A comparative evaluation of the Global Leadership Initiative on Malnutrition vs the Patient-Generated Subjective Global Assessment in assessing nutrition status in patients diagnosed with terminal cancer: A retrospective study.

This study aimed to evaluate the clinical utility of the Patient-Generated Subjective Global Assessment (PG-SGA) as a nutrition screening tool and the Global Leadership Initiative on Malnutrition (GLIM) criteria as a diagnostic framework in a cohort of patients with terminal cancer. This single-institution, retrospective cohort study included adults who were diagnosed with cancer and a predicted life expectancy <3 months intolerant to anticancer treatment who received palliative care between October 2023 and March 2024. Of 104 patients screened, 78 (54% male) were included in the analysis and 26 were excluded because of a terminal condition that precluded completion of the PG-SGA. The median age, body mass index, and survival were 73 years, 20.4, and 32 days, respectively. Weight loss occurred in 46% of patients within the previous 3 to 6 months, whereas 17% gained weight. Within the previous 2 weeks, 28% exhibited weight gain. The GLIM classified 35% of patients as well nourished, whereas the PG-SGA identified none as such. Agreement between the two tools was low (kappa coefficient = 0.037). Between the nutrition status screened by PG-SGA and assessed by the GLIM, no significant differences of all symptoms in Edmonton Symptom Assessment Systems or of survival outcomes were observed. In contrast, fluid retention and low handgrip strength emerged as significant predictors of mortality in Cox proportional hazards models. These findings suggest that, in patients with terminal cancer, PG-SGA may serve as a sensitive screening tool, whereas GLIM may have limited diagnostic applicability in end-of-life settings.