I.A. van Asperen 1, T. Mank 2, G.J. Medema 1, C. Stijnen 3, A.S. de Boer 1, J.F. Groot 4, P. Ten Ham 5, J.F. Sluiters 6, M.W. Borgdorff 1
1 National Institute of Public Health and the Environment, The Netherlands
2 General Practitioners Laboratory, Haarlem, The Netherlands
3 Ruwaard van Putten Hospital, Spijkenisse, The Netherlands
4 Regional Health Department, Spijkenisse, The Netherlands
5 Regional Health Department, Haarlem, The Netherlands
6 University Hospital, Rotterdam, The Netherlands
A hospital microbiologist in Spijkenisse, in the south west of the Netherlands – who had recently attended a parasitology course – identified cryptosporidial oocysts in stools from a patient with diarrhoea on 16 August 1995. Re-examination of 89 stool specimens received since 10 August in which no bacterial pathogen had been identified revealed 15 cases of cryptosporidiosis, which were confirmed at the parasitology laboratory of the University of Leiden (Dr A.M. Polderman). Most cases were children aged £5 years and women aged 25-35 years, none of whom had evidence of impaired immunity.
Only parasitology departments in the Netherlands routinely examine stools for cryptosporidial oocysts and baseline data on the prevalence of cryptosporidiosis are sparse. Two surveys of patients with gastroenteritis suggest that 1% to 2% have cryptosporidiosis, with a peak of 3% to 4% in June and August (1, and personal communication L.M. Kortbeek, National Institute), much smaller proportions than the 17% in Spijkenisse in August 1995. On 31 August, the Dutch Medical Inspectorate commissioned the National Institute to investigate the source of the outbreak.
Investigation of the water supply
The public water supply of Spijkenisse was a potential source of the outbreak so the water supply was sampled and analysed (2) at two sites on 29 August (1500 litres at each site). Data on the water treatment processes and maintenance or repair procedures in the water distribution system were investigated for any disruptions/failures. A 3 mm filter of the demineralised water installation of the Spijkenisse hospital laboratory, which had been in place since June 1995, was also examined for cryptosporidial oocysts on 11 September. Data from the treatment plant provided no evidence of treatment failures or contamination after treatment, and samples from tap water and the filter were negative for cryptosporidium. No preventive or corrective measures were taken either at the treatment plant or to the consumer supply. The public was informed that the public water supply was safe.
Cryptosporidiosis incidence study
A study of the incidence of cryptosporidiosis was carried out in five laboratories in the south west and north west parts of the country. Cryptosporidial oocysts were detected in 147 of 1495 successive stool specimens from patients with gastroenteritis examined in September and October 1995 (10%, range 5-14%). The incidence declined by the end of September. The survey showed that the proportion of stool specimens from patients with gastroenteritis in which cryptosporidial oocysts were found had risen in all five regions of the country. The areas investigated received their water from different public water supplies.
Case control study
A case control study was carried out in the catchment areas of the Spijkenisse and Haarlem laboratories. Data from laboratory surveillance was used to find cases. A case of cryptosporidiosis was defined as a person who became ill with diarrhoea after 15 July 1995 and in whose stools cryptosporidial oocysts were detected between 4 and 26 September. Positive results were confirmed at the parasitology departments of the National Institute and the University Hospital of Rotterdam. Ten controls of the same sex and year of birth as each case were selected at random from local populations. It was planned that two controls, selected at random from the ten identified, should be interviewed for each case but in four cases only one control could be interviewed. Interviews were carried out by telephone using a standardised questionnaire, to obtain age, sex, details of illness including onset and duration, predisposing diseases, and exposure to recognised risk factors in the two to four weeks before the onset of diarrhoea. Information from matched controls was collected for the same calendar period as for the cases. Seventy-one cases, aged 0-65 years (median 5 years) included in the study were matched to 138 controls. All cases reported diarrhoea, 51 reported stomach cramps and weight loss. A common source for the outbreak was not identified. Using conditional logistic regression, household contact with people with diarrhoea and swimming in municipal pools (no particular pool implicated) were significantly associated with illness (odds ratios (OR) 5.4; 95% confidence interval (CI) 2.0-14.7 and OR 3.9; 95% CI 1.5-10.2, respectively). An apparent association with visiting day care centres in Haarlem did not reach statistical significance (OR 2.01; 95% CI 0.01-9.9). The median duration of illness was 25 days, 40% of cases rested in bed for a median of seven days, and half of the cases reported having taken a median of four days off work or school.
This is the first outbreak of cryptosporidiosis detected in the Netherlands. As routine investigation of faecal specimens in the Netherlands does not include cryptosporidium, previous increases in the incidence of cryptosporidiosis may not have been recognised. Loose stool specimens have been examined specifically for cryptosporidium in the Haarlem laboratory since January 1993, and an increased incidence has been observed in the Haarlem area every summer. The peak in 1995, however, was higher than in previous years. Our incidence study suggests that cryptosporidiosis may have a seasonal distribution nationwide. Bathing in swimming pools or surface waters may contribute to the observed seasonal distribution. Cryptosporidial oocysts are highly resistant to most common chemical disinfectants, including chlorine, and may persist in pool water for some time, if the pool is not well maintained, increasing the risk of transmission. Several outbreaks have been linked to swimming pools (3,4). Surface waters have been identified previously as a cause of cryptosporidiosis (5). Transmission of cryptosporidiosis through pool or surface water is facilitated by the low infective dose of cryptosporidium (6). We investigated two fresh water lakes and sea water in both areas and found cryptosporidial oocysts in some samples, but the case control study showed no increased risk associated with bathing in surface waters.
We recommend strengthening cryptosporidium surveillance through the routine examination of stools of patients with gastroenteritis in selected laboratories, to investigate seasonality in cryptosporidiosis and to facilitate prompt recognition of widespread outbreaks. Further studies on the role of swimming pools and surface waters in the transmission of cryptosporidiosis are needed (7).
1. Bänffer JRJ, Duifhuis JCC. Cryptosporidiose: prevalentie in de regio Rotterdam en vergelijking van twee kleuringstechnieken. Ned Tijdschr Geneeskd 1989: 133; 2229-33.
2. Le Chavallier, MW, Norton WD, Siegel JE, Abbaszadegan M. Evaluation of the immunofluorescence procedure for detection of Giardia cysts and Cryptosporidium oocysts in water. Appl Environ Microbiol 1995; 61: 690-7.
3. McAnully JM, Fleming DW, Gonzalez AH. A community-wide outbreak of cryptosporidiosis associated with swimming at a wave pool. JAMA 1994; 272: 1597-1600.
4.Sorvillo FJ, Fujioka K, Nahlen B, Tormey MP, Kebabjian R, Mascola L. Swimming-associated cryptosporidiosis. Am J Publ Health 1992; 82: 742-4.
5. Gallaher MM, Herndon JL, Nims LJ, Sterling CR, Grabowski DJ, Hull HF. Cryptosporidiosis and water. Am J Publ Health 1989; 79: 39-42.
6. DuPont HL, Chappell CL, Sterling CR, Okhuysen PC, Rose JB, Jakubowski W. The infectivity of Cryptosporidium parvum in healthy volunteers. N Engl J Med 1995; 332: 855-9.
7. Cryptosporidium in water supplies. UK Departments of Environment and Health. Report of a Group of Experts. Chaired by Sir John Badenoch. London: HMSO, 1990