Healy's law and the pediatric liver allograft

Abstract

Although the long-term care of children who have undergone liver transplantation has a relatively short history, such care has already evolved. This evolution has largely been driven by the availability of new therapies. The biggest current problems are the seemingly conflicting requirements that graft integrity be maintained and that side effects of therapy, most notably renal toxicity, be minimized. Several strategies have shown promise with respect to reducing nephrotoxicity, although long-term data are still awaited. Given the complexity of today's pharmacopoeia, the care of the graft remains a major concern. LFT, liver function test; SPLIT, Studies of Pediatric Liver Transplantation. More fundamental than drug-choice issues, however, as addressed by Ekong et al.1 in this issue of Liver Transplantation, is this question: how should we best monitor the graft? Ekong et al.1 present a summary of their biopsy findings in 63 of 222 long-term (>3 years) liver transplant recipients. Biopsy was prompted in 32 by clinical-laboratory test result [liver function test (LFT)] abnormalities; in this group, unsurprisingly, inflammation and fibrosis were seen. Biopsy to assess liver status in the absence of LFT abnormalities (surveillance biopsy) was offered to 118 longer term (>5 years) survivors, of whom 87 declined. Among the 31 patients who underwent surveillance biopsy at >5 years, 8 had substantial fibrosis, a finding of concern for graft outcome over decades. In addition, longer intervals since transplantation were associated with increased fibrosis. These data confirm those assembled in Birmingham, United Kingdom, by Evans et al.2 The English cohort included, at 5-year survival, 148 patients still considered pediatric who were followed at the reporting institution. Of these, 3 were judged to be poor candidates for liver biopsy, and 10 declined the procedure. Among the 135 who underwent biopsy prompted not specifically by LFT abnormalities but instead by the study schedule (protocol biopsy), 58 had chronic hepatitis, and 47 of these had substantial fibrosis (4 with cirrhosis). At 10-year survival, 68 patients were eligible for protocol biopsy; 1 was judged to be a poor candidate, and 3 declined the procedure. Of the 64 biopsy specimens obtained, 41 showed chronic hepatitis, and 37 of these had substantial fibrosis (6 with cirrhosis). LFTs did not distinguish between patients with and without hepatitis and fibrosis. These 2 reports thus establish that, as in adults,3 surveillance or protocol biopsy in children will identify inflammation and scarring in the engrafted liver and that these will worsen over time. They also establish that LFT monitoring alone does not reflect the status of the allograft liver in children and that liver biopsy is required for an accurate assessment of that status. Bernadine Healy in 1997 wrote: “Don't order a test if you lack the facts to know how to interpret the result.”4 Is it time to apply Healy's Law in pediatric posttransplant liver care? What shall we do, after all, with the information on the liver's status that we obtain at protocol or surveillance biopsy? Shall we intervene when hepatitis or fibrosis is found, stepping in to augment immunosuppressive regimens, with their side effects? Can we act to improve graft outcome, and if we can, how can we? Neither Evans et al.2 nor Ekong et al.1 tell us. To answer these questions requires that, should protocol biopsy find abnormalities, assignation to intervention or to watchful waiting be at random and that allograft status, with other indices of health or morbidity, be assessed periodically thereafter. Can such a study be justified? Should such a study even be attempted, or should we proceed from the assumption that inflammation and scarring found at protocol biopsy require intervention? Probably no single care team attends enough liver-transplanted children to answer these questions. The Studies of Pediatric Liver Transplantation (SPLIT) consortium5 is at least positioned to consider the questions, and perhaps the study by Ekong et al.,1 confirming the observations of Evans et al.,2 will provide the impetus for SPLIT or another consortium to decide whether it wishes to set the agenda for further work in this field.