Gastroenterology
Volume 142, Issue 2 , Pages 207-210, February 2012
Baylor University Medical Center, Dallas, Texas
National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Maryland
published online 26 December 2011.
See “Estimates of early death, acute liver failure, and long-term mortality among live liver donors,” by Muzaale AD, Dagher NN, Montgomery RA, et al, on page 273.
Management of donor risk is the most important consideration in the decision to perform a living donor liver transplantation (LDLT). Counseling potential donors on operative morbidity and mortality requires a comprehensive understanding of the available data on outcomes. Ideally, an individual potential living donor should be informed of the potential complications based on an objective analysis of a center's own results. At present, this ideal has not been obtained. Instead, potential donors and their physicians must rely on outcome data from numerous single-center and a few multicenter studies. Reports from the three largest series from North America found that 28%–40% of donors developed ≥1 complication.1, 2, 3 Perhaps the most comprehensive of these studies, the 9-center Adult-to-Adult Living Donor Liver Transplant Cohort Study (A2ALL) noted that most complications were minor (27%) and life-threatening problems occurred in only 2% of cases.3 The most common complications were biliary leaks (9%), bacterial infections (12%), and incisional hernia (6%). Early postoperative laboratory abnormalities resolve within a few months of surgery with the exception of low platelet counts. A small minority of donors have sustained thrombocytopenia for >1 year after donation.4 Donor mortality is the most significant complication as underscored last year after 2 unexpected fatalities within months of each other at 2 of the largest LDLT programs in the country. The approximate risk of donor death is <1%, but a more precise estimation is difficult to ascertain for a number of reasons. Most important, the total number of LDLTs performed in the United States is small relative to the total number of liver transplants. Between 2000 and 2010, there were 3159 living donor liver transplants, which represented only 5.1% of all liver transplantations performed in the United States.5 In addition, unlike transplant recipients, there is no national database to register donor deaths. Consequently, reports of donor deaths are found through the media, case reports, or informal communications. Finally, there is no incentive for centers to report donor deaths owing to the negative repercussions of such disclosure.
In this context, the paper by Muzaale et al helps to provide some clarity regarding mortality following donation.6 Through the Organ Procurement and Transplantation Network, the authors acquired the social security numbers of all living liver donors and cross-referenced these with the Social Security Death Master File to determine periprocedural (within 90 days of surgery) and overall mortality rates. Comparisons were made with living kidney donors as well as participants in the third National Health and Nutrition Examination Survey (NHANES–III), 88% of which were excluded owing to comorbidities, which would have precluded living liver donation. These investigators reported that the 90-day mortality risk of living liver donors (1.7 per 1000 donors or 0.17%) was not significantly different compared with living kidney donors (0.05%). The long-term cumulative mortality risk was comparable to that of live kidney donors and NHANES participants (1.2%, 1.2%, and 1.4%, respectively) at 11 years. The authors further describe 4 cases of nonfatal acute liver failure in living liver donors. Three of these patients required deceased-donor liver transplantation. Because the Organ Procurement and Transplantation Network provides the full cohort of living liver donors in the United States and confirmation of mortality through the Social Security Death Master File is absolute, the results of this study are likely the best estimation of donor mortality. Clinicians will no doubt find this information useful as they counsel potential LDLT donors and recipients. In addition, this format of cross-referencing 2 databases provides an important study design for future estimates of donor mortality.
Although this study analyzed the largest cohort to date with the most rigorous methodology, the results are similar to previous reports. One prior analysis of publications in the medical and lay press reported donor mortality “definitely” and “definitely or possibly” related to the donor surgery at 0.15% and 0.20%, respectively.7 Another study based on survey data estimated donor mortality risk at 0.2%.8 An important distinction of the current paper from earlier reports is the use of comparison groups. However, the identification of an appropriate comparator group for living liver donors is difficult because of their unique characteristics. Living liver donors represent a highly selected group; up to 65% are rejected for the discovery of even a mild medical problem during the evaluation.9 Therefore, the use of a general population control group (NHANES) for comparison with the living liver donors is problematic. Living liver donors are likely to be healthier than the NHANES participants, despite the exclusion of 88% of that cohort for comorbidities. Consequently, the long-term health of living liver donors might be expected to be even higher than NHANES participants. The fact that there was no difference in long-term mortality between these 2 groups could potentially reflect a worse outcome for the liver donors. In addition, the nonconcurrent time of selection for the 2 cohorts, 1988–1994 for NHANES and 1994–2011 for the living donors, raises concerns about their true comparability. The use of living kidney donors is perhaps the best comparator group for living liver donors, although the risk of a kidney donor operation (0.05%) is likely much lower than liver donation (0.17%), even if not statistically different. On the other hand, as a result of the large volume of living donor kidney transplantations (approximately 6000 per year), there have been 3 times as many perioperative deaths than after living liver donation (Figure).10 Interestingly, these kidney donor deaths have received much less media attention. Because perioperative mortality is a rare event, efforts to lower the rates may need to take a safety-systems approach, as is currently being developed among A2ALL sites.11
Figure. Number of living liver and living kidney early (< 90 days) donor deaths.6, 10 While the number of living liver donor deaths per 1000 donors is higher (1.7 vs 0.5 living kidney donors), the total number of deaths for kidney donors per year (1.7) is more than four-fold higher compared with liver donors (0.4), because the number of kidney donors per year (5410) is more than 20-fold higher than liver donors (242).
The finding of living liver donors who died by their own hand is notable. The current analysis discovered 2 such deaths in the early perioperative period. The A2ALL study identified 2 other deaths from donor suicides or drug overdose and another attempted suicide, all of which occurred >90 days after surgery. Although psychiatric complications are uncommon, occurring in about 4% of living liver donors, they may be attributable indirectly to the operation by the stress associated with undergoing the complicated evaluation and surgery.12 The number of documented donor deaths by suicide and drug overdose (n = 4) rivals the total number of early perioperative deaths by other means (n = 5), highlighting the importance of the mental health of donors and prospective donors. Furthermore, the occurrence of these complications after the early postoperative period emphasizes the possibility that surgical complications may not be immediately evident. In fact, one third of rehospitalizations for living liver donors occur >90 days after donation.13
In consideration of health-related quality of life (HRQOL) of donors, a recent review concluded that physical well-being declines after surgery, but returns to predonation levels within 1 year, and mental health status remains about the same as predonation HRQOL.14 A subsequent study has confirmed these findings.15 However, these conclusions are drawn from just a handful of studies with both pre- and postoperative results that included about 300 donors in all. In addition, postdonation data were not obtained from all donors, introducing potential biases—donors with worse HRQOL may not respond to requests for information. Most important, long-term information on donor HRQOL is largely absent. Out-of-pocket expenses and lost wages have been estimated to average about $5,000, but the range is substantial.16 Importantly, the personal economic impact is among the greatest concerns for potential living donors.
Not all outcomes for the donor are negative. Among potential benefits is the intangible benefit of an altruistic act. There is often the direct benefit of having a loved one remain alive and returned to function. This is more likely with living than deceased donation owing to the increased risk of death during the often protracted time that recipients await deceased donor transplantation.17 There is also a public health benefit: Living donation allows another patient to receive a deceased donor liver, thereby decreasing the demand for deceased donor transplantation.
In summary, the report by Muzaale et al6 is the current standard for estimating donor mortality. Because the liver community has been unable to establish a national database for donor outcomes, and there are no real prospects for such, the rigorous methodology utilized in this study will allow the best means of prospectively reporting living liver donor mortality risk in the future. However, much more high-quality information is needed in LDLT, especially regarding long-term outcomes.
Acknowledgments
The opinions expressed herein by Dr. Everhart are those of the author and do not necessarily reflect the views of the National Institutes of Health or the Department of Health and Human Services.
Conflicts of interest The authors disclose no conflicts.
PII: S0016-5085(11)01700-8
doi:10.1053/j.gastro.2011.12.018
© 2012 AGA Institute. Published by Elsevier Inc. All rights reserved.
No comments:
Post a Comment