Portal Vein Thrombosis and Other Venous Anomalies

Prevalence

PVT is thought to be a rare event in the general population, resulting from a variety of causes. A population-based study of more than 23,000 consecutive autopsies has demonstrated PVT to be more common than previously thought, with a population prevalence of 1.1%. This study also found that 28% of these patients had cirrhosis, one third of whom also had a primary hepatic malignancy. Cirrhosis was present in 5% of the population, 6% of which had PVT at the time of autopsy. Most interesting was the finding of malignancy in 63% of individuals with PVT, implicating the importance of neoplasms in its development.

The prevalence of PVT among transplant candidates would presumably be close to that of cirrhotic patients in the general population. According to data from the Scientific Registry of Transplant Recipients (SRTR), the rate of PVT among patients wait-listed for liver transplant is 2.1%. This is lower than the rates previously discussed and also lower than other data suggest. However, the prevalence among transplanted patients is nearly double the rate of listed patients at 4.0%. This would suggest that perhaps there are missed diagnoses based on imaging or there is a high rate of de novo PVT among patients on the waiting list. The relatively low rate may indicate that some patients are excluded from transplant secondary to PVT or the rate of PVT at transplant is underreported.

When considering the incidence of PVT, how it is defined must be taken into account. PVT, as defined earlier, may be partially occlusive and may exist only in an intrahepatic branch or it may be completely occlusive and extend past the confluence into the splenic and mesenteric veins. Yerdel et al classified PVT from grade 1 (<50% thrombosis of the portal vein) to grade 4 (complete portal vein [PV] thrombosis and complete superior mesenteric vein [SMV] thrombosis). Starzl’s group defined grade 1 as partial thrombosis of the intrahepatic portal vein branches. Grade 4 in the Starzl classification equaled a near-complete or complete obliteration of the portal vein trunk, but with a patent SMV. These systems may seem to differ on only minor points, but this may account in part for the wide variability in the prevalence of PVT reported in the literature.

PVT may have its greatest implication when it becomes clinically relevant. PVT was initially considered a contraindication for liver transplant, and patients were excluded from listing based on its presence. As experience grew and early success was documented, the rate of PVT transplants increased significantly. Lerut et al reported on the largest experience to date in 1987, with 393 consecutive orthotopic liver transplants, a 16.3% rate of PV anomalies, 7% of which were thrombosis. The same group in 1992 reported a rate of PVT in 13.8% of transplants. However, only 9% of these were surgically significant, meaning they alter the course of the operation. As experience was gained, more patients with PVT were listed and underwent transplant, but only a portion of these were surgically significant thrombosis.

The average time from listing to transplant may vary widely throughout the world. Some transplant centers have average waiting times greater than 1 year, whereas others may transplant within 6 months of listing the patient. This may be of considerable importance with regard to the incidence and prevalence of PVT and its variability in the listed and transplanted patient population. Francoz et al found a 7.4% incidence of de novo thrombosis in patients awaiting transplant, with a mean waiting-list time of 12 months. This rate is in addition to an incidence of 8.4% PVT at the time of listing. Many of these patients were discovered to have PVT on routine surveillance imaging; however, more than half were discovered at the time of transplant and were previously though to have patent portal veins. This study found that the risk for de novo thrombosis was significantly and independently associated with the length of time from listing to transplant.

The prevalence of PVT varies depending on how it is defined and when it is discovered. It is influenced by severity of liver disease and increases with duration of disease. The next question is what factors are behind its development.

Pathophysiology

Although Rudolf Virchow helped to elucidate the mechanisms behind the development of pulmonary embolism, his name has famously become associated with the factors that contribute to the development of thrombosis: Virchow’s triad. These factors are hypercoagulability, hemodynamic changes, and endothelial injury. These same factors have proven important in the development of PVT. Alterations in portal venous flow and distortions of normal coagulation, in particular, have been shown to be critical in the formation of PVT.

The liver synthesizes coagulation factors, inhibitors, and fibrinolytic proteins. All of these unique factors exist naturally in balance to avoid the generation of excess thrombin. Protein C, protein S, and antithrombin levels are lower in the serum of patients with liver disease. In fact, patients with higher Model for End-Stage Liver Disease (MELD) scores have been shown to have significantly lower levels of protein C and antithrombin when compared to cirrhotic patients with lower MELD scores. This corollary continues to hold true, because the rate of PVT is associated with higher MELD scores. Inherited defects in protein C and S, as well as other defects such as factor V Leiden, have been associated with the development of PVT, yielding further evidence as to the importance of balance between procoagulation factors and inhibitors of coagulation.

Cirrhosis is associated with impairment of hepatic synthesis and a reduction of procoagulants and anticoagulants to an equal degree. This balance can be easily upset, as seen in the operating room with surgical bleeding in a cirrhotic patient or the development of an infection in the hospitalized patient with cirrhosis. But that balance can be disturbed in the opposite direction, resulting in thrombosis. The international normalized ratio (INR), which is used in the transplant population as part of the MELD score, may not accurately assess the bleeding or thrombotic potential of cirrhotic patients. INR was developed to measure the effect of vitamin K antagonists in patients with normal liver function and has been shown to be less reliable in the setting of cirrhosis. Several studies have demonstrated a higher rate of venous thromboembolism, including deep vein thrombosis and pulmonary embolism, in patients with severe liver disease when compared to controls.

Portal venous flow is abnormal in patients with cirrhosis, and increasing perturbation is associated with a rising MELD score. Stasis in the portal vein is caused in part by splanchnic vasodilation found in cirrhosis. This is further exacerbated by the increased resistance in the hepatic vascular bed caused by the architectural derangements of the scarred, cirrhotic liver. Reduced portal venous blood flow velocity has been shown to be independently associated with the development of PVT. This further illustrates the relationship between rising MELD score and increased incidence of PVT.