Ambivalence in living liver donors



All right hepatic lobe (RHL) donors in our program are asked to participate in a longitudinal quality-of-life study that begins at their evaluation and continues throughout the first postdonation year. Here we report the characteristics of donor candidates who completed the donation process despite ambivalence. In all, 183 RHL candidates consented, and 133 became donors. Ambivalent donors (ADs; n = 45) identified themselves through verbal statements or written comments, or they were identified by staff during the evaluation. ADs were predominantly male (73.3%), were older than unambivalent donors (UADs; >35 years: 76% of ADs versus 53% of UADs, P = 0.008), and were well educated (college graduate: 60% of ADs versus 17% of UADs, P = 0.01). Brother-to-brother and son-to-father combinations were most common among ADs. Alcohol (22% versus 11%, P = 0.04) and hepatitis C virus (51% versus 27%, P = 0.008) were more common as disease etiologies for recipients with ADs versus recipients with UADs. More ADs than UADs considered themselves to be religious (68.9% versus 43.2%, P = 0.007). Ambivalence about RHL donation was present in 33.8% of the candidates who completed the donation process. These results suggest that ambivalence should not be the sole reason for disqualifying a potential donor who otherwise satisfies program requirements. Liver Transpl 17:1226–1233, 2011. © 2011 AASLD.

Living donor adult liver transplantation is now an accepted treatment modality for end-stage liver disease and leads to patient and graft survival rates equivalent to those seen with deceased donor liver transplantation.1-4 In areas with long waiting times for deceased donor organs, living donor adult liver transplantation may be one of the only realistic opportunities for avoiding disease progression beyond parameters acceptable for deceased donor liver transplantation. The benefits of living donor transplantation for recipients are obvious, and transplant teams are faced with the need to balance these benefits against the risks for living right hepatic lobe (RHL) donors.

At our center, all candidates for RHL donation undergo an extensive physical and psychological evaluation before the actual donation. The protocol has been previously described in detail.5, 6 All donor candidates who reach the evaluation stage are asked to participate in a longitudinal quality-of-life (QOL) study. The results for the first 43 donors indicated that a subset of this group had expressed ambivalence about the donation process either directly or inferentially.

Ambivalence in living donors has been the subject of much study, but there is little consensus about whether ambivalence should disqualify a potential donor. It has been described as mixed feelings that coexist with the intention to donate, and Simmons et al.7-9 found it to be fairly common among kidney and bone marrow donors. They observed that high predonation levels were associated with psychological difficulties after donation, but they did not suggest the exclusion of ambivalent donors (ADs) otherwise judged to have made an informed choice to ultimately proceed with donation. In contrast, others have suggested that predonation donor ambivalence is so strongly associated with postdonation psychological distress that AD candidates should be disqualified.10-15 It is likely that these discrepancies arise in part from different approaches to defining and assessing ambivalence across studies.

In this study, we examined the records of 183 potential donors. Here we describe a group of actual RHL donors who displayed ambivalence during the evaluation or the postdonation follow-up. We use a more formal dictionary definition of the word ambivalence: simultaneous and contradictory attitudes and/or feelings toward a person (the recipient) or an action (RHL donation). In agreement with the work of Simmons et al.7-9, 16 and others,10 we believe ambivalence to be a process distinct from reluctance (a consistent unwillingness to donate) and coercion (the imposition of the will of another person such as a family member or the recipient on the donor candidate). Although donor reluctance and coercion are obvious reasons for rejection, ambivalence in a donor candidate may represent a process of careful consideration and should not be used as the sole basis for donor disqualification.


AD, ambivalent donor; HCV, hepatitis C virus; MCS, Mental Composite Score; NS, not significant; PCS, Physical Composite Score; QOL, quality of life; RHL, right hepatic lobe; SF-36, Short Form 36; UAD, unambivalent donor.


Donor Evaluations

All donor candidates are evaluated as previously described.5, 6 Potential donors are between the ages of 18 and 60 years and must be genetically or emotionally related to the potential recipient. Screening laboratory values are obtained, and an initial telephone assessment is conducted by the donor coordinator to rule out patients with major medical issues. The donor surgeon and the dedicated living donor coordinator conduct initial informational sessions that include a description of the evaluation process, the surgery, the postoperative care, and the incidence of complications (including death) as well as a preliminary physical examination.

Potential donors who remain interested undergo a stepwise assessment of their medical and psychosocial suitability. All donor candidates are evaluated by an independent donor advocate team, which consists of a psychiatrist, a social worker, and a gastroenterologist who have no affiliation with the recipient evaluation team. Each member attempts to ensure that coercion by family members or the medical team has not occurred and that motivations of guilt or unrealistic expectations for recipient outcome are not unduly influencing the decision to donate. The assessments of the team member and the pertinent laboratory and imaging data are presented to the weekly multidisciplinary screening committee, which evaluates the suitability of each donor-recipient pair.

The committee's decision is first communicated to the donor so that individuals who no longer wish to donate can abort the process. Donor candidates are repeatedly assured that they may stop the donation process at any point up to the induction of anesthesia and that such decisions will be kept confidential. A suitable explanation (eg, anatomic variants) is offered to those who withdraw.

Patient Population

All donor candidates whose initial meeting with the donor surgeon occurred between March 2001 and October 2008 were asked to participate. In all, 184 potential donors were asked to participate, and 183 were enrolled prospectively into the QOL study. One hundred thirty-three became RHL donors, and they formed the study cohort.

QOL Instruments

This study was approved by the institutional review board of Lahey Clinic. It included 2 surveys: the standard Short Form 36 (SF-36) health-related QOL survey and a customized, procedure-specific, 43-question survey developed by our staff. The SF-36 questionnaire is a validated survey that evaluates various measures of physical and mental health. Numerical scores in each category are used to calculate aggregate values for the Physical Composite Score (PCS) and the Mental Composite Score (MCS). For the general population of the United States, the average MCS value and the average PCS value are both 50.17, 18

The program-specific survey has been previously reported.6 The questions cover demographics, financial concerns, relationships, the willingness to participate in the process again, symptoms and pain, the psychological impact of donation, and the effect on employment status. The survey also contains a section in which donors are encouraged to comment or to write about their concerns. Donors complete both surveys before the donation and at 5 time points during the first postdonation year (at 1 week and at 1, 3, 6, and 12 months).

Determination of Ambivalence

A retrospective review of the chart notes of donors, the free-text sections of the survey, and the minutes of the weekly living donor committee meetings were reviewed. Individual records from the first evaluation to the 1-year postdonation visit were included. The period was March 2001 to October 2009, and this provided at least 1 year of follow-up for all donors. The following chart notes were used as source documents: all encounters with the donor surgeon, the psychiatrist, the social worker, the nurse coordinator, and the donor gastroenterologist. During the 8 years of the study period, the donor surgeon (n = 1) and the gastroenterologists (n = 2) remained the same. The psychiatrist, the social worker, and the donor coordinator each changed once during this time period.

The following data were considered to be indicative of ambivalence: a written comment by a donor on any survey indicating mixed feelings about donating or having donated; a verbal comment to a provider (documented in a record) indicating uncertainty about disease recurrence, recidivism in the recipient, or the death of the recipient; documentation indicating a donor's concern about donation endangering his well-being or the well-being of his family; a provider assessment indicating a donor's hesitancy about proceeding with the evaluation; and reports from family members about a donor's hesitancy that in the professional opinion of the assessor were judged to be valid.

The records were divided randomly among 3 researchers (M.A.S., J.K., and D.S.M.) and were assessed for ambivalence as described. All records for an individual patient were reviewed by a single reviewer. An additional assessment was completed by another researcher (J.E.V.) to determine interobserver variability. Twenty charts (5 from each 2-year period of the study) were randomly chosen and were assessed for ambivalence as described.

Data Analysis

Statistical analysis was performed with SPSS for Windows. Descriptive statistics and mean values were compared with the Student t test; P values less than 0.05 were considered significant. We analyzed responses to the SF-36 by fitting a linear regression with QOL measures as dependent variables and with the number of days from donation as an independent variable.



The study patients were 133 actual RHL donors. The cohort was predominantly male (85 men and 48 women), the ages ranged from 18 to 59 years, and all the donors were genetically or emotionally related to the recipients. All the donors completed 1 year of follow-up after donation.


In all, 45 ADs were identified (33.8% of the total cohort). Table 1 lists the sources for the AD designation. Eighteen donors were identified only by staff members, 14 donors were identified only by themselves, and 9 donors were identified by themselves and staff members. Another 2 donors were identified by themselves, staff members, and family members, and 2 donors were identified by staff members and family members but not by themselves. On average, 5.625 ADs (range = 5-8) were identified in each of the 8 years of the study. The characteristics of the staff-identified and self-identified groups were first analyzed separately. No differences were identified between the groups; therefore, we present the results for all 45 ADs.

Table 1. ADs: Means of Identification
  • NOTE: There were 45 donors. Chart notes, QOL surveys, and donor committee meeting minutes were reviewed.

  • *

    The percentages do not add up to 100 because of rounding.

Staff only1840.0
Donor only1431.1
Staff and donor920.0
Staff, donor, and family24.4
Staff and family24.4

Ambivalence was most often identified early in the process (38 before donation, 32 at 1 week, and 27 at 1 month), with fewer ADs noted at later times (14 at 3 months and 8 at 6 months). No differences were noted between those demonstrating ambivalence before donation and those whose ambivalence was detected for the first time after donation. The ADs were predominantly male (33/45 or 73.3%), and 52 of the 88 unambivalent donors (UADs; 59.1%) were male (P = 0.04 versus ADs).

Age Distribution

The average age for all donors in this study was 38.7 years (range = 18-59 years). ADs were 39.6 years old on average (range = 18-56 years), whereas UADs were 37.9 years old on average [range = 22-59 years, P = not significant (NS)]. However, the proportion of donors older than 35 years was significantly larger in the AD group (76%) versus the UAD group (53%, P = 0.01 for ADs versus all donors and P = 0.008 for ADs versus UADs; Fig. 1).

Figure 1.

Donor age. Actual donors who were older than 35 years and actual donors who were 35 years old or younger are compared in the following categories: all donors, ADs, and UADs. *P = 0.01 versus all donors and P = 0.008 versus UADs.

Educational Levels

All donors in this study graduated from high school. ADs were better educated than UADs, with 85% having attained some post–high school education. Specifically, 60% of ADs (27/45) had baccalaureate degrees, whereas only 17% of UADs (15/88) did (P = 0.01). Additionally, 20% of ADs (9/45) had some college education, and 6.7% (3/45) had graduate degrees. UADs had higher representation in the category of graduate degrees (14% versus 6.7%, P = NS; Fig. 2).

Figure 2.

Educational levels of the donors. The donors are grouped according to the highest level of their education. The results for all donors, ADs, and UADs are compared. *P = 0.01 versus UADs.

Recipient Liver Disease

The most common etiology of end-stage liver disease for recipients (regardless of the donor category) was hepatitis C virus (HCV), which was followed by alcohol-related disease. However, HCV was the disease etiology for 51% of the recipients with ADs but for only 27% of the recipients with UADs (P = 0.008). Alcoholic liver disease was seen twice as frequently in the recipients with ADs (22%) versus the recipients with UADs (11%, P = 0.04). Conversely, all other causes of liver disease were more common for the recipients with UADs (53/88 or 60.2%) versus the recipients with ADs (12/88 or 13.6%, P = 0.001; Fig. 3).

Figure 3.

Etiologies of recipient diseases. The end-stage liver disease etiologies of the recipients are compared for all donors, ADs, and UADs. HCV and alcohol were most common for all the groups and are listed separately. All other causes of liver disease are grouped together. *P = 0.02 versus all donors and P = 0.008 versus UADs, **P = NS versus all donors and P = 0.04 versus UADs, and ***P = 0.02 versus all donors and P = 0.001 versus UADs.

Relationship to the Recipient

The largest subcategory of ADs was brother to brother (18/45 or 40%). Sisters were less likely to be ambivalent about the process [sister to sister (2) and sister to brother (2)]. The number of child-to-parent pairs was also significant for ADs (17/45 or 37.8%), and there were more sons (11/45 or 24.4%) than daughters (6/45 or 13.3%; Table 2).

Table 2. ADs: Relationships With the Recipients
Males (n = 33)Females (n = 12)
  1. NOTE: There were 45 donors.

Brother to brother (18)Sister to sister (2)
Brother to sister (1)Sister to brother (2)
Son to father (7)Daughter to mother (4)
Son to mother (4)Daughter to father (2)
Brother-in-law (2)Friend (1)
Friend (1)Mother to adult daughter (1)


The total cohort of 133 donors was almost evenly divided between those who described themselves as religious or spiritual and those who indicated no religious preference (P = NS; Fig. 4). Thirty-one of the 45 ADs (68.9%), 38 of the 88 UADs (43.2%, P = 0.007), and 69 of all donors (51.9%, P = 0.004) considered themselves religious.

Figure 4.

Religious affiliations of the donors. The donors were asked to classify themselves as “religious” or “no preference.” The responses of all donors, ADs, and UADs are compared. *P = 0.007 versus UADs and P = 0.004 versus all donors.

Interobserver Variability

There was 95% agreement between the initial rates and the independent rater, with the independent rater assigning 1 additional donor to the ambivalent category. This resulted in a kappa value of 0.89, which indicated good interobserver correlation.


In all, 183 potential donors participated in the QOL study conducted at the time of the evaluation. ADs did not demonstrate significant differences in any SF-36 domain versus the total cohort of actual donors. Table 3 shows the composite scores for ADs and UADs in the study period.

Table 3. Mean SF-36 Composite Scores at All Study Time Points: ADs Versus UADs
 Before Donation1 Week1 Month3 Months6 Months1 Year
  1. NOTE: The composite scores for all SF-36 domains were compared for UADs and ADs at each time point. No differences were observed. In each group, the MCS remained stable throughout the study period. The PCS decreased after donation but rose steadily during the study period and returned to the baseline values by the 1-year mark.


The SF-36 results for ADs and UADs at the time of the evaluation were also compared to the results for donor candidates who did not proceed to RHL donation. Among the donor candidates who did not become actual donors, those whose recipients improved or received deceased donor transplants, those who deferred to other living donors, and those who were rejected for anatomic reasons did not significantly differ from actual donors. In comparison with actual donors, those who were rejected for medical reasons had lower scores for domains related to physical health (P = 0.02) but showed equivalent results for domains related to mental health. Donor candidates whose evaluations were cut short by the deterioration or death of a recipient had lower scores for the Vitality scale (P = 0.03), the Mental Health scale (P = 0.01), and the MCS (P = 0.04) in comparison with donors. However, the donor candidates who withdrew from the process had lower MCS values in comparison with the total donor population (P = 0.04) and the AD subgroup (P = 0.03; Table 4).

Table 4. SF-36 Scores at the Time of Evaluation: Actual Donors Versus All Other Categories
PCSMCSVitalityMental HealthP Value
  1. NOTE: The scores for all 8 SF-36 domains from the surveys given during the predonation evaluation were compared. The table shows the domains for which significant differences from the results for actual donors were observed; the remaining domains showed no differences from the results for actual donors.

 ADs (n = 45)57.553.558.854.0NS for all domains
 UADs (n = 88)57.252.957.953.7 
Reasons for stopping the evaluation (n = 50)     
 Recipient improvement (n = 3)58.254.860.558.0NS for all domains
 Deceased donor liver transplantation (n = 7)57.855.658.951.1NS for all domains
 Another donor (n = 9)57.152.958.652.9NS for all domains
 Unsuitable anatomy (n = 8)55.156.354.953.9NS for all domains
 Medical rejection (n = 10)53.153.656.253.80.02 for PCS versus donors
 Recipient sickness or death (n = 6)52.746.651.245.3NS for PCS, 0.04 for MCS, 0.03 for the Vitality scale, and 0.01 for the Mental Health scale versus donors
 Withdrawal (n = 7)58.148.656.650.90.04 for MCS versus donors


The unquestioned benefit of living donor liver transplantation for the recipient cannot be allowed to overshadow the fact that the living liver donor is a healthy person who has volunteered to undergo a complex surgical procedure that will place him or her at risk for harm. All discussions of living donor transplantation emphasize that donor safety is the overriding concern. Safeguards are put in place to ensure that an RHL donor will retain sufficient liver volume to maintain his or her health and that there are no known significant underlying medical conditions. These measures to protect the donor's physical health are relatively straightforward and uncomplicated. Assessing and safeguarding the donor's mental health is a far less precise procedure.

We performed a retrospective review of chart entries and QOL surveys completed by actual RHL donors, and we found that approximately one-third displayed ambivalence at some point. The word ambivalence is defined in this study as simultaneous and contradictory attitudes and/or feelings toward a person (the recipient) or an action (RHL donation). In the organ donation literature, this term has been defined as simply mixed feelings coexisting with an intention to donate. The definition that we have adopted is more precise and can, therefore, be better operationalized. We feel that it is important to identify our use of ambivalence and to distinguish it clearly from reluctance and coercion because ambivalence is frequently used to describe situations that more completely fit the definition of reluctance or coercion.7, 13, 19-22

The circumstances surrounding RHL donation foster ambivalence and make adherence to ideal standards for informed consent difficult.19, 20 A potential donor must balance the impending death of a loved one with the knowledge that a refusal to donate may hasten death and that successful donation will most likely be lifesaving. Potential living liver donors must make a decision that will have major effects not only on themselves and their recipients but also on the lives and well-being of their families. In making a life or death decision, they are aware of both the effort required to provide care to the recipients and the hardship that will ensue if they are injured as a result of the donation.

It is unrealistic to expect someone who has carefully considered all aspects of RHL donation to have no concerns about donation. However, voicing realistic concerns about the risks of donation is not the same as experiencing ambivalence. Donors may be well informed and understand that risks exist but still have no hesitancy or doubt about the decision to proceed with donation.10 Nevertheless, it is also likely that our methods underestimate the actual incidence of ambivalence in donors and that the donors whom we identified experienced relatively high degrees of ambivalence. The rate of ambivalence noted in our study is, however, well within the range of ambivalence rates noted in other living donor cohorts, which typically vary from 20% to 65%.8, 16, 20, 23 Donor ambivalence may be part of a decision continuum that begins with ambivalence and progresses to reluctance and ultimately to withdrawal from the donation process. It is also possible for ambivalence toward the act of donation or toward the recipient to exist in individuals who willingly proceed to become RHL donors. The retrospective nature of this study precludes the use of some of the quantitative measures of ambivalence7-9; however, we have identified some characteristics of donors who fall into this latter category.

RHL donors were more likely to be male; this was seen in both the total cohort (63.9%) and the AD subgroup (73.3%). This is in contrast to earlier reports about kidney donors7, 8, 24, 25: all donors and ambivalent or reluctant donors were more likely to be female. Several factors, including major changes in the economic structure of families in Western societies, could account for the discrepancies. It is now commonplace for wives and mothers to contribute substantially to the family finances, and this negates the frequently cited argument that their temporary donation-related recovery period will be less disruptive to the overall stability of the family structure.8, 21, 22

The age distribution of ADs was similar to the age distributions observed for the total donor cohort and the UADs. However, 76% of the ADs were older than 35 years, but only 53% of the UADs were. The greater life experience and the increased family and work responsibilities that accompany age may have contributed to this result. Some individuals both expressed a willingness to donate and were concerned that donation might negatively affect their families. Also, younger individuals tended to downplay the possibility that potential adverse outcomes would actually affect them.

The ADs tended to be well educated. All graduated from high school, and 80% were working toward or had completed baccalaureate degrees; the college graduation rate for the UAD cohort was just 17%. Like the differences in the age distributions of the groups, this observation is difficult to interpret in concrete terms. A university education tends to emphasize the need to evaluate various solutions to a given problem. If that skill set is reflexively applied to RHL donation, the likely result may be some degree of ambivalence.

The liver disease etiologies for the recipients differed between ADs and UADs. Nearly 75% of the recipients with ADs suffered from diseases that are often considered self-induced. HCV was the etiology for 51% of the recipients with ADs; there was only 1 instance of transfusion-associated disease, with the remainder attributed to substance abuse. Alcohol-related disease accounted for another 22% of the recipients with ADs. In contrast, the incidence of HCV (including 4 cases of transfusion-associated disease) was 27% for recipients with UADs, and the incidence of alcohol cirrhosis was 11%. As part of our standard donor education protocol, donors in both groups received counseling about the potential for recidivism or disease recurrence in the recipients. However, only donors in the ambivalent group expressed concerns. Similar findings have been reported by others.26 It is unclear whether the UADs had no concerns in this area, had different attitudes and beliefs about recipients' responsibility or blame for their liver disease, or simply fell into the category of candidates who are determined to donate as soon as they become aware of the need and are rarely influenced by information presented during the evaluation process.8, 26, 27

Brothers were the donors most likely to be ambivalent. In all, 19 brothers were ambivalent about donation, with 18 in the brother-to-brother category (40%) and 1 in the brother-to-sister category. The next most common donor-recipient pairing associated with ambivalence was the child-to-parent category (37.8%), which had a larger number of sons than daughters. Sons donating to their fathers composed the largest number in this subgroup. Two things are worth noting about these groups. First, the available verbal or written comments of these donors indicate some degree of residual anger directed at the recipient and specifically related to the upheaval and, in some cases, the fear that the recipient imposed on the donor and the family at large while the recipient was under the influence of drugs or alcohol. This unresolved anger was likely a major contributor to the expressed ambivalence. Second, the ambivalence related to donation to a parent is in marked contrast to the total absence of ambivalence in parents donating to their minor children, even if the likelihood of successful transplantation is remote.28-30

One unexpected finding was a mother who repeatedly expressed ambivalence verging on anger about donation to an adult daughter suffering from acute hepatic necrosis secondary to minocycline use for acne. She questioned why the daughter's husband or other family members did not step forward, but when she was reminded that donation must be voluntary, she replied, “Well, I'm volunteering.” A psychiatric consultation obtained during her evaluation suggested that her statements and actions reflected frustration and anxiety associated with the rapid onset of her daughter's critical condition rather than an unwillingness to proceed. At 1 year, she indicated no regrets and was satisfied with the donation process.

A formal religious affiliation or a spiritual lifestyle was indicated by approximately half of the donors in the study. That increased to 68.9% in the AD group, and religious or morality-based concepts appeared to guide the decision-making process for donation. Some ADs were asked why they donated despite past problems and the potential for recidivism or disease recurrence. “It is the right thing to do” was a frequent response, and so was “If he drinks again, I will have no guilt; I will have done all that I can.” Another answer that hints at the donor's acceptance of past and present realities was “because I grew up without a functioning father—I want his children to have a chance at life with a functioning father.”

Notably, the severity and incidence of postdonation complications, the time to return to work, and the degree of hepatic regeneration were similar in the 2 groups of donors (additional data not shown).6 There were no demonstrable differences between ADs and UADs in any of the physical or mental health domains measured by the SF-36 health-related QOL survey at any of the 6 time points. Similarly to other reports,31 all donors had returned to their predonation mental and physical health states according to both self-reports and assessments by the donor team 1 year after donation. Although the lack of differences between ADs and UADs was unexpected, we believe it to be valid. Our study surveys the donors once before donation and 5 times after donation and ends at the 1-year anniversary. Each donor serves as his or her own control, and the results indicate that donors return to their predonation scores (usually by month 6).

Our donor evaluation process involves multiple individuals (at least 5), and all emphasize the voluntary nature of the process and the ability of the candidate to stop it at any time. Potential donors do opt out (1 candidate did so on the day of the scheduled donation). This process appears to be efficient at weeding out those who are truly reluctant versus those who are ambivalent. The right of the donor to stop the donation process without explanation or prejudice is emphasized by all caregivers at every encounter in the evaluation. The ADs described in this study did not make this choice, whereas others who came to the program did. This suggests that they were neither reluctant nor coerced and emphasizes the need for an evaluation protocol that allows donors the time necessary to make a deliberative decision.

In most instances, comparisons of the predonation survey results for the actual donors (ADs and/or UADS) and the candidates who began but did not complete the evaluation process showed no differences between the groups. Lower scores in categories related to physical health were observed in candidates rejected for medical issues, and lower scores in categories related to mental health were noted in candidates whose recipients died. Similar reductions in measures of physical or mental health in donor candidates with these issues have been reported by others.11, 12, 31

Unlike most candidates who did not proceed to donation, candidates who removed themselves from the process had lower MCS values in comparison with ADs and UADs. The small number of candidates in this category precluded an in-depth analysis of motivations, but this group likely included individuals who were truly reluctant to donate

To standardize our approach to data collection and analysis, we applied criteria that we believe meet a formal dictionary definition of ambivalence to reviews of the donor charts, and we subjected a random sample of the charts to a review by an independent rater to guard against observer bias. Despite these efforts, our findings must be viewed with caution because the study design was retrospective and no truly quantitative measures of ambivalence were employed. However, we feel that our findings about the nature and correlates of ambivalence in RHL donors suggest the importance of conducting future prospective studies in this population. Such studies could sequentially assess prospective donors' degree of ambivalence at multiple points during the evaluation and postdonation follow-up with the brief, psychometrically validated ambivalence scale used in previously reported kidney and bone marrow donor studies.7-9, 16 This approach would allow for a better understanding of the natural history of ambivalence during the predonation period. It would also enable an examination of whether ambivalence and other factors strongly contribute to postdonation psychosocial outcomes in living liver donors. The inclusion of a systematic assessment of ambivalence in prospective liver donors would also potentially facilitate efforts to counsel these individuals if this should become necessary and to increase the likelihood that they feel satisfied with their decision to donate.

It is important to view ambivalence in living liver donors as a concept distinct from coercion and reluctance. It is also important to recognize that ambivalence may be a transient intermediary leading to a positive or negative decision about donation; alternatively, it may be a permanent finding in a candidate deemed suitable in all other respects. We have described a cohort of RHL donors who expressed ambivalence but completed the donation process despite repeated assurances that it was all right to stop. These individuals were mature, were well educated, and were influenced by religious or moral beliefs to donate even when they were conflicted by past realities or future possibilities of poor recipient behavior. Importantly, they did not express regret about donating, and most said that they would donate again if it were possible. We believe that for many donors, ambivalence represents a careful and considered examination of the donation process and should not be the sole reason for disqualification of an otherwise acceptable candidate.