Patients with cirrhosis can have abnormalities in laboratory tests reflecting changes in primary haemostasis, including bleeding time, platelet function tests, markers of platelet activation, and platelet count. Such changes have been considered particularly relevant in the bleeding complications that occur in cirrhosis. However, several studies have shown that routine diagnostic tests, such as platelet count, bleeding time, PFA-100, thromboelastography are not clinically useful to stratify bleeding risk in patients with cirrhosis. Moreover, treatments used to increase platelet count or to modulate platelet function could potentially do harm. Consequently the optimal management of bleeding complications is still a matter of discussion. Moreover, in the last two decades there has been an increased recognition that not only bleeding but also thrombosis complicates the clinical course of cirrhosis. Thus, we performed a literature search looking at publications studying both qualitative and quantitative aspects of platelet function to verify which primary haemostasis defects occur in cirrhosis. In addition, we evaluated the contribution of qualitative and quantitative aspects of platelet function to the clinical outcome in cirrhosis and their therapeutic management according to the data available in the literature. From the detailed analysis of the literature, it appears clear that primary haemostasis may not be defective in cirrhosis, and a low platelet count should not necessarily be considered as an automatic index of an increased risk of bleeding. Conversely, caution should be observed in patients with severe thrombocytopenia where its correction is advised if bleeding occurs and before invasive diagnostic and therapeutic procedures. © 2011 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.