LAUSR

Background Percutaneous Endoscopic Gastrostomy (PEG) is the preferred route for long-term enteral nutrition in dysphagic patients. However, in-hospital mortality after PEG is still a major concern with aspiration pneumonia reported as the major cause of death. Higher levels of blood urea nitrogen (BUN) have been associated with poor prognosis in patients with cardiovascular as well as cerebrovascular diseases. In this study, we investigated the prognostic utility of BUN in patients undergoing PEG. Methods 190 patients (88 men and 102 women) who received PEG for enteral nutrition at our hospital between September 2013 and September 2016 were included in this study. ROC analysis was used to evaluate BUN and other commonly used biomarkers as predictors for postoperative aspiration pneumonia and in-hospital mortality. Results Mean age of patients was 81.4±8.6 (SD) years old. 26 patients (14%) developed aspiration pneumonia and 19 patients (10%) died before discharge. Preoperative biomarkers/parameters: Body mass index (BMI) 18.5±9.2 kg/m2, Serum albumin (Alb) 2.8±0.5 g/dL, BUN 22.8±16.7 mg/dL, Total lymphocyte count (TLC) 1498±744/&mgr;L, Onodera’s Prognostic Nutritional Index (O-PNI) 35.7±6.8, Geriatric Nutritional Risk Index (GNRI) 76.7±18.6. AUC (Area under the curve) for the outcome of aspiration pneumonia after PEG: BMI 0.593 (95%CI: 0.490–0.695), Alb 0.656 (95%CI: 0.567–0.746), BUN 0.676 (95%CI: 0.565–0.787), TLC 0.612 (95%CI: 0.507–0.717), O-PNI 0.687 (95%CI: 0.585–0.789), GNRI 0.683 (95%CI: 0.580–0.787). AUC for postoperative in-hospital mortality: BMI 0.626 (95%CI: 0.482–0.770), Alb 0.642 (95%CI: 0.528–0.757), BUN 0.835 (95%CI: 0.750–0.921), TLC 0.652 (95%CI: 0.525–0.778), O-PNI 0.685 (95%CI: 0.576–0.794), GNRI 0.680 (95%CI: 0.557–0.804). A cut-off value of 24.5 mg/dL yielded a sensitivity of 78.9% and a specificity of 78.4% for in-hospital mortality. Conclusions Higher preoperative BUN levels predict aspiration pneumonia after PEG reasonably well and is a superior prognostic factor for in-hospital mortality when compared to other traditionally used biomarkers or even widely used risk indices such as O-PNI and GNRI. As BUN is also a frequently measured biomarker, clinicians should take it into account during preoperative assessment of patients undergoing PEG.