Background: Poor renal function prior to surgery is associated with increased risk for mortality in patients undergoing major vascular surgery. Traditionally, this function is assessed by serum creatinine concentration (SeCreat). However, SeCreat is also influenced by age, gender and body weight. Hence, creatinine clearance (C(Cr)) is considered to be a better reflection of renal function. This study was undertaken to explore the prognostic value of preoperative calculated Cc, compared to SeCreat for the prediction of postoperative mortality.
Patients and methods: The study group comprised 852 consecutive patients who underwent elective major vascular surgery at the Erasmus Medical Center, Rotterdam. Preoperative C(Cr) was calculated based on the Cockroft-Gault equation using preoperative SeCreat, age, body weight and gender. Univariable logistic regression analyses were used to study the relation between preoperative SeCreat, C(Cr) and postoperative mortality. Furthermore, multivariable logistic regression analysis was applied to evaluate the additional predictive value of age, body weight and gender additional to SeCreat. The receiver operating characteristic (ROC) curve was determined to evaluate the predictive power of several regression models for perioperative mortality.
Results: Postoperative mortality was 5.9% (50/852) within 30 days of surgery. In a univariable analysis, 10 micromol/l increment of SeCreat were associated with a 20% increased risk of postoperative mortality (OR = 1.2, 95% CI, 1.1-1.3) with an area under the ROC curve of 0.64 (95% CI, 0.56-0.71). If age, gender and body weight were added, the area under the ROC curve increased to 0.70 (95% CI, 0.63-0.77; p < 0.001), indicating that these risk factors had additional prognostic value. Indeed, in a separate regression analysis 10 ml/min decrease in C(Cr) was associated with a 40% increased risk of postoperative mortality (OR = 1.4,95% CI, 1.2-1.5; ROC area: 0.70, 95% CI, 0.63-0.76). ROC curve analysis showed that the cut-off value of 64 ml/min for C(Cr) yielded the highest sensitivity/specificity to predict postoperative mortality.
Conclusion: Preoperative SeCreat was strongly associated with postoperative mortality, and adding age, gender, and body weight to the model showed improved predictive power indicating that preoperative C(Cr) calculated with these data has additional prognostic value.