Risk factors for perioperative bleeding in percutaneous endoscopic gastrostomy construction: A case series and retrospective preliminary investigation.

Abstract

Percutaneous endoscopic gastrostomy (PEG) is one of the methods of tube feeding in patients who are incapable of oral intake. There are no reports on risk factors for bleeding at PEG construction. Our main objective was to investigate the risks and predictors of bleeding associated with PEG construction. This retrospective, observational study included patients who had undergone PEG construction at our institution. To investigate the risks of bleeding associated with PEG construction, we compared the baseline characteristics between bleeding and non-bleeding patients. In terms of early predictors of post-PEG bleeding, we evaluated whether there had been a decrease from baseline of >10% in hemoglobin levels on the day after the procedure. The median preoperative albumin levels were 22.5 g/L (range 20-29 g/L) and 30 g/L (range 18-40 g/L) in the bleeding and non-bleeding groups, respectively (P = 0.014, Mann-Whitney U-test). The median preoperative platelet counts were 177 500 (range 87 000-265 000) and 271 000 (83 000-749 000) in the bleeding and non-bleeding groups, respectively (P = 0.043, Mann-Whitney U-test). The number of patients for whom hemoglobin levels decreased >10% from baseline on the day after the procedure differed significantly between the bleeding and non-bleeding groups (2/4, 50% and 3/58, 5.45%), respectively (P = 0.002, Pearson's χ2 -test). Low serum albumin and preoperative platelet counts might be risk factors for bleeding. Rigorous follow up is necessary for patients showing a decrease in hemoglobin level ≥10% of their baseline the day after the procedure. Geriatr Gerontol Int 2020; 20: 932-937.