Preoperative frailty and tumor volume as determinants of primary management strategy and outcomes in petroclival meningioma patients
Objective We aimed to study how patient frailty, pre-operative tumor volume, and symptom burden collectively impact primary management choice, postoperative complication rates, and outcomes in the treatment of petroclival meningiomas. Methods We retrospectively analyzed a cohort of 73 patients diagnosed with petroclival meningiomas at Emory Healthcare from 2000 to 2022. Frailty was quantified with the 5-item modified frailty index (mFI-5). Tumor volume was measured from pre-operative MRI with 3D Slicer. Tumor-related symptoms were clustered into five domains (visual, vestibulocochlear, sensory, motor, headache) and summed into a composite Symptom Burden Score (SBS). Primary management was categorized as surgical resection versus conservative therapy (serial imaging or radiotherapy). Multivariable logistic regression evaluated predictors of treatment selection; a second model examined 30-day postoperative complications in surgical patients. Linear mixed-effects models (LMM) tracked SBS, and domain-specific generalized linear mixed-effects models (GLMM) assessed symptom resolution at 3, 6, and 12 months, controlling for both frailty and treatment approach. Sensitivity analyses dichotomized frailty (mFI-5 = 0 vs. ≥1). Significance threshold: two-sided p < 0.05. Results Surgically treated tumors were larger than conservatively managed lesions (median 14.3 cm3 vs. 8.9 cm3; p = 0.018), and management choice varied by frailty strata (χ2 = 12.86, p = 0.015). In multivariable analysis, neither tumor volume (OR 1.04, 95% CI 0.97–1.11) nor mFI-5 score (OR 0.56, 95% CI 0.20–1.82) independently predicted operative selection. Thirty day complication rates increased with frailty but mFI-5 was not an independent risk factor (OR 0.51, 95% CI 0.22–1.18). Mean patient level SBS declined in both groups at 12 months (conservative −28%; surgical −8%), with a marginally significant interaction between treatment approach and frailty (p = 0.072). Symptom domain analysis showed significant improvement in visual and vestibulocochlear deficits irrespective of management, whereas conservative therapy was associated with poorer motor deficit recovery (p = 0.025). Conclusions Preoperative tumor volume and frailty jointly influence treatment approach, yet neither metric alone dictates management decision when considered together. Perioperative risk-reduction protocols may mitigate frailty-associated morbidity. While most neurologic symptoms improve within a year, frailer patients have attenuated symptomatic improvement, in addition to domain-specific differences.