The role of diagnostic imaging and liver biopsy in the diagnosis of focal nodular hyperplasia in children.

BACKGROUND & AIMS: Focal nodular hyperplasia (FNH), a benign liver tumour, has a characteristic appearance on diagnostic imaging (DI) and histology. The role of liver biopsy in children for the diagnosis of FNH is unclear. This study investigates the diagnostic accuracy of DI for FNH in children without comorbidities, compared to liver biopsy.

METHODS

A total of 304 consecutive patients (age <18 years) with a biopsied liver mass were retrospectively ascertained (1990-2010). Individuals with a history of malignancy, liver disease or syndromes with increased malignancy risk were excluded. DI and biopsy data were reviewed.

RESULTS

After excluding 205 cases, 99 liver masses were studied. Based on histology, the most common diagnosis was hepatoblastoma (46/99, 44%) followed by FNH (23/99, 23%). The mean age at FNH diagnosis was 11.1 ± 5.2 years, with female preponderance (78%), and a median follow-up of 1.35 years (interquartile range 0.54, 4.20 years). 19/23 biopsy-proven FNH met standard criteria for FNH on DI. In 4/23 cases of biopsy-proven FNH, imaging did not suggest FNH. Two false positive cases included adenoma and fibrolamellar hepatocellular carcinoma. On review of original reports, DI had 82.6% sensitivity and 97.4% specificity for the diagnosis of FNH. On blind review, the sensitivity of DI for FNH diagnosis was 81.3% for MRI (13/16), and 53.3% for CT (8/15).

CONCLUSIONS

In this cohort of children with liver masses and no comorbidities, a diagnosis of FNH by imaging was highly specific, and MRI was the most sensitive study for its diagnosis. Liver biopsy may be deferred in selected children if the DI, particularly MRI, is indicative of FNH.