Advertisement

Association of center-level temporary mechanical circulatory support use and waitlist outcomes after the 2018 adult heart allocation policy

Open AccessPublished:April 27, 2022DOI:https://doi.org/10.1016/j.surg.2022.03.032

      Abstract

      Background

      The present study characterizes the association of center-level temporary mechanical circulatory support use with waitlist outcomes after the 2018 adult heart allocation policy change.

      Methods

      The United Network for Organ Sharing database was queried for all single-organ, adult heart transplant candidates from November 2015 to October 2021. The study population was divided into 2 cohorts, prepolicy and postpolicy, centered around the rule change on October 18, 2018. The primary study outcome was center-level rate of poor waitlist outcome, defined as death or deterioration on the waitlist. Competing-risks regression was used to generate risk-adjusted rates of poor waitlist outcome at each center, while Pearson’s correlation coefficient (r) was used to assess the significance of center-level temporary mechanical circulatory support use (defined as the proportion listed with temporary mechanical circulatory support) and poor waitlist outcome.

      Results

      Of 22,077 transplant candidates included in analysis, 50.5% were listed during postpolicy. Compared to prepolicy, postpolicy candidates were more often listed with temporary mechanical circulatory support and less commonly listed with a durable left-ventricular assist device. The proportion of hospitals not using any temporary mechanical circulatory support decreased significantly from prepolicy to postpolicy (15% to 1%, P < .001). During prepolicy, center-level temporary mechanical circulatory support use showed no correlation with adjusted poor waitlist outcome. However, center-level temporary mechanical circulatory support use showed a negative correlation with poor waitlist outcome during postpolicy (r = -0.42, P < .001).

      Conclusion

      The 2018 adult heart allocation policy appears to benefit patients listed at high temporary mechanical circulatory support using centers, with significant interhospital variation in temporary mechanical circulatory support use in the new era. Given the growing role of temporary mechanical circulatory support on the heart transplant waitlist, greater standardization of its application is warranted.

      Introduction

      The Organ Procurement and Transplantation Network implemented significant changes to the existing adult heart allocation policy on October 18, 2018.
      Organ Procurement and Transplantation. Organ Procurement and Transplantation (OPTN) policies.
      ,
      OPTN|UNOS Thoracic Organ Transplantation Committee, OPTN|UNOS Policy Department
      Proposal to Modify the Adult Heart Allocation System.
      These changes modified the existing 3-tier model into 6 categories, with prioritization of those on temporary mechanical circulatory support (tMCS) modalities such as extracorporeal membrane oxygenation (ECMO), intraaortic balloon pumps (IABP), and temporary ventricular assist devices (tVAD).
      OPTN|UNOS Thoracic Organ Transplantation Committee, OPTN|UNOS Policy Department
      Proposal to Modify the Adult Heart Allocation System.
      Early investigations following implementation of the new policy noted an initial rise in the use of tMCS to support patients on the waitlist,
      • Hanff T.C.
      • Harhay M.O.
      • Kimmel S.E.
      • et al.
      Trends in mechanical support use as a bridge to adult heart transplant under new allocation rules.
      ,
      • Parker W.F.
      • Chung K.
      • Anderson A.S.
      • Siegler M.
      • Huang E.S.
      • Churpek M.M.
      Practice changes at U.S. transplant centers after the new adult heart allocation policy.
      a phenomenon that was partially blunted by the coronavirus disease 2019 pandemic in the United States.
      • Kim S.T.
      • Hadaya J.
      • Tran Z.
      • et al.
      Impact of the coronavirus disease 2019 pandemic on utilization of mechanical circulatory support as bridge to heart transplantation.
      While the large-scale impact of the 2018 policy change continues to be monitored closely, several groups have reported a significant increase in transplant rates and a concomitant decrease in waitlist mortality under the new allocation system.
      • Kilic A.
      • Mathier M.A.
      • Hickey G.W.
      • et al.
      Evolving trends in adult heart transplant with the 2018 heart allocation policy change.
      • Cogswell R.
      • John R.
      • Estep J.D.
      • et al.
      An early investigation of outcomes with the new 2018 donor heart allocation system in the United States.
      • Kim S.T.
      • Tran Z.
      • Xia Y.
      • et al.
      The 2018 adult heart allocation policy change benefits low-volume transplant centers.
      Specifically, patients bridged to transplantation with ECMO under the new policy experience improved survival both on the waitlist and following transplantation.
      • Hess N.R.
      • Hickey G.W.
      • Sultan I.
      • Kilic A.
      Extracorporeal membrane oxygenation bridge to heart transplant: trends following the allocation change.
      ,
      • Nordan T.
      • Critsinelis A.C.
      • Mahrokhian S.H.
      • et al.
      Bridging with extracorporeal membrane oxygenation under the new heart allocation system: a united network for organ sharing database analysis.
      Conversely, overall rates of post-transplant survival appear lower than prior years.
      • Kilic A.
      • Mathier M.A.
      • Hickey G.W.
      • et al.
      Evolving trends in adult heart transplant with the 2018 heart allocation policy change.
      ,
      • Cogswell R.
      • John R.
      • Estep J.D.
      • et al.
      An early investigation of outcomes with the new 2018 donor heart allocation system in the United States.
      This observation has in part been attributed to the higher acuity of transplanted patients in the new era.
      • Kilic A.
      • Mathier M.A.
      • Hickey G.W.
      • et al.
      Evolving trends in adult heart transplant with the 2018 heart allocation policy change.
      ,
      • Khazanie P.
      • Drazner M.H.
      The blurred line between gaming and patient advocacy: heart transplant listing decisions in the modern era.
      Despite such efforts toward elucidating the impact of the new allocation scheme at the patient level, little is known regarding center-level use of tMCS under the new scheme and its association with waitlist outcomes. In the present study, we characterized the association of institutional tMCS use with waitlist mortality and transplantation after the rule change. Additionally, we characterized the association of the 2018 policy change with center-level variation in the use of tMCS. We hypothesized that use of tMCS would increase after the policy change, and that hospital tMCS use would be associated with lower waitlist mortality and higher transplantation rates under the new system.

      Methods

      Study cohort

      This was a retrospective cohort study of all adult heart transplant candidates (≥18 years) listed between November 2015 and October 2021 in the United Network for Organ Sharing (UNOS) database. Patients requiring multiorgan transplantation as well as transplant centers performing <5 heart transplants per year were excluded from analysis. Heart transplant candidates were categorized based on the date of listing into equal time cohorts of 1,080 days, with prepolicy denoting the period before the allocation policy change on October 18, 2018, and postpolicy for the interval after. Use of tMCS was defined as the presence of IABP, ECMO, or tVAD (including percutaneous ventricular assist devices and other temporary circulatory support systems)
      Scientific Registry of Transplant Recipients
      2112 Public Standard Analysis Files Data Dictionary.
      at the time of listing. The primary study endpoint was center-level rate of adjusted poor waitlist outcome (PWO). Additional analysis was performed to characterize center-level variation in tMCS use following the new allocation policy. To facilitate analysis at the patient level, transplant centers were divided into tertiles of tMCS use (low, mid, and high tMCS) based on the proportion of listed patients requiring tMCS, with candidates being stratified into the respective tMCS category based on center of listing. For center-level models, tMCS use was evaluated as a continuous variable based on the proportion of patients listed with tMCS at a given transplant center.

      Patient-level analysis

      To assess patient-level associations with waitlist outcomes, competing risk subdistribution hazard models were developed for prepolicy and postpolicy eras using Fine-Gray competing-risk regressions.
      • Fine J.P.
      • Gray R.J.
      A proportional hazards model for the subdistribution of a competing risk.
      Waitlist outcomes of interest included transplantation and PWO, which was defined as death or delisting due to deterioration. Patients listed in the prepolicy era with a waitlist endpoint occurring in the postpolicy era were censored at the date of the policy change. The cumulative incidence function for each tMCS group was plotted to assess the association of center tMCS tertile and waitlist outcomes of interest, including transplantation and PWO.

      Center-level analysis

      To assess differences in center-level tMCS use between prepolicy and postpolicy eras, multivariable logistic regressions (Supplementary Table S1) were used to generate risk-adjusted predicted probabilities (P) of tMCS use for each candidate. The elastic net regularization method described was used to select covariates for multivariable analysis (Supplementary Table S1), as described elsewhere in detail.
      • Zou H.
      • Hastie T.
      Regularization and variable selection via the elastic net.
      The true proportion of candidates listed with tMCS for each center was calculated and divided by the mean value of P to generate center-level observed-to-expected (O/E) ratios. Values were plotted against hospital rank in O/E to facilitate visual inspection of center variation in tMCS use by era.
      To generate values of center-level adjusted PWO, the cumulative incidence function was calculated for each candidate using Fine-Gray competing-risk regressions and averaged among candidates at each transplant center, with separate values generated for each hospital during prepolicy and postpolicy. The mean PWO was then plotted against the proportion being listed with tMCS (as a continuous variable) to determine the association between center-level PWO and tMCS use across eras. Pearson’s correlation coefficient (r) was used to assess the significance of risk-adjusted PWO with the unadjusted proportion listed with tMCS for prepolicy and postpolicy hospitals.
      Baseline characteristics between prepolicy and postpolicy transplant candidates were compared using the Mann-Whitney U and χ2 tests, where appropriate. Hospitals were divided into low-, medium-, and high-volume tertiles based on annual heart transplant volume for multivariable analysis. Patient functional status was assessed using the Karnofsky performance status scale,
      • Mor V.
      • Laliberte L.
      • M J.N.
      • Wiemann M.
      The Karnofsky Performance Status Scale.
      with lower numbers denoting more severe illness. Regional changes in tMCS use were visualized and mapped based on geographic allocations set by the Organ Procurement and Transplantation Network.
      Organ Procurement and Transplantation. Organ Procurement and Transplantation (OPTN) policies.
      Adjusted logistic regression outcomes are reported as adjusted odds ratios (AOR) with 95% confidence intervals (95% CI), while outputs of competing-risks models were reported as subdistribution hazard ratios (SHR) with 95% CI. An α-level of <.05 was considered significant for all statistical analyses. The study was deemed exempt from full review by the institutional review board at the University of California, Los Angeles due to patient deidentification in the UNOS database. All statistical analyses were performed using Stata 16.0 (StataCorp LP, College Station, TX).

      Results

      Baseline characteristics of candidates by era

      Of 22,077 transplant candidates included in analysis, 11,159 (50.5%) were listed during postpolicy (Table I). Postpolicy transplant candidates were more often listed with tMCS (18.1% vs 7.9%, P < .001) compared to prepolicy, including ECMO (3.7% vs 1.9%, P < .001), IABP (12.9% vs 5.2%, P < .001), and tVAD (2.8% vs 1.3%, P < .001). Conversely, postpolicy patients were less likely to be supported by durable left ventricular assist device (26.1% vs 32.6%, P < .001) and had a lower functional status (4 [interquartile range 3–7] units vs 5 [2–6] units, P < .001) at the time of listing. Postpolicy patients also had a higher incidence of cerebrovascular disease (7.0% vs 6.2%, P = .01) but lower rates of prior cardiac surgery (37.7% vs 39.4%, P < .01), compared to prepolicy.
      Table IBaseline characteristics of pre-policy vs post-policy heart transplant candidates
      VariablePrepolicy, n = 10,918Postpolicy, n = 11,159P value
      tMCS at listing7.918.1
      Age, years56 (46–63)56 (45–63).54
      Female26.726.6.84
      BMI27.8 (24.2–31.6)27.9 (24.4–31.8)<.01
      Race<.001
       White63.860.1
       Black23.225.2
       Hispanic8.610.0
       Asian3.23.7
       Other1.21.1
      tMCS modality
       ECMO at listing1.93.7<.001
       IABP at listing5.212.9<.001
       TVAD at listing1.32.8<.001
       LVAD at listing32.626.1<.001
      Diagnosis<.001
      Nonischemic dilated Cardiomyopathy53.552.7
       Ischemic cardiomyopathy27.825.5
       Congenital heart defect3.33.9
       Restrictive cardiomyopathy3.34.1
       Valvular heart disease1.31.1
       Retransplantation2.72.4
       Hypertrophic cardiomyopathy2.73.1
       Other5.47.2
      Functional status, U
      Karnofsky functional status; lower numbers denote sicker patients.
      5 (3–7)4 (2–6)<.001
      Diabetes28.328.0.65
      Pulmonary hypertension
      Pulmonary hypertension defined as mean PA pressure ≥25 mm Hg.
      61.060.7.62
      Cerebrovascular disease6.27.0.01
      Prior cardiac surgery39.437.7<.01
      Hemodynamics
       Systolic PA pressure, mm Hg40 (30–51)39.5 (30–50).04
       Mean PA pressure, mm Hg27 (20–35)27 (20–35).07
       Cardiac output, L/min4.2 (3.4–5.1)4.2 (3.4–5.0).03
      Hospital volume tertile.02
       Low volume3.62.8
       Medium volume28.428.2
       High volume68.069.0
      Values are expressed as % or median ± interquartile range unless otherwise specified.
      BMI, body mass index; ECMO, extracorporeal mechanical oxygenation; IABP, intra-aortic balloon pump; TVAD, temporary ventricular assist device; LVAD, left ventricular assist device; PA, pulmonary artery.
      Karnofsky functional status; lower numbers denote sicker patients.
      Pulmonary hypertension defined as mean PA pressure ≥25 mm Hg.
      Factors associated with tMCS use are shown in Table II. After adjustment, postpolicy status remained strongly associated with increased odds of tMCS use (AOR 3.48, 95% CI 3.08–3.94) compared to prepolicy.
      Table IIFactors associated with temporary mechanical support use
      VariableOdds ratio95% CIP value
      Transplant era
       PrepolicyReferenceReferenceReference
       Postpolicy3.483.08–3.94<.001
      Female1.000.88–1.14.97
      Age, per 1 y0.990.99–1.00.03
      Non-White race0.930.83–2.85.22
      Body mass index, per 1 kg/m20.980.96–0.99<.001
      Functional status, U
      Karnofsky functional status; lower numbers denote sicker patients.
      0.610.58–0.63<.001
      DiagnosisReferenceReferenceReference
       Nonischemic dilated cardiomyopathyReferenceReferenceReference
       Ischemic cardiomyopathy1.140.99–1.32.07
       Congenital heart defect0.460.29–0.72<.01
       Restrictive cardiomyopathy0.580.43–0.77<.001
       Valvular heart disease0.550.30–1.00.05
       Retransplantation1.340.92–1.94.13
       Hypertrophic cardiomyopathy0.400.27–0.60<.001
       Other0.860.66–1.12.26
      Hospital volume tertile
       Low volumeReferenceReferenceReference
       Medium volume1.851.20–2.85<.01
       High volume2.451.60–3.73<.001
      Prior cardiac surgery0.730.64 - 0.83<.001
      Systolic PA pressure, mm Hg0.990.98 - 1.07<.01
      Mean PA pressure, mm Hg1.061.04 - 1.08<.001
      C-statistic = 0.83; covariates not shown above include pulmonary hypertension,
      Pulmonary hypertension defined as mean PA pressure ≥25 mm Hg.
      diabetes, cerebrovascular disease, dialysis status, total bilirubin, and cardiac output
      Karnofsky functional status; lower numbers denote sicker patients.
      Pulmonary hypertension defined as mean PA pressure ≥25 mm Hg.

      Patient-level waitlist outcomes

      Transplant centers were divided into tertiles of tMCS use (low, mid, and high tMCS) based on the proportion of listed patients requiring tMCS. Among prepolicy transplant candidates, listing at mid-tMCS was associated with higher adjusted odds of transplantation (SHR 1.17, 95% CI 1.08–1.25) compared to low-tMCS, while high-tMCS did not alter these odds (SHR 0.93, 95% CI 0.87–1.01) (Figure 1). After the policy change, patients listed at both mid- and high-tMCS experienced higher adjusted odds of transplantation compared to those at low-tMCS centers (mid-tMCS SHR 1.11, 95% CI 1.04–1.19; high-tMCS SHR 1.12, 95% CI 1.05–1.20).
      Figure thumbnail gr1
      Figure 1Competing risk of transplantation at low, mid, and high temporary mechanical circulatory support (tMCS) using centers by era.
      Prepolicy patients listed at mid- and high-tMCS did not exhibit differences in adjusted PWO compared to low-tMCS (mid-tMCS SHR 0.96, 95% CI 0.80–1.15; high-tMCS SHR 1.09, 95% CI 0.91–1.29) (Figure 2). However, postpolicy patients at high-tMCS experienced lower adjusted odds of PWO compared to those at low-tMCS (SHR 0.81, 95% CI 0.65–1.00).
      Figure thumbnail gr2
      Figure 2Competing risk of poor waitlist outcome (PWO) at low, mid, and high temporary mechanical circulatory support (tMCS) using centers by era.

      Center-level variation in tMCS use

      Center-level PWO was not correlated with tMCS use during prepolicy (r = -0.21, P = .11), but it was negatively correlated with tMCS use during postpolicy (r = -0.42, P < .001) (Figure 3). Similar relationships with adjusted PWO were observed with center-level ECMO and IABP utilization as shown in Supplementary Figure S1.
      Figure thumbnail gr3
      Figure 3Center-level temporary mechanical circulatory support (tMCS) use (X-axis) versus center adjusted rate of poor waitlist outcome.
      The median O/E of tMCS use was higher among hospitals during postpolicy (O/E 1.00 vs 0.77, P = .05) compared to prepolicy (Figure 4). The proportion of hospitals not using any tMCS (O/E = 0) also decreased significantly from prepolicy to postpolicy (15% to 1%, P < .001).
      Figure thumbnail gr4
      Figure 4Interhospital variation in observed-to-expected values for temporary mechanical circulatory support (tMCS) use by era.

      Regional changes in tMCS use as bridge-to-transplant

      Regional changes in tMCS use as bridge-to-heart transplantation from prepolicy to postpolicy can be seen in Figure 5. All regions experienced a statistically significant increase in tMCS use, with region 9 showing the greatest increase (from 5.6% to 22.3% of listings, P < .001), followed by regions 8 (7.6% to 21.6%, P < .001) and 5 (4.9% to 18.0%, P < .001). The smallest increases in tMCS use were seen in regions 4 (10.0% vs 14.9%, P < .001), 6 (5.6% to 12.4%, P < .01), and 7 (13.1% to 21.4%, P < .001).
      Figure thumbnail gr5
      Figure 5Regional changes in temporary mechanical circulatory support (tMCS) use as bridge to transplant from prepolicy to postpolicy. Region 1 (Connecticut, eastern Vermont, Maine, Massachusetts, New Hampshire, Rhode Island); region 2 (Delaware, District of Columbia, Maryland, New Jersey, Pennsylvania, West Virginia, Northern Virginia); region 3 (Alabama, Arkansas, Florida, Georgia, Lousiana, Mississippi, Puerto Rico); region 4 (Texas, Oklahoma); region 5 (Arizona, California, Nevada, New Mexico, Utah); region 6 (Alaska, Hawaii, Idaho, Montana, Oregon, Washington); region 7 (Illinois, Minnesota, North Dakota, South Dakota, Wisconsin); region 8 (Colorado, Iowa, Kansas, Missouri, Nebraska, Wyoming); region 9 (New York, western Vermont); region 10 (Indiana, Michigan, Ohio); region 11 (Kentucky, North Carolina, South Carolina, Tennessee, Virginia).

      Discussion

      The impact of the heart allocation policy change continues to be monitored closely, with several studies showing drastic improvements in transplant rates and waitlist mortality in the new era.
      • Hanff T.C.
      • Harhay M.O.
      • Kimmel S.E.
      • et al.
      Trends in mechanical support use as a bridge to adult heart transplant under new allocation rules.
      ,
      • Cogswell R.
      • John R.
      • Estep J.D.
      • et al.
      An early investigation of outcomes with the new 2018 donor heart allocation system in the United States.
      However, adjusted analyses of center-level tMCS use and its association with waitlist outcomes under the new policy are lacking. In the present study, we found that patients listed at high-tMCS centers experienced higher transplantation rates and lower adjusted rates of PWO compared to those at low-tMCS. At the center level, tMCS use was associated with lower adjusted rates of PWO among hospitals during postpolicy, but not prepolicy. Significant interhospital variation in tMCS use was noted in the new era, despite the vast majority of hospitals (99%) using some form of tMCS on the waitlist. More broadly, all UNOS regions demonstrated increased use of tMCS as a bridge to transplantation, with regions 5, 8, and 9 exhibiting the largest increases.
      Prior patient-level investigation on the policy change noted that those on tMCS modalities such as ECMO appear to benefit most under the new scheme, with drastic improvements in waitlist and post-transplant survival in this cohort.
      • Hess N.R.
      • Hickey G.W.
      • Sultan I.
      • Kilic A.
      Extracorporeal membrane oxygenation bridge to heart transplant: trends following the allocation change.
      ,
      • Nordan T.
      • Critsinelis A.C.
      • Mahrokhian S.H.
      • et al.
      Bridging with extracorporeal membrane oxygenation under the new heart allocation system: a united network for organ sharing database analysis.
      In accordance with this observation, others have noted an early increase in patients bridged to transplantation with tMCS modalities in the new era.
      • Hanff T.C.
      • Harhay M.O.
      • Kimmel S.E.
      • et al.
      Trends in mechanical support use as a bridge to adult heart transplant under new allocation rules.
      Interestingly, some have suggested that these findings may be a manifestation of “gaming” the new system and patient advocacy, with physicians choosing bridging modalities that optimize the odds of transplantation for their patients.
      • Khazanie P.
      • Drazner M.H.
      The blurred line between gaming and patient advocacy: heart transplant listing decisions in the modern era.
      ,
      • Ran G.
      • Chung K.
      • Anderson A.S.
      • et al.
      Between-center variation in high-priority listing status under the new heart allocation policy.
      In light of these changes, we found patients listed at high-tMCS centers to exhibit higher transplantation rates and lower rates of PWO compared to those listed at low-tMCS centers. These findings build on prior literature, showing that center-level tMCS use may profoundly impact patient-level outcomes regardless of the patient bridging strategy.
      In the present work, center-level tMCS use persisted despite adjusting for total hospital transplant volume, highlighting its independent role in predicting outcomes. Prior findings have shown that hospitals with low total transplant volume have benefited most under the new policy, with patients in this setting experiencing lower waitlist mortality and no detriment in post-transplant survival compared to the prior era.
      • Kim S.T.
      • Tran Z.
      • Xia Y.
      • et al.
      The 2018 adult heart allocation policy change benefits low-volume transplant centers.
      Conversely, patients at high transplant volume centers have faced worse post-transplant survival in the new era. The previous findings may highlight the evolving roles of hospital tMCS use and total transplant volume as markers of performance in the new era. Given the changes brought forth by the rule change, center tMCS use may serve as a useful supplement to total hospital volume when assessing hospital performance under the new policy.
      Our study builds on prior center-level analyses that have noted increases in transplantation rates and tMCS use under the new allocation policy, despite significant variation among centers for both metrics.
      • Kilic A.
      • Mathier M.A.
      • Hickey G.W.
      • et al.
      Evolving trends in adult heart transplant with the 2018 heart allocation policy change.
      ,
      • Cascino T.M.
      • Stehlik J.
      • Cherikh W.S.
      • et al.
      A challenge to equity in transplantation: Increased center-level variation in short-term mechanical circulatory support use in the context of the updated U.S. heart transplant allocation policy.
      Recent analysis by Cascino et al noted increased center-level variation in unadjusted rates of tMCS under the new policy.
      • Cascino T.M.
      • Stehlik J.
      • Cherikh W.S.
      • et al.
      A challenge to equity in transplantation: Increased center-level variation in short-term mechanical circulatory support use in the context of the updated U.S. heart transplant allocation policy.
      We observed center-level variation to persist even upon adjusting for hospital-level and patient-level factors not accounted for by previous studies. Additionally, we observed increased center-level tMCS use to be associated with lower adjusted odds of PWO in the new era, an association that was not apparent before the rule change and has yet to be reported in literature. Specifically, center-level use of ECMO and IABP were correlated with lower odds of PWO. Several reasons may account for the change in PWO observed in our study. First, broader center-level integration of multidisciplinary cardiogenic shock teams in recent years may be leading to improved outcomes at centers where tMCS is more readily available. Second, the observed benefit at high-tMCS centers may reflect an improvement in tMCS devices, with newer generations of various tMCS modalities becoming available on an annual basis. Finally, center-level tMCS use may be a proxy for the overall level of resources available at a center, leading to lower rates of PWO in the new era. Regardless, these findings may further validate center-level tMCS use as a potentially useful predictor of institutional performance under the new allocation policy.
      Taken together, our findings have important implications regarding access to care and equitable organ sharing under the new policy. Despite a national increase in the use of tMCS, we noted great variation at the regional level, a finding that has yet to be reported. While the South (region 4: Texas, Oklahoma) exhibited only a 5% increase in tMCS use under the new policy, New York and western Vermont (region 9) showed a 17% increase in use, suggesting the presence of geographical variation in the use of lifesaving tMCS modalities. As the number of donor organs available has not significantly increased in recent years,
      • Dharmavaram N.
      • Hess T.
      • Jaeger H.
      • et al.
      National trends in heart donor usage rates: are we efficiently transplanting more hearts?.
      ,
      • Khush K.K.
      • Ball R.L.
      Great variability in donor heart acceptance practices across the United States.
      the relative increase in transplantation rate among some centers will likely lead to decreased transplantation at others. Thus, improved standardization of tMCS use may help to promote more equitable organ sharing under the new allocation policy.
      Along these lines, our data set was limited in the ability to assess patient-level severity of illness at the time of initiation of tMCS. Therefore, the degree to which this contributed to PWO was not able to be specifically elucidated. However, it is plausible that center-level variation also exists with respect to indications for initiating each distinct tMCS modality. Recent investigation by Ran et al supports the presence of such a phenomenon, noting increased center-level variation in high-priority listing status in the new era.
      • Ran G.
      • Chung K.
      • Anderson A.S.
      • et al.
      Between-center variation in high-priority listing status under the new heart allocation policy.
      Such findings are in line with the variation in center-level tMCS use noted in our study and may give credence to the idea that a broad spectrum of patients may be listed as high-priority status following the policy change. To better assess equity in the effect of the policy change, standardization of criteria for the initiation of tMCS could play a critical role. Furthermore, center-level integration of a multidisciplinary cardiogenic shock team may help to aid in clinical decision-making and promote the judicious application of tMCS. Future national guidelines may consider promoting this team-based approach to cardiogenic shock and tMCS utilization.
      We acknowledge several notable limitations in this study, including those that are inherent to its nature as a retrospective cohort study. The UNOS database is a large-scale, national database that lacks clinical granularity. We were thus unable to adjust for clinical factors including echocardiographic abnormalities such as interventricular septal thickness and specific patient comorbidities including renal and liver disease. Further, potential variation within centers among individual cardiac surgeons and cardiologists was unable to be ascertained. Given that the UNOS database does not include information on costs, we were unable to include any analyses regarding the financial impact of the policy change. In order to maintain sufficient power for analysis, the study period included the duration of the coronavirus disease 2019 pandemic, which was shown to dampen national heart transplant volume during its initial stages.
      • DeFilippis E.M.
      • Sinnenberg L.
      • Reza N.
      • et al.
      Trends in US heart transplant waitlist activity and volume during the coronavirus disease 2019 (COVID-19) pandemic.
      However, heart transplant volume has since largely recovered,
      • Kim S.T.
      • Hadaya J.
      • Tran Z.
      • et al.
      Impact of the coronavirus disease 2019 pandemic on utilization of mechanical circulatory support as bridge to heart transplantation.
      and inclusion of this period was not expected to significantly alter the study findings.
      In conclusion, we observed increased center-level use of tMCS after the heart allocation policy change, with significant variation among transplant centers and regions. Under the new policy, patients listed at high-tMCS-using centers experienced increased transplant rates and lower rates of adjusted death and deterioration on the waitlist. Furthermore, center-level tMCS use was correlated with lower institutional waitlist death or deterioration following the policy change. These findings suggest that high-tMCS-using centers may be benefiting most under the new allocation policy, with greater institutional tMCS use now serving as an important marker for improved waitlist outcomes. Given the growing role of tMCS on the heart transplant waitlist, greater standardization of its application is warranted.

      Funding/Support

      None declared.

      Conflict of interest/Disclosure

      None declared.

      Supplementary materials

      References

      1. Organ Procurement and Transplantation. Organ Procurement and Transplantation (OPTN) policies.
        • OPTN|UNOS Thoracic Organ Transplantation Committee, OPTN|UNOS Policy Department
        Proposal to Modify the Adult Heart Allocation System.
        2016: 1-74
        • Hanff T.C.
        • Harhay M.O.
        • Kimmel S.E.
        • et al.
        Trends in mechanical support use as a bridge to adult heart transplant under new allocation rules.
        JAMA Cardiol. 2020; 5: 728-729
        • Parker W.F.
        • Chung K.
        • Anderson A.S.
        • Siegler M.
        • Huang E.S.
        • Churpek M.M.
        Practice changes at U.S. transplant centers after the new adult heart allocation policy.
        J Am Coll Cardiol. 2020; 75: 2906-2916
        • Kim S.T.
        • Hadaya J.
        • Tran Z.
        • et al.
        Impact of the coronavirus disease 2019 pandemic on utilization of mechanical circulatory support as bridge to heart transplantation.
        ASAIO J. 2021; 67: 382-384
        • Kilic A.
        • Mathier M.A.
        • Hickey G.W.
        • et al.
        Evolving trends in adult heart transplant with the 2018 heart allocation policy change.
        JAMA Cardiol. 2021; 6: 159-167
        • Cogswell R.
        • John R.
        • Estep J.D.
        • et al.
        An early investigation of outcomes with the new 2018 donor heart allocation system in the United States.
        J Hear Lung Transplant. 2020; 39: 1-4
        • Kim S.T.
        • Tran Z.
        • Xia Y.
        • et al.
        The 2018 adult heart allocation policy change benefits low-volume transplant centers.
        Clin Transplant. 2021; : 1-11
        • Hess N.R.
        • Hickey G.W.
        • Sultan I.
        • Kilic A.
        Extracorporeal membrane oxygenation bridge to heart transplant: trends following the allocation change.
        J Card Surg. 2021; 36: 40-47
        • Nordan T.
        • Critsinelis A.C.
        • Mahrokhian S.H.
        • et al.
        Bridging with extracorporeal membrane oxygenation under the new heart allocation system: a united network for organ sharing database analysis.
        Circ Hear Fail. 2021; 14: e007966
        • Khazanie P.
        • Drazner M.H.
        The blurred line between gaming and patient advocacy: heart transplant listing decisions in the modern era.
        Circulation. 2019; 140: 2048-2050
        • Scientific Registry of Transplant Recipients
        2112 Public Standard Analysis Files Data Dictionary.
        • Fine J.P.
        • Gray R.J.
        A proportional hazards model for the subdistribution of a competing risk.
        J Am Stat Assoc. 1999; 94: 496-509
        • Zou H.
        • Hastie T.
        Regularization and variable selection via the elastic net.
        J R Stat Soc Ser B Stat Methodol. 2005; 67: 301-320
        • Mor V.
        • Laliberte L.
        • M J.N.
        • Wiemann M.
        The Karnofsky Performance Status Scale.
        Cancer. 1984; 53: 2002-2007
        • Ran G.
        • Chung K.
        • Anderson A.S.
        • et al.
        Between-center variation in high-priority listing status under the new heart allocation policy.
        Am J Transplant. 2021; 21: 3684-3693
        • Cascino T.M.
        • Stehlik J.
        • Cherikh W.S.
        • et al.
        A challenge to equity in transplantation: Increased center-level variation in short-term mechanical circulatory support use in the context of the updated U.S. heart transplant allocation policy.
        J Hear Lung Transplant. 2022; 41: 95-103
        • Dharmavaram N.
        • Hess T.
        • Jaeger H.
        • et al.
        National trends in heart donor usage rates: are we efficiently transplanting more hearts?.
        J Am Heart Assoc. 2021; 10
        • Khush K.K.
        • Ball R.L.
        Great variability in donor heart acceptance practices across the United States.
        Am J Transplant. 2020; 20: 1582-1596
        • DeFilippis E.M.
        • Sinnenberg L.
        • Reza N.
        • et al.
        Trends in US heart transplant waitlist activity and volume during the coronavirus disease 2019 (COVID-19) pandemic.
        JAMA Cardiol. 2020; 5: 1048-1052