SEARCH

SEARCH BY CITATION

Keywords:

  • Hepatitis;
  • high-risk donors;
  • HIV;
  • NAT testing

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

The use of Public Health Service/Centers for Disease Control and Prevention (PHS/CDC) high-risk donor (HRD) organs remains controversial, especially in light of a recent high-profile case of human immunodeficiency virus (HIV) and hepatitis C virus (HCV) transmission. Nucleic acid testing (NAT), while more expensive and time consuming, reduces infectious risk by shortening the period between infection and detectability. The purpose of this study was to characterize HRDs and disposition of their organs by organ procurement organization (OPO), to measure NAT practices by OPO and to examine associations between NAT practices and use of HRD organs. We analyzed 29 950 deceased donors (2574 HRDs) reported to UNOS since July 1, 2004 and May 8, 2008. We then surveyed all OPO clinical directors about their use of NAT, average time to receive NAT results, locations where NAT is performed and percentage of the time NAT results are available for allocation decisions. In total, 51.7% of OPOs always perform HIV NAT, while 24.1% never do. A similar pattern is seen for HCV NAT performance, while the majority (65.6%) never perform HBV NAT. AIDS prevalence in an OPO service area is not associated with NAT practice. OPOs that perform HIV NAT are less likely to export organs outside of their region. The wide variation of current practice and the possibility that NAT would improve organ utilization support consideration for a national policy.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

The organ supply is limited, with a median waiting time of 3.2 years for a kidney transplant, 1.6 years for a liver, 1 year for a pancreas, 7 months for a heart and 2.5 years for a lung, according to the United Network for Organ Sharing/Organ Procurement and Transplantation Network (UNOS/OPTN) 2007 Annual Report (1). As such, efforts have been made to expand the pool of available organs without adversely affecting recipient outcomes. One solution that remains controversial is the use of donors classified by the Public Health Service (PHS) and published by the Centers for Disease Control and Prevention (CDC) as high risk for infectious transmission of HIV (2). The PHS/CDC has defined an adult high-risk donor (HRD) as someone falling into one of the seven behavioral categories or three laboratory categories outlined in Table 1. According to PHS guidelines, organs from HRDs should be discarded unless the risk to the recipient from not performing the transplant outweighs the risk of possible HIV infection (2,3).

Table 1.  PHS/CDC guidelines for high-risk donor behavior and laboratory/other criteria
Behavior:
 1. Men who have had sex with another man in the preceding 5 years
 2. Persons who report nonmedical intravenous, intramuscular or subcutaneous injection of drugs in the preceding 5 years
 3. Persons with hemophilia or related clotting disorders who have received human-derived clotting factor concentrates
 4. Men and women who have engaged in sex in exchange for drugs or money in the preceding 5 years
 5. Persons who have had sex in the preceding 12 months with any person described in items 1–4 above or with a person known or suspected to have HIV infection
 6. Persons who have been exposed in the preceding 12 months to known or suspected HIV-infected blood through percutaneous inoculation or through contact with an open wound, nonintact skin or mucous membrane.
 7. Inmates of correctional systems
Laboratory/other:
 1. Persons who cannot be tested for HIV infection because of refusal, inadequate blood samples (e.g. hemodilution that could result in false negative tests) or any other reasons.
 2. Persons with a repeatedly reactive screening assay for HIV-1 or HIV-2 antibody regardless of the results of supplemental assays.
 3. Persons whose history, physical examination, medical records or autopsy reports reveal other evidence of HIV infection or high-risk behavior, such as a diagnosis of AIDS, unexplained weight loss, night sweats, blue or purple spots on the skin or mucous membranes typical of Kaposi's sarcoma, unexplained lymphadenopathy lasting >1 month, unexplained temperature >100.5°F (38.6°C) for >10 days, unexplained persistent cough and shortness of breath, opportunistic infections, unexplained persistent diarrhea, male-to-male sexual contact, sexually transmitted diseases or needle tracks or other signs of parenteral drug abuse.

The biggest disincentive to using these organs is the potential to transmit certain infectious diseases, most commonly human immunodeficiency virus (HIV), hepatitis C virus (HCV) and hepatitis B virus (HBV) (4–9). A recent high-profile case highlights this disincentive, as four transplant patients contracted HIV and HCV from a donor who was antibody negative for both diseases in the Fall of 2007. Upon discovery of the infections, the donor was retested for both diseases using a nucleic acid test (NAT), and the results were positive by this method (8). All potential donors (those from HRDs and non-HRDs) are tested for the presence of these diseases by enzyme-linked-immunosorbent assay (ELISA) as required by UNOS and detailed in the minimum procurement standards for an organ procurement organization (OPO) (10). Although ELISA is highly sensitive for these diseases, it will not detect recent infections because it depends on antibody formation that occurs days to months following infection (11). The time between infection and serologic detectability is called the window period (WP). Using an ELISA, HIV has a mean WP of approximately 22 days (95% CI 6–38), HCV approximately 66 days (95% CI 38–94) and HBV 44 days (no confidence interval reported) (12).

NAT is a method of detection that can reduce the WP significantly (12–17). The WP is approximately 9 days for HIV NAT (13 days shorter than ELISA) and 22 days for HBV NAT (22 days shorter than ELISA) (12). The WP for HCV NAT is 7.4 days, a full 59 days shorter than the average WP for ELISA. Unfortunately, NAT is expensive and takes longer to perform compared to ELISA. False positive results can occur (18), and it can be argued that the mitigation of infectious risk by NAT might be outweighed by the discarding of viable organs due to a false positive NAT. In the United States, the decision of whether or not to perform NAT for potential donors is currently made independently by each OPO. For example, in the November 2007 case, NAT was not performed nor was it the standard within the OPO (8). The goals of this study were to characterize the HRD population, to examine disposition of HRD organs by OPO, to measure NAT practices by OPO and to determine if associations exist between NAT practices and disposition of HRD organs.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

HRD characteristics

Our study population included 29 950 deceased donors (2574 HRDs and 27 376 non-HRDs) from whom at least one organ was recovered between July 1, 2004 (when the HRD designation began to be captured in UNOS) and May 8, 2008, as reported to the United Network for Organ Sharing (UNOS). Donors were categorized by their OPOs as either: PHS/CDC high risk (Table 1) or non-PHS/CDC high risk. The classification was taken from the UNOS deceased donor registration form question reading, ‘Does the donor meet CDC guidelines for “High Risk” for an organ donor?’ Because the guidelines were written by the PHS, we refer to donors as ‘PHS/CDC high risk’ for the sake of clarity. PHS categorization was available only as a dichotomous variable, as the social history contributing to this designation was not captured in the UNOS donor data set. Characteristics between the two groups were compared using t-tests for continuous variables and chi-squared tests for categorical variables.

OPO use of HRDs

The proportion of OPO volume from HRDs was calculated by dividing the total number of HRDs in an OPO by the total number of donors for that OPO during the study period. For all results, only donors where at least one organ was recovered are included. The percentage of kidneys, livers, pancreata, hearts and lungs transplanted from HRDs was calculated for each organ as follows: total number of HRDs where at least one organ of a given type was transplanted/ total number of HRDs where at least one organ of any type was recovered.

Survey of OPO NAT practices

Between January 17 and March 30, 2008, we conducted a written survey assessing current OPO NAT practices for HIV, HCV and HBV; the survey was administered at a meeting of OPO clinical directors and supplemented electronically for those not attending the meeting, with a response rate of 100%. All OPO directors in the United States were surveyed and responded. We asked the OPO clinical directors to categorize their NAT practice within each disease category (HIV, HCV and HBV) as follows:

  • 1
    Do you perform NAT: (1) always, (2) for donors classified as ‘CDC high risk’, (3) only for certain categories of ‘CDC high risk’ behaviors, but not for others, (4) when requested by a transplant provider or (5) never.
  • 2
    If 1 is not ‘Never’: Choose the location where NAT is performed from the following categories: (1) In-house, (2) one of the transplant center laboratories, (3) an outside laboratory in the same city, (4) an outside laboratory in the same state or (5) an outside laboratory in a different state.
  • 3
    If 1 is not ‘Never’: Estimate the time between ordering NAT and receiving results, categorized as follows: (1) 2–4 h, (2) 4–8 h, (3) 8–12 h, (4) 12–24 h or (5) > 24 h.
  • 4
    If 1 is not ‘Never’: Choose the category describing the percentage of the time NAT results are available before a transplant decision is made: (1) always, (2) 75–99%, (3) 50–75%, (4) 25–50%, (5) 10–25%, (6) <10% or (7) never.

Geographic patterns in NAT practice by OPO

Maps were generated to depict geographic patterns in OPO performance of NAT for HIV, HCV and HBV. NAT performance was categorized as follows (1) Always; (2) Sometimes (if NAT was performed for all HRDs, for some categories of HRDs, or only when requested); and (3) Never. Each category was depicted as a separate color on the map, and the map was divided by counties, with each county's categorization corresponding to the NAT practices of the OPO servicing it. The maps were produced using ARC GIS version 9.1 (ESRI, Redlands, CA).

OPO-level demographics by HIV NAT practice

We dichotomized performance of HIV NAT as performing HIV NAT always as compared to performing HIV NAT sometimes or never. Characteristics of OPOs stratified by this HIV NAT categorization were compared using t-tests. All t-tests were two-sided.

HIV NAT practice and OPO disposition of organs from HRD

Generalized linear models were built to estimate the associations between local use of organs (defined as using the organ locally or regionally) and national export of organs (defined as exporting the organ outside of the region) with HIV NAT performance, using methods previously described (19). Local use referred to use within the OPO service area, regional to use outside the OPO but within the UNOS region and national to use outside the UNOS region (20). From these models, relative risks (RR) are interpreted as the relative association of HIV NAT use and each increasing quintile of a covariate. Furthermore, the influence that each OPO contributed to these models was weighted by the volume of that OPO, so that overall inferences represented national estimates. Weights were assigned based on organ type (so that for kidney analysis, kidney volume was used) and donor type (so that for HRD analyses, HRD volume was used), and normalized so as to avoid false reductions of the standard errors. All analyses were performed using STATA 10/MP for Linux (StataCorp, College Station, TX). AIDS prevalence data for each state were obtained from the CDC's 2006 HIV/AIDS surveillance report (21). To calculate an estimate of the number of AIDS cases in an OPO service area, we multiplied the proportion of counties in each state served by the OPO by the number of AIDS cases in the respective state, then summed the contribution of each state serviced by the OPO to estimate number of cases in each OPO. Prevalence was determined by dividing the number of AIDS cases by the total population of the OPO service area. We recognize that this method is limited because it ignores within-state differences in AIDS prevalence that likely exist; unfortunately, county-level AIDS data were unavailable for many states.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

HRD characteristics

Of deceased donors where at least one organ was recovered, HRDs were on average younger than non-HRDs (mean age 37.5 and 40.8 years, respectively) and were most likely to fall into the 18–30 age group (Table 2). HRDs were predominately males (71.0% vs. 57.9%), and were more likely to be African Americans as compared to non-HRDs (20.4% vs. 14.6% by chi-squared test of independence for race). The most common cause of death among non-HRDs was cerebrovascular accident or stroke (43.3%), whereas head trauma was the most common cause of death among HRDs (40.8%). HRDs were less likely to meet the criteria for expanded criteria kidney donor (22) (12.0% vs. 25.6%). HRDs and non-HRDs had comparable creatinine levels, and a similar proportion had a history of myocardial infarction (MI). HRDs had significantly higher proportions of exposure to HCV (anti-HCV+), and HBV (HBcAb+), but had similar proportions of HBsAg positivity compared with non-HRDs (0.3% vs. 0.2%, respectively).

Table 2.  Donor characteristics, by PHS/CDC high-risk donor (HRD) status
CharacteristicHRDs1 (n = 2574)Non-HRDs (n = 27 376) p-Value2
  1. 1PHS high-risk donor, as defined in the UNOS deceased donor registration form.

  2. 2The p-values were obtained using two-sided chi-squared tests (categorical variables), and two-sided t-tests (continuous variables).

  3. 3Expanded criteria donor, defined as (1) a donor > 60 years old, or (2) a donor > 50 years old with any two of the following characteristics: hypertension, serum creatinine level >1.5, or cause of death cerebrovascular accident.

Age (mean)37.540.8<0.001
Age category (%)
 <183.212.8<0.001
 18–3034.020.2 
 31–4019.611.9 
 41–5025.119.8 
 51–6013.619.6 
 >604.515.7 
Male (%)71.057.9<0.001
BMI (mean)26.330.10.7
Ethnicity (%)
 White64.968.3<0.001
 African20.414.6 
   American   
 Hispanic12.513.8 
 Other2.33.4 
Cause of death (%)
 Anoxia24.016.3<0.001
 CVA/stroke32.443.3 
 Head trauma40.837.3 
 CNS tumor0.40.7 
 Other2.42.4 
ECD (kidney12.025.6<0.001
 definition, %)3   
Serum creatinine1.51.4<0.001
   (mean)   
Serum Cr category (%)
 <0.51.75.2<0.001
 0.5–1.045.146.1 
 1.0–1.528.926.8 
 1.5–29.19.3 
 >215.212.5 
BUN (mean)18.818.10.04
Comorbidities (%)
 History of2.73.80.006
    cancer   
 History of MI4.44.60.8
 Hypertension27.533.9<0.001
Urine infection (%)7.37.30.9
Pulmonary25.322.5<0.001
  infection (%)   
Viral serologies (%)
 HIV0.00.00.5
 HBcAb+14.64.9<0.001
 HBsAg+0.30.20.3
 Anti-HCV18.32.8<0.001
 HTLV0.50.30.1
 CMV66.465.90.6

OPO use of HRDs

Recovery: Figure 1 depicts the percentage of OPO volumes from HRDs. The mean OPO annual HRD volume was 10.1 organs, with OPO volumes ranging from 1 to 63.6 organs per year. Only five OPOs reported two or fewer organs per year from HRDs, while only three OPOs reported volumes greater than 30. On average, HRDs comprised 7.7% of an OPO's total donor volume, but there was wide variation, ranging from 2.3% to 26.1% (Figure 1, upper panel). This did not appear to correlate with the AIDS prevalence within the OPO donor service area, as some of the OPOs with lowest proportion of volume from HRDs had some of the highest AIDS prevalences, while OPOs with some of the highest proportion of volume from HRDs had some of the lowest AIDS prevalences (Figure 1, lower panel).

image

Figure 1. Proportion of deceased donors in a given OPO that are classified as CDC high risk (upper panel), and AIDS prevalence in the general population of that OPO (cases per 100 000 population; lower panel). To show the degree correlation between high-risk donor (HRD) proportion and AIDS prevalence in each OPO, both histograms were sorted by proportion of HRDs, such that the X-axes of upper and lower panels correspond.

Download figure to PowerPoint

Transplantation:  Kidneys and livers from HRDs were more likely than pancreata, lungs and hearts to be transplanted (Figure 2). Of all HRDs where at least one organ was recovered, at least one kidney was transplanted from an average of 72.4% (range 31.8–100% among OPOs), and the liver from 78.3% (range 56.3–100%). In contrast, pancreata were only transplanted from an average of 16.0% of HRDs (range 0–47.4%), hearts from 24.9% (range 0–57.1%) and lungs from 11.9% (range 0–30.7%).

image

Figure 2. Distribution across OPOs of the proportion of high-risk donors where at least one kidney (upper left), liver (upper right), pancreas (middle left), heart (middle right) or lung (lower left) was transplanted. Density (Y-axis of histogram) indicates the number of OPOs where the proportion is that which was indicated on the X-axis.

Download figure to PowerPoint

Survey of OPO NAT practices

There was wide variation in NAT use by OPO and virus type. OPOs were more likely to always perform NAT for HIV and HCV compared to HBV: 51.7% of all OPOs performed HIV NAT for all donors, compared with 48.3% for HCV NAT and only 24.1% for HBV NAT. Twenty-four percent of OPOs never NAT tested for HIV, compared with 31% for HCV and 65.5% for HBV. The remaining OPOs fell somewhere in the middle: 10.3% performed HIV NAT for HRDs only, 12.1% for selected HRDs only and 1.7% only when requested. The patterns for HCV and HBV were similar (Table 3).

Table 3.  Summary of national OPO-level reported NAT practices
 HIVHCVHBV
No.%No.%No.%
  1. 1Of OPOs that perform NAT.

Perform NAT
 Always3051.7%2848.3%1424.1%
 For all CDC high-risk donors 610.3% 712.1% 3 5.2%
 For some CDC high-risk donors 712.1% 4 6.9% 2 3.4%
 When requested 1 1.7% 1 1.7% 1 1.7%
 Never1424.1%1831.0%3865.5%
Average time between order of NAT and receipt of results1
 2–4 h 2 4.5% 2 5.0% 0 0.0%
 4–8 h2147.7%1947.5%1260.0%
 8–12 h 818.2% 717.5% 420.0%
 12–24 h 613.6% 615.0% 1 5.0%
 >24 h 715.9% 615.0% 315.0%
Location where NAT testing performed1
 In-house 715.9% 615.0% 525.0%
 Transplant center laboratory 511.4% 512.5% 210.0%
 Outside lab in same city 818.2% 717.5% 1 5.0%
 Outside lab in same state, different city 4 9.1% 410.0% 1 5.0%
 Outside lab in different state2045.5%1845.0%1155.0%
Percent of time NAT results are available before allocation decisions are made1
 Always1636.4%1537.5% 735.0%
 75–99%1431.8%1230.0% 735.0%
 50–75% 2 4.5% 1 2.5% 0 0.0%
 25–50% 3 6.8% 410.0% 210.0%
 10–25% 4 9.1% 410.0% 210.0%
 <10% 2 4.5% 1 2.5% 1 5.0%
 Never 3 6.8% 3 7.5% 1 5.0%

Of the 76% of OPOs that performed HIV NAT, only 36.4% reported always having the NAT results available in time for a transplant decision. Interestingly, 6.8% of OPOs reported that they never had HIV NAT results available in time for a decision, despite choosing to perform NAT. A similar pattern was seen for HCV, where 37.5% of OPOs using NAT reported that they always had the results in time for a decision, while 7.5% reported they never had them in time. For HBV, 35% of OPOs using NAT reported they always had results in time, while 5% reported they never had them. The remaining proportions are somewhere between these extremes, with 31.8%, 30.0% and 35% of OPOs that perform NAT reporting results available 75–99% of the time (HIV, HCV and HBV). A majority of OPOs reported that receiving the results for an HIV, HCV or HBV NAT took between 4 and 8 h on average. However, a significant number of OPOs reported that receiving NAT results took more than 24 h (15.9%, 15.0% and 15.0% for HIV, HCV and HBV).

There was also variation in the location of NAT testing for each OPO. For many, NAT was performed in outside labs located in different states from the OPO headquarters (45.5%, 45.0% and 55.0% for HIV, HCV and HBV). Around 15.9% of OPOs that performed HIV NAT performed it at an on-site location, this figure was 25% for OPOs performing HBV NAT, and 15.0% for OPOs performing HCV NAT. The remainder performed the tests in a transplant center laboratory (11.4%, 12.5% and 10.0% for HIV, HCV and HBV), in an outside lab in the same city (18.2%, 17.5% and 5.0%) or an outside lab in the same state (9.1%, 10.0% and 5.0%).

Geographic patterns in NAT practice by OPO

Maps depicting NAT performance practices across the United States show significant geographic variation (Figure 3). In the Western and Southwestern regions, HIV and HCV NAT are almost always performed, whereas the majority of the Midwest and parts of Texas never perform HIV and HCV NAT. In the south, HIV and HCV NAT tend to be performed only for HRDs, or when requested. The Northeast shows high heterogeneity in NAT testing practices for HIV and HCV. HBV NAT trends are very different from HIV and HCV, with a majority of the West, Midwest and Southern regions never performing HBV NAT.

image

Figure 3. Performance of NAT testing by OPO, at county level.

Download figure to PowerPoint

Association between HIV NAT practice and HRD disposition

Total donor volume (number of donors where at least one organ was recovered), and total transplant volume also did not differ significantly between OPOs that always performed NAT and OPOs that did not (Table 4). The percentage of OPO volume from HRDs was slightly higher in OPOs that always performed HIV NAT (7.9% vs. 7.4%), but this difference was not statistically significant. There was also minimal difference in the percentage of transplant volume derived from HRDs (7.6% in OPOs that always performed HIV NAT vs. 7.3% in those that did not).

Table 4.  OPO-level demographics and use of NAT for HIV
 HIV NAT testing policy
Sometimes/neverAlwaysp-Value1
  1. 1p-Values calculated using two-sided t-tests.

AIDS prevalence   
 AIDS prevalence (per 100 000 pop.) 12.2 11.80.9
OPO volume (no. of cases per year)
 OPO donor volume118.2115.10.9
 OPO transplant volume111.1107.30.8
Percent OPO volume from HRDs
 Donor volume  7.4  7.90.7
 Total transplant volume  7.3  7.60.7
 Kidney transplant volume  7.0  7.40.8
 Liver transplant volume  7.4  7.80.7
 Pancreas transplant volume  6.8  5.20.4

HIV NAT was associated with a decrease in the percentage of total livers and kidneys exported nationally by an OPO (Table 5). An average of 5.8% of all livers and 7.2% of HRD livers were exported nationally when HIV NAT was always performed, versus 11.6% of all livers and 12.6% of HRD livers when HIV NAT was not always performed. An association was found between 31% less HIV NAT use and each increasing quintile of liver national export volume, and 26% less HIV NAT use and increasing quintile of HRD liver national export volume (p < 0.001). An association was also found between 25% less HIV NAT and each increasing quintile of kidney national export volume (p < 0.001). There was an association between 8% less HIV NAT use with each increasing quintile of total pancreas national export (p = 0.04), but no statistically significant association between lowered HIV NAT and increased national export of pancreata from HRDs.

Table 5.  HIV NAT testing policy and national export of organs
 HIV NAT testing policy
Sometimes/neverAlwaysRR2p-value
  1. 1All averages are weighted by OPO volume.

  2. 2Per quintile, weighted by OPO total volume of organ type and OPO HRD volume of organ type, as appropriate for each analysis.

Percent of total organs exported nationally1
 Kidneys21.619.80.75<0.001
 Livers11.65.80.69<0.001
 Pancreata36.933.70.94  0.04
Percent of HRDs exported nationally
 Kidneys20.719.40.960.1
 Livers12.67.20.74<0.001
 Pancreata28.229.81.260.2

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

While the PHS/CDC guidelines were designed to minimize HIV transmission, we found high rates of HBV and HCV infection in HRDs compared to non-HRDs, suggesting that these guidelines may be useful for preventing transmission of these diseases as well. Although HRDs carry extra infectious risks, we found they are less likely than non-HRDs to be ECD by UNOS kidney definition, suggesting that these organs could be very beneficial to potential recipients, provided the infectious risks were mitigated. We found wide variation in both NAT practice patterns and disposition of HRD organs by OPO. We did not find any association between universal NAT practice and AIDS prevalence in the OPO, nor did we find any associations between universal NAT practice and OPO volume or percentage of volume derived from HRDs. We did find that OPOs with universal HIV NAT practices were less likely to export livers and kidneys nationally.

An OPO's decision to perform NAT is not straightforward. While ELISA is inexpensive and rapid, its performance depends on antibody formation, and NAT outperforms ELISA for detecting recent infections. Furthermore, NAT use can lead to higher rates of false positive results and the potentially inappropriate discard of viable organs. It is also important to note that NAT does not eliminate the need for antibody testing, as transmission of some infections can still occur in the setting of a NAT-negative, antibody-positive result. Providers might consider using organs from donors who are NAT negative, ELISA positive for certain infections, particularly HBV, as the risk of transmission with proper prophylaxis is likely very small for all organs except the liver (23,24). While NAT offers a significant reduction in the WP, it is more expensive and time consuming than ELISA, and some have argued that the additional cost and potential increase in cold ischemic time might outweigh the benefits from the few infections that might be prevented (25).

Although many HRDs have been transplanted in the United States, there have been very few reports of transmission of HIV and HCV. Given the legal and regulatory implications of such reports, and the lack of seroconversion tracking by UNOS, this may reflect a strong reporting bias. However, even with such a reporting bias, the paucity of reported seroconversions likely reflects a relatively good ability of antibody testing to detect transmission risk. While universal NAT may not prevent a large number of infections compared to ELISA (12), the impact on organ disposition could be significant. The increase in national export by OPOs without HIV NAT is likely partially explained by the geographic homogeneity in NAT practices. If an OPO does not perform NAT, it is unlikely the OPOs in the surrounding region perform NAT, and thus the organ has to be exported nationally if an OPO performing NAT is more likely to accept the organ. Although we cannot say whether the relationship between NAT practice and national export of organs is causal, there is evidence to suggest that infectious risk is a major disincentive for organ use. A negative NAT allows a provider to say with more certainty that the infectious risk of an organ is very low, and thus might increase the probability that the organ be used locally rather than exported nationally.

We acknowledge several limitations to our study. The UNOS donor data base only includes information on donors where at least one organ was recovered. If HRDs were less likely to reach this step in the donation process because of OPO worries that their organs would not be accepted for transplantation, this could bias our results to underestimate the potential for expanding the donor pool. It is possible that centers in an OPO are aware of their OPO's NAT practices, and thus less likely to refer HRDs in an OPO when NAT is not performed. If this were the case then our results would underestimate the association between NAT and HRD disposition. This bias may also exist on the level of the center: the final decision of whether or not to use the organ is made by the transplant surgeon; surgeons in OPOs that do not perform NAT may be less likely to use an organ from an HRD than surgeons from OPOs who do use NAT. A further limitation is that all data were based on reports and estimations made by the OPO clinical director rather than actual data, thus, it is possible that OPO directors may have over- or underestimated the average time and availability of NAT results. It is also possible that there is heterogeneity in OPO interpretation of donors as high risk; however, we think this is unlikely as the question on the deceased donor registration form explicitly refers to the PHS/CDC guidelines. Finally, our analysis of the association between AIDS prevalence and OPO NAT use is limited in two ways: (1) the use of AIDS prevalence as a surrogate for HIV prevalence is limited given the obvious difference between end-stage HIV disease diagnosis and screening, but the lack of data on HIV prevalence effected this decision, and (2) our method of calculating OPO-level AIDS prevalence by OPO ignored within-state differences in prevalence that likely exist. Our analysis was purely exploratory and further research is needed using county-level data to establish the relationship between OPO NAT use and AIDS prevalence.

Universal NAT allows both patients and providers to make better informed decisions about use of organs from both HRDs and non-HRDs, as NAT significantly decreases the window between infection and detectability. We have shown that NAT practices differ drastically among OPOs. Furthermore, NAT may increase an OPO's likelihood of placing the organ locally as opposed to exporting it nationally, a practice that might decrease the amount of cold ischemic time and as a result lead to better posttransplant outcomes. The wide variation of current practice and the possibility that NAT would improve organ utilization support consideration for a national policy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References