PT - JOURNAL ARTICLE AU - Karande, S AU - Bhatt, M AU - Kelkar, A AU - Kulkarni, M AU - De, A AU - Varaiya, A TI - An observational study to detect leptospirosis in Mumbai, India, 2000 AID - 10.1136/adc.88.12.1070 DP - 2003 Dec 01 TA - Archives of Disease in Childhood PG - 1070--1075 VI - 88 IP - 12 4099 - http://adc.bmj.com/content/88/12/1070.short 4100 - http://adc.bmj.com/content/88/12/1070.full SO - Arch Dis Child2003 Dec 01; 88 AB - Background: Leptospirosis is relatively uncommon in children. Following torrential rains and flooding an outbreak of leptospirosis was suspected in Mumbai. Aims: To investigate the possibility of an outbreak of leptospirosis and describe the clinical illness. Methods: From 24 July to 14 September 2000, children with a history of abrupt onset of high fever (>39°C), who presented to our hospital, were admitted and tested serologically for anti-Leptospira antibodies by a quantitative enzyme linked immunosorbent assay (ELISA) test. An IgM titre of more than 20U/ml confirmed the diagnosis of leptospirosis. Clinical features in the confirmed leptospirosis and leptospirosis negative groups were analysed. Results: Of 53 children screened, 18 (34%) had leptospirosis. In all 18, the disease was anicteric and responded well to intravenous penicillin. Four clinical features present at the time of admission were significantly associated with leptospirosis: a history of contact with flood water (18/18 v 16/35), conjunctival suffusion (5/18 v 1/35), abdominal pain (9/18 v 5/35), and skin rash (5/18 v 1/35). As the number of these four features concomitantly present increased, the chances of the child having leptospirosis also increased significantly. A history of contact with flood water had a sensitivity of 100%, and the presence of conjunctival suffusion, abdominal pain, and skin rash had a specificity of 97%, 86%, and 97%, respectively, for identifying children with leptospirosis. Conclusion: Leptospirosis should be suspected in febrile children with contact with flood water.