Factors affecting growth after pediatric liver transplantation

Poor linear growth after pediatric orthotopic liver transplantation (OLT) is a well-described phenomenon. We have undertaken a bivariate and multivariate analysis of multiple factors that might effect postOLT growth in all children who underwent transplantation at a single center, with survival > 1 year and adequate follow-up.

Standardized height score (Z score) and height deficit (centimeters below the 50th percentile) were computed for each patient over time. The variables assessed were (i) age at OLT, (ii) gender, (iii) pretransplantation diagnosis, (iv) Z score and height deficit at OLT, (v) tacrolimus versus cyclosporine as primary immunosuppressive therapy, (vi) retransplantation, (vii) graft disease, (viii) chronic illness, (ix) posttransplant lymphoproliferative disease, (x) intractable rejection, and (xi) prednisone withdrawal.

A total of 236 children met the inclusion criteria, with a mean follow-up of 3.8+/-1.9 years. For the population as a whole, the baseline Z score was -1.72 (fourth percentile) with a significant improvement to - 1.37 (ninth percentile) at 2 years, but with no additional gain at 5 years (Z score -1.4). The baseline height deficit was -6.4 cm, with no improvement at 2 years (-6.52 cm), and was significantly worse at 5 years (-7.87 cm). In the bivariate analysis, the most important variables affecting growth were age at OLT, Z score at OLT, and diagnosis. In general, children <2 years with biliary atresia and those with the most growth delay at OLT showed the best posttransplantation growth. In the multivariate analysis, 18 factors were considered, of which 9 were significant. These were (i) Z score at baseline, (ii) follow-up time, (iii) age at OLT, (iv) diagnosis of tumor, (v) diagnosis of fulminant hepatic failure, (vi) retransplantation, (vii) graft disease, (viii) posttransplant lymphoproliferative disease, and (ix) stoppage of prednisone. Multivariate models using these nine variables accounted for 84% of the variation in standardized height.

In general, children after OLT show some potential for catch-up growth but do not achieve normal height compared with their age and sex-matched peers. A multivariate analysis was necessary to investigate the interdependent effects of the many variables that can affect growth after OLT. The most important detrimental affects were older age at time of OLT, Z scores greater than -2.0 at OLT, fulminant hepatic failure, tumor, and postOLT complications causing graft dysfunction.