Abstract
Purpose: The recurrence rate and related risk factors of infantile hemangiomas after treatment discontinuation remain concerns. We aim to evaluate the risk of recurrence after termination of oral propranolol for IHs and its associated risk factors. Methods: The Embase, PubMed, Web of Science, and Cochrane Central databases and clinicaltrials.gov were searched comprehensively for relevant studies from the inception of this study to November, 2024. Two independent reviewers conducted the data extraction and quality assessment. This review protocol was registered in the PROSPERO database (CRD42024589110). Results: A total of 1662 patients in 10 studies met the eligibility criteria, which was predominantly retrospective in design. All participants were infants diagnosed with infantile hemangiomas who received oral propranolol therapy; the majority of patients received propranolol treatment for at least six months. The results revealed that the pooled recurrence rate was 20% (95% CI: 15-24%), and 11% of patients required retreatment with propranolol (95% CI: 9-14%). Female sex (OR = 1.76, 95% CI: 1.20-2.59) and IHs located on the head and neck (OR = 2.40, 95% CI 1.59-3.63) increased the risk of recurrence. In contrast, IH type, lesion distribution, duration of therapy, and treatment initiation age showed no significant associations. Additionally, one trial included in this review reported that continued medication for one month after the lesion reaches its maximum degree of regression might increase the risk of recurrence as compared to three months of maintenance (OR = 1.86, 95% CI 0.98-3.5); however, the evidence is limited and preliminary. Conclusions: Female sex and IHs located on the neck or head contribute to the recurrence of IHs after termination of treatment. In addition, the type of IH and withdrawal criteria may influence recurrence risk, although evidence remains limited. Thus, optimizing treatment protocols, including individualized therapy duration and discontinuation strategies, may help reduce recurrence rates.