Portal de Periódicos da CAPES

An Evolution of Organ Allocation: Principles, Processes, and Innovations (Con)

2023; Wolters Kluwer; Volume: 107; Issue: 11 Linguagem: Inglês

10.1097/tp.0000000000004513

1534-6080

Darren R. Cullinan, Ola Ahmed, Joseph R. Scalea, William C. Chapman,

Liver Disease and Transplantation

Solid organ transplant allocation has changed from an allocation system based on geographic donation service areas (DSAs) to a concentric circle surrounding the donor hospital, a model that uses newly defined acuity circles. This change initially began with a lung transplant in November 2017 following a court challenge in New York. This process was subsequently adopted for liver allocation in February 2020 and most recently with kidney and pancreas allocation in March 2021. The goal of the change was to improve parity across DSAs by using a standard distance from the donor hospital rather than geographic boundaries. However, we feel that this change does not affect the major factor underlying the geographic disparities within transplant—transplant center aggressiveness and organ procurement organization (OPO) effectiveness at organ recovery. The newly adopted allocation system does not account for the significant variance between transplant centers’ likelihood of organ acceptance, as some centers are more risk averse than others. Additionally, the new system has tremendously increased the cost of organ allocation and created significant logistical hurdles that have negative effects on the system as a whole. THE VARIANCE PROBLEM One of the primary goals of the acuity circle model was to decrease the variance of the median MELD at transplant (MMaT) across DSAs. Although these changes in allocation did result in some changes in the distribution of donated livers, they did little to address the variance among transplant centers. Analysis from the most recent United Network for Organ Sharing (UNOS)/Organ Procurement and Transplantation Network (OPTN) report evaluating the first 18 mo of data following the implementation of the acuity circle allocation demonstrates that the MMaT has remained unchanged when analyzed by region, DSA, state, or program level (Table 1).1 TABLE 1. - Distribution of median adult deceased donor liver-alone recipient allocation MELD score at transplant by geographic units and era (Table 28, page 55, OPTN/UNOS report) MTS Unit of MTS Era N Minimum 25th Percentile Median Mean 75th percentile Maximum OPTN region Pre 11 24 26.5 29 28.4 30 33 Post 11 24 26 29 27.9 30 30 DSA Pre 51 20 24.75 28 26.8 29 34 Post 51 18 24 27 26.9 29 34 State Pre 39 20 24 27 26.5 29 33 Post 39 18 24 27 26.8 29 33 Transplant program Pre 128 17 25.375 28 28.6 31.6 40 Post 126 6 25 28 27.6 30 40 DSA, donation service area; MELD, model for end-stage liver disease; MTS, median transplant score; OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing. More importantly, analysis of the variance over the same period shows that regional, DSA, and state variation decreased slightly, but the majority of the variance lies at the center level (Table 2).1 This suggests that despite efforts to decrease MMaT variability between DSA most of the variation lies at the center level and no change in allocation will lead to an adjustment in a center’s tolerance for risk. These data suggest that centers have largely self-segregated into aggressive and nonaggressive phenotypes and the change in allocation has only exaggerated the discrepancy between them. One of the main goals of the acuity circle model was to decrease the variation across the system so that patients on the waiting list would not be disadvantaged because of the region where they live or the transplant center where they are listed, and the data do not show an improvement in the variation on any level. TABLE 2. - Variance and SD of median adult deceased donor liver-alone recipient allocation MELD score at transplant by era (Table 29, page 55, OPTN/UNOS report) Unit of median transplant score Prepolicy Postpolicy Variance SD Variance SD OPTN region 7.05 2.66 5.29 2.30 DSA 12.05 3.47 11.84 3.44 State 11.18 3.34 9.51 3.08 Transplant center 18.14 4.26 27.70 5.26 DSA, donation service area; MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing. Since MMaT has remained stable and center variation has increased, what has happened with center volume? A recent article by Chan and colleagues evaluated how the change in the allocation has affected individual transplant centers. They compared high-MELD centers, whose MMaT was above the national average, and low-MELD centers, whose MMaT was below the national average. They determined that high-MELD centers had the same MMaT, 30.3, in both preacuity and postacuity circle implementation. Low-MELD centers saw their MMaT increase by 1.6 points. As expected with broader sharing, high-MELD centers saw an increase in ideal and standard donor livers, but they did not increase their usage of nonideal donors. Low-MELD centers saw slight decreases in ideal and standard donors and a significant increase in nonideal donors.2,3 The new allocation system does little to promote maximal utilization of the donor pool. High-MELD centers have seen their volume increase without changing the quality of donor livers they use, thus creating a disincentive to maximize the use of nonideal donors from their region. Lower-MELD programs have either seen their volume decrease or have increased their usage of nonideal donors. We agree that higher-MELD patients should have access to higher-quality organs as their individual risk at surgery is significantly higher. However, lower-MELD patients who are listed at high-MELD centers currently have little chance of getting transplanted until their MELD score increases. We believe that lower-MELD patients should also have access to liver transplants, even if that means they receive livers from nonideal donors. Although UNOS is not involved in transplant center practices at a direct level, one of the major goals of the allocation change was to level the playing field with regard to access to transplants for patients in need of this therapy. However, when the system has such divergent center practices, these changes in allocation, with their associated substantial expense and logistical challenges, are likely to fail the original goal. WAITLIST MORTALITY Another goal of the acuity circle allocation system was to reduce waitlist mortality by prioritizing donor livers for the highest-MELD patients. This was somewhat successful as the rate of transplants increased for recipients with an MELD >291 (Figure 1). Despite the increased rate of transplants for MELD >29, there has been very little change in waitlist mortality. The net difference in waitlist removal for death or too sick for transplant was 121 fewer removals after the change in allocation (Figure 2). Interestingly, there was an increase in the number of removals for the MELD 15–28 group. This is the largest group of new patients added to the waiting list; 73% of patients added to the liver waiting list have an MELD 37 and adult patients with an MELD 29–32, MELD 33–36, and Status 1 groups (Figure 4). An important consideration, especially for death or removal from the waitlist, is that the COVID-19 pandemic was declared on March 13, 2020, and the new allocation had just started on February 4, 2020. It will be important to follow this data closely now that COVID deaths have been significantly reduced with vaccination.FIGURE 1.: Liver alone transplant rates per 100 active person-years waiting by MELD or PELD score or status and era (Figure 12, page 24, OPTN/UNOS report). MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; PELD, pediatric end-stage liver disease; UNOS, United Network for Organ Sharing.FIGURE 2.: Adult liver-alone registrations removed for death/too sick by MELD score or status group (Figure 17, page 36, OPTN/UNOS report). MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing.FIGURE 3.: Adult registrations added to the liver waitlist by MELD score or status at listing and era (Figure 2, page 9, OPTN/UNOS report). MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing.FIGURE 4.: Liver-alone waitlist removal because of death or too sick per 100 person-years waiting by MELD or PELD score or status, age at listing, and era (Figure 10, page 16, OPTN/UNOS report). MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; PELD, pediatric end-stage liver disease; UNOS, United Network for Organ Sharing.DONOR LIVER UTILIZATION AND OFFER RATES A likely unintended consequence of the change in the allocation has been a significant increase in organ offers. Organ offer rates have more than doubled at every MELD score (Figure 5).1 This has created a significant increase in offers that must be handled by the on-call surgeon. When combined with the increase in offers with the new kidney allocation, OPOs have had to add staff to coordinate the calls and import organ logistics. Multiple transplant centers have added call teams staffed by nurses or other trained staff to field the initial organ offers, adding another layer of complexity and cost to the system.FIGURE 5.: Number of offers per patient-year wating by allocation MELD or PELD score or status and era (Figure 60, page 89, OPTN/UNOS report). MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; PELD, pediatric end-stage liver disease; UNOS, United Network for Organ Sharing.Nonutilization of donated livers has remained relatively stable after the change in allocation. The national nonutilization rate was 9.3% both before and after policy change (Table 3). However, over the same time, the liver utilization rate (the number of livers transplanted/the number of livers recovered) has decreased from 71.5% to 66.4%. A trend of decreased utilization was seen in every OPTN region (Table 4).1 These data taken together suggest that broader sharing does not necessarily translate to more transplants. Similar results were seen in kidney transplantation when broader sharing was introduced and utilization of marginal grafts decreased.4 The decrease in liver utilization may partially be explained by pursuing marginal liver donors, which are initially offered locally to minimize ischemic time on marginal organs and promote increased utilization of these grafts. TABLE 3. - Liver discard rate by OPTN region and era (Table 58, page 98, OPTN/UNOS report) OPTN region Prepolicy Postpolicy Recovered Discarded % Recovered Discarded % 1 425 46 10.82 369 28 7.59 2 1587 246 15.50 1673 243 14.50 3 2232 110 4.93 2421 114 4.71 4 1473 107 7.26 1409 133 9.44 5 2156 291 13.50 2265 328 14.48 6 523 76 14.53 557 59 10.59 7 1042 80 7.68 1020 65 6.37 8 906 68 7.51 1062 88 8.29 9 538 26 4.83 486 33 6.79 10 1219 101 8.29 1369 111 8.11 11 1440 104 7.22 1645 120 7.29 National 13 541 1255 9.27 14 276 1322 9.26 OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing. TABLE 4. - Liver utilization rate by OPTN region and era (Table 60, page 100, OPTN/UNOS report) OPTN region Prepolicy Postpolicy % % 1 63.01 59.66 2 65.36 63.25 3 79.78 75.16 4 73.94 65.81 5 71.18 65.42 6 62.90 61.63 7 70.97 67.67 8 66.20 63.00 9 69.75 60.72 10 76.63 68.49 11 71.57 64.80 National 71.53 66.42 OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing. INCREASED ORGAN COST One of the significant challenges of the new organ allocation system is increased cost. Since the adoption of the acuity circle model, two-thirds of donors are from outside the local DSA (Figure 6). The primary sources for this increased cost come from increased organ recovery fees including import fees, surgeon fees, acquisition fees, and flight fees. When an organ is accepted from outside the recipient DSA this results in a double hit of increased charges of both acquisition fees that are fees charged by the donor OPO (including surgeon fees, tissue typing, operating room cost, etc), and import fees that are charged by the recipient OPO for coordinating the care of the donor outside of their DSA.5FIGURE 6.: Adult deceased donor liver-alone transplants by donor share type and era. (Figure 33, page 56, OPTN/UNOS report). OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing.The fees charged by OPOs are variable and unregulated and have led to a substantial increase in cost. A review by Wall et al showed a per-organ cost increase from $45 725 to $52 966, a difference of $7241.5 A review of our center data showed a similar per organ cost increase of $38 459–$49 654, a difference of $11 195. If these cost increases are extrapolated to the nearly 8000 liver donors per year, this results in a cost increase in excess of $70 million across the liver transplant system.6 A significant cost increase was also seen at our center following the adoption of acuity circles in lung transplantation. The median organ cost doubled from $34 000 to $70 203 after the adoption of the new allocation policy. The factors cited for the increased cost were similar: organ acquisition costs from the donor OPO, import costs from our local OPO, and transportation costs.7 Our center is particularly sensitive to increased travel costs since local organ procurements are performed at our donor recovery center that facilitates both decreased travel costs and increases the flexibility of the recovery time.8 INCREASED TRAVEL AND COORDINATION In addition to the cost associated with the travel and coordination required to recover organs from outside the DSA, recovering these organs utilizes more time for both the transplant team and OPO staff. Under the new allocation, the median travel distance to the donor hospital has nearly doubled from 72 NM to 142 NM with almost 50% of organ procurements requiring travel of >150 NM (Figure 7). Increased travel was significantly increased for all MELD scores >29. (Figure 8).1 In the new liver transplant allocation system, the number of procurements with a flight-consistent distance increased from 42.5% to 60.5% with 82.7% of DSA seeing an increase in the number of flight-consistent donors.2FIGURE 7.: Adult deceased donor liver-alone transplants by classification distance and era (Figure 34, page 57, OPTN/UNOS report). OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing.FIGURE 8.: Adult deceased donor liver-alone transplants by allocation MELD score or status, classification distance, and era (Figure 35, page 58, OPTN/UNOS report). MELD, model for end-stage liver disease; OPTN, Organ Procurement and Transplantation Network; UNOS, United Network for Organ Sharing.Coordination of organ recovery between multiple DSAs increases overnight transplants. A review of transplant timing by our lung transplant group showed that they were able to perform 100% of their lung transplants during daytime hours when they used our local DSA and when an outside DSA was involved one-third of the transplants were performed at night. Daytime transplants utilize fewer resources and cost less.7 LOGISTICAL CHALLENGES RESULTING FROM ORGAN SHARING It only makes sense to adopt models of organ allocation that optimize acceptance rates and reduce nonutilization rates. However, the biologic, physiologic, and immune parameters alone should not be misconstrued as the sole determinants of organ acceptance appropriateness. Embedded in the above discussion of increased time and distance are the resulting challenges of organ recovery, shipment, and acceptance logistics. Indeed, accepting a donor from a short distance from the transplant center is very different than accepting the same donor from a great distance—beyond the simple calculation of cold ischemic time (CIT). As livers are accepted at further distances, new complexities are introduced. For example, before the most recent allocation changes, it was common for liver teams to fly out and recover the majority of their organs. Since the adoption of the acuity circle allocation methodologies, this trend has changed. There is an increased likelihood that livers fly unattended and are thus not recovered by the transplanting surgeon. Although this is likely safer and cheaper for recovery teams, such a change in behavior adds complexity to the current system that was not designed with an infrastructure to synthesize the communication, data sharing, and time sensitivity of organ shipment. As organs travel further distances, more personnel and stakeholders are involved. More people touch the process and the organ itself. Because our system of organ allocation was not designed to accommodate travel complexities and was rather more focused on the appropriateness of biologic, physiologic, and immunologic matching, these logistical challenges of organ procurement, shipment, and acceptance were not included in the most recently developed models. As a case study, it makes clear sense to transplant a high-MELD patient with a locally recovered organ with a short CIT, even if the donor is imperfect. However, when we introduce time and logistical complexity, we introduce significant additional, previously absent, uncertainty. When we combine traditional organ acceptance risk with new logistical risks, we increase the total risk for our high-MELD patients and subsequently lead to organ decline. This now happens more commonly with kidneys, particularly high-kidney donor profile index kidneys. As such, it is important to remain circumspect about the changes we introduce. The use of acuity circles may be a smart approach; however, we only optimize and realize the acuity circle benefit when the other elements of the process (procurement, logistics, and organ acceptance) are incorporated into the model. Had critical logistical data been included in the newest allocation models, many of these current challenges may have been obviated. Unfortunately, the historic and relevant data regarding procurement, shipment, and recovery team management do not exist. What is the solution? Many fields beyond transplant surgery have smartly managed complex workflows and logistics. What makes transplant special is the time sensitivity and the enormous risk taken by patients and surgical recovery teams to execute procurement, shipment, and subsequent transplantation. A comprehensive organ management system that provided transparency, communication, data cataloging, and team management tools would help generate the data necessary to inform future allocation strategies because the logistical data surrounding prior experience with each stage of the transplant process could be built into the model. Indeed, we can only fix outcomes and processes when we have the data to support our changes. Said simply, we cannot fix what we cannot measure. In this regard, it is incumbent upon our entire field to begin learning from our current challenges to inform a smarter, more effective future in organ allocation. What are the benefits of improving logistics? Many innovative and exciting new methods for organ procurement, shipment, tracking, and communication have been recently discussed and published in the literature.9-12 By reducing the complexity of the organ recovery and shipment phase of organ allocation, organs can likely enjoy reduced cold ischemia times, and thus recipients can enjoy improved quality and outcome. Further, the reduction in complexity from optimized shipment can be reasonably expected to reduce the costs of organ shipment. Perhaps more importantly, with improved efficiency and transparency, early organ decline could be converted to reallocation of in-transplant organs, allowing for reductions in organ nonutilization—this has particular relevance for high kidney donor profile index kidneys being shipped longer distances, but also for some liver transplants. OPPORTUNITIES FOR IMPROVEMENT Although the current allocation system has significant challenges, there are opportunities to standardize the organ recovery process that will lead to improvements that would be effective under any allocation system. First, we would strongly advocate for a requirement that OPOs utilize a donor care unit (DCU). These facilities have been shown to increase donor organ yield while decreasing cost.8 The establishment of a national system wherein each OPO manages its own DCU was identified in a recent National Academies of Science, Engineering, and Medicine report on the organ transplant system. The report calls for Health and Human Services to mandate OPO-run DCUs and for the Centers for Medicare and Medicaid Services to revise the payment incentives so that hospitals are not disadvantaged by transferring a donor to a DCU.13 To decrease the travel of donor recovery teams, we would advocate for standard donors to be recovered by local recovery teams. This worked well when it was strongly recommended by UNOS/OPTN and later by the American Society of Transplant Surgeons during the COVID pandemic. The reliance on local teams requires improvements in telecommunications so recipient teams can adequately assess the organ. Standardization of digital imaging, digital microscopy, and video sharing through a secure platform like DonorNet would be facilitated by having standardized DCUs with these technologies in place. As many programs utilize fellows for donor recovery, it would be the responsibility of the local center to provide adequate training and ensure the competency of the recovery surgeon. Although it may still be necessary to send a recovery team in specialized situations, like pediatric recoveries or particularly marginal organs, the vast majority of organs could be recovered by local teams, which would significantly decrease travel for recovery surgeons and OPO staff. A critical underlying hurdle for the process of organ shipment is cold ischemia time. Because time is so important during the transplant process, we move mountains to reduce cold ischemia time, frequently driving chaotic recovery and transplant surgical procedures. Using machine perfusion with either hypothermic or normothermic perfusates, or even super cooling techniques could change the way we think about CIT, and thus allocation strategies. Shipment itself could also be innovated. There has been great excitement in transplantation around the use of unmanned aircraft, or drones to facilitate organ shipment. Indeed, as these technologies advance it will become increasingly appealing to move an organ directly from donor hospital to recipient hospital, unmanned, at high speeds. This would reduce CIT, improve team safety by taking recovery surgeons and teams out of the skies, and it would likely improve access to organs. Indeed, the next decade of technology development in this space will be quite exciting, and each of the technologies will change how we think about organ allocation and its evolution. It is time for the industry to focus on critical technology and infrastructure development. The entire transplant process includes organ donor identification, procurement, shipment, team management, transplantation, and posttransplant care. An area of clear opportunity for our industry stakeholders is to partner with experts in workflow, efficiency, and supply chain management to inform current and future practices. By collecting the data today necessary for organ allocation strategies tomorrow, our future patients will be more likely to benefit from the models that we implement. Further, by building comprehensive data models that include more than biologic, physiologic, and immunologic data, we will improve access to organs and help more patients conquer organ failure. Through the standardization of organ recovery with DCUs and local recovery teams, a fair and transparent import fee can be applied to organs that are transferred between DSAs. This combined with innovations in cold preservations technologies and a comprehensive improvement in supply chain coordination we can modernize the donor recovery process and help mitigate some of the cost and increased travel that has come with the shift in allocation to acuity circles. CONCLUSION The changes in liver allocation to the acuity circles model have not improved MMaT, death on the waiting list, or organ utilization. Despite this, the acuity circle model has led to a significant increase in the cost of organ allocation because of increased flights needed to transport organs over longer distances and other logistical hurdles. We believe that the prior region-based system was more efficient and cost-effective and that the acuity circle allocation system has not demonstrated any of its proposed benefits.