Reply to: "Predicted Volume or Actual Weight for Graft Selection Policy in Living-donor Liver Transplantation".

Abstract

We are grateful to Dr Ikegami and his colleagues for their interest and useful comments regarding our recently published article.1 We claimed that small grafts with 0.6% ≤ actual graft-to-recipient weight ratio (GRWR) < 0.8% should be used as safely as large grafts with 0.8% ≤ actual GRWR regardless of lobe selection, and that 0.6% in GRWR was reasonable as the minimum requirement of graft volume (GV) in living-donor liver transplantation. However, small grafts should be indicated carefully for high-risk cases. We extracted data on estimated and actual GRWR from the electronic medical record system and demonstrated their relationship in the article (Figure S1A). However, the dataset may not be sufficient to fully answer their questions. This is because missing data on estimated GRWR were not negligible, and the time of measurement of the recipients’ preoperative weight was inconsistent. Therefore, we recollected preoperative volumetric imaging data for as many cases as possible and rechecked the estimated GV for each case. Estimated and actual GRWRs were calculated by dividing graft weight by recipient weight just before living-donor liver transplantation. We retrieved the data of 384 donor grafts of 417 cases in the original cohort. The following formula was used for the volume-to-weight conversion of grafts: estimated graft weight (g) = 0.91 × estimated GV (mL). The coefficient has been used at our institution based on our experience from previous cases. As demonstrated in Figure 1, the actual GRWR was strongly correlated with the estimated GRWR (actual GRWR = 0.95 × estimated GRWR + 0.07, P < 0.001). However, 10 cases with an actual GRWR <0.6% were identified, whose GRWR was overestimated preoperatively as ≥0.6%.FIGURE 1.: Correlation between the estimated GRWR and the actual GRWR. The solid line represents the best-fit line. Gray upper and lower shaded areas indicate 95% confidence intervals for the estimates. ρ, Spearman rank correlation coefficient; GRWR, graft-to-recipient weight ratio.We next compared the safety of small grafts with 0.6% ≤ estimated GRWR < 0.8% against that of large grafts with 0.8% ≤ estimated GRWR. Multivariate analysis demonstrated that small grafts were not a significant risk factor for overall graft survival (P = 0.90; hazard ratio: 0.90; 95% confidence interval: 0.62-1.70). However, Kaplan-Meier curve of small grafts was slightly inferior to that of large grafts (Figure 2). This may partly be due to the aforementioned overestimation of small graft size. The comparison between the extra-small grafts with estimated GRWR <0.6% and small grafts was not realistic because only 7 extra small grafts were included.FIGURE 2.: Unadjusted Kaplan-Meier graft survival curves between extra small grafts with estimated GRWR <0.6%, small grafts with 0.6% ≤ estimated GRWR < 0.8%, and large grafts with 0.8% ≤ estimated GRWR. GRWR, graft-to-recipient weight ratio; LDLT, living-donor liver transplantation.To solve the issue of graft size discrepancy, especially the overestimation of small grafts, we recently changed the software for preoperative analysis of GV. The parenchymal cutting plane has been strictly determined for each case before the procedure by paying attention to the middle hepatic vein/inferior vena cava, Spiegel lobe, and peripheral segmental branches of the portal vein in 3D imaging. At this point, we have not yet encountered any small graft cases with an actual GRWR <0.6%. Regardless of this fact, we are aware that graft selection should be cautiously determined if the estimated GRWR is close to 0.6%, and our correction coefficient of 0.91 should be reevaluated from a considerable number of cases because of volume analysis modification. Furthermore, as they stated, adjusting factors such as donor age (responsible for causing estimation errors in liver volumetry) is essential to accurately predict GV.2 Once again, we greatly appreciate the helpful insights provided by Dr Ikegami and his colleagues.