TAFI deficiency in liver cirrhosis: relation with plasma fibrinolysis and survival.

INTRODUCTION: TAFI (thrombin-activatable fibrinolysis inhibitor) is a potent anti-fibrinolytic and anti-inflammatory factor of liver origin. It is markedly reduced in liver cirrhosis but its effect on fibrinolysis remains controversial and no data are available on its prognostic value. We evaluated the relationship of TAFI level with plasma fibrinolysis and survival in cirrhotic patients. PATIENTS AND METHODS: Sixty-five patients with liver cirrhosis were studied. TAFI antigen, plasma fibrinolysis and other laboratory variables were assayed at study entry and their association with mortality was assessed during a 3-year follow-up. RESULTS: TAFI level and fibrinolysis time were markedly reduced in liver cirrhosis as compared to healthy subjects (p<0.0001) and TAFI deficiency was strongly correlated with fibrinolysis time (p=0.0002). TAFI level at entry, but not fibrinolysis time, was significantly lower in non-survivors (n=25) than in survivors (n=40, p=0.0001). By Cox regression analysis, after adjustment for possible confounding factors, TAFI, but not fibrinolysis time, was identified as an independent predictor of mortality. TAFI assay, moreover, showed a clinically relevant accuracy in assessing patients' survival (ROC curve analysis, p<0.0001) achieving a sensitivity of 92%, a specificity of 55%, and a negative predictive value of 91.7%. CONCLUSIONS: Our data indicate that TAFI deficiency in liver cirrhosis is associated with enhanced plasma fibrinolysis. Moreover, they suggest that TAFI, but not fibrinolysis time, is a strong predictor of survival and thus TAFI assay might prove useful to select candidates for liver transplantation.

Hyperprothrombinaemia-induced APC resistance: differential influence on fibrin formation and fibrinolysis.

The prothrombin gene mutation G20210A is a common risk factor for thrombosis and has been reported to cause APC resistance. However, the inhibition of thrombin formation by APC not only limits fibrin formation but also stimulates fibrinolysis by reducing TAFI activation. We evaluated the influence of prothrombin G20210A mutation on the anticoagulant and fibrinolytic activities of APC (1 microg/ml). Thirty-two heterozygous carriers and 32 non carriers were studied. APC anticoagulant activity was assessed by aPTT prolongation whereas APC fibrinolytic activity was determined by a microplate clot lysis assay. APC-induced aPTT prolongation was markedly less pronounced in carriers than in non carriers. On the contrary, fibrinolysis time was shortened by APC to a comparable extent in both groups. Accordingly, prothrombin levels were strongly correlated with APC-induced aPTT prolongation but not with APC-induced shortening of lysis time. The addition of purified prothrombin to normal plasma (final concentration 150%) caused APC resistance in the clotting assay over the whole range of tested APC concentrations (0.125-1.5 microg/ml). In the fibrinolytic assay, instead, prothrombin supplementation made the sample resistant to low but not to high concentrations of APC (>0.5 microg/ml). Thrombin and TAFIa determination in the presence of 1 microg/ml APC revealed that hyperprothrombinemia, although capable of enhancing thrombin generation, was unable to induce detectable TAFIa formation. It is suggested that APC resistance caused by hyperprothrombinaemia does not translate in impaired fibrinolysis, at least in the presence of high APC levels, because the increase in thrombin formation is insufficient to activate the amount of TAFI required to inhibit plasminogen conversion. These data might help to better understand the relationship between thrombin formation and fibrinolysis down-regulation.

Hyperprothrombinemia associated with prothrombin G20210A mutation inhibits plasma fibrinolysis through a TAFI-mediated mechanism.

The prothrombin gene mutation G20210A is a common risk factor for thrombosis and is associated with increased prothrombin levels. However, the mechanism whereby hyperprothrombinemia predisposes to thrombosis remains unclear. Because thrombin is the physiologic activator of TAFI (thrombin activatable fibrinolysis inhibitor), the precursor of an antifibrinolytic carboxypeptidase (TAFIa), we evaluated the influence of hyperprothrombinemia on fibrinolysis. Thirty-two heterozygous carriers of the G20210A mutation and 30 noncarriers were studied. Plasma fibrinolytic factors and TAFI levels were similar in the 2 groups. Mean lysis time of tissue factor-induced plasma clots exposed to 25 ng/mL exogenous tissue-type plasminogen activator (t-PA) was significantly longer in 20210A carriers than in control donors. This difference disappeared on addition of a specific inhibitor of TAFIa. Determination of thrombin and TAFIa activity, generated during clot lysis, revealed that G20210A mutation was associated with a significant enhancement of late thrombin formation and an increase in TAFI activation. Plasma prothrombin level was highly significantly correlated with both clot lysis time and TAFI activation. The addition of purified prothrombin, but not of factors X or VIIa, to normal plasma caused a concentration-dependent, TAFI-mediated inhibition of fibrinolysis. These findings provide a new mechanism that might contribute to the thrombotic risk in prothrombin 20210A carriers.

Deficiency of thrombin activatable fibrinolysis inhibitor in cirrhosis is associated with increased plasma fibrinolysis.

Hyperfibrinolysis is thought to contribute to bleeding associated with advanced cirrhosis. Thrombin activatable fibrinolysis inhibitor (TAFI) is a plasma precursor of a carboxypeptidase (TAFIa) with antifibrinolytic activity and was recently shown to be reduced in cirrhosis. In this study, we evaluated the influence of TAFI deficiency on in vitro fibrinolysis in cirrhotic patients. Fifty-three patients with cirrhosis and 43 healthy controls were studied. TAFI antigen was measured by enzyme-linked immunosorbent assay and TAFI activity by chromogenic assay. Fibrinolysis was evaluated as tissue plasminogen activator-induced plasma clot lysis time in the absence and in the presence of a specific inhibitor of TAFIa. TAFI antigen and activity levels were markedly reduced in cirrhotic patients (P <.0001). In these patients, the lysis time of plasma clots was shorter than in controls (median, interquartile range: 25 minutes, 21-36 minutes vs. 48 minutes, 40-60 minutes, respectively; P <.0001) and was poorly influenced by the TAFIa inhibitor. Accordingly, TAFIa and thrombin activity, generated in cirrhotic samples during clot lysis, were significantly lower than in control samples. Addition of purified TAFI to cirrhotic plasma prolonged the lysis time and enhanced the response to TAFIa inhibitor in a dose-dependent manner. In conclusion, our results indicate that in vitro plasma hyperfibrinolysis in cirrhosis is largely due to a defective TAFIa generation resulting from low TAFI levels and probably from impaired thrombin generation. Impairment of the antifibrinolytic TAFI pathway might contribute to bleeding associated with this disease.

Effect of heparin on TAFI-dependent inhibition of fibrinolysis: relative importance of TAFIa generated by clot-bound and fluid phase thrombin.

Heparin has been proposed to enhance thrombolysis by inhibiting thrombin-dependent generation of activated TAFI (thrombin activatable fibrinolysis inhibitor), a carboxypeptidase that inhibits fibrinolysis. We evaluated the effect of heparin in an in vitro thrombolysis model consisting of a radiolabelled blood clot submerged in defibrinated plasma. Fibrinolysis was induced by adding t-PA (250 ng/ml) and calcium to the plasma bath. Control experiments indicated that thrombin generation induced by recalcification caused significant TAFI activation and inhibited clot lysis. Heparin (up to 1 U/ml), added to the plasma bath, failed to enhance clot lysis. Thrombin generation in the fluid phase was totally inhibited by heparin at concentrations > 0.5 U/ml. In contrast, thrombin generation on the clot surface was not inhibited by heparin (1 U/ml). TAFIa generation did occur in heparin-containing samples (1 U/ml) and amounted to about 10% of TAFIa formed in control samples. This low amount of TAFIa did exert antifibrinolytic activity as indicated by the observation that the addition of a specific TAFIa inhibitor (PTI) along with heparin enhanced clot lysis. Hirudin (10 micrograms/ml), at variance with heparin, inhibited clot-bound thrombin and enhanced clot lysis. These data show that heparin is unable to stimulate fibrinolysis through a TAFI-dependent mechanism, most likely because of its inefficiency in inhibiting thrombin generation on the clot surface. Moreover, they suggest that clot-bound thrombin plays a major role in TAFI-mediated inhibition of fibrinolysis through "localized" TAFIa generation.