How Can We Predict Bi-Ventricular Assist Device Requirement?

S380

The Journal of Heart and Lung Transplantation, Vol 35, No 4S, April 2016

1( 061)

Methods: In total, 116 patients (mean age, 38 ± 14 years), who underwent hemodynamic assessments preoperatively including 12 BiVAD patients, and had been followed at our institute from 2003 to 2015, were included. We performed logistic regression analyses to find predictors of BiVAD requirement among preoperative baseline variables. Results: Uni/multivariate logistic regression analysis indicated that RVSWI and pulmonary vascular resistance (PVR) were independent predictors of BiVAD requirement (p < 0.05 for both). In addition, all patients were classified into 4 groups: (1) normal (N group: RVSWI > 5 g/m, PVR < 3.7 WU), (2) pulmonary hypertension (P group: RVSWI > 5, PVR > 3.7), (3) RV failure (R group: RVSWI < 5, PVR < 3.7), and (4) both pulmonary hypertension and RV failure (RP group: RVSWI < 5, PVR > 3.7), and examined (see Fig. N indicates the number of BiVAD patients). Most of the patients in the RP group (75%), with acutely depressed hemodynamics and inflammatory responses in the myocardium, required BiVAD. Overall, patients with BiVAD had a worse survival rate as compared with those with LVAD alone. Conclusion: High PVR in addition to low RVSWI effectively predicts BiVAD requirement. These predictors must be useful to construct a therapeutic strategy using VAD.

Understanding Perceived Racial and Gender Differences in LVAD Use on the Transplant Waitlist A.R. Alejos ,1 D. Schaubel,2 J. Magee,3 M. Colvin,3 R. Sung.3  1University of Michigan Medical School, Ann Arbor, MI; 2University of Michigan School of Public Health, Ann Arbor, MI; 3University of Michigan Health System, Ann Arbor, MI. Purpose: Previous studies have shown that among heart failure patients, African Americans and women are less likely to have Left Ventricular Assist Devices (LVAD). LVADs confer 30 days of status 1A and indefinite status 1B to individuals on the transplant waitlist yet the demographics of LVAD use and how they compare to those waitlisted without LVAD has not been fully characterized. We evaluate the demographics of LVAD use at time of listing focusing on racial and gender differences. Methods: Using data from the Scientific Registry of Transplant Recipients, we conducted a retrospective review of 33,362 adult heart transplant candidates who were on the wait list between January 1st 2000 and December 31st 2012. Of chief interest was the relationship between having an LVAD and race or gender. Logistic regression was used to model the covariate-adjusted odds of having an LVAD at listing. The resulting logistic model fit well (C index =  0.82). Results: LVADs were present at time of listing in 4,714 candidates listed for transplant during the study period. African-Americans and Hispanics were more likely, and females less likely, to have an LVAD in univariate analyses. However, after adjusting for height, weight, region, prior cardiac surgery, cardiac diagnosis and insurance status there was no statistically significant difference in likelihood of having an LVAD at listing by race or gender. Conclusion: Among individuals waitlisted for heart transplant, apparent race and gender disparities to LVAD access are mediated by other factors, including geography. This contrasts findings for heart failure patients not listed for transplant, and should be further investigated to see how LVAD affects access to and movement on the waitlist.

Table 1. Odds of having a LVAD at wait listing by race and gender. * denotes p< 0.05

Factor

Group

Unadjusted

Covariate Adjusted (Not geographic)

Race

African American Hispanic Asian Other White Female Male

1.29*

1.06

1.08

1.07

0.93 1.43* 0.87 1 (reference) 0.88* 1 (reference)

0.84* 1.21* 0.66* 1 0.91* 1

0.92 1.2 0.7 1 0.90* 1

1 1.24 0.74 1 0.92 1

Gender

Covariate Adjusted (UNOS region)

Covariate Adjusted (DSA)

1( 062) WITHDRAWN 1( 063) How Can We Predict Bi-Ventricular Assist Device Requirement? T. Imamura ,1 K. Kinugawa,1 D. Nitta,2 M. Hatano,2 O. Kinoshita,3 K. Nawata,3 M. Ono.3  1Department of Therapeutic Strategy for Heart Failure, Graduate School of Medicine, University of Tokyo, Tokyo, Japan; 2Department of Cardiovascular Medicine, Graduate School of Medicine, University of Tokyo, Tokyo, Japan; 3Department of Cardiac Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan. Purpose: Patients receiving biventricular assist device (BiVAD) have very poor prognosis. Although the right ventricular stroke work index (RVSWI) is a good index for RV function, a low RVSWI alone is not simply an indicator for the need for a right ventricular assist device (RVAD) at the time of left VAD implantation. In the clinical setting, some patients with low RVSWI can avoid RVAD therapy at the timing of left VAD implantation. What is a more precise indicator for the need for a BiVAD?

1( 064) Pre-Extra Corporeal Membrane Oxygenation (ECMO) MELD Score May Predict Meaningful Survival M. Byku ,1 R. Piotrowski,1 A. Itoh,2 S.J. Larue,1 J.M. Vader,1 C.L. Holley,1 G.A. Ewald.1  1Cardiology, Washington University in St. Louis, Saint Louis, MO; 2Cardiothoracic Surgery, Washington University in St. Louis, Saint Louis, MO. Purpose: Cardiogenic shock is associated with high morbidity and mortality. ECMO has become an important means of acute support in the setting of cardiogenic/respiratory shock, serving as a bridge to hemodynamic stability and more definitive management. Despite marked increase in its use, little is known about patient selection and outcomes. The goal of this study was to perform a retrospective analysis of ECMO cases at a single institution to identify factors associated with recipient 1-year survival. Methods: We identified 94 cases of ECMO between January 2011 and December 2013. 80% of the cases were implanted for management of cardiac conditions, (cardiac arrest, cardiogenic shock, severe right ventricular failure). 20% of patients had respiratory failure. Patient demographics, Model for End stage Liver Disease (MELD) score, requirement of hemodialysis and ECMO related complications were analyzed with regards to survival. Results: Overall, 50% of the patients were alive at 30 days and 30% were alive at 1 year. Survivors at 1 year were younger than non-survivors (45 vs 52 years, p≤  0.004). Patients receiving ECMO for cardiac conditions had a higher mortality compared to those with a primary pulmonary problem (80% alive at 1 year). There was a significant association between peri-procedural need for dialysis, (within two weeks of ECMO), with death at 1 year (p ≤ 0.004). Among cardiac patients, the pre-ECMO MELD score was lower in those who were alive at 1 year compared to those that were not (16.7 vs 21.3, p ≤ 0.03), with an upper limit score of 20 on the