Introduction

I read with great interest the report by Trujeque et al. describing multimodal imaging and transcatheter embolisation of a large interventricular septal mass causing severe outflow tract obstruction in a 14-year-old boy, presumed to represent a cardiac haemangioma. ^{Reference Trujeque, Hernandez and Tavera1} The case elegantly shows how echocardiography, tissue-characterising MRI and coronary CT angiography can together substitute for a biopsy that would have carried unacceptable risk. Two points, however, deserve emphasis before the diagnosis and the therapeutic success are taken as fully settled.

First, the label “cardiac haemangioma” deserves more caution than imaging alone can supply. Infantile haemangioma is a specific, GLUT1-positive proliferative endothelial tumour, whereas congenital and deep visceral vascular tumours and most vascular malformations are GLUT1-negative and behave differently, neither reliably proliferating nor predictably involuting. ^{Reference North, Waner, Mizeracki and Mihm2} A prominent feeding vessel on colour Doppler and a raised extracellular volume, however striking, confirm high vascularity but do not distinguish between these biologically distinct entities. Given that histopathological confirmation was, understandably, deferred because of bleeding risk, the more defensible working label may be “vascular tumour of the interventricular septum, favouring haemangioma,“ since natural history and any rationale for adjunct medical therapy differ between the two categories.

Second, coil embolisation interrupts the dominant feeding artery but does not resect the lesion itself and vascular tumours are recognised for recruiting collateral supply once their primary feeder is occluded. Recanalisation with symptomatic recurrence has been reported as late as five years after apparently successful embolisation of a bladder haemangioma ^{Reference Safar, Al-Qahtani and Al-Qahtani3} and coil recanalisation rates of up to 11.7% have been documented in pulmonary arteriovenous malformations followed over two decades. ^{Reference Andersen, Duvnjak, Gerke and Kjeldsen4} A 12-month follow-up with a residual gradient of 37 mmHg is encouraging, but may be too short to exclude late recruitment of collateral septal perforators in a growing adolescent heart. Serial imaging echocardiography with low Nyquist colour Doppler, supplemented by CT angiography at intervals extending well beyond the first year, would strengthen confidence that true occlusion, rather than flow redistribution, accounts for the improving gradient.

A related, testable question is whether the propranolol this patient had already been receiving prescribed for the incorrect diagnosis of hypertrophic cardiomyopathy contributed to the improvement independently of embolisation, whether through its negative chronotropic and inotropic effect on outflow dynamics or, more speculatively, an antiproliferative action of the kind described in GLUT1-positive infantile haemangiomas. ^{Reference Drolet, Frommelt and Chamlin5} Reporting serial mass dimensions, rather than gradient alone, in future reports would help separate a true anatomical response from a purely haemodynamic one.

None of this diminishes what the authors achieved: a technically elegant, minimally invasive solution to an anatomically forbidding tumour that avoided high-risk surgery in a symptomatic adolescent. It simply argues for watchful, imaging-based vigilance beyond the first postoperative year and for restraint in histological labelling until tissue or considerably longer follow-up speaks for itself.