Body mass index and mortality in patients with gastric cancer: A large cohort study.

Abstract

16 Background: Although excess body weight has been known to be an important risk factor for mortality from many cancers including colorectal, endometrial and breast cancers, the prognosis of gastric cancer (GC) in obese patients seems controversial. Methods: We aimed to evaluate the association between body mass index (BMI) and mortality in GC in a large cohort. A single institute cohort of 7,765 GC patients undergoing curative gastrectomy between October 2000 and June 2016 were categorized into 6 groups; underweight (< 18.5 kg/m2), normal (18.5 to < 23 kg/m2), overweight (23 to < 25 kg/m2), obese I (25 to < 28 kg/m2), obese II (28 to < 30 kg/m2), and severely obese (≥30 kg/m2). The hazard ratios (HRs) for overall survival and disease-specific survival were calculated using Cox proportional hazard model. Results: We identified 1,279 all-cause and 763 disease-specific deaths among 7,765 patients, and the median follow-up period was 83.05 months (range, 1.02-186.97 months). In multivariable analyses adjusted by age, sex, tumor stage, comorbidity and operation methods, preoperative BMI was associated with all-cause mortality in a nonlinear pattern. As compared with patients who were normal weight, underweight BMI showed increasing mortality risk (hazard ratio (HR), 1.42, 95% confidence interval (CI), 1.15-1.77). In contrast, patients who were overweight (HR, 0.84; 95% CI, 0.73-0.97), obese I (HR, 0.77; 95% CI, 0.66-0.90) and obese II (HR, 0.77; 95% CI, 0.59-1.01) had lower risk of mortality. disease-specific mortality also had a similar pattern to overall survival showing the lowest mortality in obese II group (HR, 0.59; 95% CI, 0.40-0.88). There was no significant difference in severely obese patients in both all-cause and disease-specific mortalities. In spline analyses illustrated by a bell-shaped curve, risk for all-cause mortality was the lowest in patients with 26.67 kg/m2. Conclusions: Preoperative overweight and obese patients (23 to < 30 kg/m2) had lower all-cause and disease-specific mortalities compared to those with normal weight in GC patients who underwent curative surgical resection.