Thrombotic Ringed Polytetrafluoroethylene Graft With Infection After Living-Donor Liver Transplantation

The safety of the living donor in living-donor liver transplantation (LDLT) is always the first priority, meanwhile, the graft-to-recipient weight ratio (GRWR) and the anatomy of the liver allograft must also not be compromised in order to warrant tranplatation success. When it comes to the allograft of the right lobe of the liver without the middle hepatic vein (R-M), the outflow and adequate drainage for the territory of middle hepatic vein (MHV) is one critical concern. Despite publications in some high-volume transplant centers on the positive results of using expanded polytetrafluoroethylene (ePTFE) grafts to substitute those of autologous veins, complications related to the ePTFE graft have not been well discussed.

From July 2012 to June 2016, 129 adult patients who underwent living donor liver transplantation in Taipei Veterans General Hospital were analyzed. There were 3 cases of adjacent organ erosion with gas bubbles in the lumen of an ePTFE graft, including gastrointestinal (GI) tract penetration in 2 out of the first 15 cases that used the venous graft of ringed expanded polytetrafluoroethylene (rPTFE). The patient survival rate during this period was compared and radiological findings of rPTFE function and clinical signs of erosion with infection were also examined to raise the concerns of safety as well as early detection of complications of rPTFE.

The overall 1-year patient survival rate was 90%, of which the right lobe wih MHV (R+M) group was 93.5% and the R-M group was 91.9%. For the mean of GRWR, the R+M group was 1.05 ± 0.19 and R-M group was 1.19 ± 0.27, while those who needed reconstruction with vein grafts was 0.96 ± 0.11. Among the R-M group, 24 out of 88 cases (27.3%) needed reconstruction of MHV tributaries. Of the 24 cases, 15 cases were done with rPTFE and the 1-year patient survival rate of the rPTFE group was 73%, which is significantly worse (P = .008) than the non-rPTFE (89%) and non-reconstructed (97%) groups. The mean GRWR is significantly higher (P = .001) in the non-reconstructed group (1.19 ± 0.27) than in the rPTFE (0.99 ± 0.11) and non-rPTFE (0.94 ± 0.11) groups. The venous grafts patency rate between the different graft types is no different, and there is also significance in warm ischemic time (P = .009) between the non-reconstructed (49 ± 15), rPTFE (81 ± 51), and non-rPTFE (56 ± 18) groups in the mean minutes.

In cases of fever of unknown cause in patients receiving LDLT with rPTFE graft, a regular computed tomography (CT) scan with contrast and gas bubbles within the graft lumen is the best way for early detection of graft related infection and suspicious GI tract penetration. To decrease the risks of tissue reaction induced by ePTFE graft in LDLT, omentum patches or other inert agents can be introduced as a buffer between the graft and adjacent organs, especially the GI tracts. However, research in material science shall be explored to solve the problem in the future.