Research Article
Leblond J, Beauchesne MF, B
Abstract
Background: Insulin is commonly prescribed to treat hyperglycemia in the hospital setting, but is associated with a risk of hypoglycemia. The objective of this study was to determine the incidence rate and risk factors for hypoglycemia and hyperglycemia in hospitalized patients receiving insulin. Method: Retrospective cohort study analysing 58,496 patient-days of insulin exposure from 7780 hospitalizations of 5537 adult subjects at a teaching hospital between July 2009 and June 2011. The incidence rate of hypoglycemia (glycemia ≤ 3.9 mmol/L) and hyperglycemia (glycemia >16.7 mmol/L) were evaluated. Glycemia was measured by point-of-care blood-glucose. The association between risk factors and hypoglycemia/hyperglycemia events was determined using a Cox model. Results: The incidence rates for days with hypoglycemia were 11.1 per 100 patient-days for subcutaneous (s.c.) insulin and 10.4 per 100 patient-days for continuous intravenous insulin (CII). The incidence rates for days with hyperglycemia were 10.2 and 4.6 per 100 patient-days for s.c. insulin and CII, respectively. Clinically relevant risk factors associated with hypoglycemia for subjects on s.c. insulin were: creatinine clearance ≤ 60 mL/min: adjusted hazard ratio (HR) 1.14 [95% CI: 1.03-1.27]; surgery: HR 1.23 [95% CI: 1.04-1.46]; and diabetes: HR 1.79 [95% CI: 1.44-2.23]. For hyperglycemia, the risk factors were diabetes: HR 5.10 [95% CI: 3.65-7.12]; use of systemic corticosteroids: HR 2.13 [95% CI: 1.90-2.38]; and prescription of scheduled with sliding scale insulin: HR 1.89 [95% CI: 1.62-2.21]. ] Conclusion: The identified risk factors indicate areas for targeted improvement initiatives for glycemic control and should help reduce the rate of hyperglycemic and hypoglycemic events, thereby decreasing the occurrence of adverse outcomes.