Outcome Features Analysis in Intramedullary Tumors of the Cervicomedullary Junction: A Surgical Series

Abstract

Abstract Object The aim of this study is to investigate the impact of surgery for different cervicomedullary lesions on symptomatic pattern expression and postoperative outcome. We focused on specific outcome features of the early and late postoperative assessments. The former relies on surgery-related transient and permanent morbidity and feasibility of radicality in eloquent areas, whereas the latter on long-term course in lower grade tumors and benign tumorlike lesions (cavernomas, etc.). Material and Methods We retrospectively analyzed 28 cases of intramedullary tumors of the cervicomedullary junction surgically treated at our institution between 1990 and 2018. All cases were stratified for gender, histology, macroscopic appearance, location, surgical approach, and presence of a plane of dissection (POD). Mean follow-up was 5.6 years and it was performed via periodic magnetic resonance imaging (MRI) and functional assessments (Karnofsky Performance Scale [KPS] and modified McCormick [MC] grading system). Results In all, 78.5% were low-grade tumors (or benign lesions) and 21.5% were high-grade tumors. Sixty-one percent underwent median suboccipital approach, 18% a posterolateral approach, and 21% a posterior cervical approach. Gross total resection was achieved in 54% of cases, near-total resection (>90%) in 14%, and subtotal resection (50–90%) in 32% of cases. Early postoperative morbidity was 25%, but late functional evaluation in 79% of the patients showed KPS > 70 and MC grade I; only 21% of cases showed KPS < 70 and MC grades II and III at late follow-up. Mean overall survival was 7 years in low-grade tumors or cavernomas and 11.7 months in high-grade tumors. Progression-free survival at the end of follow-up was 71% (evaluated mainly on low-grade tumors). Conclusions The surgical goal should be to achieve maximal cytoreduction and minimal postoperative neurologic damage. Functional outcome is influenced by the presence of a POD, radicality, histology, preoperative status, and employment of advanced neuroimaging planning and intraoperative monitoring.