RA03.02: ACUTE KIDNEY INJURY AFTER ESOPHAGEAL CANCER SURGERY: INCIDENCE, RISK FACTORS, AND IMPACT ON ONCOLOGIC OUTCOMES

Background Surgery remains the cornerstone of treatment for esophageal cancer. Esophagectomy serves as an exemplar of major operative trauma, yet there is a paucity of literature regarding postoperative renal outcomes. We aimed to study the incidence of acute kidney injury (AKI) after esophageal cancer surgery, and to determine independent risk factors for postoperative renal impairment. Methods Consecutive patients undergoing potentially curative surgery for esophageal cancer from 2006–2016 were studied. AKI was defined according to AKIN criteria. Complications were recorded prospectively and comprehensive complications index (CCI) was determined. Multivariate linear and logistic regression were performed to determine factors independently predictive of postoperative AKI. Results 661 patients (72.9% male), a mean age of 63.5 ± 9.7 years, underwent surgery (2-stage esophagectomy, 307 [46.4%]; 3-stage esophagectomy, 117 [17.7%]; transhiatal esophagectomy, 112 [16.9%]; extended total gastrectomy, 122 [18.5%)], pharyngolaryngoesophagectomy, 3 [0.5%]), with a CCI of 21.3 ± 19.7 and an in-hospital mortality of 1.4%. Baseline prevalence of chronic kidney disease was 1.8%. Postoperative AKI occurred in 174 (26.3%) patients, with AKIN 1, 2 and 3 in 122 (18.5%), 41 (6.2%) and 11 (1.7%), respectively. Of these, 5 (2.9%) required renal replacement therapy during admission. Preoperatively, greater BMI (P = 0.02, OR 1.05 [95% CI 1.01–1.10]), male sex (P = 0.05, OR 1.68 [1.01–2.78]), age (P = 0.001, OR 1.04 [1.02–1.07]), hypertension (P = 0.005, OR 1.85 [1.20–2.84]), and transthoracic approach (P = 0.01, OR 1.82 [1.14–2.89]) independently predicted postoperative AKI. After surgery, while CCI was predictive of AKI on univariable analysis (P < 0.001 OR 1.03 [1.02–1.04]), atrial fibrillation (P = 0.001 OR 3.25 [1.57–6.72]) and prolonged intubation (P = 0.016 OR 3.61 [1.28–10.21]) were independently associated with AKI on multivariable logistic regression. On multivariable linear regression, neoadjuvant treatment (surgery only, 13.5 ± 35.8%; chemotherapy, 25.3 ± 42.2%; chemoradiation 18.7 ± 31.9%, P = 0.02), transthoracic approach (P = 0.006), and baseline BMI (P = 0.02) predicted postoperative %Δcreatinine. Postoperative AKI did not impact survival outcome on univariable or multivariable analysis. Conclusion Major AKI is rare after esophageal cancer surgery, with risk independently associated with increasing age, features of the metabolic syndrome, and postoperative morbidity. Disclosure All authors have declared no conflicts of interest.