Outcome of Living-Donor Liver Transplantation Using Grafts from Donors Treated for Fatty Liver

Background

Liver transplantation is the only definitive treatment for end-stage liver disease. Graft selection is an important factor in achieving a good result after transplantation. In the field of cadaveric donor liver transplantation, a liver graft with moderate to severe steatosis (>30–60%) has been reported to be associated with an increased risk of primary nonfunction (PNF), impaired early graft function, and graft loss [1,2]. Furthermore, such macrosteatosis (MaS) is associated with a significantly lower 3-year overall survival among recipients of MaS (>6%) grafts compared to recipients with nonsteatotic grafts (57% vs. 95%; P=0.026) [3]. Because of the shortage of cadaveric donors, however, the use of grafts with more than mild hepatosteatosis depends on the surgeon’s decision and/or recipient’s status [1,2].

In the setting of living-donor liver transplantation (LDLT), fatty infiltration in donor candidates is also encountered during their evaluation. Experiences in using living donor grafts with mild to moderate MaS have been reported in a few studies. Hayashi et al. reviewed 41 LDLTs using liver grafts with various degrees of fatty infiltration and concluded that mild (0–30% steatosis) to moderate (30–60% steatosis) fatty liver (FL) grafts provided comparable graft survival compared with non-FL grafts, although early graft function was slightly disturbed when an FL graft was implanted [4]. Another retrospective study reported that FL grafts with moderate (20–50% steatosis) showed comparable 1-year patient and graft survival compared with less steatotic grafts [5]. However, peak alanine transaminase (ALT) was significantly higher in the moderate FL graft group compared with less steatotic groups, possibly as a result of accelerated ischemia-reperfusion injury [5]. These findings suggested that up to a certain degree of FL would be acceptable for LDLT, paying attention to postoperative ischemia-reperfusion injury.

In this context, fatty infiltration in living donors is considered to be treatable, and a less steatotic graft might result in better posttransplant graft function [6]. Hwang et al. attempted to treat FL donors and reported the efficacy of short-term weight reduction for living donors with moderate to severe FL to alleviate excessive hepatic steatosis [6]. However, the graft function after LDLTs from these donors was not evaluated in the study.

The aim of the present study was to assess the efficacy of treatment for FL in living-donor candidates who were otherwise considered eligible, and to explore its effect on the safety of donor hepatectomy, as well as on early graft function, in recipients implanted with treated grafts.

Material and Methods

TREATMENT PROTOCOL FOR DONORS WITH FATTY LIVER:

The treatment protocol for FL primarily comprised diet–exercise therapy and pharmacological intervention (Figure 1) [10,11]. Diet therapy included restriction of calorie intake to fewer than 1600 kcal/day regardless of body weight, and cessation of alcohol consumption. Donors were instructed to perform aerobic exercise (such as walking, jogging, and/or riding a stationary bicycle) for more than 20 min at least 3 times a week. Statins were prescribed when FL candidates had concurrent hyperlipidemia. When a poor response to diet–exercise therapy was recognized, essential phospholipid (1500 mg/day) was added. Treatment efficacy was evaluated by CT study and laboratory examination once per month. When CTL/S was greater than or equal to 1.1, and CTL was more than or equal to 55 HU, a liver biopsy was performed. Finally, when MaS ≤10% was histopathologically confirmed, the donor was considered to be eligible and the operation was planned. Frozen biopsy of a wedge-cut specimen just after laparotomy was also performed in all donors for reconfirmation.

PROTOCOL FOR IMMUNOSUPPRESSION:

The immunosuppression protocol in our institute is basiliximab induction with a triple regimen of tacrolimus, mycophenolate mofetil, and steroids. Tacrolimus was started on day 4 after transplant, with a target trough level of 10–12 ng/ml during the first postoperative month. Mycophenolate mofetil 500 mg per day was started on day 1 after transplant and gradually increased to 1500 mg per day in 2–3 weeks. The steroid component was completely withdrawn by 4 weeks after transplant.

TREATMENT ASSESSMENT IN DONORS WITH FL:

To determine the effectiveness of treatment in the donors with FL, laboratory findings, CT_L, CT_L/S, and BMI were assessed before and after treatment. The influence of the treatment on surgical parameters (surgical time and estimated blood loss) was determined. Short-term outcome after donor hepatectomy was evaluated using peak ALT, peak total bilirubin (T-Bil), and the incidence of morbidity greater than or equal to Clavien-Dindo grade IIIa.

ASSESSMENT OF GRAFT FUNCTION:

To evaluate graft function, the following laboratory data were collected: ALT, T-Bil, platelet count (Plt), and prothrombin time-international ratio (PT-INR). The following clinical parameters were also assessed: ascites volume, bile production in recipients determined through a biliary drainage tube, duration of hospitalization, incidence of PNF and acute cellular rejection (ACR), and 3-month graft and patient survival. Laboratory data, ascites volume, and bile production were noted on postoperative days (PODs) 1, 3, 7, 10, and 14.

STATISTICAL ANALYSIS:

Continuous variables are expressed as mean±standard deviation (SD) or median with range. The significance level for all statistical testing was set at P<0.05 for a two-tailed test. Continuous variables were compared using the t test or the Mann-Whitney U test. Categorical variables were compared using the chi-squared test or Fisher’s exact test. Survival analysis was performed using the Kaplan-Meier method. All statistical analyses were performed using JMP Pro^®, version 12 (SAS Institute, Inc., Cary, NC, USA).

Results

TREATMENT EFFICACY IN DONORS WITH FL:

Baseline characteristics of donors in the FL and non-FL groups, as well as recipient demographic characteristics for each group, are shown in Table 1. Of 30 candidates, 8 donors were diagnosed with FL by CT evaluation during the study period and received diet–exercise therapy (FL group). All were successfully treated and eventually proceeded to donor surgery. Of those, 3 were treated with diet and exercise therapy only, while the other 5 needed additional pharmacological treatment. While heavy alcohol consumption was not observed in any of the donors in the FL group, donors were instructed to completely abstain during the treatment period. The median treatment duration in the FL group was 58 days (range 35–109 days). After the treatment, significant body weight and BMI reduction and CTL and CTL/S improvements were observed. Confirmatory liver biopsies revealed MaS ≤10% in all donors in the FL group (Table 2). There was no evidence of fibrosis, ballooning, or lobular inflammation on histopathological examination of the biopsy samples.

Changes in physical status, CT value of the liver, LFTs, and the other laboratory findings in the FL group before and after treatment are shown in Table 3. These parameters, other than total cholesterol (T-Cho) and albumin (Alb) after treatment in the FL group, were comparable to those in the non-FL group. However, T-Cho and Alb were significantly lower in the FL group after treatment compared with those in the non-FL group.

INFLUENCE OF TREATMENT ON DONOR SURGERY AND POSTOPERATIVE COURSE:

Surgical parameters and short-term outcome of donor hepatectomy are shown in Table 4. Surgical time and estimated blood loss in the FL and non-FL groups were comparable: 404.8±53.1 min vs. 371.5±56.4 min (P=0.17) and 180.6±134.7 mL vs. 200.2±193.7 mL (P=0.8), respectively. The postoperative evaluation of maximum ALT and T-Bil in the FL and non-FL groups showed no significant differences: 189.6±94.7 IU/L vs. 196.8±57.4 IU/L (P=0.81) and 2.2±1.1 mg/dL vs. 1.7±0.5 mg/dL (P=0.26), respectively. Morbidity greater than or equal to the Clavien-Dindo classification of grade IIIa was not experienced in either group.

EARLY GRAFT FUNCTION IN RECIPIENTS:

Early graft function results in recipients is presented in Figure 2. There were no significant differences in postoperative laboratory findings during the first 2 weeks between the FL and non-FL groups, as shown on POD 7: ALT (38.6±9.8 IU/L vs. 58.6±52 IU/L; P=0.11), T-Bil (5.1±2.5 mg/dL vs. 4.7±2.3 mg/dL; P=0.71), Plt (56625±26934.8/mm3vs. 79636.4±27453.3/mm3; P=0.059), PT-INR (1.3±0.11 vs. 1.4±0.19; P=0.68), ascites volume (2479.6±1208.8 mL/day vs. 1764.2±1211 mL/day; P=0.18), and bile production (100.9±51.4 mL/day vs. 81.3±36.8 mL/day; P=0.28), respectively. Although Plt on POD 1 was significantly lower in the FL group than in the non-FL group (77375±15834.6/mm3vs. 100272.7±37665/mm3; P=0.033), the count reached comparable levels soon thereafter. There was no significant difference in the length of the hospital stay: 79.1±39.1 days in the FL group vs. 83.4±37.9 days in the non-FL group, P=0.79. PNF was not observed in either group. ACR within 1 month was not observed in the FL group, while it was observed in 18.2% of the non-FL group (P=0.55). Graft and patient survival at 3 months were 100% in both groups.

Discussion

Our results show that treatment for donors with FL was effective, and they accomplished the goal in a median period of 58 days. The duration of treatment was almost the same as previously reported, in which the combination therapy of diet and exercise with or without medication was applied for treatment [6,12–14]. For pharmacological treatment, only lipid-lowering drugs such as statins and bezafibrate were used in some reports [12,14]. Because there is evidence showing the efficacy of in reducing LFTs in patients with FL and in improving liver morphology on imaging studies, we applied it in our treatment protocol [15,16].

Donor safety is one of the most important considerations in LDLT. Donor hepatectomy in those with fatty infiltration should be performed carefully because some investigators have reported that hepatosteatosis has a detrimental effect on the postoperative course after major hepatectomy [17–19]. A retrospective study including 386 patients who underwent hepatectomy for colorectal metastases found that increased morbidity, a higher incidence of infective complications, and significantly elevated postoperative LFT results were associated with the degree of underlying liver steatosis [17]. A meta-analysis of the major hepatectomies including living donation concluded that fatty infiltration of the liver directly influenced postoperative complications and mortality [18]. In our study, the donors in the FL group showed a comparable perioperative outcome to those in the non-FL group for operative time, estimated blood loss, postoperative T-Bil and ALT, and incidence rates of mortality and severe morbidity. Thus, the treatment for donor FL might contribute to donor safety.

We also evaluated early graft function using postoperative LFTs, ascites volume, and bile production. These parameters and the duration of hospitalization, incidence of PNF, and 3-month graft and patient survival were also similar between the FL and non-FL groups. These results suggest that the treatment for FL donors is a practical option for non-urgent recipients who can wait several months for LDLT. In terms of the impact of MaS on ACR, a 2.5-fold increased risk of ACR in donors with MaS >30% has been reported [20]. In our study, however, no ACR was experienced in the FL group, suggesting that amelioration of fatty infiltration in the graft might be attributable, at least in part, to inhibiting ACR.

In previous studies, the efficacy of treatment for FL was assessed by body weight, BMI, and preoperative liver biopsy [6,12–14]. Radiological evaluations including CTL and CTL/S were rarely used. Although the confirmation of histological improvement after the treatment is mandatory, inevitable complications should be considered when liver biopsy is performed, especially in the case of steatotic liver. Indeed, the most recent study by Novruzov et al. demonstrated that CT scans can provide detailed evaluation of liver parenchyma to avoid unnecessary liver biopsy [21]. Thus, we selected imaging (CTL and CTL/S) for screening rather than liver biopsy, to avoid the risk of biopsy-related complications. Several studies have revealed the correlation between CTL or CTL/S and histological severity of fatty infiltration [8,22,23]; they reported that a cut-off value of CTL of 35–58 HU and CTL/S of 0.9–0.98 provided a sensitivity of 79%–100% and 79%–91% and a specificity of 61.1–97% and 55.6–97% for detecting moderate MaS (>20–30%), respectively. These discrepancies in the sensitivity and specificity, likely due to the differences of cohort number, patients’ characteristics, and performance of modality, can cause difficulty in determining an optimal cut-off value. However, in a large retrospective cohort study in Japan that evaluated the relationship between degree of fatty infiltration and CTL/S, the authors argued that the optimal value of CTL/S to predict >30% hepatosteatosis was 1.1 [7]. In addition, Park et al. reported that the cut-off value of 58 HU for CTL provided an excellent sensitivity of 100% and a specificity of 95% for the diagnosis of MaS ≥30% [8]. Furthermore, a very high reproducibility of CT attenuation value was demonstrated in the largest cohort study, including 6814 participants [9]. The authors of the study concluded that CT scans can be a reliable diagnostic tool for FL. Therefore, we selected CTL 55 and CTL/S 1.1 as a cut-off value by which donor candidates were considered to have equal to or more than moderate MaS. Actually, only less than 10% steatosis was observed by confirmatory liver biopsy, once candidates achieved CTL >55 HU and CTL/S >1.1. Therefore, these CT values might be useful as a noninvasive preoperative evaluation and could be a surrogate modality to liver biopsy.

Fatty infiltration of the liver is frequently encountered in the setting of LDLT. It is a common reason for ineligibility in living donation, which is reported in up to 38.4% of donors [24] Non-alcoholic fatty liver disease (NAFLD) is a liver disorder with an increasing worldwide incidence rate, and the prevalence of this disease is increasing in parallel with obesity and metabolic syndrome. A cohort study in Japan showed that the rate of NAFLD identified by ultrasonography increased from 13% to 30% during the last decade [25]. As the population who have FL increases, management of donor candidates with FL becomes a more important issue.

Conclusions

In conclusion, our study showed the efficacy of short-term treatment for FL for both donors and recipients. Because metabolic syndrome will become more prevalent in the future, management of donors with a FL is an important issue to consider. Diet and exercise with or without pharmacological therapy as well as noninvasive evaluation by CTL and CTL/S would be an option to resolve this issue.