Table 4.

Subgroup analysis of postoperative underweight patients

No. at riskNo. of eventsUnadjusted HR (95% CI)Multivariate adjusted HR (95% CI)a
Postoperative BMI (n = 1,845) Overall survival
 <18.5 and preoperative BMI <18.550101.32 (0.70–2.50)1.23 (0.64–2.36)
 <18.5 and preoperative BMI ≥18.5158451.98 (1.43–2.75)1.90 (1.34–2.69)
 18.5 to <23.01,0811751 [Reference]
 23.0 to <25.0364320.51 (0.35–0.75)0.65 (0.44–0.95)
 25.0 to <27.5145120.49 (0.27–0.88)0.64 (0.35–1.16)
 ≥27.54730.39 (0.12–1.22)0.82 (0.26–2.60)
Cancer-specific survival
 <18.5 and preoperative BMI <18.55061.32 (0.58–3.01)1.00 (0.43–2.34)
 <18.5 and preoperative BMI ≥18.5158241.76 (1.13–2.74)1.69 (1.05–2.73)
 18.5 to <23.01,0811031 [Reference]
 23.0 to <25.0364140.39 (0.22–0.67)0.53 (0.30–0.93)
 25.0 to <27.514560.42 (0.18–0.95)0.91 (0.37–2.21)
 ≥27.54710.22 (0.03–1.56)1.08 (0.15–7.93)
  • aCox regression models adjusted for age at diagnosis in years, sex (male [reference], female), ASAs classification (I [reference], II, or ≥III), smoking status (never [reference], past, or current), T-stage (1 [reference], 2, 3, or 4), N-stage (0 [reference], 1, 2, or 3), venous invasion (no [reference], yes), perineural invasion (no [reference], yes), histology (differentiated [reference], undifferentiated, or mixed), chemotherapy (no [reference], yes), and extent of gastric resection (distal gastrectomy [reference], total gastrectomy, proximal gastrectomy, or pylorus-preserving gastrectomy).