Antibiotic use and infection in snakebite victims.
OBJECTIVES: To determine the incidence of infection in snakebite patients, the bacterial species involved, and the indication for antibiotics. METHOD: A prospective trial was undertaken at Eshowe Hospital, KwaZulu-Natal, involving 363 snakebite patients (records available for 310 patients). It was protocol not to give antibiotics unless necrosis was present or anticipated. Bacterial species were identified from necrotic areas and abscesses. A swelling classification was devised. RESULTS: The syndromic presentation included no clinical envenomation (12%), painful swelling (85%), weakness (1.8%), venom ophthalmia (0.6%) and other presentations (0.6%). There was 1 death, of an 11-year-old with gross swelling and thrombocytopenia. 15.2% of patients received antibiotics. Protocol was broken in 23 patients. None of the patients who did not receive an antibiotic became infected, and hospital stay was not prolonged in this group. Of 20 bacteria isolated, 18 were Gram-negative aerobic enterobacteriaceae. No anaerobes were cultured. CONCLUSIONS: Antibiotics should be reserved for those snakebite patients with necrosis (< 10%) and should cover Gram-negative aerobic bacilli and Gram-positive aerobic cocci.