Focal Nodular Hyperplasia

Fig. 12.1

(a) Left lateral sectionectomy performed for FNH, surgical specimen. (b, c) Central fibrous scan with fibrotic arterial vessel in surgical specimen

Despite its polyclonal origin, FNH is usually considered a benign hepatocellular tumor due to its mass-forming presentation. Besides this typical form of FNH (80 %), several variants are described and commonly classified as “atypical FNH” (20 %). Abnormal histoarchitecture or vascular malformations may be absent in atypical forms, but bile duct proliferation is always present [5]. Atypical FNH is a heterogeneous group including FNHs without central fibrous scar, FNH with prominent steatosis, or lesions presenting telangiectasia or features of adenoma. Interestingly, molecular studies demonstrated that among “atypical FNHs,” lesions displaying telangiectatic changes are clonal processes [6].

Most important risk factors for developing FNH are the following:

Gender. Female to male ratio is between 8:1 and 12:1 [2].
Oral contraceptives and estrogens. Several studies have been conducted to rule out a possible association of focal nodular hyperplasia with oral contraceptives. It was first proposed that oral contraceptives may not promote de novo formation of focal nodular hyperplasia [7] but just increase tumor size. Mathieu et al. showed that neither the size nor the number of focal nodular hyperplasia lesions could be influenced by oral contraceptive use [8]. In a recent study, Rifai et al. [9] followed 20 pregnant women with FNH: during pregnancy, growth of FNH was observed in only 3 patients, while FNH size was stable in 7 patients and even reduced by 11 ± 6 % in 10 patients. So authors concluded that women with FNH are not at risk of significant growth progression or FNH-related complications.
Chemotherapy and radiations. Focal nodular hyperplasia may be a late complication of high doses of alkylating agents and liver radiotherapy, probably as a consequence of vascular endothelium injury and subsequent circulatory disorders. In a recent study, Bouyn et al. [10] showed that incidence of FNH in people who underwent chemotherapy or radiotherapy at a young age is significantly higher compared to the general population (0.45 % vs. 0.02 %; p < 0.001).
Extrahepatic congenital portosystemic shunt (Abernethy malformation). It is a very rare congenital condition: venous flow from splanchnic district bypasses the liver and drains directly into the inferior vena cava, left renal vein, left hepatic vein, iliac veins, or right atrium. This condition is frequently associated to hepatic tumors (40 %) including focal nodular hyperplasia, adenoma and hepatocellular carcinoma. The development of focal nodular hyperplasia may be related to an alteration of liver vascular perfusion [11].
Hereditary hemorrhagic telangiectasia (HHT, Rendu–Osler–Weber disease). It is a genetic disease characterized by widespread cutaneous, mucosal, and visceral telangiectasias that can involve the lung, brain, and liver. Livers with HHT show hepatocellular regenerative activity, and it is a trigger for development of FNH. Prevalence of FNH in patients with HHT is 100-fold greater than in general population [12]. Association between HHT and focal nodular hyperplasia further reinforces the theory of vascular etiology of FNH.

12.3 Clinical Presentation

The majority of patients with FNH are asymptomatic, and FNH is usually detected incidentally during radiologic examinations, autopsies, or operations performed for other indications. Most FNH are solitary, although 20 % of patients may have multiple FNH nodules at diagnosis. Mean size at diagnosis is 3.3 cm (range 1–9.9 cm) [13] even though in literature cases of FNH larger than 20 cm are reported [1]. In blood tests, liver chemistry is usually unaltered, and the serum tumor markers alpha-fetoprotein (AFP), Ca19.9 (GICA), and carcinoembryonic antigen (CEA) are invariably negative, showing that blood tests are not accurate for FNH diagnosis. If FNH is large, it can lead to compression of intrahepatic bile ducts resulting in mildly elevated serum gamma-glutamyl transferase level. When symptoms occur, they may include palpable mass, hepatomegaly, and loss of appetite. Large subcapsular lesions may cause stretching of Glisson’s capsule and therefore produce vague abdominal pain. Large FNH may cause displacement of adjacent organs even though this happens very rarely; indeed, Eris et al. reported a case of gastric obstruction due to focal nodular hyperplasia of the liver [14]. There are several reports describing presentation of FNH with acute abdominal pain due to intratumoral hemorrhage and subsequent hemoperitoneum [15]: this risk, even rare, may occur in the presence of a large exophytic tumor. However, FNH is detected incidentally in most of cases (74 % in the series of Cherqui et al. [16]). Diagnostic confirmation of FNH relies solely on imaging findings if typical features are shown. When clinical, biochemical, or imaging data are not typical of FNH, histological diagnosis is mandatory.

12.3.1 Ultrasound and Contrast Ultrasound

Most of FNH are fortuitously discovered on an ultrasound routine exam. FNH usually appears as a round mass hypoechoic or slightly hyperechoic compared with liver parenchyma. Sometimes it may be detected only because of displacement of the surrounding vessels. The central scar is rarely seen at baseline ultrasound, but it is easily visualized by color Doppler ultrasound [17]. In contrast-enhanced ultrasound (CEUS), FNH presents an early homogeneous filling in the arterial phase and subsequently becomes isoechoic.

12.3.2 Computed Tomography

Computed tomography (CT) can be used to characterize lesions of medium to large size, but it has limitations in the diagnosis of small lesions. On non-contrast examination, FNH is hypodense or isodense and may show occasional mass effect and lobulated contours. After contrast, FNH shows strong homogeneous enhancement during the arterial phase with rapid change to isodensity of the lesion during portal and delayed phases [18]. The central scar on the contrast is typically hypoattenuating during the arterial phase. On CT, the detection rate of the central scar is 35 % for nodes smaller than 3 cm and 65 % for those exceeding 3 cm [19]. An atypical FNH may show less arterial enhancement, absence of a central scar, and the presence of an enhancing pseudocapsule. In these situations, other diagnosis should always be considered, and sometimes biopsy or excision may be indicated.

12.3.3 Magnetic Resonance Imaging (MRI)

Contrast-enhanced MRI is the most accurate imaging method for characterization of FNH. FNH is usually hypo-isointense with respect to normal liver parenchyma on T1-weighted images and slightly hyper-isointense on T2-weighted images. Owing to hypervascularization, the central stellate scar is hypointense on T1-weighted images and hyperintense on T2-weighted images. Because its vascular supply is predominately arterial, following the administration of gadolinium, FNH typically shows intense uniform enhancement during arterial phase sparing the central scar [20]; the lesion usually fades to become isointense to the liver on portal venous and equilibrium phases.

Atypical findings that can be misleading include the following:

Heterogeneous T1 and T2 signal intensity
Hyperintense signal on precontrast T1-weighted images
Hypointense signal relative to the liver during hepatic venous and equilibrium phases suggesting “washout”
Presence of a pseudocapsule
Presence of fat accumulation (a finding usually suggestive of hepatic adenoma or hepatocellular carcinoma [21])

Small FNH lesions often lack a visible center. Grazioli et al. showed that the 80 % of FNH lesions smaller than 3 cm lacked a central scar [22]. Conversely, central scars may be seen in other lesions, including fibrolamellar carcinoma and HCC [23].

12.3.4 Biopsy

When conventional imaging does not allow to determine a certain diagnosis, needle biopsy can be considered. If an adequate specimen is obtained, the diagnosis of FNH can usually be established on hematoxylin and eosin sections [24]. Fabre et al. [25] identified four major diagnostic criteria for focal nodular hyperplasia (including fibrous septa without portal tracts, thick-walled blood vessels, ductular proliferation, and hepatocellular nodules) along with two minor criteria (sinusoidal dilatation and perisinusoidal fibrosis). Fabre classified the results of biopsy in the following:

Definite FNH: 3 or 4 major features
Possible FNH: 2 major and 2 minor features
Negative for FNH 2 or fewer major features without minor features

Fibrous septa were recognized in needle biopsies in 78 % of FNH, abnormal vessels in 61 %, ductular proliferation in 67 %, and hepatocellular nodules in 56 % [25]. In the study of Fabre et al., 22 % of biopsies lacked all major criteria for diagnosis, although the lesions later proved to be atypical FNH. Special stains as well as immunostains can be carried out to confirm the diagnosis of FNH. Specific stainings of copper and copper-binding protein in hepatocytes (i.e., rhodanine and Victoria blue stains) have very high specificity [24]. Immunostaining with monoclonal antibodies to biliary type CK (i.e., CK7 and CK19) also supports the diagnosis of FNH because the presence of ductular structures is among atypical features of FNH.

12.4 Evolution

Focal nodular hyperplasia is in most of cases an asymptomatic incidental finding. Once the nature of FNH has been established with imaging technique or biopsy, no specific treatment is required if the patient is asymptomatic, and most authors advocate a “wait and see” approach [16, 26, 27]. Natural history of FNH is to remain asymptomatic in most of cases. A change of size is relatively frequent, while occurrence of complications is rare.

Change in size. Changes in the size of FNH lesions are common: a study that followed patients with FNH for a mean period of 42 months showed that 71 % of nodules are stable in size, 3 % increase in diameter by more than 30 %, and 27 % even regresses in size over the follow-up period [28]. It has been postulated that lesion regression occurs as a result of thrombosis of the feeding artery, a pathological finding that was confirmed in one large autopsy study [3]. Even though low dose of estrogen is safe to be used in patients with FNH [8], anyway clinicians generally recommend treatment with oral contraceptives discontinuation. Women with FNH are not considered at risk of growth progression during pregnancy [9], but the nodules should be closely monitored during pregnancy.

Transformation in malignant lesion. In literature, there are no documented cases of transformation of a histologically proven FNH into a malignant lesion. Lesions evolving into malignancies should be considered for an initial different diagnosis.

Hemorrhage. While hepatic adenoma has a high risk of rupture, patients with focal nodular hyperplasia have a very low bleeding risk, and rupture is reported only in few case reports [29, 30].

12.5 Therapy

12.5.1 Watchful Waiting

The stability in size of most FNH lesions, the lack of potential for malignant transformation, and the extremely low risk of rupture and hemorrhage support a conservative approach for management of most patients with radiologically convincing or histologically proven FNH. Watchful waiting is currently the most diffused approach if the patient is asymptomatic [16, 27]. Ultrasound examination every 6 months for the first 3 years after diagnosis is generally suggested. The frequency can be reduced once the disease has stabilized with no change in nodule size or number.

Only gold members can continue reading. Log In or Register to continue