Abstract
Background.
Understanding factors that contribute to liver discards and nonusage is urgently needed to improve organ utilization.
Methods.
Using Scientific Registry of Transplant Recipient data, we studied a national cohort of all US adult, deceased brain dead donor, isolated livers available for transplantation from 2003 to 2016, including organ-specific and system-wide factors that may affect organ procurement and discard rates.
Results.
Of 73686 available livers, 65316 (88.64%) were recovered for transplant, of which 6454 (9.88%) were ultimately discarded. Livers that were not procured or, on recovery, discarded were more frequently from older, heavier, hepatitis B virus (HCV)+, and more comorbid donors (P < 0.001). However, even after adjustment for organ quality, the odds of liver nonusage were 11 % higher on the weekend (defined as donor procurements with cross-clamping occurring from 5:00 PM Friday until 11:59 AM Sunday) compared with weekdays (P < 0.001). Nonuse rates were also higher at night (P < 0.001), defined as donor procurements with cross-clamping occurring from 5:00 PM to 5:00 AM; however, weekend nights had significantly higher nonuse rates compared with weekday nights (P = 0.005). After Share 35, weekend nonusage rates decreased from 21.77% to 19.51% but were still higher than weekday nonusage rates (P = 0.065). Weekend liver nonusage was higher in all 11 United Network of Organ Sharing regions, with an absolute average of 2,00% fewer available livers being used on the weekend compared with week-days.
Conclusions.
Although unused livers frequently have unfavorable donor characteristics, there are also systemic and operational factors, including time of day and day of the week a liver becomes available, that impact the chance of liver nonprocurement and discard.
INTRODUCTION
Despite the shortage of organ donors, a substantial number of organs recovered for transplantation are not used in a recipient (discards). Solid organ discard rates have risen over the last decade from ~11% in 2000 to ~14% in 20121; discards in liver transplantation have followed similar trends, recently reported as ~10% per year for all recovered livers.1 Efforts to understand these trends in kidney transplantation suggest that resource limitations related to weekends may be a contributing factor.2 However, it is unclear if other organs are similarly affected by this weekend effect.
A number of approaches have been used to expand the organ donor pool and increase organ procurement. Although these attempts have included efforts to increase organ donation via initiatives such as organ donor registration drives, the establishment of better online registries, and best practice guidelines for organ procurement organizations (OPOs), they may have resulted in higher procurement rates of organs with an elevated predicted risk of graft failure or disease transmission (extended criteria donors [ECDs]) that may not be accepted for transplantation.3 But whether and to what degree organs from such donors are used or not and what factors impact their utilization trends is not currently known.4 In addition, the relationship between the recovery of ECD livers and the increase in discards has not been fully explored,5,6 and many of the criteria for ECD organs are not well defined.7–9
For any given liver, there are a variety of donor, recipient, and organ-specific factors that likely play a role in why a center chooses to discard a particular organ. But given the wide OPO and regional variation in utilization and discard rates, it is likely that reasons other than organ quality may be contributing factors.5,10–12 Utilizing centers vary widely in their acceptance criteria, in which Garonzik-Wang et al13 has described an “aggressiveness” phenotype for both liver and kidney programs. Recognizing that liver selection is a complex, multifactorial process, we attempt to understand utilization trends in liver transplantation given that 2456 people die waiting for transplant and 482 people are removed from the waiting list annually because they were too sick to undergo transplant.1 We also seek to discern whether livers, like kidneys, are similarly affected by potential resource limitations as suggested by the weekend effect.
MATERIALS AND METHODS
Study Design and Participants
We utilized data from the Scientific Registry of Transplant Recipients (SRTR) standard analytical file (2016 Quarter 1) to identify all adult, deceased brain dead donor, isolated livers available for transplant in the United States in the post-Model for End-stage Liver Disease (MELD) era from 2003 to 2016. We separately analyzed the effects of Share 35, an effort in June 2013 to increase the availability of livers to the sickest recipients by requiring that deceased donor liver offers are first made to regional candidates with MELD scores >35 before other local and regional candidates with MELD scores <35. In our Share 35 analysis, we compared livers recovered from 2003 to 2012 with livers recovered July 1, 2013 through March 2016, which was the end date for our first Quarter 2016 data.
Specifics of the SRTR data file have been described elsewhere.1 Briefly, the SRTR data system includes data on all donor, waitlisted candidates, and transplant recipients in the United States, submitted by the members of the Organ Procurement and Transplantation Network. The Health Resources and Services Administration, US Department of Health and Human Services provides oversight to the activities of the Organ Procurement Transplantation Network and SRTR contractors.
For this analysis, livers from living and pediatric donors (aged <18 years), and nonisolated liver transplant offers were excluded. Donation after cardiac death (DCD) donors were also excluded since they are felt to be sufficiently different from brain dead donors and only account for ~5.6% of transplants in the study period; however, a brief analysis of these donors is included in an SDC. A procured liver was defined as a liver that was recovered for transplant (procurement rate = No. of procured/No. of available), and a discarded liver was defined as an organ that was procured but was not ultimately transplanted (discard rate = No. of discarded/No. of procured). Liver nonusage was defined as a liver that was either not procured or, if procured, discarded (combining nonprocurement and discard into a single category; nonuse rate = (No. of discarded + No. of not procured)/No. of available for procurement).
Donor, organ-specific, and systemic factors that could potentially contribute to the discard or nonuse of a liver were analyzed in both univariable and multivariable fashion. The impact of these variables on the procurement and discard rate of available organs was then studied in our cohort.
For our day of the week analysis, we had to create a methodology that would allow us to estimate what day of the week a transplant would have occurred on for those livers that were not ultimately transplanted. To do this, we looked at cases where transplants actually happened and found that procurement cross-clamps after 5:00 pm were more likely to have an SRTR transplant date listed as the day after the procurement (this also corresponded to the cold ischemic time over our study period; median = 6.5 hours, mean = 6.9 hours). Thus, the methodology we chose defined “likely day of transplant” according to whether the cross-clamp time for the procurement took place before or after 5:00 pm. So, if the cross-clamp time took place, for example, before 5:00 pm on a Tuesday, then the likely day of transplant was considered to be Tuesday as well. However, if the cross-clamp time was after 5:00 pm on Tuesday, then the likely day of transplant was considered to be Wednesday. In cases where the liver was offered but not procured, we used the cross-clamp time for the organs that were procured. For our weekend definition, we used a cross-clamp time from 5:00 pm on Friday until 11:59 pm on Sunday night, which accounted for nearly ~95% of transplants that occurred on Saturday or Sunday. Nighttime transplant were then defined as 5:00 pm in the evening until 5:00 am in the morning; conversely, daytime transplants were from 5:00 am until 5:00 pm. This minimized the misclassification of liver discards with regard to when they were intended to be transplanted (ie, weekday versus weekend), and on average correctly identified ~89% of transplants on any given day of the week.
Statistical Analysis
Data were analyzed using Stata MP version 14,1 (Stata Corp., College Station, TX). Categorical variables were reported as frequencies and percentages, and were analyzed using chi-squared analysis. Continuous variables were reported as means with SDs and were analyzed using Student t test/ANOVA as appropriate or their nonparametric counterpart as indicated. After univariable analysis, a multivariable logistic regression model was built determining the odds of liver nonusage compared with transplantation. All variables that were significant in univariable analysis (P < 0.20) for a liver being not used were also examined in the multivariable analysis. Odds of nonuse were presented with corresponding 95% confidence intervals and P values for each variable examined, with a significance level set at 0.05. Cases with missing values for any potential covariate were excluded from the final analysis (no included variable was missing in >1% of cases). The Columbia University Medical Center Institutional Review Board approved the study.
RESULTS
Liver Allograft Utilization in Our Cohort
From 2003 to 2016, there were 73 686 adult, deceased brain dead donor, isolated livers available for procurement; of these, 65316 were procured for transplant (88.64% of livers available), but only 58 862 (90.12% of livers procured) were ultimately transplanted. As a result, over the study period 9.88% (6454) of livers procured for transplantation were discarded (Figure 1). The rate of liver discard ranged from a low of 7.62% in 2003, just at the beginning of the MELD era, to a peak of 11.45% in 2007—rates have been stable at approximately ~9% to 10% in more recent years.
FIGURE 1.
Flow diagram for our study cohort–all adult, deceased donor, isolated livers available for transplant (2003–2016).
Liver Quality Between Procured/Nonprocured and Discarded/Transplanted Organs
Table 1 provides a descriptive analysis of available livers stratified by both procurement and transplantation status over 2003–2016 (see also Table S1, SDC, http-//links.lww.com/TP/B657). Of available livers, nonprocured livers had more risk factors for graft failure than procured livers: donors with nonprocured livers were older (mean age 47.8 versus 44.8 years; P < 0.001), had a higher body mass index (BMI) (29.6kg/m2 versus 27.7kg/m2; odds ratio (OR), 1.55 for those with a BMI >30 kg/m2 compared with those with a BMI <30kg/m2), were more likely to be HCV+ (11.4% versus 4.4%; OR, 2.76), had a higher mean aspartate transaminase (323 versus 86; OR, 3.52 for an aspartate transaminase >250), and were more likely to have a total bilirubin >2.0 (2.8% versus 7.6%; OR, 3.58) (all P < 0.001; Table S1, SDC, http Minks.Iww.com/TP/B657). Similarly, of procured livers, discarded livers had more markers of an elevated predicted risk of graft failure or disease transmission than transplanted livers: discarded livers were more likely to be from donors who were older (OR, 2.10 for those who were >60 years old compared with those <60 years), had a higher BMI (OR, 2.05 for those with a BMI >30kg/m2 compared with those with a BMI <30 kg/m2), were more likely to be HCV+ (9.6% versus 3.9%; OR, 2.64), and be more comorbid, among other factors (Table 1). Collectively, these data show that donors whose livers were either never procured or, if procured, were discarded had more risk factors than donors who had their livers transplanted.
TABLE 1.
Description of livers available for adult, deceased brain dead donor, isolated liver transplants by recovery status and, if procured, transplanted vs discarded (2003-2016)
| Livers procured for transplant | Likelihood of discard | ||||
|---|---|---|---|---|---|
| Donor characteristics | Total procured (N = 65316) | Transplanted (n = 58862) | Discarded (n = 6454) | Crude OR | P |
| Age, mean (SD) | 44.37 (16.02) | 43.56 (16.00) | 51.83 (14.30) | 1.034 (1.032–1.035) | <0.001 |
| Age >60 y (%) | 12287 (18.81) | 10302 (17.50) | 1985 (30.76) | 2.09 (1.98–2.22) | <0.001 |
| Sex (% female) | 27372 (41.91) | 24496 (41.62) | 2876 (44.56) | 1.13 (1.07–1.19) | <0.001 |
| Race (%) | |||||
| White | 43471 (65.56) | 38963 (66.19) | 4508 (69.85) | 1.18 (1.12–1.25) | <0.001 |
| African American | 11352 (17.38) | 10492 (17.83) | 860 (13.33) | 0.71 (0.66–0.76) | <0.001 |
| Other | 10492 (16.06) | 9406 (15.98) | 1086 (16.83) | 1.06 (0.99–1.14) | 0.079 |
| BMI, mean (SD) | 27.70 (6.39) | 27.44 (6.17) | 30.10 (7.75) | 1.56 (1.53–1.60) | <0.001 |
| Obese (% with BMI >30 kg/m2) | 18176 (28.05) | 15489 (26.48) | 2687 (42.52) | 2.05 (1.95–2.17) | <0.001 |
| Cause of death (%) | |||||
| Head trauma | 21049 (32.23) | 19809 (33.65) | 1240 (19.21) | 0.47 (0.44–0.50) | <0.001 |
| Anoxia | 13577 (20.79) | 12064 (20.50) | 1513 (23.44) | 1.19 (1.12–1.26) | <0.001 |
| Stroke | 29129 (44.60) | 25593 (43.48) | 3536 (54.79) | 1.58 (1.50–1.66) | <0.001 |
| Other | 1561 (2.39) | 1396 (2.37) | 165 (2.56) | 1.08 (0.92–1.27) | 0.356 |
| Donor AST, mean (SD) (OR is for increments of 50) | 86.19 (212) | 80.20 (157) | 141 (478) | 1.009 (1.008–1.01) | <0.001 |
| AST >250 | 4,074 (6.24) | 3398 (5.77) | 676 (10.47) | 1.91 (1.75–2.08) | <0.001 |
| Donor ALT, mean (SD) (OR is for increments of 50) | 76.25 (189) | 71.39 (161) | 120 (349) | 1.009 (1.007–1.01) | <0.001 |
| Donor sodium, mean (SD) | 147 (8.48) | 147 (8.51) | 147 (8.27) | 0.999 (0.096–1.002) | 0.613 |
| Donor total bilirubin, total (mean [SD]) | 0.94 (1.13) | 0.93 (1.12) | 1.04 (1.21) | 1.06 (1.04–1.08) | <0.001 |
| Total bilirubin >2 (%) | 4941 (7.62) | 4280 (7.32) | 661 (10.31) | 1.45 (1.33–1.58) | <0.001 |
| Donor diabetes (%) | 8364 (12.88) | 6881 (11.75) | 1483 (23.13) | 2.26 (2.12–2.41) | <0.001 |
| Donor hypertension (%) | 26189 (40.34) | 22502 (38.46) | 3687 (57.56) | 2.17 (2.06–2.29) | <0.001 |
| HCV (%HCV Ab positive) | 2901 (4.44) | 2281 (3.88) | 620 (9.61) | 2.64 (2.40–2.89) | <0.001 |
| HBV (%Hb core Ab positive) | 3867 (5.92) | 3290 (5.59) | 577 (8.94) | 1.66 (1.51–1.82) | <0.001 |
| CDC high risk (%) | 7396 (12.59) | 6724 (12.72) | 672 (11.35) | 0.88 (0.81–1.95) | 0.002 |
| Split (%) | 790 (1.19) | 740 (1.26) | 40 (0.62) | 0.49 (0.36–0.67) | <0.001 |
| Biopsy (% yes) | 26768 (41.14) | 21729 (37.05) | 5039 (78.54) | 6.22 (5.84–6.61) | <0.001 |
| Macrosteatosis (>30%) if biopsy | 3188 (14.15) | 1503 (8.25) | 1685 (39.02) | 7.11 (6.56–7.71) | <0.001 |
| ET0H dependency in donor (%) | 9463 (16.31) | 8247 (15.81) | 1216 (20.77) | 1.40 (1.31–1.49) | <0.001 |
| Donor OPO volume, by quartile (%) | |||||
| Lowest quartile | 5910 (9.05) | 5306 (9.01) | 604 (9.36) | 1.04 (0.94–1.14) | 0.36 |
| Second quartile | 11816 (18.09) | 10782 (18.32) | 1034 (16.02) | 0.85 (0.79–0.91) | <0.001 |
| Third quartile | 17501 (26.79) | 15776 (26.80) | 1725 (26.73) | 0.99 (0.94–1.06) | 0.899 |
| Highest quartile | 30089 (46.07) | 26998 (45.87) | 3091 (47.89) | 1.08 (1.03–1.14) | 0.002 |
| Transplant era | |||||
| 2003-2006 | 20185 (30.90) | 18314 (31.11) | 1871 (28.99) | 0.90 (0.85–0.96) | <0.001 |
| 2007-2009 | 15045 (23.03) | 13367 (22.71) | 1678 (26.00) | 1.20 (1.13–1.27) | <0.001 |
| 2010-2012 | 15567 (23.83) | 14003 (23.79) | 1564 (24.23) | 1.03 (0.96–1.09) | 0.427 |
| 2013-2016 | 14519 (22.23) | 13178 (22.39) | 1341 (20.78) | 0.91 (0.85–0.97) | 0.003 |
| Weekend procurement events (%) | 20642 (31.60) | 18490 (31.41) | 2152 (33.34) | 1.09 (1.03–1.15) | 0.002 |
Descriptive statistics for adult, deceased brain dead donor, isolated liver offers for transplant: broken down by recovery status and then, if recovered, transplanted/used vs discarded organs. CDC high risk = donation considered CDC high risk. ETOH dependency = whether the donor consumed >2 alcoholic drinks daily. Donor OPO volume = quartile of OPOs for the number of transplants per year averaged over the study period; weekend defined as donor cross-clamp time from Friday 5:00 PM to Sunday 11:59 PM.
Ab, antibody; ALT, alanine transaminase; AST, aspartate transaminase; BMI, body mass index; CDC, Centers for Disease Control and Prevention; Hb, hemoglobin; HBV, hepatitis B virus; HCV, hepatitis C virus; ETOH, ethanol; OR, odds ratio; OPO, organ procurement organization.
The Weekend Effect
In addition to differences in organ quality noted above, there was also significant variation in the percentage of livers not procured or discarded over the course of a calendar week. Table 2 displays the effect of likely day of transplant on the available, procured, transplanted, and discarded livers. Liver nonuse rates were lowest on Monday (19.33%), with rates increasing throughout the week and ultimately peaking on Saturday (21.09%) and Sunday (21.34%). Comparing weekends versus weekdays, nonprocurement (12.09% versus 11.02%; P < 0.001), discard (10.43% versus 9.63%; P = 0.002), and nonuse (21.26% versus 19.58%; P < 0.001) rates were all higher on the weekend compared with weekdays.
TABLE 2.
Adult deceased brain dead donor, isolated livers available for transplant by day of week versus weekend (2003–2016): likely day of transplant
| Likely day of week of transplant | Weekday versus weekend | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Liver availability and utilization | Total | Monday | Tuesday | Wednesday | Thursday | Friday | Saturday | Sunday | Weekdaya | Weekendb | P |
| Available for procurement | 73686 | 9649 | 10708 | 11170 | 11079 | 10470 | 10640 | 9970 | |||
| Percentage available for procurement (row %) | 100 | 13.09 | 14.53 | 15.16 | 15.04 | 14.21 | 14.44 | 13.53 | 68.13 | 31.87 | – |
| Procured for transplant | 65316 | 8597 | 9536 | 9945 | 9808 | 9306 | 9378 | 8746 | |||
| Procurement rate (no. of procured/no. of available) | 88.64 | 89.1 | 89.05 | 89.03 | 88.53 | 88.88 | 88.14 | 87.72 | 88.98 | 87.91 | <0.001 |
| Percentage procured for transplant each day/period (row %) | 100 | 13.16 | 14.60 | 15.23 | 15.02 | 14.25 | 14.36 | 13.39 | – | – | |
| Discarded | 6454 | 813 | 918 | 949 | 972 | 916 | 982 | 904 | |||
| Percentage discarded from each day/period (discard rate) | 9.88 | 9.46 | 9.63 | 9.54 | 9.91 | 9.84 | 10.47 | 10.34 | 9.63 | 10.43 | 0.002 |
| Transplanted | 58862 | 7784 | 8618 | 8996 | 8836 | 8390 | 8396 | 7842 | |||
| Percentage of transplantation rate (no. of transplanted/no. of procured) | 90.12 | 90.54 | 90.37 | 90.46 | 90.09 | 90.16 | 89.53 | 89.66 | 90.37 | 89.57 | 0.002 |
| Not used | 14824 | 1865 | 2090 | 2174 | 2243 | 2080 | 2244 | 2128 | |||
| Percentage of organs either discarded or not procured | 20.12 | 19.33 | 19.52 | 19.46 | 20.25 | 19.87 | 21.09 | 21.34 | 19.58 | 21.26 | <0.001 |
Likely day of transplant defined in reference to the cross-clamp time in the donor operation.
Weekday defined as likely day of transplant from Monday to Friday (donor cross-clamp time from 12:00 AM Monday morning through 5:00 PM Friday afternoon.
Weekend defined as donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM.
Given these differences in organ usage by likely day of transplant, we then examined whether there were differences in organ quality for livers available for weekday compared with weekend transplants (Table 3; see also Table S2, SDC, http://links.lunv.com/TP/B657). Based on a variety of factors, there were no major differences in organ quality for livers available during the week compared with the weekend: livers were of similar age, sex, race, comorbidities, laboratory values, Centers for Disease Control and Prevention high risk status, and rates of macrosteatosis on biopsy. Weekend donors were also less likely to be used as split livers (P < 0.001) and were less likely to have their livers biopsied if they were not procured, possibly indicating less interest in evaluating marginal allografts on the weekend. Although nonprocured livers during the weekday did have higher HCV+ rates (P < 0.001) (Table S3, SDC, http://links.ltvw.com/TP/B657).
TABLE 3.
Description of livers from adult, deceased brain dead donor, isolated livers by weekday vs weekend procurement (2003-2016)
| Livers procured for transplant | ||||
|---|---|---|---|---|
| Donor characteristics | Total (N = 65316) | Weekdaya (n = 44674) | Weekendb (n = 20642) | P |
| Age, mean, (SD) | 44.37 (16.02) | 44.43 (16.08) | 44.25 (15.89) | 0.1648 |
| Age >60 y (%) | 12287 (18.81) | 16864 (19.05) | 3778 (18.30) | 0.024 |
| Sex (% female) | 27372 (41.91) | 18700 (41.86) | 8672 (42.01) | 0.713 |
| Race (%) | ||||
| White | 43471 (66.56) | 29691 (66.46) | 13780 (66.76) | 0.459 |
| African American | 11352 (17.38) | 7765 (17.38) | 3587 (17.38) | 0.988 |
| Other | 10492 (16.06) | 7217 (16.16) | 3275 (15.87) | 0.349 |
| BMI, mean (SD) | 27.70 (6.39) | 27.71 (6.39) | 27.68 (6.39) | 0.5637 |
| Obese (% with BMI >30 kg/m2) | 18176 (28.05) | 12444 (28.07) | 5732 (27.99) | 0.834 |
| Cause of death (%) | ||||
| Head trauma | 21 049 (32.23) | 14365 (32.1) | 6684 (32.38) | 0.567 |
| Anoxia | 13577 (20.79) | 9343 (20.91) | 4234 (20.51) | 0.239 |
| Stroke | 29129 (44.60) | 19937 (44.63) | 9192 (44.53) | 0.816 |
| Other | 1561 (2.39) | 1029 (2.30) | 532 (2.58) | 0.033 |
| Donor AST, mean (SD) | 86 (212) | 86 (218) | 86 (199) | 0.9192 |
| AST >250 | 4074 (6.24) | 2772 (6.20) | 1302 (6.31) | 0.614 |
| Donor ALT, mean (SD) | 76 (188) | 76 (190) | 76 (186) | 0.9911 |
| Donor sodium, mean (SD) | 147 (8.48) | 147 (8.47) | 147 (8.51) | 0.0237 |
| Donor bilirubin, total (mean [SD]) | 0.93 (1.13) | 0.93 (1.06) | 0.93 (1.26) | 0.9758 |
| Total bilirubin >2 (%) | 4941 (7.62) | 3369 (7.59) | 1572 (7.68) | 0.709 |
| Donor diabetes (%) | 8364 (12.88) | 5761 (12.97) | 2603 (12.80) | 0.309 |
| Donor hypertension (%) | 26189 (40.34) | 18011 (40.56) | 8178 (39.88) | 0.099 |
| HCV (%HCV Ab positive) | 2901 (4.44) | 1985 (4.44) | 916 (4.44) | 0.974 |
| HBV (%Hb core Ab positive) | 3867 (5.92) | 2664 (5.96) | 1203 (5.83) | 0.496 |
| CDC high risk (%) | 7396 (12.59) | 4998 (12.46) | 2398 (12.85) | 0.184 |
| Split (%) | 780 (1.19) | 590 (1.32) | 190 (0.92) | <0.001 |
| Biopsy (% yes) | 26768 (41.14) | 18451 (41.45) | 8317 (40.47) | 0.018 |
| Macrosteatosis (>30%) if biopsy | 3188 (14.15) | 2204 (14.20) | 984 (14.04) | 0.739 |
| ETOH dependency in donor (%) | 9463 (16.31) | 6508 (16.43) | 2955 (16.05) | 0.249 |
| Donor 0P0 volume, by tercile (%) | ||||
| Lowest quartile | 5910 (9.05) | 4055 (9.08) | 1855 (8.99) | 0.708 |
| Second quartile | 11816 (18.09) | 8102 (18.14) | 3714 (17.99) | 0.658 |
| Third quartile | 17501 (26.79) | 11941 (26.73) | 5560 (26.94) | 0.58 |
| Highest quartile | 30089 (46.07) | 20576 (46.06) | 9513 (46.09) | 0.948 |
| Transplant era | ||||
| 2003-2006 | 20185 (30.90) | 14005 (31.35) | 6180 (30.62) | <0.001 |
| 2007-2009 | 15045 (23.03) | 10263 (22.97) | 4782 (23.17) | 0.585 |
| 2010-2012 | 15567 (23.83) | 10612 (23.75) | 4955 (24.00) | 0.485 |
| 2013-2016 | 14519 (22.23) | 9794 (21.92) | 4725 (22.89) | 0.006 |
| Discarded livers (%) | 6454 (9.88) | 4302 (9.63) | 2152 (10.43) | 0.002 |
Descriptive statistics for adult, deceased brain dead donor, isolated liver offers: broken down by likely weekday vs weekend transplant for both procured and not procured livers. Weekday defined as likely day of transplant Monday to Friday (donor cross-clamp time from 12:00 AM Monday morning through 5:00 PM Friday afternoon. Weekend defined as donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM. CDC high risk = donation considered CDC high risk; ETOH dependency = whether the donor consumed >2 alcoholic drinks daily; donor 0PO volume = quartile of OPOs for the number of transplants per year averaged over the study period.
Weekday defined as likely day of transplant from Monday to Friday (donor cross-clamp time of 12:00 AM Monday morning through 5:00 PM Friday afternoon.
Weekend defined as donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM.
Ab, antibody; ALT, alanine transaminase; AST, aspartate transaminase; BMI, body mass index; CDC, Centers for Disease Controland Prevention; Hb, hemoglobin; HBV, hepatitis B vims; HCV, hepatitis C virus; ETOH, ethanol; OPO, organ procurement organization.
Univariable and Multivariable Analysis
Despite no major differences in risk factors (Table 3), livers procured for weekend transplants were less likely to be used for transplant compared with weekday livers (Table 2). Next, given that the decision to accept and transplant a liver is a complicated choice involving organ, donor, and systemic factors, we performed a univariable and multivariable analysis to assess how these variables impacted organ utilization over the course of a calendar week (see Tables 4–6 and Figure 2).
TABLE 4.
Odds of liver nonprocurement, discard, and nonuse by likely day of transplant for adult, deceased brain dead donor, isolated livers (2003-2016)
| Odds of not procuring an organ (n = 73686) | Odds of discard If procured (n =65316) | Odds of not using an organ (n = 73686) | ||||
|---|---|---|---|---|---|---|
| Likely day of transplant | Crude OR | P | Crude OR | P | Crude OR | P |
| Monday | Ref | – | Ref | – | Ref | – |
| Tuesday | 1.08 (0.96–1.20) | 0.189 | 1.07 (0.97–1,18) | 0.178 | 1.08 (0.99–1.16) | 0.06 |
| Wednesday | 1.07 (0.96–1.19) | 0,251 | 1,07 (0.97–1,18) | 0.161 | 1.07 (0.99–1.16) | 0.071 |
| Thursday | 1.11 (0.99–1.24) | 0.058 | 1,13 (1.02–1,24) | 0.015 | 1.13 (1.04–1.22) | 0.002 |
| Friday | 1.10 (0.99–1.23) | 0.081 | 1.09 (0.99–1,20) | 0.091 | 1.10 (1.02–1.19) | 0.015 |
| Weekend | 1.18 (1,07–1.30) | 0.001 | 1,20 (1.10–1.31) | <0.001 | 1.20 (1.12–1.29) | <0.001 |
Odds of discard, nonprocurement, and nonuse by day of the week a liver was procured. Monday as reference day of procurement. Likely day of transplant was defined in relation to the donor cross-clamp time during the procurement operation: each likely day of transplant is from donor cross-clamp time from 5:00 PM the day before until5:00 PM of that day. Weekend = donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM. Odds of nonuse defined as either nonprocurement or discard vs transplant (combining nonprocurement and discard into a single category).
OR, odds ratio; ref, reference.
TABLE 6.
Odds of liver nonuse by weekday vs weekend procurement: adult, deceased brain dead donor, isolated livers: univariable and multivariable Analyses (2003-2016)
| Univariable (n = 73686) | Multivariable (n = 69205) | |||
|---|---|---|---|---|
| Variable | Crude OR | P | Adjusted OR | P |
| Likely day of transplant (ref = Monday) | ||||
| Tuesday | 1.08 (0.99–1.16) | 0.06 | 1.10 (1.01–1.22) | 0.034 |
| Wednesday | 1.07 (0.99–1.16) | 0.071 | 1.05 (0.96–1.15) | 0.304 |
| Thursday | 1.13 (1.04–1.22) | 0.002 | 1.16 (1.06–1.28) | 0.001 |
| Friday | 1.10 (1.02–1.19) | 0.015 | 1.07 (0.97–1.17) | 0.179 |
| Weekend | 1.20 (1.12–1.29) | <0.001 | 1.22 (1.12–1.33) | <0.001 |
| Donor age >60 y | 1.50 (1.44–1.56) | <0.001 | 1.33 (1.26–1.41) | <0.001 |
| Donor gender (female) | 1.11 (1.07–1.16) | <0.001 | 0.98 (0.94–1.03) | 0.398 |
| Donor race (ref = Caucasian) | ||||
| African American | 0.72 (0.68–0.76) | <0.001 | 0.59 (0.55–0.63) | <0.001 |
| Other | 1.29 (1.23–1.35) | <0.001 | 1.30 (1.23–1.38) | <0.001 |
| Donor BMI >30 kg/m2 | 1.83 (1.76–1.90) | <0.001 | 1.62 (1.55–1.70) | <0.001 |
| Donor cause of death (ref = HT) | ||||
| Anoxia | 1.50 (1.43–1.58) | <0.001 | 1.13 (1.06–1.21) | <0.001 |
| Stroke | 1.67 (1.60–1.75) | <0.001 | 1.41 (1.33–1.50) | <0.001 |
| Other | 1.67 (1.48–1.86) | <0.001 | 1.54 (1.33–1.78) | <0.001 |
| Donor AST > 250 | 2.94 (2.78–3.11) | <0.001 | 3.35 (3.11–3.61) | <0.001 |
| Donor Total bilirubin >2 | 2.65 (2.52–2.80) | <0.001 | 2.81 (2.63–3.00) | <0.001 |
| Donor diabetes | 1.85 (1.77–1.94) | <0.001 | 1.35 (1.27–1.43) | <0.001 |
| Donor HTN | 1.73 (1.67–1.80) | <0.001 | 1.31 (1.24–1.38) | <0.001 |
| HCV+ | 2.94 (1.74–3.15) | <0.001 | 2.94 (2.69–3.20) | <0.001 |
| HBV core Ab+ | 1.67 (1.57–1.68) | <0.001 | 1.18 (1.08–1.28) | <0.001 |
| CDC high risk | 0.88 (0.83–0.94) | <0.001 | 0.75 (0.70–1.81) | <0.001 |
| Liver biopsy | 2.84 (2.72–2.96) | <0.001 | 2.11 (2.01–2.21) | <0.001 |
| Donor 0P0 volume | ||||
| Lowest quartile | Ref | – | Ref | – |
| 2nd quartile | 0.91 (0.85–0.98) | 0.013 | 0.91 (0.83–0.99) | 0.026 |
| 3rd quartile | 0.90 (0.84–0.96) | 0.003 | 0.92 (0.85–0.99) | 0.047 |
| Highest quartile | 1.08 (1.01–1.15) | 0.023 | 0.94 (0.87–1.02) | 0.116 |
Odds of nonuse by day of the week a liver was procured for likely weekday vs weekend transplant. Weekend defined as Friday 5:00 PM to Sunday 11:59 PM.Adjusted for donor age >60 y, sex, race, BMI >30 kg/m2, cause of death, donor AST >250 and donor bilirubin >2, comorbidity (DM, HTN), HCV seropositivity, HBV core Ab+, CDC high risk status, liver biopsy, and donor 0P0 volume. Donor 0P0 volume = quartile of OPOs for the number of transplants per year averaged over the study period. Donor ETOH abuse not included since 13.5% of data missing, allother variables <1 % missing values. Ab, antibody; AST, aspartate transaminase; BMI, body mass index; CDC, Centers for Disease Control and Prevention; DM, diabetes mellitus; ETOH, ethanol; HBV, hepatitis B virus; HCV, hepatitis C virus; HT, head trauma; HTN, hypertension; OPO, organ procurement organization; OR, odds ratio; ref, reference.
FIGURE 2.
The percentage and odds of nonuse of an adult, deceased brain dead donor, isolated liver procurement in the United States over the course of a calendar week (2003-2016). Likely day of transplant defined as cross-clamp time in the donor operation from 5:00 PM the day before to 5:00 PM of that day (eg, Tuesday = donor cross-clamp from 5:00 PM Monday afternoon to 5:00 PM Tuesday afternoon). Weekend = donor cross-clamp 5:00 PM on Friday afternoon to 11:59 PM Sunday night.
Tables 4 and 5 display the univariable model for the odds of liver nonprocurement, discard, and nonuse by likely day of transplant. Here we note that each of these odds was higher on the weekend compared with weekdays. Overall, compared with Monday the odds of nonuse was lowest on Tuesday (OR, 1.08 compared with Monday) and highest on the weekend (OR, 1.20), with the second highest rate of nonuse on Thursday (OR, 1.13). Table S3, SDC, http://links.lunv.com/TP/B657 shows that, compared with the weekend, the odds of nonuse were statistically significantly lower for each day of the week.
TABLE 5.
Odds of liver nonprocurement, discard, and nonuse by likely day of transplant for adult, deceased brain dead donor, isolated livers (2003-2016)
| Odds of not procuring an organ (n = 73686) | Odds of discard If procured (n = 65316) | Odds of not using an organ (n = 73686) | ||||
|---|---|---|---|---|---|---|
| Likely day of transplant | Crude OR | P | Crude OR | P | Crude OR | P |
| Weekday | Ref | – | Ref | – | Ref | – |
| Weekend | 1.09 (1,03–1.15) | 0.002 | 1 11 (1.06–1 17) | <0.001 | 1.11 (1.07–1.15) | <0.001 |
Odds of discard, nonprocurement, and nonuse by day of the week a liver was procured. Weekdays as reference days of procurement. Likely day of transplant was defined in relation to the donor cross-clamp time during the procurement operation: Weekday = My day of transplant from donor cross-clamp time from 12:00 AM Monday until 5:00 PM Friday. Weekend = likely day of transplant from donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM. Odds of nonuse defined as either nonprocurement or discard vs transplant (combining nonprocurement and discard into a single category).
OR, odds ratio; Ref, reference.
In the final model of liver nonuse (Table 6), we show that donor age, higher donor BMI, donor cause of death other than head trauma, donor comorbidities (hypertension and diabetes), elevated liver function tests, requiring a liver biopsy, and hepatitis B virus/HCV seropositivity were all associated with a higher likelihood of nonuse (all P-value < 0.05). This was unsurprising given the findings in Table 1. Nonetheless, in multivariable analysis the odds of a liver being not used on the weekend was 22% higher, even after adjustment for the above characteristics (OR, 1.22, P < 0.001; Table 6). See Figure 2 for a graphical display of the nonuse rate and the odds of liver nonuse by likely day of transplant. Unfortunately, since no recipient OPO or center was recorded when a liver was discarded (ie, this fact was only recorded when a liver was taken for transplant), we were unable to compare discard and nonuse rates for either recipient centers or OPOs. Table S4, SDC, http://links.ltvw.com/TP/B657 shows a similar analysis for DCD livers: the odds of nonprocurement and nonuse were significantly higher on the weekend compared with weekdays (adjusted OR, 1.32, P < 0.001, and OR, 1.20, P < 0.001, respectively). However, once procured, DCD livers were no more likely to be discarded on the weekend compared with weekdays (OR, 0.96, P = 0.578).
Liver Utilization and the Impact of Share 35 on the Weekend Effect
In June of 2013, via the Share 35 initiative, liver allocation underwent major changes in an effort to increase the availability of livers to the sickest recipients. Given this change to the allocation system, we examined whether the weekend effect was affected by the implementation of Share 35 (Tables 7–9). We found that after the implementation of Share 35 the overall liver nonuse rate decreased from 20.52% to 18.50%, and that while weekend liver nonprocurement, discard, and nonuse rates were still higher than their weekday counterparts, none reached statistical significance (odds of nonuse post-Share 35 OR, 1.08, P = 0.105; Table 9).
TABLE 7.
Pre- and post-Share 35 liver utilization by weekday vs weekend
| Pre-Share 35 | Post-Share 35 | |||||
|---|---|---|---|---|---|---|
| Weekdaya | Weekend | P | Weekday | Weekend | P | |
| Nonprocurement rate | 11.24 | 12.44 | <0.001 | 10.2 | 10.92 | 0.159 |
| Discard rate (if procured) | 9.8 | 10.66 | 0.003 | 9.04 | 9.65 | 0.231 |
| Nonuse rateb | 19.94 | 21.77 | <0.001 | 18.31 | 19.51 | 0.065 |
Pre- and post-Share 35 liver utilization. Pre-Share 35 = 2003–2012. Post-Share 35 = July 2003–2016.
Weekday = likely day of transplant from donor cross-clamp time of 12:00 AM Monday until 5:00 PM Friday. Weekend = likely day of transplant from donor cross-clamp time from Friday 5:00 PM through Sunday 11 :59 PM.
Nonuse defined as either nonprocurement or discard vs transplant (combining nonprocurement and discard into a single category).
TABLE 9.
Post-Share 35 odds of liver nonprocurement, discard, and nonuse by likely day of transplant for adult, deceased brain dead donor, isolated livers (July 2013-2016)
| Odds of not procuring an organ (n = 16210) | Odds of discard if procured (n = 14519) | Odds of not using an organ (n = 16210) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Crude | Adjusted | Crude | Adjusted | Crude | Adjusted | |||||||
| Likely day of transplant | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P |
| Weekday | Ref | – | Ref | – | Ref | – | Ref | – | Ref | – | Ref | – |
| Weekend | 1.08 (0.97–1.20) | 0.159 | 1.06 (0.94–1.19) | 0.355 | 1.08 (0.95–1.21) | 0.231 | 1.09 (0.96–1.24) | 0.176 | 1.08 (0.99–1.18) | 0.065 | 1.08 (0.98–1.19) | 0.105 |
Post-Share 35: Odds of discard, nonprocurement, and nonuse by day of the week a liver was procured. Weekdays as reference days of procurement. Likely day of transplant was defined in relation to the donor cross-clamp time during the procurement operation: weekday = likely day of transplant from donor cross-clamp time from 12:00 AM Monday until 5:00 PM Friday. Weekend = likely day of transplant from donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM. Odds of nonuse defined as either nonprocurement or discard vs transplant (combining nonprocurement and discard into a single category).
Cl, confidence interval; OR, odds ratio; ref, reference.
United Network of Organ Sharing (UNOS) Region Analysis
We next looked at the weekend effect as it related to UNOS region. Here, we found that during our study period of 2003–2016 the weekend nonuse rate was higher in all 11 UNOS regions, with weekend nonuse rates ranging from 4.46% 0.21% higher compared with weekday nonuse rates, depending on UNOS region (Figure 3). The average absolute nonuse rate was 2.00% higher on the weekend than for weekday livers across all UNOS regions. If the rates of liver nonuse were the same for weekday and weekend transplant across all regions, there would have been an additional 394 liver transplants over the course of the study period.
FIGURE 3.
UNOS region and the weekend effect. This figure shows how the weekend effect impacts liver utilization in all 11 UNOS regions (2003-2016).
Time of Day Analysis
We next analyzed whether there were hourly differences in organ utilization over the course of a day by examining nonuse rates by cross-clamp times in the donor operation (Figure 4A, and B). Figure 4A shows that nonprocurement, discard, and nonuse rates fall during daytime and rise during nighttime hours, whereas Figure 4B and Table S5, SDC, http://links.lww.com/TP/B657 show that this nighttime effect was higher on the weekend compared with weekdays.
FIGURE 4.
A, Liver utilization by time of cross-clamp in the donor operation for all adult, deceased brain dead donor, isolated liver procurement in the United States 2003-2016. Time listed by hour in military time. B, Weekday vs weekend liver nonuse by time of cross-clamp in the donor operation for all adult, deceased brain donor, isolated liver procurement in the United States 2003-2016. Time listed by hour in military time.
Figure SI, SDC, http://Unks.lww.com/TP/B657 shows a similar time of day analysis for DCD liver utilization.
Biopsy Results
Finally, we examined the biopsy reports with regard to liver discards and the reasons documented for discarding a given liver. Of the 6454 discarded livers, 99.1% of them had a reason listed for discard. There were many reasons cited for the discarding of a particular liver with the most common being findings on liver biopsy (53.49%). Other explanations for discard included prolonged cold ischemic time (1.67%), reported anatomical reasons (7.55%), vascular damage (1.44%), because of a “diseased organ” or “poor organ function” (11.89%), changes in the recipient status (2.44%), an inability to find a suitable recipient (1.58%), and another 15.71% for “other” reasons, among other factors.
Given “biopsy findings” was the most commonly listed reason for discard, we more closely examined liver biopsy reads with regard to discard. Discarded livers were more likely to be biopsied than transplanted livers (78.5% versus 37.1%, P < 0.001). Of the livers that were biopsied, discarded livers were more likely to have >30% macrosteatosis than transplanted livers (39.0% versus 8.2%, P < 0.001; Table 1). For discarded livers in which the reason for discard was cited as “biopsy findings,” 48.2% of livers had macrosteatosis rates <30%, and 74.5% had macrosteatosis rates <50%. With regard to the weekend effect, livers on the weekend had similar macrosteatosis rates compared with weekday livers (P = 0.126; Table 3).
DISCUSSION
Organ discard rates remain high despite a growing organ shortage.14–17 This analysis suggests that while non-procured and discarded livers frequently have unfavorable donor characteristics, there are also systemic and operational factors, including the time of day and day of the week a liver becomes available, that impact the chance of organ nonuse. Weekends are typically periods of staffing shortages, reduced surgeon coverage or cross-coverage, and have other indicators of both personnel and resource limitations. As such, a variety of medical and surgery-specific outcomes have been shown to be worse for patients treated on the weekend compared with weekdays.18–22 However transplant outcomes seem to be an exception to this finding, with 1 study showing that transplants occurring on nights and weekends had similar outcomes to those taking place during the week.23
In our study, we observed that livers available for likely weekend transplants were 11% more likely to be not used than their weekday counterparts, despite no apparent differences in organ quality. Although small, this effect size was consistent and remained even after adjusting for other variables relating to donor, organ, and other factors. Rates of liver biopsy, macrosteatosis on biopsy, HCV+ seropositivity, and donor risk factors were similar for weekend and weekday transplants. There was also a clear trend of an increasing nonuse rate beginning on Monday, when rates were lowest, that peaked on Saturday and Sunday, when rates were highest. Both weekday and weekend transplants had higher nonuse rates for livers at nighttime compared with daytime procurements. However weekend nights had higher discard rates than weekday nights (Figure 4B and Table S5, SDC, http://links.Iww.com/TP/B657; P = 0.005).
These results are similar to our recent analysis of discards in a cohort of kidney transplant procurements.2 This time variation in nonuse and discard trends, in addition to the wide variation in liver discards between UNOS regions, suggests that factors other than organ quality may be driving centers to not procure or discard recovered organs. This is supported by the fact that, according to SRTR, ECD usage for livers in 2012 ranged between 11% and 32%, a variability greater than for any other organ.1 This may also be related to the “aggressiveness” phenotype, in which Garonzik-Wang et al13 found that candidate disease severity, geographic differences in organ availability, and transplant volume were the main factors associated with the aggressive utilization of higher risk livers. However we were unable to examine this relationship since recipient and recipient center-level data were only recorded for transplanted and not for discarded or nonprocured livers.
Liver biopsies have been shown to be a surrogate for liver quality (ie, biopsied livers on average have higher donor risk indices),24 and it has been argued that livers with >60% macrosteatosis should not be utilized unless there is an urgent indication.25 However, grafts with macrosteatosis rates of <30% have been reported to perform similarly to grafts without macrosteatosis if their donor and recipient risk factors remain the same 25 To this point, although full liver biopsy reports were not available, we do know that 61% of livers that were discarded had <30% macrosteatosis when biopsied—biopsies were performed in ~78% of all discards—and that macrosteatosis rates were no different between weekday and weekend transplants/discards.
There are several reasons why nonuse rates may increase on the weekend. Given weekend resource limitations, centers may be less inclined to use “marginal organs” secondary to concerns about operating room, blood bank, nursing, interventional radiology, and intensive care unit staffing capabilities. Junior attending surgeons may also be more risk averse without senior backup/advice or supervision. There is also the potential that OPOs may be short-staffed on the weekends. Some centers also have dedicated liver and kidney surgeons while others must rely on a single surgeon responsible for dual liver/kidney coverage on the weekend. This may preclude these donation service areas and centers from using a liver they were originally interested in transplanting secondary to other service obligations, procurement responsibilities, emergencies with existing patients, and so on. Similarly, it is not surprising that weekend donor livers are less likely to be from DCD donors or used for splits. Both DCD and split liver procurements tend to be more labor intensive, generally require a more experienced surgeon and require tighter cold ischemic times. Given this relationship between possible marginal allografts and nonuse, it is possible that machine perfusion could help reduce marginal graft discards.
Limitations to our study include its retrospective nature and certain limitations inherent to the variables collected or not collected. SRTR only includes donors in whom ≥1 organ was procured, so we are lacking information on organ offers in which no organs were procured and it is unclear if these would vary by time of day or day of the week. We were unable to examine recipient OPO/center effects or the impact of individual recipient factors (eg, MELD score) secondary to how data were recorded (ie, recipient data were not collected unless an organ was actually transplanted). We were also not able to collect donor risk indices for discarded organs since cold/warm ischemic time and regional/national share type could not be determined for discarded livers. In addition, only 1 reason is recorded for discarding a given liver, but the decision to discard a liver is often likely multifactorial. It is also possible that OPOs have a different definition of what constitutes a discard, which would need to be standardized if true to facilitate further studies. Finally, our definition of the weekend still leaves ~11% of livers unclassified. Of note, livers that were transplanted over the weekend are procured over a different time frame from kidneys that are transplanted over the weekend due to differences in cold ischemia time, necessitating a difference in the definition from our previous publication on weekend discards of kidneys.2
In recent years, peri- and postoperative advancements in the care of transplant recipients have significantly reduced the variability of postliver transplant outcomes, making death on the waitlist the largest contributor to mortality of potential liver transplant patients.25,26 However, a number of disincentives related to measuring transplant outcomes (defined narrowly as 1-year graft and patient survival) may produce an aversion to risk, leading to withholding transplantation from the highest risk patients.27 Recently, Goldberg et al28 made a persuasive case to add organ offer acceptance behavior to the evaluation of transplant center performance. Adding discard/utilization trends and wait-list mortality to center or OPO performance metrics could further change practice leading to more equitable utilization of livers in the patients who need it most.
Given the stagnant donor pool, there is an urgent need to ensure that we are efficiently, ethically, and safely maximizing transplantation rates. Examining reasons why we discard nearly 15% of all organs, and 10% of livers, recovered could lead to dramatic increases in utilization rates. An analysis aimed at examining high OPO and recipient center weekend utilizers—as well as high marginal organ utilizers/low discard centers—and their outcomes could provide a best practice model as to how to safely expand the donor organ pool.
Supplementary Material
TABLE 8.
Pre-Share 35 odds of liver nonprocurement, discard, and nonuse by likely day of transplant for adult, deceased brain dead donor, isolated livers (2003-2012)
| Odds of not procuring an organ (n = 57476) | Odds of discard if procured (n = 50797) | Odds of not using an organ (n = 57476) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Crude | Adjusted | Crude | Adjusted | Crude | Adjusted | |||||||
| Likely day of transplant | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P | OR with 95% Cl | P |
| Weekday | Ref | – | Ref | – | Ref | – | Ref | – | Ref | – | Ref | – |
| Weekend | 1.12 (1.06–1.18) | <0.001 | 1.13 (1.06–1.22) | <0.001 | 1.10 (1.03–1.17) | <0.001 | 1.12 (1.04–1.20) | <0.001 | 1.12 (1.07–1.17) | <0.001 | 1.14 (1.08–1.20 | <0.001 |
Pre-Share 35: odds of discard, nonprocurement, and nonuse by day of the week a liver was procured. Weekdays as reference days of procurement. Likely day of transplant was defined in relation to the donor cross-clamp time during the procurement operation: weekday = likely day of transplant from donor cross-clamp time from 12:00 AM Monday until 5:00 PM Friday. Weekend = likely day of transplant from donor cross-clamp time from Friday 5:00 PM through Sunday 11:59 PM. Odds of nonuse defined as either nonprocurement or discard vs transplant (combining nonprocurement and discard into a single category).
Cl, confidence interval; OR, odds ratio; ref, reference.
ACKNOWLEDGMENTS
The data reported here have been supplied by the Minneapolis Medical Research Foundation as the contractor for the SRTR. The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy of or interpretation by the SRTR or the US Government.
This work was supported in part by Health Resources and Services Administration contract 234-2005-37011C. S.M. was supported by grant (R01-DK114893-01); S.M. and M.C.C. was supported by grant (U01-DK116066-01) and received funding from the American Society of Transplantation as well as the Laura and John Arnold Foundation. D.J.C. is supported by National Institutes of Health training grant (T32HL007854).
Footnotes
The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.
D.J.C. wrote and edited first and subsequent drafts of the paper as well as performed all statistical analyses. M.C.C. helped acquire data, helped with critical revision of drafts of the paper as well as helped perform and guide statistical analyses. E.C.V. helped guide study design and the analysis and interpretation of data, as well as performed critical revision of drafts of the paper. K.J.H. helped conceptualize portions of the study and selection of patients and covariates. He also contributed critical analysis to key facets of the paper. J.C.E performed critical revision of drafts of the paper and helped with the analysis and interpretation of data. L.E.R. conceived of the idea of the paper, performed critical revision of drafts of the paper, and helped with the analysis and interpretation of data. S.M. helped with study concept and design, oversaw all statistical analyses, and performed critical revision of drafts of the paper.
Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com).
The authors declare no conflicts of interest.
REFERENCES
- 1.Kim WR, Smith JM, Skeans MA, et al. OPTN/SRTR 2012 Annual Data Report: liver. Am J Transplant. 2014; 14(Suppl 1):69–96. [DOI] [PubMed] [Google Scholar]
- 2.Mohan S, Foley K, Chiles MC, et al. The weekend effect alters the procurement and discard rates of deceased donor kidneys in the United States. Kidney Int. 2016;90(1): 157–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gordon Burroughs S, Busuttil RW. Optimal utilization of extended hepatic grafts. Surg Today. 2009;39(9):746–751. [DOI] [PubMed] [Google Scholar]
- 4.Routh D, Naidu S, Sharma S, et al. Changing pattern of donor selection criteria in deceased donor liver transplant: a review of literature. J Clin Exp Hepatol. 2013;3(4):337–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.McCormack L, Quinonez E, Rios MM, et al. Rescue policy for discarded liver grafts: a single-centre experience of transplanting livers ‘that nobody wants’. HPB (Oxford). 2010;12(8):523–530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wolfe RA, LaPorte FB, Rodgers AM, et al. Developing organ offer and acceptance measures: when ‘good’ organs are turned down. Am J Transplant. 2007;7(5 Pt 2):1404–1411. [DOI] [PubMed] [Google Scholar]
- 7.Azancot MA, Moreso F, Salcedo M, et al. The reproducibility and predictive value on outcome of renal biopsies from expanded criteria donors. Kidney Int. 2014;85(5):1161–1168. [DOI] [PubMed] [Google Scholar]
- 8.Cadillo-Chávez R, Santiago-Delpín EA, Gonzalez-Caraballo Z, et al. The fate of organs refused locally and transplanted elsewhere. Transplant Proc. 2006;38(3):892–894. [DOI] [PubMed] [Google Scholar]
- 9.Feng S, Goodrich NP, Bragg-Gresham JL, et al. Characteristics associated with liver graft failure: the concept of a donor risk index. Am J Transplant. 2006;6(4):783–790. [DOI] [PubMed] [Google Scholar]
- 10.Loss J, Drewitz KP, Schlitt HJ, et al. Accept or refuse? Factors influencing the decision-making of transplant surgeons who are offered a pancreas: results of a qualitative study. BMC Surg. 2013;13:47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Tanriover B, Mohan S, Cohen DJ, et al. Kidneys at higher risk of discard: expanding the role of dual kidney transplantation. Am J Transplant. 2014; 14(2):404–415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wullstein C, Woeste G, de Vries E, et al. Acceptance criteria of pancreas grafts: how do surgeons decide in Europe? Transplant Proc. 2005;37(2):1259–1261. [DOI] [PubMed] [Google Scholar]
- 13.Garonzik-Wang JM, James NT, Arendonk KJV, et al. The aggressive phenotype revisited: utilization of higher-risk liver allografts. Am J Transplant. 2013;13(4):936–942. [DOI] [PubMed] [Google Scholar]
- 14.Klein AS, Messersmith EE, Ratner LE, et al. Organ donation and utilization in the united states, 1999–2008. Am J Transplant. 2010; 10(4 Pt 2):973–986. [DOI] [PubMed] [Google Scholar]
- 15.Ojo AO, Heinrichs D, Emond JC, et al. Organ donation and utilization in the USA. Am J Transplant. 2004;4(Suppl 9):27–37. [DOI] [PubMed] [Google Scholar]
- 16.Sung RS, Galloway J, Tuttle-Newhall JE, et al. Organ donation and utilization in the United States, 1997–2006. Am J Transplant. 2008;8(4 R 2):922–934. [DOI] [PubMed] [Google Scholar]
- 17.Tuttle-Newhall JE, Krishnan SM, Levy MF, et al. Organ donation and utilization in the united states: 1998–2007. Am J Transplant. 2009;9(4 Pt 2):879–893. [DOI] [PubMed] [Google Scholar]
- 18.Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med. 2001. ;345(9):663–668. [DOI] [PubMed] [Google Scholar]
- 19.Groves EM, Khoshchehreh M, Le C, et al. Effects of weekend admission on the outcomes and management of ruptured aortic aneurysms. J Vasc Surg. 2014;60(2):318–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kumar G, Deshmukh A, Sakhuja A, et al. ; Milwaukee Initiative in Critical Care Outcomes Research (MICCOR) Group of Investigators. Acute myocardial infarction: a national analysis of the weekend effect over time. J Am Coll Cardiol. 2015;65(2):217–218. [DOI] [PubMed] [Google Scholar]
- 21.Nanchal R, Kumar G, Taneja A, et al. ; from the Milwaukee Initiative in Critical Care Outcomes Research (MICCOR) Group of Investigators. Pulmonary embolism: the weekend effect. Chest. 2012; 142(3):690–696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pathak R, Karmacharya P, Aryal MR, et al. Weekend versus weekday mortality in myocardial infarction In the United States: data from healthcare cost and utilization project nationwide inpatient sample. Int J Cardiol. 2014;174(3):877–878. [DOI] [PubMed] [Google Scholar]
- 23.Orman ES, Hayashi PH, Dellon ES, et al. Impact of nighttime and weekend liver transplants on graft and patient outcomes. Liver Transpl. 2012;18(5):558–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Spitzer AL, Lao OB, Dick AA, et al. The biopsied donor liver: incorporating macrosteatosis into high-risk donor assessment. Liver Transpl. 2010;16(7):874–884. [DOI] [PubMed] [Google Scholar]
- 25.Busuttil RW, Tanaka K. The utility of marginal donors in liver transplantation. Liver Transpl. 2003;9(7):651–663. [DOI] [PubMed] [Google Scholar]
- 26.Emond JC. Measuring access to liver transplantation: an overdue metric for center quality and performance. J Hepatol. 2016;64(4):766–767. [DOI] [PubMed] [Google Scholar]
- 27.Axelrod DA. Balancing accountable care with risk aversion: transplantation as a model. Am J Transplant. 2013;13(1):7–8. [DOI] [PubMed] [Google Scholar]
- 28.Goldberg DS, French B, Abt PL, et al. Increasing the number of organ transplants in the united states by optimizing donor authorization rates. Am J Transplant. 2015;15(8):2117–2125. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.