Multidimensional outcomes of initial treatment strategies for pediatric head and neck lymphatic malformations

To compare real-world initial treatment strategies for pediatric head and neck lymphatic malformations using a multidimensional outcome framework that distinguishes complete radiological response, objective lesion response, progression control, and treatment burden, and to examine how lesion complexity influences treatment selection and outcomes.

We conducted a retrospective cohort study of pediatric patients with head and neck lymphatic malformations treated at a tertiary referral center between January 2016 and December 2025. Patients were categorized into surgery alone, interventional therapy alone, medical therapy alone, or combined therapy according to the initial treatment strategy. Radiological outcomes were classified as complete response, partial response, stable disease, or progression/recurrence. Primary outcomes included complete response, objective response rate, progression/recurrence, and progression-free survival. Lesion complexity was classified using an exploratory, literature-informed framework based on De Serres stage, cystic subtype, number of involved anatomical regions, and critical structure involvement. Multivariable logistic regression, generalized propensity score-based overlap weighting, propensity score matching, event-density analysis, and follow-up sensitivity analyses were used to address confounding by indication and unequal follow-up.

Among 210 patients, 40 received surgery alone, 110 interventional therapy alone, 23 medical therapy alone, and 37 combined therapy. Surgery alone achieved the highest complete response rate (75.0%). Interventional therapy alone showed the highest objective response rate (87.3%), highest disease control rate (95.5%), lowest progression/recurrence rate (4.5%), and lowest additional-treatment rate (4.5%). Medical therapy alone had the highest progression/recurrence rate (39.1%), a pattern that may at least partly reflect the selection of systemic therapy for more complex disease, although a contribution from limited treatment effect in this heterogeneous group cannot be excluded. High-complexity lesions were less likely to achieve complete response and had numerically higher progression/recurrence. The main outcome patterns remained directionally consistent after multivariable adjustment, overlap weighting, propensity score-matched sensitivity analysis, and follow-up sensitivity analyses.

Initial treatment strategies for pediatric head and neck lymphatic malformations demonstrated distinct outcome profiles rather than a simple hierarchy of effectiveness. Surgery alone was most strongly associated with complete radiological response, whereas interventional therapy alone showed a favorable objective-response and progression-control profile. A multidimensional framework integrating eradication, objective response, lesion complexity, treatment burden, and progression control may better support individualized treatment selection.