EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Objectives: Donor safety is paramount in living donor liver transplantation. However, there remains a risk of postoperative complications for some donors. Here, we provide a comprehensive assessment of donor morbidity by a single team with 17 years of experience at a single center.
Materials and Methods: We retrospectively reviewed 453 donor hepatectomies of living donor liver transplants at Kumamoto University from August 2000 to March 2017. Posterior segment graft cases were excluded in this study.
Results: The donors were classified by graft type as follows: right lobe (n = 173), left lobe (n = 149), and left lateral segment (n = 131). The overall complication rate was 29.8%, and the severe complication (Clavien-Dindo grade IIIa or higher) rate was 9.1%. The most frequent complication was bile leakage, with an overall incidence of 13.9% and severe incidence of 4.6%. Among the 3 types of graft, there were no significant differences in bile leakage with any Clavien-Dindo grade. However, upper gastrointestinal complications, such as a duodenal ulcer and gastric stasis, were related to left lobe donation.
Conclusions: There were no significant differences in the incidence of postoperative donor complications, except upper gastrointestinal complications, among the 3 types of graft.

Key words : Bile leakage, Duodenal ulcer, Gastric stasis, Graft type, Living donor liver transplantation

In Japan, living donor liver transplantation (LDLT) started as a pediatric program 3 decades ago and has since expanded to include adults.^1-4 Although the safety of donors is paramount in LDLT, some donors remain at risk for postoperative complications.^5-9

Previous studies have reported that the risk factors associated with postoperative complications are intraoperative hypotension, intraoperative blood transfusion, a prolonged operative time, and body mass index (BMI).^5,9-11 Regarding the type of donor hepatectomy, previous studies have often reported that the incidence of postoperative complications, including biliary complications, is more frequent in right hepatic donation versus left hepatic donation.^7,10,12,13 However, recent studies have repor-ted similar outcomes for right and left hepatic donations.^5,6,14

Here, we provide a comprehensive assessment of donor morbidity by a single team with 17 years of experience of LDLT at a single center.

Materials and Methods

Donors
From August 2000 to March 2017, 466 living donors underwent donor hepatectomies at Kumamoto University Hospital. Nine donors who underwent posterior sectionectomy and 4 donors who were < 20 years old at the time of donor surgery were excluded.

Donors were classified by graft type as follows: right lobe (RL) group (n = 173), left lobe (LL) group (n = 149), and left lateral segment (LLS) group (n = 131). The characteristics and postoperative complications of each group were retrospectively examined.

This study was conducted in accordance with the Declarations of Helsinki and Istanbul. Written informed consent was obtained from all donors, and each donor operation was approved by the institutional review board of Kumamoto University.

Donor and graft selection
In our institution, LDLT donors are selected from spouses or from relatives up to the third degree of kinship (eg, parents, children, siblings, grandparents, uncles, aunts). The age of eligibility for an LDLT donor is limited to <70 years. If the RL graft is the target for procurement, then the donor age is limited to ≤65 years. The standard preoperative routine work-up at our institution includes cardiac and respiratory function. Additionally, our program includes imaging studies such as dynamic computed tomography (including drip infusion cholecystocholangiography), laboratory tests such as liver function, and mental health screening of competencies for autonomy by a clinical psychologist. We use a 3-dimensional image analysis system (Synapse Vincent, Fujifilm) for preoperative evaluation of the size of graft and size of the remnant; the preoperative estimated ratio of the liver remnant must be ≥30%, and the estimated graft-to-recipient weight ratio must be ≥0.7%.

Surgical procedure
Briefly, after isolation of the hepatic pedicle, the hepatic parenchyma was transected using an ultrasonic dissector or surgical aspirator and electrical cautery, without inflow occlusion of blood supply. The bile duct was divided based on the image from intraoperative cholangiography using a C-arm fluoroscopy. After systemic administration of heparin, the hepatic artery, portal vein, and hepatic vein were divided, and the graft liver section was excised. The remnant biliary stump was closed by continuous 6-0 absorbable monofilament sutures. Cholangiogram was performed to ensure the absence of bile leakage and stricture. An adhesion barrier sheet (Seprafilm) was placed behind the abdominal wound and on the parenchymal stump of the liver only in cases after LL donation to prevent the postoperative adhesion. Biliary drainage tube placement via remnant of the cystic duct and peritoneal drainage tube placement were a step in our protocol until 2005 when both were judged inadequate for prevention of the bile leakage. If a drain was placed, then it was removed within 1 week unless bile leakage was detected at that time. After 2006, the peritoneal drain was not routinely placed.

Postoperative management
Laboratory tests, such as complete blood counts, C-reactive protein, liver function, and coagulation activity, were monitored on postoperative days (POD) 0, 1, 2, 3, 6, 8, 10, and 13. Antibiotics were continued until POD 1. Oral intake was started on POD 2. Discharge was generally permitted after POD 14. Postoperative follow-ups were performed with laboratory tests and abdominal ultrasonography at 1 month, 3 months, 6 months, and 1 year.

Statistical analyses
Quantitative data are shown as mean values with SD. Qualitative data are shown as the frequency and rate. Categorical variables were analyzed using the χ2 test and the Fisher test. Continuous variables were analyzed using 1-way analysis of variance or t tests. Variables with P < .20 identified by univariate analysis were further assessed with a binomial logistic regression analysis. We considered P < .05 as statistically significant. All calculations were performed with SPSS software (version 25.0).

Variables
The following variables were compared among the 3 hepatectomy groups. Preoperative variables included age, sex, and BMI. We divided the study period into 3 eras as follows. Era 1 (n = 107) was from August 2000 to December 2005, era 2 (n = 189) was from January 2006 to December 2011, and era 3 (n = 157) was from January 2012 to March 2017. The operative and postoperative variables were operative time, blood loss, length of postoperative hospital stay, and posto-perative complications categorized in accordance with the Clavien-Dindo (CD) classification.¹⁵ Bile leakage was defined as the presence of bile-stained fluid in the abdominal drain or in the intra-abdominal fluid collection. The CD I bile leakage required no replacement of the operatively installed abdominal drain, or no treatment for uninfected intra-abdo-minal fluid collection detected by ultrasonography in the case of no drain. The CD II bile leakage required antimicrobial therapy. The CD IIIa bile leakage required interventional drainage or endoscopic retrograde cholangiopancreatography. The CD IIIb bile leakage required relaparotomy. Complications such as pleural effusion not requiring treatment and temporal hyperbilirubinemia without biliary complication were excluded.

Changes in number of graft types
The changes in number of graft types in our institution are shown in Figure 1. Although the number of each graft type each year slightly changed, there was little difference among the 3 groups since the first case of RL donation in 2001.

Donor characteristics according to type of graft
The donor characteristics among the 3 groups are shown in Table 1. Age (P < .001), sex (P < .001), operative time (P < .001), and blood loss (P < .001) were significantly different in each group. There was no significant difference in BMI, postoperative hospital stay, era, or the incidence of complications with any CD grade among the 3 groups. The overall complication rate and severe complication (CD grade IIIa or higher) rates were 29.8% and 9.1%, respectively.

Details of donor complications
The details of donor complications among the 3 groups are shown in Table 2. The most frequent complication was bile leakage, with an overall incidence of 13.9% and a severe incidence of 4.6%. The second most frequent complication was a duodenal ulcer, followed by gastric stasis and skin wound problems. Multivariable analysis of significant higher risk factors for postoperative complications is shown in Table 3. Male sex was a significant risk factor for any complication (P = .01). When each complication was analyzed, the only risk factor for bile leakage with a CD grade II or higher was male sex (P = .02). As a reference, 2 cases of bile leakage with CD grade IIIb were observed in the LL group. Biliary stricture was not observed in the study period. The LL donation was a significant risk factor for upper gastrointestinal complications, which comprised a duodenal ulcer (P = .01) and gastric stasis (P = .03). One pulmonary embolism occurred after ascending and descending stairs at 1 month postoperatively. Therefore, a temporary inferior vena cava filter was placed, and anticoagulant therapy with heparin and warfarin was administered. At least 4 donors had a diagnosis of depression by a psychiatrist after donor hepatectomy.

Donor characteristics and changes in the postoperative complication rates for the various grafts
Donor characteristics and changes in the posto-perative complication rate in each group are shown in Table 4, Table 5, and Tables 6. Although preo-perative variables, such as age, sex, and BMI, of the donors for each type of graft were similar among the eras, we observed that blood loss, operative time, and postoperative hospital stay were greater in era 1 versus the other eras. Among these 3 factors, the postoperative hospital stay was affected by the change in policy of the team to shorten the stay in 2005. We observed no major change in the complication rate with any CD grade for any type of hepatectomy for any era.

The number of donors in each group in our study was similar. Many previous reports have been based on unbalanced proportions of graft types, such as a comparatively low number of a graft type or unification as a single group comprising LLS grafts and LL grafts according to the preference of each LDLT program.^{5,6,9,10,14,16,17} Because our team had performed pediatric LDLT and adult LDLT, we were able to obtain useful donor data for which the proportions of the various graft types were similar, which facilitated our evaluation of the differences in donor morbidity among the 3 graft types.

In this study, the overall morbidity rate was 29.8% and the severe morbidity rate was 9.1%. No donor mortality was observed. Previously published systematic reviews have reported donor morbidity rates ranging from 0% to 100%, with donor mortality rate up to 0.5% after RL donation and from 0.1% to 0.3% after LL donation.16,18-20 A recently published meta-analysis has shown that the incidence of donors who developed any postoperative complication was not significantly different between the RL group (33%) and the LL group (23%).14 Our study also showed no significant differences in the overall complication rate with any CD grade among the 3 types of graft.

Previous studies have reported that the most frequent complication in donor was bile leakage, and bile leakage occurred more frequently in RL donors versus LL donors.^7,10,12,13 Our study, however, showed no significant difference in the frequency of postoperative bile leakage among the 3 types of hepatectomy. Many cases of bile leakage following donor hepatectomy are derived from the transection surface of hepatic parenchyma, which can be treated with conservative management. Interestingly, 2 cases of bile leakage with CD grade IIIb were observed in the LL group. One of these donors was treated by closure of right hepatic duct leakage site with interrupted 6-0 absorbable monofilament sutures. The other patient received Roux-en-Y choledocho-jejunostomy because of the refractory intractable bile leakage from the complete transected right posterior superior segmental duct.

In the present study, the upper gastrointestinal complications of duodenal ulcer and gastric stasis were more frequently observed in the LL group versus the RL and LLS groups. We previously reported that left-sided hepatectomy, including LL donation and LLS donation, was a risk factor for duodenal ulcers.²¹ We hypothesize that adhesion to the transection surface of the liver may result in a condition of dislocation and deformity of the stomach and the duodenum, which obstructs the passage of food and disrupt blood flow to the duodenum. In relation to this hypothesis, gastric stasis, including the concept of delayed gastric emptying, is related to LL donation.

In Japan, LDLT procedures account for 95% of all liver transplants according to the latest survey.²² In the aging population in Japan, the use of older donors is crucial when no eligible younger candidates are available among family and relatives. We previously reported that older donors aged ≥60 years were similar to younger donors regarding postoperative recovery of liver function, postoperative complication rate, and graft survival rate of the recipients.²³ Similarly, the donor age was not a significant factor for any donor complications in the present study.

The limitation of this study is that it is a retrospec-tive study from a single center. Accumulation of further experience and continuous improvement are necessary.

Our study, which contained a similar number of RL, LL, and LLS donations, shows that postoperative complications of the donors were not related to the type of graft, except for upper gastrointestinal complications such as a duodenal ulcer and gastric stasis.

Volume: 22 Issue: 3 March 2024

FULL TEXT