Hospitalization for trichinellosis and echinococcosis in Canada, 2001-2005: the tip of the iceberg?
Gilbert, Nicolas L. ; Dare, Oluwayemisi K. ; Libman, Michael D. 等
This study was undertaken to quantify geographic variations in the
occurrence of echinococcosis and trichinellosis, two infections that are
to a large extent autochtonous (i.e., acquired within the country).
Currently, no helminth infection is reportable nationally in
Canada. Trichinellosis used to be but was removed from the list in 2000,
though it is still notifiable in most provinces and territories. For
both diseases, non-specific clinical signs make diagnosis difficult and
result in under-reporting. (1,2) Most cases of echinococcosis and
trichinellosis are treated without hospitalization but some
complications may require hospitalization for investigation or
treatment.
Echinococcosis is caused by Echinococcus spp. tapeworms (class
Cestoda). The definitive hosts of the tapeworms are wild or domesticated
canids, and humans acquire infection from contact with contaminated
faeces from these species. (3) Natural intermediate hosts range from
rodents to cervids, but in Canada echinococcosis is thought to be most
commonly caused by the sylvatic variant of Echinococcus granulosus, for
which caribou, elk and moose are the main natural intermediate hosts.
(4) Infection with E. granulosus eggs results in the development of
characteristic hydatid cysts in the liver, lungs or other organs, while
infection with eggs of the less common E. multilocularis results in the
more invasive 'alveolar echinococcosis' that may spread
metastatically through major organs. (5) Treatment is difficult,
combining careful surgical removal of cysts and anthelmintic therapy,
(6) but is usually successful for infections occurring in Canada. (2)
Trichinellosis in Canada is attributable to three Trichinella
species of nematodes--T. spiralis, T. pseudospiralis and T. nativa--that
can cause two distinct syndromes. Encapsulation of larvae in muscles
causes inflammatory responses leading to pain in muscles, eyelids and
the face, and in some instances more severe complications such as
myocarditis.5 Reinfection in sensitized individuals causes
immune-mediated gastroenteritis, a syndrome referred to as
"secondary trichinellosis", which has been observed in the
Canadian Arctic in individuals infected by T. nativa. (7,8)
The annual number of human cases of trichinellosis reported in
Canada between 1970 and 1997 ranged between 3 to 49, with a mean of
18.2. (1) Several outbreaks that occurred in Inuit communities of
Nunavik and Nunavut in the 1980s and 1990s were traced to consumption of
raw or fermented walrus meat, which are common meals in these
communities. (7-10) Domestic swine, the main reservoir of Trichinella
worldwide, is no longer a source of infection in Canada where domestic
swine is virtually free from this para site. (11,12)
<01_TB034>
Most of the current knowledge of the epidemiology of echinococcosis
and trichinellosis is based on outbreak investigations and case studies,
and therefore restricts the possibility of estimating or comparing
disease rates. The objectives of this study were to quantify the
occurrence of those trichinellosis and echinococcosis cases sufficiently
severe to require hospitalization in Canada, and to compare the
incidence of hospitalization between residents of northern regions and
the rest of the Canadian population. A higher incidence in the north was
expected for both diseases based on published case and outbreak reports.
(1,2,4,6-10)
METHODS
Case records were extracted from the hospital morbidity database
(HMDB) held by the Canadian Institute for Health Information (CIHI). All
hospitalizations occurring in Canada are included in this database and
discharge diagnoses are coded using either the ninth or the tenth
revision of International Classification of Diseases (ICD-9 and ICD-10)
of the World Health Organization, (13,14) depending on jurisdictions.
All cases for which the first discharge diagnosis was either
trichinellosis (ICD-9 code 124, ICD-10 code B75) or echinococcosis
(ICD-9 codes 122.0 to 122.9, ICD-10 codes B67.0 to B67.9) were included.
In addition, cases whose second discharge diagnosis was one of these
diseases were included if their first discharge diagnosis was a symptom
or condition likely caused by one of these infections (Table 1).
The HMDB contains encrypted health card numbers that were used to
find duplicate cases (two or more hospitalizations for the same
individual in the same year). Such duplicate entries were merged, i.e.,
lengths of hospital stays were summed up. When this encrypted number was
missing, two individuals with the same age, sex and postal code and
hospitalized for the same disease in the same year were assumed to be
the same individual (this only occurred for two echinococcosis cases).
The province of residence of cases was determined based on their
postal code or the province that issued their health card. When both of
these were missing, the province of hospitalization was assumed to be
the province of residence. The latitude at the centroid of forward
sortation areas (FSA), i.e., the first three digits of cases'
postal codes, was determined using commercial digitalized maps (DMTI
Spatial Inc, Markham, Ontario), and then latitude categories were
created (<50[degrees], 50[degrees]-54[degrees], and [greater than or
equal to] 55[degrees]).
Data were analyzed using SAS Enterprise Guide 4 (SAS Institute
Inc.). Crude and standardized annual incidence rates were determined by
province and territory and by latitude categories, using population
numbers from the 2001 Census (15) as denominators. Relative risks and
binomial 95% confidence intervals (CI) were calculated. For
echinococcosis, comparisons were made between latitude categories; for
trichinellosis, Northern Quebec (defined as postal code starting with
J0M, which includes Nunavik and James Bay) and Nunavut, i.e., regions
where Inuit represent a large proportion of the population, were
compared to the rest of Canada.
RESULTS
In 2001-2005, 108 hospitalizations related to echinococcosis were
found. Nineteen cases (15%) were aged less than 20, including two
children aged less than 2 years. Most cases (67%) were women, and their
risk of hospitalization was significantly greater than that of men (RR
1.92, 95% CI 1.29-2.87). The incidence of echinococcosis hospitalization
across Canada was 0.72 per million per year. There was a clear
south-north gradient in the incidence of hospitalization for
echinococcosis, the highest incidence (2.9 per million per year) being
measured north of the 55th parallel (Table 2).
There were 14 hospitalizations related to trichinellosis. All cases
were adults ([greater than or equal to] 21 years) and most of them
(9/14) were aged 40 or over. Most cases were males and the risk of
hospitalization was greater in males than in females (RR 3.81, 95% CI
1.06-13.67). The incidence of trichinellosis hospitalization across
Canada was 0.09 per million per year. However, a much higher incidence,
42 per million per year, was found in Nunavut and northern Quebec: the
risk of hospitalization for trichinellosis was more than two orders of
magnitude higher in those regions compared to the rest of the country
(Table 3).
DISCUSSION
This study confirms that while the national incidence of
echinococcosis and trichinosis may be low, the inhabitants of northern
regions of Canada are at much higher risk of hospitalization due to
these infections than are other Canadians. These two infections rarely
require hospitalization, so hospitalized cases likely represent only the
most severe cases. Therefore, incidence based on hospitalization data is
most likely a considerable underestimation of the true incidence, and
the number of cases identified in our study may represent only the tip
of the iceberg in terms of burden of disease.
As trichinellosis usually presents as an acute disease within two
weeks after ingestion of infective larvae, geographical and temporal
variations in hospitalization may reflect those of disease incidence. In
contrast, as Echinococcus cysts typically grow for years or decades
before they come to medical attention, variations in hospitalization
rates may rather reflect past trends in disease incidence.
We included only cases with either trichinellosis or echinococcosis
as the first discharge diagnosis, or cases for which these diseases were
identified as the second discharge diagnosis and for which the first
diagnosis was a condition most likely related to these diseases. We
therefore excluded several cases with echinococcosis as a secondary
diagnosis, in which the disease was either already known, or
asymptomatic and discovered by accident. We may therefore have excluded
some cases for which the primary cause of hospitalization was caused by
Echinococcus infection.
<01_TB035>
<01_TB036>
On the other hand, the use of a national hospital database allowed
us to search cases hospitalized across the country for a given period,
which would not have been possible through chart reviews, and to
investigate geographic variations in the incidence of hospitalization.
In the absence of national surveillance data, these results may provide
a baseline that will allow the monitoring of temporal variations in the
occurrence of these diseases. It is anticipated that climate change may
impact the epidemiology of zoonotic infectious diseases in the Arctic
through a variety of mechanisms, in particular changes in the ecology of
natural hosts. (16)
Echinococcosis is not uncommon in Canada. Seventeen patients with
symptomatic E. granulosus infection apparently acquired in Manitoba or
in Northern Ontario were recorded in two Winnipeg hospitals from 1987 to
1997, (6) while 22 symptomatic cases were recorded in Edmonton hospitals
from 1991 to 2001. (2) Our findings that women and inhabitants of
northern regions are at higher risk of being hospitalized for
echinococcosis are consistent with previous observations: of the 22
symptomatic echinococcosis cases investigated in Edmonton, (2) 77% were
female, 41% were resident of the Northwest Territories (Edmonton
hospitals serve as reference centre for NWT), and 41% were either
self-declared Aboriginals or resident of First Nation reserves. The
higher incidence of echinococcosis hospitalization in residents of
northern communities may result from more frequent contact with the
parasite. Country foods are particularly important for Dene/Metis and
Inuit communities in the Canadian Arctic, (17) so hunting, house or sled
dogs are likely to have a diet with a high proportion of offal from
hunted game. Indeed, E. granulosus infection prevalence in these dogs is
high. (3)
Echinococcus multilocularis is endemic in Central Europe, Central
Asia and China, (18) and E. granulosus is endemic in parts of South
America, East Africa, Central Asia and China. (19) Therefore, some of
the cases may be immigrants from or people who traveled to these areas.
Cystic echinococcosis control programs have been put in place in
various parts of the world, and some were indeed successful. (20) All of
these programs took place in countries where the main intermediate hosts
of E. granulosus were livestock species, therefore approaches targeting
these species may not be applicable in Canada where intermediate hosts
are wildlife species. On the other hand, in Iceland, a control strategy
centered on the education of dog owners and the prevention of dogs
feeding from offal led to the elimination of cystic echinococcosis.
That northern communities, particularly Inuit, are at risk of
trichinellosis is well known: several trichinellosis outbreaks caused by
the consumption of raw or undercooked walrus meat have been reported in
Inuit communities in Nunavut and Nunavik in the 1990s. (7-10) Findings
on trichinellosis hospitalization in this study confirm those from
outbreak investigations: the fact that cases are mostly older adults is
consistent with similar observations by MacLean et al. (8) and with
their finding that the frequency of walrus meat consumption among Inuit
increases with age.
Seven patients from northern Quebec and Nunavut were hospitalized
for trichinellosis in 2001-2005, compared to 11 from only two villages
of Nunavik in 1982-1984. (7) This decrease is likely attributable to the
trichinellosis prevention program initiated in Nunavik in 1992. This
program is based on the testing of walrus before distributing its meat.
(10)
Most trichinellosis hospitalizations identified in this study
occurred in people who were not resident of northern regions. The source
of these infections remains unknown. Trichinella has been detected in
several wildlife species, (1) including black bear which has also been
identified as the source of an outbreak in Saskatchewan in 2000. (21) As
mentioned earlier, pork meat is an unlikely source of Trichinella
infection in Canada.
In conclusion, the incidence of hospitalization for echinococcosis
and trichinellosis is low at the national level. However, significantly
higher rates have been measured in northern regions of Canada despite
the fact that both diseases are theoretically preventable and that a
Trichinella control program is in place in Nunavik. Further efforts,
probably educational in nature, will be required to reduce the incidence
of these infections in high-risk areas.
Received: December 4, 2009
Accepted: March 8, 2010
REFERENCES
(1.) Appleyard GD, Gajadhar AA. A review of trichinellosis in
people and wildlife in Canada. Can J Public Health 2000;91(4):293-97.
(2.) Somily A, Robinson JL, Miedzinski LJ, Bhargava R, Marrie TJ.
Echinococcal disease in Alberta, Canada: More than a calcified opacity.
BMC Public Health 2005;5(1):34.
(3.) Moro P, Schantz PM. Cystic echinococcosis in the Americas.
Parasitol Int 2006;55:S181-S186.
(4.) Miller MJ. Hydatid infection in Canada. CMAJ1953;68(5):423-34.
(5.) Villeneuve A. Les zoonoses parasitaires. L'infection chez
les animaux et chez l'homme. Montreal : Presses de
l'Universite de Montreal, 2003.
(6.) Al-Saghier M, Taylor MC, Greenberg HM. Canadian-acquired
hydatid disease: A case report. Can J Infect Dis 2001;12(3):178-82.
(7.) MacLean JD, Viallet J, Law C, Staudt M. Trichinosis in the
Canadian Arctic: Report of five outbreaks and a new clinical syndrome. J
Infect Dis 1989;160(3):513-20.
(8.) MacLean JD, Poirier L, Gyorkos TW, Proulx JF, Bourgeault J,
Corriveau A, et al. Epidemiologic and serologic definition of primary
and secondary trichinosis in the Arctic. J Infect Dis
1992;165(5):908-12.
(9.) Serhir B, MacLean JD, Healey S, Segal B, Forbes L. Outbreak of
trichinellosis associated with arctic walruses in northern Canada, 1999.
Can Commun Dis Rep 2001;27(4):31-36.
(10.) Proulx JF, MacLean JD, Gyorkos TW, Leclair D, Richter AK,
Serhir B, et al. Novel prevention program for trichinellosis in inuit
communities. Clin Infect Dis 2002;34(11):1508-14.
(11.) Gajadhar AA, Bisaillon JR, Appleyard GD. Status of
Trichinella spiralis in domestic swine and wild boar in Canada. Can J
Vet Res 1997;61(4):256-59.
(12.) Appleyard GD, Forbes LB, Gajadhar AA. National serologic
survey for trichinellosis in sows in Canada 1996-1997. Can Vet J
2002;43(4):271-73.
(13.) WHO. International Classification of Diseases. Manual of the
International Statistical Classification of Diseases, Injuries, and
Causes of Death, 9th Revision. Geneva, Switzerland: World Health
Organization, 1977.
(14.) WHO. International Classification of Diseases, 10th Revision.
Geneva: World Health Organization, 1992.
(15.) Statistics Canada 2002. Profile of Age and Sex, for Canada,
Provinces, Territories and Forward Sortation Areas, 2001 Census.
Catalogue no. 95F0486XCB2001003.
(16.) Parkinson AJ, Butler JC. Potential impacts of climate change
on infectious diseases in the Arctic. Int J Circumpolar Health
2005;64(5):478-86.
(17.) Kuhnlein HV, Receveur O. Local cultural animal food
contributes high levels of nutrients for arctic Canadian indigenous
adults and children. J Nutr 2007;137(4):1110-14.
(18.) Vuitton DA, Zhou H, Bresson-Hadni S, Wang Q, Piarroux M,
Raoul F, Giraudoux P. Epidemiology of alveolar echinococcosis with
particular reference to China and Europe. Parasitology
2003;127:S87-S107.
(19.) Craig PS, McManus DP, Lightowlers MW, Chabalgoity JA, Garcia
HH, Gavidia CM, et al. Prevention and control of cystic echinococcosis.
Lancet Infect Dis 2007;7(6):385-94.
(20.) Craig PS, Larrieu E. Control of cystic
echinococcosis/hydatidosis: 1863-2002. AdvParasitol 2006;61:443-508.
(21.) Schellenberg RS, Tan BJK, Irvine JD, Stockdale DR, Gajadhar
AA, Serhir B, et al. An outbreak of trichinellosis due to consumption of
bear meat infected with Trichinella nativa in 2 northern Saskatchewan
communities. J Infect Dis 2003;188(6):835-43.
Nicolas L. Gilbert, MSc, [1,2] Oluwayemisi K. Dare, PhD, [3]
Michael D. Libman, MD, [4] Pia K. Muchaal, MSc, [3] Nicholas H. Ogden,
DPhil [5]
Author Affiliations
[1.] Environmental Issues Division, Public Health Agency of Canada,
Ottawa, ON
[2.] Departement de medecine sociale et preventive, Universite de
Montreal, Montreal, QC
[3.] Surveillance and Targeted Studies Division, Public Health
Agency of Canada, Guelph, ON
[4.] Division of Infectious Diseases, McGill University Health
Centre, Montreal, QC
[5.] Zoonoses Division, Public Health Agency of Canada,
Saint-Hyacinthe, QC Correspondence: Nicolas Gilbert, Public Health
Agency of Canada, 200 Eglantine Driveway, Tunney's Pasture, Ottawa,
ON K1A 0K9, E-mail: nicolas.gilbert@phacaspc.gc.ca
Conflict of Interest: None to declare.
Table 1. Compatible Conditions for Trichinellosis and Echinococcosis
Infection Compatible Condition
Condition ICD-9 ICD-10
Trichinellosis (ICD-9 Diarrhoea and 009.1 A09
code 124, ICD-10 gastroenteritis of
code B75) presumed infectious
origin
Infective myocarditis 422.92 I40
Echinococcosis (ICD-9 Cholangitis 576.1 K83.0
code 122, ICD-10
code B67)
Note: only compatible conditions found in our cases are listed here.
Table 2. Echinococcosis Hospitalization by Latitude
Latitude * Population Echinococcosis
in 2001 Hospitalization
2001-2005
<50 [degrees]) 24,931,045 74
50 [degrees])-54 [degrees]) 4,518,930 24
>55 [degrees]) 556,785 10
Latitude * Crude Incidence Standardized RR
(per 1,000,000) Incidence
(per 1,000,000)
<50 [degrees]) 0.59 0.59 1.00
50 [degrees])-54 [degrees]) 1.06 1.08 1.83
>55 [degrees]) 3.59 2.87 4.88
Latitude * 95% CI
<50 [degrees])
50 [degrees])-54 [degrees]) 1.16-2.90
>55 [degrees]) 2.52-9.44
* At the centroid of the forward sortation area, i.e., the first three
digits of postal code (e.g., K1A).
Table 3. Trichinellosis Hospitalization: Nunavut and Northern Quebec
Compared to Rest of Canada
Region Population Trichinellosis
in 2001 Hospitalization
2001-2005
Nunavut and northern Quebec * 44,030 6
Rest of Canada 29,963,075 8
Region Crude Standardized
Incidence Incidence
(per 1,000,000) (per 1,000,000)
Nunavut and northern Quebec * 27.25 41.65
Rest of Canada 0.05 0.05
Region RR 95% CI
Nunavut and northern Quebec * 780 271-2248
Rest of Canada 1
* Northern Quebec is defined as postal code starting with J0M (which
includes Nunavik and James Bay).
