been supported with the HMII device. LVAD groups were matched for age, etiology, and support duration. Data from hearts supported with the HM device replicate what we have previously shown, demonstrating recovery of SERCA protein to levels mea- sured in non-failing controls. Data from patients supported with the HM II device failed to show recovery of SERCA protein, with levels remaining the same as those measured in non-supported failing hearts. These data suggest that hemodynamic un- loading of the failing heart with a pulsatile LVAD is able to reverse the down-regu- lation of SERCA, a key calcium regulatory protein, and thus move the heart in a direction compatible with recovery. Continuous flow LVAD support caused no mea- surable reversal of SERCA protein. Lack of SERCA recovery may be limiting if the goal of LVAD support is reversal of heart failure. 050 Multiple Left Ventricular Assist Device-Associated Hospital Readmissions Are Associated with Poorer Survival: A Single Center Experience Andrea M. Cordero-Reyes, Rey P. Vivo, Carlos M. Orrego, Erick E. Suarez, Matthias Loebe, Brian Bruckner, Guillermo Torre-Amione, Barry Trachtenberg, Jerry D. Estep, Arvind Bhimaraj; The Methodist Hospital, Houston, TX Introduction: Repeat hospital admissions strongly predict higher mortality in pa- tients with heart failure (HF) but this relationship is not clearly established among patients implanted with continuous-flow left ventricular assist devices (LVAD) with prolonged support. Hypothesis: We hypothesize that LVAD-associated readmissions have a negative impact on overall survival in this population. Methods: A total of 144 patients received an LVAD from 2008-2011 in a single center. Among patients who were discharged alive after LVAD implant, LVAD-associated hospital admis- sions were analyzed up to 1 year. Non-LVAD associated conditions and elective pro- cedures were excluded. Patients who received a heart transplant within 1 year of LVAD were censored. The primary endpoint was death up to 1 year from last admis- sion (survival was reset to a new time zero after each admission). Kaplan Meier sur- vival curves were constructed for patients with 0, 1-2 and O3 readmissions, p-values ! 0.05 were considered significant. Results: The final study population was com- posed of 117 patients of these 61% were categorized as Destination Therapy (DT). Mean age was 56 6 13 years, 73% were male. Mean time to 2nd readmission was 126 6 90 days. There were no readmissions in 44 (38%) patients; 46 (39%) and 27 (23%) had 1-2 and O3 readmissions, respectively. The average length of stay was 15 6 21 days. Leading causes for readmissions were device-related infections (33%), bleeding events (26%), residual HF (11%), stroke (7%), pump malfunction (6%) and arrhythmias (6%). There was a significant and indirect association between frequency of readmissions and overall survival: mean survival in patients with 0, 1-2 and O3 readmissions was 89%, 72% and 67%, respectively (p 5 0.03) (Figure). Conclusions: Based on our center’s experience in patients supported with continu- ous-flow LVAD primarily as destination therapy, more frequent hospital readmissions are associated with poorer prognosis. Measures to curb multiple hospitalizations may help improve long-term survival. Table. Proposed LVAD Clot Score to detect device malfunction. Echocardiographic parameter Point D LVEDd !0.6 cm 1 D AVOD !8.6 ms 1 D Mitral DT !73 ms 1 D 5 difference from lowest to highest pump speed. 051 Race and Worsening Renal Function after Solid Organ Transplantation David Lanfear 1 , Ali Shafiq 2 , Edward Peterson 3 , Karen Wells 3 , Yong Hu 3 , L. Keoki Williams 1 ; 1 Henry Ford Hospital, Detroit, MI; 2 Henry Ford Hospital, Detroit, MI; 3 Henry Ford Hospital, Detroit, MI Background: Calcineurin-inhibitors like cyclosporine (CYA) and Tacrolimus (TAC) have stimulated a new era in transplantation medicine resulting in substantial im- provement in success rates after solid organ transplantation. However calcineurin-in- hibitor induced nephrotoxicity frequently complicates transplantation. Objective: African American (AA) patients suffer a disproportionate burden of renal failure in a variety of settings. We wished to explore whether calcineurin-inhibitor induced nephrotoxicity differed by race. Methods: We identified 2,855 patients undergoing initial solid organ transplantation (cardiac, lung, liver and renal) between January 2000 and June 2012. TAC, CYA and serum creatinine levels were screened through electronic records. Creatinine level at discharge from transplantation was considered baseline. The average creatinine within subsequent 6 month intervals was calculated. Change in creatinine was calculated for each new creatinine value. We tested the as- sociation between cumulative TAC or CYA exposure and serial changes in creatinine after adjusting for age, gender, hypertension and diabetes. P values !0.05 were con- sidered positive for main effects. P!0.1 were considered positive for interactions. Results: There were 2,162 patients treated with TAC (1342 white, 820 AA) and 693 patients treated with CYA (426 white, 267 AA). TAC exposure was associated with rising creatinine (p!0.001) and this association was more pronounced among AA patients compared to white patients. Among AA patients a 1 ng/ml higher TAC level corresponded to 0.09 mg/dl average rise in creatinine, whereas in white patients TAC was associated with a numerical improvement of 0.06/ mg/dl creatinine (interaction p50.001). No overall effect of CYA exposure on creatinine was identi- fied (p50.9) nor any significant difference by race (p50.17). Conclusion: African American patients may be particularly vulnerable to TAC-induced renal dysfunction when compared to white patients. These racial differences may help guide choice of therapy in different racial groups. 052 Ongoing Substance Abuse at the Time of Left Ventricular Assist Device Implantation Is Associated with Poor Outcomes Rebecca J. Cogswell, Lillian Bauman, Elisa Smith, Aimee Hamel, Angela Herr, Sirtaz Adatya, Justin Roberts, Monica Colvin-Adams, Marc Pritzker, Ranjit John, Peter Eckman; University of Minnesota, Minneapolis, MN Introduction: Advanced heart failure teams are often faced with the decision of whether or not to implant a left ventricular assist device (LVAD) in patients who have end-stage heart failure but ongoing substance abuse. The outcomes of these pa- tients after LVAD implantation are unknown. Hypothesis: Ongoing substance abuse at the time of LVAD implantation will be associated with higher mortality and loss to follow-up compared to controls. Methods: Cases were defined as patients with a his- tory of substance abuse or dependence by DSM IV criteria who continued to use within 6 months prior to LVAD implantation. Controls were then pulled in a 2:1 fash- ion from University of Minnesota LVAD database matched by age, INTERMACS profile and year of implant. The primary outcome was combined death or loss to fol- low-up. Secondary outcomes included ever being listed for cardiac transplantation or successfully undergoing cardiac transplantation. Results: The cohort consisted of 21 cases and 42 controls matched by age (43.5 vs. 43.3 years, p5 0.94) and INTER- MACS profile (3.75 vs. 3.58, p 50.73). During an average follow-up period of 2.75 years (+/- 1.48), the active substance abuse group had a 66% lower rate of being listed for transplant, (5/21 vs. 26/42, rate ratio 0.34, p !0.0001) and a 73% lower rate of undergoing cardiac transplant (2/21 vs. 15/42, rate ratio 0.27, p ! 0.05). The sub- stance abuse group had 3.2 times the rate (hazard) of death or loss to follow-up The 17 th Annual Scientific Meeting  HFSA S19