Abstract
The hook effect is a well-recognized problem that can occur in assays of most tumour markers, including alpha feto-protein (AFP). We present a case of hepatoblastoma in a baby. The diagnosis was delayed as a result of unrecognized 'hooking' of a very high AFP concentration in the automated immunoassay method used. The falsely low result obtained was considered normal for the patient's age and supported the diagnosis of benign haemangioendothelioma. Liaison between clinical and laboratory staff was critical in obtaining an accurate AFP result, proceeding to liver biopsy and establishing the definitive diagnosis of hepatoblastoma. While the reasons and solutions for hook effect have been well researched and published, we believe the presence of extremely high serum AFP concentration in some hepatoblastoma patients means that the hook effect remains a problem and can generate erroneously low AFP results despite assay reformulation by manufacturers. Therefore constant vigilance by laboratory staff is still needed.