OBJECTIVE: Cardiac complications are the most important source of morbidity and mortality during and after non-cardiac procedures, especially among diabetic patients. At present, the Philippine General Hospital (PGH) uses the 1977 Goldman Cardiac Risk Index, which includes nine independent variables. In 1999, the Lee Revised Cardiac Index was developed, which incorporated only six clinical variables, including diabetes mellitus. This study compared the accuracy of the two indices in predicting major cardiac complications among patients with diabetes mellitus undergoing major elective non-cardiac surgery.

METHODOLOGY: Non-probability convenience sampling was used. All subjects were interviewed, examined, and risk-stratified using the Goldman and Lee indices. Pre-operative and post-operative 12-lead electrocardiography (ECG) was done. All patients ^ere followed up until the day of discharge and the incidence of major cardiac complications was defined in a checklist.

RESULTS: Major cardiac complications occurred in 13 (14 percent) of the 90 subjects included in the study. Lee Class III was shown to be the best cut-off point for the clinical prediction of major cardiac complications in major non-cardiac surgery, having a positive LR of 2.54. The area under the receiver operating characteristics (ROC) curve for the Lee index was 0.76 (95 percent CI 0.66-0.85, p=0.001), while that for the Goldman index was 0.56 (95 percent CI 0.45-0.66, p=0.53). The difference of 0.21 between the areas under the ROC curve of the two indices was significant (95 percent CI 0.027-0.39, p= 0.025).

CONCLUSION: The Lee index is superior to the Goldman index in predicting major cardiac complications among diabetic patients undergoing major elective non-cardiac surgery who do not have major clinical predictors of increased perioperative cardiovascular risk.