A prospective cohort study of the effectiveness of the primary hospital management of all snakebites in Kurunegala district of Sri Lanka

INTRODUCTION Sri Lanka records substantial numbers of snakebite annually. Primary rural hospitals are important contributors to health care. Health care planning requires a more detailed understanding of snakebite within this part of the health system. This study reports the management and epidemiology of all hospitalised snakebite in the Kurunegala district in Sri Lanka. METHODOLOGY The district has 44 peripheral/primary hospitals and a tertiary care hospital-Teaching Hospital, Kurunegala (THK). This prospective study was conducted over one year. All hospitals received copies of the current national guidelines on snakebite management. Clinical and demographic details of all snakebite admissions to primary hospitals were recorded by field researchers and validated by comparing with scanned copies of the medical record. Management including hospital transfers was independently assessed against the national guidelines recommendation. Population rates were calculated and compared with estimates derived from recent community based surveys. RESULTS There were 2186 admissions of snakebites and no deaths in primary hospitals. An additional 401 patients from the district were admitted directly to the teaching hospital, 2 deaths were recorded in this group. The population incidence of hospitalized snakebite was 158/100,000 which was significantly lower than community survey estimates of 499/100,000. However there was no significant difference between the incidence of envenomation of 126/100,000 in hospitalised patients and 184/100,000 in the community survey. The utilisation of antivenom was appropriate and consistent with guidelines. Seventy patients received antivenom. Anaphylactic reactions to antivenom occurred in 22 patients, treatment reactions was considered to be outside the guidelines in 5 patients. Transfers from the primary hospital occurred in 399(18%) patients but the majority (341) did not meet the guideline criteria. A snake was identified in 978 cases; venomous snakebites included 823 hump-nosed viper (Hypnalespp), 61 Russell's viper, 14 cobra, 13 common krait, 03 saw scaled viper. CONCLUSIONS Primary hospitals received a significant number of snakebites that would be missed in surveys conducted in tertiary hospitals. Adherence to guidelines was good for the use of antivenom but not for hospital transfer or treatment of anaphylaxis. The large difference in snakebite incidence between community and hospital studies could possibly be due to non-envenomed patients not presenting. As the majority of snakebite management occurs in primary hospitals education and clinical support should be focused on that part of the health system.