The Revised Cardiac Risk Index in the new millennium: a single-centre prospective cohort re-evaluation of the original variables in 9,519 consecutive elective surgical patients

Abstract

Purpose: Cardiac complications following non-cardiac surgery are major causes of morbidity and mortality. The Revised Cardiac Risk Index (RCRI) has become a standard for predicting post-surgical cardiac complications. This study re-examined the original six risk factors to confirm their validity in a large modern prospective database.

Methods: Using the definitions in the original risk index, this study included 9,519 patients aged ≥ 50 undergoing elective non-cardiac surgery with an expected length of stay ≥ two days at two major tertiary-care teaching hospitals. The validity of the original predictors was tested in this population using binomial logistic regression modelling, area under the receiver operator curve (ROC) analysis, and the net reclassification index.

Results: Rates of major cardiac complications with 0, 1, 2, ≥ 3 of the predictors were 0.5%, 2.6%, 7.2%, and 14.4%, respectively, in our patient cohort compared with 0.4%, 1.1%, 4.6%, and 9.7%, respectively, in the original cohort. Similar to the original report, binary logistic regression analysis showed that both preoperative treatment with insulin (odds ratio [OR] 1.4; 95% confidence interval [CI] 0.7 to 2.6) and preoperative creatinine > 176.8 mmol·L(-1) (OR 1.7; 95% CI 0.8 to 3.6) did not improve the predictive ability of the index. Analysis of the remaining four factors resulted in an area under the curve (AUC) identical to that seen for the reconstructed six-factor RCRI (AUC = 0.79). We found that a glomerular filtration rate (GFR) < 30 mL·min(-1) was a better predictor of major cardiac complications (OR 2.2; 95% CI 1.2 to 4.3) than creatinine > 176.8 mmol·L(-1). The receiver operating characteristic analysis of this resultant 5-Factor model resulted in an AUC of 0.79, with 0, 1, 2, ≥ 3 of the predictors representing 0.5%, 2.9%, 7.4%, and 17.0% risk, respectively, among our patient cohort.

Conclusion: Compared with the RCRI, a simplified 5-Factor model using a high-risk type of surgery, a history of ischemic heart disease, congestive heart failure, cerebrovascular disease, and a preoperative GFR < 30 mL·min(-1) results in superior prediction of major cardiac complications following elective non-cardiac surgery.