Out of Sequence Heart Transplants: Why, How Many, and to Whom

Abstract

Background:

Heart transplantation (HT) is governed by the Organ Procurement and Transplantation Network, which commissions 56 organ procurement organizations (OPO) to perform allocation. Each allocation follows a match list of eligible donors ranked by urgency. We sought to describe the prevalence of OOS HT, characterize donor and recipient characteristics, and assess if the centers and recipients of OOS donors differ from non-OOS donors.

Methods:

Using the Scientific Registry of Transplant Recipients all adult (≥18 years), single organ HT recipients between 1/18/2018–1/1/2024 were identified. We performed network analysis and compared in sequence vs OOS donors with multivariable logistic regression.

Results:

Among 18,394 HT, 880 (4.8%) were involved an OOS allocation for any reason (52% were OPO initiated). Four transplant centers received one-quarter (4.3%-8.5% each) of hearts allocated OOS. Six OPOs accounted for 27% of OOS allocations (4.0%-6.6% each). Graft survival was similar at 1-year between groups (90.6% vs 90.9%, p=0.79). Recipient characteristics most associated with OOS heart allocation included blood type O, UNOS Status 4 or 6, and the absence of surgical mechanical circulatory support (MCS). The most common reason for OOS allocation was expedited heart placement and the frequency was stable over time.

Conclusions:

OOS heart allocation occurs in 4.8% HT. Nearly all centers and OPOs participated in OOS allocation, but certain were more prevalent. Recipients of OOS hearts more commonly had a lower priority status (Status 4 or 6), blood type O, and did not have surgical MCS. Further study is needed to ensure a balance between utility and equity.

Introduction:

Allocation of organs to individuals in need of transplantation is governed by the United Network for Organ Sharing (UNOS), an organization that administers the Organ Procurement and Transplantation Network in the United States. The country is separated into 11 regions and subdivided into 56 donor service areas that are serviced by local organ procurement organizations (OPO) that work to coordinate the recovery and allocation of donor hearts. Prior to the 2018 heart allocation policy change, patients within a transplant region were served by a local OPO that allocated hearts to local Status 1A or 1B candidates preferentially, followed by candidates with UNOS 1A or 1B designation in an adjoining 500-mile concentric circular zone. When the adult heart transplantation allocation policy was amended in 2018, the stated goal was to improve access to individuals with the greatest medical need through improved stratification of risk.¹ Patients are now assigned a status based on urgency ranging from Status 1 (most urgent, e.g., on ECMO) to Status 6 (least urgent, e.g., at home on oral medical therapy; Status 7 candidates are inactive). Potential donor hearts are allocated to patients of the highest urgency (Status 1 & 2) within 500 nautical miles of the donor, irrespective of the donor service area, followed by Status 3 candidates within 250 nautical miles. The donor heart is then offered to Status 1 and Status 2 candidates within 1,000 nautical miles, then Status 4 candidates within 250, nautical miles, followed by Status 3 candidates within 500 nautical miles, and continuing through a possible 68 stages.

When a donor heart is allocated, a match run of ranked candidates is generated according to the allocation algorithm, and the listing center for each candidate may either accept or decline an offered heart. Of potential heart transplant (HT) donors, 44% are accepted for transplantation.² Since the 2018 allocation policy change, hearts have been allocated at a median sequence number of 8 (interquartile range 3–24). Nonuse of a recovered heart is uncommon but occurs in 1.3% of cases (6.4% of donors >55 years of age, <1% of donors <18 years of age).³ There are instances when donor, recipient, or administrative circumstances arise that lead to a candidate being bypassed in the allocation sequence, often to limit nonuse, resulting in an out of sequence (OOS) allocation. OOS allocation has been known to exist, but the extent of these events was not well quantified. This was reported in kidney transplant in 2022⁴, and more recently has been reported to occur in nearly 20% of kidney transplants.⁵ OOS allocation gained widespread attention following a February 2025 New York Times story headlining, “The sickest patients are supposed to get priority for lifesaving transplants. But more and more, they are being skipped over.”⁶ With this in mind, we sought to describe the prevalence of OOS heart transplantation, characterize donor and recipient characteristics most often associated with OOS allocation, and assess if the centers and recipients of OOS donors differ from non-OOS allocated donors.

Methods:

Study Population and Data Source

Data for the study were obtained from the Scientific Registry of Transplant Recipients (SRTR). The SRTR data system includes data on all donor, wait-listed candidates, and transplant recipients in the US, submitted by the members of the Organ Procurement and Transplantation Network. The Health Resources and Services Administration, U.S. Department of Health and Human Services provides oversight to the activities of the Organ Procurement and Transplantation Network and SRTR contractors. The data reported here have been supplied by the Hennepin Healthcare Research Institute as the contractor for the SRTR. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the SRTR or the U.S. Government. All adult (Age≥18 years), single organ transplant recipients between October 18^th, 2018, and December 31^st, 2023, were included (the current allocation system), with follow-up through April 1^st, 2024. The SRTR standard analysis files for deceased donors and HT were merged with potential transplant recipient data to confirm that offers marked accepted in the match-run were transplanted by matching recipient-donor ID pairs in the potential transplant recipient and transplant files. Data regarding the donor acceptance practices for individual centers were obtained from the July 2021 SRTR program-specific reports. The overall acceptance ratio was utilized (which included the acceptance of donors with Public Health Service increased infectious risk, ejection fraction <60%, donor age ≥ 40 years, hard-to-place hearts [Sequence ≥ 50], and donor more than 500 miles away).

Donors that were allocated OOS were identified if a candidate on the waitlist was bypassed in the allocation sequence. OPO-initiated allocation exceptions were identified if the following refusal codes were included in the match-run: Operational OPO (861), donor medical urgency (862), offer not made due to expedited placement attempt (863), or other (798 or 898) where the free text explicitly specified one of the aforementioned reasons.⁴

Statistical Analysis

The primary aim of this study was to describe the prevalence of OOS HT, characterize donor and recipient characteristics most often associated with OOS allocation, and assess if the centers and recipients of OOS donors differ from non-OOS allocations. Missing data were assumed to be missing completely at random and handled with multiple imputation using a Markov Chain Monte Carlo method to generate ten imputations. Demographic and clinical variables were expressed as median (interquartile range [IQR]) for continuous variables and count (%) for categorical variables. Group comparisons were made with Chi-squared and Mann-Whitney U tests when appropriate. Relative risk was used to compare offer acceptance practices and calculated by comparing the donor heart acceptance rate to the expected donor heart acceptance rate. Kaplan-Meier survival analysis was performed to compare post-transplant survival. Univariable and multivariable logistic regression was performed to identify donor and recipient characteristics associated with OOS allocation. Multivariable models included covariates that had a p<0.10 on univariable assessment. A 2-tailed p-value of <0.05 was considered significant. Analyses were performed using SAS software version 9.4 (SAS Institute, Inc.). Network diagrams were constructed in Gephi (version 0.10.1) for all transplants during the study period, for transplants allocated OOS, and for transplants allocated OOS with an OPO-initiated allocation exception. Studies involving this de-identified dataset have been determined to be exempt from review by the Institutional Review Board of Columbia University Irving Medical Center.

Results:

There were 39,928 match runs from 38,595 donors, with 56 OPOs allocating hearts to 145 transplant centers resulting in 18,394 adult HT during the study period. Among the hearts that were transplanted, 880 (4.8%) involved a recipient being bypassed on the match list for any reason. During the study, the annual OOS transplant volume varied without a specific pattern: 207 (6.8%) in 2019, 161 (5.0%) in 2020, 138 (4.1%) in 2021, 130 (3.6%) in 2022, and 198 (4.9%) in 2023 (p<0.001 for any difference).

Network Analysis

Network analysis found that during the study period, a median of 100 HT were performed by each center (Interquartile range 22–198), with a median of 283 (IQR 148–442) hearts originating from each OPO (Figure 1A). Ninety-nine different transplant centers (68.3%) received hearts where there was an OOS allocation from forty-nine different OPOs (91%, Figure 1B). Among the centers receiving an OOS heart, the median number of OOS transplants was 6 (IQR 3–11). The median number of hearts allocated OOS per OPO was 11 (IQR 6–23).

Figure 1.

Panel diagram summarizing the main findings of the study including the overall frequency of OOS allocation, how donor characteristics were similar and differenced, and the frequency over time of OOS allocation focusing on both centers and OPOs.

There were OOS outliers for both transplant centers and OPOs. Four transplant centers received one-quarter (4.3%-8.5% each) of hearts allocated OOS (Supplemental Figure 1). An additional eleven centers, combined, received 22.7% of OOS allocated hearts (1.5–3.0% each); the remaining 131 centers received 52.3% (Figure 2A). The high frequency group, which accounted for 8.8% of all heart transplants, consisted of busier transplant centers (Median 432 transplants, IQR 163–623) than the moderate-frequency (Median 248 transplants, IQR 180–421; 17.2% of all heart transplants), and low-frequency centers (median 75 transplants, IQR 6–163; 74% of all transplants, p<0.001). Using the SRTR defined criteria, the high-frequency group was more likely to accept an offer compared to national offer acceptance practices (relative risk [RR] 1.37, 95% CI 1.14–1.63, group RR range 0.26–2.96). Surprisingly, the moderate-frequency group demonstrated a slightly more aggressive offer acceptance pattern (RR 1.74, 95% CI 1.50–2.02, group RR range 0.61–2.97).

Figure 2:

Bar chart showing the annual number of out of sequence transplants over the study period. The overall volume changed during the study period, but without a consistent trend. (A) The groups were separated into high-volume OOS centers (each center receiving 4.3%-8.5% of all OOS hearts), moderate-volume OOS centers (each center receiving 1.5%-3.0% of all OOS hearts), and low-volume OOS centers (each center receiving <1.5% of all OOS hearts). The proportion of OOS by each group remained relatively consistent over time. (B) The groups were separated into high-volume OOS OPOs (each OPO allocating 4.0%-6.6% of all OOS hearts), moderate-volume OOS OPOs (each OPO allocating 2.1%-3.4% of all OOS hearts), and low-volume OOS OPOs (each OPO allocating <1.8% of all OOS hearts) based on the proportion of overall OOS transplants. During the study period the high-volume group decreased the proportion of OOS transplants, while the low-volume group increased.

Among OPOs, six accounted for 27% of OOS heart allocations (4.0%-6.6% each; cumulatively allocating 17.2% of all heart transplants). A second group of fifteen allocated 40% of OOS hearts (2.1–3.4% each; 35.1% of all heart transplants), with the remaining 37 OPOs allocating 33% of OOS hearts (<1.8% each; 47.8% of all heart transplants). The distribution changed throughout the study period; when grouped according to overall study period volume, there was an inversion in the frequency of OOS allocations by high and low frequency OPOs over time, with a consistent proportion of moderate frequency OPOs (Figure 2B).

Donor Characteristics

The donors who were transplanted OOS were a similar age (32.5 years (IQR 25–41) vs. 32 years (IQR 25–40), p=0.22), similar size (body surface area: 1.98 m² (IQR 1.84–2.16) vs. 1.99 m² (1.83–2.17), p=0.56), were more commonly female (31.8% vs. 27.9%, p=0.01), and had a similar racial composition when compared with standard allocation donors (Table 1). The median donor ejection fraction was the same and there was no difference in the prevalence of donor coronary artery disease (4.0% vs. 3.6%, p=0.54). Donor comorbid conditions were similar, with only hypertension being more prevalent in the OOS group (18.1% vs. 15.3%, p=0.0008). Active donor infection or bacteremia were similar, as was the use of DCD, Public Health Service increased infectious risk, or HCV positive donors. The timing of transplant was also similar as OOS transplants were equally likely to happen on weekends (28.9% vs. 28.3%, p=0.70). OOS hearts were allocated at a higher sequence number (22 [IQR 7–63] vs. 8 [IQR 3–22], p<0.0001) and travelled a greater distance (347.2 nautical miles [IQR 154.6–476.6] vs. 302.4 nautical miles [IQR 184.5–547.7], p<0.0001). Individual donor characteristics that increased the odds of OOS allocation are listed in Table 2, and after multivariable adjustment blood type O, a higher sequence number, a lower ejection fraction, greater distance to donor hospital, donor inotrope use, and a history of cancer remained associated with an OOS heart allocation.

Table 1.

Baseline donor characteristics of hearts allocated with in-sequence allocation and out of sequence.

	Standard	OOS	p-value
n	17,514	880
Age (years)	32 (25–40)	32.5 (25–41)	0.22
Female	4,890 (27.9)	280 (31.8)	0.01
Body surface area (m2)	1.99 (1.83–2.17)	1.98 (1.84–2.16)	0.56
Ejection fraction (%)	60 (57–65)	60 (55–65)	0.17
Distance Heart Travelled (nautical miles)	302.4 (154.6–476.6)	347.2 (184.5–547.7)	<0.0001
Sequence Number	8 (3–22)	22 (7–63)	<0.0001
Race			0.86
Asian	307 (1.8)	13 (1.5)
Black	2,840 (16.2)	144 (16.4)
Other	286 (1.6)	12 (1.4)
White	14,071 (80.4)	711 (80.8)
Blood Type			<0.0001
A	6,777 (38.7)	316 (35.9)
AB	895 (5.1)	23 (2.6)
B	2,719 (15.5)	97 (11.0)
O	7,123 (40.7)	444 (50.5)
PHS Increased Infectious Risk	4,950 (28.4)	267 (30.7)	0.16
DCD	1,256 (7.3)	53 (6.0)	0.16
HCV Ab+	1,974 (11.3)	91 (10.3)	0.39
Weekend Transplant	4,951 (28.3)	254 (28.9)	0.70
Donor Infection	13,839 (79.0)	695 (79.0)	0.98
Donor Blood Infection	2,049 (11.7)	107 (12.2)	0.68
Smoker	2,122 (12.1)	111 (12.6)	0.66
Cocaine use	4,328 (24.7)	225 (25.6)	0.57
Inotrope Use	5.259 (30.0)	311 (35.3)	0.0008
Hypertension	2,633 (15.3)	157 (18.1)	0.02
History of cancer	426 (2.4)	30 (3.4)	0.07
Donor cardiac arrest after death	1,088 (6.8)	58 (7.0)	0.83
Coronary artery disease	628 (3.6)	35 (4.0)	0.54
>50% stenosis	14 (0.1)	1 (0.1)	0.82
Heavy Alcohol use	3,433 (19.6)	177 (20.1)	0.71

DCD: Donation after circulatory death; HCV: Hepatitis C virus; PHS: Public health service; OOS: Out of sequence

Table 2.

Unadjusted and adjusted logistic regression for donor characteristics associated with the odds of a donor being allocated out of sequence.

	Univariable	p-value	Multivariable	p-value
Age (years)	1.005 (0.99–1.01)	0.17
Female	1.21 (1.04–1.39)	0.01	1.08 (0.93–1.26)	0.32
Body surface area (m2)	0.95 (0.73–1.24)	0.69
Ejection fraction (per 5% decrease)	1.07 (1.02–1.12)	0.006	1.07 (1.02–1.12)	0.008
Distance Heart Travelled (per 10 nautical miles)	1.004 (1.002–1.006)	0.0001	1.004 (1.002–1.006)	0.0004
Sequence Number (per 5 sequence increase)	1.05 (1.017–1.023)	<0.0001	1.018 (1.015–1.021)	<0.0001
Race		0.85
Asian	0.84 (0.48–1.47)
Black	1.003 (0.84–1.21)
Other	0.83 (0.46–1.49)
White	Reference
Blood Type		<0.0001		<0.0001
A	0.36 (0.13–0.98)		0.40 (0.15–1.10)
AB	0.67 (0.57–0.77)		0.74 (0.63–0.86)
B	0.44 (0.33–0.58)		0.50 (0.37–0.66)
O	Reference		Reference
PHS Increased Infectious Risk	1.11 (0.96–1.29)	0.16
DCD	0.82 (0.61–1.08)	0.16
HCV Ab+	0.91 (0.73–1.13)	0.39
Weekend Transplant	1.03 (0.89–1.20)	0.71
Donor Infection	1.002 (0.85–1.18)	0.98
Donor Blood Infection	0.96 (0.78–1.18)	0.68
Smoker	10.5 (0.85–1.28)	0.66
Cocaine use	1.05 (0.90–1.22)	0.57
Inotrope Use	1.27 (1.11–1.47)	0.0008	1.26 (1.09–1.45)	0.002
Hypertension	1.23 (1.03–1.47)	0.02	1.10 (0.92–1.32)	0.31
History of cancer	1.41 (0.97–2.04)	0.07	1.75 (1.11–2.78)	0.02
Donor cardiac arrest after death	1.03 (0.78–1.35)	0.83
Coronary artery disease	1.11 (0.79–1.58)	0.54
Heavy Alcohol use	1.03 (0.87–1.22)	0.71

DCD: Donation after circulatory death; HCV: Hepatitis C virus; PHS: Public health service; OOS: Out of sequence

Recipient Characteristics

Recipients of OOS donor hearts were more commonly female (33.0% vs. 26.8%, p<0.0001), but had similar age and body surface area. There were no meaningful differences in candidate race or ethnicity. There were no significant differences in recipient heart failure etiology or comorbid conditions. Out of sequence recipients were more often blood type O (50.5% vs. 40.7%, p<0.0001). Additionally, UNOS priority status was notably different, as OOS hearts were more commonly allocated to Status 4 (34.2% vs. 17.0%) and Status 6 candidates (14.8% vs. 4.5%, p<0.0001). Patients receiving an OOS heart were less likely to be on an inotrope, receive any form of surgical mechanical circulatory support (MCS), be on extracorporeal membrane oxygenation, or have an intra-aortic balloon pump (Table 3). There was no clinically meaningful difference in hemodynamics between the two groups. After adjustment, recipient characteristics that were most associated with an OOS heart allocation included blood type O, UNOS Status 4 or 6, and the lack of surgical MCS (Table 4). Post-transplant survival was similar at 90 days (94.4% vs. 93.8%, p=0.53) and 1-year (90.6% vs 90.9%, p=0.72).

Table 3.

Baseline recipient characteristics of hearts allocated with in-sequence allocation and out of sequence.

	Standard	OOS	p-value
n	17,239	880
Age (years)	56 (43–63)	57 (45.5–63)	0.13
Female	4,689 (26.8)	290 (33.0)	<0.0001
BMI	27.4 (24.0–31.3)	27.8 (24.2–31.8)	0.04
BSA (m2)	2.00 (1.83–2.19)	2.01 (1.84–2.19)	0.84
Race			0.04
Asian	677 (3.9)	21 (2.4)
Black	4,420 (25.2)	219 (24.9)
Other	2,005 (11.5)	87 (9.9)
White	10,412 (59.5)	553 (62.8)
Latino	1,880 (10.7)	83 (9.4)	0.22
Blood Type			<0.0001
A	6,777 (38.7)	316 (35.9)
AB	895 (5.1)	23 (2.6)
B	2,719 (15.5)	97 (11.0)
O	7,123 (40.7)	444 (50.5)
HF Etiology			0.15
Ischemic	4,774 (27.3)	239 (27.2)
Retransplant	531 (3.0)	22 (2.5)
Restrictive	792 (4.5)	43 (4.9)
Congenital	763 (4.4)	24 (2.7)
Non-ischemic	10,654 (60.8)	552 (62.7)
Cerebrovascular disease	1,350 (7.8)	59 (6.7)	0.25
Prior Malignancy	1,576 (9.1)	93 (10.6)	0.13
Prior cardiac surgery	3,317 (19.5)	876 (4.9)	0.70
Albumin	4.1 (3.7–4.6)	4.1 (3.3–4.2)	0.26
Estimated GFR	70.0 (51.3–93.5)	69.0 (51.1–86.9)	0.04
UNOS Status			<0.0001
Status 1	1,959 (11.2)	30 (3.4)
Status 2	8,980 (51.3)	257 (29.2)
Status 3	2,643 (15.1)	147 (16.7)
Status 4	2,969 (17.0)	301 (34.2)
Status 5	172 (1.0)	15 (1.7)
Status 6	791 (4.5)	130 (14.8)
Any Surgical MCS	6,549 (37.4)	267 (30.4)	<0.0001
LVAD	5,738 (33.5)	253 (28.9)
Inotrope use	6,827 (39.0)	276 (31.4)	<0.0001
ECMO	1,106 (6.3)	23 (2.6)	<0.0001
IABP	4,548 (26.0)	140 (15.9)	<0.0001
Hemodynamics
PA Systolic	39 (30–50)	36 (28–48)	<0.0001
PA Diastolic	19 (13–25)	17 (12–23.5)	<0.0001
PA Mean	27 (20–34)	24 (18–33)	<0.0001
PCWP	18 (12–25)	16 (10–23)	<0.0001
Cardiac Output	4.1 (3.7–4.6)	4.1 (3.4–5.0)	0.10

BMI: Body mass index; BSA: Body surface area; ECMO: Extracorporeal membrane oxygenation; GFR: Glomerular filtration rate; HF: Heart failure; IABP: Intra-aortic balloon pump; LVAD: Left ventricular assist device; MCS: Mechanical circulatory support; PA: Pulmonary artery; PCWP: Pulmonary capillary wedge pressure; UNOS: United Network for Organ Sharing

Table 4.

Unadjusted and adjusted logistic regression for donor characteristics associated with the odds of a recipient receiving a heart allocated out of sequence.

	Univariable	p-value	Multivariable	p-value
Age (years)	1.004 (0.998–1.009)	0.18
Female	1.35 (1.16–1.56)	<0.0001	0.99 (0.86–1.16)	0.94
BMI	1.00 (0.99–1.001)	0.89
BSA (m2)	1.01 (0.78–1.31)	0.93
Race		0.04		0.02
Asian	0.58 (0.38–0.91)		0.68 (0.43–1.07)
Black	0.93 (0.80–1.10)		1.11 (0.94–1.31)
Other	0.82 (0.65–1.03)		0.79 (0.62–0.997)
White	Reference
Latino	0.87 (0.69–1.09)	0.22
Blood Type		<0.0001		<0.0001
A	0.41 (0.27–0.63)		0.25 (0.16–0.38)
AB	0.75 (0.65–0.87)		0.55 (0.47–0.64)
B	0.57 (0.46–0.72)		0.42 (0.33–0.53)
O	Reference		Reference
HF Etiology		0.15
Ischemic	0.97 (0.83–1.13)
Retransplant	0.80 (0.52–1.24)
Restrictive	1.05 (0.76–1.44)
Congenital	0.61 (0.40–0.92)
Non-ischemic	Reference
Cerebrovascular disease	1.17 (0.89–1.53)	0.25
Prior Malignancy	1.19 (0.95–1.48)	0.13
Prior cardiac surgery	0.97 (0.81–1.15)	0.70
Albumin	0.71 (0.28–1.83)	0.48
Estimated GFR	0.997 (0.996–1.009)	0.33
UNOS Status		<0.0001		<0.0001
Status 1	0.28 (0.19–0.41)		0.20 (0.12–0.34)
Status 2	0.52 (0.42–0.63)		0.42 (0.33–0.53)
Status 3	Reference		Reference
Status 4	1.82 (1.49–2.24)		1.82 (1.48–2.24)
Status 5	1.5 (0.90–2.73)		1.50 (0.85–2.64)
Status 6	2.96 (2.30–3.79)		2.78 (2.13–3.64)
Any Surgical MCS	0.73 (0.63–0.85)	<0.0001	0.72 (0.60–0.85)	0.0001
Inotrope use	0.72 (0.62–0.83)	<0.0001	0.99 (0.84–1.17)	0.94
ECMO	0.40 (0.26–0.61)	<0.0001	1.33 (0.75–2.36)	0.33
IABP	0.54 (0.45–0.65)	<0.0001	1.01 (0.80–1.29)	0.91

BMI: Body mass index; BSA: Body surface area; ECMO: Extracorporeal membrane oxygenation; GFR: Glomerular filtration rate; HF: Heart failure; IABP: Intra-aortic balloon pump; LVAD: Left ventricular assist device; MCS: Mechanical circulatory support; PA: Pulmonary artery; PCWP: Pulmonary capillary wedge pressure; UNOS: United Network for Organ Sharing

Causes of bypass

There were 880 total patients that were bypassed for any reason. OPO initiated bypasses ([Code 861] Operational OPO, [862] donor medical urgency, and [863] offer not made due to expedited placement attempt, [799/898] Other Specify where an open offer or expedited placement was specified) constituted more than half of all bypasses (464/880) and 2.5% of all transplants (Supplemental Table). When analysis was restricted to these bypasses only, the number of centers (n=80) and OPOs (n=54) remained similar. Three centers, which were among the four high-volume OOS centers for all OOS allocations, received 30.5% of all OPO-initiated bypasses (12.0%, 11.6%, 6.9% respectively, Supplemental Figure 2). At the OPO level there were three high-volume OPOs that allocated 17.8% of OPO-initiated OOS hearts (8.6%, 4.7%, 4.5% respectively), while a second group of 17 allocated 49.2% of OPO-initiated OOS hearts (2.1–3.9% each). Two of the three high-volume OPOs were high-volume OPOs for all OOS hearts, while the third was in the moderate-volume group. Relationships between OPOs and individual centers are shown in Figure 3.

Figure 3.

Network graph showing the relationship between individual heart transplant centers and OPOs for hearts that were allocated with an OPO initiated bypass. The blue dots represent OPOs, and the coral dots represent transplant centers, with the size of the dot proportional to the volume during the study period. The black lines and arrows represent connections between the OPO and transplant centers, with the thickness corresponding to magnitude of the relationship. Panel A, B, and C represent the relationship between OPOs (blue dots) and the high-frequency OOS transplant centers (coral dots). Panel D shows the relationship between all OPOs and transplant centers for comparison.

Discussion:

The number of adult HT in the United States continues to grow with a 72% increase since 2014, while maintaining very good outcomes. The ideal system to allocate hearts equitably minimizes waitlist mortality, maximizes post-transplant survival, and avoids nonuse of donor hearts. Regrettably, during the same period nonuse of hearts has doubled to 1.3%.³ Out of sequence solid organ allocation has been described in kidney transplantation and is sometimes deemed necessary for efficient organ placement and minimization of nonuse.⁴ In this analysis, we examined instances where donor hearts were allocated OOS and found: 1) OOS allocation for any reason occurs in nearly 5% of heart transplants. 2) Four centers received 25% of OOS transplants, and 15 centers received nearly half of all OOS transplants. 3) Outcomes for OOS hearts were similar to those allocated in sequence. 4) OOS donors were more often blood type O, allocated at a higher sequence number, had a lower ejection fraction, had a greater distance to the recipient hospital, utilized inotropic support, and had a history of cancer. 5) Recipients of OOS hearts were more commonly blood type O, UNOS Status 4 or 6, and were less likely to have surgical MCS.

This analysis found that two-thirds of centers and more than 90% of OPOs were involved with OOS heart allocations. While the overall volume remained consistent over time, it varied across OPOs. Those that initially more frequently allocated a heart OOS allocated fewer OOS hearts as the study period progressed (Figure 2B), while OPOs that were less frequent at the beginning increased their frequency by the end. The opposite was true for centers receiving OOS hearts, where the proportion of OOS transplant across high-, moderate-, and low-frequency centers remained fairly fixed. More than half of OOS allocated hearts were due to time constraints, which may happen when there are certain donor characteristics that are less desirable (older age, lower ejection fraction, coronary artery disease, etc.), when there is a strict time deadline, if an organ is declined in the operating room (and potentially another organ e.g. heart/lung), or potentially when the initial recipient’s condition changes. On average, these were larger centers (three were in the top 15 in overall heart transplant volume), which consequently would be more likely to have more patients on the waitlist and available in a timely manner. An added consideration was that these were more aggressive centers, which again, on average, was true. Though there was variation within the high-frequency group (ranging from 74% less likely to accept an offer to 196% more likely), it is possible that the SRTR reported metric does not fully capture a center’s patient selection tendencies. Similarly, while there were certain relationships between OPOs and centers that were greater in magnitude, each high-frequency center was allocated OOS hearts from multiple OPOs (range 13–27). Centers with more experience may, based on prior experience, be more apt to accept what may be considered by some a marginal graft. Further, the diligent and thorough evaluation of a donor for a particular recipient (e.g., coronary angiography, prospective crossmatch, etc.) can impact on expedient allocation and may hasten subsequent allocation if they heart is accepted for that particular recipient. There are unmeasured confounders that are contributing to this pattern, as center size, specific relationships, and center acceptance practice do not provide a complete explanation.

Recent data from the Donor Heart Study has helped identify donor characteristics that contribute to decreased utilization.^7–9 When looking for specific donor characteristics that increased the likelihood of OOS allocation, many of those associated with an OOS allocation were not surprising: higher sequence number (2% increased odds per 5 addition sequences), donor inotrope use (25% increased odds), and a donor history of cancer (75% increased odds). However, previously identified reasons for hearts being declined (older age, hepatitis C infection, left ventricular dysfunction, coronary artery disease, and blood type AB)¹⁰ were not associated with an increased risk of an OOS allocation. What stood out was the marked increased odds of an OOS allocated heart being blood type O. Type O is the most common blood type in the United States, and 60.5% of candidates on the waitlist have type O blood,¹¹ more than 40% of transplant recipients have type O blood, and donors with type O blood are more likely to be accepted for transplant.¹⁰ Why donors with type O blood were more likely to be allocated OOS is not clear. On the recipient side, it was notable that recipients of an OOS heart were more likely to be of a lower priority status (UNOS Status 4 [82% increased odds] or Status 6 [178% increased odds]) and without durable LVAD. These data would suggest that the OOS hearts more commonly went to candidates at home who typically have a lower risk of waitlist mortality.¹²

The Final Rule issued by the United States Department of Health and Human Services in 2002 altered transplant allocation in the United States. Designing a perfect allocation system that maximizes patient survival and organ stewardship is challenging, with dedicated healthcare professionals continually seeking optimization. When organs are allocated out of sequence, especially when OPO initiated, the prevailing reason is time, and the goal is reducing organ nonuse. OOS kidney allocation has been shown to do just that.¹³ However, the flip-side is that the organ then may go to any patient, irrespective of their priority status on the waitlist. In kidney transplantation OOS organs have been shown to favor older, female, and preemptive patients,^4,5 while in the present study we observed that OOS hearts more commonly went to patients with a lower priority status (Status 4 & Status 6) and type O blood. While it is encouraging to see potentially increased utilization of hearts that were at risk for nonuse, this needs to be balanced with the maintenance of equity. Universal trust in the organ allocation system must be unimpeachable and gain in utility cannot come at the expense of equity.

Non-standard abbreviations:

LVAD

Left ventricular assist device

MCS

Mechanical circulatory support

OOS

Out of sequence

OPO

Organ procurement organization

SRTR

Scientific Registry of Transplant Recipients

UNOS

United Network for Organ Sharing