J. Pu, X. Zhou, R. Ullah, G. Dong, W. Wu, K. Huang, X. Chen, et al. (2022). Optimized simplified pediatric diabetes severity warning system for the early identification of diabetic ketoacidosis in children. Pediatric diabetes, 23(5), 569-577. https://doi.org/10.1111/pedi.13345

PubMed 35419919

[Abstract]
OBJECTIVE: Diabetic ketoacidosis (DKA) is the leading cause of mortality in children with type 1 diabetes. Diagnosis of DKA is difficult in resource-limited areas owing to the unavailability of blood gas test, the gold standard for DKA diagnosis. The Simplified Pediatric Diabetes Severity Warning System (SPDSWS) has been developed to identify high-risk DKA patients with limited resources in China. Here we optimized and validated this system.

METHODS: This study included 835 children admitted between January 2011 and June 2020 with the principal diagnosis of type 1 diabetes. Data were collected based on demographic and clinical characteristics. DKA and its severity were defined according to the criteria of ISPAD. SPDSWS was optimized based on logistic regression analyses and then was validated in a validation cohort.

RESULTS: The 20-point optimized SPDSWS included strong positive urine ketone, young age, dehydration, fatigue, anorexia, vomiting, abdominal pain, abnormal pulse, and high blood glucose. The optimized SPDSWS predicted DKA with an AUC value of 0.882 in the derivation cohort. When the cut-point score ≥7 was used, the sensitivity and specificity were 75.5% and 86.0%, respectively, in the derivation cohort and were 90.0% and 85.8%, respectively, in the validation cohort. The optimized SPDSWS also predicted the moderate/severe DKA with an AUC value of 0.911 in the derivation cohort and 0.937 in the validation cohort. A score > 11 was associated with an extremely high incidence of DKA.

CONCLUSIONS: The optimized SPDSWS could assist health care practitioners in underdeveloped remote areas to identify the children at high risk of DKA as early as on admission.