Associations of patient-generated subjective global assessment (PG-SGA) and NUTRISCORE with survival in gastric cancer patients: timing matters, a retrospective cohort study.

ABSTRACT

BACKGROUND: The timing of nutritional assessment may be important to treat cancer patients and predict their prognosis. This study examined whether Patient-Generated Subjective Global Assessment (PG-SGA) and NUTRISCORE scores were associated with survival among gastric cancer patients who underwent surgery and chemotherapy and whether the timing of the assessment after surgery mattered. METHODS: A total of 952 gastric cancer patients (622 men and 330 women) were included in this retrospective cohort study. The PG-SGA and NUTRISCORE scores were calculated at 1 month (n = 952), 2 months (n = 657), and 3 months (n = 294) after surgery. Cox proportional hazards model was used to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: The PG-SGA scores assessed at 1 month after gastrectomy were not associated with survival. However, high PG-SGA scores at 2 months after gastrectomy (median = 65 days) were associated with an increased risk of mortality; the HR (95% CI) was 2.26 (1.22-4.21) for 9-11 vs. ≤ 5 of PG-SGA scores. When we included patients who received all three consecutive consultations, HR (95% CI) was 2.56 (1.02-6.42) for ≥ 9 (malnutrition) vs. ≤ 8 of PG-SGA scores assessed at 3 months after surgery (median days = 98 days). Likewise, high NUTRISCORE scores assessed at the 3-month follow-up were associated with higher mortality; the HR (95% CI) was 3.84 (1.18-12.55) for ≥ 7 vs. ≤ 4 of NUTRISCORE scores. CONCLUSION: Malnutrition assessed with the PG-SGA and NUTRISCORE at 2 to 3 months after gastrectomy was associated with poor survival among gastric cancer patients. Our findings suggest that the timing of the nutritional evaluation may be important in identifying and treating malnutrition related to gastric cancer prognosis.