Predictors and treatment outcome of hyperglycemic emergencies at Jimma University Specialized Hospital, southwest Ethiopia

Predictors and treatment outcome of hyperglycemic emergencies at Jimma University Specialized Hospital, southwest Ethiopia

BMC Research Notes, October 2015

26455633

Tigestu Alemu Desse, Tesfahun Chanie Eshetie, Esayas Kebede Gudina

Diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS) commonly known as hyperglycemic emergencies are the two most common life-threatening acute metabolic complications of diabetes. The objective of this study is to assess predictors and treatment outcome of hyperglycemic emergencies (HEs) among diabetic patients admitted to Jimma University Specialized Hospital (JUSH). It is a three year retrospective review of medical records of patients admitted with HEs at JUSH. Patient demographics, admission clinical characteristics, precipitants, insulin used and treatment outcomes were extracted. Statistical analysis was done using student's t test, Chi square test, and binary logistic regression with level of α set at 0.05. Statistical significance was considered for variables with p < 0.05. Complete data was available for 163 out of 421 patients admitted with HEs. The majority (62.6 %) were males. Mean age of patients was 36.6 ± 15.9 years. About 64 % of patients had type 1 diabetes. About 93 % of the participants developed DKA. The most common precipitants of HEs were infections 95 (59 %), non-compliance to medications 52 (32.3 %), and newly diagnosed diabetes 38 (23.6 %). Recurrent hyperglycemia, hypoglycemia and ketonuria occurred in 88 (54 %), 34 (20.9 %) and 31 (20.5 %) patients respectively. Mean amount of insulin used and duration of treatment till resolution of DKA were 136.85 ± 152.41 units and 64.38 ± 76.34 h respectively. The median length of hospital stay was 6 days. Mortality from HEs was 16 (9.8 %). Admission serum creatinine >1.2 mg/dL (P = 0.018), co-morbidity (P < 0.001) and sepsis (P = 0.014) were independent predictors of HEs mortality. Infections, non-compliance and new onset diabetes were the most common precipitants of HEs. Length of hospital stay and mortality were high. High use of insulin, recurrent hyperglycemia, hypoglycemia, and ketonuria were common during HEs management. Elevated serum creatinine, sepsis and co-morbidity are independent predictors of HEs mortality.