Comparison Of LVAD Outcomes In Patients With Non-ischemic Cardiomyopathy And Ischemic Cardiomyopathy: Insight From National Readmission Database

Abstract

Background : Left ventricular assist device (LVAD) therapy remains a well-established therapy as either bridge to heart transplant or as a destination therapy for non-transplant candidates. Data related to the characteristics of readmissions by heart failure etiology (ischemic vs non-ischemic) after left ventricular assist device (LVAD) implantation is limited. We sought to compare readmissions between LVAD patients with non-ischemic cardiomyopathy (NICM) versus ischemic cardiomyopathy (ICM). Methods We identified patients from the Nationwide Readmission Database (NRD) who underwent LVAD placement between 2014-2017 using ICD-9 and 10 codes. Patients were classified into 2 groups based on the etiology of heart failure whether NICM or ICM. Outcomes were all-cause and cause-specific 30-day readmissions. Multivariable regression was conducted for 30 days readmission adjusting for patient demographics, hospital characteristics, and Elixhauser Comorbidity Index. Results We identified 7,001 recipients, 3292 having NICM as etiology of end-stage heart failure, and 3709 ICM. 28% of NICM and 30% of ICM LVAD recipients were readmitted within 30 days. NICM group was younger with more female patients compared with the ICM group (Table 1). For NICM versus ICM LVAD recipients; LOS of the index admission was 45 vs 39 days, P <0.001, cost of index admission was 995,739 ±38,692 vs 918,877 ±33,899 $, p=0.01, LOS of first readmission was 10.8±0.75 vs 11.3±0.75 days, p=0.67, cost of first readmission was 135,223 ±13,590 vs 117,270 ±10,115$,p=0.26. The most common cause of readmissions for NICM vs ICM were (ventricle tachycardia: 32 vs 25%, P=0.05), (Coagulation disorders: 21 vs 14%, P=0.001), (GI bleed: 15 vs 23%, P = 0.013), (acute heart failure: 29 VS 27%, P >0.05), (Atrial fibrillation/flutter: 40 Vs 37%, P >0.05), (AKI: 25 vs 24%, P >0.05) and (Infections: 12 vs 14%, P>0.05). ICM LVAD recipients have similar odds of 30-day readmission compared to NICM (OR=1.08, P=0.4). Renal failure patients have higher odds of 30-day readmission compared to non-renal failure (OR=1.23, P=0.005). Conclusion : Despite similar 30-day readmissions between patients with ICM vs NICM, post LVAD implantation, there are significant differences in the causes of readmissions. Further studies to explore these differences may help tailor preventive efforts aimed at preventing readmissions based on etiology of cardiomyopathy. : Left ventricular assist device (LVAD) therapy remains a well-established therapy as either bridge to heart transplant or as a destination therapy for non-transplant candidates. Data related to the characteristics of readmissions by heart failure etiology (ischemic vs non-ischemic) after left ventricular assist device (LVAD) implantation is limited. We sought to compare readmissions between LVAD patients with non-ischemic cardiomyopathy (NICM) versus ischemic cardiomyopathy (ICM). We identified patients from the Nationwide Readmission Database (NRD) who underwent LVAD placement between 2014-2017 using ICD-9 and 10 codes. Patients were classified into 2 groups based on the etiology of heart failure whether NICM or ICM. Outcomes were all-cause and cause-specific 30-day readmissions. Multivariable regression was conducted for 30 days readmission adjusting for patient demographics, hospital characteristics, and Elixhauser Comorbidity Index. We identified 7,001 recipients, 3292 having NICM as etiology of end-stage heart failure, and 3709 ICM. 28% of NICM and 30% of ICM LVAD recipients were readmitted within 30 days. NICM group was younger with more female patients compared with the ICM group (Table 1). For NICM versus ICM LVAD recipients; LOS of the index admission was 45 vs 39 days, P <0.001, cost of index admission was 995,739 ±38,692 vs 918,877 ±33,899 $, p=0.01, LOS of first readmission was 10.8±0.75 vs 11.3±0.75 days, p=0.67, cost of first readmission was 135,223 ±13,590 vs 117,270 ±10,115$,p=0.26. The most common cause of readmissions for NICM vs ICM were (ventricle tachycardia: 32 vs 25%, P=0.05), (Coagulation disorders: 21 vs 14%, P=0.001), (GI bleed: 15 vs 23%, P = 0.013), (acute heart failure: 29 VS 27%, P >0.05), (Atrial fibrillation/flutter: 40 Vs 37%, P >0.05), (AKI: 25 vs 24%, P >0.05) and (Infections: 12 vs 14%, P>0.05). ICM LVAD recipients have similar odds of 30-day readmission compared to NICM (OR=1.08, P=0.4). Renal failure patients have higher odds of 30-day readmission compared to non-renal failure (OR=1.23, P=0.005). : Despite similar 30-day readmissions between patients with ICM vs NICM, post LVAD implantation, there are significant differences in the causes of readmissions. Further studies to explore these differences may help tailor preventive efforts aimed at preventing readmissions based on etiology of cardiomyopathy.