医学
胃切除术
内科学
胃肠病学
危险系数
癌症
食管胃交界处
单变量分析
风险因素
多元分析
病态的
比例危险模型
外科
腺癌
置信区间
作者
Yoshihiko Kakiuchi,Shigetoshi Kuroda,Yasuhiro Choda,S. Otsuka,Satoshi Ueyama,Norimitsu Tanaka,Atsushi Muraoka,Shinji Hato,Yasuaki Kamikawa,Toshiyoshi Fujiwara
标识
DOI:10.1016/j.suronc.2023.101990
摘要
Although proximal gastrectomy (PG) is commonly used in patients with upper gastric cancer (GC) and esophagogastric junction (EGJ) cancer, long-term prognostic factors in these patients are poorly understood. The double-flap technique (DFT) is an esophagogastrostomy with anti-reflux mechanism after PG; we previously conducted a multicenter retrospective study (rD-FLAP) to evaluate the short-term outcomes of DFT reconstruction. Here, we evaluated the long-term prognostic factors in patients with upper GC and EGJ cancer. The study was conducted as a secondary analysis of the rD-FLAP Study, which enrolled patients who underwent PG with DFT reconstruction, irrespective of disease type, between January 1996 and December 2015. A total of 509 GC and EGJ cancer patients were enrolled. Univariate and multivariate analyses of overall survival demonstrated that a preoperative prognostic nutritional index (PNI) < 45 (p < 0.001, hazard ratio [HR]: 3.59, 95% confidential interval [CI]: 1.93–6.67) was an independent poor prognostic factor alongside pathological T factor ([pT] ≥2) (p = 0.010, HR: 2.29, 95% CI: 1.22–4.30) and pathological N factor ([pN] ≥1) (p = 0.001, HR: 3.27, 95% CI: 1.66–6.46). In patients with preoperative PNI ≥45, PNI change (<90%) at 1-year follow-up (p = 0.019, HR: 2.54, 95%CI: 1.16–5.54) was an independent poor prognostic factor, for which operation time (≥300 min) and blood loss (≥200 mL) were independent risk factors. No independent prognostic factors were identified in patients with preoperative PNI <45. PNI is a prognostic factor in upper GC and EGJ cancer patients. Preoperative nutritional enhancement and postoperative nutritional maintenance are important for prognostic improvement in these patients.
科研通智能强力驱动
Strongly Powered by AbleSci AI