Assessment of factors affecting mortality in patients with percutaneous endoscopic gastrostomy tube placement in the intensive care unit

Abstract

Aim: It was aimed to evaluate the clinical outcomes, complications, and factors affecting mortality of percutaneous endoscopic gastrostomy (PEG) applied to patients in the intensive care unit (ICU).Material and Method: PEG procedures which were performed in the ICU between January 2016 and January 2021 and patients' age, gender, comorbidities, trauma history, serum CRP, albumin levels, CRP albumin ratios (CAR), and PEG-related complications were reviewed. Patients were divided into two groups, patients without mortality (Group 1) and patients with mortality (Group 2), and a comparison between groups was made.Results: Of all patients, 49 (39.2%) patients had mortality. The mean age of the patients in group 2 was 69.57±16.78 years, which was higher than the other group (p<0.001). Nephrological diseases and diabetes were significantly more common in Group 2, whereas neurologic diseases were less common (p=0.005, p=0.005, 0.044, respectively). The median length of stay (LOS) of the patients in Group 1 was 50 days, while the median LOS of the patients in Group 2 was found to be significantly higher, with 81 days (<0.001). The mean CRP of Group 2 was 81.63±54.06, which was higher than the other group, while the mean of albumin was found to be 2.29±0.5 and was lower than Group 1 (p<0.001, p<0.001). The mean CAR of Group 1 was 15.96±16.81, which was significantly lower than that of Group 2 (p<0.001). The optimal CAR cut-off value for mortality discrimination was found to be 20,216 with a sensitivity of 73.5%, a specificity of 78.9%, a positive predictive value of 69.2%, a negative predictive value of 82.2%, and 76.8% test accuracy. A CAR value of ≥ 20.216 increased the odds of death 9.3-fold (OR 10.385, CI 95% 4.481-24.065, p<0.001). Conclusion: We suggest that CAR ratio, low albumin, and high CRP levels could be predictors of early mortality. Considering that PEG is an elective procedure, we believe that it can be a safe and effective procedure when nutritional support is provided by alternative means and appropriate conditions are met.