Organ recipients have acquired a variety of diseases transmitted through the donated organ, occasionally with significant adverse outcomes.1-4 As such, systematic risk assessment in living and deceased donors has become a required practice of organ transplantation.5 Testing for human immunodeficiency virus (HIV), hepatitis C virus (HCV), hepatitis B virus (HBV), cytomegalovirus, Epstein-Barr virus, and syphilis is required by Organ Procurement and Transplantation Network (OPTN) policy (2.9.2). In 1994, the public health service (PHS) issued guidelines stipulating that specific donor behaviors, exposures or social situations were “high risk” for potential HIV transmission for an organ (and tissue) recipient.6 In 2008, OPTN policy mandated that potential organ recipients be informed of any organ offer from a PHS high risk donor (Policy 15.3). In 2013, the PHS guidelines7 were modified (dialysis within prior 12 months, new sexually transmitted disease within prior 12 months and time clarifications of incarceration and sexual activity) to identify organ donors with an increased risk for transmitting HIV, HBV, and HCV. OPTN policy 2.3.1 states that medical and behavioral histories must be obtained to identify conditions that may “affect the decision to use the donated organ.” This is salient, because up to 24% of potential recipients would reject an increased risk kidney.8
The impact of this information upon the system performance has not been provided. The purpose of this report is to (1) delineate the demographics of the PHS risk-identified donors and organs affected by the PHS identification, (2) discern the current utilization of organs from these donors contrasted to standard risk donors, and (3) transplant outcomes for the recipients of these organs.
MATERIALS AND METHODS
The OPTN database was queried for standard (no PHS designation) and PHS risk-identified deceased donors. Donor demographics, donor testing results (HCV+, hepatitis B surface antigen (HBsAg)+ and antibody reacting to hepatitis B core antigen [HBcAb]+), organ utilization, and transplant recipient outcomes were identified. Standard OPTN donor and recipient groupings by age, ethnicity, and cause of death were maintained. Kidney quality was calculated using the Kidney Donor Profile Index (KDPI).9
To assess the impact of the 2013 change in the PHS guidelines, the OPTN database was queried for calendar years 2012 (old PHS “high risk” donors) and 2014 (the first complete year of identification of PHS “increased risk” organ donors and near universal use of nucleic acid test (NAT) screening). The methodology used for attribution of disease transmission by Disease Transmission Advisory Committee (DTAC) (proven, probable, possible, intervention without documented transmission, and so on) has been previously described,10,11 and results were reviewed in cumulative, yearly format.
Statistical analyses were performed using SAS Enterprise Guide 5.1 (SAS Institute, Inc., Cary, NC). For categorical characteristics (eg, ethnicity, sex, cause of death), unless expected cell counts were small (≤5), a χ2 statistic was used to compare the distributions within the risk groupings. When expected cell counts were small, Fisher exact test was used. The same conventions were used to compare the distributions of HCV and HBV status and data on disease transmissions within risk groupings.
For continuous characteristics (eg, age, KDPI, organs transplanted per donor [OTPD]), a t-test assuming unequal variances was used to compare the distributions within risk groups. Survival rates were computed using the Kaplan-Meier method and were stratified by risk group. The outcome curves were compared for statistical differences using the log-rank test statistic.
Demographics of PHS-Identified (High/Increased) Risk Donors
PHS-identified risk donors are a substantial percentage of the deceased donor pool. From 2005 to 2010, 7.5% to 8.9% of US deceased donors were identified as “high risk” by the 1994 PHS criteria. By 2012, the percentage had risen to 11.9% (966/8143). In 2013, the change in the risk classification expanded the numbers of PHS-identified (increased risk) donors, and by the end of 2014, 20.6% (1772/8596) of all deceased organ donors were classified by PHS criteria as “increased risk.” Year-end 2015 data show that the number and percentage of PHS-identified risk donors has still grown to 22.2% (2015/9080) (Figure 1).
The PHS-identified donors on the whole are younger than the standard risk organ donor; approximately half were between ages 18 and 34 years and over 75% of all PHS-identified donors were between ages 18 and 49 years. Still, in 2012, 5.9% of organ donors 50 years and older had PHS-identified risk factors and that rose to 11.4% in 2014. Of all the young adult deceased donors (18-34 years), PHS-identified donors comprised 20.7% in 2012 and 34% in 2014 after the PHS revisions (Figure 2).7
The PHS-identified donors were more likely white, male, to die from anoxia, and less likely to have diabetes or hypertension as contrasted to the standard risk organ donor (Table 1). Fewer PHS-identified donors became donation after circulatory death donors (12.0% vs 14.8% of standard donors).
Organ Quality of PHS-Identified Donors
Overall donor “quality” is heterogeneous, organ-specific, and often difficult to measure. For a simplistic approach to donor comparison, the KDPI was calculated for the PHS and standard risk donors. In aggregate, the mean KDPI of the PHS-identified donor was lower than the standard risk donor (44.5 vs 54.9, P < 0.0001), consistent with the impact of donor age in the calculation (Figure 3). To minimize comorbidities and normalize age upon the calculation, 18- to 34-year-old deceased donors were analyzed. The impact of HCV upon KDPI is evident; 25.6 PHS risk versus 21.7 standard (P < 0.0001). By removing HCV+, from 18- to 34-year-old donors, the KDPI became nearly equivalent (21.8 PHS vs 21.3 standard; P = 0.3604) (Figure 3).
Testing for HBV and HCV
Within the PHS-identified group, a higher percentage of donors tested positive for HCV and HBcAb (HIV+ individuals were excluded from donation during this period and not in the OPTN database) contrasted to the standard risk deceased organ donor. In 2012, 15.9% of PHS-identified donors were HCV+ and 8% HBcAb+. These percentages were not demonstrably altered with the change in PHS guidelines. In 2014, the PHS-identified donors were 17.7% HCV+ and 7.6% HBcAb+. This contrasts to the relatively steady identification of HCV+ and HBcAb+ within the standard risk deceased donor of 2% to 3% and 4% to 5%, respectively. HBsAg+ individuals were infrequently used as organ donors. Only 0.1% of both standard risk (n = 11) and PHS-identified risk (n = 2) donors were HBsAg + in 2012 and 2014 with organs used for transplantation (Table 1).
Of 18- to 34-year-old PHS-identified risk deceased donors, 83 (17.1%) of 484 in 2012 and 147 (17.5%) of 842 in 2014 were HCV+, contrasted to only 27 (1.5%) of 1851 in 2012 and 22 (1.3%) of 1632 in 2014 of the standard risk, 18- to 34-year-old donors.
Organ Utilization From the PHS Risk-Identified Donor Contrasted to the Standard Risk Donor
Despite the lower KDPI of kidneys from PHS-identified donors, offers of accepted kidneys needed to be made to twice as many potential recipients to find a recipient for the PHS versus standard risk kidney. The median 2014 match run sequence for final kidney acceptance was 14.0 for PHS donor-identified and 7.0 for standard donor kidneys. Additionally, the recipients were slightly older for PHS versus standard donor kidneys. Of all other organ match runs, only with liver match runs was the median sequence number for the last acceptance equivalent between PHS and standard risk organs (5.0 for both).
Balanced against the recipient benefit from organ transplantation is donor risk assessment that the organ will cause harm to the recipient. To assess the impact of PHS designation as a solitary factor (and limit the impact of donor preexisting medical comorbidities) upon acceptance and transplantation of an organ, organ utilization from 18- to 34-year-old donors was analyzed. Although these standard and PHS-designated donors are mostly similar, there were a few notable differences: 18- to 34-year-old PHS-identified risk donors were more likely to die from anoxia than from head trauma, the PHS donor was more likely to be white and had 2% diabetes, whereas the standard donor group had 3% diabetes-identified (statistically significant, but clinically irrelevant) (Table 2).
The major determinant for organ acceptance and usage was not PHS risk status; rather a positive HCV laboratory test. In 2014, from standard risk, 18- to 34-year-old donors, OTPD significantly decreased from 4.20 to 1.91 (P < 0.0001) when the standard risk donor was found to be HCV+. This trend was also observed in the PHS risk-identified donor, although fewer organs were transplanted from the PHS donor, whether identified as HCV negative or positive, 3.90 HCV− versus 1.73 OTPD HCV+ (P < 0.0001). When HCV+ was eliminated as a variable in the 18- to 34-year-old donors, PHS risk identification had minimal impact upon liver utilization; however, fewer kidneys, hearts, lungs, and pancreata were transplanted from these young adult HCV− donors. The monthly use of organs transplanted/donor from the young adult HCV− standard versus PHS-identified donor demonstrated on average that 0.1 fewer of each organ type was transplanted from the HCV−, PHS increased risk donor resulting in 0.3 to 0.4 fewer organs transplanted (Figure 4).
From HCV+ donors, kidneys and livers were essentially the only organs transplanted, either from standard or PHS-identified donors. Fewer kidneys were transplanted from the donor identified by PHS risk criteria (0.85 vs 1.18 kidneys transplanted for PHS vs standard). Other than kidneys and livers, only a single heart was transplanted from an HCV+, 18- to 34-year-old organ donor (PHS risk-identified) in 2012 and 2014.
HBcAb+ as a solitary variable did not seem to alter utilization of organs from the 18- to 34-year-old donor, although the number of donors with only HBcAb+ was quite small.
It was striking that organs from donors with PHS-identified risk factors were infrequently used for pediatric recipients, irrespective of the donor HCV status. In 2014, after the new PHS guidelines were active, children received 11.6% of kidneys transplanted from standard risk 18- to 34-year-old HCV− donors, yet only 3.5% of the kidneys from PHS-identified HCV− donors. Pediatric recipients of more size-dependent organs, such as liver and heart, had nonstatistically reduced utilization of young adult organs from HCV−, PHS-identified donors (Figure 5). The new kidney allocation system, which favors children, was not implemented until December 2014, so its impact is not reflected in these numbers.
Outcomes of Transplant Recipients With Organs From PHS-Identified Risk and Standard Risk Donor
Patient and Graft Survivals
Transplant outcomes with organs from standard versus PHS-identified risk donors follows donor age and HCV status. One-year patient and graft survivals after deceased donor kidney and liver transplantation from PHS-identified risk donors are higher than that observed after organs from the standard risk donor, emphasizing the impact of donor age.
By limiting analysis to those recipients of organs from 18- to 34-year-old HCV− donors (2012-2014; before and after the PHS definition change), PHS-identified and standard risk donor outcomes were equivalent (Table 3). However, as follow-up was short, outcomes from 2005 to 2010 recipients of 18- to 34-year-old HCV− standard or PHS high risk organs were assessed. Although the donor population differed over the period and risk criteria varied, the 1, 3 and 5-year patient and graft survival of organ recipients were equivalent for all organ recipients, irrespective of PHS (high risk) status (Figure 6).
To assess the impact of change in PHS guidelines, the OPTN DTAC reports were reviewed for transplants from deceased donors recovered in 2012 and 2014. In 2012, 2.23% (160/8143) of standard and 3.31% (32/966) of PHS high-risk donors were reported to DTAC for concern of a potential disease transmission event. Fifteen (9.4%) of 160 standard and 8 (25%) of 32 PHS high-risk donor reports were for possible HIV, HBV, and HCV. The 2012 DTAC attribution was 4 proven/probable/possible HIV, HBV, or HCV transmissions from standard risk and 2 from PHS high risk donors. For total proven/probable/possible other transmissions (fungus, bacteria, and so on), 25% (41/161 referrals) from standard and 27% (9/33) PHS donors were proven/probable/possible transmissions. In 2014, after the PHS modifications, 3.2% (213/6602) standard and 2.9% (52/1772) PHS increased risk donations were referred to DTAC. Eleven (5.2%) of 213 standard and 12 (21.6%) of 51 PHS-identified were concerns for HIV, HBV, or HCV. In 2014, DTAC attributed 1 proven/probable/possible transmission from standard and 1 from PHS-identified donor referrals. Total unexpected proven/probable other disease transmissions were 9.8% (21/213) standard and 13.7% (7/51) PHS-identified referrals. The DTAC attribution of unexpected disease transmission in the 8 cases of HCV/HBV (no HIV); 5 from standard risk donors (of 13 981 donors and a 0.036%/transplant risk) and 3 from PHS risk-identified donors (of 2735 donors for a 0.10%/transplant risk), without statistical difference (P = 0.13). Although the transmission numbers are very small, the attribution criteria were made as loose as possible to favor transmission identification.
From 2009 to 2015, from 58 162 deceased organ donations (50 066 standard and 8056 PHS risk-identified), 140 cases were reviewed by DTAC (88 standard and 51 PHS risk, 1 unknown) over concerns of unexpected HIV, HBV, or HCV transmissions. Twenty-five donations were attributed by DTAC to have HIV, HBV, or HCV transmission, 18 from standard risk and 7 from PHS-identified donors, a statistically insignificant difference.
Additionally, OPTN center attributions for patient or graft loss from 2005 to 2010 did not demonstrate a difference in infectious, liver disease, or other causes for patient or graft loss for recipients of standard or PHS-identified risk organs. However, because over 25% of the causes of death were “unknown” and 25% of reported causes for the loss of kidney grafts were “unknown,” the query results are limited.
Organ use for transplantation balances recipient need and organ quality (benefit) against potential unacceptable surgical risk, insufficient organ function, and disease transmission (risk). The OPTN database does not capture how these decisions are made. The analysis in this article oversimplifies many of these issues by homogenizing a heterogeneous donor landscape,12 organ availability, and recipient need.13,14 The PHS guidelines for HIV “high risk” in 1994 and the “increased risk” 2013 modifications were intended to identify donors which were more likely to transmit HIV (and then HBV/HCV), thereby allowing providers and organ candidates unwilling to take such a risk the opportunity to defer. However, identifying a donor as PHS high/increased risk may alter decision making about organ acceptance without sufficient information about “degree” of risk.15 Much of the literature about organ donor testing for HIV, HBV, and HCV infection and associated recipient risk has been generated through mathematical calculations using assumptions about prevalence and incidence of new infection derived from nonorgan donor populations and the time sensitivities of detection methods.16-18 This report compiles the US experience, as reported through the OPTN and its DTAC and assesses the impact of the PHS criteria upon the utilization and outcomes of donated organs.
There are limitations of the OPTN database. The specific PHS risk factors used to identify the “high/increased” risk donor was not searchable. Patient and provider perceptions are likely influenced by mode of donor death and it is possible that some of the perceived “really” high risk donors (death with a needle in the arm) were outright declined and with no recognized transmissions. The number of potential PHS donors is not available in the database, only those with at least 1 organ transplanted. The data field for HCV during the analysis specified only serologic (anti-HCV) identification. Before 2013, HCV NAT was used inconsistently for donor screening, but after the 2013 PHS revisions included universal donor screening with HCV NAT, most US organ donors were screened with NAT and anti-HCV (antibody). The impact of NAT upon organ acceptance is unknown during this period as only HCV serologic data was captured. A recent CDC report of 3 instances of anti-HCV and NAT negative donors resulting in HCV transmissions highlight methodologic issues for disease detection, attribution, and capture of a disease transmission event.19 First, no detection process is foolproof, and disease transmission occurs despite sensitive tests. However, the attribution information used by the CDC and DTAC differs from what is available to the clinician/patient at the time of decision. Behaviors and incidents relevant for disease transmission determination are not always clear when an organ is offered.4
This review demonstrates that: (1) PHS-identified donors are numerous (20.6% of all deceased donors in 2014) and disproportionately young. In 2014, 34% of all young adult (18-34 years) deceased donors in the United States were identified as PHS increased risk for HIV, HBV, and HCV. In other words, in the absence of disease transmission risk, these donors are a significant proportion of the highest potential functioning organs. (2) The revised 2013 PHS classification identifies a substantial percentage of deceased donors that test positive for HCV; 72% of all 2014 HCV+ deceased donors were identified by the 2013 PHS criteria. However, specificity is low; 82% of 2014 PHS risk-identified donors were HCV−. To a lesser (but statistically significant) extent, the revised PHS criteria identify donors with an increased prior exposure to HBV, HBcAb+ (7.6% vs 4.2% standard risk). The PHS criteria (incarceration, sexual activity, drug usage, and so on) associated with the increased prevalence of HCV or HBV is not available within the OPTN database. (3) Pathogen testing impacts organ utilization decision making/triage. Less than 0.1% transplanted organs came from an HBsAg+ donor. Donor HCV detection was a major determinant of organ utilization. In the absence of HCV detection, over 4 organs were transplanted from a standard risk, 18- to 34-year-old donor. However, when the standard risk 18- to 34-year-old donor was HCV+, the numbers of organs transplanted decreased to less than 2. The PHS designation had a lesser, but additive effect upon organ use; organs transplanted decreased 0.3 to 0.4 organs/18- to 34-year-old HCV− donor. From the HCV−, PHS risk young adult donors, fewer organs were transplanted and when accepted, more organ offers were made before successful placement. The latter was expected as the intent of PHS identification is intended to inform and identify only those recipients willing to take the increased risk of acquiring HIV, HBV and HCV through organ transplantation. (4) The raw patient and graft survival after transplantation of organs from PHS-identified donors equaled or surpassed the outcomes of transplants from standard risk donors. However, this only confirms that younger donor organs function better than older organs. Patient and graft outcomes after age-matched organ transplantation (18- to 34-year-old donors) varies by donor HCV detection, but not PHS risk identification. (5) The experience of OPTN/DTAC, even when using the loosest positive attribution criteria, revealed that clinical recognition of unanticipated HIV, HBV, HCV disease transmission, although not zero, was not statistically different for organ recipients from PHS and standard risk donors. After the 2013 PHS guideline modification, greater than 20% of deceased donors were PHS risk (from 11.7% in 2012), but without demonstrable change in transmitted HIV, HBV, or HCV. Given the higher prevalence of HCV within the PHS population, it was assumed that a higher frequency of HCV transmission would occur from undetectable infection (and also HIV and HBV) from PHS risk organs. Although this is still probably true, the numbers of organ donors and methods of attribution have failed to demonstrate a statistical difference from more than 58 000 donors.
This study was designed to assess the benefit and impact of PHS (increased and high) risk identification upon organ use and outcome. Donor (organ) age is an independent element determining organ use. In 2014, 1.74 kidneys/donor were transplanted from 18- to 34-year-old donors, dropping to 1.52 in the 35- to 49-year-old donors and 1.23 in donors aged 50 to 65 years with increasing comorbidities. Comorbidity matching of recipients and standard and PHS risk organ donors is impractical with the existing database. Additionally, the c-statistic for kidney outcome prediction is quite low (0.6582).20 Because HCV detection significantly diminishes organ use, it was elected to contrast transplant use and outcomes from donors with high quality organs, but minimal preexisting medical comorbidities and low objective risk for disease transmission; the HCV−, 18- to 34-year-old deceased donor. Although the demographics of 18- to 34-year PHS and standard donors differ, a similar KDPI and equivalent transplant outcomes corroborated their use as comparators.
Decision making uses objective and subjective elements. When the recipient, family, or physician is told that an organ is from an “increased risk donor,” how does this information impact decision making?21 Experience demonstrates that the PHS label decreases organ utilization, but less so than objective pathogen testing. Organs from PHS donors are infrequently transplanted into the pediatric population and are more commonly transplanted into older adults with more comorbidity. This becomes problematic because the recipients of these organs do not appear to have a statistical increased risk of disease acquisition. In 2014, there were 2305 18- to 34-year-old, HCV− deceased donors (695 PHS increased risk/1610 standard risk). 3.90 organs per donor were transplanted from the “increased” risk donor, 0.3 fewer OTPD than the 4.20 OTPD from the standard risk group. Had PHS donor organs been utilized at the same rate as standard donors, an additional 209 of the highest functioning organ transplants would have occurred. An additional 205 transplants would have been possible from other donor age groups (22 in 0-17 years, 132 in 35-49 years, and 51 in 50 years or older). The attribution of the PHS risk is associated with a decreased utilization of organs, with no statistically discernible benefit of identifying organs with an increased frequency of disease transmission.
The “precautionary principle”22 cautions against aggressive attempts to reduce risk, as it can result in system dysfunction and undesired consequences. It is intuitive that disease transmission risk is inconsistent for organs from different donors. The PHS “high” or “Increased risk” donor designation for HIV, HBV, and HCV implies that there are real differences in disease transmission events, when contrasted to organs from a standard donor. The current data suggest that over 200 people in 2014 alone were not transplanted with the “best” quality organs from young, HCV− deceased donors, at least in part due to PHS donor designation. Labels and categories have significant and sometimes unintended consequences. If there is an enduring need to stratify donors for risk of HIV, HBV, and HCV transmission, it will be important to define the degree of increase required to meet the threshold for recipient notification and then refine/delete categories that fail to identify at that level. Otherwise, the continued identification of these donors appears to put potential recipients in harm’s way through avoidance of high-quality organs without a demonstrated increased risk of disease transmission.
1. Miceli MH, Gonulalan M, Perri MB, et al. Transmission of infection to liver transplant recipients from donors with infective endocarditis: lessons learned. Transpl Infect Dis
2. Centers for Disease C, Prevention. Balamuthia mandrillaris transmitted through organ transplantation—Mississippi, 2009. MMWR Morb Mortal Wkly Rep
3. Basavaraju SV, Kuehnert MJ, Zaki SR, et al. Encephalitis caused by pathogens transmitted through organ transplants, United States, 2002-2013. Emerg Infect Dis
4. Vora NM, Basavaraju SV, Feldman KA, et al. Raccoon rabies virus variant transmission through solid organ transplantation. JAMA
5. Health Resources and Services Administration. Policies. Organ Procurement & Transplantation Network website. https://optn.transplant.hrsa.gov/governance/policies
. Accessed 2016.
6. Rogers MF, Simonds RJ, Lawton KE, et al. Guidelines for preventing transmission of human immunodeficiency virus through transplantation of human tissue and organs. Centers for Disease Control and Prevention. MMWR Recomm Rep
7. Seem DL, Lee I, Umscheid CA, et al. PHS guideline for reducing human immunodeficiency virus, hepatitis B virus, and hepatitis C virus transmission through organ transplantation. Public Health Rep
8. Reese PP, Tehrani T, Lim MA, et al. Determinants of the decision to accept a kidney from a donor at increased risk for blood-borne viral infection. Clin J Am Soc Nephrol
9. Kucirka LM, Bowring MG, Massie AB, et al. Landscape of deceased donors labeled increased risk for disease transmission under new guidelines. Am J Transplant
10. Ison MG, Hager J, Blumberg E, et al. Donor-derived disease transmission events in the United States: data reviewed by the OPTN/UNOS Disease Transmission Advisory Committee. Am J Transplant
11. Ison MG, Nalesnik MA. An update on donor-derived disease transmission in organ transplantation. Am J Transplant
12. Saunders MR, Lee H, Alexander GC, et al. Racial disparities in reaching the renal transplant waitlist: is geography as important as race? Clin Transplant
13. Axelrod DA, Vagefi PA, Roberts JP. The evolution of organ allocation for liver transplantation: tackling geographic disparity through broader sharing. Ann Surg
14. Kumar D, Humar A, Kim SJ, et al. A survey of increased infectious risk donor utilization in Canadian transplant programs. Transplantation
15. Ros RL, Kucirka LM, Govindan P, et al. Patient attitudes toward CDC high infectious risk donor kidney transplantation: inferences from focus groups. Clin Transplant
16. Ellingson K, Seem D, Nowicki M, et al. Estimated risk of human immunodeficiency virus and hepatitis C virus infection among potential organ donors from 17 organ procurement organizations in the United States. Am J Transplant
17. Lai JC, Kahn JG, Tavakol M, et al. Reducing infection transmission in solid organ transplantation through donor nucleic acid testing: a cost-effectiveness analysis. Am J Transplant
18. Humar A, Morris M, Blumberg E, et al. Nucleic acid testing (NAT) of organ donors: is the ‘best’ test the right test? A consensus conference report. Am J Transplant
19. Suryaprasad A, Basavaraju SV, Hocevar SN, et al. Transmission of hepatitis C virus from organ donors despite nucleic acid test screening. Am J Transplant
20. SRTR. Risk Adjustment Model Documentation: Waiting List and Post-Transplant Outcomes [C Statistic]. http://www.srtr.org
21. Chow EK, Massie AB, Muzaale AD, et al. Identifying appropriate recipients for CDC infectious risk donor kidneys. Am J Transplant
22. Kirkland M. The precautionary principle: a double edged sword? Cell Tissue Bank