Aim: To compare the fluid management of hypernatraemic dehydration in acute gastroenteritis in those who developed cerebral oedema (cases) versus those who did not (controls).
Methods: A retrospective study of 97 cases of hypernatraemic dehydration at a tertiary children's hospital in China over five years, in which rehydration regimes of 49 children who developed cerebral oedema were compared with 48 children who made an uneventful recovery.
Results: Risk factors for cerebral oedema (vs. no cerebral oedema) were an initial fluid bolus (29/49 vs. 15/48, P = 0.006), the mean rate of bolus infusion (14.7 ± 2.2 vs. 10.8 ± 1.4 mL/kg/hr, P < 0.001), the severity of hypernatraemia (serum sodium 167.7 ± 7.8 vs. 161.3 ± 7.9 mmol/L, P < 0.001) and the overall rehydration rate (8.2 ± 1.1 vs. 6.4 ± 0.6 mL/kg/hr, P < 0.001). On logistic regression, a rapid rehydration rate was the most significant contributor to cerebral oedema. From receive operating characteristic (ROC) curve analysis, the safe rate of rehydration was <6.8 mL/kg/hr.
Conclusion: The key risk factors for the development of cerebral oedema during recovery from hypernatraemic dehydration were too rapid a rate of rehydration, an initial fluid bolus to rapidly expand plasma volume and the severity of the hypernatraemia. Thus, we conclude that a uniformly slow rate of rehydration is the best way of preventing cerebral oedema.