Objectives: To describe the epidemiology, clinical features, management and outcome of children with burns admitted to a rural district hospital.
Design: A retrospective analysis of the case notes of consecutive cases of paediatric burns.
Setting: Hlabisa Hospital, KwaZulu-Natal-a 450-bed rural district hospital serving approximately 200,000 people.
Subjects: All cases of paediatric burns (age < 12 years) admitted to Hlabisa Hospital in 1994.
Main outcome measures: Number of admissions, month of admission, age, sex, time to presentation, site of burn, complications, number of surgical procedures, adherence to management protocol and outcome.
Results: One hundred and forty-nine children presented to the outpatient department in 1994 and 88 (59%) were admitted. The median age of those admitted was 36 months with 66 (75%) aged less than 5 years; 42 (48%) were boys. Thirty-nine children (44%) were admitted during the four winter months of May to August. The average interval from the time of the burn to presentation at hospital was 42 hours (range 1-120). Sixty-eight burns (77%) were due to hot fluid or food burning the legs, trunk or arms. There was a high level of morbidity. Nineteen (22%) children developed wound infections, 5 (6%) developed contractures and 20 (23%) required a total of 32 surgical procedures. There was 1 death. Burns were responsible for more paediatric patient days spent in hospital than any condition other than malnutrition, and a longer length of stay was associated with delay in presentation. Children presenting within 24 hours of the burn had a mean length of stay of 12.8 days, compared with 25.2 days (P = 0.03) for children presenting 24 hours or more after the burn. Twenty of the 22 children who stayed for longer than 3 weeks or who required transfer were judged to have been managed inadequately in at least one respect compared with 3 of 48 who were discharged within 2 weeks or not transferred (P < 0.001).
Conclusion: This study shows that paediatric burns are an important cause of morbidity and contribute significantly to inpatient stay in this rural setting. The lengthy delay from time of burn to presentation at hospital is of serious concern and our results show that this delay is associated with increased hospital stay. As most burns were due to spillage of hot fluids or food there seems to be significant potential for preventive interventions. Community-based studies would help to estimate the true incidence of burns and would contribute to an understanding of the reasons for delay in presentation. The information gathered is being used to inform the development of a burns prevention programme.