Cryptosporidium infections in Denmark, 2010-2014

Cryptosporidiosis is an infection caused by single-celled parasites of the Cryptosporidium genus. Infected patients may present with watery diarrhoea lasting for up to weeks [1, 2]. Even in otherwise healthy individuals, pronounced symptoms may arise. The parasites are primarily transmitted through contaminated drinking or recreational water; however, infection may occasionally arise through contaminated food or contact to animals or infected individuals [3, 4].

1-10 oocysts (spores) are sufficient to establish infection. The cryptosporidia invade and multiply in the gut epithelium, which is usually accompanied by development of symptoms; oocysts are shed in the faeces throughout this period. The longevity of oocyst shedding varies. Oocysts are environmentally resilient and may survive for days even in chlorinated water; meanwhile, ultraviolet (UV) irradiation and boiling of water is efficient in terms of limiting waterborne transmission.

Cryptosporidiosis is self-limiting within 1-3 weeks in otherwise healthy individuals, whereas immunosuppressed patients may develop chronic diarrhoea with prolonged oocyst shedding. An effective treatment regimen remains to be identified, although some effect of nitazoxanide, a nitrothiazolyl-salicylamide derivative, has been reported [5, 6].

On a global scale, cryptosporidia are a frequent cause of outbreaks. In Milwaukee, USA, an outbreak in 1993 caused by contaminated drinking water affected more than 400,000 people [7]. In 2010, approximately 27,000 people became infected in Östersund, also through contaminated drinking water [8], and waterborne outbreaks are quite common in Sweden [4, 8-11]. In Sub-Saharan Africa and other regions, cryptosporidiosis is one of the most common and serious causes of
diarrhoea-related morbidity and mortality in infants and toddlers [12].

The first known outbreak in Denmark was recorded at Hvidovre Hospital in 1989 [13], and in 2005 an outbreak was described that affected employees in a large Danish company due to oocyst-contaminated raw carrots served in the company’s canteen [3]. Since then, minor outbreaks have been described related to the handling of calves, a significant reservoir of Cryptosporidium parvum infection. C. parvum is possibly the species most commonly associated with cryptosporidiosis in humans; another common species is primate-adapted C. hominis. Zoonotic transmission involving other species, including C. meleagridis, C. canis, and C. felis, is seen occasionally [9, 14]. C. parvum infection is associated with more severe symptoms than cases of cryptosporidiosis caused by C. hominis [9]. Multiple subtypes and subtype variants have been described for several species, and molecular typing is critical to successful outbreak investigations.

Contrary to the situation in our neighbouring countries, no national surveillance of cryptosporidiosis exists in Denmark. Hence, the incidence of the infection is unknown, and no guidelines as to which patients should be suspected of and tested for cryptosporidiosis are avail-able. The aim of this article is to increase awareness of cryptosporidia as a cause of diarrhoea. Based on currently available national data, we summarised the number of cases detected between January 2010 and April 2014, and generated preliminary data that indicate which species account for cryptosporidiosis detected in Denmark. By introducing sensitive diagnostic methods and by increasing the focus on cryptosporidia as a potential cause of diarrhoea, a more complete and accurate picture of the incidence and transmission patterns will be revealed.

METHODS

Data from the Danish Microbiology Database and molecular typing

Data were retrieved from the Danish Microbiology Database (MiBa) [15, 16], which receives laboratory test results from all microbiology departments in Denmark. Data comprised test results on cryptosporidia detected by microscopy or polymerase chain reaction (PCR) between 1 January 2010 and 30 April 2014. As the national reference laboratory for parasitology at Statens Serum Institut (SSI) did not comply with the national standard protocol for transferring electronic microbiological reports to MiBa during the first two years of the study, data from this laboratory were extracted directly from the SSI database and merged with the MiBa data set to obtain complete nationwide data.

Samples identified as positive by the SSI were furthermore submitted to species and subtype analysis by conventional PCR and sequencing of ribosomal (SSU rRNA) and gp60 genes, respectively, as assays for gp60-based subtyping are now available for the three most common species, C. parvum, C. hominis and C. meleagridis [17]. Data were standardised and analysed using Stata v. 13. Results are indicated as number of cases, a case-episode being defined as an individual with one or more positive stool samples, excluding multiple positive test results obtained within a 60-day period.

Trial registration: not relevant.

RESULTS

Incidence and distribution

From January 2010 to April 2014, a total of 689 Cryptosporidium-positive stool samples had been submitted by 387 patients, several patients testing positive more than once. Limiting case-episodes to two months (60 days), only one patient had more than one episode. Hence, a total of 388 case-episodes representing 387 patients were identified, including 210 episodes in females and 178 episodes in males. Analysing the patient age distribution (Figure 1), it appears that cases of cryptosporidiosis were most common among infants and toddlers. Moreover, a peak in incidence was observed among younger adults aged 23-24 years. In older adults and seniors, only few positive samples were seen.

Among children less than five years old, boys were more prone to developing infection than girls, whereas in adolescents and younger adults, more cases of cryptosporidiosis were found among females than males (Figure 2). In terms of seasonal variation, there was a general trend towards identifying more cases in late summer and autumn, which is in agreement with observations from other countries [18].

GP60 data

For 43 Cryptosporidium-positive faecal samples analysed from 2010 to 2014, parasites were identified to species and subtype level. C. parvum was found in 34 samples, C. hominis in eight, and C. meleagridis was identified in a sample from a patient with a recent history of kidney transplantation. With regard to C. parvum, we observed a relatively high number of two subtypes that are common in calves in Scandinavia, IIaA15G2R1 (n = 10) and IIaA16G3R1 (n = 5). Some of the affected individuals were veterinarians who had been handling calves prior to developing the infection.

DISCUSSION

Diagnostic considerations

Traditional parasitology uses microscopy of Ziehl-Neelsen (ZN)-stained smears of faecal concentrates to detect cryptosporidia or commercial kits based on antigen detection. In Denmark, DNA-based methods are being introduced in several routine clinical microbiology laboratories for detection of parasites such as Cryptosporidium spp. and Giardia intestinalis. A study including 889 prospective and random faecal samples identified a total of 16 cases of cryptosporidiosis by real-time PCR, while no cases were detected by microscopy of ZN-stained faecal smears [19]. At SSI, a genus-specific real-time PCR is employed to enable the detection of species other than C. hominis and C. parvum. Even when real-time PCR is used, one of two samples from an infected patient may test negative, and it is recommended to test at least two samples to reduce the risk of overlooking cases.

In Denmark, no national guidelines are available regarding the referral of samples for Cryptosporidium testing. In Halland, Sweden, PCR has been implemented in the routine screening of patients with diarrhoea regardless of type of diarrhoea and region of exposure (domestic or travel-related). Interestingly, Halland has the highest incidence of cryptosporidiosis seen throughout Sweden, 17/100,000 (2013; [20]). Meanwhile, in Denmark, testing for cryptosporidiosis is primarily performed in cases of travel-associated or persisting diarrhoea, or in cases of diarrhoea in immunocompromised individuals. Otherwise healthy individuals with acute
diarhoea and no history of travelling are therefore not tested for cryptosporidiosis on a routine basis. Preliminary research data from the SSI indicate that the incidence of cryptosporidiosis in patients with acute
diarrhoea acquired in Denmark is comparable to the incidence among patients suspected of cryptosporidiosis, including patients with chronic and/or travel-associated diarrhoea. Based on observations from Halland and the SSI, ongoing studies will address whether testing for Cryptosporidium should be implemented consistently in the routine screening of stool samples from patients with diarrhoea acquired in Denmark.

The samples that were subtyped had been collected over a period of five years. Therefore, the probability of an unrecognised outbreak of, e.g., IIaA15G2R1 is small, though still possible.

Six out of eight cases of C. hominis infection were registered in the winter season. In Sweden, the vast majority of C. hominis cases are imported [9].

The typing data may indicate that mostly sporadic cases or cases related to exposure to calves are detected in Denmark. A few cases of travel-associated C. hominis are seen, and cases in which the disease is particularly debilitating (as in infants and toddlers) are likely to be identified, which again may explain the relatively higher incidence among infants and toddlers. However, since cryptosporidiosis is probably under-diagnosed, these data should be interpreted with caution.

Moving towards surveillance

For several reasons, cryptosporidiosis is likely underdiagnosed in Denmark. Most often cryptosporidiosis is a self-limiting disease with only few patients seeking medical assistance. Furthermore, there is often a low index of suspicion of cryptosporidiosis and specific examination of faeces for Cryptosporidium species is rarely requested. Finally, some methods currently in use for the detection of cryptosporidia in patient samples have only a low or moderate sensitivity. The disease burden may therefore be heavier than anticipated, and the chances of detecting outbreaks are not optimal.

It is not very likely that the risk of waterborne outbreaks is as high in Denmark as in countries such as Sweden and England. In Denmark, surface water is used as a source of drinking water only to a very limited extent, which reduces the risk of domestic zoonotic transmission. However, the general lack of awareness in Denmark explains why the outbreak in 2005 was detected late [3], since testing for cryptosporidia was performed only after the exclusion of bacteria and virus as potential outbreak causes.

Since sensitive DNA-based tests are currently being introduced into routine diagnostic clinical microbiology laboratories in Denmark, it is now relevant to implement national surveillance. MiBa was established in 2010 by combined efforts of the regional clinical microbiology laboratories and the SSI with a view to developing a state-of-the-art digital surveillance system. If general awareness regarding cryptosporidia as a potential cause of diarrhoea is increased, it will be possible to efficiently monitor infections in real-time, which will enable fast and targeted action to contain and manage suspected outbreaks as a cause of exposure to contaminated water and foods, or infected animals.

CONCLUSION

In Denmark as well as in other countries, cryptosporidia constitute an important pathogen causing diarrhoeal disease in otherwise healthy children and adults. Outbreaks are frequently seen in our neighbouring countries and are also likely to occur in Denmark although probably less frequently. However, outbreaks, if they exist, would likely not be easily recognised because of lack of awareness, lack of national guidelines on testing and because of the continued use of diagnostic methods with limited sensitivity. As more sensitive methods are being introduced in diagnostic laboratories, we propose establishing national surveillance of cryptosporidiosis.

Correspondence: Christen Rune Stensvold, Infektionsepidemiologi, Statens Serum Institut, Artillerivej 5, 2300 Copenhagen S, Denmark. E-mail: run@ssi.dk

Accepted: 25 March 2015

Conflicts of interest:Disclosure forms provided by the authors are available with the full text of this article at www.danmedj.dk