2026 Poster Presentations
P102: CHOOSING THE CORRIDOR: FAR LATERAL VERSUS RETROSIGMOID APPROACHES TO THE FORAMEN MAGNUM
Eric M Cohen, BS1; Alec Dallas, BS1; Yosef Nafii, BS1; Chloe Chose, BS1; Daniel Colome, BS1; Brandon Dyer, BS1; Alexander Kueffer, MD, PhD, DMD2; Mohammad Hassan A Noureldine, MD, MSc3; Kunal Vakharia, MD3; Harry Van Loveren, MD3; Siviero Agazzi, MD, MBA3; Davide Croci, MD2; 1USF Health Morsani College of Medicine; 2Department of Neurosurgery, Lakeland Regional Health, Lakeland, Florida, United States; 3Department of Neurosurgery and Brain Repair, University of South Florida Morsani College of Medicine, Tampa, Florida, USA
Background: Selecting a skull-base approach requires balancing exposure against morbidity. We assessed whether the operative corridor for foramen magnum lesions, specifically comparing far-lateral with retrosigmoid approaches, independently influences perioperative outcomes after accounting for lesion anatomy and pathology.
Methods: This was an IRB-approved, retrospective multicenter study of consecutive skull-base cases managed with the far-lateral or retrosigmoid approaches. Inclusion required an operative note description consistent with the recorded approach and independent screening by a skull-base fellowship-trained neurosurgeon to confirm the concordance of lesion anatomy and pathology. All operating surgeons were skull-base specialists. Outcomes included operative time, ICU length of stay, hospital length of stay, overall complications, pseudomeningocele formation, surgical site infection, and disposition. Descriptive statistics were performed to summarize medians and IQRs. Hypothesis testing was conducted using Shapiro-Wilks to assess for normality and followed by Mann-Whitney U or t-tests for two group comparisons. Categorical variables were analyzed with chi-square or Fisher exact tests. Correlations were assessed univariately using Spearman’s rank correlation and in multivariate models with logistic regression adjusting for approach and lesion characteristics. Variables with less than 5 instances were excluded.
Results: Forty-nine patients were included (mean age 57.7 years; 63% female). Approaches used were far lateral in 36/49 (73.5%), retrosigmoid in 11/49 (22.4%), and lateral suboccipital in 2/49 (4.1%). Comparative analyses were done between far lateral and retrosigmoid only due to small sample of lateral suboccipital approaches. Median operative time was 364 minutes (IQR 257–451), ICU length of stay was 3 days (IQR 2–10), and hospital length of stay was 8 days (4–21). Overall complications occurred in 45%, with pseudomeningocele in 27%. Condylar drilling was done more with the far lateral approach (14/36, 38.9%) than retrosigmoid (0/11, 0%; p=0.029). Incision type differed by the upper-most and lower-most cranio-caudal lesion extent (p=0.0033 and p=0.0083), anterior–posterior diameter (p=0.0069), ICU length of stay (p=0.045), and any complication (p=0.0448). Lesion size varied by histology (p=0.0034) and was associated with longer follow-up (ρ=0.418, p=0.012). Larger overall lesion size was associated with surgical-site infection on univariate analysis (5.00 cm vs 2.65 cm, p=0.018); this association did not persist after adjustment (all p>0.05). After adjustment, neither surgical approach nor incision type (C-shaped vs Reverse Hockey stick) independently predicted overall complications, pseudomeningocele formation, surgical-site infection, ICU length of stay, or hospital length of stay (all p>0.05).
Conclusion: In this multicenter series restricted to far lateral and retrosigmoid approaches and excluding groups with fewer than five patients, between-group differences reflected case selection based on lesion anatomy and pathology rather than the chosen corridor. Treating lesion at level of the foramen magnum carries a high risk of overall complications especially of pseudomeningocele. Adjusting for lesion characteristics, approach, and incision were not independent drivers of complications, pseudomeningocele formation, surgical-site infection, or length of stay. While recognizing the limits of small sample size and group imbalance, these findings support an approach tailored to lesion anatomy and pathology without an inherent perioperative penalty in outcome.