Correction of hemostatic abnormalities and portal pressure variations in patients with cirrhosis

We read with great interest the articles by Seijo and GarciaPagan, and Tripodi and Primignani recently published in Hepatology. We feel that these commentaries address an important and timely issue in patients with advanced liver disease, as they focus on the complex problem of hemostatic alterations in patients with cirrhosis, and on their possible therapeutic management. Indeed, patients with cirrhosis have a rebalanced coagulation equilibrium, although this balance is quite feeble and may be tipped either on the bleeding or on the thrombotic side by the occurrence of infections, derangement in renal function, and by the performance of invasive maneuvers. In order to correct the alterations of hematological and coagulation parameters, it is common practice among clinicians to order platelet or plasma transfusions, especially before invasive procedures. A recent survey has shown that, although patients with liver disease represented 7.7% of the inand out-patients receiving transfusions, they used 32% of administered plasma and 13% of the platelets. These findings are relevant because cirrhosis patients seldom obtain an adequate correction of thrombocytopenia with platelet transfusions and numerical improvement in prothrombin time requires transfusion of at least 2-6 units of fresh-frozen plasma. We feel that transfusing cirrhosis patients with several units of plasma to correct alteration in coagulation tests is not useful and may do harm by triggering portal hypertension-related bleeding. In fact, in 1974 Zimmon and Kessler found that the correlation between portal pressure and blood volume in cirrhosis patients was approximately linear. They demonstrated that every 100 mL of blood volume expansion, in a relatively short period of time, predicted a mean increase in portal pressure of 1.4 cm H20, equivalent to 1.03 mmHg. Using data obtained by the Holland and Brooks study, we found that to reach a target International Normalized Ratio (INR) value of 1.5 starting from various INR values (Table 1) required the transfusion of several liters of fresh-frozen plasma, and predicted an expected increase in portal pressure following transfusions that ranges from 15.5 to 25.8 mmHg. Reaching an INR target often reported as safe for invasive procedures or surgery in liver disease patients may actually paradoxically trigger bleeding. These data clearly show that research in the field of coagulation and liver disease should be fostered, and that the use of global coagulation tests such as thromboelastography should be further investigated as alternative measures in order to improve our management of patients with cirrhosis. EDOARDO G. GIANNINI, M.D., PH.D. R. TODD STRAVITZ, M.D. STEPHEN H. CALDWELL, M.D. Dipartimento di Medicina Interna Unit a di Gastroenterologia IRCCS-Azienda Ospedaliera Universitaria San Martino-IST Universit a di Genova Genova, Italy Section of Hepatology Division of Gastroenterology, Hepatology and Nutrition Hume-Lee Transplant Center Virginia Commonwealth University Richmond, VA, USA Division of Gastroenterology and Hepatology Digestive Health Center University of Virginia Charlottesville, VA, USA