Predictors of postoperative acute kidney injury in robotic-assisted pulmonary lobectomy.

Abstract

e20090 Background: Acute Kidney Injury (AKI) has been extensively documented as a predictor of poor perioperative outcomes in thoracic surgery [1,2]. Even minor increases in creatinine are associated with longer hospital stays, mechanical respiratory ventilation, progression to chronic kidney disease, and even postoperative mortality. Here we explore the predictive factors of AKI for patients who undergo robotic-assisted pulmonary lobectomy (RAPL). Methods: We retrospectively analyzed 797 patients who underwent RAPL by one surgeon over a 13-year period. We define AKI using the modified Kidney Disease Improving Global Outcomes (KDIGO) definition [3] of an increase in serum creatinine of at least 0.3 mg/dL within 48 hours after surgery when compared to values collected within 30 days of preoperatively. Those with end-stage renal disease prior to lobectomy were excluded. Differences in demographics, preoperative comorbidities, and intraoperative data between the AKI and non-AKI cohorts were compared using chi-square tests of independence and multivariate logistic regression. Results: Demographic predictors of AKI include older age (p=0.043), male sex (p=0.013), and higher BMI (p=0.002). Preoperative hypertension (p=0.005) and diabetes mellitus (p=0.010) were also significant risk factors. However, there was no significant difference in the Charles Comorbidity Index (CCI) between AKI and non-AKI cohorts (p=0.055). Additionally, preoperative CKD (p=0.410), CHF (p=1.000), and smoking history (p=0.295) did not significantly affect risk of AKI. Intraoperative predictors included longer operative time (p=0.013), intraoperative blood loss (p < 0.001), and presence of surgical complications (p=0.049). There was no difference in lowest intraoperative mean arterial pressure (p=0.656) or total intraoperative fluid balance (p=0.525). After adjusting for significant variables, estimated blood loss remained the only independent predictor of AKI (p=0.007). Conclusions: Our data suggests that patients without outright CKD, but with significant risk factors, develop postoperative AKI and should be monitored closely. Secondly, this risk may not be fully captured by CCI, which does not include hypertension and BMI. We should also note the importance of minimizing intraoperative blood loss, which is the strongest predictor for AKI.