New Insights in the Management of Hepatocellular Adenoma

Anne J. Klompenhouwer; Robert A. de Man; Marco Dioguardi Burgio; Valerie Vilgrain; Jessica Zucman-Rossi; Jan N. M. Ijzermans


Liver International. 2020;40(7):1529-1537. 

In This Article

Follow-up of HCA - When to Treat?

Treatment of HCA should be reserved for patients with a high risk of haemorrhage and development of HCC. Subgroups at risk are men with HCA, patients with β-(I)HCA, and patients with HCA showing progressive growth.[13,21,42] Additionally, the 2016 EASL guideline advised to implement lifestyle changes for all female patients with HCA, irrespective of baseline diameter.[11] The most important lifestyle change is cessation of oral contraceptives, as it has been shown even in the 1970s that this can lead to regression of HCA.[43] In the past decade, several studies have also shown that weight loss may lead to regression of HCA in obese patients.[4,44] The EASL guideline states to perform a surgical resection if HCA exceed 5 centimetres six months after implementation of these lifestyle changes, given the higher risk of complications in HCA > 5cm.[42,45] Two recent studies however advocate to await the effect of cessation of oral contraceptives for a longer period of time (at least 12 months and longer for larger HCA) as the majority of HCA will regress over time and no complications occurred during follow up in these cohorts.[46,47]

The chance of regression of HCA to <5 cm appears to be lower in H-HCA as compared to I-HCA.[47] Interestingly, a recent study even showed that H-HCA have a higher rate of progression from <5 cm to >5 cm as compared to I-HCA, despite a lower and shorter oral contraceptive intake.[48] A hypothesis may be the varying oestrogen sensitivity of the different HCA subtypes (higher in I-HCA as compared to H-HCA), but this has yet to be proven.[21,49]

Haemorrhage is the most frequent complication in HCA, and has been reported in up to 25% of cases.[45] Not all patients are symptomatic, especially intratumoral haemorrhage may go unnoticed. When lifestyle changes are implemented, treatment is often not required as the hematoma will resorb over time and the tumour will regress. However, in some cases massive bleeding may occur resulting in intraparenchymal haemorrhage, subcapsular hematoma or even hemoperitoneum caused by rupture of the liver capsule. When massive bleeding occurs patients may present with hemodynamically unstable conditions. In the acute phase, conservative management is justified when hemodynamic stabilization can be reached.[50] In case of persistent hemodynamic instability or active bleeding, transarterial embolization (TAE) is the preferred management.[51] Liver resection is not advisable in the acute phase as it is associated with increased morbidity and mortality.[50,52]

Pregnancy used to be discouraged in patients with unresected HCA, because of the risk of hormone induced growth and rupture during pregnancy. In 2004, a study was published reporting the mortality risk of ruptured HCA during pregnancy: 44% for the mother and 38% for the foetus.[53] The majority of cases included in this review dated from the 1970s and 1980s. However, in 2011 it was shown that a large number of patients who were diagnosed with HCA, already had been pregnant and had uneventful pregnancies.[54] This initiated a study that was recently published, assessing the risk of growth and haemorrhage of HCA <5 cm during pregnancy. In this study, growth occurred in a quarter of cases but no haemorrhage occurred.[55] No subgroups at risk for growth could be identified in this cohort. Given the fairly high proportion of patients with growing HCA, close monitoring during pregnancy with ultrasound is recommended, enabling an intervention in case of progressive growth.[55] Future research should focus on trying to identify subgroups at risk of pregnancy-related complications.

In addition, given the sensitivity of HCA to hormones in fertile women, a study was performed questioning whether surveillance of HCA is still required in post-menopausal women.[56] The study showed that HCA become smaller after menopause and that routine follow-up of small HCA (<5 cm) is not required.[56]

When a patient is diagnosed with HCA, a great number of factors should be taken into account when deciding whether the patient should undergo a resection or whether a wait-and-see policy is legitimized. The most important patient related factors to be considered include sex, age and co-morbidity. For instance, resection is advised in men given the far higher risk of malignant transformation (estimated 50% in men as compared to 5% in females).[23,42,57] Additionally, although rare, resection may be advised in patients with HCA and hepatitis B or C infection, given the a priori higher risk of HCC. The key tumour factor that should be considered is the HCA subtype: patients with b-(I)HCA and sh-HCA are at greater risk of complications and therefore surgical resection is preferred. When malignancy is suspected on imaging a resection should be performed. A wait-and-see policy is legitimized in H-HCA and I-HCA without ß-catenin mutation that show regression with lifestyle changes.

As for every type of liver surgery, the anatomical location of the tumour should be taken into account as well as the quality and volume of the future liver remnant. As a large number of patients with HCA are overweight or obese, the presence and degree of steatosis should be taken into account as these factors impact the perioperative complication- and mortality rates of liver resections.[58] A model combining all these factors would be an ideal solution. A decision curve analysis (DCA) is an example of a model requiring a binary decision, in our case this would be surgery versus wait-and-see. We would have to decide how many patients we would be willing to operate although they will not have tumour-related complications, to avoid one patient getting haemorrhage or malignant transformation. To perform such an analysis we would require detailed information on surgical complications in those treated with a resection and the incidence of malignant transformation and haemorrhage in those treated conservatively. A very large population would be needed to estimate the risks and benefits of both treatment strategies. Unfortunately, considering the low incidence of HCA and the rare indications for surgery, it will be hard to realize a prospective study. Furthermore, the surgical expertise will definitely play a major role in the outcome of such a study. Referring these patients to expert liver centres may offer the best-case scenario for diagnostic and surgical management at present.