End-stage Renal Disease but Not Chronic Kidney Disease Is a Predictor for Worse Outcomes in Patients With Severe Sepsis and Septic Shock

Abstract

PURPOSE: Severe sepsis and septic shock are the most common causes of mortality in noncoronary intensive care units. Acute renal failure in these patients has shown to be an independent risk factor for poor outcomes. ESRD in animal models of sepsis has shown to increase vascular permeability, and splenocyte apoptosis. These animals also have higher plasma vascular endothelial growth factor (VEGF) & serum IL-10 levels. Though a few small studies have documented that increased prevalence and mortality from severe sepsis in the ESRD population, the same remains largely unknown for patients with CKD. We sought to examine the outcomes of severe sepsis (SSE) and septic shock (SSH) in patients with ESRD and CKD. METHODS: Using the NIS 2007, patients aged 18 or more who had discharge diagnosis of SSE and SSH were identified through appropriate ICD-9 codes. The secondary diagnoses of ESRD and CKD were also identified by ICD-9-CM codes. Multivariate logistic regression was performed using appropriate survey commands in STATA to find the outcomes of SS patient in ESRD and CKD. The analysis was adjusted for age, sex, race and other relevant co-morbid conditions. RESULTS: In 2007 NIS, there were an estimated 338,456 discharges with SSE and 232,848 discharges with SSH. The in hospital mortality from SSE was significantly higher in dialysis patients (OR 1.45; 95%CI 1.38-1.52) after adjusting for co-morbidities, but did not reach significance for CKD patients. Similarly, the in-hospital mortality of SSH was significantly higher in dialysis (OR 1.52;95%CI 1.44-1.60) and did not reach significance for CKD. Lengths of stay were significantly shorter in patients with CKD developing SSE or SSH and significantly longer in patients with ESRD. CONCLUSION: This observational study shows that in-hospital mortality of severe sepsis and septic shock is higher in ESRD patients but surprisingly unchanged in CKD when compared to the general population. Prospective studies are needed to investigate the causal relationship. CLINICAL IMPLICATIONS: ESRD but not CKD is a predictor for worse outcomes in patients with severe sepsis and septic shock.