POS-068 RED BLOOD CELL DISTRIBUTION WIDTH AS PREDICTOR OF IN-HOSPITAL MORTALITY IN PATIENTS WITH SEPSIS INDUCED ACUTE KIDNEY INJURY: 5 YEAR RETROSPECTIVE REVIEW STUDY FROM 2014-2018

Abstract

Sepsis is the leading cause of Acute Kidney Injury (AKI). Patients with sepsis induced AKI have an increased risk of death and a longer duration of hospitalization. Red cell distribution width (RDW) measure the range of variation in the red cell volume and variability in size of erythrocytes. Inflammation and oxidative stress are essential components of sepsis cascade, and they have been suggested to reduce Red Blood Cell survival and suppress their maturation leading to a release of increased premature RBCs into circulation, thus, an elevated RDW. The study adapt a retrospective observational analytical cross-sectional review of patients admitted at University of Santo Tomas Hospital from January 2014 to December 2018. Information was gathered across the departments of Surgery, Obstetrics and GynecologyInternal Medicine that were referred to the Section of Nephrology. Total of 421 patients were analyzed, with median age of 67 years, 53% being female, and majority with diabetes. Patients had low median hemoglobin (118 g/dl) and elevated WBC (14.54 x 1000/µL); low mean serum albumin (2.98); and elevated median serum creatinine (2.12 mg/dl). The prevalence of stages 1, 2, and 3 AKI were 70%, 13%, and 17%, respectively. One in 10 patients underwent dialysis. There were 32 patients who had RDW of at least 15.1%. This group had higher median age (77.5 years), serum creatinine (3.78 mg/dL), increase hematologic comorbidity (9%), cardiac disease (53%), septic shock (87.5%), stage 3 AKI (78%), need for Renal Replacement Therapy (44%); higher SOFA score (12); and longer hospital stay (21 days) compared to those with lower baseline RDW levels. Optimal cut-off value for RDW was 15.1%. At this level, sensitivity and specificity were 82.14 and 97.71%, respectively. On bivariate analysis, factors associated with increase in-hospital mortality were higher RDW, age, WBC, serum creatinine, SOFA score, and presence of AKI stages >1 and septic shock. With adjustment for other factors, only RDW and SOFA remained as the significant predictors, with every point increase resulting to 97.5% and 68.4% rise in odds of death, respectively. In the final multivariate Cox’s regression (after adjusting for all other factors such as RDW, age, and other factors in the initial model), only RDW and SOFA score are significant. For every unit increase in RDW, the hazard for mortality increases by approximately 15% (aHR 1.15, 95% CI 1.06 to 1.25, p = 0.001), after adjusting for age, sex and co-morbidities. For every unit increase in SOFA scores, the hazard for mortality increases by approximately 43% (aHR 1.431, 95% CI 1.25 to 1.64, p < 0.001). This model explained 28.47% in the variance of time to mortality and was statistically significant at p<0.00. The current study found that an elevated RDW is an independent predictor of in hospital mortality among patients with sepsis induced AKI. Patients with RDW of > 15.1 are approximately twice more likely to die among patients with sepsis induced AKI. This finding suggests that RDW may aid in risk stratification and can represent an inexpensive marker to predict mortality among sepsis induced AKI.