Telephone: 82-2-2072-3789; FAX: 82-2-766-3975
Article first published online: 30 MAY 2007
Copyright © 2007 American Association for the Study of Liver Diseases
Volume 13, Issue 6, pages 797–806, June 2007
How to Cite
Yi, N.-J., Suh, K.-S., Cho, J. Y., Lee, H. W., Cho, E.-H., Yang, S. H., Cho, Y. B. and Lee, K. U. (2007), Three-quarters of right liver donors experienced postoperative complications. Liver Transpl, 13: 797–806. doi: 10.1002/lt.21030
See Editorial on Page 781
Presented at the 11th Annual Congress of International Liver Transplantation Society, Los Angeles, CA, July 20–July 23, 2006.
- Issue published online: 30 MAY 2007
- Article first published online: 30 MAY 2007
- Manuscript Accepted: 28 SEP 2006
- Manuscript Received: 6 JAN 2006
A donor right hepatectomy (RH) is associated with a higher rate of morbidity than a left hepatectomy. Therefore, the precise morbidity should be known to improve the success of donor RH implementation. However, the rate of complication varies according to the individual definition of morbidity. This study prospectively analyzed the outcomes of 83 consecutive living donor RHs between January 2002 and July 2004 using a standardized classification of the severity of complications. The morbidity was classified using the modified Clavien system: grade I for minor complications; grade II for potentially life-threatening complications requiring pharmacological treatment; grade III for complications requiring invasive intervention; grade IV for complications causing organ dysfunction requiring intensive care unit management; and grade V complications resulting in the death of the patient. The donors were followed-up regularly for at least 12 months. No donor death or relaparotomy was noted. Overall, 65 out of 83 donors (78.3%) experienced postoperative complications: grades I, II, III, IV, and V complications in 64 (77.1%), 11 (13.3%), 1 (1.2%), 0, and 0 patients, respectively. The most common grade I complications were hyperbilirubinemia (n = 31) and pleural effusion (n = 31), and bile leakage in grade II (n = 7). The bilirubin and alanine aminotransferase levels were normal in 92.7% of donors at the 1-year follow-up. In conclusion, although most of these adverse events were minor and self-limited, 78% of right liver donors still experienced morbidity. Therefore, continuous standardized reporting of the donor morbidity as well as meticulous surgery and intensive care are essential for the success of donor RH implementation. Liver Transpl 13:797–806, 2007. © 2006 AASLD.
Since the first case of living donor liver transplantation (LDLT) was reported in 1989,1, 2 many transplantation centers have performed LDLT due to the shortage of cadaveric donors.3 LDLT is now common practice in many transplantation centers worldwide and achieves results comparable to those of deceased donor liver transplantation. However, the donor morbidity and size mismatch are a major obstacle to the expansion of LDLT to adult recipients.4 The transplantation of the right liver from a living donor was demonstrated to be technically feasible in the mid-1990s. A right liver graft has been used in many centers to meet the metabolic demands of large recipients. Despite the rapid implementation of this procedure, analysis of the outcome of the right liver donor is still incomplete.
There were extensive ethical discussions when right liver LDLT was first performed, because a right liver LDLT involves the most complicated and technically demanding surgical procedure. In the worldwide reports of LDLT, the rate of complication from a right liver donor was approximately 31%, ranging widely from 0% and 67%, depending on the individual definition and recognition of morbidity.5–8 The latest results on mortality from Europe, Asia, and the United States suggest that a donation of the right liver carries a high risk of mortality.6 The lack of a standardized assessment of the perioperative complications and a retrospective review is a serious limitation of the analysis of donor morbidity and mortality. Liu et al.8 reported a prospective evaluation of the live liver donor morbidity but did not use a uniform definition and recognition of these complications. Broering et al.5 analyzed live liver donor morbidity according to a modified systemic classification of the donor complications originally suggested by Clavien et al.9, 10 that was adapted to a living donor situation. However, their study was a retrospective historical review.
This study prospectively collected and analyzed 83 consecutive living donor right hepatectomy (RH) procedures since 2002 using the standardized classification of the severity of complications.
PATIENTS AND METHODS
Between January 2002 and July 2004, 83 consecutive donors underwent RH at the Seoul National University Hospital. The donors' mean age was 30.8 ± 8.8 years, and their mean body mass index was 23.5 ± 5.8 (kg/m2). The patients' offspring were the most common donors (39.6%) followed by other relatives (20.5%), siblings (19.3%), spouses (10.8%), and unrelated donors (9.6%). There was no ABO incompatibility in this study (Table 1).
|Total (n = 83)||RH Group (n = 57)||MERH Group (n = 26)||P Value|
|Age at donation (years)||30.8 ± 8.8||31.2 ± 8.7||30.4 ± 9.6||0.316|
|Body weight (kg)||65.9 ± 5.8||65.9 ± 12.0||65.7 ± 7.9||0.066|
|Body mass index (kg/m2)||23.5 ± 5.8||23.3 ± 2.9||23.8 ± 4.0||0.105|
|Relation to the recipient||0.712|
|Offspring||33 (39.6%)||22 (38.6%)||11 (42.3%)|
|Other relatives||17 (20.5%)||13 (22.9%)||4 (15.4%)|
|Sibling||16 (19.3%)||11 (19.3%)||5 (19.2%)|
|Spouses||9 (10.8%)||4 (7.0%)||5 (19.2%)|
|Unrelated donors||8 (9.6%)||7 (12.3%)||1 (3.8%)|
|Graft macrovesicular steatosis (%)||4.8 ± 2.9||4.8 ± 7.0||5.0 ± 7.1||0.833|
|<10%||74 (89.2%)||51 (89.5%)||23 (88.5%)|
|10–20%||5 (6.0%)||3 (5.3%)||2 (7.7%)|
|21–30%||4 (4.9%)||3 (5.3%)||1 (3.8%)|
|Remnant liver volume (%)||35.3 ± 5.9||34.9 ± 6.4||36.4 ± 4.3||0.201|
|Graft-vs.-recipient weight ratio (%)||1.28 ± 1.26||1.16 ± 0.34||1.53 ± 2.17||0.029|
|Operative time (min)||287.1 ± 42.4||292.5 ± 43.7||275.0 ± 37.5||0.490|
|Estimated blood loss (mL)||491.9 ± 225.2||440.7 ± 220.6||578.1 ± 213.7||0.732|
|Postoperative hospital stay (days)||13.0 ± 3.2||12.8 ± 3.4||13.5 ± 2.6||0.553|
|Retention of an abdominal drain (days)||9.7 ± 3.2||9.1 ± 1.9||10.8 ± 4.6||0.004|
Outcome analysis included a postoperative liver function test, computed tomography (CT) scans, and clinical data. Two surgeons and 1 physician assistant performed the clinical surveillance prospectively. The mean follow-up period was 21.5 ± 8.9 months.
Classification of Donor Morbidity After LDLT
The donor morbidity was classified according to the new version of the Clavien classification suggested by Dindo et al.11 and the donor morbidity classification adapted by Broering et al.5 Broering et al.5 adapted the original Clavien classification of complications designed for liver transplantation recipients7, 9, 10 to a living donor situation. In this classification, the morbidity associated with immunosuppressive therapy was disregarded. The complications that were graded low in the original classification of morbidity after liver transplantation were graded higher in donor morbidity classification.5 The new version of Clavien classification was also adapted and modified to a living donor situation. Donor complications were classified from grades I to V.5, 11 A grade I complication was defined as any deviation from the normal postoperative course without the need for intervention and any non-life-threatening complication that did not result in residual disability. A grade II complication was defined as any complication that required pharmacological treatment or more than 1 foreign blood unit but did not require invasive intervention and did not result in residual disability. Complications higher than grade II were classified as major potentially life-threatening complications. A grade III complication was defined as any complication requiring invasive intervention, and/or readmission to the intensive care unit but did not result in residual disability. A grade IV complication was defined as any complication leading to any lasting disability. A grade V complication was defined any complication resulting in the death of the patient. Table 2 provides detailed examples of complications at our institution according to this classification.12, 13
|I||Any deviation from the normal postoperative course without the need for intervention, and a non-life-threatening complication that does not result in residual disability and does not require therapeutic intervention or the use of drugs, except for antipyretic, analgesic, anti-inflammatory, or antiemetic drugs||Superficial wound infection treated without antibiotics Postoperative bleeding not requiring transfusion Transient bile leak (total bilirubin of a drain that is twice that in serum) that was treated conservatively and diminished within 1 week, or hyperbilirubinemia (total bilirubin >3mg/dL more than 3 days or exceeding 1.3 mg/dL 7 days after operation)|
|Pleural effusion, abdominal fluid collection treated conservatively, or an asymptomatic but significant finding on an imaging study|
|Postoperative gastrointestinal problem not requiring readmission|
|Postoperative voice change|
|II||Any complication that is potentially life threatening that requires pharmacological treatment or >1 foreign blood unit, but that does not require a therapeutic invasive intervention and does not result in residual disability||Bacterial, viral, or fungal infection requiring antibiotic, antiviral, or antifungal therapy Postoperative bleeding without a relaparotomy but requiring transfusion Transient postoperative psychological problem|
|Ileus requiring readmission after discharge|
|Hypersensitivity to dyes or drugs|
|Bile leak requiring reposition of the drain but not requiring endoscopic or surgical intervention|
|III||Any complication that is potentially life-threatening that requires therapeutic intervention, and/or leads to readmission to an intensive care unit but does not result in residual disability||Bile leak requiring transhepatic, endoscopic or surgical procedures Postoperative bleeding requiring relaparotomy Deep infection requiring relaparotomy or interventional drainage and antibiotic therapy|
|Complicated venous thrombosis with pulmonary embolism|
|Gastrointestinal bleeding treated endoscopically or surgically|
|IV||Any complication with residual or lasting disability||Progressive liver failure requiring liver transplantation|
|Renal failure requiring persistent hemodialysis or renal transplantation|
|De novo human immunodeficiency virus, hepatitis B, or hepatitis C infection|
|V||Death of a patient||Any disease leading to death|
Donor Selection Criteria
A donor surgeon informed the families considering LDLT of the medical risks and benefits of the procedure for both the donor and recipient. The ethical issues and potential psychological and social consequences were discussed with the donor, coordinator, medical social worker, and psychologist. Apart from informed consent, potential right liver donors needed to fulfill the following criteria: (1) a volunteer with an ability and willingness to comply with a long-term follow-up; (2) age between 16 and 55 years old; (3) blood group compatibility with the recipient; (4) a relationship between the recipient within the third degree of consanguinity or an intense emotional relationship; and (5) no known medical disorders that would significantly increase perioperative risks. Candidates fulfilling these criteria were evaluated according to the evaluation protocol for a living liver donor.4, 12, 13 This included an assessment of the general health and an accurate estimation of the liver volume, quality, and anatomy. The evaluation was divided into in 3 phases. During phase 1, the general condition of the potential donor was evaluated. The assessment included medical history, body size, psychosocial circumstance, liver function test, biochemistry, hematology, coagulation profile, urine analysis, serology for hepatitis A, B, and C, chest radiograph, and electrocardiography. During phase 2, the potential donor underwent a complete medical evaluation and imaging study for liver quality. The laboratory assessment included alpha-fetoprotein, carcinoembryonic antigen, and serology for cytomegalovirus, Epstein-Barr virus, Venereal Disease Research Laboratory (VDRL), and human immunodeficiency virus. Other examinations included an esophagogastroduodenoscopy for all, mammography and Pap smear for female donors >35 years old, and a sigmoidoscopy for donors >40 years old. A CT scan and magnetic resonance imaging were performed to evaluate the liver size and quality, and vascular and biliary anatomy. Phase 3 was devoted to invasive procedures required to investigate the potential problems discovered during phases 1 and 2. This included a liver biopsy, or additional consultations and procedures. A preoperative liver biopsy was taken from donors suspected of having steatosis from imaging studies in order to exclude those with severe hepatic steatosis. Conventional hepatic angiography and endoscopic retrograde cholangiography were not routinely performed.
The donors were advised to stop smoking and drinking. However, the presence of mild systemic diseases, such as well-controlled mild hypertension or diabetes, was not necessarily a contraindication.
A RH was considered if the estimated remnant liver volume was >30% of the whole liver and the liver biopsy showed <30% macrovesicular steatosis.12–14 A modified extended RH (MERH), first introduced by our group,14 that harvested the middle hepatic vein to the graft remaining segment 4 to the remnant liver side, was considered for donors whose remnant liver volume was estimated to be >35% of the whole liver as well as those free of steatosis. Donors with a graft-vs.-recipient weight ratio >0.8% were accepted. They were admitted 2 days before surgery and were finally interviewed by an anesthesiologist and donor surgeon. No autologous blood was preserved.
Donor Right Hepatectomy
The detailed surgical technique is described elsewhere.4, 13, 14 An intraoperative liver biopsy was done to exclude patients with moderate to severe steatosis and any pathology finding not found preoperatively. A transection plane was drawn after transient clamping of the hepatic artery and portal vein on the right liver. A liver parenchymal transection was carried out using an ultrasound aspirator and a bipolar coagulator without a vascular inflow occlusion to either side of the liver. Conventional RH and MERH were performed in 57 (68.7%) and in 26 (31.3%) cases, respectively. There were no differences between the 2 groups in terms of the mean donor age (P = 0.316), the gender ratio (P = 0.207), mean body weights (P = 0.066), mean body mass index (P = 0.105), and relationship to the recipients (P = 0.712). The preoperative estimated volume of the remnant liver was 34.9 ± 6.4% of the whole liver volume in the RH group and 36.4 ± 4.3% in the MERH group (P = 0.201) (Table 1).
All of the donors were admitted to the intensive care unit for monitoring at least overnight. The prophylactic antibiotics were stopped and the nasogastric tube and Foley catheter were removed one day after surgery. Low-molecular-weight heparin was not routinely used. The abdominal drain was removed if there were no abnormal findings by either a physical examination or biochemical analysis, and the amount of abdominal drainage per day was <150 mL. The alanine aminotransferase (ALT), total bilirubin, and prothrombin time were checked. Of the 83 donors, 76 (91.6%) underwent a protocol liver CT scan during postoperative day (POD) 7 to 14.13, 14 After discharge, the donors were followed up at 1, 4, and 12 months after donation, and liver function tests was performed for up to 1 year.
All the values are expressed as a mean ± SD. The categorical variables were compared using a Fisher exact test. The continuous variables were compared using a nonparametric Mann-Whitney U test. The statistical analyses were performed using SPSS (Release 10.0; SPSS Inc., Chicago, IL). P values <0.05 were considered significant.
Preoperative Morbidity Associated With Evaluation
A total of 94 patients fulfilled the essential donor criteria for suitable right liver donors. Of the 94 potential right liver donors, 11 (11.7%) were rejected for the following reasons: a more suitable potential donor (n = 4), medical problems found during the evaluation (n = 4), and refusal to undergo surgery (n = 3). The medical problems were cytotoxic antibody positivity between the donor and recipient, moderate degree steatosis, advanced gastric cancer, and multiple liver abscess. There was no evaluation morbidity or intraoperative abortion of the donor operation
The mean time from the donor's first visit to surgery was 36.0 ± 54.9 (range, 1–359) days. In emergency cases (n = 6), the median duration from inception to surgery was 7 (range, 1–26) days. The donor evaluation period was shortened in those emergency cases (P = 0.039).
The mean graft–vs.-recipient weight ratio was 1.28 ± 1.26%, and the mean donor remnant liver volume percentage was 35.3 ± 5.9%. The mean macrovesicular steatosis in the intraoperative liver biopsy was 4.8 ± 7.0%. The level of graft steatosis was <10% in 74 donors (89.2%). The mean operative time from the skin incision to closure was 287.1 ± 42.4 minutes. The mean estimated blood loss was 491.9 ± 225.2 mL (Table 1). No patient received any blood transfusions or underwent a relaparotomy. An autotransfusion device, Cell Saver 4 System (Haemonetics Corp., Braintree, MA), was not used.
There were no significant differences in the surgical data between the RH and MERH groups (Table 1) or between the elective and emergency procedures (P > 0.05).
Postoperative Liver Function and Recovery
The postoperative serum ALT (36.1 ± 19.2 IU/L) and total bilirubin (0.88 ± 0.38 mg/dL) levels evolved appropriately and returned to normal within 1 month after surgery. The prothrombin time returned to normal (81.5 ± 9.6%) within 1 week after surgery (Fig. 1).
Of the 83 donors, 6 (7.2%) had a persistently asymptomatic abnormal liver function test at the 12th month. Hyperbilirubinemia was noted in 4 cases, ranging from 1.5 to 2.2 mg/dL, but the direct bilirubin levels were normal. Among them, 2 had preoperative Gilbert syndrome. Two donors showed a mild elevation in the ALT level. A 20-year-old male donor whose ALT level was 99 IU/L 1 year after and 58 IU/L 2 years after donation had been consuming alcohol and suffered from a recurrent ileus. Counseling was arranged, but he refused treatment. The other donor was 38 years old, with an ALT level of 78 IU/L 1 year after donation. The ALT level improved to normal 2 years after surgery. The abdominal drain was retained for 9.7 ± 3.2 (range, 5–22) days postoperatively. The mean hospital stay was 13.0 ± 3.2 (range, 8–22) days.
There were no obvious differences between the RH and MERH groups in terms of their postoperative biochemical profiles (P > 0.05). The postoperative abdominal drainage volume on day 5 was larger in the MERH group (210.3 ± 151.2 mL) than in the RH group (109.1 ± 78.5 mL) (P = 0.007). In addition, the drain was retained for 9.1 ± 1.9 days in the RH group and for 10.8 ± 4.6 days in the MERH group (P = 0.004). However, the postoperative hospital stay was similar in the RH (12.8 ± 3.4 days) and MERH groups (13.5 ± 2.6 days) (P = 0.553). There were no significant differences in the postoperative liver function or patients recovery between the elective and emergency procedures (P > 0.05).
Postoperative Liver CT Scan
Of the 83 donors, 76 (91.6%) underwent a liver CT scan during POD 7 to 14, according to the routine protocol. A CT scan showed no postoperative abnormal findings in 50% of donors (n = 42, 55.3%). The others (n = 34, 44.7%) had the following problems: pleural effusion (n = 25), intra-abdominal fluid collection including biloma (n = 21), wound problem (n = 2), pneumonia (n = 1), and intra-abdominal abscess (n = 1).
Of the 34 donors, 20 (58.9%) had no symptoms associated with abnormal CT findings and did not require intervention. The other 14 donors (41.2%) had symptoms associated with abnormal CT findings. However, 7 of these 14 donors did not require further treatment. The other 7 required additional medication for fluid collection accompanying fever (n = 3), reposition of an abdominal drain against biloma (n = 2), and additional antibiotics against culture-proven pneumonia and an intra-abdominal abscess (n = 2) (Table 3).
|Complication||Number (n = 83)||Early Postoperative CT Finding (n = 76)||Management||1-Year Outcome|
|Hyperbilirubinemia||31 (37.4%)||Observation||Remained in 4 cases|
|Pleural effusion||31 (37.4%)||25 (32.9%)||Observation||Disappeared in all|
|Fluid collection||17 (20.5%)||17 (22.4%)||Medication (n = 3)||Remained in 2 cases|
|Wound problem||7 (8.4%)||2 (2.6%)||Observation||Cured|
|Transient bile leakage||4 (4.8%)||Observation||Cured|
|Transient bleeding on drain||3 (3.6%)||Observation||Cured|
|Bowel habit change||2 (2.4%)||Medication (n = 2)||Cured|
|Voice change||2 (2.4%)||Observation||Cured|
|Bile leakage||7 (8.4%)||4 (5.3%)||Drain reposition (n = 2)||Cured|
|Pneumonia||2 (2.4%)||1 (1.3%)||Antibiotics (n = 2)||Cured|
|Ileus||2 (2.4%)||Conservative management||Recurrent symptoms in 1 case|
|Psychological problem||2 (2.4%)||Interview with psychologist (n = 2)||Cured|
|Hypersensitivity to dye or drug||2 (2.4%)||Conservative management||Cured|
|Intra-abdominal abscess||1 (1.2%)||1 (1.3%)||Antibiotics||Cured|
|Sepsis due to bile leakage||1 (1.2%)||Readmission to intensive care unit||Cured|
|Total||114 cases among 65 donors (78.3%)||50 cases among 34 donors (44.7%)|
Of these 34 donors with abnormal CT findings, 23 (67.6%) underwent a follow-up CT scan 4 months after surgery. Most of the abnormal CT findings had either resolved (n = 19) or improved (n = 2). The other 2 showed no interval changes in fluid collection at the liver resection site; 1 case of fluid collection was improved on the follow-up CT scan taken 1 year after surgery, and the other had no clinical symptoms and did not undergo a follow-up CT scan.
No donor died or suffered from any life-long complications during the study period. Overall, 65 (78.3%) of the 83 consecutive donors experienced complications. There were 97 cases of grade I complication among 64 (77.1%) donors. Twelve donors (14.5%) had major complications classified as either grade II (16 cases in 11 donors; 13.3%) or grade III (1 case; 1.2%). However, there were no grade IV or V complications. Table 3 gives a complete list of the observed morbidity in the whole series. All complications except for 3 occurred during the 30-day perioperative period: Two donors showed frequent loose stools for 4 and 6 months and were was treated with medication, and the other donor was readmitted due to an adhesive ileus 14 months after donation.
There were no differences in the incidences of donor complications according to age (P = 0.817), gender (P = 1.000), emergency surgery (P = 0.074), types of hepatectomy (P = 0.489), operative time (P = 0.982), estimated blood loss (P = 0.872), graft steatosis (P = 0.972), remnant liver volume (P = 0.993), or follow-up duration (P = 0.970).
The most frequent grade I complications were hyperbilirubinemia (n = 31; 37.4%) and pleural effusion (n = 31; 37.4%) followed by intra-abdominal fluid collection (n = 17; 20.5%). Superficial wound problems including a keloid (n = 7; 8.4%), transient bile leakage treated conservatively within 1 week (n = 4; 4.8%), transient bleeding on drain that improved without transfusion (n = 3; 3.6%), and changes in bowel habits (n = 2; 2.4%) and voice (n = 2; 2.4%) were observed.
The most common grade II complications were minor bile leakage (n = 7, 8.4%), which lasted longer than 1 week but resolved spontaneously, followed by pneumonia (n = 2; 2.4%), adhesive ileus (n = 2; 2.4%), psychiatric problems such as postoperative depression and delirium (n = 2; 2.4%), and drug hypersensitivity (n = 2; 2.4%). Another grade II complication was an abscess of an unknown cause in the pelvic cavity in a 43-year-old male patient. His immediate postoperative recovery was uneventful, but he developed fever and mild abdominal discomfort on POD 5. A CT scan performed on POD 7 revealed a pelvic abscess. He was treated with antibiotics.
There was 1 case of a grade III complication: sepsis associated with bile leakage. This 17-year-old male donor had biliary anatomical variation. The right posterior duct drained to the left hepatic duct, but the donor hepatectomy was uneventful. During the early postoperative period, bile leakage was noted on the drain but resolved spontaneously (Fig. 2A). The patient was discharged on POD 17 but visited the emergency room on POD 19 due to fever and hypotension caused by gram-negative sepsis (Fig. 2B). He was readmitted to the intensive care unit and was managed conservatively. He recovered his general health with normal liver function. Figure 2C shows the follow-up CT scan.
There were no significant differences in the types and incidences of morbidities between the RH and MERH groups (Table 4). Morbidity developed in 43 out of 57 (75.4%) donors in the RH group and in 22 out of 26 (84.6%) donors in the MERH group (P = 0.489). In the RH group, 57 cases of grade I complications developed among 42 donors (73.6%), and 13 cases of grade II complications developed among 9 donors (15.8%). In the MERH group, 40 cases of grade I complications developed among 22 donors (84.6%), 3 cases of grade II complications developed among 3 donors (11.5%), and 1 case of grade III complication developed in one donor (3.8%).
|Complication||RH group (n = 57)||MERH group (n = 26)|
|Grade I||57 cases among 42 donors||40 cases among 22 donors|
|Hyperbilirubinemia (n = 31)||21 (36.8%)||10 (38.5%)|
|Pleura effusion (n = 31)||21 (36.8%)||10 (38.5%)|
|Fluid collection (n = 17)||7 (12.3%)||10 (38.5%)|
|Wound problem (n = 7)||4 (7.0%)||3 (11.5%)|
|Transient bile leakage (n = 4)||2 (3.5%)||2 (7.7%)|
|Transient bleeding on drain (n = 3)||1 (1.8%)||2 (7.7%)|
|Bowel habit change (n = 2)||–||2 (7.7%)|
|Voice change (n = 2)||1 (1.8%)||1 (3.8%)|
|Grade II||13 cases among 9 donors||3 cases among 3 donors|
|Bile leakage (n = 7)||6 (10.5%)||1 (3.8%)|
|Pneumonia (n = 2)||2 (3.5%)||–|
|Ileus (n = 2)||2 (3.5%)||–|
|Psychological problem (n = 2)||1 (1.8%)||1 (3.8%)|
|Hypersensitivity to dye or drug (n = 2)||1 (1.8%)||1 (3.8%)|
|Intra-abdominal abscess (n = 1)||1 (1.8%)||–|
|Grade III||–||1 donor|
|Sepsis due to bile leakage (n = 1)||–||1 (3.8%)|
|Total||70 cases among 43 donors||44 cases among 22 donors|
Blood products were transfused in only 2 cases. In the first case, 1 pint of fresh-frozen plasma was inadvertently transfused by an on-call doctor 1 day after hepatectomy; in the second case, several pints of fresh-frozen plasma and platelet concentrate were transfused to correct disseminated intravascular coagulation, which was previously described as a grade III complication.
The reported incidence of right liver donor complications showed a marked disparity ranging from 0 to 67%.4–7, 10, 15–22 This disparity might be related to the different definitions of “complication” and the limitations associated with a retrospective review based on incomplete medical records. A reliable and accurate assessment of morbidity is an important task for any transplantation center, because the safety of the donor is the most important issue in LDLT. Therefore, potential live donors must undergo a complete preoperative medical evaluation. The donor morbidity must be collected prospectively, and any abnormal finding constituting a deviation from a normal postoperative course must be counted as morbidity.9, 11 In addition, the well-standardized classification of morbidity9–11 must be adapted to the requirements of the living donor, as reported by Broering et al.5 Therefore, both the new version of Clavien standardized negative outcome of liver surgery suggested by Dindo et al.11 and the modified Clavien classification for live liver donor morbidity suggested by Broering et al.5 was adapted and tuned up in this study, and donor morbidity was prospectively analyzed (Table 2).
The potential risk for injury or death necessitates a complete preoperative medical and anatomical evaluation of potential donors. Therefore, the donor evaluation protocol must include a general health examination of the donor as well as the liver volume and quality. Our donor evaluation protocol was designed to maximize donor safety and included a national guide cancer screening test of Korea, a gastroscopy, and a breast examination including mammography for those >40 years old, a Pap smear for those >30 years, and a colonoscopy for those >50 years. A single case of malignancy was encountered among the 27 donors older than 40 in this study. It was reported that between 15% and 45% of potential donors who present for an evaluation might be suitable candidates,5, 15, 20, 23 a finding similar to that of our data (11.7%). All donor RHs procedures in this study were performed with no aborted donor operation, no residual disability, and no deaths.
The donor workup must include sufficient evaluation time in any urgent recipient circumstance. However, in this study, the evaluation period was shortened in 6 emergency cases. The median time from its inception to surgery was 7 days, and 4 out of 6 donors were evaluated within 2 weeks. In Korea, there is little chance of obtaining a deceased donor organ even in urgent situations such as United Network for Organ Sharing status 1 and 2A. Fortunately, the outcomes of these donors were similar to the outcomes of others. Of the 6 donors, 2 experienced a grade I complication, hyperbilirubinemia. The hospital stay was similar to that of the nonemergency cases.
The rate of donor morbidity may correlate with types of donor hepatectomy.5, 6, 15, 16 In MERH, segment 4 congestion can occur because MERH harvests the middle hepatic vein to the graft remaining the segment 4 to the remnant liver side.14 However, MERH is considered for donors in good condition whose estimated remnant liver volume is >35% of the whole liver as well as patients who are free of steatosis. Recently, this procedure was applied to more donors with a scissural vein, which maintained some of the outflow of segment 4 and avoided congestion of segment 4 in the early postoperative period.14 Therefore, there were no obvious differences in intraoperative data, postoperative recovery, or rate of morbidity between the conventional RHs and MERHs in this study.
Donor morbidity after LDLT is strongly correlated to each center's experience.5 Therefore, even though our institution's LDLT program began in 1999, we decided to study the period after 2002. We reported the safety of a donor right hepatectomy based on the experiences with 100 cases of LDLT in 2002.4 In that study, no donor deaths were reported, there was no major morbidity, and only several transient minor complications (3%) occurred: abdominal fluid collection (n = 2), catheter-related pneumothorax (n = 1), and prolonged ileus (n = 1). This experience enabled the team to pass the learning curve. However, that study was a retrospective review and did not follow a systemic classification for donor complications. Therefore, the donor morbidity was monitored prospectively. It was expected that the perioperative morbidity would be reduced by ongoing improvements in the surgical technique, donor selection, and postoperative care. However, results were far from what we expected. Using the systemic classification,5, 11 a total of 114 postoperative complications were experienced in 78.3% of the donor population, a rate higher than that reported in the literature.4–7, 15–22 The reason for this is speculative and multifactorial, but it might be because the surveillance was prospectively designed and a systemic classification for morbidity analysis was applied.
The complications were defined as any deviation from a normal postoperative course, which was differentiated from the sequela or “aftereffect” of surgery inherent to the procedure, and failure to cure, which means the original aim of surgery had not been achieved.11 This definition considered subclinical abnormal findings such as pleural effusion.5, 11–13, 22 Broering et al.5 graded pleuralcardial or pericardial effusion without the necessity of pleuralcardial or pericardial centesis being potentially life-threatening complications. This grading method was adapted by another study of right liver donor complications in Japan.22 In the present study, significant pleural effusion detected on a protocol CT scan was recorded as a complication. Because a protocol CT scan had been taken during POD 7 to 14, significant pleural effusion at that time was not a normal finding of recovery.
Hyperbilirubinemia after a liver resection was also classified as a complication, because it was one of the important findings of liver dysfunction after liver resection. Therefore, it must be recorded as an abnormal postoperative recovery finding under defined criteria in a prospective study setting. The definition of hyperbilirubinemia after a liver resection varied; a 100% increase in the bilirubin level from the preoperative level in a conventional liver resection (32.1%),23 or more than 3 mg/dL postoperatively (30.5–33.3%),12, 13 more than 5 mg/dL during the early postoperative periods (5.8–7.3%),6, 17, 22 or transient cholestasis8 after a donor right hepatectomy. In this study, the bilirubin level peaked on POD 3 (3.73 ± 2.01 mg/dL) and then declined to less than 3 mg/dL within 3 days. Previously healthy donors sustaining an increased bilirubin level for more than 3 days after a right hepatectomy was not a normal course of recovery. In addition, jaundice becomes clinically detectable when the serum bilirubin exceeds 3 mg/dL.24 Therefore, hyperbilirubinemia of more than 3 mg/dL for more than 3 days postoperatively or 1.3 mg/dL for 7 days after the hepatectomy was included in the list of donor postoperative complications.12, 13
Most of the complications encountered were minor and self-limited or were silent in that they were noted only in laboratory and protocol imaging studies. However, several patients experienced potentially life-threatening complications requiring additional treatment.
Since 2002, at our institution, a protocol volumetric liver CT scan is done for all donors who give informed consent. A protocol CT scan provides a great deal of information during postoperative donor care. First, complications can be detected before the development of clinical symptoms. Therefore, the scan directs appropriate donor management. Second, although this study did not evaluate the liver's regenerative power, it provided some information in this area. We previously reported12, 14 that regardless of the degree of steatosis and the type of donor hepatectomy, the remnant donor liver grew from 35 to 60% of the original liver volume within 2 weeks after surgery and then to 80% by 3 months.
The types of donor complications were different from those reported in Western countries.5, 16 There was no positioning damage, incisional hernia, deep vein thrombosis, or pulmonary embolism during the study. The lean body mass in individual Asian donors might be a contributing factor.13, 16 The body mass indices of all patients were <28 kg/m2 in this study. However, a multicenter survey conducted at 5 Asian liver transplantation centers17 along with the Japanese Liver Transplantation Society16 reported positional damage (0.1–0.3%), incisional hernia (0.1–0.2%), and pulmonary embolism (0.3–0.5%). Therefore, special attention was paid to pulmonary embolism and to deep-vein thrombosis.25 The cause of the first reported donor death in the world. We did not routinely use perioperative low-molecular-weight heparin, and Doppler ultrasound was not routinely used to exclude deep-vein thrombosis. At our institution, low-molecular-weight heparin was used if the patient had more than 2 of the following factors: history of heavy smoking, age >40 years, and obesity (>20% of the ideal body weight).26 Fourteen donors (16.1%) received perioperative low-molecular-weight heparin.
The incidence of biliary complications was more frequent in right liver donors than in left liver donors.15, 16, 18, 27 In this series, the rate of biliary complication—bile leakage from the cut liver surface—was 13.8%. Although it did not require surgical or endoscopic intervention, bile leakage can be a potentially life-threatening complication. Therefore meticulous dissection of the hilar plate without undrained segments is essential for reducing the risk of bile leakage.8 In addition, careful monitoring against further complications associated with bile leakage, such as an infected biloma, is an important task.
The long-term complications after donor RH are essentially unknown because many donors may be lost to follow-up or be difficult to contact years after the LDLT.17 Although no mortality among 1,860 live liver donors has been reported since the inception of the LDLT program in Korea in December 1994, worldwide reports of late donor deaths and residual morbidity have suggested that long-term follow-up and care might be needed.15–17 Donors must be informed that a long-term follow-up is important.
In conclusion, our study demonstrated that donor RH is a relatively safe procedure, because most of the adverse events were minor and self-limiting and occurred during the perioperative period. However, 78.3% of donors in this study still experienced postoperative morbidity. This finding highlights that meticulous surgery and intensive perioperative care of donors is essential for reducing the incidence of major potentially life-threatening complications. Furthermore, continuous standardized reporting and a comprehensive database to precisely define true donor morbidity is crucial for the safe implementation of this procedure.
- 24Diagnostic techniques. In: BlumgartLH, FongY, eds. Surgery of the Liver and Biliary Tract. London, UK: WB Saunders; 2000: 218., .
- 25Living donor liver transplantation: evaluation of donor and recipient. Nephrol Dial Transplant 2004; 19( Suppl 4): S11–S15., , , .