Feasibility, safety and outcome of endoscopic gastrostomy in patients with esophageal cancer

Background and aims: esophageal cancer (EC) is an important health problem worldwide with high morbidity and mortality. EC patients are likely to develop malnutrition. The aim of this study was to assess the feasibility and safety of endoscopic gastrostomy (PEG) feeding in EC cancer, and to identify risk factors associated with poor prognosis. Methods: a retrospective observational study was performed using records from EC patients referred for PEG. Age, gender, cancer histologic subtype, indication for gastrostomy, and mortality data were recorded. NRS 2002, body mass index (BMI), hemoglobin, serum albumin, transferrin and total cholesterol were collected at the day of PEG. An association between anthropometric, clinical and laboratorial data with patient survival was assessed. Results: data were obtained for forty-one EC patients (36 men and 5 women) aged 39-88 years (mean, 62 years). Gastrostomy was possible in all patients referred to PEG (27 patients selected for curative treatment and 14 patients for palliative nutrition). No major complications occurred. Mean survival after PEG was 18.1 months, and mortality rate at 3 months was 31.7 %. Most patients (34; 82.9 %) died under PEG feeding. Mean BMI was 21.3 kg/m² and 14 patients (34.1 %) patients had low BMI. Serum albumin, transferrin and total cholesterol were low in 10 (24.4 %), 20 (48.8 %) and 18 (43.9 %) patients, respectively. Higher BMI (R = 0.30), serum albumin (R = 0.41) and transferrin (R = 0.47) tended to be positively correlated with survival (p < 0.005). Conclusions: PEG is a feasible and safe technique for enteral feeding in EC patients. Higher BMI, serum albumin and transferrin levels at admission predict a better outcome. Enteral feeding through PEG should be considered early in EC patients due to their higher risk of malnutrition, which is associated with shorter survival.