Campylobacter species: don't put all your eggs in one chicken.

Full text

Page 1

CORRESPONDENCE • CID 2002:34 (1 March) • 719
curred, ELISA reactivity had decreased,
SIA reactivity against HCV core protein
had disappeared, and SIA reactivity
against NS3 and NS4 proteins had de-
creased (table 1). Evolution of the HCV
status showed that the nurse was recov-
ering from her infection.
In the case report by Morand et al. [1],
complete seroconversion never occurred,
despite weak ELISA reactivity at week 5.
A short period (1 month) of viral rep-
lication was associated with a high ALT
level. In our experience, seroconversion
occurred and was followed by serore-
version that included a 2–3-month rep-
lication period and a transiently mod-
erate elevation of the ALT level. These 2
cases outline the fact that absence of se-
roconversion or rapid seroreversion is
strongly related to quick virus eradica-
tion, either spontaneously or with treat-
ment [1–3, 5]; a recent report by Jaeckel
et al. [4] supports this finding. Further
studies of T cell response in such treated
patients might help us improve our un-
derstanding of the mechanisms of early
virus elimination. In our experience, fol-
low-up with SIA is a useful tool for prog-
nosis. Finally, the sustained response ob-
tained with IFN-a therapy inourpatient,
despite the persistence of virus replica-
tion at 1 month of treatment and a ge-
notype 1a virus, outlines the different vi-
rological response profile of primary and
chronic infection with HCV. All param-
eters—virological, biochemical, and im-
munological—are of importance when
making decisions regardingadaptationof
postexposure treatment (i.e., type, dose,
and duration of treatment).
In conclusion, acute hepatitis C is of-
ten associated with the absence of sero-
conversion or seroreversion, as demon-
strated by the case reports presentedhere
and elsewhere [1, 3, 4]. This may lead to
underestimation of the global prevalence
of hepatitis C,and,moreover,thechronic
evolution of hepatitis C may be over-
estimated.
Sophie Alain,1Ve ´ronique Loustaud-Ratti,2
Fre ´de ´ric Dubois,3Marie-Dominique Bret,4
Sylvie Rogez,1Elizabeth Vidal,2
and Franc ¸ois Denis1
Departments of
2Internal Medicine, Teacher Hospital Dupuytren,
Limoges,
Bretonneau, Tours, and
Department, Gueret Hospital, Gueret, France
1Bacteriology-Virology-Hygiene
and
3Department of Virology, Teacher Hospital
4Health Workers
References
1. Morand P, Dutertre N, Minazzi H, et al. Lack
of seroconversion in ahealth careworkerafter
polymerase chain reaction–documentedacute
hepatitis C resulting from a needlestickinjury.
Clin Infect Dis 2001;33:727–9.
2. Takagi H, Uehara M, Kakizaki S, et al. Acci-
dental transmission of HCV and treatment
with interferon. J Gastroenterol Hepatol
1998;13:238–43.
3. Genesca J, Esteban JI, Quer J, et al. Hepatitis
C virus markers in patients with acute post-
transfusion hepatitis treated with interferon
alfa-2b. Gut 1993;34(Suppl):S62–3.
4. Jaeckel E, Cornberg M, Wedemeyer H, et al.
Treatment of acute hepatitis C with interferon
alfa-2b. N Engl J Med 2001;345:1452–7.
5. Poignet JL, Degos F, Bouchardeau F, Chau-
veau P, Courouce AM. Complete response to
interferon-alpha for acute hepatitis C after
needlestick injury in a hemodialysis nurse. J
Hepatol 1995;23:740–1.
Reprints or correspondence: Dr. Sophie Alain, Laboratoire de
Virologie, CHU Dupuytren, 2 avenue Martin Luther King,
87042 Limoges cedex, France (sophie.alain@unilim.fr).
Clinical Infectious Diseases
? 2002 by the Infectious Diseases Society of America. All
rights reserved. 1058-4838/2002/3405-0028$03.00
2002;34:717–9
Campylobacter
Species: Don’t Put
All Your Eggs
in One Chicken
Sir—We read with interest the recent ar-
ticle “Campylobacter jejuniInfections:Up-
date on Emerging Issues and Trends” by
Allos [1], which raised the profile of this
most important yet frustratingly elusive
pathogen. However, we believe that a
number of statements are not supported
by the references cited bytheauthor.First,
it is stated that “the reported incidence of
Campylobacterinfectionamonghomosex-
ual men is almost 40 times greater than
inthegeneralpopulation”(p.1204),citing
a study by Sorvillo et al. [2]. In fact, this
study from Los Angeles compared the in-
cidence of reported gastroenteritis due to
Campylobacter species in the general pop-
ulation with the incidence amongpatients
with AIDS, who were not exclusively ho-
mosexual men. The figure quoted by the
author is likely to be an overestimation,
because persons with AIDS would be ex-
pected to present to the health services
(and, hence, be reported) more than
would the general population.
It is also stated that, “depending upon
the population studied, as many as 50%
of persons who are infected during out-
breaks are asymptomatic” (p. 1202). The
study cited by the author [3] reported a
ratio of illness to infection for Campy-
lobacter species of 1:2 among Mexican
children !5 years of age who were fol-
lowed up for 1 year, but there is no in-
dication in that study that the cases were
related to an outbreak of infection. In-
deed, the ratio of illness to infection de-
pends upon the prevalence of immunity
and is likely to be influenced by under-
lying endemicity and the degree of pre-
vious exposure, regardless of whether in-
fection is acquired sporadically or as part
of an outbreak. This was supported in
the study by the positive relationship be-
tween age and proportion of asymptom-
atic infections.
Most importantly, the author states,
“In studies in many parts of the United
States, Europe, and Australia, 50%–70%
of all Campylobacter infections have been
attributed to consumption of chicken”
(p. 1203). One of the studies cited is that
by Adak et al. [4], which was performed
in England in 1990. The results of this
study appear to have been misinterpre-
ted. The study did notreportpopulation-
attributable fractions for the exposures
identified, and, contrary to what was im-
plied, the study demonstrated an inverse
relationship between handling or con-
sumption of chicken prepared in the
home and risk of gastroenteritis due to
Campylobacter species (OR, 0.44; 95%
CI, 0.24–0.79). Among the risk factors

Page 2

720 • CID 2002:34 (1 March) • CORRESPONDENCE
identified, however, were occupationalex-
posure to raw meat, presence of a house-
hold pet with diarrhea, and ingestion of
untreated water from lakes, rivers, and
streams. More recent studies have shown
that consumption of commercially-pre-
pared chicken (but not chicken that is
prepared in the home) is a risk factor
[5–7]. Moreover, although the consump-
tion of chicken accounts for a minority
of cases, most infections remain unex-
plained by recognized risk factors [8].
Thus, the role of chicken consumption
as a route of transmission of Campylo-
bacter species to humans remains un-
clear, and while it is plausible that cross-
contamination of other foods occurs
after the handling of raw poultry, thishas
not been demonstrated convincingly in
case-control studies. Twenty years of em-
phasizing the importance of poultry in
Campylobacter transmission has yielded
little success in terms of its control and
prevention. There may well be much
more to Campylobacter infection than
chickens. We need to keep an openmind.
Clarence C. Tam, Sarah J. O’Brien,
Goutam K. Adak, and Iain A. Gillespie
Gastrointestinal Diseases Division, Communicable
Disease Surveillance Centre, London,
United Kingdom
References
1. AllosBM.Campylobacterjejuniinfections:up-
date on emerging issues and trends. Clin In-
fect Dis 2001;32:1201–6.
2. Sorvillo FJ, Lieb LE, Waterman SH. Incidence
of campylobacteriosis among patients with
AIDS in Los Angeles County. J Acquir Im-
mune Defic Syndr 1991;4:598–602.
3. Calva JJ, Ruiz-Palacios GM, Lopez-Vidal AB,
Ramos A,Bohalil R.Cohortstudyofintestinal
infection with campylobacterinMexicanchil-
dren. Lancet 1988;1(8584):503–6.
4. Adak GK, Cowden JM, Nicholas S, Evans HS.
The Public Health Laboratory Service na-
tional case-control study of primary indige-
nous sporadic cases of campylobacter infec-
tion. Epidemiol Infect 1995;115:15–22.
5. A report of the study of infectious intestinal
disease in England. London: The Stationery
Office, 2000.
6. Friedman C, Reddy S, Samuel M, et al., and
the EIP Working Group. Risk factors for
sporadic Campylobacter infections in the
United States: a case-control study on
FoodNet sites. Proceedings of the 2nd In-
ternational Conference on Emerging Infec-
tious Diseases (Atlanta, Georgia). 2000.
Available at: http://www.cdc.gov/foodnet/
pub/icedid/2000/friedman_c.htm.
7. Effler P, Ieong M, Kimura A, et al. Sporadic
Campylobacter jejuni infections in Hawaii:
associations with prior antibiotic use and
commercially prepared chicken. J Infect Dis
2001;183:1152–5.
8. Neal KR, Slack RC. Diabetes mellitus, anti-
secretory drugs and other risk factors for
Campylobacter gastroenteritis in adults: a
case-control study. Epidemiol Infect 1997;
119:307–11.
Reprints or correspondence: Clarence C. Tam,Gastrointestinal
Diseases Division, PHLS Communicable Disease Surveillance
Centre, 61 Colindale Ave., London NW9 5EQ,UnitedKingdom.
Clinical Infectious Diseases
? 2002 by the Infectious Diseases Society of America. All
rights reserved. 1058-4838/2002/3405-0029$03.00
2002;34:719–20
Reply
Sir—Dr. Tam and colleagues raise 3 in-
teresting points about the epidemiology
of Campylobacter infections [1]. First,
they point out that Campylobacter infec-
tions that occur in a group of persons
with AIDS who are largely homosexual
men are more likely to be reported than
are such infections in healthier persons
[2]. I agree with Tam et al. [1] that these
studies may overestimate the excess risk.
Furthermore, as stated in my paper, the
most recent analyses [3] suggest that the
rate of Campylobacter infection in HIV-
negative homosexual men is no higher
than the rate in heterosexual men of a
similar age.
Second, my colleagues dispute the
statement that, in some studies, up to
50% of persons infected with Campylo-
bacter species are asymptomatic. They
point out that such high rates of asymp-
tomatic infection are typical only among
young children residing in areas of en-
demicity. Tam and colleagues are quite
correct both in their statement and in
their interpretation of the cited refer-
ences. High rates of asymptomatic Cam-
pylobacter infectionsdetectedduringout-
breaks in Western countries, although
reported [4–6], are distinctly unusual.
Most persons infected with Campylobac-
ter species, especially in industrialized
nations, are likely to exhibit symptoms.
Finally, Tam and colleagues argue
against the importance of poultry that is
not eaten in restaurants as a leading
source of Campylobacter infection in hu-
mans. My reviewof therelevantliterature
leads me to disagree. The following facts
bear repeating:
1.US Department of Agriculture
statistics indicate that ∼90% of broiler
chicken carcasses are contaminated with
Campylobacter species[7].Althoughother
studies haveshownlowerratesofcontam-
ination, dependingontheseasonandgeo-
graphic location, in virtually all investi-
gations, the contamination rate is ?50%.
2.The quantity of Campylobacter or-
ganisms on the surface of a fresh chicken
carcass is estimated to be 103to 106per
chicken [8].
3. The infective dose of Campylo-
bacter species is not known precisely, but
as few as 500 organisms can produce ill-
ness in some healthy persons [9].
4.The importance of cross-contam-
ination of food by Campylobacter species
in domestic kitchens has been investi-
gated and documented [10]. During the
preparation of ordinary meals, the bac-
teria can be widely disseminatedtochop-
ping boards, counter tops, knives, and
human hands.
5.Although numerous case-control
studies of Campylobacter-infected per-
sons, including the studies cited by the
authors, have documented a variety of
potential sources, when one reviews the
literature in its totality, the evidence is
rather overwhelming that poultry con-
sumption and preparation is implicated
in most infections in humans [3].
6. The emergence of antibiotic resis-
tance among Campylobacter isolates re-
covered from humans has directly fol-
lowed and mirrored such resistance in
Campylobacter isolates recovered from
poultry flocks [11]. Additionalmolecular
subtyping of strains recovered from
chickens and humans has confirmed the
association between Campylobacter con-