Eastern Cardiothoracic Surgical Society

Regional Availability of Low-Risk Donors Does not Drive Organ Acceptance in Heart Transplantation
Ahmet Kilic, Asia McDavid, Larry Hromalik, Bryan Whitson, Don Hayes, Dmitry Tumin, Ohio State University Wexner Medical Center, Columbus, OH


OBJECTIVE: Decisions to accept hearts from donors with known risk factors may be influenced by the regional availability of low-risk donors. In this study, we sought to test associations between donor risk factors and acceptance of donor hearts in the context of availability of low-risk donors in each United Network for Organ Sharing (UNOS) region.

METHODS: The UNOS registry was queried for deceased donors recovered between 2006-2015. The outcome was acceptance of donor hearts for transplantation. Six donor risk factors were included: age >50 years; female gender; hypertension; diabetes; cerebrovascular accident as cause of death; and left ventricular (LV) ejection fraction <50%. Donors with none of the 6 risk factors were considered low-risk. Regional monthly percentages of low-risk donors were interacted with donor risk factors in multivariable mixed-effects logistic regression.

RESULTS: Hearts were recovered for transplantation from 23,522 (29%) of 80,002 donors. Region-specific monthly percentages of low-risk donors were 25 ±6%. The most and least common risk factors were female gender (41%) and LV dysfunction (11%), respectively. Each donor risk factor, but not the availability of low-risk donors in the prior month, was associated with odds of heart utilization (Table SA-19-1). LV dysfunction most strongly predicted heart non-utilization. LV dysfunction was more predictive of heart non-utilization when there have been more low-risk donors in the prior month, as indicated by the significant interaction term.

CONCLUSIONS: Decisions regarding organ acceptance in heart transplantation were mostly not sensitive to recent shortages of low-risk donors. The single most important factor for organ utilization was left ventricular function.

Table SA19-1
Multivariable mixed-effects logistic regression of heart utilization for  transplantation, 2006-2015 (N = 80,002)

Model 1a Model 2a
Variable OR

(95% CI)

P OR

(95% CI)

P
Donor risk factors
  Age >50 0.12

(0.12, 0.13)

<0.001 0.12

(0.12, 0.13)

<0.001
  Female gender 0.72

(0.69, 0.75)

<0.001 0.72

(0.69, 0.75)

<0.001
  Hypertension 0.37

(0.35, 0.39)

<0.001 0.37

(0.35, 0.39)

<0.001
  Diabetes 0.39

(0.35, 0.42)

<0.001 0.39

(0.35, 0.42)

<0.001
  Cerebrovascular COD 0.74

(0.71, 0.77)

<0.001 0.74

(0.71, 0.78)

<0.001
  LVEF <50% 0.08

(0.07, 0.09)

<0.001 0.08

(0.07, 0.09)

<0.001

a Mixed effects logistic regression models include random effects for UNOS region and month.

OR, odds ratio; CI, confidence interval; COD, cause of death; LVEF, left ventricle ejection fraction; UNOS, United Network for Organ Sharing