Meningiomas represent approximately 15% of adult intracranial neoplasms. Surgical resection is the preferred treatment whenever total removal can be accomplished with acceptable morbidity Large series on meningioma surgery have found gross total resection possible for 38–80% of patients, depending primarily on tumor location  However, the intimate relationship between some meningiomas and critical neurovascular structures makes complete resection impossible with acceptable risk Still, even when gross total resection has been achieved, tumor recurrence rates at 5 and 10 years have been reported to be 4–14% and 18–25%, respectively

Recently, stereotactic radiosurgery has been performed for an increasing number of patients with small- to moderate-size meningiomas as an alternative to surgical excision In this study, we compared the tumor control rates for adult patients undergoing either surgical resection or radiosurgery as primary management for benign meningiomas <35 mm in average diameter. Between 1990 and 1997, 198 adult meningioma patients treated at our center underwent either surgical resection (n = 136) or radiosurgery (n = 62) as primary management for benign meningiomas <35 mm in average diameter. Tumor recurrence or progression rates were calculated by the Kaplan-Meier method according to an independent radiographic review. The mean follow-up was 64 months.For patients undergoing surgical resection, the operative reports were reviewed and the degree of tumor removal classified according to Simpson grade In particular, a patient was found to have a Grade 1 resection only when the operative note clearly stated that all tumor was removed along with the involved dura and bone. For cases in which the surgeon stated complete tumor removal was achieved but did not remark on the involved dura and bone, Grade 2 resection was assigned. The extent of tumor removal was confirmed by follow-up neuroimaging performed within 6 months for most patients (n = 128, 94%). The tumor resections were Grade 1 in 57 (42%), Grade 2 in 57 (42%), and Grade 3–4 in 22 (16%). The number of patients having a Grade 1 resection by tumor location was skull base in 30 (31%) of 96, falx/tentorium in 4 (25%) of 16, and convexity in 23 (96%) of 24.

Stereotactic radiosurgery
Radiosurgical procedures were performed using the Leksell Gamma Knife (Elekta Instruments, Norcross, GA), with MRI as the imaging modality for dose planning. An average of 8 isocenters (range 1–15) was used to cover a mean prescription isodose volume of 7.4 cm³ (range 0.6–23.5). The mean margin and maximal radiation dose was 17.7 and 34.9 Gy, respectively.

The tumor resections were Simpson Grade 1 in 57 (42%), Grade 2 in 57 (42%), and Grade 3–4 in 22 (16%). The mean margin and maximal radiation dose at radiosurgery was 17.7 Gy and 34.9 Gy, respectively. Tumor recurrence/progression was more frequent in the surgical resection group (12%) than in the radiosurgical group (2%; p = 0.04). No statistically significant difference was detected in the 3- and 7-year actuarial progression-free survival (PFS) rate between patients with Simpson Grade 1 resections (100% and 96%, respectively) and patients who underwent radiosurgery (100% and 95%, respectively; p = 0.94). Radiosurgery provided a higher PFS rate compared with patients with Simpson Grade 2 (3- and 7-year PFS rate, 91% and 82%, respectively; p <0.05) and Grade 3–4 (3- and 7-year PFS rate, 68% and 34%, respectively; p <0.001) resections. Subsequent tumor treatments were more common after surgical resection (15% vs. 3%, p = 0.02). Complications occurred in 10% of patients after radiosurgery compared with 22% of patients after surgical resection (p = 0.06).
The PFS rate after radiosurgery was equivalent to that after resection of a Simpson Grade 1 tumor and was superior to Grade 2 and 3-4 resections in our study. If long-term follow-up confirms the high tumor control rate and low morbidity of radiosurgery, this technique will likely become the preferred treatment for most patients with small- to moderate-size meningiomas without symptomatic mass effect.

Simpson in 1957, published his landmark paper documenting the direct correlation between the degree of meningioma resection and later tumor recurrence. In this report, he identified five grades of meningioma removal. A Grade 1 resection is the complete removal of all macroscopic tumor with excision of its dural attachments and any abnormal bone. Grade 2 is macroscopically complete tumor removal with coagulation of its dural attachments. Grade 3 is macroscopically complete removal of the intradural tumor without resection or coagulation of its dural attachments or extradural extensions. Grade 4 and 5 denote partial tumor resection and decompression, respectively. With a follow-up interval extending >20 years, the rate of symptomatic tumor recurrence was 9%, 19%, and 29% for patients with Grade 1, 2, and 3 resection, respectively. Simpson concluded that meningioma surgery should be as radical as is safe, recognizing that total tumor removal is typically not feasible for many skull-based tumors or for patients with tumors attached to a great venous sinus. During the past five decades, physicians have embraced this philosophy and validated its basic tenets, creating the foundation used in the treatment of meningioma patients.

The results of the present study have significant implications for the treatment of patients with intracranial meningiomas. In a direct comparison, radiosurgery provided tumor control equivalent to a Simpson Grade 1 resection and had a higher rate of tumor control compared with patients having less than this degree of tumor removal. Moreover, the radiosurgery patients underwent fewer additional tumor treatments during the follow-up period. As expected, the ability to achieve a Simpson Grade 1 resection strongly correlated with tumor location in the surgical resection group. More than 95% of patients with convexity meningiomas had Simpson Grade 1 resections compared with less than one-third of patients with tumors involving the skull base, falx, or tentorium. Thus, in a relatively large series of patients with a mean follow-up >5 years, radiosurgery provided better tumor control than surgical resection for patients with small- to moderate-size intracranial meningiomas.

A number of factors must be considered when evaluating the results of this paper. First, several questions could be raised regarding the quality and appropriateness of the analyzed data. The information regarding patients undergoing surgical resection was retrospectively obtained, and there were significant differences in the patient characteristics of the two groups. However, because a randomized trial comparing these two techniques is not likely, only by thoughtfully evaluating the available information will a consensus opinion be possible. Also, although most patients undergoing surgical resection could have undergone radiosurgery, a number of patients had tumors causing symptomatic mass effect so that surgical resection was clearly indicated as the best available treatment. For example, patients with impaired vision related to compression of the optic apparatus frequently show improvement in their visual fields and acuity after surgical resection. Therefore, our results do not reflect a true comparison of surgical resection and radiosurgery for a matched patient cohort. However, the comparison of tumor control rates between radiosurgery and surgical resection according to Simpson grade is meaningful in the discussion of the best primary treatment for patients with intracranial meningioma. Second, the length of follow-up (5.3 years) is inadequate to conclude that one treatment is preferable to another for patients with benign meningiomas. It remains possible that treatments that appear effective today may be viewed differently as more patient follow-up data are obtained. Still, significant differences in tumor control were detected during this limited observation period between radiosurgery and patients having anything less than a Simpson Grade 1 resection. The surgical resection data also confirm the concept that anything less than complete tumor removal is likely to be only temporizing for patients with extended life expectancies Finally, the potential long-term radiation-related complications after radiosurgery must be considered when discussing this option with patients. Radiation-induced neoplasms have now been reported after radiosurgery of meningiomas and arteriovenous malformations . Nonetheless, the relative incidence of radiation-induced tumors will probably be at least a log factor less than the 1–3% typically quoted after fractionated radiotherapy Another late complication is injury to the major intracranial arteries: occlusion of the internal carotid artery has been noted after radiosurgery of cavernous sinus meningiomas  and pituitary adenomas

Despite these criticisms, our results are strengthened by their similarity to prior studies on surgical resection and radiosurgery of meningiomas. We noted a 5-year PFS rate of 96%, 82%, and 34% for Simpson Grade 1, 2, and 3–4 resection, respectively. In the original paper by Simpson , absence of tumor progression was noted in 91%, 81%, and 71% of patients with Grade 1, 2, and 3 resection, respectively. More recently, Adegbite et al. had a 5-year PFS rate of 86% for Grade 1and 82% for Grade 2 resection. In a series restricted to patients with cranial base meningiomas, Mathiesen et al. reported a 5-year PFS rate of 96%, 96%, and 75% for patients with Simpson Grade 1, 2, and 3 resection, respectively. Our PFS rate was calculated on the basis of radiographic progression, and earlier studies primarily relied on symptomatic progression to calculate the recurrence rates. With regard to radiosurgery, 1 patient had progressive tumor enlargement on serial imaging, for a crude tumor control rate of 98%. Other centers have reported tumor control rates between 84% and 98% after meningioma radiosurgery with a range of follow-up intervals Significantly, most prior studies included patients with atypical or malignant tumors who had lower control rates after radiosurgery The precise pathologic features of our radiosurgical patients are unknown because none had undergone prior surgery. Therefore, because our tumor recurrence/progression rates are well within the expected range for both our patient groups, we believe our findings are valid and likely to be reproducible at other centers.

On the basis of these data, should radiosurgery be recommended as the primary treatment for all patients with small- to moderate-size meningiomas? The answer is no. As discussed earlier, a minimal follow-up of 10 years is needed to assess adequately any technique used to treat meningioma patients. Also, patients with symptomatic mass effect rarely show significant neurologic improvement after radiosurgery and should undergo tumor removal to decompress the compromised regions. However, if after such surgery, tumor remains or recurs at a later date, radiosurgery or fractionated radiotherapy could be performed to improve PFS. At our center, patients with extended life expectancies and subtotal resection of a meningioma generally undergo radiosurgery if the remaining tumor is a discrete nodule; if the region of concern is more diffuse, fractionated radiotherapy is recommended. What our analysis does support is the concept that radiosurgery should be strongly considered as the primary treatment option for patients with tumors involving the skull base, where a Simpson Grade 1 resection often cannot be accomplished with acceptable risk The morbidity of radiosurgery for meningiomas of the cavernous sinus and petroclival regions has been specifically analyzed and is considered to be <10% . Patients with meningiomas of the falx and tentorium are another group who may benefit from radiosurgery compared with surgical resection if the tumor is not too large  Finally, on the basis of our current knowledge, radiosurgery of convexity meningiomas should be performed only in rare instances. The vast majority of these tumors can be completely excised with minimal morbidity. Kinjo et al. had no surgical morbidity and found no tumor recurrences in a series of 37 patients with convexity meningiomas. Thus, whenever radical resection of a meningioma is possible with minimal risk, the chance of delayed radiation-related complications favors tumor removal for such patients.