermans S., Karman H, Engels G.B. Origin and Prevalence of
Campylobacter jejuni in Poultry Processing. J Food Prat, 1983; 46 (4): 339-344.

Oosterom J, Dekker R, deWilde GJA, vanKempen—deTroye F, Engels GB. Prevalence of
Campylobacter jejuni and Salmonella during pig slaughtering. Vet Q 1985; 7: 31-34.

20

Orr KE, Lightfoot NF. Sisson PR, Harkis BA, Tweddle JL, Boyd P, et al. Direct milk
excretion of C ampylabacter jejuni in a dairy cow causing cases of human enteritis.
Epidemiol Infect 1995; l 14: 15-24.

Pi CH, Sorger S, Lackman L, Sinai RE. Marks MI. C ampylobacter gastroenteritis in
children. J Pediatr 1979: 94: 589-91.

Pearson AD, Greenwood MH, Donaldson J, Healing TD, Jones DM, Shahamat M, et al.
Continuous Source Outbreak of Campylobacteriosis Traced to Chicken. J F oad Prat
2000: 63(3); 309-314.

Popovic-Uroic T. Campylobacter jejuni and Campylobacter coli diarrhoea in rural and
urban populations in Yugoslavia. Epidemiol Infect 1989; 102: 59-67

Potter ME, Blaser MJ, Sikes RK, Kaufmann AF, Wells JG. Human Campylobacter
infection associated with certified raw milk. Am J Epidemiol 1983; 117(4): 475-483.

Rice BE, Rollins DM, Mallinson ET, Carr L Joseph SW. Campylobacter jejuni in broiler
chickens: colonization and humoral immunity following oral vaccination and
experimental infection. Vaccine 1997: 15: 1922-32.

Rivoal M, Denis M, Saivat G, Colin P, Errnel G. Molecular characterization of the
diversity of Campylobacter spp. isolates collected from poultry slaughterhouse: analysis
of cross-contamination. Lett Appl Microbiol 1999; 29: 370-4.

Robinson DA. Infective dose of Campylobacter jejuni in milk. BMJ 1981; 282: 1584.
Rohrbach BW, Draughon FA, Davidson PM, Oliver SP. Prevalence of Listeria
monacytogenes, Camypylabacterjejuni, Yersinia enterocalitica, and Salmonella in Bulk

Tank Milk: Risk Factors and Risk of Human Exposure. J Food Prat 1992; 55: 93-97.

Saeed AM, Harris NV, DiGiacomo RF. The Role of Exposure to Animals in the Etiology
of Campylobacterjejuni / coli Enteritis. AmJ Epidemiol 1993; 137; 108-114.

Salfield NJ, Pugh EJ. Campylobacter enteritis in young children living in households
with puppies. BMJ 1987; 294: 21-22.

Schmid GP, Schaefer, RE, Plikaytis BD, Schaefer JR. Bryner JH. Wintermeyer LA, et al.
A One-Year Study of Endemic Campylobacteriosis in a Midwestern City: Association

with Consumption of Raw Milk. J Infect Dis 1987; 156(1): 218-222.

Schorr D, Schmid H, Rieder HL, Baumgartner A, Vorkauf H, Bumens A. Risk Factors
for Campylobacter enteritis in Switzerland. Zbl Hyg, 1994: 196: 327-337.

Sibbald CJ, Sharp CM. Campylobacter infection in urban and rural populations in
Scotland. Journal of Hygiene, 1985; 95: 87-93.

21

Sillero M, Almirall J. C ampylobacter jejuni and Hemolytic-Uremic Syndrome.
Nephran, 1999; 82: 363-4.

Skirrow MB, Blaser MJ. Clinical Aspects of Infection. In: Nachamkin I, Blaser MJ
editors. Campylobacter. 2nd ed. Washington, DC: ASM Press; 2000: pg. 79.

Sorvillo FJ, Lieb LE, W atennan SH. Incidence of campylobacteriosis among patients
with AIDS in Los Angeles County. J Acquir Immune Defic Syndr Hum Retrovirol, 1991;
598-602.

Southern JP, Smith RMM, Palmer SR. Bird attack on milk bottles: possible mode of
transmission of Campylobacter jejuni to man. Lancet. 1990; 336: 1425-1427.

Stanley K, Cunningham R, Jones K. Isolation of Campylobacterjejuni from groundwater.
JAppl Microbiol 1998: 85: 187-191.

Studahl A, Andersson Y. Risk factors for indigenous campylobacter infection: a Swedish
case-control study. Epidemiol Infect 2000; 125: 269-275.

Tauxe RV, Pegues DA. Hargrett-Bean N. Campylobacter infections: the emerging
national pattern. Am J Public Health, 1987; 77: 1219-1221.

Tauxe, RV. Epidemiology of Campylobacter jejuni infections in the United States and
other industrial nations. In: Nachamkin, Blaser, Tompkins, editiors. Campylobacter
jejuni: current and future trends. Washington: American Society for Microbiology; 1992.
P. 9-12

Taylor NS, Ellenverger MA, Wu PY, Fox JG. Diversity of serotypes of Campylobacter
jejuni and C ampylobacter coli isolated in laboratory animals. Laboratory Animal
Science, 1989; 39: 219-221

Thompson J S, Cahoon FE, Hodge DS. Rate of Campylobacter spp. Isolation in Three
Regions of Ontario, Canada, from 1978 to 1985. J Clinical Microbial 1986; 24(5): 876-
878.

Torre E, Tello M. Factors influencing fecal shedding of Campylobacter jejuni in dogs
without diarrhea. Am J Vet Res 1993; 54(2): 260-262.

Uyttendaele M, De Troy P, Debevere J. Incidence of Salmonella, Campylobacter jejuni,
Campylobacter coli, and Listeria monocytogenes in Poultry Carcasses and Different
Types of Poultry Products for Sale on the Belgian Retail Market. J F oad Prat, 1999:
62(7): 735-740.

Wallace J S, Stanley KN, Jones K. The colonization of turkeys by thermophilic
campylobacters. J Appl Microbiol 1998; 85: 224-230.

22

Wood RC, MacDonald KL, Osterholm MT. Campylobacter enteritis outbreaks
associated with drinking raw milk during youth activities. JAMA 1992; 268(22): 3228-
30.

23

CHAPTER 2

CAMPYLOBA CTER JEJUNI. GUILLAIN-BARRE SYNDROME, AND THE CAUSAL
CRITERIA

Abstract

Guillain-Barré Syndrome (GBS) is the most common cause of acute generalized
paralysis since the decline of polio. Current estimates suggest that approximately 30% of
these cases are triggered by C ampylabacter jejuni, the most frequently isolated cause of
bacterial diarrhea in the United States. Because the costs of C. jejuni associated GBS are
very high due to treatment, loss of productivity and life, and because of the moderate
potential for exposure to C. jejuni, research investigating the association between C.
jejuni and GBS has become a priority. In this paper Hill’s criteria are used to evaluate the
evidence for a causal relationship between C. jejuni and GBS and to identify gaps in
current understanding of the association. While there is increasing evidence for an
association, we conclude that additional research needs to be done before C. jejuni is
considered a cause of GBS.
Introduction

Guillain-Barré Syndrome (GBS) is the most common cause of acute, generalized
paralysis since the decline of polio (Blaser et al., 1997). It is a subacute polyneuropathy
that affects the motor, sensory and autonomic nerves supplying the limbs and may extend
to involve the respiratory muscles and cranial nerves. C urrently. between 2,628 to 9,575
new cases of GBS occur every year in the United States (Buzby and Roberts, 1997).
About 10% of these patients will die, while another 10% will not completely recover

(Hughes and Rees, 1997).

24

Since Guillain, Barre. and Strohl described the syndrome in 1916, it has been known
as a post-infectious condition. If C. jejuni enteritis is a cause of GBS, it will be the most
common antecedent infection, accounting for between 532 and 3.830 cases per year
(Buzby and Roberts, 1997). The risk of GBS following C. jejuni infection has been
estimated to be between one in 1,058 (Allos, 1997) and one in 3.285 (McCarthy and
Giesecke, 2001). The cost of C. jejuni associated GBS is estimated to be between $0.2 to
$1.8 billion each year in the United States alone (Buzby et al., 1997). When one
considers the cost of GBS, the ubiquitous nature of C. jejuni in the environment and that
it is the most common cause of bacterial gastroenteritis in the United States (CDC, 2001),
it is apparent that determining if C. jejuni is a true cause of GBS should be a priority.

Epidemiologists use established criteria to help determine if an association is causal
(Hill, 1965). These include strength of the association, consistency, specificity,
temporality, biologic gradient, plausibility, coherence, and analogy. Evaluating the
association with these criteria allows one to determine if the putative association between
C. jejuni and GBS is causal and, if it is not, to direct targeted research. Therefore, the
objective of this chapter is to evaluate the published literature to determine whether there
is sufficient evidence for a causal association between C. jejuni and GBS.

Strength of the Association

In 1982, the first case of GBS that was preceded by a gastrointestinal illness
caused by C. jejuni was reported. Campylobacterjejuni had been cultured from
the patient’s stool and a serum antibody response to the organism was also
demonstrated (Rhodes and Tattersfield, 1982). Numerous case reports and case

series from other investigators followed (Molnar et al., 1982: Sovilla et al., 1988).

25

Serosurveys of GBS patients also showed that C. jejuni antibody titers were
higher than would be expected in the general population (Speed et al., 1987;
Kuroki et al.. 1991). Though these studies were not comparing cases and
controls, they suggested a potentially strong association between C. jejuni and
GBS because it appeared that C. jejuni enteritis would be one of the most, if not
the most, common antecedent illnesses in GBS patients. Strength of association
cannot be assessed by case series, however. It must be measured in an analytical
epidemiologic study.

In 1993, Gregson, et al. compared GBS patients with healthy controls
(Gregson et al., 1993). Elevated antibody titers to C. jejuni were detected in 15 of
the 42 cases (35%) and none of the 41 controls. Another 1993 study also found
that GBS cases were more than five times as likely (95% CI 2.4. 12.5) to have C.
jejuni seropositivity compared to controls (Mishu et al., 1993). In a later study, C.
jejuni antibody titers in GBS patients were compared to those in healthy controls,
multiple sclerosis patients, myasthenia gravis patients, and neuroborreliosis
patients. The GBS patients were significantly more likely (p<0.05) to be C. jejuni
specific IgA positive than all other control groups, except for the myasthenia
gravis group (Enders et al., 1993). Kaldor and Speed found evidence of C. jejuni
in 21 of 56 GBS patients and none of 30 healthy controls (Kaldor and Speed,
1984). In 1995, Rees, et al., found 26% of GBS patients were diagnosed as C.
jejuni positive compared to two percent of household controls and one percent of
hospital controls, a highly significant difference (Rees et al., 1995). Ho et al. also

found a large odds ratio for evidence of preceding C. jejuni infections in GBS

26

patients when compared with controls. Their point estimate was 10.2 with 95%
confidence interval 3.8 to 27.8 (Ho et al., 1995). In a study by Jacobs, et al.
(Jacobs et al., 1996), C. jejuni was found to be the most common antecedent
infection among 154 GBS patients. It was more common than infection with the
other common antecedent infections; cytomegalovirus, Epstein-Barr virus,
Mycoplasma pneumaniae. Haemophilus influenzae, parainfuluenza 1 virus,
influenza A, influenza B, adenovirus, herpes simplex virus, or varicella zoster
virus. The odds ratio for infection with C. jejuni was 3.1 (95% CI 1.7, 5.9)
between GBS cases and other neurological disease controls and 5.4 (95% CI 1.9,
15.1) between cases and healthy subjects controls.

Numerous studies have shown that the association between C. 'ejuni and GBS
is strong and this finding enhances our suspicion that the relationship between the
two may be causal. The reason for this is that it is hard to imagine a potential
confounding biological or environmental variable that is so strongly associated
with both C. jejuni infection and GBS that it could drive this association.
Consistency

Consistency refers to the repeated observation of an association by different
investigators in different populations, places, circumstances, and times. This
means that for a given association to be considered causal, the association should
be found in a wide variety of situations and analytical designs. Consistency does
not imply repeatability per se, because a flaw in study design may also be

repeated.

27

A review of the strength criterion above will show that the association has
been found repeatedly by different investigators. We are unaware of any
investigations that have failed to find an association between GBS and C. jejuni,
but this may only reflect the under-publishing of negative results. The association
has also been found in different populations including those in Italy, China, Japan,
the United States, the United Kingdom, and South Africa.

The analytical epidemiologic investigations to date, however, have been
limited to retrospective or prospective case-control study designs with
hospitalized cases. The consistency of the association has, therefore, not been
established. The case-control design is well suited to the study of GBS because
the disease is rare and because multiple etiologic factors can be assessed.
Additionally, it is relatively quick and inexpensive when compared to cohort
designs. The limitations of this study design, however. are several. First, it is
inefficient for the study of rare exposures, such as some C. jejuni serotypes,
unless the attributable risk is high. Second, as will be discussed later, the
temporal relationship between exposure and outcome is difficult to establish.
Third, case-control studies are particularly prone to biases, particularly selection
and recall bias.

Selection bias is defined as an error due to systematic differences in the
characteristics of those who are selected for study and those who are not.
Selection of study participants may be biased due to the exposure, the disease, or
a confounder. One possible example of selection bias that could be occurring on

exposure in the hospital based case-control studies is Berkson’s bias. Berkson‘s

28

bias occurs when the combination of exposure and disease under study increase
the risk of admission to a hospital which leads to a systematically higher exposure
rate among the hospital cases than the hospital controls. Evidence for this is that
C. jejuni seems to lead to a more severe form of GBS. acute motor axonal
neuropathy (AMAN), and may have a more rapid onset and delayed recovery
(Hadden et al., 2001). Between 4.7 and 28% of GBS patients have a milder
disease, retaining their ability to walk throughout their illness (Green et al. 2001)
and these mildly affected patients are less likely to have evidence of C. jejuni and
other antecedent infections (VanKoningsveld et al., 2000). Thus, GBS patients
with antecedent C. jejuni infection may be more likely to be admitted to a referral
center or be included in a clinical trial and become a study participant than GBS
patients without prior C. jejuni infection. Several studies have been limited to
cases in referral centers. infectious disease hospitals, or clinical trials (Enders et
al., 1993; Kaldor and Speed, 1984; Jacobs et al., 1998). Admitted hospital
patients with other neurological diseases have often been used as controls. Prior
C. jejuni infection would play no role in the risk of their admission to the hospital.
If this bias were operating, it would lead to an over-estimation of the odds ratio.
Additional studies should be done to determine the effect of this bias and how
many, if any, GBS patients never come to the attention of a neurologist.

Recall bias is a systematic error due to differences in accuracy of recall of past
events between cases and controls. Cases are more likely to remember antecedent
events than are healthy controls. When C. jejuni exposure is defined by elevated

titers, recall bias is not a problem, but recall of diarrheal illness should not be used

29

as a proxy for C. jejuni infection because it would lead to an overestimate of the
association.

Another difficulty with case-control studies is choosing an appropriate control
group. Controls should represent the exposure experience of the source
population. Because C. jejuni enteritis has a strong gender, age, and seasonal
distribution. controls should be matched to GBS patients on these criteria. A
failure to do so could bias study results in any direction. Failure to adhere to
these principles leads not only to the selection bias discussed above, but it also
decreases the internal and external validity of the study results.

Though the association has been observed by different investigators in
different places and circumstances, the exclusive use of the case-control design
with hospitalized cases means that the causal criterion of consistency has not been
satisfied. Consistency is not repeatability.

Specificity

The criterion of specificity may be reviewed in two different ways. It may be
looked at in terms of the specificity of the relationship between the exposure and
outcome or in terms of the specificity of the definition of exposure and outcome.

The first way is not very useful because diseases often have more than one
cause. GBS itself is a syndrome of multiple causes. It is possible to get GBS
without ever having had C. jejuni, just as it is possible to have C. jejuni without
ever developing GBS. Indeed, many possible causes of GBS have been identified
including Epstein-Barr virus, Mycoplasma pneumoniae, and cytomegalovirus

infections. There are case-control studies to support these associations (Jacbos et

30

al., 1998, Winer et al.. 1988). Other infectious agents. such as T axoplasma gandii
(Bossi et al., 1998), Helicabacter pylori (Chiba et al., 1998), hepatitis A (Ono et
al., 1994), Rocky Mountain Spotted Fever (Toemer et al., 1996), Cyclaspora
(Richardson et al., 1998), and West Nile Virus (Ahmed et al., 2000) have been
incriminated in GBS case reports. These findings do not mean that the
relationship is not causal; it only means that C. jejuni is neither a necessary nor a
sufficient cause of GBS and that is an indication to apply the causal criteria.

A more useful application of this criterion is to the definition of exposure and
outcome. If both exposure and outcome are very specifically defined and the
relationship holds, causality is enhanced. In terms of GBS and C. jejuni, this
means specificity of serotype of the organism and form of GBS.

Campylobacterjejuni must be cultured in order for it to be typed. During the
two week period between C. jejuni infection and GBS the organism is often
cleared from the GI tractor is shed in such low numbers that repeated cultures are
needed to detect it (Nachamkin, 1997). When C. jejuni has been cultured from
the stools of GBS patients and typed, serotypes 0:2, 0:4, 0211, O: 19, and 0:41
are predominantly found (Allos et al., 1998, Goddard et al., 1997). In
retrospective studies, O:l9 (Kuroki et al., 1993) and 0:41 (Lastovica et al., 1997)
is particularly predominant in isolates from GBS patients compared to
uncomplicated C. jejuni enteritis isolates. Perhaps there may be other serotypes
that also precede GBS, but these serotypes may tend to resist clearance long
enough for the patient to develop GBS and for the organism to be cultured.

Strains that induce GBS but are rapidly cleared would not be available for culture.

31

GBS may manifest itself in one of several ways. It may be the demyelinating
form (AIDP), or cause axonal degeneration of exclusively motor nerves (AMAN)
or both motor and sensory nerves (AMSAN). The AMAN and AMSAN forms
are associated with poor or delayed recoveries. In the early 19905, case reports
started to suggest that C. jejuni was associated with a specific, severe form of
GBS. Case reports have revealed evidence of antecedent C. jejuni in AMAN
patients (Lugaresi et al., 1997). Case-comparison studies have also been done.
One retrospective study of GBS patients found that poor recovery was
significantly more frequent in 10 patients with high anti-C. jejuni titers (3 of 10)
than in patients without this infection (0 of 48) (Kaldor and Speed, 1984;
Vriesendorp etal., 1995). Nerve conduction studies of these patients were
initially consistent with the AIDP form of the disease but as their disease
progressed, axonal degeneration was evident. In 1996, Jacobs, et al., found that
patients who were C. jejuni and anti-GMI antibody positive were clinically
distinct from patients without this antibody profile. They more often had the
AMAN form of GBS (Jacobs et al., 1996). Additionally, C. jejuni positive
Chinese patients with GBS were more likely to exhibit the AMAN form of the
disease than C. jejuni negative patients (N evo and Pestronk, 1997). Another study
of 55 GBS patients in Japan, however, found no association between the type of
GBS and the presence of specific anti-ganglioside antibodies (Yuki et al., 1999).
Additionally, C. jejuni is not the only organism associated with AMAN;
Haemophilus influenzae is also capable of eliciting this pathophysiologic response

(Mari et al., 2000).

32

At this time, the authors are not aware of any study that attempts to directly link
a specific C. jejuni serotype with severity or form of GBS. It has been indirectly
investigated, however. and this will be discussed as evidence for the biological
plausibility criterion. If future studies find that a specific serotype of C. jejuni
causes an elevation of some specific anti-ganglioside antibodies and produces a
specific form of GBS. then that would be strong evidence for causality.
Temporality

Because the C. jejuni and GBS association has been primarily investigated
through case-control studies that measure exposure and outcome simultaneously.
the temporal relationship between the two has not been well established.
Exposure has been measured by means of antibody titers, culture, and recall of
gastrointestinal illness. Elevated antibody titers have been very commonly used
to determine exposure and are most troublesome for this criterion.

Serum antibody levels are typically elevated for several weeks and even
months after infection. so anti-C. jejuni antibody in the serum of a GBS patient
may reflect an exposure that occurred several months prior, and may have no
relevance to the onset of GBS. It could also represent an exposure that occurred
after the disease process had begun. Specifically, IgA titers are known to decline
rapidly about one week after infection, while lgG levels peak about two weeks
after onset of symptoms and decline after three or more weeks. IgG antibody may
persist at high levels for several months in some patients (Kaldor et al., 1983,
Blaser and Duncan, 1984). Studies that rely on recall of symptoms of

gastroenteritis two to three weeks before onset of GBS are trying to establish

33

time-order, but they are prone to recall bias, and some C. jejuni cases may be mild
or subclinical. In other case-control studies, exposure is measured by culture and,
though this method is less sensitive than serology, it is a better measure for very
recent antecedent infection than antibody titers because the organism is typically
cleared from the GI tract two to three weeks after infection.

Similarly, the role of anti-ganglioside antibodies has not been established.
They, too, have been measured after onset of GBS but does their presence
indicate a response to the disease or a cause of it? The best way to establish time
order for this disease is through a prospective cohort study or with an animal
model.

Biologic Gradient

This criterion refers to a dose-response relationship. In this case, it could be
interpreted either as seeing an increased incidence of GBS in locations with
higher incidence of C. jejuni, or as an increasing risk for GBS with increasing
dose of C. jejuni. The descriptive epidemiology of the disease in China argues
more strongly for an infectious cause then does the descriptive epidemiology of
the disease in developed countries. In China, where C. jejuni is a common enteric
pathogen and significant numbers of children are asymptomatic carriers (Young et
al., 1986), epidemics of the AMAN form of Guillain-Barré syndrome occur in the
rural areas of northern China every summer. The primary involvement of
children from rural areas with peak incidence in the summer is especially
intriguing. Rural Chinese have been reported to drink unboiled well water and

have contact with a variety of farm animals, including chickens, pigs, goats, and

34

dogs - all known risk factors for C. jejuni (McKhann et al., 1993). In a case-
control study of GBS patients from these rural areas, significantly more had
elevated titers to C. jejuni (66%) than did village controls (16%). The authors
calculated that GBS patients were 10.2 (95% CI 3.8, 27.8) times more likely to
have elevated anti-C. jejuni antibody titers than village controls (Ho et al., 1995).
In an area of the world where C. jejuni infection is common, there is a pattern of
GBS that is more severe (AMAN) and has the epidemiologic features of an
infectious disease.

It is very difficult for an epidemiologic study to estimate the dose of pathogen
ingested by cases. so traditional information on dose-response will have to wait
for an appropriate animal model.

Plausibility

Fulfillment of this criterion requires that the hypothesized causal relationship
between C. jejuni and GBS is biologically sound. There is circumstantial
evidence to support the role of C. jejuni in GBS because secretory antibody is one
of the primary barriers to adherence of the organism and the timing of peak
antibody response to infection corresponds with GBS onset. This criterion has
been extensively investigated in the literature.

It is believed that C. jejuni infection stimulates an autoimmune response
through the process of molecular mimicry. In molecular mimicry, a host acquires
an infection with an agent that has antigens that are immunologically similar to a
host antigen, but differs sufficiently to induce an immune response. As a result,

the tolerance to autoanti gens breaks down and the pathogen-specific immune

35

response cross-reacts with host structures to cause tissue damage and disease
(Albert and Inman, 1999). If this describes the pathogenesis of GBS following C.
jejuni infection, then C. jejuni antigen must be similar to neurological host
antigen, antibodies to host antigen should be produced in the immune response to
infection, and these autoantibodies should cross-react with C. jejuni antigen.
Lipopolysaccharides (LPS), also known as endotoxins, are parts of the outer
membrane of most gram-negative bacteria. Structurally, the LPS is composed of
a polysaccharide or oligosaccharide covalently bound to a lipid component.
Gangliosides are glycospingolipids that tend to occur in regional patterns in the
nervous system. Gangliosides are referred to by an alphanumeric abbreviation. G
stands for ganglia and the M, D, T, or Q following indicate the number of sialic
acid residues. The numbers and lowercase letters following refer to the sequence
of migration that is determined by thin-layer chromatography (Willison and
O’Hanlon, 2000). The ganglioside GMI is widely distributed in the nervous
system. It has been found on axonal membranes at the node of Ranvier and the
neuromuscular junction and on myelin, especially around the nodes (Feasby and
Hughes, 1998). It has been shown that the core oligosaccharide (OS) of C. jejuni
O:19 LPS contains tetra— and pentasaccharide moieties identical to those of GM]
and GDla gangliosides. respectively (Moran et al.,1996). The LPS of 0:41
isolates bear a GMl-Iike epitope and the core OS of 0:4 mimics GDIa and GM]
epitopes (Yuki et al., 1997). These findings indicate that there are structural

similarities between C. jejuni antigen and neurological host antigen.

36

It has also been shown that there is strong cross-reactivity between anti-GM]
IgM and the LPS from C. jejuni serotypes 0:4 and 0:19 (Wirguin et al., 1994).
Additionally, cholera toxin, which specifically recognizes the GM]
oligosaccharide has shown reactivity with the LPS fraction of C. jejuni O:19
(Yuki et al.. 1990).

Case control studies have found anti-GMI antibodies in C. je/‘uni-preceded
GBS patients, but not in controls (Walsh et al., 1991). In a study of GBS patients
with anti-GMI and / or anti-Gle antibodies, these patients were more likely to
have serological evidence of antecedent C. jejuni enteritis than GBS patients
without these antibodies (Schwartz et al., 1982). In contrast, other studies have
found no correlation between GMI binding and C. jejuni, but did find the
infection was associated with poor recovery and severe disease (Mishu et al.,
1993). The association between C. jejuni and antiganglioside antibodies has not
consistently been shown.

An alternate explanation of the role of antiganglioside antibodies is that they
block neuronal sodium channels causing AMAN-like clinical signs, but with rapid
resolution. There is some experimental evidence that GM] can do this (Weber et
al., 2000). It has not been determined how commonly this occurs. and what host
factors, C. jejuni strains, or dose determines when this occurs.

The timing of the occurrence of antiganglioside antibodies is significant
because, if they are induced by C. jejuni, they should precede the onset of GBS
clinical signs. The implication here is that degenerating and damaged nerves may

be the source of the anti-ganglioside response in GBS. It has been shown that

37

anti-ganglioside antibodies are produced following nerve injury (Winer et al.,
1988). Future prospective studies should measure these antibodies in C. jejuni
patients before they develop GBS. The presence of these antibodies before nerve
damage was evident would strongly enhance the plausibility of C. jejuni-induced
GBS.

The plausibility of a causal relationship between C. jejuni and GBS would be
further enhanced if the putative host-environment interaction were better
understood. The major histocompatibility complex (MHC), or human leukocyte
antigen (HLA), has been implicated in the susceptibility to disease and to the
development of autoimmunity because HLA molecules function as antigen-
presenting structures. GBS is a heterogenous syndrome characterized by multiple
etiologies and pathologies and it has therefore been difficult for investigators to
identify a single HLA type associated with the syndrome (Ma et al., 1998; Rees et
al., 1995). Studies that have been restricted to or have divided the GBS group into
those with common antecedent infections or common outcomes (AIDP, AMAN)
have more often found associations (Monos et al., 1997; Koga et al., 1998)

From an epidemiologic standpoint, it is not clear what kind of variable HLA
type may be in the causal pathway between C. jejuni and GBS. It may function as
an antecedent variable that causally precedes the C. jejuni - GBS association.
This type of variable does not affect the strength of the association.

Alternatively, HLA type may be a marker for an immune response to C. jejuni
infection, in which case it would play no role in the causal chain. HLA may also

be a modifying variable causing a more severe type of GBS when in combination

38

with Campylobacterjejuni. Further studies are needed to elucidate the role of
HLA as a third variable in the causal pathway.
Analogy

This criterion is satisfied when the proposed causal relationship is comparable
to some other established causal relationship. Two diseases and a few animal
models have been proposed to satisfy this criterion.

In humans, acute disseminated encephalomyelitis (ADEM) and acute
hemorrhagic leukoencephalitis (AHLE) are other post-infectious demyelinative
diseases, but the demyelination occurs in the central nervous system rather than
the peripheral nervous system. These diseases represent an immune-mediated
complication often related to Mycoplasma pneumoniae infection, which parallels
the hypothesized mechanism for C. jejuni induced GBS.

A disease of dogs, acute polyradiculoneuritis, or Coonhound Paralysis, is
similar to GBS both clinically and pathologically. It usually occurs shortly after a
raccoon bite, but has also occurred after rabies vaccination. The condition has
been reproduced experimentally by injection of raccoon saliva into a dog that had
recovered from two earlier spontaneous attacks.

Experimental allergic neuritis (EAN) in the Lewis rat is considered the in vivo
model of GBS. EAN is quite different from the AMAN form of GBS.
Unfortunately, EAN cannot be induced in the rat by C. jejuni or gangliosides
(Vriesendorp et al., 1997). It can, however, provide some insight into the process

of immune-mediated peripheral nerve demyelination.

39

Based on the examples from other human diseases and animal models, it is
apparent that the criterion of analogy has been satisfied.
Conclusion

Whether or not these criteria should be used to determine if a relationship can
be elevated from association to causation is a debate that is beyond the scope of
this paper. These criteria can be used, however, to look at an association in a new
way and to reveal the gaps in the current understanding of that association. We
conclude that the criteria of temporality, consistency, and plausibility have not yet
been satisfied, nor can they be with retrospective epidemiological studies. A
prospective study or animal model is required to determine when in the course of
the disease antiganglioside antibodies are developed, the role of HLA, what C.
jejuni strains are associated with the disease, and the environmental source of
these strains.

If, indeed, C. jejuni does cause GBS, then public health measures aimed at
reducing the incidence of C. jejuni exposure could decrease the number of GBS

cases by almost one third.

40

References

Ahmed S, Libman R, Wesson K, Ahmed F, Einberg K. Guillain-Barre Syndrome: An
unusual presentation of West Nile virus infection. Neurology 2000; 55: 144-146.

Albert LJ and Inman RD. Molecular Mimicry and Autoimmunity. N Engl J Med 1999;
341(27): 2068-2074.

Allos, Ban Mishu. Association between Campylobacter Infection and Guillain-Barré
Syndrome. J Infect Dis 1997; 176 Suppl 2: 8125-128.

Allos BM, Lippy FT, C arlsen A, Washburn RG, Blaser MJ. Campylobacter jejuni
strains from patients with Guillain-Barré syndrome. Emerg Infect Dis 1998; 4(2):
263-268.

Aspinall GO, Fujimoto S, McDonald AG, Pang H, Kurjanczyk LA, Penner JL.
Lipopolysaccharides from Campylobacter jejuni associated with Guillain-Barré

Syndrome Patients Mimic Human Gangliosides in Structure. Infect Immun 1994;
62(5): 2122-2125.

Blaser MJ, Duncan DJ. Human serum antibody response to Campylobacter jejuni
infection as measured in an enzyme-linked immunosorbent assay. Infect Immun 1984;
44(2): 292-298.

Blaser MJ, Allos BM, Lang D. Development of Guillain-Barré Syndrome following
Campylobacter infection. J Infect Dis, 1997; 176 Suppl 2: S92.

Bossi P, Caumes E, Paris L, Darde M, Bricaire F. Toxoplasma gandii-associated
Guillain-Barré syndrome in an immunocompetent patient. J Clin Microbiol 1998; 36(12):
3724-3 725.

Buzby JC, Roberts T. Guillain—Barre' Syndrome Increases F oodborne Disease Costs.
Food Review 1997: 36-42

Buzby JC, Allos BM, Roberts T. The economic burden of Campylobacter-Associated
Guillain-Barre' syndrome. J Infect Dis 1997; 176 Suppl 2: 8192-8197.

CDC. Preliminary F oodNet Data on the Incidence of Foodbome Illnesses ---Selected
Sites, United States, 2000. MMWR 2001; 50: 241-246.

Chiba S, Sugiyama T, Matsumoto H, et a1. Antibodies against Helicabacter pylori Were
Detected in the Cerebrospinal Fluid Obtained from Patients with Guillain-Barre
Syndrome. Ann Neural 1998; 44: 686-688.

41

Enders U, Karch H, Toyka K, et al. The spectrum of immune responses to Campylobacter
jejuni and glycoconjugates in Guillain-Barré syndrome and in other neuroimmunological
disorders. Ann Neural 1993; 34(2): 136-144.

F easby TE, Hughes RAC. Campylobacter jejuni, antiganglioside antibodies, and
Guillain-Barré syndrome. Neurology 1998; 51: 340-342.

Goddard EA, Lastovica AJ, Argent AC . Campylobacter 0:41 isolation in Guillain—Barré
syndrome. Arch Dis Child 1997; 76: 526-8.

Green DM. Rapper AH. Mild Guillain-Barré Syndrome. Arch Neural 2001; 58: 1098-
1101.

Gregson NA, Koblar S, Hughes RAC. Antibodies to gangliosides in Guillain-Barré
syndrome: specificity and relationship to clinical features. QJM 1993; 86: 111-117.

Hadden RDM, Karch H. Hartung HP, et al. Preceding infections. immune factors, and
outcome in Guillain-Barré syndrome. Neurology 2001; 56: 758-765.

Hill AB. The environment and disease: association or causation? Proceedings, R Soc
Med, 1965; 58: 295-300.

Ho TW, Mishu B, Li CY, et al. Guillain-Barre’ syndrome in Northern China: Relationship
to Campylobacter jejuni infection and anti-glycolipid antibodies. Brain 1995; 118: 597-
605.

Hughes RAC, Rees J H. Clinical and Epidemiologic Features of Guillain-Barré
Syndrome. J Infect Dis 1997; 176 Suppl 2: 892-98.

Jacobs BC, van Doom PA, Schmitz PIM, et al. Campylobacterjejuni infections and anti-
GMI antibodies in Guillain—Barré syndrome. Ann Neurol 1996; 40: 181-187.

Jacobs BC, Rothbarth PH, van der Meche FGA, Herbrink P, Schmitz PIM, de Klerk MA,
van Doom PA. The spectrum of antecedent infections in Guillain-Barré syndrome: a
case-control study. Neurology 1998; 51: 1 110-11 15.

Kaldor J, Pritchard H, Serpell A, Metcalf W. Serum antibodies in Campylobacter
enteritis. J Clin Microbial 1983; 18(1): 1-4.

Kaldor J, Speed BR. Guillain-Barré syndrome and Campylobacter jejuni: a serological
study. BMJ 1984; 288: 1867-1870.

Koga M, Yuki N, Kashiwase K, Tadokoro K, Juji T, Hirata K. Guillain-Barre and
Fisher’s syndromes subsequent to Campylobacter jej uni enteritis are associated with

HLA-BS4 and CW] independent of anti-ganglioside antibodies. J Neuroimmunol 1998;
88: 62-66.

42

Kuroki S, Haruta T, Yoshioka M, Kobayashi Y, Nukina M, Nakanishi H. Guillain-Barré
syndrome associated with Campylobacter infection. Pediatr Infect Dis J 1991; 10(2):
149-151.

Kuroki S, Saida T, Nukina M, Haruta T, Yoshika M, Kobayashi Y, et al. Campylobacter
jejuni strains from patients with Guillain-Barré syndrome belong mostly to Penner

serogroup 19 and contain fi-N-Acetylglucosamine Residues. Ann Neural I993; 33: 243-
247.

Lastovica AJ, Goddard EA, Argent AC. Guillain-Barré syndrome in South Africa
associated with Campylobacter jejuni O:4l strains. J Infect Dis 1997; 176 Suppl 2: 8139-
S143.

Lugaresi A, Ragno M, Torrieri F, Di Gugliemo G, Fermani P, Uncini A. Acute motor
axonal neuropathy with high titer IgG and IgA anti-GDla antibodies following
Campylobacter enteritis. J Neurol Sci 1997; 147: 193-200.

Ma JJ, Nishimura M, Mine H, et al. HLA and T-cell receptor gene polymorphisms in
Guillain-Barré syndrome. Neurology 1998; 51: 379-384.

McCarthy N, Giesecke J. Incidence of Guillain-Barré Syndrome following infection with
Campylobacter jejuni. Am J Epidemiol 2001; 153 (6): 610-614.

McKhann GM, Comblath DR, Griffin J W, et al. Acute Motor Axonal Neuropathy: A
Frequent Cause of Acute Flaccid Paralysis in China. Ann Neural 1993; 33: 333-342.

Mishu B, Ilyas AA, Koski Cl, Vriesendorp F, Cook SD, Mithen FA, Blaser MJ. Serologic
evidence of previous Campylobacter jejuni infection in patients with the Guillain-Barré
syndrome. Ann Intern Med 1993; 118(12): 947-953.

Molnar GK, Mertsola J, Erkko M. Guillain—Barré syndrome associated with
campylobacter infection. BMJ 1982; 285: 652.

Monos DS, Papaioakim M, Ho TW, Li CY, McKhann GM. Differential Distribution of
HLA Alleles in Two F arms of Guillain-Barre' Syndrome. J Infect Dis 1997; 176 (Suppl
2): 8180-2.

Moran AP, Prendergast MM, Appelmelk BJ. Molecular mimicry of host structures by
bacterial lipopolysaccharides and its contribution to disease. FEMS Immunol Med

Microbiol 1996; 16: 105-115.

Mori M, Kuwabara S, Miyake M, et al. Haemophilus influenzae infection and Guillain-
Barré syndrome. Brain 2000; 123: 2171-2178.

Nachamkin I. Microbiologic Approaches for Studying Campylobacter species in patients
with Guillain-Barré syndrome. J Infect Dis 1997; 176 Suppl 2: S106-14.

43

Nevo Y, Pestronk A. Acute Immune Polyneuropathies: Correlations of Serum Antibodies
to Campylobacter jej uni and Helicobacter pylori with Anti-GM] Antibodies and Clinical
Patterns of Disease. J Infect Dis 1997; 176 Suppl 2: $154-$156.

Ono S, Chida K, Takasu T. Guillain-Barré syndrome following fulminant hepatitis A.
Internal Medicine 1994: 33: 799-801.

Rees JH, Soudain SE. Gregson NA, Hughes RAC. Campylobacter jejuni infection and
Guillain-Barré syndrome. N Engl J Med 1995; 333: 1374-1379.

Rees JH, Vaughan RW. Kondeatis E, Hughes RAC. HLA-class II alleles in Guillain-
Barre syndrome and Miller Fisher syndrome and their association with preceding
Campylobacter jejuni infection. J Neuroimmunol 1995; 62: 53-57.

Rhodes KM, Tattersfield AE. Guillain-Barré syndrome associated with Campylobacter
infection. BMJ1982; 285: 173-174.

Richardson, RF , Remler BF, Katirji B, Hated Muram M. Guillain-Barre Syndrome after
Cyclospora Infection. Muscle Nerve 1998; 21: 669-671.

Schwartz M, Sela BA, Eshhar N. Antibodies to Gangliosides and Myelin Autoantigens
Produced in Mice Following Sciatic Nerve Injury. J Neurochem 1982; 38: 1192-1195.

Sovilla J, Regli F, Francioli P. Guillain-Barré syndrome following Campylobacter jejuni
enteritis. Arch Intern Med 1988; 148: 739-741

Speed BR, Kaldor J, Watson J, et al. Campylobacterjejuni / Campylobacter coli-
associated Guillain-Barre syndrome: Immunoblot confirmation of the serological
response. MedJAust 1987; 147: 13-16.

Toemer JG. Kumar PN. Garagusi VF. Guillain-Barré Syndrome Associated with Rocky
Mountain Spotted Fever: Case Report and Review. C [in Infect Dis 1996; 22: 1090-1091.

Van Koningsveld R, Van Doom PA, Schmitz PIM, Ang CW, Van der Meche' FGA. Mild
Forms of Guillain-Barre’ syndrome in and epidemiologic survey in the Netherlands.
Neurology 2000; 54: 620-625.

Vriesendorp FJ, Triggs WJ, Mayer RF, Koski CL. Electrophysiological studies in
Guillain—Barré syndrome: correlation with antibodies to GMI, Gle and Campylobacter
jejuni. J Neurol 1995; 242: 460-465.

Vriesendorp FJ. Insights into Campylobacter jejuni-Induced Guillain-Bare Syndrome
from the Lewis Rat Model of Experimental Allergic Neuritis. J Infect Dis 1997; 176
(Suppl 2): 8164-168.

44

Walsh FS, Cronin M. Koblar S, Doherty P, Winer J, Leon A, Hughes RAC. Association
between glycoconjugate antibodies and Campylobacter infection in patients with
Guillain-Barré syndrome. J Neuroimmunol 1991; 34: 43-51.

Weber F, Rudel R, Aulkemeyer P, Brinkmeier H. Anti-GMI antibodies can block
neuronal voltage-gated sodium channels. Muscle Nerve 2000: 23: 1414-1420.

Willison HJ, O’Hanlon GM. Antiglycosphingolipid antibodies and Guillain-Barre
Syndrome. In: Nachamkin I, Blaser MJ editors. Campylobacter. 2nd ed. Washington,
DC: ASM Press; 2000: pg 261.

Winer JB, Hughes RAC. Anderson MJ, Jones DM, Kangro H, Watkins RPF. A
prospective study of acute idiopathic neuropathy. II. Antecedent events. J Neurol
Neurosurg Psychiatry 1988; 51: 613-618.

Winer JB, Briggs D, Welsh K, Hughes RAC. HLA antigens in the Guillain-Barré
syndrome. J Neuroimmunol 1988; 18: 13-16.

Wirguin I, Suturkova-Milosevic L, Della-Latta P, Fisher T, Brown RH, Latov N.
Monoclonal IgM Antibodies to GMI and Asialo-GMI in Chronic Neuropathies Cross-
React with Campylobacter jejuni Lipopolysaccharides. Ann Neurol 1994; 35: 698-703.

Young DM, Biao J, Zheng Z, Hadler J, Edberg SC. Isolation of C ampylobacter jejuni in
Hunan, China: Epidemiology and comparison of Chinese and American methodology.
Diagn Microbial Infect Dis 1986; 5(2): 143-149.

Yuki N, Yoshino H, Sato S, Miyatake T. Acute Axonal Neuropathy associated with high
titer IgG and IgA anti-GDIa antibodies following Campylobacter enteritis. Neurology
1990; 40: 1900-1902.

Yuki N. Molecular mimicry between gangliosides and lipopolysaccharides of
Campylobacter jejuni isolated from patients with Guillain-Barré syndrome and Miller
Fisher Syndrome. J Infect Dis 1997; 176 Suppl 2: 8150-8153.

Yuki N, Ho TW, Tagawa Y, Koga M, Li CY, Hirata K, et al. Autoantibodies to GMlb

and GalNAc-GDI a: relationship to Campylobacter jej uni infection and acute motor
axonal neuropathy in China. J Neural Sci 1999; 164: 134-138.

45

CHAPTER 3

CAMPYLOBACTER JEJUNI ENTERITIS IN MICHIGAN 1992-1999:
A COMPARISON OF INCIDENCE RATES IN AREAS OF HIGH AND LOW
POULTRY DENSITY
Abstract
Objectives. To compare the incidence of Campylobacterjejuni enteritis in high and low
poultry density counties in Michigan between the years 1992 and 1999.
Methods. An ecological study was conducted in the state of Michigan to evaluate the
incidence of C ampylobacter jejuni during an eight—year period in regions of high and low
poultry density. A loglinear model was fit with Poisson regression methods to determine
the association between incidence rate and poultry density controlling for age, gender,
season, and year.
Findings. The overall incidence of C. jejuni enteritis was higher in counties with a high
poultry density than in those without, particularly among children and young adults.
Conclusions. The findings suggest that high poultry density is associated with high rates
of C. jejuni enteritis. Excess cases may be due to elevated risk among poultry workers or
manure management and transport practices. Future studies should be conducted to
address these issues.
Introduction
Previous descriptive studies have compared the incidence rates for C. jejuni enteritis

between urban, rural, and farming communities. In general, these studies have found that
rates are higher in rural areas (Sibbald and Sharp, 1985), particularly rural farm areas
(Thompson et al. 1986). Case-control studies have revealed that living on or visiting a

farm (Friedman. 2000). having contact with farm animals (Friedman, 2000; Engleberg,

46

1984), and poultry contact (Studahl and Andersson, 2000; Potter, et al. submitted 2002)
are major risk factors in rural areas. Rural is usually defined by htunan population
density. If exposure to animals the cause of higher rates in rural areas, then rates should
be higher in areas with high animal density than in those with low animal density. We
hypothesized that the odds of developing C. jejuni enteritis would be higher in counties
with high poultry density than in those with low density. Poultry were selected because
very high isolation rates have been obtained from chickens. Using data collected from
the National Agricultural Statistics Service (NASS), the counties of the state of Michigan
were considered either high or low poultry density. Because there is a large variation in
poultry density between counties, an ecological study design was considered.
Methods
Study Design

Ecological studies may be conducted when a group-level variable is of interest, when
data on individuals is not available, or when convenience and cost are a concern.
Ecological studies are most appropriate, however, when the role of a group property that
cannot be measured at the individual level is being investigated. In this study, poultry
density per county, a proxy for both occupational exposure to C. jejuni and the
environmental burden of the organism, was a group-level variable hypothesized to be the
antecedent cause of disease and not measurable in individuals.
Source of Data

Incidence rates for Campylobacter enteritis were obtained from cases reported to the
Michigan Department of Community Health (MDCH) for the eight-year period between

1992 and 1999 and population counts from the 1990 census. Incidence rate data were

47

then categorized into strata for the covariates age (19 strata), gender (2 strata), year (8
strata), and season (4 strata) for each region It was assumed that the age distribution of
the population remained constant throughout the eight-year period and that disease
occurrence was independent.

Poultry population estimates were obtained from the National Agricultural Statistics
Service (NASS), which conducts a complete count of livestock inventory every five
years. Additional data. collected yearly, are based on sample surveys. Data are not
always published for all counties. If a county has less than 500 head or if birds are
concentrated in only one operation within a county, data are not published for that county
but provided as a summary number for the district. For its poultry count, Michigan only
collects data on layers 20 weeks of age and over. Counties with less than one percent of
the layers in the state were grouped into the low-density region. High-density counties
had between one percent and 27% of the yearly inventory. In all, the high-density region
had between 71% and 97% of the layers in the state between 1992 and 1999. Because of
the way data is reported counties in which more than one percent of the state total
chicken inventory was concentrated in one operation were misclassified as unexposed.
This misclassification may bias results to the null because of the mixing of the exposure
effect among the unexposed. High and low density counties by year are shown in table 1.
Statistical Analyses

Estimates of the incidence rate were calculated by dividing the munber of new cases by
population counts from the 1990 census.

A loglinear model was fit with Poisson regression methods to estimate incidence rate

ratios (IRR). The model assumes independence and homogeneity (Flanders and

48

Kleinbaum, 1995). Independence was assumed because person-to-person transmission of
C. jejuni is rare. If one person has C. jejuni enteritis, it does not change the risk of others
in the group of developing C. jejuni enteritis. Homogeneity means that disease risks are
the same across people and time. We assumed that the dependent variable, the case
count, followed a Poisson distribution. Independent variables included in the model were
five-year age group, gender, season, year, and poultry density. Epi Info, Microsoft Excel
and SAS Version 8 were used for data entry and analysis.
Results

The risk (estimated by the rate ratio) for C. jejuni enteritis was 1.31 (95% CI 1.04,
1.42) times higher in the high-density region than the low-density region after controlling
for age, gender, season, and year. The incidence rates for each region with 95%
confidence intervals (95% C1) are 12.01 cases/ 100,000 population (95% CI; 11.09,
12.98) in high poultry density counties and 8.60 cases / 100,000 population (95% CI;
8.38, 8.82) in low-density counties (Table 2). Counties with a high poultry density had a
significantly higher incidence every year during the eight-year study period except 1997
and 1998. In unexposed counties and the state as a whole, the incidence of
Campylobacter enteritis has steadily decreased between 1992 and 1999. In counties with
a large poultry inventory, however, the rate was lowest in 1997 but rose again in 1998
and 1999 (Figure 1).

Age-specific and gender-specific rates varied by region. Rates were significantly
higher for one to nine year-olds, 25 to 29 year-olds, 35 to 39 year-olds and 45 to 49 year-
olds in counties with high poultry density compared to the referent (Figure 2). Males

were at higher risk in the state as a whole and in the referent region, but there was no

49

significant difference in gender-specific rates in the region with high poultry inventory
(Figure 3).

The high poultry prevalence region had the highest rate every season of the year, but
the rate ratio was largest in winter. Incidence rates were 1.36 times higher (95% CI; 1.18,
1.79) in winter in counties with a large poultry inventory than those with a low inventory
(Figure 4).

Discussion

Ecological studies are necessary for describing group-level properties but are
criticized for not being applicable to individuals. Group-level variables do, however,
cause disease (Schwartz. 1994). We hypothesized that poultry density was a variable
particular to a group, much like “contagion” (Susser, 1994). A high prevalence of
Campylobacter-excreting chickens in a county may increase the risk of disease for
everyone in that county whether they have direct contact with poultry or not because C.

jejuni may be highly prevalent in the environment and in other animal species. It may
modify the effect of individual-level risk factors.

There may be at least two mechanisms functioning to increase C. jejuni enteritis rates
in the high-density region. First, poultry workers may be responsible for the excess
cases. Second, poultry may be shedding Campylobacters into the environment causing
other animals, including sheep, swine, cattle, dogs, and cats, to become colonized. This
would increase the risk for carnpylobacteriosis associated not only with direct contact
with these animals but also with drinking unpasteurized milk and untreated water. In this

case, the immediate cause of carnpylobacteriosis may be contact with a dog, cow, cat or

50

consumption of untreated water or raw milk, but the ultimate cause is the large
environmental pool of Campylobacters in the area.

Exposure of poultry workers to C. jejuni has been established. Campylobacterjejuni
organisms were isolated from the hands of processing-line poultry workers in a study in
The Netherlands (Oosterom et al., 1983) and antibody to C. jejuni was higher in poultry
processing workers than in antenatal patients in Great Britain (Jones and Robinson,
1981). A study of chicken abattoir workers in Sweden found evidence of higher mean
lgG antibody levels in long term workers than short term workers and blood donors
(Cawthraw et al., 2000). Evidence for a direct relationship between antibody titers and
length of employment was suggested from the data, but it was not statistically significant.
Interestingly, fecal cultures were positive for eight workers in the study, but only one, a
short-term worker, had symptoms of enteritis (Cawthraw et al., 2000). These findings
and others (Blaser et al., 1987) suggest that the low antibody titers are probably
protective against symptomatic C. jejuni enteritis. Infection may occur among new
employees. This has been supported by a protective effect for occupational contact with
livestock in a case-control study (Adak et al., 1995). In contrast, daily contact with
poultry has been associated with increased odds for enteritis (Studahl and Andersson,
2000), as has poultry husbandry (Potter et al. submitted, 2002) in case-control studies.
The increased risk among young adults in high-density areas would suggest that workers
are responsible for at least some of the excess cases. The three poultry processing plants
in Michigan are not located in high poultry density counties. They are located in Kent,
Barry, and Livingston Counties. Kent and Barry counties are adjacent counties to the

east of the high poultry density counties of Allegan and Ottawa. This may contribute to

51

the non-differential misclassification of poultry density in the study. Additional study
should be undertaken to elucidate the epidemiology of C. jejuni infection in farm and
abattoir workers.

Warm-blooded animals are good reservoirs for C. jejuni because they rapidly become
colonized by the organism but do not develop illness. Animals may acquire the infection
in one of several ways. They may become infected by direct contact with other members
of their flock or herd who are excreting the organism. F omites, such as boots, may
transmit the organism from infected animals to non-infected animals. Water can carry
the organism to susceptible animals (Kapperud et al., 1993, Stanley et al., 1998).
Vectors, such as flies (Khalil et al., 1994) or wild birds (Riordan et al., 1993, Craven et
al., 2000) may also introduce the organism into a flock or herd. Companion animals may
also carry the organism (Gondrosen et al., 1985) so they might also act as vectors,
moving the organism between farm animal species and humans.

These findings may also be due to biases. If there are differences in reporting in the
high and low density region or if there is a high percentage of uninsured in the low
density counties, for example in the urban centers, then the incidence rate differences we
found may be due to these biases. Future studies should investigate the potential
influence of these biases on the study results. One method would be to compare the
distribution of another reportable disease, not associated with chicken exposure, in the
high and low poultry density counties.

Additional work is needed to discover the source of the excess cases in the high-
density region. The measurement of chicken density by county is a crude way to

determine exposure but we were limited by the data that were available. Future studies

may look at poultry per person per area as a better measure of exposure. Because only
data on layers per county was available, we did not measure exposure in this way.
Additionally, the exposure to chickens within counties may vary considerably. For
example, Monroe County was considered high density but is an urban county near
Detroit, the largest city in Michigan, and has a very large population. However, the
poultry houses must have been in a sparsely populated area within that county. The
epidemiology of C. jejuni infection in poultry workers should be investigated to
understand if they are experiencing higher or lower than expected rates of infection.
Investigations of the way manure is managed and its impact on the environment may also
reveal the cause of the excess cases. A comparison of C. jejuni colonization in farm,
wild, and companion animals in the two regions would also be of interest. Changes in
manure management practices or hygiene in poultry workers may be needed to decrease
C. jejuni enteritis in the high poultry density counties. These interventions should be

implemented at the ecological level.

53

Table 3-1: High and low density counties by year 1992 - 1999

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L922 1.913. fi & % Bfl 12% &
Ottawa Ottawa Ottawa Ottawa Ottawa Ottawa Ottawa Ottawa
Allegan Allegan Allegan Allegan Allegan Allegan Allegan Allegan
Ionia Ionia Ionia

Huron Huron Huron Huron Huron Huron Huron Huron
K-zoo* K-zoo K-zoo K-zoo
Isabella Isabella Isabella Isabella

Tuscola Tuscola Tuscola Tuscola Tuscola Tuscola Tuscola Tuscola
VanBuren VanBuren VanBuren

Newaygo Newaygo Newaygo Newaygo Newaygo

Hillsdale Hillsdale Hillsdale Hillsdale Hillsdale Hillsdale Hillsdale

Gratiot Gratiot

Monroe Monroe Monroe

 

 

K-zoo* = Kalamazoo county

Table 3-2: Distribution of cases between areas of high and low poultry density 1992

 

 

 

 

— 1999.
Region Number of cases Population Incidence Rate per
100,000 population
High Density 639 665,297 12.01
Low Density 5935 8,630,000 8.37
Total 6574 9,295,297 8.63

 

 

 

 

 

54

 

 

Figure 3-1: Yearly incidence rates in high and low poultry density regions:
Michigan 1992-1999

 

 

 

 

 

 

    

 

 

‘_IR high 3.3,. IR low +IR total ‘
16
14 .-
12 -
1o .-
t! 8
6
4 ,
2
o . ,
1992 1993 1994 1995 1996 1997 1998 1999
year

 

 

 

IR" = incidence rate / 100,000 population

Figure 3-2: Gender-specific incidence rates in high and low poultry density regions:
Michigan 1992-1999

 

 

‘Efemale .malJ

 

IR"

 

 

 

high low state
poultry density

 

 

 

IR" = incidence rate / 100,000 population

55

Figure 3-3: Age-specific incidence rates in high and low poultry density regions:
Michigan 1992-1999

 

 

 

 

 

 

 

 

451 — — —
40, L L L LL L L L L
35.“ LL L L L L L L L L
30.3%,, -LLLL--- LL_-_L_
. an + --+ ~ + ~—
1 1 ‘ *’ z ”a “mm-f“
15.. L -L -L
1o.LL \ L J.“ \ fig! L 'r I-) {-
5: _ x .7

 

 

 

Dr V" firV '* T— #i‘m‘mfi“ V1’V1fi'

318188339155888

81881893818818

1
5
10-14
15-19
70-74
79 .

399M

 

 

 

IR" = incidence rate / 100,000 population

56

Figure 3-4: Seasonal incidence rates in high and low poultry density regions:
Michigan 1992-1999

 

ithii Iow

 

 

 

 

fall summer spn'ng winter

 

 

 

IR‘ = Incidence rate / 100,000 population

57

References

Adak GK, Cowden J M. Nicholas 8, Evans HS. The Public Health Laboratory Service
national case-control study of primary indigenous sporadic cases of campylobacter
infection. Epidemiol Infect 1995; 115: 15-22.

Blaser MJ, Sazie E, Williams P. The Influence of Immunity on Raw Milk-Associated
Campylobacter infection. JAIIM 1987; 257: 43—46.

Cawthraw SA, Lind L, Kaij ser B, Newell DG. Antibodies, directed towards
Campylobacterjejuni antigens, in sera from poultry abattoir workers. C lin Exp Immunol
2000; 122: 55-60.

Craven, S.E., N.J. Stern, E. Line, J .S. Bailey, N.A. Cox, and P. Fedorka-Cray.
Determination of the Incidence of Salmonella spp., Campylobacter jejuni, and

Clostridium perfringens in Wild Birds near Broiler Chicken Houses by Sampling
Intestinal Droppings. Avian Dis 2000; 44: 715-20.

Engleberg NC, Correa-Villasenor A, North CQ, Crow T, Wells JG, and Blake PA.
Campylobacter Enteritis on Hopi and Navajo Indian Reservations. West J Med 1984;
141(1): 53-56.

Flanders WD and Kleinbaum DG. Basic Models for Disease Occurrence in
Epidemiology. IntJ Epidemiol 1995; 24(1): 1-7.

Friedman C, Reddy S, Samual M, Marcus R, Bender J, Desai S, Shiferaw B, Helfrick D,
Carter M, Anderson B, Hoekstra M, and the EIP Working Group. Risk Factors for
Sporadic Campylobacter Infections in the United States: A Case-Control study on

FoodNet Sites. 2nd International Conference on Emerging Infectious Diseases. Atlanta,
GA, July 2000.

Gondrosen, B., T. Knaevelsrud, and K. Dommarsnes. Isolation of thermophilic
Campylobacters from Norwegian dogs and cats. Acta vet. Scand 1985. 26: 81-90.

Jones DM, Robinson DA. Occupational Exposure to Campylobacter jejuni infection.
Lancet 1981; i : 440-441.

Kapperud G., E. Skjerve, L. Bik, K. Hauge, A. Lysaker, I. Aalmen, S.M. Ostroff, and M.
Potter. Epidemiological investigation of risk factors for campylobater colonization in
Norwegian broiler flocks. Epidemiol Infect 1993; 111: 245-255.

Khalil, K., G.-B. Lindblom, K. Mazhar, and B. Kaijser. Flies and water as reservoirs for

bacterial enteropathogens in urban and rural areas in and around Lahore, Pakistan.
Epidemiol Infect 1994; 113: 435-444.

58

Oosterom J, Notermans S., Karman H, Engels G.B. Origin and Prevalence of
Campylobacter jejuni in Poultry Processing. J Food Prat 1983; 46 (4): 339-344.

Potter RC, Kannene J B, Hall W. Risk Factors for Campylobacterjejuni Infections in
Rural Michigan: A Prospective Case-Control Study. Submitted to American Journal of
Public Health, 2002.

Riordan, T., T.J. Humphrey, and A. Fowles. A point source outbreak of campylobacter
infection related to bird-peeked milk. Epidemiol Infect 1993; 110: 261-265.

Schwartz, Sharon. The Fallacy of the Ecological Fallacy: The Potential Misuse of a
Concept and the Consequences. Am J Public Health 1994; 84(5): 819-824.

Sibbald CJ and Sharp J CM. Campylobacter infection in urban and rural populations in
Scotland. Campylobacter infection in urban and rural populations in Scotland J Hyg,
Camb 1985;95: 87-93

Stanley, K., R. Cunningham, and K. Jones. Isolation of Campylobacterjejuni from
groundwater. J Appl Microbiol 1998; 85: 187-191.

Studahl A and Andersson Y. Risk factors for indigenous campylobacter infection: a
Swedish case-control study. Epidemiol Infect 2000; 125: 269-275.

Susser, M. The Logic in Ecological: I. The Logic of Analysis. Am J Public Health 1994;
84(5): 825-829.

Thompson J S, Cahoon FE, Hodge DS. Rate of Campylobacter spp. Isolation in Three
Regions of Ontario, Canada from 1978 to 1985. J Clin Microbial 1986; 24(5): 876-878.

59

CHAPTER 4

DESCRIPTIVE EPIDEMIOLOGY OF GUILLAIN-BARRE SYNDROME IN
MICHIGAN: 1992 - 1999

Introduction

Guillain-Barre’ Syndrome (GBS) is an acute inflammatory demyelinating
polyneuropathy that is usually post-infectious. Campylobacterjejuni is the most common
cause of bacterial gastroenteritis in the United States (CDC, 2001). A previous
ecological study of the distribution of reported C. jejuni enteritis cases in the state of
Michigan between 1992 and 1999 revealed significantly elevated incidence rates in
counties with high poultry density (Potter, et al. submitted 2002). A recent C. jejuni
infection has been estimated to increase the risk of developing GBS approximately 100-
fold (McCarthy and Giesecke, 2001). We hypothesized that because C. jejuni is strongly
associated with GBS (Enders et al., 1993, Mishu et al.. 1993, Rees et al., 1995), the
distribution of GBS cases would follow that of C. jejuni cases. Reported cases were used
to test this hypothesis
Methods
Study Design

Ecological studies may be conducted when a group-level variable is of interest, when
data on individuals is not available, or when convenience and cost are a concern.
Ecological studies are most appropriate, however, when the role of a group property that
cannot be measured at the individual level is being investigated. In this study, an area of

high C. jejuni incidence was compared with a low incidence region. These regions have

60

been described (Potter et al., submitted 2002). The level of contagion in a community is
a group level variable.
Sources of Data

Reported cases of GBS were collected the Michigan Department of Community
Health (MDCH). Population denominator data were obtained from 1990 census data.
Statistical A nalyses

The number of new cases per 1,000,000 population was used to estimate incidence
rates. Exact 95% Poisson confidence intervals were calculated around the point estimate
of the rate.

Results

Overall, 479 cases of GBS were reported to MDCH between 1992 and 1999. The
records for eight of these cases were incomplete, so they were dropped, leaving 471 cases
for analysis. This gives a crude annual incidence rate of 6.3 / 1,000,000 population. This
rate is lower than has been reported in other descriptive studies (Hughes and Rees, 1997),
and may indicate under-reporting.

Men were affected more often than women. The overall incidence rate for men was
7.4 cases per 1,000,000 population (95% Confidence Interval; 6.6, 8.4) and 5.3 cases per
1,000,000 population (95% CI; 4.6, 6.1) for women. There was no significant difference
in the incidence rates of GBS in high and low Campylobacterjejuni incidence regions.

In the high incidence region the crude rate was 5.8/1,000,000 population (95% CI: 3.9,
8.3) compared to 6.4/1.000,000 population (95% CI: 5.8, 7.0) in the low incidence

region.

61

For all years and both sexes, the age-specific incidence rate showed a bimodal

distribution with a low peak in the 10-14 year old age group and a higher peak in the 70-

74 year old group (Table 4-1).

Table 4-1: Age-specific incidence rates for GBS in Michigan: 1992-1999

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age Cases Population Incidence Rate
(1992 — 1999) / 1,000,000
population
< 1 1 129255 1.0
1-4 17 573299 3.7
5-9 17 692247 3.1
10-14 29 666370 5.4
15-19 24 696803 4.3
20-24 23 705318 4.1
25-29 19 764262 3.1
30-34 27 810291 4.2
35-39 31 749062 5.2
40-44 24 657087 4.6
45-49 24 523730 5.7
50-54 24 424389 7.1
55-59 44 392787 14.0
60-64 39 401936 12.1
65-69 44 36911 1 14.9
70-74 44 286727 19.2
75-79 23 212494 13.5
80-84 14 133222 13.1
85 and over 3 106907 3.5

 

 

The incidence of GBS decreased between 1992 and 1999. In 1992 the incidence rate
per 1,000,00 population was 8.8. It decreased to 5.5 in 1992 (Figure 4-1).

There were no significant differences in the age-specific incidence rates between the
two regions, but point estimates were unusually elevated in 5-14 year-olds and 75-79 year
olds in the high C. jejuni incidence region (Figure 4-2). These estimates are unstable due

to low numbers, however.

62

Similarly, the twelve months of the year were divided into four three-month seasons.
The highest rate occurred in winter (November, December, and January) but the
difference was not significant.

Discussion

The age, gender, and seasonal distributions of GBS in Michigan are similar to what has
been found in other developed countries. Men are affected more often than females, most
cases occur in the elderly, and there is no change in incidence by season. The
epidemiology of Campylobacterjejuni enteritis is only similar to that of GBS in its
gender distribution. Possible explanations for this lie in the multiple causes of GBS.
Many infections can precede GBS and their relative preponderance in different seasons of
the year, e. g., Campylobacter in summer and flu in winter, may contribute to the lack of
seasonality. Host factors, such as a diminished ability to distinguish self from non-self
may explain why most cases of GBS occur in the elderly, while its most common
antecedent C. jejuni usually occurs in infants and young adults, but the reasons are not
clear. Reported cases declined between 1992 and 1999, but the reasons for this decline
are not known. The reported incidence of C. jejuni also declined during this period
(Potter et al., submitted 2002). The incidence of GBS did not vary between the
previously identified high and low C. jejuni incidence regions. Indeed, the point estimate
of the incidence rate was higher in counties with a lower C. jejuni incidence rate. This
may be due to the multiple causality of GBS. Other infectious diseases antecedent to
GBS may be more common in the urban areas within the low C. jejuni incidence region.
Campylobacterjejuni serotype and human leukocyte antigen (HLA) type of the host are

also thought to play an important role in the pathogenesis of C. jejuni-associated GBS.

63

We could not reject the null hypothesis of no difference in GBS rates between high
and low C. jejuni incidence regions. A more robust method of data analysis may have
been able to detect small differences in rates between regions, while controlling for
potentially confounding variables but the data was too sparse for both Poisson regression
models and zero-inflated Poisson regression models. If, in the future, validated hospital

discharge data are available, these models may be used for the analysis.

64

Figure 4-1: Crude annual incidence rates for GBS in Michigan 1992-1999

 

 

 

f I +cases p—ef 17.06.111.000I

cases I 1 ,000,000

o-aruwa-mmxroocoo

 

 

 

1992 1993 1994 1995 1996 1997 1998 1999
year

 

 

 

Figure 4-2: Age-specific incidence rates for GBS in high and low C. jejuni enteritis
regions in Michigan 1992-1999

 

 

 

 

 

 

 

 

60
50' l“.I-Iigh C.jejuni counties 1 * a m w:
8 4o iAELowCJejunicounties‘l g“ _
3'
q 30. a - .--L-.u--. -. 3‘. a- - . - i _l
1-
m
a 20 LLLLLLLLLL - L,
m ..
8 m L-LLLLLLLL E1 ,
a I E i f I
o 3 1' it: t» I
V3$§$§$§$§$9$5$§g$3
omomomomomomomo"
*- v- N N or) to <1- <1- 10 IO co (0 rx 1» co A
agegroup

 

 

65

 

References

CDC. Preliminary F oodNet Data on the Incidence of Foodbome Illnesses ---Selected
Sites, United States, 2000. MMWR 2001; 50: 241-246.

Enders U, Karch H, Toyka KV, Michels M, Zielasek J, Pette M, et al. The Spectrum of
Immune Responses to C ampylobacter jejuni and Glycoconjugates in Guillain-Barre’

Syndrome and in Other Neuroimmunological Disorders. Ann Neural 1993; 34(2): 136-
144.

Jacobs BC, van Doom PA, Schmitz PIM, Tio-Gillen AP, Herbrink P, Visser LH, et al.
Campylobacterjejuni Infections and Anti-GMI Antibodies in Guillain-Barré Syndrome.
Ann Neurol 1996; 40: 181-7.

McCarthy N and Giesecke J. Incidence of Guillain-Barré Syndrome following Infection
with Campylobacter jejuni. Am J Epidemiol 2001; 153(6): 610-614.

Mishu B, Ilyas AA, Koski CL, Vriesendorp F, Cook SD, Mithen FA, et al. Serologic
Evidence of Previous Campylobacter jejuni Infection in Patients with Guillain-Barré
Syndrome. Ann Intern Med 1993; 118: 947-953.

Potter RC, Kaneene J BK, Gardiner J. Campylobacter jej uni enteritis in Michigan 1992-
1999: A comparison of incidence rates in areas of high and low poultry density. Vector-
Bome and Zoonotic Diseases. Submitted 2002.

Rees JH, Gregson NA, and Hughes RAC. Anti-Ganglioside GM. Antibodies in Guillain-

Barre' Syndrome and Their Relationship to Campylobacterjejuni Infection. Ann Neurol
1995; 38: 809-16.

66

CHAPTER 5

RISK FACTORS FOR CAMPYLOBAC T ER JEJUNI INFECTIONS IN RURAL
MICHIGAN: A PROSPECTIVE CASE-CONTROL STUDY
Abstract
Objectives. To investigate the risk factors for carnpylobacteriosis in a rural population.
Exposure to live farm animals was hypothesized to increase the risk for Campylobacter
jejuni enteritis.
Methods. In this prospective case-control study, incident cases from rural counties
reported to the Michigan Department of Community Health and matched controls were
asked to complete a self-administered postal questionnaire.
Results. Persons engaged in activities related to the care and raising of poultry,
husbandry, had an increased odds of campylobacteriosis, odds Ratio (OR) 6.884, 95%
confidence interval (C I) 1.438, 32.954. There was evidence for dose effect between the
number of types of poultry contact and carnpylobacteriosis.
Conclusions. In rural populations, contact with live poultry is a significant risk factor for
campylobacteriosis.
Introduction
Campylobacterjejuni is the most common cause of bacterial gastroenteritis in the
United States (CDC, 2001). Including undiagnosed and unreported cases it is estimated
to affect over two million people every year. The annual cost of carnpylobacteriosis has
been estimated to be between $1.3 to $6.2 billion dollars (Buzby et al., 1997). This cost

increases when sequelae, such as Guillain-Barré Syndrome and reactive arthritis are

67

considered. Of additional concern, Campylobacterjejuni and C. coli are becoming
increasingly resistant to some antimicrobials (Engberg et al., 2001).

Risk factors for outbreak cases and sporadic cases differ. Outbreaks are typically due
to raw milk (Komblatt et al., 1985; Harris et al., 1987; Evans et al., 1996) or
contaminated water consumption (St. Louis, 1988; CDC, 1999) but the vast majority of
cases are sporadic. Identified risk factors for sporadic cases include consumption of
undercooked chicken (Hopkins et al., 1984; Deming et al., 1987; Friedman et al., 2000),
contact with pets, especially puppies and kittens (Adak et al., 1995; Neal and Slack,
1997), and contact with diarrheic animals (Adak et al., 1995; Saeed et al., 1993). Some
ecological descriptive studies have shown that rates of infection are higher in rural areas
than urban areas and that, among rural areas, farming regions have the highest rates
(Thompson et al., 1986). Living on or visiting a farm has been shown to increase the
odds of infection (Friendman etal., 2000), but a multicenter study in England and Wales
showed a decreased odds associated with occupational contact with livestock or their
feces (Adak et al., 1995). A study of carnpylobacteriosis on Hapi and Navajo Indian
reservations, rural areas, showed an increased risk with ownership of farm animals
(Engleberg et al., 1984).

We conducted a study to determine the risk factors for C. jejuni enteritis in rural
communities. We hypothesized that exposure to food animals is a major risk factor and

that the odds of infection changes with exposure to different species.

68

Methods
Design

A prospective matched case-control study design was implemented. The study
duration was one year, from October 2000 to October 2001. Incident cases reported to
the Michigan Department of Community Health (MDCH) were identified weekly. Cases
were contacted and invited to participate if they met the following inclusion criteria.
First, the cases must have been residents of a rural county. Rural counties were defined
as those having a population less than 70,000 by the 1990 census and not adjacent to a
major metropolitan center. Fifty-eight of Michigan’s 84 counties met this criterion.
Second, the cases must have been reported to MDCH within 30 days of the onset of
symptoms. This was to avoid potential problems with recall bias. Third, the cases must
not have been part of an identified outbreak and fourth, must have been the first in a
household. Only first cases in a household were considered because the effect of person
to person transmission (though uncommon) was not of interest. If these criteria were
satisfied, the case was contacted by phone about the study and if they expressed an
interest in participating. a postal questionnaire was sent. Two controls were matched to
each case by age group, gender, and county of residence. These variables were chosen
for matching to eliminate them as potential confounders. Age groups were divided into
those less than one year old, 1 to 2 year-olds, 3 to 4 year-olds, 5 to 12 year-olds, 13 to 19
year-olds, then by 20 year age groups, and finally those 60 years and over. After a case
was contacted by phone. two controls were identified by random digit dialing (RDD)

within the same area code and three-digit prefix as the case. If the potential control was

69

of the same age group and gender as the case and had not had any gastrointestinal
symptoms during the two weeks prior to contact, they were sent a postal questionnaire.
Questionnaire

The questionnaires sent to cases and controls were identical except that where cases
were asked about behaviors two weeks prior to onset of symptoms, controls were asked
about behaviors two weeks prior to contact by the investigators. After an informed
consent form was signed, information was collected on demographic characteristics,
foreign travel, antibiotic and antacid use, animal contact and husbandry behaviors, and
food consumption habits. If the case or control was a child, the parent or guardian was
asked to complete the questionnaire. A case or control was considered to engage in
animal husbandry if that person participated in feeding, cleaning. or raising an animal for
milk, eggs, or meat, or had housed an animal in their home.
Data Analysis

Microsoft Access and Excel were used to enter and organize data. Statistical analysis
was performed using SAS version 8. A conditional logistic regression model was used
for tmivariable and multivariable analyses. Independent variables associated with the
outcome (p<0.15) were then tested for association using a chi-square test. If two or more
independent variables were significantly associated with each other (p<0.05) the more
biologically important variable was included in the model and the other discarded.
Variables that showed an association at the p<0.15 level were then entered into a
conditional logistic regression model for the multivariable analysis. The final model was
obtained through hierarchical backward elimination of variables using statistical and

epidemiological criteria for assessment of interaction and confounding. Interaction terms

70

were considered significant if the p-value for the term was less than 0.05 by the Wald
test. Confounding was assessed by the impact of the potential confounder on the
parameter estimate for the main effects, i.e. poultry husbandry. If removal of a
confounding variable caused a 10% or greater change in the value of the parameter
estimate, that variable was considered a confounder and left in the model.
Modeling

The questionnaire asked about contact with a particular species of animal and then
specific questions about the type of contact. These questions included feeding, cleaning,
raising an animal for meat, eggs, or milk, keeping an animal in the home or garage, and
having clothing contaminated with fecal material. (Table 5-1) When analyzed separately,
these variables were highly associated with each other, as expected. Raising animals for
eggs, milk, or meat almost always involves feeding, cleaning, and fecal contact. As a
result, these variables were combined into one dichotomous summary variable,
husbandry, for each species. Cases and controls were considered to be positive for the
husbandry exposure if they indicated exposure to any of the independent variables
described above. This modeling of independent variables is biologically plausible
because husbandry, or the care and raising of livestock, is a special kind of contact. It
involves repeated, at least daily, direct contact with the animals including contact with
the fecal material of species that have the potential to carry Campylobacter.
Results

Of the 193 C. jejuni enteritis cases reported to the MDCH from rural counties during
the year of our study, 48 did not meet the inclusion criteria described above, leaving 143

cases. F arty-three of the 48 who did not meet the inclusion criteria did not do so because

71

they were reported more than 30 days after the onset of their illness. Twenty-one of the
cases reported too late to participate were from very rural counties with a population of
35,000 or less by 1990 census counts. We were not able to contact 31 of the eligible
cases. There was contact with 110 cases, among which there were six refusals and 21
who initially agreed to participate in the study but later changed their minds, were not
sure they could remember the period prior to their illness, or did not return their postal
questionnaire. Using the method described by Slattery, et al., we calculated a cooperation
rate of 75% (Slattery et al., 1995). The cooperation rate is the percentage of people
interviewed of those who were contacted. All cases and controls were contacted two
weeks after the questionnaire had been sent to ensure that it had been received and to
answer any questions that had arisen. If the case or control could not be contacted by
phone, a second copy of the questionnaire was sent. Questionnaires were completed for
83 cases and 122 controls. We calculated a response rate of 58% for cases. Response rate
is defined as the percentage of people interviewed of those who were selected for and
eligible for study. Similarly, the response rate for controls was calculated at 48%. There
was no significant difference in age, county of origin, or week of report between those
cases included in the study and the baseline available rural case population of 193 cases
(Wilcoxon rank sum test). Males were under-represented in our study group, however
(Chi Square p=0.0237).
Univariable Analyses

Contact with any food producing animal (bovines, swine, or poultry) was significantly
associated with illness (OR, 4.722; 95% CI, 1.737, 12.833). Of factors considered

significant in the univariable analysis (p < 0.15), there were increased odds of illness if

72

there was contact with adult domestic poultry (OR, 3.216; 95% CI, 0.811, 12.763) and
participation in the care and raising of poultry (OR, 8.454; 95% CI, 1.877, 38.081). The
care and raising of cattle was also associated with illness (OR, 3.058; 95% CI, 0.907,
10.307) as was the care and raising of swine (OR, 7.358; 95% CI, 0.845, 64.079) and
horses (OR 3.380; 95% CI, 0.860, 13.294). Contact with foals also conferred increased
odds for infection (OR. 6.275; 95% CI, 0.689, 57.165).

We inquired about consumption and preparation of poultry, ground beef, and pork. Of
these, only the consumption of undercooked pork and poultry were significant in the
univariable analysis, but their effect was protective (OR, 0.333; 95% CI, 0.110, 1.013 and
OR, 0.180, 95% CI, 0.052, 0.622, respectively). Raw milk was not associated with
illness. Eleven percent of controls and ten percent of cases reported raw milk
consumption in the two weeks prior to contact or illness.

Of the other exposures, including foreign travel, living on a farm, taking antibiotics or
antacids, and having problems with rodents or houseflies in the home, only living on a
farm was associated with illness (OR, 2.484; 95% CI, 1.041 , 5.930). (Table 5-1)

We considered the interactions of antacid use, antibiotic use, and poultry exposure
with the other significant animal contact and food consumption terms. None were
statistically significant.

Assessment of associations between independent variables

Equine husbandry and exposure to foals were strongly associated with poultry
husbandry and bovine husbandry. Because chickens and cows are known to be important
reservoirs of C. jejuni, while horses are not (Prescott and Bruin-Mosch, 1981) the horse

exposure variables were dropped from the model. Similarly, farm exposure was very

73

highly associated with animal husbandry but of the two, the specific animal exposure is
more biologically important. Farm exposure was, therefore, dropped from the model.
Farm exposure was not as strongly associated with outcome as the animal husbandry
variables.
Multivariable model

Factors significant in the univariable model were entered into the multivariable model.
These included: poultry husbandry; cattle husbandry; swine husbandry; consuming
poultry that was pink at the center, had red juices running from the meat, or was raw; and
consuming park that was pink at the center, had red juices running fiom the meat, or was
raw. The consumption of undercooked poultry and pork were not statistically significant
nor was there any evidence of confounding, so they were removed from the model. In
the final model, only poultry husbandry (OR, 6.884; 95% CI, 1.438, 32.954) was
associated with C. jejuni enteritis. Swine and cattle husbandry showed increased risk but
it was not significant over and above poultry husbandry (Table 5-2).
Dose-Eflect

Because we found a strong association with poultry, swine, and bovine husbandry
exposures, we further investigated these variables to look for a dose-effect relationship.
From the seven questions asked about husbandry exposure, three categories were created
and modeled as an ordinal variable. The lowest husbandry category was one or two
exposures, the middle category was three to five exposures, and the highest was six or
seven. A dose-response relationship was observed for poultry and bovine husbandry, but
no cases had six or seven of the exposure variables for these species. Because the

variable was modeled as an ordinal one, the odds for the second and third category were

74

calculated from the parameter estimate for the first category. The results are shown in
Figure 5-1.
Analysis of Non-Responders

The response rate for cases was 58%. We compared the age distribution, week of
onset, and gender of responders and non-responders. There was no significant difference
between the two groups. Gastrointestinal illness case investigation reports were available
for 17 of the 28 non-responders. These reports are the results of a telephone interview
between a public health nurse at the local health department and the case. Of the 17
reports, two were blank and one had declined to give information, leaving 14 for analysis.
Two of the 14 reported food-animal contact. One case had contact with poultry and one
had contact with sheep. There was no significant difference between responders and non-
responders in contact with farm animals (Fisher’s exact test, p = 0.1280). Non-
responders were less likely to have contact with companion animals, however, than were
responders (Fishers exact test, p = 0.0014). These findings suggest that there was not a
response bias on food animals, but we may have been under-powered to find such a bias
if it does exist. The response rate for controls was 48%. There was no significant
difference in age or gender for those controls who responded and those who did not.
Data were not collected on farm or companion animal exposure for controls that did not
respond.
Discussion

We found that contact with farm animals was a significant risk factor for C. jejuni
enteritis in rural areas. Specifically, the care and raising of poultry increased the risk for

disease by seven-fold over and above husbandry for other species known to be reservoirs

75

for Campylobacterjejuni. Additionally, we found evidence for a dose-effect relationship
for increased odds of infection with increasing kinds of food animal contact.

Ecological studies have indicated that the incidence rate for C. jejuni infection is
higher in rural areas than urban areas (Brieseman, 1990), especially farming communities
(Thomopson et al., 1986). Raw milk consumption (Thompson et al., 1986; Schmid et al.,
1987) contact with farm animals (Engleberg et al., 1984), and daily contact with chickens
or hens (Studahl and Andersson, 2000) have been identified as potential risk factors in
rural populations. This study confirms the findings of previous research implicating farm
animal contact and further defines that contact as activities related to the care and rearing
of poultry.

We did not find that raw milk consumption was a risk factor in our study, however.
This may be due to over-matching on location or because raw milk consumption was a
more common practice among controls than anticipated in power calculations. In a
survey of milk producers, 35% reported drinking raw milk (Rohrbach et al., 1992).
Additionally, regular raw milk consumption has been shown to cause an elevated anti-C.

jejuni antibody titer that protects against symptomatic infection (Blaser et al., 1987).

We did not find any association with cat or kitten exposure. This is in contradiction
with other studies that have found exposure to cats and kittens to be significantly
positively associated with illness (Hopkins et al., 1984; Deming et al., 1987). This may
be explained by the fact that the previous studies took place in urban environments and
that, although cats are fairly ubiquitous on farms, their presence is discouraged around

the chicken coop and there may be little actual contact with them.

76

This study may have been under-powered to find other significant associations with
livestock. Many of the cases from very rural counties were excluded because they were
not reported within 30 days of onset of disease. A study with a two-year duration or
active case ascertainment from local health departments may enroll enough cases to find
these associations. Additionally, our questionnaire was long so a postal questionnaire
was used. This method has the disadvantage of a lower response rate. The findings of
this study, however, confirm previous investigations and provide a basis for additional
research.

Cases who responded to the questionnaire were more likely to have association with
companion animals than those who did not respond to the questionnaire. Although
questions were asked about a variety of known risk factors for C. jejuni, most of the
questions were on animal contact. It is possible that cases who felt the questionnaire did
not apply to them, i.e.. they had no animal contact, did not return it. We did not collect
the data to measure this bias in controls. A large number of controls had contact with
pets (88%), and this finding may indicate that controls were also motivated to return their
questionnaire if they had companion animal contact.

This study illustrates that, in rural areas, the care and raising of farm animals,
particularly poultry, confers an increased risk for C. jejuni enteritis. Public health

measures aimed at prevention should be population-specific.

77

Table 5-1: Univariable analyses of covariates included in the summary husbandry
variables.

 

 

 

Variable No. Exposed No. Unexposed m OR* 95% CIt
wCases Controls Cases Controls

Poultry housed in 5 2 78 120 3.812 0.726, 20.014

home or garage

Poultry hatched 4 2 79 120 5.767 0.621, 53.603

on property

Poultry raised for 7 2 76 120 7.922 0.930, 67.472

meat

Poultry raised for 13 4 70 1 18 10.902 1.339, 88.744

eggs

Feeding poultry 9 8 74 114 2.169 0.619, 7.592

Poultry associated 6 3 77 1 19 3.412 0.672, 17.339

cleaning

Clothing 6 4 77 118 2.234 0.516, 9.678

contaminated with

poultry feces

Calves born on 1 1 82 121 0 0

property

Bovines housed in 0 0 83 122 0 0

home or garage

Bovines kept for l 0 82 122 0 0

milk

Bovines kept for 6 2 77 120 2.637 0.434, 16.013

beef

Feeding bovines 6 3 77 1 19 2.189 0.513, 9.342

Bovine associated 7 1 76 121 8.421 0.998, 71.041

cleaning

Clothing 7 4 76 118 2.610 0.613, 11.119

contaminated with

bovine feces

Pigs housed in 0 0 83 122 0 0

home or garage

Pigs born on 3 0 80 122 0 0

property

Pig kept as pet 2 0 81 122 0 0

Pigs raised for l 0 82 122 0 0

meat

Feeding Pigs 2 1 81 121 2.561 0.225, 29.120

Pig associated 1 0 82 122 0 0

cleaning

Clothing 4 1 79 121 6.275 0.689, 57.165

contaminated with

pig feces

 

*mOR = matched odds ratio

1C1 = confidence interval

78

Table 5-2: Matched odds ratios for animal contact, demographic characteristics,
and food consumption habits for cases of C. jejuni enteritis: exposures significant in
univariable analysis (p < 0.15)

 

 

 

 

 

Exposure No. exposed NoLunexposed mOR* 95% CH
Cases Controls Cases Controls
Adult 9 4 74 118 3.216 0.811, 12.763
Poultry
Poultry 18 8 65 114 8.454 1.877, 38.081
Husbandry
Bovine 12 5 71 117 3.058 0.907, 10.307
Husbandry
Swine 5 1 78 121 7.358 0.845, 64.079
Husbandry
Equine 9 5 74 117 3.380 0.860, 13.294
Husbandry
Foal 4 2 79 120 6.275 0.689, 57.165
Farm 21 12 62 110 2.484 1.041, 5.930
Undercooke 6 26 77 96 0.180 0.052, 0.622
d Poultry
Undercooke 4 18 79 103 0.333 0.110, 1.013
d Pork

 

‘mOR = matched Odds ratio 'ICI = Confidence interval

79

Table 5-3: Final multivariable model

 

 

Exposure mOR* 95% Cl'l p Value

Poultry 6.884 1.438, 32.954 0.0158
Husbandry

Bovine 2.447 0.657. 9.1 14 0.1822
Husbandry

Swine 2.149 0.178, 25.995 0.5477
Husbandry

 

*mOR = matched Odds ratio

fCI = Confidence interval

80

Figure 5-1: Demonstration of a linear dose-response in number of types of contact
with poultry or bovines and odds ratios

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

*mOR = matched odds ratio

81

References

Adak GK, Cowden JM. Nicholas S, Evans HS. The Public Health Laboratory Service
national case-control study of primary indigenous sporadic cases of campylobacter
infection. Epidemiol Infect 1995; 115: 15-22.

Brieseman MA. A further study of the epidemiology of Campylobacterjejuni infections.
N Z Med J 1990: 103: 207-209.

Buzby JC, Allos BM, Roberts T. The Economic Burden of Campylobacter-Associated
Guillain-Barre Syndrome. J Infect Dis 1997; 176 (Suppl 2); Sl92-7.

CDC. Outbreak of Escherichia coli 0157:H7 and Campylobacter Among Attendees of
the Washington County Fair - New York, 1999. MMWR 1999; 48(36): 803-805

CDC. Preliminary F oodNet Data on the Incidence of F oodbome Illnesses ---Se1ected
Sites, United States, 2000. MMWR 2001; 50: 241-246.

Deming MS, Tauxe RV, Blake PA, Dixon SE, Fowler BS, Jones S, Lockamy EA, Patton
CM, Sikes RO. Campylobacter Enteritis at a University: Transmission from Eating
Chicken and from Cats. Am J Epidemiol 1987; 126(3): 526-534.

Engberg J, Aarestrup FM, Taylor DE, Gemer-Smidt P, Nachamkin I. Quinolone and
Macrolide Resistance in Campylobacter jejuni and C. coli: Resistance Mechanisms and
Trends in Human Isolates. Emerg Infect Dis 2001; 7(1): 24-34

Engleberg NC, Correa-Villasenor A, North CQ, Crow T, Wells JG, Blake PA.
Campylobacter Enteritis on Hopi and Navajo Indian Reservations: clinical and
epidemiologic features. WestJ Med 1984; 141: 53-56.

Evans MR, Roberts RJ, Ribeiro CD, Gardiner D, Kembrey D. A milk-bome
campylobacter outbreak following and educational farm visit. Epidemiol Infect 1996;
117: 457-462.

Friedman C, Reddy S, Samual M, Marcus R, Bender J, Desai S, Shiferaw B, Helfrick D,
Carter M, Anderson B, Hoekstra M, and the EIP Working Group. Risk Factors for
Sporadic Campylobacter Infections in the United States: A Case-Control Study on

FoodNet Sites. 2nd International Conference on Emerging Infectious Diseases, 2000.
Atlanta, GA, July 16-19, 2000.

Harris NV, Kimball TJ. Bennett P, Johnson Y, Wakely D, Nolan CM. Campylobacter
jejuni Enteritis Associated with Raw Goat’s Milk. Am J Epidemiol 1987; 126(2): 179-
186.

Hopkins, RS, Olmsed R, Istre GR. Endemic Campylobacterjejuni Infection in Colorado:
Identified Risk Factors. Am J Public Health 1984; 74(3): 249-250.

82

Komblatt AN, Barrett T. Morris GK, Tosh FE. Epidemiologic and Laboratory

Investigation of an Outbreak of Campylobacter Enteritis Associated with Raw Milk. Am
JEpidemiol 1985; 122(5): 884-889.

Neal KR, Slack RCB. Diabetes mellitus, anti-secretory drugs and other risk factors for
campylobacter gastro-enteritis in adults: a case-control study. Epidemiol Infect 1997;
119: 307-311.

Prescott JF, Bruin-Mosch CW. Carriage of Campylobacter jejuni in Healthy and
Diarrheic Animals. Am J Vet Res 1981; 42(1): 164-5.

Rohrbach, B.W., Draughon F .A., Davidson P.M., Oliver S.P. Prevalence of Listeria
monocytogenes, Campylobacter jejuni, Yersinia enterocolitica, and Salmonella in Bulk
Tank Milk: Risk Factors and Risk of Human Exposure. J F oad Prat 1992: 55; 93-97.
Blaser MJ, Sazie E, Williams P. The Influence of Immunity on Raw Milk-Associated
Campylobacter Infection. JAMA 1987; 257: 43-46.

Saeed AM, Harris NV, DiGiacomo RF. The Role of Exposure to Animals in the Etiology
of Campylobacterjejuni / coli Enteritis. Am J Epidemiol 1993; 137(1): 108-114.

Schmid GP, Schaefer RE, Plikaytis BD, Schaefer JR, Bryner JH, Wintermeyer LA, et al.
A one-year study of endemic carnpylobacteriosis in a midwestem city: association with

comsumption of raw milk. J Infect Dis 1987; 156: 218-222.

Slattery ML, Edwards SL1, Caan BJ, Kerber RA, Potter JD. Response Rates among
Control Subjects in Case-Control Studies. Ann Epidemiol 1995; 5: 245-249.

St. Louis, Michael E. Water-Related Disease Outbreaks, 1985. MMWR 1988; 37 (SS-2):
15-24.

Studahl, A. and Andersson Y. Risk factors for indigenous campylobacter infection: a
Swedish case-control study. Epidemiol Infect 2000; 125: 269-275.

Thompson J S, Cahoon FE, Hodge DS. Rate of Campylobacter spp. Isolation in Three
Regions of Ontario, Canada, from 1978 to 1985. J Clin Microbial 1986; 24(5): 876-878.

83

SUMMARY AND CONCLUSIONS

In the first chapter. the current knowledge on risk factors for C. jejuni enteritis was
reviewed. C ampylobacter jejuni enteritis is known to be more common in rural
communities than urban communities, but the reasons for this are not clear. Some studies
had suggested that raw milk or untreated water consumption and animal contact may be
the causes. Additional information on species of animal, age of animal, and type of
animal contact was needed to create appropriate prevention strategies. The case-control
study in Chapter 5 was designed to address this gap in the current understanding of the
epidemiology of C. jejuni in rural areas.

The second chapter reviewed the association between C. jejuni and GBS in the
framework of the causal criteria. Through this review it became clear that future studies
on the association should be prospective. A prospective or cohort study is necessary to
satisfy the criterion of time order (the sine qua non of causality) and to elucidate the
pathophysiological events that lead from C. jejuni infection to GBS. It was also noted
that mild cases of GBS may be more common than previously thought and that these
cases have not been investigated because they may not go to the hospital and may not be
reported. For these reasons, therefore, GBS cases reported to the MDCH will need to be
supplemented in order to have usage in studying the relationship between GBS and C.
jejuni.

The third chapter was an ecological study of C. jejuni in high and low poultry density
regions of Michigan between 1992 and 1999. It was found that the incidence of C. jejuni
enteritis was significantly higher in high poultry density counties than in counties with

less than one percent of the poultry in the state. Children and young adults had higher

84

rates in the high-density region than the low-density region. The reasons for these
findings are not known. but several hypotheses were generated. Transport or manure
management practices may be causing environmental contamination, which is a risk
factor unto itself. and may also cause other species of animals in the region to become
carriers and excreters of the organism. Workers in the poultry industry may account for
the increased incidence. Additional studies to investigate manure management practices,
environmental contamination with C. jejuni, and disease in poultry workers in the high
poultry density region are recommended.

In the fourth chapter. the rates of GBS in the high and low C. jejuni incidence regions
identified in chapter three were compared. Data on GBS reported to the MDCH were too
sparse to conduct any meaningful analytical evaluation so descriptive methods were used.
We found no significant difference in the rates between the two regions. The limited data
that were available to us from the hospital discharge database indicated that GBS is
probably under-reported in Michigan. Because it was not possible to verify the diagnoses
in this database we were not able to quantify the magnitude of under-reporting. This
further underlines the fact that MDCH reported cases will need to be supplemented with
additional information in future studies.

The case-control study in chapter five showed that poultry husbandry was a significant
risk factor in rural areas. Additionally, we showed an increasing number of types of
poultry husbandry increased the odds of C. jejuni enteritis. This information can be used
to educate poultry producers and hobby farmers in rural areas and prevent future cases. It
also indicates that poultry workers may be the cause of excess cases in the high-poultry

density region identified in Chapter 3.

85

APPENDIX

86

Campylobacterjejuni Consent Form

Campylobacterjejuni is a type of bacteria that can cause illness characterized by
diarrhea, fever, nausea, and abdominal pain. Most people recover from the illness in a
few days, but a very small number (less than 1%) will go onto develop severe secondary
diseases such as arthritis or paralysis shortly after their illness. Because of the severity of
this disease and the potential for complications, all persons in Michigan who are
diagnosed with Campylobacterjejuni must be reported to the Michigan Department of
Community Health by the physician or laboratory that made the diagnosis. Your name
has been identified from this list of reported cases. We hope that you will join us in a
research study of people who have had Campylobacterjejuni infections.

The purpose of this study is to investigate the events that occur prior to infection with the
goal of determining why some people get sick and others do not. This research is
important because Campylobacter infections are becoming more common in the United
States and in Michigan. Your participation in this study will help us to learn more about
this disease. Dr. John B. Kaneene and Dr. Rachel C. Potter of Michigan State University,
East Lansing, Michigan are conducting this study in cooperation with the Michigan
Department of Community Health.

As a member of the study you are asked only to participate in the completion of a
detailed questionnaire. This questionnaire, which is included, asks for information on
your health history, occupational history, food preparation habits. and other demographic
characteristics. The questionnaire takes approximately 15 to 20 minutes to complete. A
self-addressed, stamped envelope has been provided for you to return the questionnaire.

All information collected will be kept confidential and your privacy will be protected to
the maximum extent allowable by law. As a study participant, you will not be identified
by name on any of the information you provide, but will be assigned a study number. At
no time will individuals be identified in the study results or summaries. Information

identified by your study number will be kept separate from your name in a secure, locked
file.

If you have any particular questions about the study, your participation, or if, after having
volunteered, you wish to discontinue with the study, you may call Dr. John B. Kaneene at
(517) 355-2269 or Dr. Rachel C. Potter at (517) 355-1745. If you have any questions or
concerns regarding your rights as a study participant, or are dissatisfied at any time with
any aspect of this study, you may contact — anonymously if you wish — Ashir Kumar,
Chair of the University Committee on Research Involving Human Subjects (UCRIHS) by
phone: (517) 432-2222, fax: (517) 353-2976, e-mail: ucrihstiiimsuedu, or regular mail: 246
Administration Bldg, East Lansing, MI 48824.

 

To take part in the study, you need to sign the form on the following page and complete
the questionnaire. If you choose not to take part in this study, or if you withdraw after
you have started, you will not be penalized in any way.

87

I have had an opportunity to ask questions about the study and was given sufficient time
to consider my decision to participate. I agree to be in the study.

Name (please print)

 

Signature Date

 

 

If the participant is under the age of 18 years, the signature of a parent or guardian is
required.

Name (please print)

 

Signature Date

 

 

Once you have read and signed this consent form, please place it in the postage paid
envelope and mail it to us with the questionnaire. We have enclosed a copy of the
consent form for you to keep.

Thanks!

88

CAMPYLOBA C TER JEJUNI QUESTIONNAIRE

ALL OF THE FOLLOWING QUESTIONS PERTAIN TO YOUR DIARRHEAL
ILLNESS THAT OCCURRED AROUND

 

Travel History: The following questions pertain only to travel in the two
weeks before your diarrheal illness:

1. Did you travel outside the continental United States and Canada? (circle
only one response)

Yes No

2. If you answered yes to question 1., where did you travel?

 

 

Occupational History: The following question pertains only to your job in
the two weeks before your diarrheal illness:

3. Briefly describe your job:

 

89

Residence: The following questions pertain to the area where you lived
during the two weeks prior to your illness:

4. Did you live on a farm?

Yes No

5. Did you consider houseflies to be a problem at that time?

Yes No

6. Did you have a problem with rodents in your home at that time?

Yes No

Medications: The following questions pertain to any medications you were
taking during the two weeks prior to your illness:

7. Were you taking any antibiotics?
Yes No
8. Were you taking any antacids?

Yes No

90

History of Contact with Animals: the following questions pertain only to
any contact that you had with live animals two weeks before your illness.

9. In the two weeks before you became ill did you have any exposure to
chickens, ducks, geese, turkeys, or other domestic poultry (not wild
ducks, geese, or other wild poultry or pet birds) of any age?

Yes No

10.If you answered yes to question 9, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:

1

2

10

11

direct physical contact with adult poultry

direct physical contact with young poultry or chicks
no physical contact but poultry are on your property
poultry have been inside your house or garage
poultry were hatched on your property

poultry are raised for meat on your property

poultry are kept for eggs on your property

you fed poultry

you cleaned-up after poultry

clothing contaminated with fresh poultry
droppings/manure

other, specify

 

91

11.In the two weeks before your illness did you have M exposure to
cows or cattle of any age?

Yes No

12.If you answered yes to question 11, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:

1

2

10

11

direct physical contact with adult cattle or cow(s)
direct physical contact with young cattle or calves

no physical contact but cattle or cow(s) are on your
property

cattle or cow(s) were born on your property

cattle or cow(s) have been inside your house or garage
(inc. calf)

cows are kept for milking on your property
cattle are raised for beef on your property
you fed cattle or cow(s)

you cleaned-up after cattle or cows(s)

clothing contaminated with fresh cattle or cow manure

Other, specify

 

92

13.1n the two weeks before you became ill did you have m exposure to
any pigs / swine of any age?

Yes No

14.If you answered yes to question 13, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:

1

2

10

11

direct physical contact with adult pig(s)

direct physical contact with young or juvenile pig(s)
no physical contact but pig(s) is on your property
pig or baby pig was in your house or garage

pi g(s) was born on your property

pig kept as a pet on your property

raised for meat on your property

you fed pig(s)

you cleaned-up after pig(s)

clothing contaminated with fresh pig manure

other, specify

 

93

15.Did you have a_ny exposure to horses of any age in the two weeks
before you became ill?

Yes No

16.If you answered yes to question 15, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:

1

2

direct physical contact with adult horse(s)

direct physical contact with young horses or foal(s)
no physical contact but horse(s) is on your property
horse(s) was born on your property

horse (or foal) was in your house or garage

you fed horse(s)

you cleaned-up after horse(s)

clothing contaminated with fresh horse manure

other, specify

 

94

17.Did you have m exposure to a dog(s) of any age in the two weeks
before you were ill?

Yes No

18.If you answered yes to question 17, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:

1

2

direct physical contact with an adult dog

direct physical contact with a juvenile dog or a puppy
no physical contact but dog(s) was on your property
a dog(s) was in you house or garage

a puppy(ies) was born on your property

you fed dog(s)

you cleaned-up after dog(s)

clothing contaminated with dog droppings

other, specify

 

95

19.Did you have any exposure to a cat(s) of any age in the two weeks
before you became ill?

Yes No

20.If you answered yes to question 19, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:

1

2

direct physical contact with adult cats

direct physical contact with kittens

no physical contact but cats are on your property
cats in house or garage

a kitten(s) was born on your property

you fed cat(s)

you cleaned-up after cat(s) or emptied a “cat box”
clothing contaminated with cat droppings

other, specify

 

96

21.Did you have any exposure to pet birds (not chicken, ducks, turkeys,
other poultry, or wild birds) of any age in the two weeks before you
became ill?

Yes No
22.If you answered yes to question 21, indicate the kind of exposure in
the two weeks before you became ill by circling each that applies:
1 Direct physical contact with an adult pet bird(s)
2 physical contact with young or juvenile pet bird(s)
3 no physical contact but a pet bird(s) is on your property
4 a pet bird(s) in house
5 a pet bird(s) was hatched on your property
6 you fed a pet bird(s)
7 you cleaned-up after a pet bird(s) or cleaned a bird cage
8 clothing contaminated with fresh pet bird droppings

9 other, specify

 

97

Food consumption habits before illness: The following questions pertain
only to your food consumption habits for the time period two weeks
before your diarrheal illness.

23. How did you usually eat ground beef (hamburger)? (circle only one
response)

y—d

Well-done (cooked throughout / no pink showing)

Medium (pink at the center) 2
Rare (red juices running from the meat) 3
Raw 4

24. Did you ever eat ground beef that was medium (pink at the center), rare
(red juices running from the meat) or raw? (circle only one response)

Yes No

25. How did you usually eat poultry (chicken or turkey)? (circle only one
response)

p—d

Well-done (cooked throughout / no pink showing)
Cooked, pink at the center or on the bone 2

Cooked, with red juices running 3

Raw 4

98

26. Did you consume any poultry (chicken or turkey) that was prepared
in the following ways? (circle all that apply)

Fried 1
Barbecued 2
Roasted 3
Baked 4

Food consumption habits (continued):
27. Did you ever eat poultry (chicken or turkey) that was cooked and pink at
the center or on the bone, or cooked with red juices running, or raw?

(circle only one response)

Yes No

28. How did you usually eat pork? (circle only one response)

Well-done (cooked throughout / no pink showing) 1

Cooked, pink at the center or on the bone 2
Cooked, with red juices running 3
Raw 4

99

29. Did you ever eat park that was cooked and pink at the center or on the
bone, or cooked with red juices running, or raw? (circle only one
response)

Yes No
30. Did you ever drink untreated water in the two weeks prior to illness
(water that was not chlorinated or boiled)? (circle only one response)
Yes No
31. Did you ever drink raw (unpasteurized) milk? (circle only one response)
Yes No

32. Are you the person in your household that usually prepares meals?

Yes No

100

REFERENCES

Abeyta C, Deeter FG, Kaysner CA, Stott RF, Wekell MM. Campylobacter jejuni in a
Washington State Shellfish Growing Bed Associated With Illness. J Food Prat 1993; 56:
323-325.

Achen M, Morishita TY, Ley EC. Shedding and Colonization of Campylobacter jejuni in
broilers from day-of-hatch to slaughter age. Avian Dis 1998; 42: 732-7.

Adak GK, Cowden JM, Nicholas S, Evans HS. The Public Health Laboratory Service
national case-control study or primary indigenous sporadic cases of campylobacter
infection. Epidemiol Infect, 1995; 115: 15-22.

Ahmed S, Libman R, Wesson K, Ahmed F, Einberg K. Guillain-Barre Syndrome: An
unusual presentation of West Nile virus infection. Neurology, 2000; 55: 144-146

Alary M and Nadeau D. An Outbreak of Campylobacter Enteritis Associated with a
Community Water Supply. Can J Public Health, 1990; 81: 268-271.

Albert LJ and Inman RD. Molecular Mimicry and Autoimmunity. N Engl J Med 1999;
341(27):2068-2074.

Allos, Ban Mishu. Association between Campylobacter Infection and Guillain-Barre’
Syndrome. J Infect Dis 1997; 176 (Suppl 2): 8125-128.

Allos BM, Lippy FT, Carlsen A, Washburn RG, Blaser MJ. Campylobacter jejuni
strains from patients with Guillain—Barré syndrome. Emerg Infect Dis, 1998; 4(2):
263-268.

Altekruse SF, Hunt JM, Tollefson LK, Madden J M. Food and animal sources of human
Campylobacterjejuni infection. J Am Vet Med Assoc 1994; 204(1); 57-61.

Aspinall GO, F ujimoto S, McDonald AG, Pang H, Kurjanczyk LA, Penner JL.
Lipopolysaccharides from Campylobacter jejuni associated with Guillain—Barré

Syndrome Patients Mimic Human Gangliosides in Structure. Infect Immun 1994;
62(5): 2122-2125.

Atabay HI and Carry J EL. The prevalence of campylobacters and arcobacters in broiler
chickens. J Appl Microbial 1997; 83: 619-26.

Baker J, Barton MD, Lanser J. Campylobacter spp. in Cats and Dogs in South Australia.
Aust VetJ 1999; 77: 662-666.

101

Blaser M, Cravens J, Powers BW, Wang WL. Campylobacter enteritis associated with
canine infection. Lancet 1978; 2(8097): 979-981.

Blaser MJ, Laforce M. Wilson NA, Wang WLL. Reservoirs for Human
Campylobacteriosis. J Infect Dis 1980; 141: 665-669.

Blaser MJ, Weiss SH, Barrett TJ. Campylobacter enteritis associated with a healthy cat.
JAMA 1982: 247; 816.

Blaser MJ, Duncan DJ. Human serum antibody response to Campylobacter jejuni

infection as measured in an enzyme-linked immunosorbent assay. Infect Immun 1984;
44(2): 292-298.

Blaser, MJ., Sazie E., Williams P. The Influence of Immunity on Raw Milk-Associated
Campylobacter Infection. JAMA 1987; 257: 43-46.

Blaser MJ, Allos BM, Lang D. Development of Guillain-Barre' Syndrome following
Campylobacter infection. J Infect Dis 1997; 176 (Suppl 2): S92

Blaser, Martin J. Epidemiologic and Clinical Features of Campylobacter jejuni
infections. J Infect Dis 1997; 176 (Suppl 2): 8103-5.

Bolton F J , Coates D, Hutchinson DN, Godfree AF. A study of thermophilic
campylobacters in a river system. Journal of Applied Bacteriology 1987; 62: 167-176.

Bossi P, Caumes E, Paris L, Darde M, Bricaire F. Toxoplasma gandii-associated

Guillain-Barre' syndrome in an immunocompetent patient. J Clin Microbiol 1998; 36(12):
3724-3725.

Bremell T, Bj elle A, Svedhem A. Rheumatic symptoms following an outbreak of
campylobacter enteritis: a five year follow up. Ann Rheum Dis 1991; 50: 934-938.

Brieseman MA. A further study of the epidemiology of Campylobacter jejuni infections.
NZ Med J 1990; 103: 207-209.

Bumens AP, Angeloz-Wick B, Nicolet J. Comparison of Campylobacter carriage rates in
diarrheic and healthy pet animals. J. Vet. Med. B 1992; 39: 175-80.

Busato A, Hofer D, Lentze T, Gaillard C, Bumens A. Prevalence and infection risks of
zoonotic enteropathogenic bacteria in Swiss cow-calf farms. Vet Microbial 1999; 69:

251-263.

Buzby JC, Allos BM, Roberts T. The Economic Burden of Campylobacter-associated
Guillain-Barre Syndrome. J Infect Dis1997; 176 (suppl 2): 8192-7.

1 02

Buzby JC, Roberts T. Guillain—Barré Syndrome Increases F oodbome Disease Costs.
Food Review, 1997: 36-42

Casemore D.P., E.G. Jessop, D. Douce, and F .8. Jackson. Cryptasporidium plus
campylobacter: an outbreak in a semi-rural population. Journal of Hygiene 1986; 96: 95-
105.

Cawthraw SA, Lind L, Kaij ser B, Newell DG. Antibodies, directed towards

Campylobacterjejuni antigens, in sera from poultry abattoir workers. Clin Exp Immunol
2000; 122: 55-60.

CDC. Outbreak of Campylobacter Enteritis Associated with Cross-Contamination of
Food — Oklahoma, 1996. MMWR 1998; 47(07): 129-131.

CDC. Outbreak of Escherichia coli 01 57:H7 and Campylobacter among attendees of the
Washington County Fair — New York, 1999. MMWR 1999; 48(36); 803-4.

CDC. Preliminary F oodNet Data on the Incidence of F oodbome Illnesses --- Selected
Sites, United States, 2000. MMWR 2001: 50(13): 241-6.

Chiba S, Sugiyama T, Matsumoto H, et al. Antibodies against Helicabacter pylori Were
Detected in the Cerebrospinal Fluid Obtained from Patients with Guillain-Barre
Syndrome. Ann Neural 1998; 44: 686-688.

Craven SE, Stern NJ, Line E, Bailey J S, Cox NA, Fedorka-Cray P. Determination of the
Incidence of Salmonella spp., Campylobacter jejuni, and Clostridium perfringens in Wild

Birds near Broiler Chicken Houses by Sampling Intestinal Droppings. Avian Dis 2000;
44: 715-20.

Davidson RJ, Sprung DW, Park CE, Rayman MK. Occurrence of Listeria
monocytogenes, Campylobacter spp. and Yersinia enterocolitica in Manitoba Raw Milk.
Canadian Institute of F aod Science and Technology Journal 1989; 22: 70-74.

Deming MS, Tauxe RV, Blake PA, Dixon SE, Fowler BS, Jones TS, et al.
Campylobacter enteritis at a university: transmission from eating chicken and from cats.
Am J Epidemiol 1987; 126(3): 526-534.

Desmasures, N., F. Bazin, and M. Gueguen. Microbiological composition of raw milk
from selected farms in the Camembert region of Normandy. J Appl Microbial 1997; 83:
53-58.

Dilworth CR, Lior H, Belliveau MA. Campylobacter Enteritis Acquired from Cattle.
Can J Public Health 1988; 79: 60-62.

Doyle MP and Roman DJ. Prevalence and Survival of Campylobacterjejuni in
Unpasteruized Milk. App] Environ Microbiol 1982; 44: 1154-1158.

103

Duffy EA, Belk KE, Sofos JN, Bellinger GR, Pape A, Smith GC. Extent of Microbial
Contamination in United States Pork Retail Products. J F aad Prat 2001; 64: 172-178.

Eberhart-Philips J, Walker N, Garrett N, Bell D, Sinclair D, Rainger W, et al.
Campylobacteriosis in New Zealand: results of a case-control study. J Epidemiol
Community Health, 1997; 51: 686-691.

Effler P, Ieong M-C, Kimura A, Nakata M, Burr R, Cremer E, et al. Sporadic
Campylobacter jejuni Infections in Hawaii: Associations with Prior Antibiotic Use and
Commercially Prepared Chicken. J Infect Dis 2001; 183: 1152-5.

Enders U, Karch H, Toyka K, et al. The spectrum of immune responses to Campylobacter
jejuni and glycoconjugates in Guillain—Barré syndrome and in other neuroimmunological
disorders. Ann Neural 1993; 34(2): 136-144.

Engberg J, Aarestrup FM, Taylor DE, Gemer-Smidt P, Nachamkin I. Quinolone and
Macrolide Resistance in Campylobacter jejuni and C. coli: Resistance Mechanisms and
Trends in Human Isolates. Emerg Infect Dis 2001; 7(1): 24-34

Engleberg NC, Correa-Villasenor A, North CQ, Crow T, Wells JG, Blake PA.
Campylobacter enteritis on Hopi and Navajo Indian Reservations. West J Med 1984;
141(1): 53-56.

Evans MR, Roberts RJ, Ribeiro CD, Gardner D, Kembrey D. A milk-bome
campylobacter outbreak following an educational farm visit. Epidemiol Infect 1996; 117:
457-462.

Evans SJ. The seasonality of canine births and human carnpylobacteriosis: a hypothesis.
Epidemiol Infect 1993; 110: 267-272.

Feasby TE, Hughes RAC. Campylobacter jejuni, antiganglioside antibodies, and
Guillain-Barré syndrome. Neurology 1998; 51: 340-342.

Flanders WD and Kleinbaum DG. Basic Models for Disease Occurrence in
Epidemiology. IntJ Epidemiol 1995; 24(1): 1-7.

Fox JG, Ackerman JA, Newcomer CE. The Prevalence of Campylobacterjejuni in
Random Source Cats Used in Biomedical Research. J Infect Dis 1985; 151(4): 743.

Friedman C, Reddy S, Samual M, Marcus R, Bender J, Desai S, et al. Risk Factors for
Sporadic Campylobacter Infections in the United States: A Case-Control Study on

F oodNet Sites. 2nd International Conference on Emerging Infectious Diseases. Atlanta,
GA, July 2000.

104

Goddard EA, Lastovica AJ, Argent AC . Campylobacter O:41 isolation in Guillain-Barré
syndrome. Arch DisChild 1997; 76: 526-8.

Gondrosen B, Knaevelsrud T, Dommarsnes K. Isolation of Thermophilic
Campylobacters from Norweigan Dogs and Cats. Acta vet Scand 1985; 26: 81-90.

Green DM, Rapper AH. Mild Guillain-Barré Syndrome. Arch Neurol 2001; 58: 1098-
1101.

Gregson NA, Koblar S. Hughes RAC. Antibodies to gangliosides in Guillain-Barré
syndrome: specificity and relationship to clinical features. QJM 1993; 86: 111-117.

Hadden RDM, Karch H, Hartung HP, et al. Preceding infections, immune factors, and
outcome in Guillain-Barré syndrome. Neurology 2001; 56: 758-765.

Hald B and Madsen M. Healthy Puppies and Kittens as Carriers of Campylobacter spp.,
with Special Reference to Campylobacter upsaliensis. J Clin Microbiol 1997; 35(12):
3351-2.

Harris NV, Weiss NS. Nolan CM. The Role of Poultry and Meats in the Etiology of
Campylobacter jejuni/coli enteritis. Am J Public Health 1986; 76: 407-11.

Harris NV, Kimball TJ, Bennett P, Johnson Y, Wakely D, Nolan CM. Campylobacter
jejuni enteritis associated with raw goat’s milk. Am J Epidemiol 1987; 126(2): 179-186.

Harvey RB, Young CR. Ziprin RL, Hume ME, Genovese KJ, Anderson RC, Droleskey
RE, Stanker LH, Nisbet DJ. Prevalence of Campylobacter spp isolated from the

intestinal tract of pigs raised in an integrated swine production system. J Am Vet Med
Assoc 1999; 215: 1601-1603.

Hill AB. The environment and disease: association or causation? Proceedings, R Soc Med
1965; 58: 295-300

Ho TW, Mishu B, Li CY, et al. Guillain-Barré syndrome in Northern China: Relationship
to Campylobacter jejuni infection and anti-glycolipid antibodies. Brain 1995; 118: 597-
605.

Hopkins RS, Olmsted R, Istre GR. Endemic Campylobacter jejuni infection in Colorado:
Identified Risk Factors. Am J Public Health 1984; 74: 249-250.

Hudson SJ, Lightfoot NF, Coulson JC, Russell K, Sisson PR, Sobo AO. Jackdaws and
magpies as vectors of milkbome human campylobacter infection. Epidemiol Infect 1991;

107: 363-372

Hughes RAC, Rees J H; Clinical and Epidemiologic Features of Guillain-Barré
Syndrome. J Infect Dis 1997; 176 (Suppl 2): S92-98.

105

Humphrey TJ and Beckett P. Campylobacterjejuni in dairy cows and raw milk.
Epidemiol Infect 1987; 98: 263-269.

Ikram R, Mitchess P, Brieseman MA, Ikram OH. A case control study to determine risk
factors for campylobacter infection in Christchurch in the summer of 1992-3. N Z Med J
1994; 107: 430-32

Itoh T, Saito K, Maruyama T, Sakai S, Ohashi M, Oka A. An Outbreak of Acute
Enteritis Due to Campylobacter fetus Subspecies jejuni at a Nursery School in Tokyo.
Microbiol Immunol 1980; 24(5): 371-9.

Jacobs BC, van Doom PA, Schmitz PIM, et al. Campylobacter jejuni infections and anti-
GMI antibodies in Guillain-Barré syndrome. Ann Neural 1996; 40: 181-187.

Jacobs BC, Rothbarth PH, van der Meche FGA, Herbrink P, Schmitz PIM, de Klerk MA,
van Doom PA. The spectrum of antecedent infections in Guillain-Barré syndrome: a
case-control study. Neurology 1998; 51: 1 110-1 1 15.

Jones DM, Robinson DA. Occupational Exposure to Campylobacter jejuni infection.
Lancet 1981; i : 440-441. '

Jones EM, Robinson DA, and Eldridge J. Serological studies in two outbreaks of
Campylobacter jejuni infection. Journal of Hygiene 1981; 87: 163.

Jones K, Howard S, Wallace J S. Intermittent shedding of thermophilic campylobacters
by sheep at pasture. J Appl Microbial 1999; 86: 531-536.

Kaldor J, Pritchard H, Serpell A, Metcalf W. Serum antibodies in Campylobacter
enteritis. J Clin Microbial 1983; 18(1): 1-4.

Kaldor J, Speed BR. Guillain-Barre' syndrome and Campylobacter jejuni: a serological
study. BMJ 1984; 288: 1867-1870.

Kapperud G, Rosef 0. Avian Wildlife Reservoir of Campylobacter fetus subsp. jejuni,
Yersinia spp., and Salmonella spp. in Norway. Appl Environ Microbial 1983; 45: 375-
380.

Kapperud G, Skjerve E, Bean NH, Ostroff SM, Lassen J. Risk Factors for Sporadic
Campylobacter Infections: Results of a Case-control Study in Southeastern Norway. J

Clin Microbial 1992; 30(12): 3117-3121.

Kapperud G, Aasen S. Descriptive epidemiology of infections due to therrnotoleant
Camppylobacter spp. in Norway, 1979-1988. APMIS 1992; 100: 883-90.

106

Kapperud G., E. Skjerve, L. Bik, K. Hauge, A. Lysaker, I. Aalmen, S.M. Ostroff, and M.
Potter. Epidemiological investigation of risk factors for campylobater colonization in
Norwegian broiler flocks. Epidemiol Infect 1993; 111: 245-255.

Khalil, K., G.-B. Lindblom, K. Mazhar, and B. Kaijser. Flies and water as reservoirs for
bacterial enteropathogens in urban and nrral areas in and around Lahore, Pakistan.
Epidemiol Infect 1994; l 13: 435-444.

Koga M, Yuki N, Kashiwase K, Tadokoro K, Juji T, Hirata K. Guillain-Barre and
Fisher’s syndromes subsequent to Campylobacter jejuni enteritis are associated with

HLA-B54 and CW] independent of anti-ganglioside antibodies. J Neuroimmunol 1998;
88: 62-66.

Komblatt AN, Barrett T, Morris GK, Tosh FE. Epidemiologic and laboratory

investigation of an outbreak of Campylobacter enteritis associated with raw milk. Am J
Epidemiol 1985; 122(5): 884-889.

Kuroki S, Haruta T, Yoshioka M, Kobayashi Y, Nukina M, Nakanishi H. Guillain-Barré
syndrome associated with Campylobacter infection. Pediatr Infect Dis J, 1991; 10(2):
149-151.

Kuroki S, Saida T, Nukina M, Haruta T, Yoshika M, Kobayashi Y, et al. Campylobacter
jejuni strains from patients with Guillain-Barré syndrome belong mostly to Penner

serogroup 19 and contain (S-N-Acetylglucosamine Residues. Ann Neurol, 1993; 33: 243-
247.

Lander KP, Gill KPW. Experimmental infection of the bovine udder with Campylobacter
coli /jejuni. Journal of Hygiene 1980; 84: 421-428.

Lastovica AJ, Goddard EA, Argent AC. Guillain-Barré syndrome in South Afiica
associated with Campylobacter jejuni O:4l strains. J Infect Dis 1997; 176 Suppl 2: 8139-
S143.

Lighton LL, Kaczmarski EB, Jones DM. A study of Risk Factors for Campylobacter
Infection in Late Spring. Public Health 1991; 105: 199-203.

Lugaresi A, Ragno M, Torrieri F, Di Gugliemo G, Ferrnani P, Uncini A. Acute motor
axonal neuropathy with high titer IgG and IgA anti-GD 1 a antibodies following
Campylobacter enteritis. J Neural Sci 1997; 147: 193-200.

Ma JJ, Nishimura M, Mine H, et al. HLA and T-cell receptor gene polymorphisms in
Guillain-Barré syndrome. Neurology 1998; 51: 379-3 84.

Manfredi R, Nanetti A, Ferri M, Chiodo F. Fatal Campylobacter jejuni bacteraemia in
patients with AIDS. J Med Microbial 1999; 48: 601-3.

107

McCarthy N, Giesecke J. Incidence of Guillain-Barré Syndrome following infection with
Campylobacterjejuni. Am J Epidemiol 2001; 153(6): 610-614.

McKhann GM, Comblath DR, Griffin J W, et al. Acute Motor Axonal Neuropathy: A
Frequent Cause of Acute Flaccid Paralysis in China. Ann Neurol 1993; 33: 333-342.

Millson M, Bokhout M, Carlson J, Spielberg L, Aldis R, Borczyk A, et al. An Outbreak
of Campylobacter jej uni gastroenteritis linked to meltwater contamination of a municipal
well. Can J Public Health 1991; 82: 27-31.

Misawa N, Shinohara S. Sato H, Itoh H, Shinohara K, Shimomura K, et al. Isolation of
Campylobacter species from zoo animals and polymerase chain reaction-based random
amplified polymorphism DNA analysis. Vet Microbial 2000; 71: 59-68.

Mishu B, Ilyas AA, Koski Cl, Vriesendorp F, Cook SD, Mithen FA, Blaser MJ. Serologic
evidence of previous Campylobacter jejuni infection in patients with the Guillain-Barré
syndrome. Ann Intern Med 1993; 118(12): 947-953.

Molnar GK, Mertsola J, Erkko M. Guillain-Barré syndrome associated with
campylobacter infection. BMJ 1982; 285: 652.

Monos DS, Papaioakim M, Ho TW, Li CY, McKhann GM. Differential Distribution of
HLA Alleles in Two F arms of Guillain-Barré Syndrome. J Infect Dis 1997; 176 (Suppl
2): 8180-2.

Moran AP, Prendergast MM, Appelmelk BJ. Molecular mimicry of host structures by
bacterial lipopolysaccharides and its contribution to disease. FEMS Immunol Med
Microbial 1996; 16: 105-115.

Mori M, Kuwabara S, Miyake M, et a1. Haemophilus influenzae infection and Guillain-
Barré syndrome. Brain 2000; 123: 2171-2178.

Nachamkin I. Microbiologic Approaches for Studying Campylobacter species in patients
with Guillain-Barré syndrome. J Infect Dis 1997; 176 Suppl 2: 8106-14.

Neal KR, Slack RCB. Diabetes mellitus, anti-secretory drugs and other risk factors for

campylobacter gastro-enteritis in adults: a case-control study. Epidemiol Infect 1997;
119: 307-11.

Nevo Y, Pestronk A. Acute Immune Polyneuropathies: Correlations of Serum Antibodies
to Campylobacter jejuni and Helicobacter pylori with Anti-GM 1 Antibodies and Clinical
Patterns of Disease. J Infect Dis 1997; 176 Suppl 2: $154-$156.

Ono S, Chida K, Takasu T. Guillain-Barré syndrome following fulminant hepatitis A.
Internal Medicine 1994: 33: 799-801.

108

 

Oosterom J, Notermans S., Karman H, Engels G.B. Origin and Prevalence of
Campylobacter jejuni in Poultry Processing. J Food Prat 1983; 46 (4): 339-344.

Oosterom J, Dekker R, deWilde GJA, vanKempen-deTroye F, Engels GB. Prevalence of
Campylobacter jejuni and Salmonella during pig slaughtering. Vet Q 1985; 7: 31-34.

Orr KE, Lightfoot NF, Sisson PR, Harkis BA, Tweddle JL, Boyd P, et al. Direct milk
excretion of Campylobacterjejuni in a dairy cow causing cases of human enteritis.
Epidemiol Infect 1995; 114: 15-24.

Pearson AD, Greenwood MH, Donaldson J, Healing TD, Jones DM, Shahamat M, et al.
Continuous Source Outbreak of Campylobacteriosis Traced to Chicken. J Food Prat
2000: 63(3); 309-314.

Pi CH, Sorger S, Lackman L, Sinai RE, Marks MI. Campylobacter gastroenteritis in
children. J Pediatr 1979; 94: 589-91.

Popovic-Uroic T. Campylobacter jejuni and Campylobacter coli diarrhoea in rural and
urban populations in Yugoslavia. Epidemiol Infect 1989; 102: 59-67

Potter ME, Blaser MJ, Sikes RK, Kaufmann AF, Wells JG. Human Campylobacter
infection associated with certified raw milk. Am J Epidemiol 1983; 117(4): 475-483.

Prescott JF, Bruin-Mosch CW. Carriage of Campylobacter jejuni in Healthy and
Diarrheic Animals. Am J Vet Res 1981; 42(1): 164-5.

Rees JH, Soudain SE, Gregson NA, Hughes RAC. Campylobacter jejuni infection and
Guillain-Barré syndrome. N Engl J Med 1995; 333: 1374-1379.

Rees JH, Vaughan RW, Kondeatis E, Hughes RAC. HLA-class II alleles in Guillain-
Barre syndrome and Miller Fisher syndrome and their association with preceding
Campylobacter jejuni infection. J Neuroimmunol, 1995; 62: 53-57.

Rhodes KM, Tattersfield AE. Guillain-Barré syndrome associated with Campylobacter
infection. BMJ 1982; 285: 173-174

Rice BE, Rollins DM, Mallinson ET, Carr L Joseph SW. Campylobacter jejuni in broiler
chickens: colonization and humoral immunity following oral vaccination and

experimental infection. Vaccine 1997: 15: 1922-32.

Richardson, RF, Remler BF, Katirji B, Hated Muram M. Guillain-Barre Syndrome after
Cyclospora Infection. Muscle Nerve 1998; 21: 669-671.

Riordan, T., T.J. Humphrey, and A. Fowles. A point source outbreak of campylobacter
infection related to bird-peeked milk. Epidemiol Infect 1993; 110: 261-265.

109

Rivoal M, Denis M, Saivat G, Colin P, Ermel G. Molecular characterization of the
diversity of Campylobacter spp. isolates collected from poultry slaughterhouse: analysis
of cross-contamination. Lett App] Microbial I999; 29: 370-4.

Robinson DA. Infective dose of Campylobacter jejuni in milk. BMJ 1981; 282: 1584.

Rohrbach BW, Draughon FA, Davidson PM, Oliver SP. Prevalence of Listeria
monocytogenes, Camypylobacterjejuni, Yersinia enterocolitica, and Salmonella in Bulk
Tank Milk: Risk Factors and Risk of Human Exposure. J Food Prat 1992; 55: 93-97.

Saeed AM, Harris NV, DiGiacomo RF. The Role of Exposure to Animals in the Etiology
of Campylobacterjejuni / coli Enteritis. Am J Epidemiol 1993; 137(1): 108-114.

Salfield NJ, Pugh EJ. C ampylobacter enteritis in young children living in households
with puppies. BMJ 1987; 294: 21-22.

Schmid GP, Schaefer, RE, Plikaytis BD, Schaefer JR, Bryner JH, Wintermeyer LA, et al.
A One-Year Study of Endemic Campylobacteriosis in a Midwestern City: Association
with Consumption of Raw Milk. J Infect Dis 1987; 156(1): 218-222.

Schorr D, Schmid H, Rieder HL, Baumgartner A, Vorkauf H, Bumens A. Risk Factors
for Campylobacter enteritis in Switzerland. Zbl Hyg 1994; 196: 327-337.

Schwartz M, Sela BA, Eshhar N. Antibodies to Gangliosides and Myelin Autoantigens
Produced in Mice Following Sciatic Nerve Injury. J Neurochem 1982; 38: 1192-1195.

Schwartz S. The Fallacy of the Ecological Fallacy: The Potential Misuse of a Concept
and the Consequences. Am J Public Health 1994; 84(5): 819-824.

Sibbald CJ, Sharp CM. Campylobacter infection in urban and rural populations in
Scotland. Journal of Hygiene 1985; 95: 87-93.

Sillero M, Almirall J. Campylobacter jejuni and Hemolytic-Uremic Syndrome. Nephron
1999; 82: 363-4.

Skirrow MB, Blaser MJ. Clinical Aspects of Infection. In: Nachamkin I, Blaser MJ
editors. Campylobacter. 2nd ed. Washington, DC: ASM Press; 2000: pg. 79.

Slattery ML, Edwards SL1, Caan BJ, Kerber RA, Potter JD. Response Rates among
Control Subjects in Case-Control Studies. Ann Epidemiol 1995; 5: 245-249.

Sorvillo FJ, Lieb LE, Waterman SH. Incidence of carnpylobacteriosis among patients
with AIDS in Los Angeles County. J Acquir Immune Defic Syndr Hum Retravira] 1991;
598-602.

110

Southern JP, Smith RMM, Palmer SR. Bird attack on milk bottles: possible mode of
transmission of Campylobacter jejuni to man. Lancet 1990; 336: 1425-1427.

Sovilla J, Regli F, Francioli P. Guillain-Barré syndrome following Campylobacter jejuni
enteritis. Arch Intern Med 1988; 148: 739-741

Speed BR, Kaldor J, Watson J, et al. Campylobacterjejuni / Campylobacter coli-
associated Guillain-Barre syndrome: Immunoblot confirmation of the serological
response. MedJAust 1987; 147: 13-16

Stanley K, Cunningham R, Jones K. Isolation of Campylobacter jejuni from groundwater.
JApp] Microbiol 1998; 85: 187-191.

St. Louis, Michael E. Water-Related Disease Outbreaks, 1985. MMWR 1988; 37 (SS-2):
15-24.

Studahl A, Andersson Y. Risk factors for indigenous campylobacter infection: a Swedish
case-control study. Epidemiol Infect 2000; 125: 269-275.

Susser, M. The Logic in Ecological: I. The Logic of Analysis. Am J Public Health 1994;
84(5): 825-829.

Tauxe RV, Pegues DA, Hargrett-Bean N. Campylobacter infections: the emerging
national pattern. Am J Public Health 1987; 77: 1219-1221.

Tauxe, RV. Epidemiology of Campylobacter jejuni infections in the United States and
other industrial nations. In: Nachamkin, Blaser, Tompkins, editiors. Campylobacter
jejuni: current and future trends. Washington: American Society for Microbiology; 1992.
P. 9-12

Taylor NS, Ellenverger MA, Wu PY, Fox JG. Diversity of serotypes of Campylobacter
jejuni and Campylobacter coli isolated in laboratory animals. Laboratory Animal Science

1989; 39: 219-221

Thompson J S, Cahoon FE, Hodge DS. Rate of Campylobacter spp. Isolation in Three
Regions of Ontario, Canada, from 1978 to 1985. J Clin Microbiol 1986; 24(5): 876-878.

Toemer JG, Kumar PN. Garagusi VF. Guillain-Barre’ Syndrome Associated with Rocky
Mountain Spotted Fever: Case Report and Review. Clin Infect Dis 1996; 22: 1090-1091.

Torre E, Tello M. Factors influencing fecal shedding of Campylobacter jejuni in dogs
without diarrhea. Am J Vet Res 1993; 54(2): 260-262.

Uyttendaele M, De Troy P, Debevere J. Incidence of Salmonella, Campylobacter jejuni,
Campylobacter coli, and Listeria monocytogenes in Poultry Carcasses and Different

lll

 

Types of Poultry Products for Sale on the Belgian Retail Market. J F aod Prat 1999;
62(7): 735-740.

Van Koningsveld R, Van Doom PA, Schmitz PIM, Ang CW, Van der Meche FGA. Mild
Forms of Guillain-Barré syndrome in and epidemiologic survey in The Netherlands.
Neurology 2000; 54: 620-625.

Vriesendorp F J , Triggs WJ, Mayer RF, Koski CL. Electrophysiological studies in
Guillain-Barre' syndrome: correlation with antibodies to GMI, Gle and Campylobacter
jejuni. J Neural I995; 242: 460-465.

Vriesendorp F J . Insights into Campylobacter jejuni-Induced Guillain-Bare Syndrome
from the Lewis Rat Model of Experimental Allergic Neuritis. J Infect Dis 1997; 176
(Suppl 2): S164-168.

Wallace J S, Stanley KN, Jones K. The colonization of turkeys by thermophilic
campylobacters. J App] Microbiol 1998; 85: 224-230.

Walsh F S, Cronin M, Koblar S, Doherty P, Winer J, Leon A, Hughes RAC. Association
between glycoconjugate antibodies and Campylobacter infection in patients with
Guillain-Barré syndrome. J Neuroimmunol 1991; 34: 43-51.

Weber F, Rudel R, Aulkemeyer P, Brinkmeier H. Anti-GM] antibodies can block
neuronal voltage-gated sodium channels. Muscle Nerve 2000; 23: 1414-1420.

Willison HJ, O’Hanlon GM. Antiglycosphingolipid antibodies and Guillain-Barre
Syndrome. In: Nachamkin I, Blaser MJ editors. Campylobacter. 2“"1 ed. Washington,
DC: ASM Press; 2000; pg 261.

Winer JB, Briggs D, Welsh K, Hughes RAC. HLA antigens in the Guillain-Barré
syndrome. J Neuroimmunol 1988; 18: 13-16.

Winer JB, Hughes RAC, Anderson MJ, Jones DM, Kangro H, Watkins RPF. A
prospective study of acute idiopathic neuropathy. II. Antecedent events. J Neural
Neurosurg Psychiatry 1988; 51: 613-618.

Wirguin I, Suturkova-Milosevic L, Della-Latta P, Fisher T, Brown RH, Latov N.
Monoclonal IgM Antibodies to GMI and Asialo-GMI in Chronic Neuropathies Cross-
React with Campylobacter jejuni Lipopolysaccharides. Ann Neurol 1994; 35: 698-703.

Wood RC, MacDonald KL, Osterholm MT. Campylobacter enteritis outbreaks

associated with drinking raw milk during youth activities. JAMA 1992; 268(22): 3228-
30.

112

Young DM, Biao J, Zheng Z, Hadler J, Edberg SC. Isolation of Campylobacter jejuni in
Hunan, China: Epidemiology and comparison of Chinese and American methodology.
Diagn Microbiol Infect Dis 1986; 5(2): 143-149.

Yuki N, Yoshino H, Sato S, Miyatake T. Acute Axonal Neuropathy associated with high
titer IgG and IgA anti-GDla antibodies following Campylobacter enteritis. Neurology
1990; 40: 1900-1902.

Yuki N. Molecular mimicry between gangliosides and lipopolysaccharides of
Campylobacter jejuni isolated from patients with Guillain—Barré syndrome and Miller
Fisher Syndrome. J Infect Dis 1997; 176 Suppl 2: $150-$153.

Yuki N, Ho TW, Tagawa Y, Koga M, Li CY, Hirata K, et al. Autoantibodies to GMlb

and GalNAc-GDla: relationship to Campylobacter jejuni infection and acute motor
axonal neuropathy in China. J Neurol Sci 1999; 164: 134-138.

113