Patient-generated subjective global assessment predicts cachexia and death in patients with head, neck and abdominal cancer: A retrospective longitudinal study.

BACKGROUND & AIMS: Cancer is a complex disease, with poor prognosis when associated with malnutrition. This condition can lead to Cancer Cachexia (CC), a syndrome characterized by loss of muscle mass with or without fat loss, often associated with higher risk of death. Although there are recommended screening tools to assess nutritional status in cancer patients, such as Patient-Generated Subjective Global Assessment (PG-SGA), little is known about CC prediction. The aim was to investigate the association between nutritional status of patients with head, neck and abdominal cancer, assessed by PG-SGA at the day of hospitalization, with CC, hospitalization time and death.

METHODS: This is a retrospective longitudinal study, where we collected data of 97 patients about PG-SGA nutritional classification, anthropometry, gender, age, cachexia diagnosis and death.

RESULTS: PG-SGA classification was strongly associated with all the anthropometric measures (p < 0.0001). According to PG-SGA classification, 30.61% (n = 15) of patients in group A developed pre-cachexia; 38.24% in group B developed CC (n = 13); and 60% (n = 9) in group C developed refractory cachexia (p < 0.0001). Death rate was 24.49% (n = 12), 54.55% (n = 18) and 80% (n = 12) in groups A, B and C, respectively (p < 0.0001). PG-SGA had good sensibility (89.5%) and accuracy (72%) for CC, and also good specificity (75.51%) and accuracy (69%) for death.

CONCLUSIONS: PG-SGA demonstrated a significative association with the variables measured and was able to predict CC and death. This, in addition to its simple applicability, suggests that PG-SGA can be a useful tool to screen cancer patients for CC establishment and death risk.