Abstract
Background.
Adoption of machine perfusion (MP) technology has rapidly expanded in liver transplantation without real-world data on utilization and outcomes, which are critical to understand the appropriate application of MP technology.
Methods.
The Organ Procurement and Transplant Network/Standard Transplant Analysis and Research database was used to identify all deceased donor livers procured with intent for transplant between October 27, 2015 (date of first recorded MP) and June 30, 2023 (n = 67 795). Liver allografts were cohorted by donation after brain death (DBD; n = 59 957) or circulatory death (DCD; n = 7873) and analyzed by static cold storage (SCS) or MP preservation method. Donor demographics, organ utilization, and graft survival were evaluated.
Results.
By 2023, 12.5% of all livers and 37.2% of DCD livers underwent MP preservation (82.6% normothermic, 6.7% hypothermic, and 10.8% other/unknown). Compared with SCS, MP liver donors were older (DBD: 48 versus 40 y [P < 0.001]; DCD: 43 versus 38 y [P < 0.001]) with higher body mass index (DBD: 28.8 versus 26.9 kg/m2 [P < 0.001]; DCD: 27.7 versus 26.9 kg/m2 [P = 0.004]). Donor livers had similar levels of macrosteatosis (median 5%). Graft utilization was higher for MP than SCS after DBD (96.4% versus 93.0%, P < 0.001) and DCD (91.4% versus 70.3%, P < 0.001) donation. Graft survival was similar between MP and SCS livers from DBD donors (P = 0.516), whereas MP-preserved grafts had superior survival from DCD donors at 1 and 3 y posttransplant (P = 0.013 and 0.037). Patient survival was similar across all groups at 3 y (P = 0.322).
Conclusions.
The use of MP in liver transplantation increased rates of liver utilization and improved graft survival after DCD. Further monitoring of MP outcomes is required to understand long-term benefits.
Simply put, machine perfusion (MP) is transforming liver transplantation. Landmark studies have demonstrated the benefits of MP to marginal liver allografts, reducing rates of complications, including early allograft dysfunction and ischemic cholangiopathy, compared with standard criteria organs.1-3 MP has quickly been recognized as a pathway to safely increase organ utilization to address the severity of organ shortage and an ever-expanding liver transplant waitlist. Beyond this, nearly every aspect of liver transplant is being reconsidered: expansion of donor selection criteria, including age and graft steatosis, viability testing to use organs once considered “unusable,” postprocurement therapeutics, and prolonged duration of MP to eliminate overnight transplantation altogether.4-7
Although it is understood that utilization of MP has increased, a comprehensive understanding of how widely this technology is being used and the cumulative effect on national transplant outcomes is not known. Although single-center reports from early adopters and large-volume users demonstrate increased rates of liver transplantation with the maintenance of excellent outcomes,8-10 there is a paucity of data on real-world use, and the vast majority of the United States experience remains unpublished. As such, the adoption of MP technology has been heterogenous, without data on medium- and long-term graft and patient outcomes.
This study seeks to identify the prevalence of MP usage in the United States, evaluate its effects on liver allograft utilization, and assess the impact on graft survival and patient survival. Using the Organ Procurement and Transplant Network (OPTN) Standard Transplant Analysis and Research database, this study provides a review of national MP utilization to inform ongoing practices and better understand the benefits of MP in liver transplantation.
MATERIALS AND METHODS
Potential liver donors were identified in the OPTN/STAR database using the June 2023 release. Study dates were defined from the first recorded date of MP utilization in the OPTN/STAR database to the end of follow-up (October 27, 2015–June 30, 2023). Donors were included for evaluation if they proceeded to the operating room with the intent to procure a transplant (n = 67 795) as previously described.11 Donors were separated by neurologic status into donation after brain death (DBD; n = 59 957) or circulatory death (DCD; n = 7873) cohorts. Database records were used to identify the utilization of static cold storage (SCS) or MP and the type of MP used: hypothermic MP (HMP), normothermic MP (NMP), or other/unknown. The Institutional Review Board at Cedars-Sinai Medical Center approved this study.
Donors were analyzed for demographics, including age, sex, race/ethnicity, body mass index (BMI), terminal laboratory values, and presence of graft steatosis when a biopsy was performed. Donors were further evaluated for frequency of MP use and subsequent liver utilization rates for transplantation. Recipients of all transplanted livers were identified and further analyzed for demographics, including age, sex, race/ethnicity, BMI, Model for End-stage Liver Disease (MELD) score at transplant, whether hepatocellular carcinoma exception points had been awarded, incidence of multiorgan transplantation, and waiting time, defined as the time interval from date of waitlisting to date of liver transplant.
The primary outcomes were 1- and 3-y posttransplant patient and graft survival. Utilization of donor livers intended for transplant was considered a secondary outcome, as defined by the number of livers used for transplant within the total population of donor livers recovered with intent for transplant as defined in the OPTN/SRTR database.
Statistical analysis was performed using the Kruskal-Wallis test for continuous variables and chi-square tests for categorical variables. Donor and recipient data are reported by median and interquartile range (IQR). Kaplan-Meier methods with log-rank testing were used to evaluate patient and graft survival. Multivariable analysis of 1- and 3-y graft outcomes was performed using Cox proportional hazards modeling with clinically relevant donor and recipient factors (age, sex, race/ethnicity, BMI, MELD score at transplant, and utilization of MP). A P value of <0.05 was considered significant. Data analysis was performed using JMP Pro, version 17.0.0 (JMP Statistical Discovery LLC, Cary, NC).
RESULTS
Rates of MP Use
MP use was analyzed by transplant year for all liver donors and stratified into DBD and DCD cohorts (Figure 1). A total of 1504 donor livers underwent MP during the study period, with 1236 undergoing NMP (82.1%), 100 undergoing HMP (6.6%), and 168 unknown (11.2%). When analyzing by donor type, a total of 841 DBD livers (1.4%) and 663 DCD livers (8.4%) were preserved by MP. However, rates of MP utilization increased rapidly since 2021. By 2023, 12.1% of all livers procured for transplant underwent MP preservation, constituting 37.2% of DCD livers and 6.7% of DBD livers.
FIGURE 1.
MP utilization by year as a percentage of all donors, stratified by donor type (DBD vs DCD). DBD, donation after brain death; DCD, donation after circulatory death; MP, machine perfusion.
Donor Demographics
Donor demographics are summarized in Table 1. Donors of livers undergoing MP were older for both DBD (median 48 y [IQR, 35–59] versus 40 y [IQR, 28–54]; P < 0.001) and DCD (median 43 y [IQR, 32–53] versus 38 y [IQR, 28–49]; P < 0.001) cohorts. Livers undergoing MP after DBD procurement were less likely to come from male donors (56.2% male versus 59.9% P < 0.001), whereas there was no sex difference among donors of DCD livers undergoing MP (70.4% male versus 68.0%; P = 0.202). There were no differences in donor race/ethnicity between MP and SCS livers for either DBD or DCD. Donors of MP livers had higher BMI as compared with donors of SCS livers for both DBD (28.8 kg/m2 [IQR, 24.9–34.0] versus 26.9 kg/m2 [IQR, 23.2–31.7]; P < 0.001) and DCD (27.7 kg/m2 [IQR, 23.8–32.9] versus 26.9 kg/m2 [IQR, 23.4–31.5]; P = 0.004) groups. Donors of MP livers were more likely to have anoxia or stroke as a cause of death than donors of SCS livers for both DBD (P = 0.001) and DCD (P = 0.012). Livers from donors carrying increased risk for blood-borne disease transmission were more likely to undergo MP after both DBD (33.9% versus 24.5%) and DCD (17.4% versus 9.7%; P < 0.001 for both).
TABLE 1.
Donor demographics and organ utilization rates, stratified by DBD and DCD
| Category | DBD | DCD |
||||
|---|---|---|---|---|---|---|
| Organ preservation | MP | SCS | P | MP | SCS | P |
| No. of donors | 841 (1.4%) | 59 116 (98.6%) | 663 (8.5%) | 7175 (91.5%) | ||
| Demographics | ||||||
| Age, y | 48 (35–59) | 40 (28–54) | <0.001 | 43 (32–53) | 38 (28–49) | <0.001 |
| Sex, % | <0.001 | 0.202 | ||||
| Male | 56.2% | 59.9% | 70.4% | 68.0% | ||
| Female | 43.8% | 40.1% | 29.6% | 32.0% | ||
| Race/ethnicity, % | 0.102 | 0.538 | ||||
| White, non-Hispanic | 63.4% | 61.8% | 75.7% | 75.8% | ||
| Black, non-Hispanic | 19.0% | 18.5% | 9.2% | 10.0% | ||
| Hispanic/Latino | 13.9% | 15.8% | 11.6% | 11.0% | ||
| Asian, non-Hispanic | 2.5% | 2.7% | 1.5% | 2.1% | ||
| Native American | 1.0% | 0.6% | 1.2% | 0.5% | ||
| Pacific Islander | 0.2% | 0.3% | 0.3% | 0.1% | ||
| Multiracial | 0% | 0.4% | 0.5% | 0.4% | ||
| BMI, kg/m2 | 28.8 (24.9–34.0) |
26.9 (23.2–31.7) |
<0.001 | 27.7 (23.8–32.9) |
26.9 (23.4–31.5) |
0.004 |
| Mechanism, % | 0.001 | 0.012 | ||||
| Anoxia | 47.3% | 43.4% | 54.2% | 52.1% | ||
| CVA/stroke | 30.1% | 27.5% | 20.7% | 17.9% | ||
| Head trauma | 20.9% | 26.7% | 20.5% | 26.0% | ||
| CNS tumor | 0.2% | 0.4% | 0% | 0.1% | ||
| Other | 1.4% | 2.0% | 4.7% | 3.9% | ||
| Increased risk for disease transmission, % | 33.9% | 24.5% | <0.001 | 17.4% | 9.7% | <0.001 |
| Donor terminal labs | ||||||
| Bilirubin, mg/dL | 0.6 (0.4–1) | 0.6 (0.4–1) | 0.378 | 0.5 (0.3–0.8) | 0.5 (0.4–0.8) | 0.012 |
| AST, units/L | 45 (25–96) | 41 (24–85) | 0.011 | 48.5 (30.75–81.25) | 51 (32–84) | 0.155 |
| ALT, units/L | 38 (22–85) | 38 (21–79) | 0.294 | 40 (22–74) | 41 (24–78) | 0.337 |
| Creatinine, mg/dL | 1.16 (0.8–2.1) | 1.1 (0.77–1.9) | 0.021 | 0.8 (0.6–1.31) | 0.8 (0.6–1.15) | 0.047 |
| pH | 7.42 (7.37–7.45) |
7.41 (7.36–7.45) |
0.018 | 7.43 (7.38–7.47) |
7.42 (7.37–7.46) |
0.007 |
| Steatosis | ||||||
| Biopsies performed, n | 455 (54.1%) | 25 697 (43.5%) | 336 (50.7%) | 1879 (26.2%) | ||
| Microsteatosis | 5% (0%–10%) | 5% (0%–10%) | 0.036 | 0% (0%–5%) | 1% (0%–10%) | 0.022 |
| Macrosteatosis | 5% (0%–15%) | 5% (0%–10%) | 0.214 | 5% (0%–10%) | 5% (0%–11%) | 0.009 |
| Macrosteatosis >20% | 95 (11.2%) | 5505 (9.3%) | 0.19 | 34 (5.1%) | 417 (5.8%) | 0.875 |
| Organ utilization | ||||||
| All donors | 96.4% | 93.0% | <0.001 | 91.4% | 70.4% | <0.001 |
| Macrosteatosis >20% | 89.5% | 70.6% | <0.001 | 85.3% | 25.4% | <0.001 |
Data reported as median (IQR).
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; CNS, central nervous system; CVA, cerebrovascular accident; DBD, donation after brain death; DCD, donation after circulatory death; IQR, interquartile range; MP, machine perfusion; SCS, standard cold storage.
Terminal donor laboratory values were compared between MP and SCS groups for both DBD and DCD donors. Although there were some statistically significant differences in terminal laboratory values, including creatinine, bilirubin, and aspartate aminotransferase, none of these differences were clinically meaningful (Table 1). A liver biopsy was performed on 455 DBD livers (54.1%) undergoing MP, 25 697 DBD livers (43.5%) undergoing SCS, 336 DCD livers (50.7%) undergoing MP, and 1879 DCD livers (26.2%) undergoing SCS. Overall, there was no clinically significant difference in average macrosteatosis or microsteatosis between groups, despite a statistically significant difference in microsteatosis for DBD livers (5% [0%–15%] versus 5% [0%–10%]; P = 0.036) and statistically lower degree of microsteatosis (0% [0%–5%] versus 1% [0%–10%]; P = 0.022) and macrosteatosis (5% [0%–10%] versus 5% [0%–11%]; P = 0.009) for MP livers in the DCD cohort. When evaluating donors with biopsy-confirmed macrosteatosis of >20%, there was no difference in the prevalence of macrosteatosis of >20% between MP and SCS livers for either DBD (11.2% versus 9.3%, P = 0.19) or DCD (5.1% versus 5.8%, P = 0.875) livers.
Liver Utilization Rates
Utilization of livers procured with intent for transplant was compared for both DBD and DCD procurements. For DBD procurements, donor livers undergoing MP were more likely to be used for transplantation (96.4%) than those undergoing SCS preservation (93.0%; P < 0.001). DCD livers undergoing MP preservation demonstrated increased utilization for transplant (91.4%) compared with SCS-preserved livers (70.4%; P < 0.001). As a subgroup analysis, utilization rates of all livers with biopsy-proven macrosteatosis of ≥20% were examined. In these steatotic livers, 89.5% of MP livers were used for transplant, compared with 70.6% of SCS livers after DBD donation (P < 0.001). After DCD procurement, 85.3% of livers with biopsy-proven steatosis were used for transplant after MP, whereas only 25.4% of steatotic livers undergoing SCS were used for transplant (P < 0.001).
Recipient Demographics
Recipients of MP livers were older than recipients of SCS livers after DBD donation (58 y [IQR, 50–64] versus 56 y [IQR, 45–63]; P < 0.001), with a similar age of MP and SCS liver recipients for DCD livers (59 y [IQR, 52–65] versus 59 y [IQR, 52–65]; P = 0.238; Table 2). There were no differences in sex of recipients receiving livers preserved by MP or SCS after either DBD (P = 0.274) or DCD (P = 0.381) procurement. Although there was no difference in race/ethnicity of recipients of DCD donors (P = 0.951), recipients of DBD livers preserved by MP were more likely to be White, non-Hispanic, and less likely to be Black, non-Hispanic (P = 0.017). Although patients receiving MP livers after DBD had slightly higher BMI (29.0 [IQR, 25.1–33.0] versus 28.0 [IQR, 24.1–32.6]; P < 0.001), there was no difference in BMI for recipients of DCD livers (P = 0.471).
TABLE 2.
Recipient demographics and transplant criteria, stratified by DBD and DCD
| Category | DBD | DCD | ||||
|---|---|---|---|---|---|---|
| Organ preservation | MP | SCS | P | MP | SCS | P |
| No. of recipients | 818 (1.4%) | 56 914 (98.6%) | 606 (10.5%) | 5146 (89.5%) | ||
| Demographics | ||||||
| Age, y | 58 (50–64) | 56 (45–63) | <0.001 | 59 (52–65) | 59 (52–65) | 0.238 |
| Sex, n (%) | 0.274 | 0.381 | ||||
| Male | 532 (65%) | 35 961 (63%) | 420 (69%) | 3503 (68%) | ||
| Female | 286 (35%) | 20 953 (37%) | 186 (31%) | 1643 (32%) | ||
| Race/ethnicity (%) | 0.017 | 0.951 | ||||
| White, non-Hispanic | 593 (72%) | 38 823 (68%) | 455 (75%) | 3849 (75%) | ||
| Black, non-Hispanic | 49 (6%) | 4926 (9%) | 20 (3%) | 286 (6%) | ||
| Hispanic/Latino | 136 (17%) | 9671 (17%) | 103 (17%) | 763 (15%) | ||
| Asian, non-Hispanic | 28 (3%) | 2492 (4%) | 20 (3%) | 164 (3%) | ||
| Native American | 5 (1%) | 518 (1%) | 6 (1%) | 50 (1%) | ||
| Pacific Islander | 3 (0%) | 110 (0%) | 1 (0%) | 7 (0%) | ||
| Multiracial | 4 (0%) | 374 (1%) | 1 (0%) | 27 (1%) | ||
| BMI, kg/m2 | 29.0 (25.1–33.0) | 28.0 (24.1–32.6) | <0.001 | 28.7 (25.1–32.3) | 28.6 (25.0–32.9) | 0.471 |
| MELD at transplant | 22 (14–31) | 24 (15–33) | <0.001 | 18 (12–24) | 19 (14–24) | 0.038 |
| HCC exception | 19.3% | 13.8% | 0.0013 | 26.2% | 19.0% | 0.019 |
| Multiorgan transplant | 81 (10%) | 6040 (11%) | 0.512 | 43 (7%) | 422 (8%) | 0.345 |
| Waiting time, d | 75.5 (12–278) | 8 (3–55) | <0.001 | 111 (24–296) | 109 (27–264) | 0.683 |
Data reported as median (IQR).
BMI, body mass index; DBD, donation after brain death; DCD, donation after circulatory death; HCC, hepatocellular carcinoma; MELD, Model for End-stage Liver Disease; MP, machine perfusion; SCS, standard cold storage.
Patients receiving MP livers were transplanted at lower MELD scores as compared with those receiving SCS livers. The median MELD score for recipients of DBD livers was 22 [IQR, 14–31] for MP livers as compared with 24 [IQR, 15–33] for SCS livers. Likewise, patients receiving DCD livers were transplanted at a lower MELD score after MP preservation (median MELD 18 [IQR, 12–24]) versus SCS preservation (median MELD 19 [IQR, 14–24]; P = 0.038). Patients receiving MP livers were also more likely to have received exception points for hepatocellular carcinoma listing (19.3% versus 13.8% for DBD [P = 0.001]; 26.2% versus 19.0% for DCD [P = 0.019]). There was no difference in the incidence of multiorgan transplantation between groups. Patients receiving MP livers from DBD donors had longer waiting time to transplant (median 75.5 d; [IQR, 12–278]) than those receiving livers undergoing SCS preservation (median 8 d [IQR, 3–55]). There was no difference in waiting time for recipients of DCD livers undergoing either MP or SCS preservation (P = 0.683).
Transplant Outcomes
Graft survival and patient survival were evaluated at both 1 and 3 y posttransplant by Kaplan-Meier methods (Figure 2). Livers undergoing MP preservation demonstrated improved 1-y (91.8% versus 88.4%, P = 0.013) and 3-y (80.0% versus 80.2%; P = 0.036) graft survival after DCD liver transplantation as compared with SCS-preserved livers. There was no significant difference in graft outcomes after DBD donation at either 1 y (90.1% versus 91.2%; P = 0.486) or 3 y (87.0% versus 84.4%; P = 0.516) posttransplant. Multivariable analysis of graft survival likewise demonstrated significant improvement in DCD graft survival after MP at 1 y (hazard ratio [HR] 0.57 [95% CI, 0.37-0.87]; P = 0.009) and 3 y (HR 0.65 [95% CI, 0.46-0.95]; P = 0.027) posttransplant compared with SCS livers. For DBD liver transplants, there was no difference in 1-y DBD graft survival (HR 0.89 [95% CI, 0.68-1.17]; P = 0.413) or 3-y graft survival (HR 0.91 [95% CI, 0.71-1.18]; P = 0.494) between MP and SCS cohorts on multivariable analysis.
FIGURE 2.
Three-year graft survival (left) and patient survival (right), stratified by type of organ preservation (MP vs SCS) and donor type (DBD vs DCD). DBD, donation after brain death; DCD, donation after circulatory death; MP, machine perfusion; SCS, standard cold storage.
Overall patient survival was similar at 1 or 3 y posttransplant after both DBD (1-y 92.1% versus 92.7%, P = 0.803; 3-y 89.3% versus 86.2%, P = 0.245) and DCD (1-y 93.7% versus 92.7%, P = 0.111; 3-y 82.6% versus 85.6%, P = 0.243) liver transplant. There was no difference in retransplantation rate after DBD liver transplant between MP and SCS livers (1.95% versus 1.96%; P = 0.991). However, patients receiving MP livers from DCD donors were less likely to require retransplantation (1.49% versus 5.21%; P < 0.001).
DISCUSSION
This study used the OPTN/STAR database to retrospectively evaluate current practices of MP utilization in the United States and its impact on organ acceptance and transplant outcomes in liver transplantation. We specifically chose livers procured with intent for transplant as our inclusion criteria in this study, as previously described.11 This allowed us to focus on intraoperative organ decline or acceptance to evaluate the specific implications of MP usage on organ utilization and posttransplant outcomes. Furthermore, DBD and DCD cohorts were separated for analysis, as they represent 2 distinct populations in liver transplantation with associated organ acceptance practices and risk factors. The results of this study demonstrate a rapid increase in the use of MP preservation since 2021, accompanied by increased organ utilization for both DBD and DCD livers undergoing MP preservation. This is the first national report to demonstrate improved graft survival at 1-y and 3-y posttransplant for DCD livers undergoing MP preservation.
This study adds to the small but important body of existing data describing national MP utilization in the United States. NMP utilization was previously shown to be increasing in frequency as of June 2022, with donor risk index preferentially increasing in the NMP group as compared with SCS-preserved livers.12 A recent publication analyzing a subset of the OPTN/STAR database for NMP utilization through December 2022 demonstrated higher donor risk index in livers undergoing NMP preservation, with lower organ discard rates and shorter patient length of stay after NMP preservation.13 In that study, there was no difference in 1-y graft survival between NMP and SCS-preserved livers.13 Based on the Federal Drug Administration (FDA) approval of these devices in 2021, it can be inferred that utilization of HMP and NMP before 2021 was likely performed as part of clinical trials, with NMP utilization after 2021 representing real-world use. However, given the profound expansion of MP use since 2021, with nearly 40% of all historical MP livers occurring in 2023 alone, this study provides the most comprehensive evaluation of MP practices in the United States to date.
Here, we demonstrate increased utilization of MP livers after both DBD and DCD as compared with SCS preservation. Although there was overall high utilization of DBD livers in both groups (96.4% MP versus 93.0% SCS), there was a marked increase in DCD liver utilization after MP preservation (91.4% MP versus 74.0% SCS). MP preservation had a particular impact on the utilization of liver allografts with biopsy-confirmed macrosteatosis over 20%, with an 18.9% absolute increase in DBD utilization and a 59.9% increase in DCD utilization. This should be considered in the context of MP liver donors having higher BMI for both DBD and DCD donors, as well as higher donor age of DBD livers undergoing MP. There is some evidence that MP preservation may mitigate the effects of graft steatosis on reperfusion syndrome and early allograft dysfunction14,15; however, these results must be considered in the context of lower MELD scores and longer wait times among MP liver recipients.
In particular for DCD livers, higher organ utilization may be attributable to a natural selection bias, whereas the commitment to mobilize a MP team and a nonrecoverable procurement cost may have led some centers to more selectively use MP for the highest quality donors. However, the value of viability testing in NMP provides reassurance of organ quality for marginal livers and contributes to the utilization of marginal livers that otherwise would have been declined in the absence of NMP preservation. The current utilization of more marginal allografts with MP may also reflect appropriate donor/recipient matching that facilitates access to transplants for patients with lower MELD scores, especially those who have been waiting long periods for an acceptable graft. This approach is further supported by graft and patient outcomes, with equivalent outcomes for DBD livers and superior graft survival for DCD livers undergoing MP preservation.
It is important to note that our analysis combines HMP and NMP modalities for analysis, with NMP representing the predominant MP technique currently in the United States. Currently, OrganOx Metra (OrganOx Limited, Oxford) and TransMedics Organ Care System (TransMedics, Inc, Andover, MA) are the only FDA-approved MP devices in the United States, both being NMP devices. Several hypothermic oxygenated MP (HOPE) devices are currently under evaluation by the FDA, which will further expand commercially available options for MP use. HOPE and NMP use distinct strategies to provide benefit to the liver allograft and can be used in a complementary manner depending on organ quality. HOPE decreases metabolic demands of the liver by perfusing below 12 °C, with supplemental oxygen delivery via perfusate providing for “reconditioning” of the liver by recharging and maintaining ATP stores.16 By maintaining oxygen to support the electron transport chain in liver mitochondria, HOPE prevents backward flow of electrons through complex 1, helping to mitigate the cascade of cell injury initiating ischemia/reperfusion injury, preserve endothelial function, and prevent accumulation of proinflammatory cytokines as during SCS.10,16-18 NMP uses a different approach, maintaining perfusion of the donor organ at physiologic temperatures and providing dual inflow via continuous portal flow and pulsatile arterial flow.16 Given the metabolic demands at 37 °C, perfusate in NMP is blood-based and supplemented with nutrients, trace elements, oxygen, and bile salts.16 NMP initiated at the donor site minimizes ischemic time and flushes inflammatory cytokines from the allograft, thereby removing many factors, which contribute to ischemia/reperfusion before implantation.19 Additionally, maintenance of the liver at physiologic conditions allows for real-time viability via lactate clearance and bile production, which can inform organ acceptance practices for marginal grafts.19
Currently, there are no standardized algorithms or established indications for either when to use MP or how to select between HMP, HOPE, and NMP modalities. As new devices and options continue to reach the US market, the selection of MP modality will be influenced by clinical needs, device availability, overall logistics, and cost. Given the existing evidence demonstrating improved early allograft outcomes in DCD liver transplant and data here demonstrating increased DCD liver utilization and improved 1- and 3-y graft survival in MP-preserved livers, further discussion is warranted whether SCS alone remains appropriate for DCD.1,3,20 Over 35% of DCD livers underwent MP preservation in 2023, with numbers likely to increase as continuing data emerge and more transplant programs integrate MP into clinical practice. Normothermic regional perfusion (NRP) represents 1 alternative to improve graft utilization and outcomes after DCD, but NRP serves as a procurement technique and may still be combined with MP strategies as needed to ensure utilization of marginal grafts.21 We feel that either MP or NRP should be recommended to optimize patient outcomes after DCD liver donation, and we concur with a recent publication that utilization of ≥1 advanced perfusion technologies may soon become the standard of care for DCD liver transplant.22
There are several limitations to this study. By nature, analysis of a national database is subject to a lack of granularity and an inability to capture significant posttransplant outcomes that would inform practice. Although the addition of an identifying field to denote MP utilization is a critical addition to the OPTN/STAR database, the widespread prevalence of this technology reaffirms that additional MP-specific variables should be captured to maintain relevancy as a critical resource to the transplant community. In particular, this data source lacks identifiers to capture livers that underwent either NRP or MP in a “back-to-base” strategy, where livers are procured and transported to the transplant center via SCS, at which point they are placed on either HMP or NMP before transplant. Although this historically represents a small subset of livers that are unlikely to confound these results based on the large population size, both NRP and “back-to-base” MP look to be important pathways to increase organ utilization, which should be captured on the national level for further study. The current system would also benefit from a modernization of recording ischemic times, as the current system is unable to distinguish the time and setting of MP initiation or true cold time versus MP time for MP livers. This prevents an effective analysis of ischemia times and pump times on graft outcomes. Finally, the OPTN/STAR database does not effectively capture notable outcomes related to MP, namely early allograft dysfunction and ischemic cholangiopathy. Improved tracking of these posttransplant outcomes would significantly strengthen the utility of this database in the MP era.
As there are no established or mandatory criteria surrounding MP use, it is unclear from database data what indication prompted MP use. Current MP practices reflect a combination of graft risk factors, recipient complexity, and transplant center logistics. MP of standard-quality allografts to facilitate daytime transplants represents a major shift in the practice of liver transplant,7,13 but also serves as a potential confounder to analysis seeking the true efficacy of MP on marginal allografts. National data collection associated with MP utilization should capture the indications for initiating MP, as well as full documentation of pump times and metrics. Most importantly, although this study is the most complete picture of MP practices thus far, MP utilization is rapidly evolving withing the field of liver transplantation. As this latest vintage of MP recipients matures and MP utilization continues to expand, further studies will be required to fully capture the impact of MP use on patient and graft outcomes.
MP preservation is quickly proving to be a transformational technology in liver transplantation. Continued expansion of MP must be guided by data, fully capturing the capabilities and limitations of the technology. In addition to national registry data, multicenter consortiums can provide the granular data to fully understand what features of a marginal graft can be safely overcome with MP preservation and to confirm that MP preservation ensures high-quality medium- and long-term graft outcomes. In contrast, a better understanding of the graft-related issues MP cannot improve, as well as the incidence of technical complications associated with MP—including organ discard rates after the initiation of MP, rates of procurement-related injury, and vascular complication during MP cases—will improve efficient utilization of this technology. Even with significant enthusiasm for MP increasing the donor pool, it is notable that graft failure and retransplantation still occur, and further analysis of factors leading to graft failure after MP preservation should continue to optimize future use. Ideally, these data should be collected and monitored independently by device manufacturers to compare across devices and ensure that the full scope of adverse outcomes is acknowledged. Finally, robust cost analysis must be performed across all devices to ensure responsible device utilization for the financial health of transplant institutions and the transplant network at large.
CONCLUSIONS
MP use continues to rise in the United States, with a resultant increase in utilization of livers procured with intent for transplant. In particular, MP increases the utilization of livers from DCD donors and livers with biopsy-proven steatosis. Importantly, MP is associated with improved graft survival after DCD liver transplantation at 1 and 3 y posttransplant, whereas DBD liver transplant outcomes are equivalent despite transplantation from older donors with higher BMI. As MP use continues to expand, outcomes and impact on graft utilization must be monitored across technologies to fully identify the maximal patient benefit.
Footnotes
The authors declare no funding or conflicts of interest.
S.A.W., J.A.S., and I.K.K. participated in research design, article preparation, and editing. S.A.W. wrote the article. S.A.W. and J.A.S. participated in performing the research and data analysis.
Contributor Information
Justin A. Steggerda, Email: justin.steggerda@cshs.org.
Irene K. Kim, Email: irene.kim@cshs.org.
REFERENCES
- 1.Markmann JF, Abouljoud MS, Ghobrial RM, et al. Impact of portable normothermic blood-based machine perfusion on outcomes of liver transplant: the OCS Liver PROTECT randomized clinical trial. JAMA Surg. 2022;157:189–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Nasralla D, Coussios CC, Mergental H, et al. A randomized trial of normothermic preservation in liver transplantation. Nature. 2018;557:50–56. [DOI] [PubMed] [Google Scholar]
- 3.van Rijn R, Schurink IJ, de Vries Y, et al. ; DHOPE-DCD Trial Investigators. Hypothermic machine perfusion in liver transplantation—a randomized trial. N Engl J Med. 2021;384:1391–1401. [DOI] [PubMed] [Google Scholar]
- 4.Clavien PA, Dutkowski P, Mueller M, et al. Transplantation of a human liver following 3 days of ex situ normothermic preservation. Nat Biotechnol. 2022;40:1610–1616. [DOI] [PubMed] [Google Scholar]
- 5.Dengu F, Abbas SH, Ebeling G, et al. Normothermic machine perfusion (NMP) of the liver as a platform for therapeutic interventions during ex-vivo liver preservation: a review. J Clin Med. 2020;9:1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mergental H, Laing RW, Kirkham AJ, et al. Transplantation of discarded livers following viability testing with normothermic machine perfusion. Nat Commun. 2020;11:2939. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Das I, Mathur AK, Aqel B, et al. “To sleep-perchance to dream”: daytime surgery start times for liver transplantation with ex situ normothermic machine perfusion. Liver Transpl. 2024;30:763–767. [DOI] [PubMed] [Google Scholar]
- 8.Faria I, Canizares S, Devos L, et al. Machine perfusion organ preservation: highlights from the American Transplant Congress 2023. Artif Organs. 2024;48:794–799. [DOI] [PubMed] [Google Scholar]
- 9.Hefler J, Leon-Izquierdo D, Marfil-Garza BA, et al. Long-term outcomes after normothermic machine perfusion in liver transplantation—experience at a single North American center. Am J Transplant. 2023;23:976–986. [DOI] [PubMed] [Google Scholar]
- 10.Guarrera JV, Henry SD, Samstein B, et al. Hypothermic machine preservation facilitates successful transplantation of “orphan” extended criteria donor livers. Am J Transplant. 2015;15:161–169. [DOI] [PubMed] [Google Scholar]
- 11.Wisel SA, Steggerda JA, Thiessen C, et al. Preserved 2-y liver transplant outcomes following simultaneous thoracoabdominal DCD organ procurement despite effects on liver utilization rate. Transplant Direct. 2023;9:e1528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Abu-Gazala S, Tang H, Abt P, et al. National trends in utilization of normothermic machine perfusion in DCD liver transplantation. Transplant Direct. 2024;10:e1596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wang BK, Shubin AD, Harvey JA, et al. From patients to providers: assessing impact of normothermic machine perfusion on liver transplant practices in the US. J Am Coll Surg. 2024;238:844–852. [DOI] [PubMed] [Google Scholar]
- 14.Czigany Z, Lurje I, Schmelzle M, et al. Ischemia-Reperfusion injury in marginal liver grafts and the role of hypothermic machine perfusion: molecular mechanisms and clinical implications. J Clin Med. 2020;9:846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kron P, Schlegel A, Mancina L, et al. Hypothermic oxygenated perfusion (HOPE) for fatty liver grafts in rats and humans. J Hepatol. 2017;68:82–91. [DOI] [PubMed] [Google Scholar]
- 16.Sousa Da Silva RX, Weber A, Dutkowski P, et al. Machine perfusion in liver transplantation. Hepatology. 2022;76:1531–1549. [DOI] [PubMed] [Google Scholar]
- 17.Chouchani ET, Pell VR, Gaude E, et al. Ischaemic accumulation of succinate controls reperfusion injury through mitochondrial ROS. Nature. 2014;515:431–435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Burlage LC, Karimian N, Westerkamp AC, et al. Oxygenated hypothermic machine perfusion after static cold storage improves endothelial function of extended criteria donor livers. HPB (Oxford). 2017;18:e4–546. [DOI] [PubMed] [Google Scholar]
- 19.Hefler J, Marfil-Garza BA, Dadheech N, et al. Machine perfusion of the liver: applications beyond transplantation. Transplantation. 2020;104:1804–1812. [DOI] [PubMed] [Google Scholar]
- 20.van Leeuwen OB, de Vries Y, Fujiyoshi M, et al. Transplantation of high-risk donor livers after ex situ resuscitation and assessment using combined hypo- and normothermic machine perfusion: a prospective clinical trial. Ann Surg. 2019;270:906–914. [DOI] [PubMed] [Google Scholar]
- 21.Brubaker AL, Sellers MT, Abt PL, et al. US liver transplant outcomes after normothermic regional perfusion versus standard super rapid recovery. JAMA Surg. 2024;159:677–685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Croome KP. Should advanced perfusion be the standard of care for donation after circulatory death liver transplant? Am J Transplant. 2024;24:1127–1131. [DOI] [PubMed] [Google Scholar]