High comorbidity indexes predict 30-day mortality after percutaneous endoscopic gastrostomy

Abstract

Abstract Background High postoperative mortality rates have been reported for percutaneous endoscopic gastrostomy (PEG), although guidelines recommend PEG for patients anticipated inability of oral intake for more than 4 weeks. Conversely, comorbidity indexes have not been adequately validated among patients who underwent PEG. We evaluated comorbidity indexes and clinical risk factors for 30-day mortality after PEG. Methods We retrospectively reviewed 1,395 patients who underwent PEG at our hospital from April 2008 to December 2020. We used the direct introducer procedure to perform PEG and compared preprocedural clinical factors, demographics, comorbidity indexes, (Charlson comorbidity index [CCI] and cumulative illness rating scale [CIRS]), and laboratory data of patients who survived and those who died within 30 days after PEG. Results A total of 104 (7.5%) patients died within 30 days after PEG. Multivariate logistic regression analysis revealed association of 30-day mortality with patients’ age of ≥ 80 years, total parental nutrition before PEG, CCI of ≥ 5, CIRS of ≥ 13, and serum C-reactive protein level of ≥ 2.6 mg/dl. The receiver operating characteristics analysis revealed CIRS as the best 30-day mortality predictor (area under the curve: 0.80; 95% CI: 0.75–0.85). The 30-day mortality among the highest quartile of CCI and CIRS groups were 16.9% and 20.3%, respectively. Kaplan–Meyer analysis revealed that multivariable-adjusted hazard ratios are favored lower comorbidity scores. Conclusion High comorbidity indexes were associated with 30-day mortality after PEG. Comorbidity indexes must be considered while predicting patients’ prognosis to reduce inappropriate performance of PEG.