Affiliation:
1. Division of Surgical Oncology , Department of Surgery, , Nagoya , Japan
2. Nagoya University Graduate School of Medicine , Department of Surgery, , Nagoya , Japan
Abstract
Summary
Late-onset postoperative pneumonia (LOPP) after esophagectomy is poorly understood. This study was designed to clarify the features and risk factors for this event. Patients who underwent esophagectomy for esophageal cancer between 2006 and 2016 were included. LOPP was defined as radiologically proven pneumonia that occurred over 3 months after surgery, and clinically relevant late-onset postoperative pneumonia (CR-LOPP) was defined as LOPP that required administration of oxygen and antibiotics in the hospital and/or more intensive treatment. The total psoas muscle area (TPA) was measured using preoperative and postoperative (at 3 months after surgery) computed tomography scan images. Potential risk factors for CR-LOPP were investigated. Among 175 study patients, 46 (26.3%) had LOPP, 29 (16.6%) of whom exhibited CR-LOPP with a cumulative incidence of 15.6% at 3 years and 22.4% at 5 years. Four (13.8%) of these patients died of LOPP. Univariable analysis showed that clinical stage ≥III (P = 0.005), preoperative prognostic nutritional index (PNI) <45 (P = 0.035), arrhythmia (P = 0.014), postoperative hospital stay ≥40 days (P = 0.003), and percent decrease of TPA more than 5% (P < 0.001) were associated with CR-LOPP but not early onset postoperative pneumonia. Multivariable analysis revealed that clinical stage ≥III (hazard ratio [HR] 3.01, P = 0.004), postoperative hospital stay ≥40 days (HR 2.51, P = 0.015), and percent decrease of TPA >5% (HR 9.93, P < 0.001) were independent risk factors for CR-LOPP. CR-LOPP occurred in over 20% of patients at 5 years, and early postoperative loss of TPA was a potential trigger for this delayed complication.
Publisher
Oxford University Press (OUP)
Subject
Gastroenterology,General Medicine