Outcomes of Patients Undergoing Renal Replacement Therapy in a Tertiary ICU in Nepal: A Three-year Retrospective Registry Based Cohort Study

INTRODUCTION: Renal failure that requires renal replacement therapy(RRT), whether acute or chronic, is a frequent occurrence in critically ill patients.(1,2) It has been consistently found to adversely affect clinical outcomes including length of ICU stay (LOS-ICU), mechanical ventilation(MV) and even mortality.(3,4) However, most of the evidences found in medical literature are usually not inclusive of the scenario in low and low middle income countries(LMICs), and only small number of studies are found from these regions.(5,6) This is a registry based retrospective cohort study of data from more than 3700 patients admitted during a period of three years in a tertiary level hospital in Nepal of patients undergoing RRT in ICU. OBJECTIVES: The objective of the study was to compare the characteristic of patients undergoing RRT with those not requiring RRT (NRRT) in ICU. METHODS: After ethical clearance from the Institutional Review Committee(IRC), de-identified data of patients admitted in the ICUs of the Department of Critical Care Medicine, Tribhuvan University Teaching Hospital, Kathmandu, Nepal was retrieved from ICU registry between January 1, 2020 and December 31, 2022. All the data were retrieved in Microsoft Excel. Descriptive statistics using percentages and means were used to summarize the characteristics of the patients. Comparisons between the groups were done using Chi-square test. For numerical data normality testing was done using Shapiro-Wilk test. Non-normally distributed data were described using medians and interquartile ranges(IQR). Outcomes were described in terms of ICU mortality, LOS-ICU and length of mechanical ventilation(MV). Between-group comparisons were made using odds ratio(OR). For mortality analysis patients who had limitations to full treatment i.e. patients who either left against medical advice(LAMA), were discharged on request(DOR) or were withdrawn from treatment were excluded. Additionally, for these patients, risk of having treatment limitation was analyzed with respect to having RRT. Level of statistical significance was set at a p value of <0.05 for all comparisons. All statistical analyses were done using SPSS. RESULTS AND DISCUSSION: Of the total 3733 patients admitted during the period of 3 years, 516(13.82%) underwent RRT. The baseline demographics are presented in Table 1. Of note, APACHE II scores, and requirement of MV and non-invasive ventilation(NIV) were higher in RRT group as compared to NRRT group which were statistically significant. The OR of ICU mortality in the RRT group was 3 compared to NRRT group (CI 2.42-3.70, p<0.05). LOS-ICU and length of MV were also higher for patients in RRT group (Table 3). Treatment limitations (Table 4) were more likely in the RRT group (OR 1.55, CI 1.21-1.99, p<0.05). An important finding here was that APACHE II score grossly underestimated the predicted mortality in RRT group (estimated: 25% vs observed: 46.33%). CONCLUSION: Patients who require RRT in ICU have poor prognosis in terms of length of MV, LOS-ICU and ICU mortality. They also have higher odds of getting treatment limitations (LAMA, DOR, withdrawal). Furthermore, APACHE II grossly underestimates the mortality rates of patients undergoing RRT in an LMICs.