Clinical characteristics and severity of diabetic ketoacidosis: A cross-sectional study from a tertiary hospital in Ghana

Objectives: Diabetic ketoacidosis (DKA) is a common, severe and often fatal complication of diabetes. This study aimed to investigate the clinical characteristics and precipitants of DKA, as well as factors associated with DKA severity in Ghanaian patients. Methods: Cross-sectional study of the medical records of all 70 adult patients >18 years managed for DKA in the adult emergency room of Korle-Bu Teaching Hospital in Ghana from March 2019 to July 2019. DKA diagnosis was based on hyperglycaemia >11.0 mmol/L, ketonuria (more than 2+) plus acidaemia of (pH < 7.3) or bicarbonate (HCO₃⁻) <15.0 mmol/L. However, when serum bicarbonate and pH were not available, clinical signs of acidosis, for example, Kussmaul breathing aided in the diagnosis. DKA severity was assessed based on the Joint British Diabetes Societies (JBDS) guidelines of factors suggestive of severe DKA. Multivariable logistic regression was used to determine the factors associated with DKA severity. Odds ratio and 95% confidence interval for factors associated with DKA severity were determined. Results: The mean (±standard deviation) age, diabetes duration and blood sugar at admission were 44.06 (±16.23) years, 7.19 (±6.04) years and 26.37 (±6.70) mmol/L, respectively. Females comprised 51.4% of the study population. The most common presenting symptoms were generalised weakness (30.0%) and fever (14.3%). The major precipitants were infection (70.0%) and non-compliance (22.9%). Overall, 71.4% of participants had features suggestive of severe DKA. In a multivariable regression model, Type 2 diabetes was associated with over fourfold decreased odds of severe DKA (OR 0.23, 95% CI [0.07–0.76], p = 0.016). Patient education on prevention of DKA was documented for only 18.6% of patients before being discharged. Conclusion: In this study, more than 70% of the study participants had features suggestive of severe DKA, with infection being the most common precipitant of DKA. 51.4% of patients had Type 2 diabetes which was associated with a statistically lower risk of severe DKA. Female sex tended to be positively associated with DKA severity. In a setting where the venous/arterial pH and bicarbonate levels may be inaccessible and/or unaffordable, using clinical features as found in the JBDS guidelines may help categorise patients and escalate care when needed. Indeed it may be useful to validate the use of the JBDS criteria for use in such settings.