In our 2 patients, investigations performed to display any cause of PVT were negative. As recommended for FNH, a conservative management was decided. Two years later, patients are asymptomatic. Number and size of the lesions were unchanged. This report describes 2 patients with a history of portal vein thrombosis, portal cavernoma and developing FNH- like lesion. Unfortunately, we cannot state when theses lesions occurred because our 2 patients were lost to follow up after portal cavernoma has been diagnosed. FNH is a benign tumour that is usually an incidental finding and concerns mostly young women (sex ratio 12/1). The prevalence in France is 1/3000. Macroscopically, FNH is a circular, lobulated, well-circumscribed lesion, free of fibrous capsule and characterized by a central stellate fibrous scar containing large vessels. In microscopy, it consists of a hyperplasia of the liver parenchyma with benign-appearing hepatocytes, organized in nodules separ- ated by fibrous septa, surrounding large vessels in the central scar often accompanied by bile ductular proliferation and inflammatory infiltration. Pathogenesis is not clearly elucidated. Analyses suggest that FNH is a polyclonal thus reactional lesion (rather than a tumoral one) [1]. Abnorm- alities of hepatic circulation have been suggested for many years as possible etiological factors [2]. Wanless IR et al. suggested that FNH is a hyperplastic response of the liver parenchyma to an increase or a disturbance of hepatic blood flow due to a pre-existing vascular malformation [3]. It should be noticed that FNH has been associated with other vascular malformations. However, some authors [4] con- sider that FNH is derived from acquired thrombosis. Thrombosis of the hepatic artery and/or portal vein being the cause of hepatic necrosis, reperfusion following hepatic arterial recanalization would result in nodule formation. The association of FNH with Budd–Chiari syndrom has already been described [5], but to our knowledge the association of FNH and portal vein thrombosis has never been reported. It is difficult to estimate the prevalence of FNH-like lesions in patients with portal cavernoma because in our unit, few patients are followed for more than 20 years after portal vein thrombosis has been discovered. European Network for Vascular Disorders of the Liver (En-Vie) should clarify this data. In our 2 patients, the increase in arterial blood flow secondary to PVT could take part in the development of the FNH. That supports the vascular theory of FNH genesis. The following question still remains: is there a pre-existing congenital abnormality, which results in vascular thrombosis? C. Bureau 1 , J.M. Pe ´ron 1 , E. Sirach 1 , J. Selves 2 , P. Otal 3 , J.P. Vinel 1 1 Service d’He ´pato-Gastro-Ente ´rologie, Fe ´de ´ration Digestive Purpan et Inserm U531, 31059 Toulouse Cedex, France 2 Service d’Anatomie et de Cytologie Pathologiques, CHU Purpan, Toulouse, France 3 Service de Radiologie Rangueil, CHU Rangueil, Toulouse, France E-mail address: bureau.c@chu_toulouse.fr References [1] Paradis V, Laurent A, Flejou JF, Vidaud M, Bedossa P. Evidence for the polyclonal nature of focal nodular hyperplasia of the liver by the study of X-chromosome inactivation. Hepatology 1997;26:891–895. [2] Kondo F. Benign nodular hepatocellular lesions caused by abnormal hepatic circulation: etiological analysis and introduction of a new concept. J Gastroenterol Hepatol 2001;16:1319–1328. [3] Wanless IR, Mawdsley C, Adams R. On the pathogenesis of focal nodular hyperplasia of the liver. Hepatology 1985;5:1194–1200. [4] Kumagai H, Masuda T, Oikawa H, Endo K, Endo M, Takano T. Focal nodular hyperplasia of the liver: direct evidence of circulatory disturbances. J Gastroenterol Hepatol 2000;15:1344–1347. [5] Schilling MK, Zimmermann A, Redaelli C, Seiler CA, Buchler MW. Liver nodules resembling focal nodular hyperplasia after hepatic venous thrombosis. J Hepatol 2000;33:673–676. doi:10.1016/j.jhep.2004.04.025 Prompt relapse of viremia after lamivudine discontinuation in e-minus chronic hepatitis B patients completely responders during 5 years of therapy To the Editor: In patients with e-minus chronic hepatitis B (e-CHB) a sustained biochemical and virological remission after therapy discontinuation is considered the main therapeutic goal [1]. In e-CHB lamivudine is effective to reduce hepatitis B virus (HBV) replication; in a previous study 52 weeks of lamivudine induced a virological response associated with transaminases normalization in the 65% of patients [2]. However, after lamivudine discontinuation, during long-term follow-up, disease and viremia relapsed in more than 90% of the responders [3]. In these patients, HBV reactivation can lead to liver decompensation and death [4–5]. At the moment, in most of the hepatologic centres, lamivudine is indefinitely prolonged in responders. No data are available on the risk of virological relapse after treatment discontinuation in long term complete responder patients. Letters to the Editor / Journal of Hepatology 41 (2004) 499–504 500