Decrease in cranial nerve complications after radiosurgery for acoustic neuromas: A prospective study of dose and volume

Purpose: To determine whether tumor control can be maintained, and cranial nerve complications decreased by reducing the radiosurgical dose to acousti c neuromas. Methods and Materials: Forty-two consecutive patients with acoustic neuroma s were treated prospectively using an initial standard-dose protocol in whi ch the tumor-margin dose (50% isodose) was 20, 18, and 16 Gy for tumor diam eters less than or equal to 2 cm, 2.1-3 cm, and 3.1-4 cm, respectively. Aft er analysis of tumor control and complications, the next 40 patients were t reated using a reduced-dose protocol in which the tumor-margin dose was 16, 14, and 12 Gy for tumor volumes less than or equal to 4.2 cm(3), 4.2-4.1 c m(3), and greater than or equal to 14.1 cm(3), respectively. Results: Median follow-up was 2.3 years (range 0.1-6) for 80 of 82 patients . The actuarial incidence (Kaplan-Meier) of facial neuropathy at 2 years wa s 38% (95% confidence interval [CI], 23-53%) for the standard-dose protocol and 8% (95% CI, 0-17%) for the reduced-dose protocol (p = 0.006). Univaria te analysis revealed an association between risk of facial neuropathy and u se of CT planning, higher radiosurgical dose, and neurofibromatosis, type 2 . Multivariate analysis revealed that the only factor associated with incre ased risk of post-treatment facial neuropathy was a tumor margin dose great er than or equal to 18 Gy. The incidence of trigeminal neuropathy at 2 year s was 29% (95% CI, 15-43%) for the standard-dose protocol and 15% (95% CI, 3-27%) for the reduced-dose protocol (p = 0.17). Univariate analysis reveal ed an association between maximal tumor diameter and increased risk of trig eminal neuropathy; multivariate analysis revealed no additional statistical ly significant associations between tumor and dosimetric and patient charac teristics and risk of trigeminal neuropathy. Two tumors in the standard-dos e protocol required salvage surgery for progression. To date, no tumor in t he reduced-dose protocol has shown progression. Conclusion: Our analysis suggests that a tumor margin dose of greater than or equal to 18 Gy is the most significant risk factor for facial nerve comp lications after acoustic neuroma radiosurgery. Patients receiving a minimal tumor dose of less than or equal to 16 Gy are at significantly lower risk for permanent facial neuropathy after radiosurgery. Longer follow-up is req uired before definitive conclusions can be made about the ultimate rate of tumor control using reduced radiosurgical doses. (C) 1999 Elsevier Science Inc.