OBJECTIVE Sinogenic intracranial abscess is a severe complication of sinusitis. Timely identification of patients requiring neurosurgical intervention can be challenging. This study aimed to evaluate preoperative clinical and radiographic variables to identify features that correlate with neurosurgical intervention and outcomes in pediatric patients with sinogenic intracranial infection. METHODS This was a single-center retrospective study of pediatric patients who were surgically treated for sinogenic intracranial infection. Patients were grouped into an otolaryngological treatment (ENT)-only cohort and neurosurgical treatment (NSGY) cohort, with the latter including those who underwent neurosurgery alone or neurosurgery plus otolaryngology treatments. Variables included abscess volume, peri-infection T2-weighted hyperintensity volume and thickness, inflammatory marker levels (white blood cells, neutrophils, C-reactive protein [CRP], erythrocyte sedimentation rate [ESR]), Lund-Mackay sinus disease severity score, clinical symptoms, and infectious organisms. Outcomes included radiographic residual disease, length of stay, reoperation, readmission, and neurological deficits. Univariate and multivariate logistic regression and receiver operating characteristic (ROC) analyses were performed to identify and evaluate significant factors. RESULTS Ninety patients met the inclusion criteria (mean ± SD age 11.43 ± 3.33 years; 61.11% of patients were male), with 56 in the ENT group and 34 in the NSGY group. In univariate analysis, larger abscess volume, larger perilesional T2-weighted hyperintensity, older age, elevated ESR and CRP level, and lower Lund-Mackay scores were associated with NSGY. In multivariate analysis, abscess volume (OR 1.11, 95% CI 1.03–1.19, p = 0.0055) and CRP level (OR 1.10, 95% CI 1.00–1.20, p = 0.048) remained the only independent correlates. ROC analyses showed that abscess volume had excellent discriminatory power (area under the curve 0.89) with an optimal threshold at 7.57 cm 3 . Streptococcus anginosus was the most common microorganism and was more prevalent in the NSGY group (p = 0.017). There were no significant differences in radiographic outcomes, length of stay, reoperation or readmission rates, or neurological deficits between the NSGY and ENT cohorts after adjusting for baseline disease severity. CONCLUSIONS Preoperative abscess volume was best correlated with neurosurgical intervention. While neurosurgical patients had more severe baseline disease, it was not associated with worse outcomes. Given the retrospective and single-center design, these findings likely reflect appropriate patient selection and institutional practice rather than causality. Future prospective studies are needed to further investigate the relationship among clinical variables, interventions, and outcomes.