Single-center experience of 39 patients with preoperative portal vein thrombosis among 404 adult living donor liver transplantations

Abstract

Portal vein thrombosis (PVT) is an uncommon complication of cirrhosis occurring in approximately 10% of patients.1, 2 However, the frequency in liver transplantation candidates with decompensated disease or hepatocellular carcinoma is likely higher. Consequently, patients on the liver transplantation list should routinely undergo screening for PVT with recurrent cross-sectional imaging and/or hepatic Doppler ultrasonography. While most patients with PVT may undergo successful deceased-donor liver transplantation, complications may be higher, especially in patients with extensive thrombosis.3-5 However, the effect of PVT on outcomes after adult-to-adult living donor liver transplantation (LDLT) has largely been limited to case reports. A recent study suggested that most centers, when responding to a survey, describe a conservative approach to PVT in LDLT recipients.6 Chronic PVT was reported in only 1% of LDLT recipients and 62% of centers considered PVT a relative or direct contraindication for LDLT. The current report by Egawa et al.7 is particularly helpful in that it is probably the most detailed description of PVT in LDLT recipients available in the literature. These investigators reported outcomes in 39 of 404 LDLT recipients with PVT at the time of transplantation. The authors provide a careful description of the clinical and anatomic characteristics of these patients, as well as the volume of blood loss and in-hospital mortality rate. PVT, portal vein thrombosis; LDLT, living donor liver transplantation; MELD, Model for End-Stage Liver Disease. What can be learned from this experience? First, the severity-of-illness in these patients may be significantly higher than LDLT recipients in the United States, where the mean Model for End-Stage Liver Disease (MELD) score at the time of transplant is 14. In the Egawa et al.7 series, the mean MELD in the PVT patients was 18.2. Of the recipients, 59% were Child-Turcotte-Pugh class C, 59% had a MELD score >15, and 28% had a MELD score >25. Because LDLT represents the only option for liver transplantation in Japan, sicker patients may be routinely selected for this procedure compared to the United States, where deceased-donor liver transplantation is available. Second, preoperative computed tomography scan failed to diagnose PVT in 22% of patients before surgery. However, all of the patients with “missed” PVT only had partial thrombosis. In fact, most of the patients in this report (29/39; 74%) had partial vein thrombosis. The clinical significance of a partial PVT is difficult to determine; intraoperative bleeding in these patients was not higher than patients without PVT. Next, the proportion of LDLT recipients with PVT in this series (10%) is much higher than the survey by Kadry et al.6 (1%) in 2002 and is similar to deceased-donor liver transplantation recipients. Perhaps the most important finding is the poor outcomes and high mortality rates in the patients with complete thrombosis. The volume of blood loss in patients with complete thrombosis was significantly higher by 3-fold compared to partial thrombosis. More important, the in-hospital mortality rate associated with complete thrombosis (6/10; 60%) was significantly higher by more than 3-fold compared to patients with partial thrombosis (5/29; 17%). However, there may have been significant confounding factors associated with this comparison. The authors point out that the in-hospital death was higher in recipients with a MELD score >25 (6/11; 55%), that 4 of the patients with complete occlusion had a MELD score >25, and that 3 of these 4 patients died. Based on these marginal outcomes, the authors rightly conclude that candidates with complete PVT and MELD score >25 may be poor candidates for LDLT. In summary, the authors have provided a useful contribution in selection of patients for LDLT. Based on their findings, PVT should be considered a relative contraindication for LDLT. That is, the significance of PVT in a specific patient should be considered relative to the extent and severity of the PVT as well as other medical problems in the patient. Ideal candidates with partial (or complete) PVT may be able to undergo successful LDLT. However, candidates with complete thrombosis are at higher risk of experiencing more intraoperative bleeding and in-hospital mortality, especially those with severely decompensated disease. Consequently, these patients should be considered with caution for this procedure. The ultimate decision in recipient selection rests on the experience and judgment of the transplant surgeon and team. Because of the uncommon occurrence of PVT in LDLT recipients, definitive and more detailed recommendations await additional reports from other centers.