Abstract

Seven infants with persistent neonatal hyperinsulinism were treated in Dhahran Health Centre from 1983 to 1986. The insulin:glucose ratio (serum insulin concentration pmol/l) divided by the blood glucose concentration (mmol/l) ranged from 12 to 636, mean (SD) 177 (201). To control hypoglycaemia, diazoxide (12-24 mg/kg/day) was given in a continuous intravenous glucose infusion (12-22 mg/kg/min) on 11 separate occasions, four infants twice each and three infants once each. An increase of more than one standard deviation in the heart and respiratory rates, together with other symptoms of heart failure, was considered to be evidence of diazoxide toxicity. Cardiorespiratory failure (toxicity) occurred on eight of the 11 occasions (73%) in seven infants. The average daily fluid intake, weight change, respiratory rate and heart rate before treatment were similar whether or not the infant developed toxicity. A diazoxide toxicity index was obtained by multiplying the dose of diazoxide by the insulin:glucose ratio to relate the diazoxide dose to the severity of the disease. In all instances when the toxicity index was more than 1533 (mean (SD) 3732 (2741) cardiac toxicity developed. In contrast, infants with a toxicity index of less than 675 (mean (SD) 364 (270), had no symptoms of toxicity. Symptoms were significantly related to the severity of the disease and the diazoxide dose. It is possible to use the toxicity index to predict the risk of toxicity and to calculate a safe dose of diazoxide in infants with persistent neonatal hyperinsulinism.