We read with interest the recently published clinical guidelines on idiopathic thrombocytopenic purpura [1]. Although this paper was not intended as an exhaustive review, we were surprised to find no mention of liver disease in the discussion of associated disorders. Thrombocytopenia is common in chronic liver disease because of increased platelet sequestration in the spleen [2] and increased expression of antiplatelet antibody in some patients [3]. Idiopathic thrombocytopenic purpura has also been associated with chronic hepatitis C [4]. Cirrhosis and portal hypertension may be very subtle and may go unrecognized in these patients.

We have informally reviewed our experience with 10 patients (6 women; mean age ±SD, 57 ± 10 years [range, 36 to 70 years]) recorded in the University of Virginia Liver Registry over the past 5 years in whom idiopathic thrombocytopenic purpura was diagnosed, often years before the diagnosis of cirrhosis. Eight patients had cryptogenic cirrhosis, 1 had chronic hepatitis B, and 1 had nonalcoholic steatohepatitis with cirrhosis. In 4 patients, cirrhosis was not recognized until the patient developed ascites. In 1 of these 4, ascites became clinically apparent only after abdominal surgery; this patient died of cirrhosis-related postoperative complications. Three of the 10 patients have had protal hypertension-related bleeding. We have had similar experiences with many other patients, who were referred to our hematology department with minimal physical findings, mild to moderate thrombocytopenia, and a diagnosis of idiopathic thrombocytopenic purpura. These patients were found to have portal hypertension from various causes, including cirrhosis, portal-vein thrombosis, portal sclerosis, and splenic-vein thrombosis. Antecedent liver enzyme and synthetic indices (prothrombin time and levels of albumin and bilirubin) are often minimally abnormal or even normal in these patients.

The early detection of cirrhosis requires a high index of suspicion in patients with unexplained thrombocytopenia. From a practical standpoint, cirrhosis, even in well-compensated patients, presents additional risks for disease progression, medication-related problems, and surgical complications. As pointed out in the editorial that accompanied these guidelines [5], the lack of firm data hampers diagnosis and treatment of idiopathic thrombocytopenic purpura. However, the association of thrombocytopenia with often subtle forms of chronic liver and portal vascular diseases prompts a reminder to clinicians to be aware of the potential pitfalls of unrecognized portal hypertension.