Background: Diabetes mellitus is an established independent risk factor related to significant morbidity and mortality after cardiac surgical procedures.
Methods: Data on 16,184 patients undergoing cardiac surgery with and without cardiopulmonary bypass between April 1996 and August 2001 were prospectively evaluated. Diabetes mellitus as a patient related risk factor was subjected to univariate analysis to identify potential associations to 28 intra- and postoperative outcome variables. Outcome variables having a significant association with diabetes mellitus (p < 0.05) were then subjected to a stepwise logistic regression model to identify the influence of diabetes mellitus as compared to additional 30 different patient related risk factors and treatment variables. Diabetes mellitus was defined as glucose intolerance treated either dietary, with oral hypoglycemics or with insulin.
Results: Overall prevalence of diabetes mellitus was 33.3 %. Compared to non-diabetic patients the group with diabetes mellitus was older (p < 0.0001) and had a significantly lower ejection fraction (p < 0.0001). 15 outcome variables having a significant association with diabetes mellitus were identified. Furthermore, diabetes mellitus could be identified as an independent predictor for 7 postoperative outcome variables (prolonged ICU-stay, sternal instability and/or infection, sternal revision and refixation respiratory insufficiency, postoperative delirium, perioperative stroke, renal dysfunction, postoperative reintubation).
Conclusion: Diabetes mellitus is a significant independent predictor for several postoperative outcome variables after cardiac surgery associated with higher postoperative morbidity and prolonged hospital stay.