Impact of kidney disease on left ventricular assisted device outcomes: insight from National Readmission Database

Abstract

\n \n \n Improvements in device technology, appropriate patient selection and timing of left ventricular assist device (LVAD) implantation have allowed this technique as the preferred destination therapy for advanced end-stage heart failure (HF) patients whom are not candidates for heart transplantation. Unfortunately, up to a third of these patients still experience poor outcomes following LVAD implantation. Although some clinical factors have been identified as potential risk factors for these poor outcomes, there is paucity of data regarding the impact of baseline kidney disease following LVAD implantation.\n \n \n \n We used the Nationwide Readmission Database (NRD), identifying patients who underwent implantable LVAD placement 2010–2017 using ICD-9 and 10 codes. Patients were stratified into 3 mutually exclusive groups based on presence and severity of chronic kidney disease (CKD) during the LVAD placement hospitalization: non-CKD, CKD and end stage renal disease requiring dialysis (CKD-D). Outcomes were all-cause and cause-specific 30-day readmissions as well as 90-day readmission.\n \n \n \n Within 30 days after discharge 28.5% of patients were readmitted. Of those without CKD, 27.8% were readmitted, compared to 29% of those with CKD and 31% of those with CKD-D. Compared to non-CKD (adjusted for demographics, index hospitalization and chronic comorbidity factors, and year), odds of 30-day readmission were 1.542 [95% CI 1.028–2.313] times higher for those with CKD-D, whereas those with CKD had similar 30-day readmission risk (OR 1.074 [95% CI 0.951–1.213]). Those with CKD and CKD-D had higher risk of 30-day readmission for acute (HF), bleeding and transfusion need.\n Within 90 days after discharge 48.88% of patients were readmitted. Of those without CKD, 46.2% were readmitted, compared to 50%of those with CKD and 45% of those with CKD-D. Odds of 90-day readmission were 1.203 (95% CI 1.088–1.330) times higher for those with CKD than non-CKD group, Also those with CKD-D requiring dialysis had higher 90-day readmission risk (OR 1.287 [95% CI 0.954–1.736]) when compared to non-CKD. Those with CKD had higher risk of 90-day readmission for acute (HF), gastrointestinal bleeding and arrhythmia while those with CKD-D had higher risk of 90-day readmission for DVT and anemia. There was no difference in 30-day and 90-day readmission between all groups for stroke, device related thrombosis or infection.\n \n \n \n It appears, based on our analysis, that although having CKD does not confer additional risk for either 30 or 90-days readmissions; these post LVAD patients have a greater risk of acute HF exacerbation, bleeding and arrhythmias. We believe that these results are very important and should be included into future risk models when deciding on LVAD implantation.\n \n \n \n Type of funding sources: None.\n