SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The aims of this study were to investigate the long-term effects of living donor liver transplantation (LDLT) on the health-related quality of life (HRQOL) of donors with the Short Form 36 health survey and to determine the risk factors for poor outcomes. Between June 1990 and June 2004, LDLT was performed 1000 times at Kyoto University Hospital. In July 2005, 997 of the 1000 donors were contacted by mail so that data on their HRQOL could be collected. In all, 578 donors responded (ie, there was a 58.0% response rate). The norm-based HRQOL scores for donors were better than the scores for Japanese norms across all time periods. All scores were similar for left lobe donors (n = 367) and right lobe donors (n = 211). For all donors, a multivariate logistic regression analysis revealed that age, the number of months until recovery to the preoperative health status, hospital visits due to donation-related symptoms, rest from work related to donation in the past month, and the existence of 2 or more comorbidities were significantly associated with decreased HRQOL scores. Postoperative complications and recipient mortality were not predictors of poor HRQOL. In conclusion, HRQOL was better for both right lobe donors and left lobe donors versus the Japanese norm population in the long term (mean postdonation period = 6.8 years). However, the prolongation of symptoms or sequelae related to donation lowered mental health or social functioning. The emergence of comorbidities after donation also significantly affected HRQOL in the long term. Careful follow-up and sustained counseling are required for donors with risk factors for lower HRQOL. Liver Transpl 18:1343–1352, 2012. © 2012 AASLD.

See Editorial on Page 1272

Living donor liver transplantation (LDLT) has been accepted worldwide as an alternative method for alleviating the shortage of donor organs from deceased donors. In Japan especially, LDLT has been the main treatment for end-stage liver disease, with more than 5700 LDLT procedures performed by the end of 2008.1 However, the procedure is associated with risks to the donors. The issues of postoperative morbidity and mortality for LDLT donors have been well characterized.2, 3

For healthy living donors, no medical benefit is derived from LDLT; the potential endpoint of donation is primarily psychological.4 Therefore, the transplant community has recognized the importance of understanding donor quality of life (QOL) after LDLT.5-17 According to prospective, longitudinal studies, donors typically experience decrements in physical well-being in the immediate postoperative period, but they tend to recover to baseline levels within a year. Their mental well-being appears to remain comparable to that of normative populations during the same time period.8, 9, 14 However, not all donors will follow such a general QOL pattern. Some reports have indicated that there are possible factors leading to poor donor outcomes, including the graft type (left or right lobe), donor age, urgency of the recipient's indication for transplantation, postoperative donor complications, and posttransplant recipient morbidity and mortality.16-18

At Kyoto University, the LDLT program was started for pediatric patients in 1990. With excellent graft and patient survival and proven donor safety, the application of LDLT was expanded from pediatric patients to adults. After the introduction of right lobe grafts for adult recipients in 1998,19 the number of LDLT cases increased and reached a cumulative total of more than 1000 by 2004.20 Recently, we investigated surgery-related complications in 1262 living liver donors and found that the incidence of major complications (Clavien classification grade III or greater) was 2.6% for left lobe donors and 17.0% for right lobe donors.3 At our institution (one of the high-volume transplant centers in Asia), we are aware of the need and responsibility to report the long-term QOL of donors.

In order to develop effective guidelines for donor selection and enhance postdonation follow-up and care, it is essential to identify the risk factors for poor donor QOL and to provide donor candidates with information about potential postdonation QOL difficulties as documented by rigorous studies. Unfortunately, previous studies have not successfully addressed predictors of poor outcomes, probably because of the small sample sizes and the low statistical power. Therefore, the issue of long-term outcomes remains an open question.18 The aims of this study were to investigate the long-term effects of LDLT on donor QOL with the Short Form 36 (SF-36) health survey and to determine the risk factors for poor outcomes in a large cohort.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Study Design

This was a historical cohort study. We longitudinally followed a group of donors after LDLT. A mailed survey with a cross-sectional design was used to collect data from donors concerning their health-related quality of life (HRQOL). We also analyzed the association of donor HRQOL with factors that might have predicted HRQOL. The study protocol was approved by the medical ethics committee of Kyoto University, and the study was performed in accordance with the ethical standards established by the 1975 Declaration of Helsinki.

Participants and Setting

Between June 1990 and June 2004, LDLT was performed 1000 times at Kyoto University Hospital. The donor selection criteria have been described previously.19 In July 2005, 997 of the 1000 donors were contacted by mail. The 3 donors who were excluded were lost to follow-up and included 1 death.3 The mailing included a cover letter explaining the study objectives, a consent form, and a written questionnaire. The letter informed each donor that the survey was being conducted for the purpose of research and that the donor's privacy would be protected. The letter also stipulated explicitly that the investigators would confirm the donor's intention to participate in the study and complete the questionnaire, which included an HRQOL scale and sociodemographic information.

Measures

We evaluated the HRQOL of living liver donors with the SF-36 version 2 questionnaire.21-23 The SF-36 version 2 questionnaire comprises 36 questions that are scored with 8 subscales: Physical Functioning (PF), Role–Physical (RP; ie, limitations on role functioning due to physical health), Bodily Pain (BP), General Health (GH), Vitality (VT), Social Functioning (SF), Role–Emotional (RE; ie, limitations on role functioning due to emotional problems), and Mental Health (MH). The 8 scales are summarized by 3 component summary scores: the Physical Component Score (PCS), the Mental Component Score (MCS), and the Role/Social Component Score (RCS).24 To allow for easy comparability, a norm-based scoring method was used to report the SF-36 results. The raw scores were transformed to norm-based scoring scores for each subscale and summary scale, with a mean of 50 and a standard deviation of 10 in the general Japanese population.

The questionnaire was used to obtain donor-reported sociodemographic information. Responders reported the number of months until recovery to their preoperative state, the number of months until their return to work, and any hospital visits (including visits to hospitals other than our university hospital) due to potentially donation-related problems after donation (eg, intestinal obstructions, incisional hernias, gastroduodenal ulcers, cholangitis, wound pain, diarrhea, constipation, jaundice, general malaise, and fever). Additionally, responders reported how many days of rest that they took from work for donation-related causes (eg, periodic hospital visits and rest due to donation-related symptoms) in the past month. The questionnaire also asked whether the donors currently had any of the following 22 comorbidities: hypertension; diabetes mellitus; cerebrovascular disease; myocardial infarction; hyperlipidemia; angina pectoris; heart failure; anemia; eye, pulmonary, gastrointestinal, renal, or genitourinary disease; arthritis or rheumatism; dermatological, neurological, psychiatric, endocrine, gynecological, or pancreatic disease; and cancer. With respect to comorbidities, the donors were classified as having none, only 1, or 2 or more.

Medical variables were obtained from the database of LDLT donors at Kyoto University Hospital. The medical variables included sex, birth date, height, weight, date of donation, hospital length of stay, graft type, postoperative complications, rehospitalization due to complications, relationship with the recipient (parent, sibling, child, spouse, or other), and recipient status (living or deceased). The body mass index (BMI) was calculated from the height and weight of each donor. There were 2 groups of grafts: the right-side graft (RG) group, which included right lobe grafts with or without the middle hepatic vein, and the left-side graft (LG) group, which included lateral segments, monosegments, left lobes, and extended left lobes with middle hepatic vein grafts. The severity of complications was graded with the Clavien classification system.3, 25

Statistical Analysis

We described and compared the characteristics of survey responders and nonresponders with the Student t test for numeric variables and the chi-square test for categorical variables. Next, we described the characteristics of the responders and compared the characteristics of LG donors and RG donors.

To assess the long-term HRQOL of donors, we transformed the 8 SF-36 subscale scores and 3 summary scores to norm-based scores and stratified them by the year of donation. The scores for each year of donation were compared with an analysis of variance.

To assess the factors predicting donor HRQOL after LDLT, we defined a score < 40 for each SF-36 summary score as an event of HRQOL deterioration. First, we calculated the crude odds ratio (OR) for each variable with respect to the 3 domains of HRQOL deterioration (physical health, mental health, and role/social health). Then, through the incorporation of variables whose P values for crude ORs were < 0.20, a multiple logistic regression analysis was conducted with the events of the 3 HRQOL domains as outcome variables.

Additionally, the frequency of current comorbidities and their effects on HRQOL were compared between the donors and the general Japanese population. The chi-square test was used for the comparison of the frequency of comorbidities, and the t test was used for the comparison of HRQOL scores. Japanese population data were obtained from SF-36 Japanese norm data studied in 2002.26 The mean age of the Japanese population was 51.1 ± 15.9 years.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Characteristics of the Responders

Five hundred seventy-eight of the 997 donors who were sent the questionnaire responded, and 419 donors did not respond. The proportion of responders was 58.0% (Table 1). The nonresponders included 3 who had bad health, 7 who felt a bit unmotivated, 4 who had other reasons, and 405 whose reasons were unknown. Among the responders (n = 578), the proportion of females was higher than the proportion of males (53.8% versus 46.2%). The mean age of the responders was higher than the mean age of the nonresponders at the time of response (45.6 versus 42.0 years) and at the time of donation (38.8 versus 35.1 years). Between the responders and the nonresponders, there were also significant differences in the relationship with the recipient (parents and spouses were more common for the responders), in the incidence of postoperative grade 2 or higher complications (7.9% versus 12.7%), and in the recipient mortality rate (19.4% versus 36.8%). The responders and the nonresponders did not differ with respect to the number of years after donation, BMI, graft type, or rate of rehospitalization.

Table 1. Donor Characteristics: Comparison of Study Questionnaires From Responders and Nonresponders
CharacteristicResponders (n = 578)Nonresponders (n = 419)P Value
  • *

    The data are presented as means and standard deviations.

Sex [n (%)]  0.03
 Male267 (46.2)224 (53.5) 
 Female311 (53.8)195 (46.5) 
Age at time of response (years)*45.6 ± 10.142.0 ± 9.7<0.001
Age at time of donation (years)*38.8 ± 10.035.1 ± 9.6<0.001
Years after donation*6.8 ± 3.46.9 ± 3.70.52
BMI before donation (kg/m2)*22.2 ± 3.022.6 ± 3.30.06
Year of donation [n (%)]  0.17
 19905 (0.9)2 (0.5) 
 199112 (2.1)9 (2.1) 
 199212 (2.1)10 (2.4) 
 199314 (2.4)19 (4.5) 
 199431 (5.4)24 (5.7) 
 199531 (5.4)33 (7.9) 
 199640 (6.9)24 (5.7) 
 199744 (7.6)30 (7.2) 
 199852 (9.0)40 (9.5) 
 199951 (8.8)41 (9.8) 
 200076 (13.1)35 (8.4) 
 200178 (13.5)41 (9.8) 
 200260 (10.4)43 (10.3) 
 200351 (8.8)42 (10.0) 
 200421 (3.6)26 (6.2) 
Graft type [n (%)]  0.38
 Left367 (63.5)278 (66.3) 
 Right211 (36.5)141 (33.7) 
Relationship between donor and recipient [n (%)]  0.001
 Parent406 (70.2)279 (66.6) 
 Sibling48 (8.3)47 (11.2) 
 Child43 (7.4)51 (12.2) 
 Spouse71 (12.3)28 (6.7) 
 Other9 (1.6)13 (3.1) 
 Unknown1 (0.2)1 (0.2) 
Postoperative complications [n (%)]  0.02
 None415 (71.8)304 (72.6) 
 Grade 1117 (20.2)62 (14.8) 
 Grade 217 (2.9)22 (5.3) 
 Grade 3 or higher29 (5.0)31 (7.4) 
Rehospitalization [n (%)]  0.34
 No551 (95.3)405 (96.7) 
 Yes27 (4.7)14 (3.3) 
Recipient status [n (%)]  <0.001
 Living466 (80.6)265 (63.2) 
 Deceased112 (19.4)154 (36.8) 
Current comorbidities [n (%)]  
 0297 (51.4) 
 1143 (24.7) 
 ≥2138 (23.9) 

Long-Term Donor HRQOL

The norm-based HRQOL scores for donors according to the year of donation are shown in Fig. 1. Figure 1 shows that donors after LDLT had better HRQOL scores than the Japanese norms across all time periods. In particular, the physical domain scores (PF from the 8 subscales and PCS from the component summary scores) were better than the other scales. On the other hand, the RCS values (from the component summary scores) for the donors were almost comparable to and sometimes lower than the RCS norms. When the donor HRQOL scores were stratified by the severity of postoperative complications (none, grade 1, grade 2, or grade 3 or higher), there were no significant differences in the 3 summary scores: the PCS values were 54.8 for donors with no complications, 54.8 for donors with grade 1 complications, 57.1 for donors with grade 2 complications, and 54.3 for donors with grade 3 or higher complications. The MCS scores were 51.2, 52.6, 47.7, and 52.3, respectively, and the RCS scores were 49.8, 50.2, 48.4, and 51.0, respectively.

thumbnail image

Figure 1. Norm-based HRQOL scores for donors in each year of donation. After LDLT, donors had better HRQOL scores than Japanese norms across all time periods.

Download figure to PowerPoint

Comparison of LG Donors and RG Donors

Among the responders, there were 367 donors in the LG group and 211 donors in the RG group (Table 2). All but one of the donors who underwent LDLT between 1990 and 1997 belonged to the LG group. In 1998, right lobe LDLT was started, and its frequency increased. In comparison with LG donors, the RG donors were significantly older, included a higher proportion of donors who were not parents, and had longer hospital stays, higher rates and more severe grades of postoperative complications, a higher incidence of rehospitalization, and a higher recipient mortality rate. However, norm-based HRQOL scores such as PCS, MCS, and RCS were similar in the 2 groups.

Table 2. Comparison of Donor Characteristics and HRQOL in LG and RG Donors
 Total (n = 578)LG Donors (n = 367)RG Donors (n = 211)P Value
  • *

    The data are presented as means and standard deviations.

Year of donation [n (%)]   
 19905 (0.9)4 (1.1)1 (0.5) 
 199112 (2.1)12 (3.3)0 (0.0) 
 199212 (2.1)12 (3.3)0 (0.0) 
 199314 (2.4)14 (3.8)0 (0.0) 
 199431 (5.4)31 (8.4)0 (0.0) 
 199531 (5.4)31 (8.4)0 (0.0) 
 199640 (6.9)40 (10.9)0 (0.0) 
 199744 (7.6)44 (12.0)0 (0.0) 
 199852 (9.0)38 (10.4)14 (6.6) 
 199951 (8.8)25 (6.8)26 (12.3) 
 200076 (13.1)29 (7.9)47 (22.3) 
 200178 (13.5)28 (7.6)50 (23.7) 
 200260 (10.4)23 (6.3)37 (17.5) 
 200351 (8.8)28 (7.6)23 (10.9) 
 200421 (3.6)8 (2.2)13 (6.2) 
Clinical data    
 Female sex [n (%)]311 (53.8)202 (55.0)109 (51.7)0.44
 Age at time of response (years)*45.6 ± 10.144.0 ± 9.048.4 ± 11.2<0.001
 Relationship between donor and recipient [n (%)]   <0.001
  Parent406 (70.2)343 (93.5)63 (29.9) 
  Sibling48 (8.3)6 (1.6)42 (19.9) 
  Child43 (7.4)5 (1.4)38 (18.0) 
  Spouse71 (12.3)6 (1.6)65 (30.8) 
  Other9 (1.6)7 (1.9)2 (0.9) 
  Unknown1 (0.2)0 (0.0)1 (0.5) 
 Hospital stay (days)*14.7 ± 6.313.5 ± 4.416.8 ± 8.4<0.001
 Postoperative complications [n (%)]   <0.001
  None415 (71.8)295 (80.4)120 (56.9) 
  Grade 1117 (20.2)61 (16.6)56 (26.5) 
  Grade 217 (2.9)6 (1.6)11 (5.2) 
  Grade 3 or higher29 (5.0)5 (1.4)24 (11.4) 
 Rehospitalization [n (%)]27 (4.7)8 (2.2)19 (9.0)0.002
 Recipient death [n (%)]112 (19.4)58 (15.8)54 (25.6)0.006
Donor responses to study questionnaire    
 Months until recovery to preoperative health status*3.5 ± 6.23.5 ± 7.03.6 ± 4.50.80
 Months until return to work*4.0 ± 7.44.0 ± 8.84.1 ± 4.90.90
 Hospital visits due to donation-related symptoms [n (%)]129 (22.3)78 (21.3)51 (24.2)0.47
 Rest from work related to donation in past month (days)*0.5 ± 2.20.5 ± 2.30.4 ± 1.90.81
 Current comorbidities [n (%)]   0.45
  0297 (51.4)188 (51.2)109 (51.7) 
  1143 (24.7)86 (23.4)57 (27.0) 
  ≥2138 (23.9)93 (25.3)45 (21.3) 
 HRQOL*    
  PCS54.9 ± 7.355.0 ± 7.454.6 ± 6.90.50
  MCS51.5 ± 9.951.1 ± 10.352.1 ± 9.10.22
  RCS49.9 ± 9.450.0 ± 9.749.9 ± 8.80.97

Factors Predicting Donor HRQOL After LDLT

The results for the crude ORs (Table 3) showed that age was significantly associated with PCS. The number of months until recovery to the preoperative health status and hospital visits due to donation-related symptoms were associated with RCS. Rest from work related to donation in the past month was associated with MCS. Importantly, the existence of 2 or more current comorbidities was significantly associated with all 3 HRQOL domains (PCS, MCS, and RCS). On the other hand, the occurrence or severity of postoperative complications, rehospitalization, and recipient death did not affect HRQOL.

Table 3. Crude ORs for Factors Predicting Donor HRQOL After LDLT
 PCS (<40)MCS (<40)RCS (<40)
ORP ValueORP ValueORP Value
Clinical data      
 Female sex1.750.231.330.271.470.15
 Age per 10 years2.120.0020.940.620.970.83
 Relationship between donor and recipient: not parental1.400.521.400.261.170.59
 Hospital stay (days)0.980.650.980.420.980.38
 Postoperative complications      
  None1.001.001.00
  Grade 11.560.381.040.890.940.85
  Grade 22.300.161.560.50
  Grade 3 or higher1.040.971.240.700.270.20
 Rehospitalization2.190.310.550.430.590.48
 Recipient death1.340.581.180.601.160.65
Donor responses to study questionnaire      
 Months until recovery to preoperative health status1.030.221.000.841.040.04
 Months until return to work1.010.821.020.371.010.42
 Hospital visits due to donation-related symptoms1.100.851.490.173.05<0.001
 Rest from work related to donation in past month0.410.392.290.021.980.06
 Current comorbidities      
  0      
  12.090.371.430.291.770.10
  ≥211.96<0.0013.17<0.0013.35<0.001

The multivariate logistic regression model revealed that age was significantly associated with PCS (adjusted OR = 1.84, P = 0.02; Table 4). The number of months until a donor's recovery to his or her preoperative health status was associated with both PCS (OR = 1.05, P = 0.03) and RCS (OR = 1.04, P = 0.02). Hospital visits due to donation-related symptoms were associated with RCS (OR = 2.54, P = 0.003). Rest from work related to donation in the past month was associated with MCS (OR = 2.24, P = 0.03). Finally, the existence of 2 or more comorbidities was significantly associated with PCS (adjusted OR = 8.89, P = 0.002), MCS (adjusted OR =2.65, P = 0.005), and RCS (adjusted OR = 2.77, P = 0.007).

Table 4. Adjusted ORs for Factors Predicting Donor HRQOL After LDLT: Multiple Logistic Regression Analysis
 PCS (<40)MCS (<40)RCS (<40)
ORP ValueORP ValueORP Value
Female sex1.990.180.980.951.120.72
Age per 10 years1.840.020.850.250.860.31
Months until recovery to preoperative health status1.050.031.000.961.040.02
Hospital visits due to donation-related symptoms0.870.801.180.612.540.003
Rest from work related to donation in past month0.400.382.240.032.070.07
Current comorbidities      
 01.001.001.00
 11.690.541.420.351.910.09
 ≥28.890.0022.650.0052.770.007

Effects of Comorbidities on HRQOL

The relationship between comorbidities and HRQOL is summarized in Table 5. Although the donors were slightly younger than the general Japanese population (45.6 ± 10.1 versus 51.1 ± 15.9 years), the frequencies of comorbidities were similar in the 2 groups. When they were stratified by the number of comorbidities, donors with 2 or more comorbidities showed norm-based scores lower than 50 for GH, VT, SF, RE, and MH (among the 8 SF-36 version 2 subscale scores). However, in comparison with the general Japanese population with 2 or more comorbidities, the donors demonstrated significantly better HRQOL scores for 6 of the 8 subscales and comparable scores for the other 2 subscales.

Table 5. Comorbidities and HRQOL: Comparison of the Donors and the Japanese Population in 2002
 Comorbidities (n)DonorsJapanese PopulationP Value
  • *

    The data are presented as means and standard deviations.

Patients [n (%)]0297 (51.4)1544 (52.1)0.90
1143 (24.7)740 (24.9) 
≥2138 (23.9)682 (23.0) 
PF*055.9 ± 4.652.0 ± 9.4<0.001
155.4 ± 4.548.4 ± 12.1<0.001
≥252.4 ± 6.542.7 ± 14.9<0.001
RP*053.5 ± 7.151.7 ± 8.8<0.001
152.7 ± 7.348.9 ± 11.2<0.001
≥250.3 ± 10.044.5 ± 13.8<0.001
BP*056.0 ± 8.251.7 ± 9.6<0.001
154.7 ± 8.749.1 ± 10<0.001
≥250.9 ± 11.345.7 ± 10.5<0.001
GH*055.0 ± 9.352.7 ± 9.3<0.001
153.1 ± 9.248.7 ± 9.5<0.001
≥247.5 ± 10.243.9 ± 10.5<0.001
VT*052.9 ± 9.251.7 ± 9.60.03
152.4 ± 9.149.4 ± 9.90.001
≥248.8 ± 10.547.4 ± 10.90.13
SF*053.3 ± 7.851.5 ± 8.7<0.001
151.8 ± 9.049.2 ± 10.4<0.001
≥249.2 ± 11.346.8 ± 12.40.01
RE*053.7 ± 7.051.5 ± 9.0<0.001
151.7 ± 8.948.9 ± 11<0.001
≥249.3 ± 11.146.2 ± 12.50.001
MH*052.2 ± 9.151.5 ± 9.40.17
152.0 ± 9.149.6 ± 10.10.002
≥247.5 ± 10.547.8 ± 10.80.74

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

At the beginning of the LDLT series at our institution, most cases were pediatric recipients who received left lateral segments or left lobe grafts donated by young parents. After the number of adult recipients increased, the proportion of right lobe donors increased, and donors with a diversity of relationships were selected (eg, parents, children, siblings, and spouses). In this cross-sectional study, there were 578 responders to the questionnaire and 419 nonresponders. The response rate (58.0%) seemed to be comparable to the response rates (62.5%-92%) of 7 previous studies.5-7, 13, 15-17 Although answers to the questionnaire were obtained from donors with a variety of characteristics and different postdonation times, particular care should be taken with respect to factors that might differentiate responders from nonresponders.12, 18 The percentage of females and the donor ages at the time of the questionnaire and at the time of donation were significantly higher for the responders (Table 1). The donor-recipient relationships and the severity of postoperative complications differed slightly but significantly between the 2 groups. In addition, recipient mortality was significantly higher for the nonresponders versus the responders in this study. It has already been reported by some authors15, 16 that response rates are lower among donors whose recipients have died. Although the reasons for not responding were not completely clear, it is possible that the responders were not representative of all donors. This point is one of the limitations of the present study. Nevertheless, 112 donors whose recipients died still responded to the survey.

In the present study, the donors' HRQOL scores were stratified by the year of donation and compared to those of Japanese norm populations (Fig. 1). As a result, the donor HRQOL was better than the norm HRQOL (scores > 50) across all time periods: PCS was significantly higher, and MCS and RCS were equivalent to or slightly higher than the norm values, regardless of the postdonation time. These findings are consistent with previous studies5-7, 9 and suggest that, on the whole, donation for LDLT did not negatively affect the HRQOL of these donors. On the other hand, it should be noted that LDLT donors often have higher QOL scores than normative populations even before donation.8, 9 Because this was a cross-sectional study of a single time point, changes in QOL before and after the donor operation could not be evaluated in this study. However, it was implied that in most cases, this better QOL had persisted for years, and the long-term outcomes of donors (ie, >5-10 years after donation) were comparable to the outcomes of donors shortly after donation.

Although many previous studies have focused on right lobe donors,5-12 1 US study15 and 2 Japanese studies14, 16 compared QOL in RG donors and LG donors (including left lateral segment donors). These studies found no significant differences in HRQOL scores between RG donors and LG donors despite the longer incision and larger resected liver mass in right lobe donation. In the present study, the perioperative course (including the postoperative hospital stay, postoperative complications, and rehospitalization) was significantly worse for RG donors versus LG donors (Table 2). We had hypothesized that these medical conditions would more adversely affect the physical and mental well-being of RG donors and that it would become evident in studies with larger sample sizes. In addition, as shown in Table 2, the majority of the LG donors were young parents donating for their children. It was supposed that preoperative motivations and emotional coping for donation were probably different for donors with other types of relationships with their recipients, and this led to particular effects on postdonation QOL. Interestingly, however, there were no differences in the recovery time or the current PCS, MCS, and RCS values between the RG donors and LG donors (Table 2). Furthermore, it was also clarified that the donor-recipient relationship (parental or not parental) did not affect any of the HRQOL scores (Table 3).

Because there were some donors whose QOL scores were lower than those of normal populations, we investigated the risk factors or predictors of poor outcomes from several angles (Table 3). Some studies have demonstrated that donors whose recipients had major complications scored significantly lower on the MH scale than donors with recipients without major complications.6, 12 However, other studies13, 14 have reported contradictory findings. The present study did not find recipient mortality to be a risk factor for lower QOL scores. As for donor complications, Erim et al.17 indicated that donors with severe complications demonstrated a negative correlation with physical and psychiatric QOL. However, in the present study, as in some previous studies,13, 14 the occurrence and severity of postoperative complications as recorded in our database were unrelated to long-term QOL. On the other hand, donors who reported repeated outpatient clinic visits or a current need for rest from work due to ongoing symptoms or prolonged medical complications potentially related to donation showed lower RCS or MCS scores. These results implied that if donors were distressed by the sequelae of their complications for a long time, their postdonation QOL (especially mental or social QOL) significantly deteriorated. This is an important message and not an unanticipated finding that was captured through the present long-term follow-up study.

Another significant risk factor for reduced QOL was the occurrence of comorbidities that emerged after donation. It is likely that the incidence of these diseases (eg, hypertension, diabetes mellitus, and metabolic syndrome) increases as donors grow older. This is an important concern for potential donors who express concern for their own well-being because they are anxious about whether the donation procedure might increase their risk of contracting these diseases. Indeed, in this study, donors with multiple comorbidities showed significantly lower QOL scores than donors without comorbidities. However, the incidences of comorbidities were similar for the donors and the general population, and the QOL scores of the donors with comorbidities were better than the QOL scores of the general population with comorbidities. On the basis of these results, we can provide donor candidates with helpful information about the long-term effects of donation on their own health conditions. In Japan a few years ago, the highly publicized and sensationalized morbidity of living liver donors negatively affected LDLT. We hope that this study will alleviate the problem by showing positive results. At the same time, this report suggests that we need to improve the donor follow-up and care system to establish LDLT as an effective and acceptable treatment modality.

Because the safety of the donor operation and the rapid postoperative recovery of the donors are of utmost importance, from the beginning of our LDLT program, we have made it a rule to check the postoperative conditions of donors at least 3, 6, and 12 months after the operation, even in the absence of any complications. However, the present study suggests that some donors have lowered QOL for longer periods because of prolonged symptoms or psychological distress related to the donation. To improve the QOL of such donors, postdonation follow-up and care should be continued for years. In particular, it is plausible that some donors have prolonged psychological distress but are reluctant to reveal their anxiety and depression. We may need to make an effort to identify such donors and recommend them for psychiatric counseling. It is also important for transplant team doctors and coordinators to be concerned about donor comorbidities emerging after donation and, if it is necessary, to help donors to receive appropriate therapy. Furthermore, incentives to donors such as financial insurance coverage, health benefits, and absence-from-work benefits should be considered, although this is a difficult issue from a practical point of view.

In conclusion, the present study is the largest single-center report to date on the HRQOL of living liver donors (n = 578). Although this study has some limitations (as mentioned previously) and its results should be carefully evaluated, we have demonstrated that HRQOL is better for both right lobe donors and left lobe donors than the Japanese norm population in the long term (mean postdonation period = 6.8 years). The existence of surgery-related complications itself did not reduce HRQOL; rather, it was the prolongation of symptoms or sequelae related to donation that lowered mental health or social functioning. The emergence of comorbidities after donation also significantly affected HRQOL in the long term. Careful follow-up and ongoing supportive counseling are required for donors with such risk factors for lower HRQOL.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES