Reply: The easiest way to improve the outcome of suboptimal liver grafts

Abstract

Transaminase levels on the seventh postoperative day in the single aortic perfusion (SAP) group were only minimally influenced by the high incidence of early graft losses in this group. In fact, 3 patients having primary dysfunction (PDF) died within the first post-LT week, whereas 2 other patients were successfully retransplanted for PDF. Thus, patients in the SAP group, who were alive on the seventh postoperative day, had transaminase levels similar to those of patients in the modified double perfusion (MDP) group. The study by de Ville de Goyet et al.3 was a retrospective analysis of an old series. In that study, the authors compared the SAP technique with the classical aortic and portal cooling technique without our expedient of a tourniquet surrounding the portal vein to be tightened immediately after cross-clamping (MDP technique). The authors found significant differences not in the overall immediate graft function but only in the postoperative peak of glutamic pyruvic transaminase. Moreover, they considered only optimal donors and used only a University of Wisconsin (UW) solution as a preservative fluid. Thus, several methodological and technical differences make de Ville de Goyet et al.'s conclusions not comparable with those of our study. Further prospective studies with a larger study population are necessary to explain the role of the preservation solution in suboptimal livers. However, we have recently performed a multicenter, prospective, randomized pilot study4 comparing Celsior and UW solutions in the context of LT. That study also evaluated a significant proportion of suboptimal donors (around 20%). In the subgroup of patients receiving a suboptimal liver (data not shown), the prevalence of PDF was higher in the UW group, but this difference did not reach statistical significance. Cold ischemia time (CIT) and histologically determined liver steatosis were used as major criteria to define suboptimal livers in our study, although they obviously cannot be known at the time of organ procurement. This choice of study design was aimed to avoid the potential bias deriving from excluding factors capable of significantly influencing graft performance in the analysis. This methodological aspect may explain the high mean values of CIT in the 2 groups. As suggested by Jeon and Ranjan and already remarked in our discussion, however, it is possible that the preventive adoption of a lower CIT could have lowered the difference in PDF prevalence between the SAP and MDP groups. We have no experience with organ procurement from donors after cardiac death. In such donors, the SAP technique is probably a more suitable procurement tool. The adoption of a lower CIT is advisable, especially in those cases in which a potential “at risk” donor is allocated. We are performing a randomized controlled trial (begun on December 2006) evaluating the role of a systemic (at the beginning of the harvesting procedure) and loco-regional infusion (intraportal vein infusion just before cross-clamping) of N-acetyl cysteine during liver procurement. In this study, we have maintained low CIT values, and we have defined suboptimal livers only on the basis of preoperative characteristics (major criteria: ultrasound steatosis and/or body mass index > 30, partial graft, age > 60 years, and hepatitis C virus or hepatitis B surface antigen positivity). The preliminary results of this study (after the first 12 months of enrolment) have shown a statistically significantly lower incidence of PDF in patients in the N-acetyl cysteine group. The use of graft machine perfusion preserving methods during the procurement procedure may improve the post-LT outcome of suboptimal livers. In this context, we are developing a subnormothermic machine perfusion model in pigs for the preservation of suboptimal grafts. An optimal donor-recipient match5 represents a crucial issue to overcome the risk of post-LT PDF. In conclusion, as suggested by Jeon and Ranjan, the MDP technique is only one of the possible tools to optimize the results of LT with suboptimal livers. The reduction of CIT median values, the use of non–heart-beating donors, the infusion of antioxidant agents, the introduction of a graft machine perfusion model, and the application of an optimal donor-recipient match are all potential instruments of a multimodal strategy aimed at improving the outcome of suboptimal livers and at extending and standardizing the use of such a relevant resource for LT candidates. Francesco D'Amico*, Alessandro Vitale , Alberto Brolese*, Giacomo Zanus*, Umberto Cillo*, * Unità di Chirurgia Epatobiliare e Trapianto Epatico Dipartimento Assistenziale di Chirurgia Generale e Trapianti d'Organo Azienda Ospedaliera di Padova Padova, Italy, Unità di Chirurgia Oncologica Istituto Oncologico Veneto Istituto di Ricovero e Cura a Carattere Scientifico Padova, Italy.