Lesson of the Week: Katayama fever: an acute manifestation of schistosomiasis

Case reports

Sixteen patients were admitted to the Hospital for Tropical Diseases in London between August 1994 and December 1995 with suspected acute schistosomiasis. All had travelled to sub-Saharan Africa. Fourteen had been exposed to fresh water only in Lake Malawi; the other two, travelling together, had swum in a river in a rural area of Mozambique. Symptoms at presentation were non-specific: apart from a history of fever, the dominant symptom was profound lethargy. Only two patients recalled having “swimmers' itch,” a transient pruritic rash occurring within 12 hours of cercarial penetration. Symptoms began, on average, 36 days after exposure. Nine patients were febrile on admission and two more developed a fever during their hospital stay. Fourteen had eosinophilia; total eosinophil counts on admission ranged from 0.2 to 12.9×10⁹/l. Stool microscopy gave negative results in 12 patients—one had ova of Ascaris lumbricoides. Ova of Schistosoma haematobium were visible in the urine of another. Snips of rectal mucosa were examined for eggs in two cases; both were negative. An enzyme linked immunosorbent assay (ELISA) against soluble egg antigen of S mansoni was performed in each case.3 Two cases had positive serology, and two others were weakly positive.

In most cases the diagnosis of acute schistosomiasis was made on clinical grounds—a history of recent exposure to potentially infected water, physical findings, a negative blood film for malaria, and the presence of eosinophilia in a peripheral blood sample (table 1). Only two cases had unequivocal results. All 16 cases were treated with praziquantel 20 mg/kg body weight, twice daily for three days. Eight also received prednisolone 20 mg/day for three days. All patients subsequently developed evidence to support the diagnosis—15 developed antibodies to soluble egg antigen and one was found to have non-viable ova of S haematobium in the urine. The time taken to seroconversion averaged 1.6 months (range 0–6). Seven patients required further courses of praziquantel because of continuing symptoms, persisting eosinophilia, a subsequent rise in the antibody titre, or more than one of these. In several patients symptoms persisted many months after initial treatment (maximum one year).

Discussion

The number of cases of schistosomiasis imported into the United Kingdom, as with malaria, has increased over the past 20 years, principally because of increasing numbers of travellers to sub-Saharan Africa.4 Acute schistosomiasis occurs in a few infected people; this hospital diagnosed a total of 488 patients with either egg or antibody positive schistosomiasis between August 1994 and December 1995, suggesting that acute symptoms develop in only 3%. We have recently reported a series of 344 cases of confirmed (egg positive) schistosomiasis seen at this hospital between 1991 and 1994. Only two of that series developed symptoms compatible with a diagnosis of Katayama fever.4

In each case the presenting symptoms were nonspecific—principally fever, lethargy, and myalgia. Equally, physical examination did not reveal any specific features—mild enlargement of the liver and spleen, cough, and wheeze, nine patients having no abnormal physical findings. The severity of the symptoms varied considerably. Ten of these patients were severely incapacitated, while six had minimal symptoms. Many, if not all, might well have been diagnosed as having a viral infection once malaria had been excluded. Acute schistosomiasis, like viral illness, is a self limiting condition even if untreated, and follow up of patients diagnosed as having a viral illness of unknown cause is unusual. Many patients with established schistosomiasis are asymptomatic; the typical symptoms of diarrhoea, abdominal pain, or rectal bleeding (S mansoni or S japonicum) or terminal haematuria (S haematobium) may be absent or develop very slowly.1 However, without treatment, patients may develop chronic infection leading to portal hypertension with S mansoni or transitional cell carcinoma of the bladder with S haematobium.1 We have seen one patient who presented to this hospital with viable ova of S haematobium in the urine 33 years after his last exposure to infected water.5

Few patients with schistosomiasis develop symptoms suggestive of acute disease, and, as in this series, the diagnosis can be confirmed only retrospectively. We emphasise the importance of taking a clear travel history, particularly with regard to exposure to potentially infected water, and of appropriate referral of anyone suspected of having the disease. Praziquantel is the preferred treatment for patients with schistosomiasis, but it is not widely available. It is highly effective against adult schistosomes but less effective against the immature schistosomula; repeat courses may therefore be necessary. It may exacerbate the clinical condition in acute schistosomiasis, and concurrent administration of prednisolone has been advocated to reduce the risk of reactions.6 Prednisolone, however, may reduce the efficacy of praziquantel by reducing the serum concentration of the drug.7

We suggest that all patients with acute schistosomiasis should initially be treated with a three day course of praziquantel 20 mg/kg body weight, given twice daily under steroid cover (prednisolone 20 mg/day for three days). A second three day course of praziquantel should be given three to six months later to eradicate any schistosomes that may have survived the first course of treatment.

We thank our consultant colleagues for permission to report on their patients, the staff of the department of parasitology, Hospital for Tropical Diseases, for their help, and Robin Bailey for helpful comments.