239 Hospital Readmissions Are a Common Occurrence in the Current Era with Continuous Flow (CF) LVADs F. Kamdar, 1 J. Rasmussen, 1 P. Eckman, 1 K. Liao, 1 B. Milavitz, 2 M. Colvin-Adams, 1 R. John. 11 University of Minnesota, Minneapolis, MN; 2 Fairview Health, Minneapolis, MN. Purpose: CF LVADs are an accepted option for patients with end-stage heart failure. However, morbidity and the resulting hospital admission may limit the application of CF LVADs to less sick patients. The objective was to study hospital readmissions and associated costs in patients with the HeartMate II LVAD. Methods and Materials: We reviewed all patients who received HM II at a single center from 10/05-1/11, excluding patients who died prior to discharge. Total cost of readmissions was obtained from a retrospective review of claims data. Results: Of 129 discharged patients, 108 were BTT. 45 (35%) patients had no readmissions and 84 (65%) had at least 1 readmission. There were a total of 312 readmissions in this patient group. The median readmission per patient was 1 (range:1-29) and the mean number of readmission per patient was 2.6  3.8. The mean time to readmission after LVAD implant was 139  16 days. The mean length of stay for readmission was 7  9 days. The main reasons for readmission included heart failure exacerbation (38, 12.2%), gastrointestinal bleeding (29, 9.3%), driveline infections (26, 8.3%), and elective procedures (22, 7.1%). Freedom from first readmission was 61% and 39% at 3 and 6 months. There was no significant difference between patients who were readmitted and those who were not in regards to age, gender, etiology, or duration of support (p = NS). The mean total cost per first readmission was $25,713 (interquartile range $5,935 -23,793). Conclusions: Readmission following LVAD implantation is a common and costly event. Further research focused at decreasing GI bleeding and driveline infections as well as identifying risk factors for readmission will improve the quality of life and costs associated with CF VADs. 240 Continuous Flow Left Ventricular Assist Devices Do Not Impair Central Sympathetic Nerve Traffic J. Tank, 1 D. Malehsa, 2 K. Heusser, 1 K. Hegemann, 1 A. Diedrich, 3 C. Bara, 2 J. Jordan, 1 M. Strueber. 21 Institute of Clinical Pharmacology, Hannover Medical School, Hannover, Lower Saxony, Germany; 2 Dept. of Cardiac, Thoracic, Transplantation, and Vascular Surgery, Hannover Medical School, Hannover, Lower Saxony, Germany; 3 Dept. of Medicine, Division of Clinical Pharmacology, Vanderbilt University, Nashville, TN. Purpose: The superior clinical outcome of new continuous flow left ven- tricular assist devices (LVAD) challenges the physiological dogma that cardiovascular homeostasis requires pulsatile blood flow and pressure. We hypothesized that continuous flow LVAD impair baroreflex regulation of heart rate (HR) and sympathetic nerve traffic (MSNA). Methods and Materials: We included 9 male patients with end-stage sys- tolic heart failure (age range: 26-61 years; BMI range 19-28 kg/m 2 ) who had been implanted with a continuous flow LVAD (HeartWare HVAD, time since implantation: 0.6-2.6 years). Electrocardiogram, respiration (thoracic imped- ance), beat-by-beat finger blood pressure (BP), and systolic brachial BP (Doppler ultrasound) were determined. We recorded MSNA from the right peroneal nerve using microneurography. After baseline measurements for 15 minutes we performed autonomic function testing including deep breathing and a Valsalva maneuver. Finally, we increased LVAD speed in seven pa- tients. HR variability and spontaneous baroreflex sensitivity were analyzed. Results: In 8 patients, we obtained good quality sympathetic nerve record- ings. Finger BP amplitude was 142 mm Hg (range 4-23 mm Hg). Brachial BP was 994 mm Hg. MSNA bursts showed a normal morphol- ogy, were linked to the cardiac cycle, and were suppressed during BP increases. Mean burst frequency was similar compared to age and BMI matched healthy controls in 3 patients and slightly increased in 5 patients. MSNA was not increased at maximum speed of 3200 rpm (speed at baseline: 286644 rpm) despite reduced pulse pressure (p0.01). Conclusions: Our data support the idea that continuous flow HeartWare HVAD does not impair cardiovascular reflexes and may ameliorate sym- pathetic overactivity. 241 Impact of Malnutrition on Early Outcomes Following Implantation of Continuous Flow Left Ventricular Assist Devices D.J. Goldstein, 1 W. Holman, 2 N. Moazami, 3 R. John, 4 D. Naftel, 2 F. Pagani. 51 Cardiothoracic Surgery, Montefiore Medical Center, Bronx, NY; 2 Cardiothoracic Surgery, University of Alabama, Birmingham, AL; 3 Cardiothoracic Surgery, Minneapolis Heart Institute, Minneapolis; 4 Cardiothoracic Surgery, University of Minnesotta, Minneapolis; 5 Cardiothoracic Surgery, University of Michigan, Ann Arbor. Purpose: Malnutrition is a well-recognized predictor of postoperative mor- bidity, mortality and prolonged length of stay among surgical patients. Patients with heart failure are often plagued by malnutrition related to duration of illness, hospitalization and malabsorption. The impact of mal- nutrition on early outcomes following implantation of contemporary left ventricular assist devices (LVAD) has not been evaluated. Methods and Materials: The INTERMACS National Registry was used to identify all adult recipients of implantable continuous flow LVAD between June 2006 and March 2011. Serum albumin (alb) [ or  3.5g/dL], serum pre-albumin (pre-alb) [ or  15mg/dL], hemoglobin (Hgb) [ or  10mg/dL] and body mass index (BMI) [ or  20] were used as correlates of nutritional status. Stepwise linear regression was used to assess the combined effects of the 4 correlates on ICU and total length of stay (LOS). Linear regression was used to estimate effect of nutritional parameters, after adjustment for known risk factors, on ICU and total length of stay, 30-day infection and survival. Results: 2900 patients were included in the study for a total of 2194 pt-years of support. Mean follow up time was 9.1 months. Alb  3.5g/dL, pre-alb 15 mg/dL and Hgb  10mg/dl were associated with significantly longer ICU and total LOS. When adjusted for known risk factors, only Hgb  10mg/dl was predictive of prolonged ICU stay. On univariate analysis, Alb 3.5 g/dL was associated with lower survival while Hgb  10mg/dL was associated with lower overall survival and time to first infection (Fig 1). BMI  20 [OR 1.041, p=0.01) and alb  3.5 g/dL [0.546, p=0.02) were multivariate predictors of 30 day mortality. Conclusions: Easily obtainable correlates of nutritional status can be used to estimate risk of perioperative death and infection as well as postopera- tive LOS in patients undergoing CF LVAD implantation. Interventions to nullify these factors before or immediately after surgery may diminish their effect on outcome. 242 Implantation of a Centrifugal Pump as Left Ventricular Assist Device (LVAD) through a Minimized Approach: Upper-Hemisternotomy Plus Anterior-Lateral-Thoracotomy J.D. Schmitto, M. Avsar, N. Schuetz, S. Schwabe, D. Malehsa, U. Molitoris, A. Haverich, M. Strueber. Hannover Medical School, Hannover, Germany. S87 Abstracts