Response to Comment on “Living Donor Liver Transplantation With Augmented Venous Outflow and Splenectomy: A Promised Land for Small Left Lobe Grafts”

We appreciate the comments of Ng et al.1 with regard to our recent study. The authors concluded that our series of living donor liver transplantation (LDLT), especially left lobe graft (LLG)-LDLT was highly selective based on recipient demographics including a low median model for end-stage liver disease (MELD) score. There is no doubt that careful selection is important to achieve favorable outcomes to allow for proof of concept in any innovative surgical endeavor, especially in LDLT. Nevertheless, we believe that the LLG-LDLTs in our series were not overly selective. Importantly, our LLG rate was 47% of adult LDLT performed in our program, which is one of the highest LLG utilization rates in Western countries. Our goal is to maximize overall donor safety by preferential use of LLG when a high rate of recipient survival can be reasonably assured. It is universally recognized that in Western countries with high penetrance of deceased donor grafts, a program’s median MELD score for LDLT is lower than that for deceased donor liver transplantation (DDLT).3 Therefore, a lower median MELD score in LDLT is not indicative of being highly overly selective regarding recipients. Although we do not have a firm institutional MELD score threshold that excludes LDLT evaluation, when a candidate’s MELD score is high enough to receive DDLT in a timely manner, the LDLT option may not be the best path forward. Nevertheless, we still preserve the LDLT option for certain high MELD patients when timely LDLT is preferred. In this setting, the 1-year graft survival following LDLT for patients with MELD greater than 25 remains favorable and comparable to that for MELD less than 25 in our series. It should be emphasized that we rarely exclude the LDLT option because of recipient-donor selection mismatch as Ng et al. suspected. Our LDLT accounts for 14% to 16 % of our annual transplant volume. The reason is simply the lack of willing live donors, rather than declining the donor candidates. To increase LDLT volume, more education of the public to facilitate donor recruitment is an imperative step. Expansion of minimally invasive donor hepatectomy will hopefully enhance the utilization of LDLT as it did for a kidney transplant and is appealing to the public.4 We have advocated for the importance of adjusting preoperative risk factors that include actual graft size, graft quality, recipient disease severity, and intra-operative optimization of graft inflow and outflow.5 Ng pointed out a lower MELD score and less portal hypertension in LLG compared with those for the right lobe graft group in our study. The mean graft size and graft-to-recipient weight ratio (GRWR) of the LLG group were always smaller compared with the right lobe graft group.6–8 The lower MELD score and the lesser degree of portal hypertension in our LLG group are merely reflective of the risk adjustment made for the smaller graft size of the LLG group. In our graft selection algorithm, we believe an LLG that provides for only a 0.6% to 0.7% GRWR for a case with multiple other risk factors of the small-for-size syndrome (SFSS), is insufficient. In that case, we offer the right lobe graft option. With this graft selection process, our LLG rate still reached 47%. Our program launched this LLG first choice approach when LLG use was less than 5% of adult LDLT in the Adult-to-Adult Living Donor Transplant Cohort Study consortium of North American Centers.9 With the recent Organ Procurement and Transplantation Network/United Network for Organ Sharing registry data analysis of 2009–2018, Kitajima et al.3 reported a still low but slowly increasing LLG rate, accounting for 17.5% of adult LDLT. This large data analysis of 1486 LDLTs showed a result consistent with previous studies that a combination of high MELD score, moderate ascites, and the use of LLG was associated with worse outcomes. However, the United Network for Organ Sharing registry does not include data on SFSS, or surgical techniques to optimize inflow and outflow that can contribute to the recent improvement in outcomes. Therefore, a single, large-volume center study is necessary to assess the impact of these surgical innovations on post-LDLT outcomes. Our study presents an innovative outflow augmentation technique combined with the use of splenectomy for inflow modulation. This combined technique nearly eradicated SFSS and successfully controlled post-transplant ascites with only 10% of LLG recipients experiencing intractable ascites (more than 1 L/day on post-transplant day 14). As Ng et al. mentioned that the Juntendo University group in Tokyo, showed excellent graft survival after LDLT-LLG with limited use of splenic artery ligation in 11% of cases and no outflow augmentation.10 However, they did report intractable ascites in 31/59 (53%) patients.10,11 In this setting, more use of inflow modulation or our proposed outflow augmentation could have reduced the ascites production. In summary, our study demonstrated that with our surgical refinement of inflow modulation and outflow augmentation along with preoperative risk assessment for SFSS, an LLG can be utilized far more frequently than the current rate in Western countries. This approach can improve donor safety while further improving access to transplant for the patients on the waiting list. We hope it is another important step in challenging the perceived boundaries of safe LLG utilization. ACKNOWLEDGMENTS The authors thank Ms. Nancy Lapid for English language editing.