Sturge-Weber Syndrome in a Multinational Paediatric Cohort: A Systematic Analysis of Different Types

Background Sturge-Weber Syndrome (SWS) is a rare neurocutaneous disease, characterized by cerebral capillary-venous malformation, glaucoma, and facial vascular birthmark. Different types are reflected in the Roach classification. Most previous studies have focussed on classic SWS Type I, but Type III cases, lacking facial birthmark, were mostly described in case reports. We systematically compare cases with and without facial birthmark, with a focus on epilepsy variables, cerebral involvement, use of aspirin, and overall outcome. Methods Using a cross-sectional observational study conducted through a well-established child neurologists’ network, we recruited paediatric patients with clinically diagnosed SWS from Germany, Switzerland, and Austria. The patients’ guardians and attending child neurologists filled in detailed questionnaires. All patients were classified according to the Roach classification by both attending child neurologists and the study team. Results Out of 111 paediatric SWS patients identified, 47 participated (43.2%). 35 cases (74.5%) fulfilled the criteria for classic SWS (Roach Type I); six cases (12.8%) showed no skin involvement (Roach Type III), the remaining six cases had overlap/atypical phacomatoses with capillary-venous malformation. Cases without facial birthmark were older at diagnosis (p = 0.005), and none showed ophthalmologic involvement. Age at first seizure did not differ significantly after adjustment for multiple comparisons. No significant differences were observed in seizure frequency, seizures types, number of used antiseizure medication (ASM), epilepsy surgery, cerebral involvement including atrophy and calcifications, SWS neuroscores or use of supportive therapies. Multivariable logistic regression showed seizure frequency was independent of SWS type and epilepsy surgery, but positively associated with the number of ASM required for seizure control (p = 0.0056). Half of operated patients were seizure-free at inclusion. Conclusions In our multinational cohort, Type I and Type III cases showed comparable epilepsy features, SWS neuroscores, number of required ASM and supportive therapy requirements. Type III patients were older at diagnosis and showed no ophthalmologic involvement, indicating a milder phenotype. Irrespective of SWS type, patients with uncontrolled epilepsy were 3.8-times more likely to require additional ASM. Despite uncontrolled epilepsy, only few patients underwent surgical evaluation or intervention. Larger cohorts are needed to evaluate surgical outcomes across SWS subtypes.