Introduction

Biliary complications (BC) cause major morbidity and mortality after liver transplantation (LT). Literature is replete with articles on early BC (less than 1 year post LT); however, little is known about late BC.

Methods

We reviewed retrospectively our prospectively collected data for adult LTs from Jan 1997 to Dec 2007. Multiorgan transplants were excluded. Patients with BC more than a year after LT (late BC) were compared to patients with BC in the first year (early BC) and patients who never had BC. Fisher exact tests with post hoc Bonferroni test and Kruskal-Wallis analysis with post-hoc Dunn’s test were used for comparisons. Univariate and multivariate Cox regression analysis was performed.

Results

1621 LTs were included, 92.4% primary and 7.6% re LTs. 29.3% patients had BC with 24.3% early BC (n=384) and 5% late BC (n=82). Late BC included anastomotic (53.6%) and nonanastomotic (21.9%) strictures, papillary stenosis (14.6%), stones (21.9%), biloma (10.9%) and tumors (1.2%). In the late BC group, 18.2% patients had surgical repair, 47.5% patients had resolution of BC, 9.7% had re LT and 32.9% patients died in follow-up. Main causes of death in late BC group were Hepatitis C (29%) and hepatic artery thrombosis (HAT) (14%). Late BC were associated with simultaneous acute cellular rejection (ACR) (21%), chronic rejection (7%) or recurrent primary sclerosing cholangitis (3.5%).

On stepwise multivariate Cox regression, use of T-tube (hazard ratio [HR] 2.0, 95% confidence interval [CI] =1.0-4.0), HAT (HR: 5.3, 95% CI=1.8-15.4), use of Cyclosporine (vs. Tacrolimus) (HR: 3.7; 95% CI=2.1-6.6) or steroids (vs. no use) at 4 weeks post LT (HR: 2.0; 95% CI=0.9-4.4), female donor gender (HR: 1.6; 95% CI=0.94-2.75) and black recipient race (HR: 1.9; 95% CI=0.8, 4.2) were predictors for late BC. Patients with early BC had longer ICU stay, higher rates of ACR, HAT and infections, higher recipient BMI and donor age, longer warm ischemia time and association with re LT.

Between the 3 groups there were no differences in regards to MELD score, cold ischemia time, intraoperative hepatic artery flow, intraoperative blood transfusion, donor BMI, B or T cell positive crossmatch and Sirolimus use at 3 months.

Conclusions

While early BC are related to perioperative events, late BC seem to be related to disease recurrence or an immunological event. Most late BC do not require surgical repair.

« Back to 2013 American Transplant Congress