BACKGROUND: Previous studies of the management of diabetic ketoacidosis have noted a wide range of incidence of hypoglycemia but have not studied the risk factors associated with it. METHODS: To describe the incidence of hypoglycemia in patients hospitalized with diabetic ketoacidosis, we retrospectively reviewed the charts of all adult patients with the diagnosis of diabetic ketoacidosis at three private, community hospitals in Milwaukee, Wis, between January 1, 1987, and May 31, 1990. Two hundred twenty admissions in 150 patients met our inclusion criteria. RESULTS: In 30% (66/220) of cases of diabetic ketoacidosis, a serum glucose level or Accu-Chek (Boehringer-Mannheim, Indianapolis, Ind) finding was 2.7 mmol/L or less during the first 14 days of hospitalization. No factors could be identified that were associated with a significantly increased risk of early hypoglycemia (within the first 48 hours of admission). The risk of a "late" occurrence of hypoglycemia (after 48 hours of hospitalization) was increased by fever (relative risk, 2.05; 95% confidence interval [CI], 1.16 to 3.63), "nothing orally" status (relative risk, 3.01; 95% CI, 1.88 to 4.83), hepatic disease (relative risk, 2.56; 95% CI, 1.39 to 4.70), and renal disease (relative risk, 2.07; 95% CI, 1.26 to 3.39). A logistic regression analysis showed "nothing orally" status to be associated with an increased risk of any hypoglycemia occurring during the hospitalization (relative risk, 2.39; 95% CI, 1.63 to 3.51). Physicians and nurses documented the first episode of hypoglycemia in their notes 45.5% and 80.3% of the time, respectively. CONCLUSION: Hypoglycemia is still a common complication of diabetic ketoacidosis, is associated with hepatic and renal disease as well as fever and "nothing orally" status, and is not documented well in physician notes.
BACKGROUND: Previous studies of the management of diabetic ketoacidosis have noted a wide range of incidence of hypoglycemia but have not studied the risk factors associated with it. METHODS: To describe the incidence of hypoglycemia in patients hospitalized with diabetic ketoacidosis, we retrospectively reviewed the charts of all adult patients with the diagnosis of diabetic ketoacidosis at three private, community hospitals in Milwaukee, Wis, between January 1, 1987, and May 31, 1990. Two hundred twenty admissions in 150 patients met our inclusion criteria. RESULTS: In 30% (66/220) of cases of diabetic ketoacidosis, a serum glucose level or Accu-Chek (Boehringer-Mannheim, Indianapolis, Ind) finding was 2.7 mmol/L or less during the first 14 days of hospitalization. No factors could be identified that were associated with a significantly increased risk of early hypoglycemia (within the first 48 hours of admission). The risk of a "late" occurrence of hypoglycemia (after 48 hours of hospitalization) was increased by fever (relative risk, 2.05; 95% confidence interval [CI], 1.16 to 3.63), "nothing orally" status (relative risk, 3.01; 95% CI, 1.88 to 4.83), hepatic disease (relative risk, 2.56; 95% CI, 1.39 to 4.70), and renal disease (relative risk, 2.07; 95% CI, 1.26 to 3.39). A logistic regression analysis showed "nothing orally" status to be associated with an increased risk of any hypoglycemia occurring during the hospitalization (relative risk, 2.39; 95% CI, 1.63 to 3.51). Physicians and nurses documented the first episode of hypoglycemia in their notes 45.5% and 80.3% of the time, respectively. CONCLUSION:Hypoglycemia is still a common complication of diabetic ketoacidosis, is associated with hepatic and renal disease as well as fever and "nothing orally" status, and is not documented well in physician notes.