Continuous-flow left ventricular assist device implantation as a bridge to transplantation or destination therapy: racial disparities in outcomes.

BACKGROUND

There is a paucity of data assessing racial disparities in outcomes after left ventricular assist device (LVAD) implantation. This may be due to the relatively low percentage of African American (AA) patients at a given center. Given the high proportion of AAs in our patient population, we sought to evaluate outcomes of LVAD implantation in AAs vs Caucasians.

METHODS

We stratified 88 LVAD patients by AA or Caucasian race. Variables were compared using 2-sided t-tests, chi-square tests, Cox proportional hazards models, and log-rank tests to determine whether a difference existed between AAs and Caucasians and whether race was a significant independent predictor of outcome.

RESULTS

AAs represented 36.4% (32 of 88) of our LVAD patients. The two groups did not differ significantly in the incidence of hypertension, diabetes, or chronic renal insufficiency, reoperation rates, pre-operative body mass index, left ventricular ejection fraction, central venous pressure, pulmonary capillary wedge pressure, pulmonary artery pressure, or right ventricular function. Compared with Caucasians, AAs were significantly younger (48.6 vs 54.8 years, p = 0.019), and had a significantly higher mean body surface area (p = 0.009) and a higher rate of non-ischemic dilated cardiomyopathy (61% vs 39%, p = 0.008). No significant difference was found in 30-day (p = 0.12), 180-day (p = 0.166), or 360-day (p = 0.18) survival. Analysis by univariate Cox proportional hazard models (hazard ratio [95% confidence interval]) showed race was not an independent predictor of 30-day (4.5 [0.56-35.94], p = 0.157), 180-day (3.9 [0.48-31.95], p = 0.2), or 360-day survival (1.8 [0.6-5.71], p = 0.286). Age and pre-operative renal failure were the only independent predictors of survival at 30 days (1.1 [1.02-1.19], p = 0.019; 4.99 [1.24-20], p = 0.023, respectively), 180 days (1.09 [1-1.18], p = 0.041; 4.14 [0.99-17.39], p = 0.05), and 360 days (1.05 [1-1.1], p = 0.044; 2.52 [0.94-6.75], p = 0.05). Analysis by a multivariate logistic regression model showed age and chronic renal failure were no longer statistically significant for survival at 30, 180, and 360 days.

CONCLUSIONS

Although multiple studies have demonstrated that AAs experience worse outcomes after coronary artery bypass grafting, heart transplantation, and valve surgery, we did not find similar results in our LVAD population. More rigorous pre-operative LVAD workup, including an evaluation by a multidisciplinary team, along with more intense post-operative follow-up, may explain improved outcomes in AAs after LVAD implantation compared with other cardiac surgical procedures, although additional analysis is required.