Rotavirus infections in paediatric oncology patients: a matched-pairs analysis.

OBJECTIVE

To conduct a systematic investigation of the clinical relevance of rotavirus infection in the setting of paediatric cancer patients receiving intensive chemotherapy.

MATERIAL AND METHODS

Twenty-eight paediatric cancer patients with positive rotavirus antigen tests were eligible for a retrospective case-control study (January 1995-December 2004). Rota-positive patients were compared with 28 rota-negative patients matched for age, underlying disease and chemotherapy. The National Cancer Institute Common Toxicity Criteria were used to determine clinical severity.

RESULTS

Median duration of rota-related symptoms (diarrhoea, fever and vomiting) was 7 days (range 4-34 days; 75th percentile 9 days). Median duration of viral shedding was 17 days (4-73 days; 75th percentile 39.5 days). The rota infection was nosocomially acquired in 19 patients (68%). The proportions of patients with diarrhoea > or =NCI II, fever >39 degrees C, clinically relevant dehydration, metabolic acidosis, mucositis and neutropenia were significantly higher in rota-positive patients. Rota-positive patients tended to have a prolonged period of hospitalization (median 8 versus 4 days; p=0.008). A higher proportion of rota-positive patients had to receive parenteral nutrition and tube feeding (p<0.001).

CONCLUSIONS

Rotavirus is a clinically relevant but preventable pathogen in paediatric cancer patients, since many cases seem to be nosocomial in origin. Rapid microbiological testing and contact precautions should be strictly applied to any symptomatic patient and to their immediate contacts. Prolonged viral shedding in immunocompromised paediatric patients necessitates repeated testing in order to determine the duration of isolation.