Prospective European-wide multicentre study on a blood based real-time PCR for the diagnosis of acute schistosomiasis

Schistosomiasis is a parasitic infection with trematode flatworms of the genus Schistosoma. According to the World Health Organization (WHO) 200 million people in at least 76 countries are affected, most of them located in tropical and subtropical regions [1]. Most infections are attributable to three species: Schistosoma mansoni, S. haematobium, and S. japonicum[2]. The courses of human infections can be divided into acute and chronic forms. Acute schistosomiasis, also known as Katayama syndrome [3, 4], constitutes a fulminant allergic reaction with fever and eosinophilia and may be accompanied by skin, pulmonary, abdominal or even neurological symptoms. In general, these clinical courses are seen in patients experiencing their first Schistosoma infection and reflect an early immunological reaction against developing young worms during prepatency [5]. Accordingly, acute schistosomiasis is usually seen in non-immune travellers returning from Schistosoma endemic countries. Considering the broad differential diagnosis of fever and eosinophilia, an early and rapid diagnosis is required to circumvent unnecessary costly and time-consuming diagnostic procedures. However, acute schistosomiasis constitutes a diagnostic challenge as classical tools often fail to detect the parasite at this early stage of infection. Parasite microscopy following Katz-Kato smears or concentration techniques for helminth eggs, lack sensitivity, in particular, in individuals with a low parasite burden and during the prepatent period [6]. In settings with low transmission rates or in patients with low parasite burden immunological tests may be more sensitive than parasite microscopy [7]. Nevertheless, a substantial number of patients remain serologically negative at initial presentation [8, 9]. Moreover, the inability of serology to discriminate between active and past disease limits its clinical value for follow up investigations [10]. The detection of Schistosoma specific circulating anodic and cathodic antigens (CAA and CCA) in blood or urine, released by viable adult worms, has been tested in schistosomiasis endemic regions and sporadic experiences have been reported for the use in patients with acute schistosomiasis [11‐13]. Because the sensitivity of these tests greatly depends on the worm burden, they are probably not suited for the diagnosis of schistosomiasis in travellers with rather low worm burden.

Recently, we reported on the development of a new real-time polymerase chain reaction (PCR) test for the detection of Schistosoma specific DNA in blood of infected individuals [14], which revealed excellent performance for the diagnosis of acute schistosomiasis in a small group of travellers returning from Rwanda [15]. Here we report on a European-wide multicenter study to evaluate the use of this PCR for the diagnosis of acute schistosomiasis in larger number of patients returning from various destinations.

Thirty-eight patients fulfilled the case definition (Group A). Four of them had a history of possible previous exposure to Schistosoma-contaminated water but had not developed symptoms after the first trip. The median time from the day of possible exposure to the onset of clinical symptoms and the median time from exposure to PCR were 35 days (range 12–184 days) and 61 days (range 20–302 days), respectively. Twelve of the patients presented within 6 weeks after the suspected exposure, the time span generally referred to as prepatent stage of S. mansoni. The median absolute eosinophil count was 2,290 μl^-1 (range 870–33,050 μl^-1). The various members of the study population have been travelled to 17 different African and Arabic countries with endemic schistosomiasis. The median age of the study population was 28 years (range 19 – 66 yr) and 76% were males. Reasons for travelling included: tourism (48%; n = 19), occupational travel (23%; n = 9), volunteer (21%; n = 8) and visiting friends and relatives (8%; n = 3). The patients´ characteristics are summarized in Table 1.

Initial real-time PCR testing revealed positive detection of Schistosoma DNA in 35 patients. Of the remaining 3 patients with negative PCR results, the first one was part of a four patients cluster. He had an initial eosinophil count of 2,290 μl^-1 and a positive serology. The three other patients in this cluster were positive by means of serology and PCR. All received Praziquantel early in the course of the disease and subsequently, laboratory results normalized and clinical symptoms resolved. Accordingly, the negative PCR result in this case was considered as false negative. The second patient with a negative PCR test result was considered true negative, since microscopy and serology remained negative during follow up. The third patient with a negative PCR test result provided only a small amount of serum for PCR testing of less than 2 ml. In favour of a conservative estimation we interpreted this result as false negative because serology was clearly positive and symptoms resolved after Praziquantel treatment. Thus blood based real-time PCR was able to detect acute schistosomiasis in 35 of 37 true positive patients (95%). In contrast, anti-schistosoma antibodies or Schistosoma eggs were detected in only 72% and 25%, respectively, of these patients.

For 8 patients follow-up data were available. Four patients presented with negative real-time PCR at their follow up visit. The remaining 4 demonstrated a significant decrease in Ct-values. Median follow-up times of patients with positive PCR test results versus patients with negative PCR test results were 6 and 18 months respectively (data not shown).

In addition, 17 patients (Group B) with high clinical suspect for acute schistosomiasis but without an episode of fever were tested (Table 2). The incubation time and the time from exposure to diagnostic work up was 38 days (range 14–93 days) and 64 days (range 17–96 days), respectively. Sixteen patients were males. The median age was 13 years (range: 6 – 60 yr) and the median eosinophil count was 1,960 μl^-1 (range 1,040–14,270 μl^-1). The decision for testing these patients was based on high eosinophil counts, a history of recent travelling to a schistosomiasis endemic region and at least one of the following symptoms: fatigue, gastro-intestinal symptoms, rash or exanthema. Blood based real-time PCR testing was negative in 5 of these patients. In these 5 patients, initial serological testing revealed a positive result in one patient and inconclusive results in two patients. On follow up visits, all of them were serologically negative. Microscopic examination of stool samples was negative in all patients at initial presentation and during follow up. Of the remaining 12 patients, one was returning from Togo with positive serology and microscopy and the other 11 were part of a travel group returning from Rwanda. In this cluster reported earlier [15], PCR was positive in all of them, whereas positive serology or microscopy at initial testing was detected in only 7 of them.

To our knowledge this is the first multicenter study on acute schistosomiasis in travellers residing in non-endemic countries. Moreover, it constitutes the largest prospective study on a blood based real-time PCR for the diagnosis of the disease. The high rate of positive PCR results within the study population may be in part attributed to the fact that the participating centres are well familiar with this otherwise rare disease of acute schistosomiasis as all centres are specialized in travel and tropical medicine.

The results presented here support previous findings that PCR for the detection of Schistosoma DNA in serum outperforms other diagnostic test such as serology or microscopy in the early phase of the infection, particular in the prepatent stage of human schistosomiasis [14], as demonstrated by the twelve patients presenting within 42 days after the suspected exposure. By PCR, a correct diagnosis was made in 95% of patients, despite the heterogeneous characteristics of the study population (17 endemic countries - cluster and single patients), which underlines the robustness of blood base real-time PCR testing for the diagnosis of acute schistosomiasis. In addition, PCR was able to exclude active Schistosoma infections in six patients, of whom 50% were suspected due to positive serology (1 positive, 2 inconclusive). Thus, the results support previous observations that the PCR test system used in this study is highly sensitive and specific [14]. Accordingly, this test system has substantial added value for an early and rapid decision making to circumvent cost-intensive and time-consuming diagnostic procedures and to prevent unnecessary therapies. Cost for one PCR test (materials and reagents) are roughly 2 US$, thus the test is a useful addition in the differential diagnosis in patients with fever and eosinophilia after travelling to schistosomiasis endemic countries but is not generally recommended as a screening assay for returning travelers.

A considerable disadvantage of classical Schistosoma diagnostics is the persistence of serologic markers and egg shedding after successful treatment. Previous studies in experimentally infected mice [14] suggested that blood based real-time PCR testing might be suitable not only for initial diagnosis but also for follow up monitoring. This is further supported by our findings in 8 patients, for whom follow up data were available. These results suggest a significant decline in Ct-values in response to treatment. Clearly negative PCRs are expected after a longer period of time due to the high sensitivity of PCR and the carry over of DNA from slowly degenerating eggs in the tissue. However, for a definite conclusion on the usefulness of PCR for monitoring of schistosomiasis treatment the follow-up of a larger number of patients is required.

The highly repetitive 121 bp DNA fragment used as target sequence in the real-time PCR assay constitutes about 12% of the S. mansoni genome [20]. Although this sequence is detected in S. haematobium and S. japonicum as well [20, 21] sensitivity of PCR for the latter two species might be reduced. As far as species could be determined by microscopic egg morphology, the vast majority of cases in this study were due to infections by S. mansoni. Thus, our results may not allow generalization to infections with Schistosoma species other than S. mansoni. Evaluation of blood based PCR for the diagnosis of acute schistosomiasis due to other Schistosoma species such as S. haematobium or S. japonicum requires further studies. Previous results with the assay used in this study indicated that the performance of the PCR using the 121 bp tandem target sequence on 2 ml serum samples works very well in S. mansoni infection, but not in S. haematobium and in S. mekongi infection (Jan Clerinx, personal communication). Likewise, additional studies are required to evaluate the usefulness of blood-based PCR for the diagnosis of chronic schistosomiasis, in particular, in suspected cases from endemic countries, with a positive serology but negative parasite microscopy. Moreover, recent studies on, PCR performed with urine or fecal samples to detect S. haematobium or S. mansoni infection have shown rather promising results [22‐25]. However, further studies are required to determine, whether urine and/or feces PCR is sensitive to detect early schistosoma infection as it is the case in patients with acute schistosomiasis.

For the early diagnosis of acute schistosomiasis, real-time PCR for the detection of Schistosoma DNA in serum is more sensitive than classic diagnostic tools such as serology or microscopy, irrespective of the region of infection. Generalization of the results to all Schistosoma species may be difficult as in the study presented here only eggs of S. mansoni were detected by microscopy. A minimum amount of two millilitre of serum is required for sufficient diagnostic accuracy.