Focal nodular hyperplasia (FNH) is the second most common benign liver tumor (after hemangioma), accounting for 8% of all primary hepatic tumors. Histologically it contains hyperplastic hepatocytes, malformed arterial vessels, bile ductules, and a central fibrous scar. It typically occurs in women aged 20-50 (female-to-male ratio 8:1). Oral contraceptive use may cause size increase but there is no risk of neoplastic transformation. FNH is not a premalignant lesion and surgical treatment is generally not required. Lesions are usually solitary (80-95%) and <5 cm in diameter. On multiphasic CT and MRI it shows characteristic enhancement: homogeneous hypervascular enhancement in the arterial phase, iso-to-hyperintense appearance in portal venous and delayed phases. Contrast retention in the hepatobiliary phase with gadoxetic acid is pathognomonic due to the presence of functioning hepatocytes within the lesion.
Age Range
20-50
Peak Age
35
Gender
Female predominant
Prevalence
Common
FNH is a hyperplastic response to a congenital or acquired arterial malformation — it is not a true neoplasm. At the center of the lesion are abnormal, thick-walled arterial vessels that provide increased blood flow to surrounding hepatocytes. This increased arterial perfusion leads to hepatocyte hyperplasia (increase in number). The central scar forms from the aggregation of these abnormal arteries and accompanying fibrous tissue; the scar also contains bile ductules and Kupffer cells. The reason for homogeneous hypervascular enhancement in the arterial phase on imaging is that the entire lesion is supplied by arterial blood supply — portal venous input is minimal or absent. The myxomatous/edematous stroma within the central scar causes a hyperintense scar appearance on T2-weighted MRI; this distinguishes it from the hypointense scar of fibrolamellar HCC. Hepatocytes within the lesion are functional and express organic anion transporter proteins (OATP1B1/B3); therefore gadoxetic acid (Eovist/Primovist) is actively retained by FNH hepatocytes in the hepatobiliary phase, making the lesion hyperintense — this is the pathognomonic finding that distinguishes FNH from adenoma and metastasis.
The FNH lesion appearing hyperintense in the hepatobiliary phase with gadoxetic acid through functional OATP transporter-mediated contrast retention — the most specific and pathognomonic finding for FNH diagnosis (95-100% sensitivity, 92-100% specificity). Adenomas, metastases, and most malignant lesions appear hypointense in the hepatobiliary phase.
Homogeneous, prominent hypervascular enhancement throughout the entire lesion in the arterial phase. The lesion enhances significantly more than surrounding liver parenchyma with well-defined margins. The central scar area does not enhance in this phase and appears hypodense. The enhancement pattern is homogeneous and diffuse — different from the peripheral nodular enhancement of hemangioma or heterogeneous enhancement of HCC. Even in large FNH, homogeneity of enhancement is maintained, which is an important distinguishing feature.
Report Sentence
Well-defined lesion in the liver demonstrating homogeneous hypervascular enhancement in the arterial phase, consistent with focal nodular hyperplasia (FNH).
In the portal venous phase, the lesion appears isodense or mildly hyperdense to surrounding liver parenchyma. The prominent hypervascularity from the arterial phase disappears and the lesion equilibrates with the parenchyma. The central scar may still be hypodense in this phase but begins to enhance in the delayed phase. This isodense appearance is known as FNH 'vanishing' (stealth lesion) in the portal venous phase, and diagnosis may be missed with only non-contrast or portal venous phase imaging.
Report Sentence
In the portal venous phase, the lesion becomes isodense to surrounding liver parenchyma with disappearance of enhancement difference; this pattern is consistent with FNH.
On hepatobiliary phase images obtained 20 minutes after gadoxetic acid (Eovist/Primovist) injection, the FNH lesion appears isointense or hyperintense relative to surrounding parenchyma. This occurs because FNH hepatocytes express functional OATP1B1/B3 transporter proteins, actively taking up gadoxetic acid and secreting it into bile canaliculi. The central scar does not enhance and appears hypointense, creating a 'donut-like' appearance. This finding provides 95-100% sensitivity and 92-100% specificity for FNH diagnosis.
Report Sentence
In the hepatobiliary phase, the lesion demonstrates hyperintense appearance relative to surrounding parenchyma, reflecting functional hepatocyte content and strongly supporting the diagnosis of focal nodular hyperplasia.
The central scar appears markedly hyperintense on T2-weighted images — this is due to high water content of myxomatous stroma, vascular channels, and edematous tissue within the scar. The lesion parenchyma of FNH is isointense or mildly hyperintense on T2. The T2 hyperintensity of the central scar reliably differentiates the FNH scar from the T2-hypointense scar of fibrolamellar HCC (fibrous, low water content). The scar may show a stellate (star-shaped) configuration.
Report Sentence
A hyperintense central scar is seen at the center of the lesion on T2-weighted images, consistent with myxomatous stroma; this finding strongly supports FNH.
On T1-weighted images, the FNH lesion appears isointense or mildly hypointense relative to liver parenchyma. The central scar is hypointense on T1. The lesion shows homogeneous signal and contains no hemorrhage or fat — this feature distinguishes it from adenoma. No signal drop occurs on opposed-phase (no intracellular lipid).
Report Sentence
The lesion demonstrates isointense appearance to liver parenchyma on T1-weighted images with no evidence of hemorrhage or intracellular lipid.
FNH does not show significant diffusion restriction on DWI. ADC values are close to normal liver parenchyma (typically 1.2-1.6 × 10⁻³ mm²/s). This finding reflects that FNH consists of well-differentiated hepatocytes — different from the marked restriction seen in highly cellular malignant tumors. DWI plays a supportive role in differentiating FNH from malignant lesions (HCC, metastasis).
Report Sentence
No significant diffusion restriction is detected in the lesion on DWI with ADC values close to normal liver parenchyma; this finding favors a benign lesion.
Radially branching arterial flow from the center to the periphery of the lesion on color Doppler US — spoke-wheel pattern. The central feeding artery is prominent and branches distribute in a stellate configuration throughout the entire lesion. Low resistance arterial waveform is observed on spectral Doppler. On B-mode US, the lesion is usually isoechoic or mildly hypoechoic with homogeneous internal echoes. This pattern is highly characteristic for FNH but can only be clearly demonstrated in 50-70% of cases.
Report Sentence
Color Doppler US examination demonstrates spoke-wheel pattern arterial flow with radial branching from center to periphery in the lesion, consistent with FNH.
In the delayed phase (3-5 minutes), the central scar begins to enhance and appears hyperdense. This is explained by slow contrast retention of fibrous tissue within the scar. In the delayed phase, the lesion parenchyma remains isodense to surrounding liver. Central scar enhancement is an additional finding supporting FNH.
Report Sentence
Enhancement of the central scar is observed in the delayed phase, consistent with fibrous tissue retention.
Criteria
Central scar + spoke-wheel arterial pattern + homogeneous hypervascular enhancement + hepatobiliary phase uptake (80% of all cases)
Distinct Features
Standard form showing all typical findings. Diagnosis is usually definitive by imaging, no biopsy needed. Central scar visible in 60-80% of lesions >3 cm.
Criteria
No or inconspicuous central scar, atypical enhancement pattern or absence of expected findings (20% of cases)
Distinct Features
Central scar may not be visible in small lesions (<3 cm). Hepatobiliary phase uptake is still preserved and is the most reliable criterion. Steatotic FNH or telangiectatic FNH are included in this group.
Criteria
Two or more FNH lesions (5-20% of cases); may coexist with hepatic hemangiomas (FNH-hemangioma syndrome)
Distinct Features
Each lesion independently shows typical FNH features. Develops on the background of arterial malformation. May be associated with hereditary hemorrhagic telangiectasia (HHT/Osler-Weber-Rendu).
Distinguishing Feature
Adenoma appears hypointense in the hepatobiliary phase (no OATP expression — especially HNF1α-inactivated type), FNH is hyperintense. Adenoma may contain hemorrhage and intracellular fat (T1 hyperintensity, opposed-phase signal drop); these are absent in FNH.
Distinguishing Feature
Central scar of fibrolamellar HCC is hypointense on T2 (fibrous tissue), FNH scar is hyperintense on T2 (myxomatous). Fibrolamellar HCC scar may contain calcification. Fibrolamellar HCC is hypointense in hepatobiliary phase, FNH is hyperintense.
Distinguishing Feature
Hemangioma shows peripheral nodular enhancement and centripetal fill-in; FNH shows homogeneous diffuse enhancement. Hemangioma is very bright on T2 (light-bulb sign), FNH is isointense to mildly hyperintense on T2.
Distinguishing Feature
HCC shows washout in portal venous or delayed phase after arterial enhancement; FNH does not show washout, remains isodense. HCC is hypointense in hepatobiliary phase, FNH is hyperintense. HCC usually occurs in cirrhotic liver.
Distinguishing Feature
Cholangiocarcinoma enhancement starts peripherally and increases in the delayed phase (desmoplastic reaction); FNH enhances homogeneously in arterial phase and becomes isodense in portal venous phase. Cholangiocarcinoma may show bile duct dilatation and capsular retraction; these are absent in FNH.
Urgency
routineManagement
surveillanceBiopsy
Not NeededFollow-up
no-follow-upFNH is a benign, non-premalignant lesion that requires no treatment or follow-up when typical imaging findings are present. Definitive diagnosis can be made with gadoxetic acid MRI (95-100% accuracy). Observation alone is sufficient in asymptomatic patients. Surgical resection is considered only for symptomatic lesions or diagnostic uncertainty. Risk of rupture and malignant transformation is practically nonexistent.
FNH requires no treatment. There is no risk of malignant transformation. May grow with oral contraceptive use but discontinuation is not mandatory. Surgery is considered only for large symptomatic lesions. Follow-up is generally unnecessary.