Postnatal treatment and evolution patterns of giant fetal hepatic hemangioma: a case series of 29 patients

Data analysis

The data were analyzed by IBM SPSS 22.0. Measurement data conforming to a normal distribution are expressed as (®± s), and paired-samples t tests were performed for intragroup comparisons and independent t tests for intergroup comparisons. Data that did not conform to a normal distribution were represented by the median (min ~ max), and a nonparametric test was used for comparisons between groups. Count data are expressed as the number of cases and percentage (%), and the χ2 test or Fisher’s exact probability method was used for comparisons between groups. P < 0.05 was considered statistically significant. The graphs, calculations and statistical analyses were performed using GraphPad Prism software version 9.0 for Mac (GraphPad Software, San Diego, CA).

Group management results

Follow-up and outcome

The median follow-up period was 3.1 years (range: 1.2–5.8 years), and all children survived. All lesions in our cohort eventually decreased in size, with complete or partial regression (Fig. 3). Two of the surgically treated cases completely regressed at 3 months (Fig. 4A and B) and 7 months of age (Fig. 4C and D), while four of the remaining 27 cases showed continued growth after birth (Fig. 5A; Table 2). Eleven cases eventually showed rapid involution (leaving small calcified remnants), and 16 cases showed partial involution (Fig. 5B). The conservative group was followed up for a median of 3.4 years (range: 1.3 ~ 5.8 years), and none of them received medication for significant lesion enlargement or clinical symptoms. Seven children in this group had rapid involution, and the remaining eight had partial involution. The treatment group was followed up for a median of 2.9 years (range: 1.2 ~ 5.8 years). All 12 children received oral propranolol, and 6 of them were treated with a combination of cardiac diuretics early in the course of treatment due to combined pulmonary hypertension. One of them underwent interventional treatment (4 ml iodine oil + 6 mg bleomycin) at the age of 3 years, which did not completely regress at the age of 4 years. In addition to the interventional children, the lesions eventually involuted completely in 4 children at the age of 3–5 years and partially in the remaining 7 children. There was no statistically significant difference in prognosis between the two groups.

Open in a separate window

Fig. 3

Ultrasound (US) images of the observation (A, B) and treatment groups (C, D) at birth (A, C) and 2 years old (B, D)

Open in a separate window

Fig. 4

Histopathologic images of GFHHs

Open in a separate window

Fig. 5

The lesions in four patients showed postnatal growth before involution (A). Time evolution curve of tumor volume in the treatment group and observation group (B)

Table 2

Clinical and imaging features of congenital hemangiomas that exhibited postnatal growth before involuting (n=4)

P	Sex	Birth gestation	Delivery pattern	Initial volume	Maximum volume	Complication	Treatment	Involution (age)	Residual lesion
P1	Male	37	Normal labor	50	78	None	Conservative	Complete (12)	NA
P2	Male	38	Cesarean section	31	48	ACS, Anemia, Pulmonary hypertension, CC, AVF	propranolol cardiac diuretics	Complete (40)	Calcification
P3	Male	38	Normal labor	50	168	Cutaneous hemangioma, CC	propranolol	Partial (12)	9 mL
P4	Male	40	Normal labor	10	22	CC	propranolol	Complete (18)	Calcification

Open in a separate window

P patient, Birth gestation birth gestation in weeks, Initial volume tumor volume in mL, Maximum volume maximum volume in mL, Involution (age) age at involution in months, ACS abdominal compartment syndrome, CC consumptive coagulopathy, AVF hepatic arteriovenous fistula

Histopathologic study showed variable-sized luminal structures lined with flat cells, and erythrocytes were seen in the lumen (original magnification, 40×)(A and C). The tumor cells were diffusely negative for GLUT-1 immunohistochemically. (B and D, original magnification, 200×)

Optimal treatment strategy

Observational management is advised for 15 cases, which are usually asymptomatic. Serial U/S follow-up observation is recommended to document the changes in tumor size and expected regression of the lesion. This is consistent with previous recommendations for the treatment of hepatic hemangiomas [15].

Treatment of hepatic hemangioma depends not only on the size of the tumor but also on clinical symptoms and laboratory findings. Children with clinical symptoms (such as dyspnea and abdominal distention leading to recurrent vomiting), postnatal cardiac ultrasound suggesting cardiac insufficiency, hepatic arteriovenous leakage, and significant abnormalities in coagulation received pharmacological treatment after definitive diagnosis. Previous studies have shown that GFHH may present with serious complications such as high-volume congestive heart failure, severe anemia, or KM syndrome. Without clinical intervention, the perinatal morbidity and mortality rate can be as high as 90% [16].

Clinical behavior can be innocuous or life-threatening. Thrombocytopenia, coagulation disorders, and increased liver enzymes at diagnosis seem to be the main predictors of mortality [17]. Congenital arteriovenous intrahepatic fistulas, which are hepatic hemangiomas and arteriovenous malformations (AVMs), are rare [18]. Both vascular anomalies were confused with each other. However, vascular anomalies of neoplastic origin show an increase in the cell proliferation of the soft tissue mass, whereas AVMs do not. Similarly, both hepatic hemangiomas and hepatic AVMs have nourishing dilated arteries and draining dilated veins [19]. As a result, high output heart failure, hepatomegaly, consumption coagulopathy, and hemorrhage are common complications in both neonatal periods [20]. Prophylactic treatment with steroids and propranolol was given to four children with giant hepatic hemangiomas complicated by arteriovenous fistulas; eventually, two patients showed complete regression, and two patients showed partial regression.

In this study, six patients with hypertension used cardiac diuretics in the early stage of treatment. For children with high volume congestive heart failure, severe anemia, thrombocytopenia, or KM syndrome in the early stage, methylprednisolone was used in combination and was gradually discontinued when the platelets were monitored and normalized and the condition stabilized.

Our previous study showed that corticosteroid therapy combined with active treatment for congestive heart failure and pneumonia was effective in most cases of neonatal hepatic hemangioma [21]. Rapid involution of approximately 30% of lesions has been reported after high-dose corticosteroid treatment, while the other 40% had a less pronounced effect, but congestive heart failure and coagulation dysfunction were relieved [22].

Although there is a lack of consensus regarding whether transcatheter arterial embolization (TAE) is effective in the treatment of hemangiomas and regarding the severity of complications, [23–25] a consensus exists among experts that interventional therapy is a viable option when drug therapy is not working [26]. Torkian et al. conducted a systematic review and meta-analysis suggesting that TAE with bleomycin, pingyangmycin, or ethanol, in combination with lipiodol, was safe and effective [23]. Yuan Bing et al. retrospectively reviewed 241 patients with hepatic hemangioma who had undergone TAE with lipiodol-bleomycin emulsion (LBE) and found that TAE with LBE was feasible and effective for giant hepatic hemangiomas in all patients. TAE was performed successfully in the participant without serious complications [27].

The postnatal growth of focal, congenital, GLUT1-negative hemangiomas has not been previously described in the liver. However, a recent study documenting the spontaneous evolution of untreated focal congenital hepatic lesions consistent with hepatic hemangioma has shown that some focal congenital lesions can show significant postnatal growth without evidence of intralesional hemorrhage [2]. “Although they disagree about the diagnostic criteria for congenital hepatic hemangioma (prenatal diagnosis and/or age at diagnosis ≤ 7 days and/or GLUT1-negative tumors), the age ranges for diagnosis of congenital hepatic hemangioma and infantile hepatic hemangioma overlap.” However, all the children included in our study were prenatally diagnosed, and four of them also had postnatal growth, one case in the observation group and three cases in the drug treatment group. Additionally, a case series in China also reported two children who were treated with propranolol after birth, and dynamic observation showed that the lesions did not subside but appeared to grow [11].

The therapeutic effect of oral propranolol administration in IHs was incidentally discovered in 2008 [28]. Propranolol is a nonselective beta-blocker that acts on both β-1 and β-2 adrenergic receptors. Oral propranolol actually represents the first-line approach for the pharmacological treatment of IHs, which suppress lesion proliferation [29–32]. Multiple cutaneous and infantile hepatic hemangiomas respond significantly to propranolol monotherapy in the neonatal period [33]. Positive staining for glucose transporter isoform 1 (GLUT 1) is the histologic marker that differentiates IHH from other types of congenital hepatic hemangioma. IHH does not spontaneously subside and responds to drugs. RICH usually subsides at the age of 12–14 months. Clinical management can be achieved through symptomatic treatment, observation, waiting, and close monitoring of hemangioma size. Drug treatment (corticosteroids, propranolol) has not yet shown efficacy [34, 35]. Considering the clinical outcome of GFHH, we believe that propranolol may have no effect on the volume change.

For uncomplicated fetal giant hepatic hemangioma, perinatal follow-up observation is the mainstay, and fetal hepatic hemangioma with comorbidities requires active medical treatment. The main limitations of this study include its retrospective and single-center nature, as well as the intra- and interobserver variability in imaging test measurements. The small sample size and selection bias are further limitations. In addition, the absence of similar studies makes it difficult to compare the results obtained. Despite its small sample size, it represents the largest reported series of patients with GFHH. However, multicenter studies are required to confirm these results.