  |
|
|
||||
|
||||||
| Journal of the New Zealand Medical Association,
21-May-2004, Vol 117 No 1194
|
||||||
|
Lack of association between long-term illness and infectious
intestinal disease in New Zealand
Rob Lake, Michael Baker, Carolyn Nicol, and Nick
Garrett
From 1988 to 2002, New Zealand experienced a marked increase
in the number of notified cases of infectious intestinal disease, especially
campylobacteriosis.1 It has been estimated that
2–3% of infectious intestinal disease cases develop a variety of secondary
long-term illnesses.2 The most recognised of
these are haemolytic uraemic syndrome (HUS) after infection with Shiga-like
toxin producing Escherichia coli,
reactive arthritis, and Guillain-Barré syndrome
(GBS).3
Reactive arthritis is associated with preceding infection by
a number of organisms causing intestinal disease—including
Campylobacter,
Salmonella,
Shigella, and
Yersinia; as well as genital
infections. An analysis of reactive arthritis in Otago from 1986 to 1993
suggested that enteric infections were the predominant cause, with
Yersinia enterocolitica being the
commonly isolated organism.4
Campylobacter is
the most common antecedent pathogen for GBS3,
and the attributable risk of GBS for laboratory confirmed cases of
Campylobacter jejuni infection has been
estimated as 30.1 per 100,000 cases (95% confidence interval (CI):
13.9–57.8) in a Swedish study.5 A review
of this association estimated that 30–40% of GBS cases had suffered
Campylobacter jejuni infection prior to
the onset of GBS.6
As part of a study of the number of cases of food-borne
intestinal disease in New Zealand7, we analysed
communicable disease notification (ESR, EpiSurv) and hospital discharge data
(New Zealand Health Information Service) for infectious intestinal disease and
associated secondary long-term illness cases. This study did not consider trends
in disease incidence. The number of cases of HUS has increased from an average
of 9 a year from 1990 to 1995, to 23 a year from 1996 to 2002—and much of
this increase may be attributable to the rising incidence of Shiga-like toxin
producing Escherichia coli infection in
New Zealand.8
In this paper, we consider whether the increase in notified
cases of infectious intestinal disease was associated with increases in the
number of admissions for GBS and reactive arthritis. The period between
prodromal infection and onset of GBS symptoms has been estimated as up to 3
weeks,9 while reactive arthritis symptoms begin
approximately 7 to 30 days after intestinal
illness3. Therefore the incidence of secondary
illness should follow closely any changes in infectious intestinal
disease.
MethodsNotification data were obtained
from surveillance reports.10 Hospital
admissions were obtained from the New Zealand Health Information Service for all
cases, with one of the codes of interest recorded in the first nine diagnosis
fields. The conditions of interest were GBS (357.0), arthropathy associated with
Reiter’s disease and nonspecific urethritis (711.1), postdysenteric
arthropathy (711.3), unspecified infective arthritis (711.9), and other
inflammatory spondylopathies (720.8). All readmissions were excluded over the
entire time period. Hospitalisations were categorised based on the first
diagnosis code that contained one of the above codes.
ResultsTable 1 shows the number of notified
cases of enteric disease, together with the number of hospital admissions coded
to relevant arthritis types and GBS, for the 15-year period from 1988 to
2002.
These data suggest only a weak association between the rise
in campylobacteriosis and the incidence of GBS (Pearson correlation
R2=0.40; p=0.14; for 1988 to 2002).
Penner serotyping based on the heat stable (HS)
antigen(s)11 has been conducted for 1130
Campylobacter isolates obtained from
human cases in New Zealand between 1996 and 2001. The serotypes identified
include almost all of those that have been associated with
GBS12: HS:1,44 (16% of serotyped isolates);
HS:2 (23%); HS:4 complex (15%); HS:5 (0.6%); HS:10 (0.6%); HS:19 (0.8%); HS:23
(8%); HS:35 (1.3%); HS:37 (4%); HS:41 (0.5%).
Serotypes HS:19 and HS:41 have been suggested as being
associated with GBS in some parts of the world, and the low proportion of these
serotypes amongst isolates from New Zealand could partially explain the lack of
correlation between campylobacteriosis notifications and GBS in this country.
However, this association is not consistent. Internationally, a wider range of
serotypes have been identified amongst isolates from GBS
cases.12
Table 1. Number of cases of selected infectious
intestinal diseases reported in New Zealand 1988 – 2001, and number of
discharged patients coded to reactive arthritis and Guillain-Barré
syndrome (GBS) for the same period
(to view the table,
see the PDF version of this paper)
There was generally no association between the various
enteric diseases (campylobacteriosis, salmonellosis, and shigellosis) and the
number of cases coded to post-dysenteric arthropathy, or other potential
sequealae codes. An association between campylobacteriosis and unspecified
infective arthritis admissions was more marked over the period 1988 to 1998
(Pearson correlation R2=0.75; p<0.008).
However, cases assigned to this code will be a mixture of
both infective arthritis and non-infective arthritis (Dr John O’Donnell,
Canterbury District Health Board, personal communication, 2003) and (without
examining case details more closely) this association must be considered as
suggestive only. A statistically significant association between
campylobacteriosis and arthropathy associated with Reiter’s disease and
nonspecific urethritis (NSU) (Pearson correlation
R2=0.69; p<0.02) was observed for the same
period, but this will be complicated by the potential for arthritis in response
to NSU as well as enteric infections.
Coding changes (International Classification of Diseases
ICD-9-CMA-II to ICD-10-AM) are thought to have caused the significant changes to
reported numbers of reactive arthritis cases from 1999 to 2002—so these
years were not included in the analysis.
DiscussionThis ecological analysis therefore
suggests that, in New Zealand in recent years, infectious intestinal diseases
are not making a significant contribution to the burden of hospitalisation for
reactive arthritis and GBS. If it is correct that 30–40% of GBS cases have
an antecedent infection with Campylobacter
jejuni,6 then a strong association
between the number of cases could have been expected. Using the predicted rate
from the Swedish study,5 and assuming that the
unreported to reported campylobacteriosis case ratio is 7.6:1 for New
Zealand,7 then the number of associated GBS
cases could have been expected to rise from 7.3 in 1988 (95% CI; 3.3–13.9)
to 32.7 in 2002 (95% CI; 14.9–62.0). This increase was not seen.
The coding of reactive arthritis cases makes examination of
any association difficult, but the lack of correlation between the number of
cases of any of the enteric diseases and those of post dysenteric arthropathy
suggests that any link is tenuous.
Author information: Rob Lake, Scientist, Food Group, Christchurch Science Centre, Institute of Environmental Science and Research (ESR) Ltd, Christchurch; Michael Baker, Senior Lecturer, Department of Public Health, Wellington School of Medicine and Health Sciences, Wellington; Carolyn Nicol, Senior Scientist, Enteric Reference Laboratory, Kenepuru Science Centre, Institute of Environmental Science and Research (ESR) Ltd, Porirua; Nick Garrett, Biostatistician, Faculty of Health Research Centre, Auckland University of Technology, Auckland Acknowledgements:
The New Zealand Ministry of Health support the operation of the national
communicable disease surveillance system, including collection and analysis of
notification data and organism typing. We also thank Elizabeth Sneyd (ESR,
Kenepuru Science Centre) for data analysis, and Dr John O’Donnell
(Canterbury District Health Board) for advice.
Correspondence: Rob
Lake, Christchurch Science Centre, Institute of Environmental Science and
Research (ESR) Ltd, PO Box 29-181, Christchurch. Fax (03) 351 0010; email: rob.lake@esr.cri.nz
References:
 |
|
||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|