Perioperative Allogenic Blood Transfusion in Renal Cell Carcinoma: Risk Factors and Effect on Long-term Outcomes.

BACKGROUND

We sought to create a preoperative model to predict the risk of perioperative blood transfusion (PBT) in patients with renal cell carcinoma (RCC) undergoing nephrectomy and to evaluate the effect of PBT on long-term outcomes.

PATIENTS AND METHODS

The present retrospective study included 648 consecutive patients who had undergone radical or partial nephrectomy for RCC at a single institution. The risk factors for PBT were analyzed using logistic regression analysis. Cox proportional hazards models addressed the effect of PBT on overall and RCC-specific mortality.

RESULTS

A total of 62 patients (10%) received a median of 2 red blood cell units (interquartile range, 2-3; range 1-20). On multivariable logistic regression analysis, 2 preoperative factors were independently associated with receipt of PBT: preoperative anemia (odds ratio, 6.28; P < .001) and open surgery (odds ratio, 3.40; P < .001). The risk of receiving PBT was high with both risk factors present (34%), intermediate with 1 risk factor present (7%-12%), and low with 0 risk factors present (2%). Within a median follow-up period of 63 months (interquartile range, 32-91), 108 patients (17%) had died of RCC and 177 (27%) had died of any cause. In the multivariable Cox models, PBT remained independently associated with overall mortality (hazard ratio [HR], 1.86; P = .004) and RCC-specific mortality (HR, 1.79; P = .007). A dose-dependent association of PBT with RCC-specific mortality was observed (HR, 1.14; P = .01).

CONCLUSION

In patients undergoing surgery for RCC, PBT was associated with adverse overall and RCC-specific mortality. Patients with preoperative anemia and those scheduled to undergo open surgery are at an increased risk of PBT and could be candidates for perioperative optimization techniques.