Readmission Rates and Costs after Transcatheter vs Surgical Aortic Valve Replacement in Dialysis Dependent Patients
Yas Sanaiha1, Aditya Mantha2, *Richard J Shemin1, Peyman Benharash1
1University of California, Los Angeles, Los Angeles, CA;2University of California, Irvine, Irvine, CA
OBJECTIVES: Patients with End Stage Renal Failure (ESRD) are at increased risk for developing aortic valve pathology. Diffuse systemic disease associated with ESRD also confers high morbidity with cardiovascular surgical procedures. In the present era of value-based healthcare delivery, a comparison of Transcatheter (TAVR) and surgical aortic valve replacement (SAVR) readmission performance in the dialysis dependent population is warranted. METHODS: All adult patients undergoing TAVR and SAVR from 2011-2014 were identified using the National Readmission Database containing data for nearly 50% of US hospitalizations. Patients with chronic kidney disease requiring hemodialysis were identified using the International Classification of Diagnoses (ICD-9) and categorized as dialysis TAVR versus dialysis SAVR. Patients with a diagnosis of acute on chronic renal failure were excluded from this study. A multivariate regression model was used to identify independent predictors of 30-day readmission and costs. Primary diagnoses for the readmission visit were tabulated using Diagnosis Related Group codes. RESULTS: Of the estimated 352,430 patients undergoing aortic valve replacement, 3.3% were hemodialysis dependent (TAVR 1,973, SAVR 10,224). Compared to dialysis SAVR patients, dialysis TAVR patients were significantly older (76.9 vs 65.6, P<0.0001) and had higher Elixhauser Comorbidity Indices (7.6 vs 7.3, P=0.005), rates of congestive heart failure (18.8 vs 14.5 %, P=0.014), and chronic lung disease (37 vs 22%, P<0.0001). Thirty-day readmission rate (TAVR 28.9 vs SAVR 30.1%, P=0.40) and index hospitalization mortality rate (14.4 vs 15.8 %, P=0.38) were similar between the two cohorts. The use of TAVR and readmission rates in ESRD patients increased while the associated mortality (26.0 to 14.4%, P=0.005) decreased (Figure 1 & Figure 2). After adjustment for baseline differences, approach to valve replacement did not impact odds of readmission (P=0.24) while chronic pulmonary (OR 1.4, P<0.0001) and liver disease (OR 1.6, P=0.009) as well as infectious (OR 1.3, P=0.01) and thromboembolic (OR 1.5, P<0.0001) complications were strongly associated with rehospitalization. Adjusted costs were higher for TAVR patients by $7,750(P=0.003). Cardiac causes of readmission were most common for both TAVR and SAVR groups, followed by infectious and pulmonary complications. The dialysis TAVR cohort had increased rates of readmission for heart failure exacerbation (19.4 vs 12.9%, P=0.04) and myocardial infarction (1.3 vs 0.3 %, P=0.03) compared to the dialysis SAVR group.
CONCLUSIONS: Based on this national cohort of dialysis dependent patients, SAVR resulted in similar readmission rates and lower adjusted costs of care. Within the CKD population, prevention of thromboembolic and infectious complications may serve as suitable areas for quality improvement and may reduce total resource utilization.
Back to 2018 Program