Are prothrombin time and clot waveform analysis useful in detecting a bleeding risk in liver cirrhosis?

Introduction: Prothrombin time is thought to be unreliable in cirrhotic patients to predict the risk of bleeding. We investigated whether prothrombin time ratio was an independent risk factor for bleeding alongside its clot waveform analysis. Methods: We studied 307 consecutive cirrhotic patients and 115 healthy subjects. A coagulometer was used for detecting both prothrombin time and clot waveform analysis which included velocity (1st derivative) and acceleration (2nd derivative) of clot formation, and area of parabolic segment of the 1st and 2nd derivatives of prothrombin time (entire cycle of the clot formation). Results: Logistic regression shows that prothrombin time ratio was the only variable significantly associated with the history of bleeding. Using a hemorrhagic score, the stepwise model included prothrombin time ratio and the area of parabolic segment of the 1st derivative of Prothrombin Time. Odds ratio was used to create a new score to be challenged against the hemorrhagic score in a ROC analysis. The AUC was 0.72, 95% CI: 0.67-0.77. Conclusion: Prothrombin time ratio is associated to an increased bleeding risk. Its role may be further emphasized considering clot waveform analysis. The new score, if aggregated to prothrombin time ratio, could be useful to provide a single parameter immediately ready to assess the bleeding risk in the individual cirrhotic patient.