Laboratory values and their relationship with short- and long-term outcome following transthoracic esophagectomy in patients with esophageal cancer.

Abstract

37 Background: Surgical resection in esophageal cancer patients remains the mainstay of therapy, but carries a considerable risk of post-operative complications. Deranged laboratory data in the early post-operative period could provide guidance in decision making for further treatment strategy. Therefore we related standard laboratory measurements on 90-day and 1-year mortality after esophagectomy. Methods: Esophagectomy with curative intent was performed in 226 consecutive patients with cancer of the esophagus between January 2000 and December 2009. Prognostic analyses for albumin, lactate dehydrogenase (LDH), ALAT, ASAT, gGT, urea, creatinine, C-reactive protein (CRP), white blood cell count (WBC) and platelet count were performed with a Mann-Whitney U-test (p<0.05) and ROC curve (area under the curve (AUC) >0.7) for postoperative day (POD) 1, 2, 3, 4, 5, 10 and 15. Extreme laboratory values (cutoff points) were calculated (5th or 95th percentile) in the case of significant outcomes. Results: Laboratory measurements between survivors and patients deceased within 90-days (n=16) were significantly different in the case of albumin on POD 1, 3, 5, 10 and 15 (5th percentile between 10-13.02, AUC: 0.761), CRP on POD 3, 4, 5, 10 and 15 (95th percentile between 228.8-324.7, AUC: 0.861) and LDH in the late postoperative period on POD 4, 10 and 15 (95th percentile between 387.1-512.2, AUC: 0.864). The other laboratory values showed no systematic differences. For one-year mortality (N=36), only albumin and CRP showed to have significant differences for non-survivors: albumin on POD 10 and 15 (5th percentile between13.2-13.75, AUC: < 0.7) and CRP on POD 3, 4 and 10 (95th percentile between 220.9-319.5, AUC: < 0.7). Conclusions: Albumin and CRP are the strongest prognostic measurements in predicting 90-day and one-year mortality after esophagectomy.