Hiroshi Sogawa, Thomas Schiano, Lauren Schwartz, Kishore Iyer

Intestinal Transplant Program, Mount Sinai School of Medicine, New York, NY, United States

Background: Combined liver-intestinal transplant (LSB) is the procedure of choice for intestinal failure associated with parenteral nutrition associated liver disease (PNALD). Choice of graft and need for liver replacement in patients with mild to moderate PNALD is still controversial, particularly given the extremely high wait-list mortality in that group awaiting LSB.  We analyze the UNOS database to determine the need for simultaneous liver replacement in patients with PNALD based on total bilirubin.

Methods: Patients with PNALD who had intestine transplant (ITx) with or without liver graft between 1/1990 and 9/2010 were analyzed from the UNOS database. For the purpose of this study, we defined PNALD as a total bilrubin > 3 mg/dl in an ITx recipient.  ANOVAs were used to compare outcomes of continuous variables while Chi-Square analyses, with univariate follow-up tests, were used for categorical variables. Patient and graft survival were analyzed by Kaplan-Meier model (log rank) and risk factors were analyzed by Cox Proportional Hazards model (multivariate).

Results: We divided severity of PNALD based on total bilrubin (TB) into mild (TB 3 – 7 mg/dL), moderate (7 – 11 mg/dL) and severe (> 11 mg/dL).  Of 1988 patients transplanted, 866 patients received ITx without liver graft (SB), and 1120 patients had LSB.  186 patients had mild, 157 patients moderate and 542 patients had severe PNALD.  30 patients with mild PNALD received SB and 61 had LSB. 1-year and 3-year patient survival was similar (p=0.868) in both groups (67.4% and 57.3% in SB, and 60.3% and 52.7% in LSB. 22 patients with moderate PNALD received SB and 135 patients had LSB. 1-year and 3-year patient survival was similar (p=0.850) in both groups (59.2% and 59.2% in SB,and 63.5% and 53.9% in LSB). 45 patients with severe PNALD received SB and 1943 had LSB. 1-year and 3-year survival was similar (p=0.698) in both groups (74.9% and 54.9% in SB, and 66.6% and 54.1%).

For patients with mild to severe PNALD,  lower albumin (Hazard ratio 0.849: p=0.076) and simultaneous kidney transplant (HR2.8-8.3: p=0.064) tended to be risk factors for poor survival in multivariate analysis but these did not reach statistical significance.

Conclusion: These data suggest that hyperbilirubinemia alone is not a contraindication for intestinal transplant without liver graft.  In the absence of the MELD/PELD score, platelet count, INR, and other surrogates for portal hypertension, these data need to be interpreted with caution. Collection of these variables at the time of UNOS listing or through the International registry may allow more critical analysis of this issue.