To the Editor:
Agha and colleagues1 presented the case of a 20-day-old, full-term girl who had respiratory distress, tachycardia, mild hepatomegaly, and pulmonary hypertension. There were no skin or mucosal vascular lesions, and the relevant family history was negative. A computed tomographic scan showed a large, well-circumscribed vascular structure with enlargements of the hepatic vein and internal mammary and celiac arteries.
Results of a 4-month follow-up evaluation revealed substantial clinical improvement, resolution of the pulmonary hypertension, and reduced vascularity. The authors assumed that the diagnosis was hepatic arteriovenous malformation (AVM) and treated the patient with furosemide, captopril, and digoxin.
Having reviewed the clinical features and images reported by the authors, I would like to comment on the diagnosis of this lesion.
Hepatic AVMs are exceedingly rare. The hallmark of AVM is the presence of abnormal communication between arteries and veins without a mass component. The authors reported a “large, well-circumscribed” vascular lesion within the liver with enlarged draining hepatic veins and feeding (internal mammary and celiac) arteries.
I strongly believe that the proper diagnosis in this child is a rapidly involuting congenital hemangioma (RICH), rather than an AVM. Hepatic hemangiomas are the most prevalent benign liver tumors in early infancy.2 There are 3 main types of hepatic hemangiomas: focal, multifocal, and diffuse.3 Whereas the multifocal and diffuse types are true infantile hemangiomas, focal lesions are RICH. The natural history of hepatic RICH is different from that of infantile hemangioma, because RICH does not typically grow after birth, and it regresses within months.4
The RICH is a solitary, often large hypervascular lesion that typically parasitizes flow from adjacent vascular pedicles, including the celiac, superior mesenteric, phrenic, and internal mammary arteries. Pulmonary hypertension is likely to occur, because of increased hepatic venous return.
On analysis, RICH stains negatively for glucose transporter isoform 1 (GLUT-1, a glucose transporter enzyme), and it has no gender preference, whereas infantile hemangiomas are GLUT-1–positive and preferentially affect females.5
The natural history of AVMs is eventual progression, not spontaneous resolution, particularly for such a large lesion as that reported by the authors.
Regardless of the diagnosis, the treatment of the infant in this report appears to have been satisfactory, and the authors deserve commendation for it.
With response by
This letter was sent to Dr. Hala Agha, who responds as follows:
I thank Dr. Alomari for his interest in our article.1
It is known that congenital hepatic arteriovenous malformation (AVM) is a rare and deadly anomaly; consequently, published data on outcomes are scarce. In pediatric patients who have vascular tumors and malformations—a challenging subgroup—imaging techniques are now sophisticated enough to enable accurate diagnosis and suggest approaches to treatment.2
In support of our diagnosis,1 I first want to discuss a paper about an unusual situation similar to that of our patient. In that publication,3 the authors presented the cases of 4 neonates who had AVMs (cerebral in 1, pulmonary in 1, and hepatic in 2) and reported that the two who had hepatic AVMs improved after medical therapy. Both infants had clinical evidence of high-output cardiac failure.3
Vascular anomalies can be broadly placed into 2 categories: neoplastic origin or congenital malformation.4,5 Neoplastic anomalies, such as infantile hepatic hemangioma, exhibit increased cell proliferation that constitutes a soft-tissue mass. In contrast, congenital malformations consist solely of dysplastic blood vessels, and the absence of an associated soft-tissue mass is an important diagnostic feature.5,6 Another distinguishing feature of AVMs is a nidus, a region of abnormal arteriovenous communication at the AVM core that bypasses the normal tissue capillary bed.7 Vascular malformations are further differentiated from tumors by their lack of mitoses and glucose transporter isoform-1 negativity.8 All the above considerations contributed to the diagnosis of AVM in our patient.