Early graft function after pediatric liver transplantation: comparison between in situ split liver grafts and...

The systematic application of living-related and cadaveric, in situ split-liver transplantation has helped to alleviate the critical shortage of suitable-sized, pediatric donors. Undoubtedly, both techniques are beneficial and advantageous; however, the superiority of either graft source has not been demonstrated directly. Because of the potential living-donor risks, we reserve the living donor as the last graft option for pediatric recipients awaiting liver transplantation. Inasmuch as no direct comparison between these two graft types has been performed, we sought to perform a comparative analysis of the functional outcomes of left lateral segmental grafts procured from these donor sources to determine whether differences do exist.

A retrospective analysis of all liver transplants performed at a single institution between February 1984 and January 1999 was undertaken. Only pediatric (<18 years) recipients of left lateral segmental grafts procured from either living-related (LRD) or cadaveric, in situ split-liver (SLD) donors were included. A detailed analysis of preoperative, intraoperative, and postoperative variables was undertaken. Survival was estimated using the Kaplan-Meier method, and comparison of variables between groups was undertaken using the t test of Wilcoxon rank sum test.

There were no significant differences in the preoperative variables between the 39 recipients of SLD grafts and 34 recipients of LRD grafts. The donors did differ significantly in mean age, ABO blood group matching, and preoperative liver function testing. Postoperative liver function testing revealed significant early differences in aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, prothrombin time, and alkaline phosphatase, with grafts from LRD performing better than those from SLD. SLD grafts also had significantly longer ischemia times and a higher incidence of graft loss owing to primary nonfunction and technical complications (9 vs. 2, P<0.05). However, six of these graft losses in the SLD group were because of technical or immunologic causes, which, theoretically, should not differ between the two groups. Furthermore, these graft losses did not negatively impact early patient survival as most patients were successfully rescued with retransplantation (30-day actuarial survival, 97.1% SLD vs. 94.1% LRD, P=0.745). In the surviving grafts, the early differences in liver function variables normalized.

Inherent differences in both donor sources exist and account for differences seen in preoperative and intraoperative variables. Segmental grafts from LRD clearly performed better in the first week after transplantation as demonstrated by lower liver function variables and less graft loss to primary nonfunction. However, the intermediate function (7-30 days) of both grafts did not differ, and the early graft losses did not translate into patient death. Although minimal living-donor morbidity was seen in this series, the use of this donor type still carries a finite risk. We therefore will continue to use SLD as the primary graft source for pediatric patients awaiting liver transplantation.