Discussion
This study investigated the current use of Hemangiol® in 94 children with proliferating IH. It provided new tolerance data on this treatment and the value of systematic cardiac screening before treatment. Moreover, the safety of a rapid initiation protocol was evaluated.
Overall, the population treated with Hemangiol® in this study is similar to previous reports8 and our cohort is representative of the usually described population of IH, with a predominant female representation21. However, more subjects with preterm birth and low birth weight were present in our cohort, considering they were previously described as risk factors for IH2,22. The cases with PHACES or LUMBAR syndromes included in the study underwent screening evaluation of clinical and radiological associated anomalies, as recommended in children with large segmental haemangiomas of the upper and lower body10,12. The main indications for treatment of our cohort were IH with a functional threat, ulcerated haemangiomas and IH with an aesthetic prognosis, as recommended in the current guidelines13. Children of our study were treated for an average of 7 months, which is moderately longer than in the initial trial (e.g. 6 months)18. However, recent studies have suggested treatment for 6 to 12 months due to a greater IH rebound growth in shorter treatment duration23,24. Indeed, treatment duration is mostly dependent on clinical complete lesion regression25. Off-label use of Hemangiol® was rare in this study. Two neonates <5 weeks of age were treated with a good tolerance during initiation and no reported adverse event during follow-up, which confirms a previous study on the safety of oral propranolol for IH in the neonatal period 26. Late Hemangiol® initiations >5 months of age concerned 29 subjects in our study. Usually, late treatment initiations are related to a delayed referral to the specialist, especially for superficial IH27.
The safety analysis found AEs in 26.6% patients, which is consistent with previous studies on oral propranolol, which reported AEs rates from 19.6%28 to 38.2%29. Sleep disturbances, wheezing and digestive side effects were the most frequently reported AEs in our cohort, as described in the previous studies, however peripheral coldness was not reported in our cohort28,29. SAEs were found in 8.5% patients, exclusively severe hypoglycaemia and uncontrolled bronchial hyperreactivity, which is higher than previous reported SAEs rates, ranging from 2.6%30 to 4.8%29. Cardiac SAEs, such as atrioventricular block, bradycardia and symptomatic hypotension, were not found in our cohort, as opposed to previous studies29,30 and despite a systematic paediatric cardiology pretherapeutic and follow-up assessment. Nevertheless, the causality of propranolol for cardiovascular SAEs in the literature remains unclear and suggests pre-existing conditions or incidental discovery29,31. Of note, SAEs occurred in two prematurely born children, who may be more prone to hypoglycaemia and bronchial hyperreactivity as previously reported32. Therefore, low dose of Hemangiol® (2mg/kg/day) could be of interest in such high-risk children23,28, however the dosage of 3mg/kg/day has been more investigated in pharmacokinetics and pharmacodynamics studies33,34 based on the manufacturer’s clinical trial18.
For all SAEs, causality scores concluded to possible to strong relation to Hemangiol®. As severe hypoglycaemia and uncontrolled bronchial hyperreactivity are well known SAEs, variability of causality scores came from intrinsic causation with variable chronological criteria (time to onset after taking the treatment, or evolution after lowering the dosage or stopping treatment) and the presence of confounding factors (i.e. infectious for bronchospasms). As a result, SAEs with high causality scores (4 or 5) in our study were patients with complete description of the event, compatible time to onset, regression of AE after stopping treatment, and no confounding factor. Previous studies did not provide any details on the causality between such SAEs and treatment with Hemangiol®.
From a general perceptive, those findings emphasize the importance of parental therapeutic education about these potential risks and how to manage them. During the treatment initiation period, oral information supported by an educational pack should be provided to parents, in order to identify symptoms potentially related to SAEs35. For example, parents should be fully aware that Hemangiol® needs to be temporarily suspended in case of food intolerance or limited food intake in their child, in order to avoid hypoglycaemia36. Similarly, bronchial hyperreactivity reactions were often related to respiratory tract infections and therefore require appropriate parental supervision29,37.
Overall, the cardiac tolerance was good. No AE, serious or not, was reported during drug initiation at the hospital, whatever the protocol used, as well as during the follow-up. No significant decrease in systolic blood pressure was observed, and the initial decrease in heart rate and diastolic blood pressure was moderate and not clinically relevant, as in the literature38–40. Similarly, previous studies have reported the safety of oral propranolol for the treatment of IH41 and suggested not to extend cardiac monitoring beyond initiation period except for heart rate during following consultations42. In light of these results, an outpatient Hemangiol® initiation protocol could be considered in selected and non-at risk patients, as suggested by Puttgen and al.43.
This study supports the absence of relevance for a systematic pre-therapeutic paediatric cardiology consultation. Indeed, all cardiac findings observed in this cohort were mostly non-significant and did not result in any contraindication for betablockers. Therefore, a simple physical examination made by the IH expert seems sufficient to identify patients at risk. Based on our cohort, children with IH requiring cardiological assessment are those with clinical symptoms (cardiac murmur), abnormal blood pressure or heart rate, syndromic forms (PHACES), high risk of heart failure (hepatic, military haemangiomas), or off-label use (preterm birth, neonate period). These results are consistent with previous studies reporting that systematic echocardiography44,45 or ECG40,46before propranolol initiation are not relevant in terms of contraindication assessment and correlation with SAE occurrence47.
Most patients (81%) underwent a conventional initiation protocol with a 3-week titration phase in day care hospital. This protocol seems safe and adapted to most children with IH, as reported in cost-effectiveness studies48. Nevertheless, a 3-week delay to reach the maintenance dose of 3 mg/kg/day may not be appropriate in a therapeutic emergency situation, such as in IH involving any vital risk, uncontrolled bleeding, ulceration, pain or infectious risk. In our study, nearly 20% of patients with severe proliferative IH underwent a rapid initiation protocol, with a 48-hour dose escalation in conventional hospitalization in paediatric cardiology. This rapid protocol was well tolerated and facilitated the prescription of strong analgesics and local treatment, as in ulcerated IH. In both conventional and rapid protocols, Hemangiol® was well tolerated in terms of blood pressure and heart rate adaptation.