International Tuberculum Sellae Meningioma Study: Preoperative Grading Scale to Predict Outcomes and Propensity-Matched Outcomes by Endonasal Versus Transcranial Approach

Abstract

Background and objectives: Tuberculum sellae meningiomas are resected via an expanded endonasal (EEA) or transcranial approach (TCA). Which approach provides superior outcomes is debated. The Magill-McDermott (M-M) grading scale evaluating tumor size, optic canal invasion, and arterial involvement remains to be validated for outcome prediction. The objective of this study was to validate the M-M scale for predicting visual outcome, extent of resection (EOR), and recurrence, and to use propensity matching by M-M scale to determine whether visual outcome, EOR, or recurrence differ between EEA and TCA.

Methods: Forty-site retrospective study of 947 patients undergoing tuberculum sellae meningiomas resection. Standard statistical methods and propensity matching were used.

Results: The M-M scale predicted visual worsening (odds ratio [OR]/point: 1.22, 95% CI: 1.02-1.46, P = .0271) and gross total resection (GTR) (OR/point: 0.71, 95% CI: 0.62-0.81, P < .0001), but not recurrence ( P = .4695). The scale was simplified and validated in an independent cohort for predicting visual worsening (OR/point: 2.34, 95% CI: 1.33-4.14, P = .0032) and GTR (OR/point: 0.73, 95% CI: 0.57-0.93, P = .0127), but not recurrence ( P = .2572). In propensity-matched samples, there was no difference in visual worsening ( P = .8757) or recurrence ( P = .5678) between TCA and EEA, but GTR was more likely with TCA (OR: 1.49, 95% CI: 1.02-2.18, P = .0409). Matched patients with preoperative visual deficits who had an EEA were more likely to have visual improvement than those undergoing TCA (72.9% vs 58.4%, P = .0010) with equal rates of visual worsening (EEA 8.0% vs TCA 8.6%, P = .8018).

Conclusion: The refined M-M scale predicts visual worsening and EOR preoperatively. Preoperative visual deficits are more likely to improve after EEA; however, individual tumor features must be considered during nuanced approach selection by experienced neurosurgeons.

FIGURE 1.

Selection bias is present in the management of tuberculum sellae meningiomas. Each blue dot represents a single institution in the study, with the location on the graph representing the number of cases done with a TCA or an EEA at the same site. Only 4 institutions had completed 10 TCA and 10 EEA approaches at the same site, and no institutions had done more than 20 through both approaches. EEA, expanded endonasal approach; TCA, transcranial approach.

FIGURE 2.

Study schema. The study was divided into 2 groups, the first (left lower) to validate and refine the grading scale from Magill et al, and the second to validate the refined/simplified grading scale (right lower). The patients from the Magill et al cohort were in the external validation group for the refined Magill–McDermott grading scale. EEA, expanded endonasal approach; pts, patients; TCA, transcranial approach.

FIGURE 3.

The refined Magill–McDermott scale for predicting TSM outcomes. A, Tumor score is 1 if the tumor diameter at the tuberculum is less than 17 mm, and B, score is 2 if greater than 17 mm. C, The canal score is 1 if there is no optic canal invasion, and D, the score is 2 if there is invasion into either canal of 3 mm or more. E, The artery score is 1 if the tumor abuts, but does not envelop the adjacent arteries more than 180°. F, The artery score is 2 if the ICA, ACA, and/or MCA are encased more than 180°. Published with permission. Copyright Kenneth X. Probst. ACA, anterior cerebral artery; ICA, internal carotid artery; MCA, middle cerebral artery.

FIGURE 4.

Nomogram application of refined Magill–McDermott grading scale and significant preoperative variables to predict A, gross total resection and B, risk of visual worsening. GTR, gross total resection.

FIGURE 5.

Propensity-matched visual outcomes by preoperative deficit. A, For patients without a preop visual deficit, there was no difference in rates of visual worsening between EEA and TCA (5.0% vs 4.7%). B, More patients had visual improvement after EEA (72.9%) compared with TCA (58.4%) with no difference in visual worsening (EEA 8.0% vs TCA 8.6%). P-values = Pearson's χ². EEA, expanded endonasal approach; TCA, transcranial approach.