EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Objectives: Living-donor liver transplant with small-for-size grafts (graft-to-recipient weight ratio < 0.8) may provide satisfactory results. We compared outcomes between right and left donor lobe in living-donor liver transplant.

Materials and Methods: Patients who had living-donor liver transplant from 2006 to 2008 with graft-to-recipient weight ratio < 0.8 (graft: right lobe, 24 patients; left lobe, 26 patients) were reviewed retrospectively.

Results: There were no significant differences in demographic and preoperative clinical data between patients who received a right or left lobe liver graft. Duration of surgery was longer, cold ischemia time was shorter, and mean baseline portal vein flow was greater in transplants performed with left than right donor lobes. Portal vein flow modulation with splenectomy was performed when portal flow was > 250 mL/min/100 g graft. Small-for-size syndrome was observed in 6 recipients (14%), but no patient who developed small-for-size syndrome developed liver failure or required revision transplant. The frequency of small-for-size syndrome was significantly greater in patients who had left lobe (4 patients [15%]) than right lobe transplant (2 patients [8%]; P ≤ .05). Graft dysfunction-free survival was significantly greater with right than left lobe grafts. In multivariate analysis, graft side was the only significant risk factor for small-for-size syndrome.

Conclusions: In patients having living-donor liver transplant with small-for-size grafts, outcome was better with right than left lobe grafts.

Key words : Hepatic failure, Small-for-size syndrome, Portal vein flow modulation, Complications

Liver transplant is a common treatment option for patients who have end-stage liver disease. The increased frequency of adult-to-adult living-donor liver transplant has resulted in an improvement in the donor organ pool.¹ However, the ideal volume of the partial liver graft is an issue because a small graft may not meet the metabolic demands of the recipient and may fail. Therefore, it is important to determine graft-size matching before transplant.^2-6 A transplant with graft-to-recipient weight ratio < 0.8 has poorer graft function and outcome than transplant with a higher ratio.² Grafts that have graft-to-recipient weight ratio < 0.8 are known as small-for-size grafts.

The minimum graft volume necessary to ensure optimal outcomes in living-donor liver transplant recipients is unknown. Most transplant centers in North America and some centers in Asia use a cutoff graft-to-recipient weight ratio = 0.8.^7-9 This lower limit was re-evaluated by other transplant centers, and there was no significant correlation between graft volume and patient outcomes.^10-12 In our center, we have cautiously adopted the policy of using grafts with graft-to-recipient weight ratio < 0.8 to avoid donor rejection on the basis of graft volume and to increase the donor pool. The clinical problems that may occur with small-for-size grafts are termed small-for-size syndrome.^2-6 Factors that predispose grafts to small-for-size syndrome include poor pretransplant condition of the recipient, prolonged ischemia times, high portal flow rates, and graft steatosis.^2-6

In many transplant centers, a right lobe liver graft is used to overcome size limitation.^7,13-18 However, the right lobe may have a graft-to-recipient weight ratio < 0.8 and cause concern about the development of small-for-size syndrome. Nevertheless, the outcomes of transplant with right lobe grafts have improved because of improved techniques for outflow reconstruction.

In our center, we have achieved good results with right lobe liver transplants that have graft-to-recipient weight ratio < 0.8. The purpose of this study was to compare outcomes after living-donor liver transplant with left or right lobe partial liver grafts that had graft-to-recipient weight ratio < 0.8.

Materials and Methods

Subjects
We retrospectively analyzed the data for 50 recipients who had living-donor liver transplant with graft-to-recipient weight ratio < 0.8 at Kaohsiung Chang Gung Memorial Hospital, Taiwan, from 2006 to 2008. All potential donors who had estimated graft-to-recipient weight ratio < 0.8 were evaluated for eligibility for the procedure when the remnant liver volume in the donor was > 30%. The recipients were informed about the possible risk of graft dysfunction because of small graft size. The living-donor liver transplants were grouped by side of the liver graft excised from the donor (right lobe, 24 transplants; left lobe, 26 transplants), and all grafts had graft-to-recipient weight ratio < 0.8. The study was approved by the Ethical Review Committee of the institute. All protocols conformed with the ethical guidelines of the 1975 Helsinki Declaration. Informed consent was obtained from all subjects.

Evaluation and procedures
Preoperative data included donor age, recipient age, Model for End-Stage Liver Disease score, Child-Turcotte-Pugh score, and indications for transplant. Liver volume was calculated with computed tomographic volumetric analysis. Recipient standard liver volume was calculated according to the Urata formula.¹⁹ The donor and recipient evaluation and hepatectomies for liver transplant were performed as previously described.^20-22 There were no transplants that had ABO incompatibility. In all right lobe transplants, the grafts were excised without the middle hepatic vein. If major middle hepatic vein tributaries (> 5 mm) were encountered, they were reconstructed on the back table with cryopreserved vascular grafts. The grafts were perfused with cold histidine tryptophan ketoglutarate solution (Custodiol, Essential Pharmaceuticals; Newtown, PA, USA). The weight of the liver grafts was assessed immediately after perfusion with preservation solution. This actual graft weight (g) was used to calculate the graft-to-recipient weight ratio.

Doppler flow hemodynamics
After vascular reconstruction and before biliary reconstruction, intraoperative Doppler ultraso-nography (Acuson; Mountain View, CA, USA) was performed to confirm good triphasic hepatic venous outflow, confirm good intrahepatic arterial flow, and avoid hepatic venous outflow block or portal vein anastomotic strictures that could potentially cause high portal pressure. Portal vein flow rate was measured by recording the angle with the corrected flow velocity and the cross-sectional area of the portal vein (mL/min) and was expressed as mL/min/100 g graft. Serial determinations of portal vein flow rate were performed, and the mean of 3 recordings was determined. When portal vein flow rate was > 250 mL/min/100 g graft, splenectomy was performed to modify the portal vein flow to < 250 mL/min/100 g graft.²³ Ligation of the portosystemic shunts was not required in any patient.24 Patients who had right or left donor lobe were compared for initial portal vein flow rates and incidence of portal vein thrombosis.

Posttransplant outcomes
Median follow-up was 13.5 months (range, 1-48 mo). Short-term outcomes (1 mo) were determined including in-hospital mortality, primary nonfunction of the graft, and graft loss. Serial testing included determination of levels of aspartate amino-transferase, alanine aminotransferase, total bilirubin, and international normalized ratio. The length of intensive care unit and hospital stay and postoperative complications such as infection, acute cellular rejection, and biliary complications were recorded.

Small-for-size syndrome
Small-for-size syndrome was diagnosed when there were 2 of 3 parameters present on 3 consecutive days (international normalized ratio > 2; total bilirubin > 100 μmol/L; and encephalopathy, grade 3 or 4) as previously described.6 This diagnosis was made only after exclusion of other causes of graft dysfunction such as infection, vascular complications, and acute cellular rejections.

Statistical analyses
Graft dysfunction-free survival was determined with Kaplan-Meier method, and factors were compared with log-rank test. Frequencies were compared with Kruskal-Wallis 1-way analysis of variance. Categorical data were compared with the chi-square test. Univariate analysis was performed with Kaplan-Meier method to identify risk factors for occurrence of small-for-size syndrome, and multivariate analysis of significant factors was performed with logistic regression. Statistical significance was defined by P ≤ .05.

Preoperative data
Most recipients were men (Table 1). Median Model for End-Stage Liver Disease score was 10 (95% interquartile range, 7-13) and Child-Turcotte-Pugh score was 7 (95% interquartile range, 5-9) (Table 1). The median graft weight was 550 g (95% interquartile range, 487-612 g). The median standard liver volume of the recipient liver was 1084 mL (95% interquartile range, 920-1358 mL). The median graft-weight-to-standard-liver-volume ratio was 41% (95% interquartile range, 38%-44%). The mean graft-to-recipient weight ratio was 0.71 ± 0.06. There were no significant differences in demographic and preoperative clinical data between patients who received a right or left lobe donor liver graft (Table 1). The most common indication for transplant was hepatitis B-related hepatocellular carcinoma, which was noted in 19 patients (38%).

Operative data
Duration of surgery was longer, cold ischemia time was shorter, and mean baseline portal vein flow rate was greater in transplants performed with left lobe grafts than with right liver lobes (Table 2). There was no donor mortality. The mean duration of hospital stay for all donors was 6 ± 2 days. In the 24 recipients who received a right lobe graft, middle hepatic vein reconstruction with cryopreserved vascular grafts was performed in 12 recipients (50%). All biliary reconstructions were performed with the duct-to-duct method. Frequency of portal flow modulation was similar between recipients who received right or left grafts (Table 2).

The mean baseline portal vein flow of all the patient population before modulation was 275 ± 18 mL/min/100 g graft and after modulation was 190 ± 11 mL/min/100 g graft (P ≤ .001; 95% CI: 71-97), indicating successful modulation (Figure 1). The preoperative spleen size was directly proportional to portal vein flow before modulation (Figure 2).

Small-for-size syndrome
Small-for-size syndrome was observed in 6 recipients (14%), but no patient who developed small-for-size syndrome developed liver failure or required revision transplant. The frequency of small-for-sizesyndrome was significantly greater in patients who had left lobe (4 patients [15%]) than right lobe transplant (2 patients [8%]; P ≤ .05). Graft dysfunction-free survival was significantly greater with right than left lobe grafts (Figure 3). In univariate analysis, graft side and baseline portal vein flow were significant risk factors for small-for-size syndrome; in multivariate analysis, graft side was the only significant risk factor for small-for-size syndrome (Table 3).

The incidence of small-for-size syndrome was similar among patients with right liver lobes with portal vein flow modulation and those without portal vein flow modulation (Figure 4) Also, the incidence of small-for-size syndrome was similar among patients with left lobe liver grafts with portal vein flow modulation and those without modulation (Figure 4).

Postoperative complications and outcomes
The overall mean hospital stay after transplant for recipients was 30 ± 3 days with right lobe transplant, similar to 32 ± 2 days with left lobe transplant (difference not significant). Recipients of left lobe transplant had a longer posttransplant intensive care unit stay (25 ± 7 d) than recipients of right lobe transplant (15 ± 3 d; P ≤ .004). The enzyme elevation, total bilirubin level, and international normalized ratio were greater with left than right lobe grafts but reached normal baseline within 1 month after transplant with left or right lobe grafts (Figure 5).

There were 24 complications that occurred in 15 patients (63%) who had right lobe transplant and 26 complications in 20 patients (77%) who had left lobe transplant (P = NS) (Table 4). There were no patients who had primary nonfunction, graft failure, or 30-day in-hospital death.

This study demonstrated that the outcomes of grafts with low graft-to-recipient weight ratio were better with right than left lobe grafts; right lobe grafts were associated with lower frequency of small-for-size syndrome than were left lobe grafts, similar to previously reported findings.²⁵ The median graft weight/standard liver volume ratio was 37% with right lobe grafts and 36% with left lobe grafts, and outcomes were better with right lobe grafts. In another study, median graft-to-recipient weight ratio was 0.74 with right lobe grafts and 0.67 with left lobe grafts (difference not significant), and outcomes were better with right lobe grafts.²⁶

A statistically better 1-year graft survival was reported in transplants with graft-to-recipient weight ratio < 1.2 However, another study showed excellent survival in recipients who had either left or right lobe grafts with graft-to-recipient weight ratio < 0.8 and creation of hemiportocaval shunt.²⁷ Another study showed excellent graft survival in recipients of left lobe grafts with graft-to-recipient weight ratio between 0.6 and 0.8.28 The main concern in using small-for-size grafts is the problem of small-for-size syndrome, which can cause graft loss and necessitate revision transplant.

In small-for-size syndrome, dysfunction or failure of the small partial liver graft (graft-to-recipient weight ratio, 0.8) occurs during the first postoperative week in the absence of technical, immunologic, or infectious causes.6 Graft dysfunction also is defined by the presence of 2 of 3 features (total bilirubin > 5.8 mg/dL; international normalized ratio > 2; and encephalopathy, grade 3 or 4) on 3 consecutive days. The recipient’s condition is an important variable because a larger graft volume may be necessary when the recipient’s peritransplant liver function is severely impaired.^29-32 In the present study, recipients selected to receive small-for-size grafts had low median Model for End-Stage Liver Disease and Child-Turcotte-Pugh score (Table 1) to ensure good outcomes with small-for-size grafts.
A major cause of small-for-size syndrome is portalhyperperfusion.^6,33,34 Portal venous pressure during living-donor liver transplant is affected by many factors such as central venous pressure, hepatic venous outflow and hepatic vein reconstruction procedures, graft type, quality, compliance, congestion volume, portal inflow volume, and portosystemic shunts. These factors affect portal venous pressure and flow. Portal vein pressure is correlated with flow rate, and we did not measure portal pressure in any patient. Portal flow modulation was performed when portal vein flow rate was > 250 mg/mL/100 g graft.^23,33,35 This modulation may be done by performing splenectomy, splenic artery ligation, or hemiportocaval shunt.^35-43 The frequency of portal flow modulation was similar with left and right lobe grafts. The mean baseline portal vein flow was greater with left than right lobes (Table 2).

In the present study, small-for-size syndrome was less frequent with right than left lobe grafts. Graft side was the only risk factor in multivariate analysis associated with the development of small-for-size syndrome (Table 3). The 2 groups in this study were not perfectly matched because of the retrospective study design and small sample size. The duration of surgery was significantly greater with left than right lobe grafts (Table 2), possibly because of greater technical difficulty implanting the left than right lobe.⁴⁴ The trimming of the shape and length of the left portal vein may be difficult because the graft may not be located at the intended site. In addition, the vertical course of the left hepatic artery and the short stump may complicate microsurgical reconstruction in a restricted operative space and cause cephalic traction of the graft, graft compression, and early graft dysfunction.

The cold ischemia time was significantly greater with right than left lobe grafts (Table 2), possibly because 12 right lobe grafts (50%) had reconstruction of middle hepatic vein tributaries with cryopreserved vascular grafts on the back table. This enabled better drainage of the anterior sector and possibly contributed to better outcomes with right small-for-size grafts. During liver regeneration, the left lobe grows in a ventrolateral direction and may rotate over the middle hepatic vein, causing interference with smooth venous outflow and early graft dysfunction. In addition, the regenerating left lobe graft may compress the inferior vena cava and block graft outflow, causing early graft dysfunction. In contrast, the right liver graft lies in the natural anatomical site natural body cavity and does not rotate on the middle hepatic vein or compress the inferior vena cava with a change in position. These differences between right and left lobe grafts may explain the better outcomes noted with right than left lobe small-for-size grafts.

The mean duration of intensive care unit stay was greater with left than right lobe grafts because of the high elevation of liver function tests with left lobe grafts early after surgery (Figure 5). However there was no difference in the mean duration of hospital stay between the 2 groups because liver function returned to normal in both groups by 1 month after transplant.

Limitations of the present study include the retrospective design and small sample size, which possibility may have caused type 1 error. The present finding may be evaluated in another study with greater sample size and prospective design. In addition, there may be some demographic factors that were not evaluated that may affect the results. Based on the present findings, we recommend rigorous perioperative care and a low threshold for intervention, especially with left lobe livertransplants and graft-to-recipient weight ratio < 0.8.

Volume: 12 Issue: 4 August 2014

FULL TEXT