Relationship between effective osmolality changes and neurological status during treatment for severe paediatric diabetic ketoacidosis

Introduction

Cerebral oedema is a common, life-threatening complication of paediatric diabetic ketoacidosis (DKA). Several risk factors at presentation are known, including urea, and pCO₂. It has recently been suggested that changes in effective osmolality during treatment may also convey risk, but this has not been confirmed due to a lack of prospective data.

Methods

A prospective observational study over 24 hours of 50 children admitted to a regional ICU with severe DKA (median pH at presentation 6.90). Development of cerebral oedema was defined clinically as a deterioration in Glaser score of >2 points during the 24 hours after presentation (equating to a Glasgow coma scale <8 to 10). Changes in effective osmolality, uncorrected and corrected sodium were modelled using linear (random coefficients) mixed models, with adjustment for baseline covariates of urea and pCO₂.

Results

Consistent with previous studies, baseline urea was higher and pCO₂ lower in patients who subsequently showed neurological deterioration. All three osmolality variables (effective osmolality, uncorrected and corrected sodium) were similar at baseline between the two groups. The best-fitting model utilized corrected sodium (lowest Akaike information criterion). This showed a significant interaction effect (P = 0.01), in that corrected sodium increased over time at a rate of 0.35 mmol/l/hour in patients who did not develop neurological symptoms, but did not change in those who did.

Conclusion

Lack of change in corrected sodium is associated with neurological deterioration in the treatment of severe DKA in children. This variable may be useful if incorporated into treatment guidelines.