The primary management of the majority of symptomatic

Transcription

1 J Neurosurg 116: , 2012 Cranial nerve dysfunction following Gamma Knife surgery for pituitary adenomas: long-term incidence and risk factors Clinical article Christopher P. Cifarelli, M.D., Ph.D., David J. Schlesinger, Ph.D., and Jason P. Sheehan, M.D., Ph.D. Department of Neurological Surgery, University of Virginia Health System, Charlottesville, Virginia Object. Gamma Knife surgery (GKS) has become a significant component of neurosurgical treatment for recurrent secretory and nonsecretory pituitary adenomas. Although the long-term risks of visual dysfunction following microsurgical resection of pituitary adenomas has been well studied, the comparable risk following radiosurgery is not well defined. This study evaluates the long-term risks of ophthalmological dysfunction following GKS for recurrent pituitary adenomas. Methods. An analysis of 217 patients with recurrent secretory (n = 131) and nonsecretory (n = 86) pituitary adenomas was performed to determine the incidence of and risk factors for subsequent development of visual dysfunction. Patients underwent ophthalmological evaluation as part of post-gks follow-up to assess for new or worsened cranial nerve II, III, IV, or VI palsies. The median follow-up duration was 32 months. The median maximal dose was 50 Gy, and the median peripheral dose was 23 Gy. A univariate analysis was performed to assess for risk factors of visual dysfunction post-gks. Results. Nine patients (4%) developed new visual dysfunctions, and these occurred within 6 hours to 34 months following radiosurgery. None of these 9 patients had tumor growth on post-gks neuroimaging studies. Three of these patients had permanent deficits whereas in 6 the deficits resolved. Five of the 9 patients had prior GKS or radiotherapy, which resulted in a significant increase in the incidence of cranial nerve dysfunction (p = ). An increased number of isocenters (7.1 vs 5.0, p = 0.048) was statistically related to the development of visual dysfunction. Maximal dose, margin dose, optic apparatus dose, tumor volume, cavernous sinus involvement, and suprasellar extension were not significantly related to visual dysfunction (p > 0.05). Conclusions. Neurological and ophthalmological assessment in addition to routine neuroimaging and endocrinological follow-up are important to perform following GKS. Patients with a history of radiosurgery or radiation therapy are at higher risk of cranial nerve deficits. Also, a reduction in the number of isocenters delivered, along with volume treated, particularly in the patients with secretory tumors, appears to be the most reasonable strategy to minimize the risk to the visual system when treating recurrent pituitary adenomas with stereotactic radiosurgery. ( Key Words Gamma Knife pituitary surgery vision cranial nerve Abbreviations used in this paper: GKS = Gamma Knife surgery; RTSS = repeat transsphenoidal surgery. The primary management of the majority of symptomatic pituitary adenomas remains either endoscopic or microneurosurgical resection. Advances in preoperative imaging and intraoperative guidance systems have improved the current standards by which these lesions are identified and the extent of their safe resection. However, despite the technical and diagnostics improvements, a subset of pituitary adenomas, both secretory and nonsecretory, requires further therapy following incomplete resection or recurrence. 2,14 The decision to perform RTSS has been the subject of considerable debate for both functional and nonfunctional lesions, with an obstacle to treatment being the potential for increased operative morbidity associated with secondary surgery. 2,4,14,18 More recently, with respect to functional lesions, such as adrenocorticotropic hormone secreting tumors in Cushing disease, several groups have advocated RTSS as low risk but also report only a 61% 71% remission rate following the procedure, requiring further adjuvant therapy, usually involving radiotherapy. 21,27 The use of conventional external-beam radiotherapy in residual or recurrent lesions has the relative deterrents of resultant hypopituitarism, cranial neuropathies, cerebrovascular accidents, or radiation-induced tumor formation as both short-term and long-term sequelae to treatment. 3,8 Hence, radiosurgery has emerged as the preferred method of radiation delivery. 5 In the present study, we retrospectively examined the effects of GKS on cra- This article contains some figures that are displayed in color on line but in black-and-white in the print edition J Neurosurg / Volume 116 / June 2012

2 Cranial nerve dysfunction after GKS for pituitary adenomas nial nerve dysfunction in patients with recurrent pituitary adenomas, including both functional and nonfunctional adenomas. Methods Study Design This is a retrospective study of patients with recurrent pituitary adenomas who underwent GKS at the University of Virginia between 1991 and The research study was done in accordance with institutional policies at the University of Virginia. Patients were treated with GKS at the sign of progression or recurrence of a nonfunctioning pituitary adenoma. Those with a secretory adenoma were treated with GKS when they demonstrated persistence of a hypersecretory state. A total of 438 patients with pituitary adenomas were captured in our prospective GKS database, and of those patients, 217 had reliable imaging and ophthalmological follow-up. Clinical Evaluation Patient age, sex, clinical history, tumor histology, and treatment details were obtained by chart review. Analysis of these factors was undertaken to determine their potential influence on the progression of pretreatment visual deficits or development or new visual impairments. Evaluation following the initial resection of the pituitary adenoma consisted of a complete neuroophthalmological examination, including assessment of visual acuity, extraocular muscle movement, and visual fields performed by an ophthalmologist and often one specializing in neuroophthalmology. Patients were examined for cranial nerve deficits after GKS at intervals cotemporaneous with radiological assessment at 6-month intervals for 2 years, at 1-year intervals during Years 2 5 post-gks, and then every 2 years thereafter. If a patient had a complaint of a new neurological problem, that patient was reevaluated clinically and radiographically at the time the complaint was raised. Postradiosurgery, patients were assessed by the treating neurosurgeon and/or an ophthalmologist. Gamma Knife Surgery Technique The Gamma Knife (Elekta AB) technique used has been previously described. 24 In brief, each patient underwent placement of a stereotactic Leksell frame under monitored anesthesia. The patient then underwent thinslice stereotactic MRI (or CT if MRI was contraindicated) through the pituitary, using contrast enhancement and fat-suppression techniques. Neuroimaging studies were imported into the Gamma Knife workstation and a conformal dose planned was rendered. The dose to the optic nerve was defined as the maximum dose to a 1% volume of the optic apparatus. The optic apparatus volume was computed by contouring the optic nerves and chiasm on thin-slice MR images. Segmentation of the optic apparatus was done using GammaPlan and contours were drawn by the treating neurosurgeon. In the current study, the mean maximal dose was 44.6 Gy (range Gy). The mean marginal dose was 19.9 Gy (range 1 30 Gy). J Neurosurg / Volume 116 / June 2012 Statistical Analysis The data were analyzed using S-PLUS 8.0 statistical software (MathSoft, Inc.). Testing for statistically significant associations between visual dysfunction following GKS and variables including maximal dose, margin dose, optic dose, volume, and number of isocenters was carried out using the 2-tail Student t-test. An F-test was used for determination of equality of variance for use of appropriate T-test parameters. Chi-square analysis was used to test for associations between visual dysfunction and categorical predictors including cavernous sinus invasion, suprasellar extension, and prior radiotherapy. Statistical significance was regarded as a p < 0.05 for all methods. The time to development of cranial nerve deficit was demonstrated via step-graph based on the total number of patients studied (n = 217) as well as an adjusted value expressed as the estimate of deficit incidence incorporating all patients lost to follow-up during that period. Results Patient Demographics Between 1991 and 2009, 418 patients with pituitary adenomas were captured in our prospective Gamma Knife Center database. Of these, 217 patients with recurrent pituitary adenomas and complete ophthalmological assessment of cranial nerve function post-gks were included in the study: 131 patients with recurrent secretory lesions and associated persisting hormonal abnormalities, and 86 patients presenting with nonfunctioning pituitary adenomas. The distribution of pituitary adenoma types is depicted in Fig. 1. The median clinical follow-up period following GKS was 32.5 months (range months). Cranial Nerve Dysfunction In 9 patients (4%) new or worsened cranial nerve dysfunction developed following GKS. Of these 9, 3 had previously undergone GKS and 2 had previously undergone external-beam radiotherapy. Of the deficits after GKS, 3 affected cranial nerve II, 5 affected cranial nerve III, 1 affected cranial nerve IV, and 3 affected cranial nerve VI. Of these, 2 of the second cranial nerve dysfunctions and 1 third nerve palsy failed to resolve within the study period. All other cranial nerve injuries related to visual dysfunction appeared to resolve over time and with a brief course of corticosteroid administration. Moreover, none of the patients who experienced the aforementioned cranial nerve injuries required further treatment, based on resolution of the underlying endocrinopathy or lack of tumor growth. Of the 9 patients in whom transient or permanent cranial nerve dysfunction was found to have developed, 8 had secretory lesions (Fig. 2). Statistical analysis revealed that an increasing number of isocenters (mean 7.1) in patients with visual dysfunction compared with patients without new visual findings (mean 5.0) was the only significant predictor of the development of visual dysfunction during the study period (p = 0.048). Among the secretory lesions treated, the average volume in patients with visual dysfunction was 3.8 ± 1.2 cm 3 compared with 1.9 ± 0.2 cm 3 in patients with secretory lesions 1305

3 C. P. Cifarelli, D. Schlesinger, and J. P. Sheehan Fig. 1. Distribution of tumors in patients included in the study. All patients underwent a full endocrinological evaluation to determine the secretory or nonsecretory status of the tumor. Sixty percent of patients had endocrinologically active tumors while the remaining 40% were nonfunctional or null cell tumors. NFA = nonfunctioning adenoma. in whom no visual symptoms developed. Although this difference did not reach statistical significance by T-test (p = 0.15), the volume of 3.8 ± 1.2 cm 3 for secretory lesions treated with resultant visual dysfunction fell outside of the 95% confidence interval (CI cm 3 ) (Fig. 3). Patients treated with prior radiation (either previous fractionated therapy and/or GKS) appeared to have an increased incidence of cranial nerve deficits, with 5 of the 9 patients with cranial nerve injury comprising this group (p = ). All other variables tested, including maximal dose, margin dose, dose received by the optic apparatus, cavernous sinus invasion, and suprasellar extension, failed to clearly demonstrate any statistically significant relationships to the development of cranial nerve dysfunction following GKS in this pituitary adenoma patient population (Table 1). Time to Onset of Cranial Nerve Dysfunction Cranial nerve dysfunction resulting in a visual deficit occurred between 6 hours and 34 months after GKS treatment. The majority (67%) of these patients presented within the first 12 months of follow-up, whereas 3 patients developed a visual deficit in the absence of tumor progression between 25 and 34 months following radiosurgery. These data are expressed as a function of the total 217 patients included in the study (solid line, Fig. 4) and as an adjusted value based on the assumed incidence among patients lost to follow-up being the same as the population with intact follow-up (dashed line, Fig. 4). At 36 months, the observed visual deficit was 4.2% (n = 9) of the total 217 patients, whereas the estimated incidence, based on the assumption that those patients lost to followup had at least the same rate of visual deficit development, was 8.1% (n = 17.5). The observed values and estimated values are summarized in Table 2. Discussion As to the scope of neoplasms encountered within the CNS, pituitary adenomas pose a unique set of issues with Fig. 2. Incidence of cranial nerve (CN) deficits based on primary diagnosis. A total of 9 patients (4%) developed cranial nerve deficits as demonstrated by ophthalmological examination in all 217 patients. Eight of these patients with a cranial nerve deficit had functional tumors, and only 1 patient with a nonfunctional tumor developed a cranial nerve deficit during the study period. regard to natural history, clinical course, surgical management, and follow-up examination. The slow growth rates and recurrences (both early and late) in patients with pituitary adenoma underscore the importance of close follow-up and repeat intervention should symptoms recur or fail to resolve. Along these lines, we also have the duty to offer potential patients a reliable estimate of the associated potential morbidities, including development of cranial nerve dysfunction. Fig. 3. Volume of tumors treated and incidence of cranial nerve deficits. Statistical analysis of all patients based on volume of tumor (treatment area) demonstrated a quantitative difference of the mean between those patients with cranial nerve deficit versus those without (3.8 cm 3 and 1.9 cm 3, respectively), which did not reach significance (p = 0.15) using the Student t-test, but both values fall outside the 95% confidence interval for total tumor volume among all study patients J Neurosurg / Volume 116 / June 2012

4 Cranial nerve dysfunction after GKS for pituitary adenomas TABLE 1: Summary of significant factors related to the development of cranial nerve deficits following GKS* Patient Group No Visual Deficit Visual Deficit (n = 9) Factor p Value 95% CI mean maximal dose (Gy) mean margin dose (Gy) maximum optic apparatus dose (Gy) cavernous sinus invasion suprasellar adenoma extension prior radiotherapy mean vol (cm 3 ) mean no. of isocenters * Statistical analysis was performed on all variables collected that related to treatment conditions (mean maximal dose, dose to optic system, and so on) as well as historical factors, such as prior radiotherapy, including previous GKS. Chi-square analysis of the structural characteristics of the tumors (that is, cavernous sinus invasion or suprasellar extension) failed to demonstrate a significant (p < 0.05) relationship with respect to cranial nerve deficits. Conversely, prior radiotherapy was a significant factor in the development of cranial nerve deficits among our patient population (p = ). The mean number of isocenters was increased in patients in whom cranial nerve deficits developed (p = 0.048). Calculated using the chi-square test. p < Focusing on new cranial nerve deficits for pituitary adenomas following GKS, we demonstrated that the overall risk to this population is low for all lesion subtypes, based on incidence and time-to-event analysis. An independent variable associated with increased risk to the optic apparatus appears to be an increasing number of isocenters. It is possible that the increased number of isocenters translated to more dose heterogeneity within the treatment volume and could have resulted in hot spots over certain segments of cranial nerves, thus translating into deterioration in neurological function. In addition, the exposure to prior radiation, either with GKS or fractionated radiotherapy, has proven to be a risk factor Fig. 4. Time to development of cranial nerve dysfunction. Patients were followed by neuroophthalmological assessment from the time of GKS (Day 0 on the x axis) up to a mean of 32.5 months. The percentage of the total number of study patients (n = 217) with unaffected vision is demonstrated as a function of time (solid line). An adjusted estimate of the percentage of patients with unaffected vision was determined using the incidence of visual deficit applied to the patients lost to follow-up (dashed line). J Neurosurg / Volume 116 / June 2012 for cranial deficits in our patient population, resulting in increased vigilance with patients considering repeated radiosurgery. In the current series, patients were re-treated with GKS only if they had a functioning adenoma and experienced an initial remission and later recurrence of their hypersecretory state. Re-treatment of such patients may still prove reasonable but dose reduction or a longer period between initial GKS and later re-treatment, thereby allowing for repair and recovery, may be advisable. One could also consider a multisession radiosurgery approach upon re-treatment with the Gamma Knife Extend or LINAC-based radiosurgery devices such as CyberKnife or Novalis. The dose of radiation delivered to the margin, extension of the lesion outside the boundaries of the sella, and increased maximal dose did not appear to have any significant relationship to the development of post-gks cranial nerve decline. It is quite likely that the failure to demonstrate a significant relationship between dose and postradiosurgery cranial nerve dysfunction is a result of the low incidence of nerve deficits and the relatively low statistical power achieved in this study cohort of 217 patients. Radiosurgical Efficacy and Cranial Dysfunction Long-term analyses from several large institutions have shown the overall efficacy of stereotactic radiosurgery for benign lesions, including pituitary adenomas, as means of tumor control, with rates ranging from 86% to 97%, depending on whether the end point was either stabilization or a decrease of tumor size. 12,13,17 Here, we expand the available literature to include a large consecutive series consisting of both functional and nonfunctional pituitary lesions, specifically addressing the risk of developing new cranial nerve deficits following radiosurgery. Several recent studies have attempted to systematically analyze both the natural history of nonsecretory adenomas and the surgical outcomes through meta-analysis. In- 1307

5 C. P. Cifarelli, D. Schlesinger, and J. P. Sheehan TABLE 2: Observed number of patients with visual dysfunction versus adjusted incidence based on patients lost to follow-up* Time (mos) Total w/ Visual Dysfunction No. of Patients (% total) Adjusted No. w/ Visual at Follow-Up Dysfunction 0 to <6 2 (1) 205 (94.5) 2.2 (1) 6 to <12 5 (2.3) 183 (84) 4.8 (2.2) 12 to <18 6 (2.7) 156 (72) 8.2 (3.8) 18 to <24 6 (2.7) 140 (65) 9.1 (4.2) 24 to <30 7 (3.2) 123 (57) 12.2 (5.6) 30 to <36 9 (4.1) 110 (51) 17.5 (8.1) * Based on 6-month intervals following GKS, the observed incidence of visual deficits is compared to the adjusted incidence based on the same rate of visual dysfunction occurrence in patients lost to follow-up. Absolute values are reported along with percentages of the total number of 217 patients. The adjusted value provides a range of visual dysfunction incidence from 4.1% to a projected value of 8.1% at 36 months after GKS. terestingly, the conclusions of these works have revealed an advantage for a combined surgical and radiotherapy approach for tumor control and improvement of visual symptoms, while the natural history of nonsecretory adenomas suggested a favorable risk-to-benefit ratio for the treatment of macroadenomas with greater than 3.5 mm growth over an average of approximately 4 years. 7,20 Overall, the observed incidence of new visual dysfunction in our patient population remained quite low (4.1%) based on long-term follow-up at 36 months. Once again, our patient population harbored both secretory and nonsecretory lesions, but the data follow a trend seen for benign skull base tumors, in which combined surgical and radiosurgical treatment affords control of disease with a low risk to surrounding neural structures. Balancing Risk of Radiosurgery Compared With RTSS The treatment of pituitary adenomas with radiation has undergone a significant revolution since its inception. The current capabilities of modern radiosurgical devices have shifted the treatment paradigm whereby radiosurgery is frequently used after recurrent or residual disease in patients with pituitary adenomas. 20 Our study demonstrates an acceptable risk (4%) to the cranial nerve structures compared with the historical risk of approximately 5% 15% in patients undergoing RTSS. 14 Even with advances in the endoscopic resection of recurrent lesions over the past 20 years, several recent studies of patients with persistent Cushing disease still only report a 60% 70% remission rate with RTSS, with 1 study noting a complication rate of 57% in these procedures, including 1 stroke and 1 sphenopalatine artery hemorrhage. 21,27 Moreover, Patil et al. 21 increased the remission rate to 83% by providing radiosurgery after failed RTSS. Thus, the algorithm of initial transsphenoidal resection, be it endoscopic, microscopic, or a combined method, followed by GKS for residual or recurrent disease, appears to be a reasonable approach based on our findings compared with the best available historical controls. Tolerance of Cranial Nerves to Radiation Much has been written about the tolerance of cranial nerves to ionizing radiation. The optic nerves are believed to be the most sensitive of the cranial nerves to radiation. They generally can tolerate an upper limit of 8 12 Gy. 19,23 Some authors have even argued for a higher tolerable dose of up to 18 Gy. 9 Multisession radiosurgical approaches, albeit with the Gamma Knife or LINACbased radiosurgical devices, may allow higher maximum doses to be delivered to the optic apparatus as long as they are delivered over 2 5 fractions. 1,11 This tolerable dose likely depends on the volume irradiated and the pretreatment function of the nerves. 15,19,25 It may also depend on the nonlethal systemic comorbidities such as diabetes or collagen vascular disease. 6 Far less has been written about the tolerance of cranial nerves in the cavernous sinus to radiosurgery. Leber and colleagues 15 reported no signs of neuropathies in patients undergoing radiosurgery in which doses of 5 30 Gy were delivered to the cavernous sinus. However, Tishler and associates 26 noted that 8 of 62 radiosurgery-treated patients with cavernous sinus pathology developed cranial nerve injury manifesting with deficits of cranial nerves III VI. Despite this, they could not find a clear maximum tolerable dose to cavernous sinus cranial nerves within the range of Gy delivered in this study. We have observed a disproportionately high rate of cranial nerve deficits in patients re-treated with GKS. In the present series 3 of the 9 patients with deficits were retreated with GKS for a recurrence of their hypersecretory state after a period of documented endocrine remission following initial GKS. In 2 of these 3 patients, a second GKS was performed within 3 years of the original GKS. This observation was previously reported by our group. 10 It was only after careful consideration of the therapeutic alternatives and patient counseling regarding the increased risk of cranial nerve dysfunction that we now perform repeat GKS for patients with recurrent secretory adenomas. In such patients, a reduction in dose at the second GKS is recommended. Optimally, there should be a period of greater than 3 years between the initial GKS and a second GKS. Timing of Onset of Cranial Nerve Dysfunction After Radiosurgery In the current study, the majority of patients developed cranial nerve dysfunction within 12 months of radiosurgery, but in some patients onset was as late as 34 months. A review of the available literature addressing the long-term incidence of cranial nerve dysfunction following radiosurgery reveals a general paucity of data. Kuo et al. 13 report the longest interval between treatment and incidence of cranial nerve palsy in the region of the sella/cavernous sinus (48 months in 2 patients with sixth cranial nerve palsies treated for benign cavernous sinus tumors). Although these deficits were temporary, 2 additional patients in the same study presented with permanent sixth cranial nerve palsies at 43 and 45 months following treatment. Taking into consideration all lateeffect radiation-related cranial nerve palsies described in the literature, including those encountered with tradi J Neurosurg / Volume 116 / June 2012

6 Cranial nerve dysfunction after GKS for pituitary adenomas tional fractionated modalities, hypoglossal dysfunction in patients treated for nasopharyngeal carcinoma has the longest reported latency period, ranging from 12 to 240 months, further emphasizing the need for long-term follow-up. 16,22 Study Limitations The retrospective nature of the present study presents some obstacles for interpretation. First, the number of patients lost to follow-up was minimal for the first 12 months following GKS, but the number increased to 49% at 36 months. Even accounting for those lost to follow-up in the estimated incidence of visual deficit, there remains the possibility that this is an underestimate, although this is unlikely because the onset of visual dysfunction would probably have prompted re-presentation to our clinical team. Next, the argument that hot spots along segments of affected cranial nerves may cause the dysfunction through an increased number of isocenters is impossible to prove, as it would require prospective detailed analysis of the treatment plans, which is not possible with our center s database. Additionally, the follow-up period for the current study, although long term based on comparable radiosurgery studies from other groups, may not be long enough to determine the absolute risk to cranial nerves following dose delivery in the region of the sella. Finally, although a complete neurological assessment was performed as part of the routine follow-up care, formal automated visual field testing was not conducted in all patients. Thus, the possibility exists that subtle changes in cranial nerve function may have gone unnoticed, although the alternative study, in which automated visual fields are performed in all patients prospectively, may prove to be logistically difficult to complete. Conclusions The risk of new or worsened cranial nerve dysfunction following radiosurgery of a pituitary adenoma is low. Most of the cranial nerve dysfunction in our series was temporary in nature. The onset of cranial dysfunction ranged from within 6 hours to 34 months of treatment. Patients need long-term neurological and ophthalmological follow-up after radiosurgery. Decreasing the number of isocenters may help to decrease the risk of postradiosurgical cranial nerve deficits. Gamma Knife surgery after previous radiation therapy or radiosurgery for a pituitary adenoma may also convey a higher risk of cranial nerve dysfunction. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Sheehan, Cifarelli. Ac quisition of data: Sheehan, Cifarelli. Analysis and interpretation of data: Sheehan, Cifarelli. Drafting the article: all authors. Critical ly revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Sheehan. Statistical analysis: Cifarelli. Study supervision: all authors. J Neurosurg / Volume 116 / June 2012 References 1. Adler JR Jr, Gibbs IC, Puataweepong P, Chang SD: Visual field preservation after multisession cyberknife radiosurgery for perioptic lesions. Neurosurgery 62 (Suppl 2): , Benveniste RJ, King WA, Walsh J, Lee JS, Delman BN, Post KD: Repeated transsphenoidal surgery to treat recurrent or residual pituitary adenoma. J Neurosurg 102: , Boelaert K, Gittoes NJ: Radiotherapy for non-functioning pituitary adenomas. Eur J Endocrinol 144: , Chang EF, Sughrue ME, Zada G, Wilson CB, Blevins LS Jr, Kunwar S: Long term outcome following repeat transsphenoidal surgery for recurrent endocrine-inactive pituitary adenomas. Pituitary 13: , Chen L, White WL, Spetzler RF, Xu B: A prospective study of nonfunctioning pituitary adenomas: presentation, management, and clinical outcome. J Neurooncol 102: , Chon BH, Loeffler JS: The effect of nonmalignant systemic disease on tolerance to radiation therapy. Oncologist 7: , Fernández-Balsells MM, Murad MH, Barwise A, Gallegos- Orozco JF, Paul A, Lane MA, et al: Natural history of nonfunctioning pituitary adenomas and incidentalomas: a systematic review and metaanalysis. J Clin Endocrinol Metab 96: , Gittoes NJ: Pituitary radiotherapy: current controversies. Trends Endocrinol Metab 16: , Hasegawa T, Kobayashi T, Kida Y: Tolerance of the optic apparatus in single-fraction irradiation using stereotactic radiosurgery: evaluation in 100 patients with craniopharyngioma. Neurosurgery 66: , Jagannathan J, Sheehan JP, Pouratian N, Laws ER, Steiner L, Vance ML: Gamma Knife surgery for Cushing s disease. J Neurosurg 106: , Kim JW, Im YS, Nam DH, Park K, Kim JH, Lee JI: Preliminary report of multisession gamma knife radiosurgery for benign perioptic lesions: visual outcome in 22 patients. J Korean Neurosurg Soc 44:67 71, Kondziolka D, Nathoo N, Flickinger JC, Niranjan A, Maitz AH, Lunsford LD: Long-term results after radiosurgery for benign intracranial tumors. Neurosurgery 53: , Kuo JS, Chen JC, Yu C, Zelman V, Giannotta SL, Petrovich Z, et al: Gamma knife radiosurgery for benign cavernous sinus tumors: quantitative analysis of treatment outcomes. Neurosurgery 54: , Laws ER Jr, Fode NC, Redmond MJ: Transsphenoidal surgery following unsuccessful prior therapy. An assessment of benefits and risks in 158 patients. J Neurosurg 63: , Leber KA, Berglöff J, Pendl G: Dose-response tolerance of the visual pathways and cranial nerves of the cavernous sinus to stereotactic radiosurgery. J Neurosurg 88:43 50, Lin YS, Jen YM, Lin JC: Radiation-related cranial nerve palsy in patients with nasopharyngeal carcinoma. Cancer 95: , Liscák R, Vladyka V, Marek J, Simonová G, Vymazal J: Gamma knife radiosurgery for endocrine-inactive pituitary adenomas. Acta Neurochir (Wien) 149: , Long H, Beauregard H, Somma M, Comtois R, Serri O, Hardy J: Surgical outcome after repeated transsphenoidal surgery in acromegaly. J Neurosurg 85: , Mayo C, Martel MK, Marks LB, Flickinger J, Nam J, Kirkpatrick J: Radiation dose-volume effects of optic nerves and chiasm. Int J Radiat Oncol Biol Phys 76 (3 Suppl):S28 S35, Murad MH, Fernández-Balsells MM, Barwise A, Gallegos- Orozco JF, Paul A, Lane MA, et al: Outcomes of surgical treatment for nonfunctioning pituitary adenomas: a systematic review and meta-analysis. Clin Endocrinol (Oxf) 73: ,

7 C. P. Cifarelli, D. Schlesinger, and J. P. Sheehan 21. Patil CG, Veeravagu A, Prevedello DM, Katznelson L, Vance ML, Laws ER Jr: Outcomes after repeat transsphenoidal surgery for recurrent Cushing s disease. Neurosurgery 63: , Schinagl DA, Marres HA, Kappelle AC, Merkx MA, Pop LA, Verstappen SM, et al: External beam radiotherapy with endocavitary boost for nasopharyngeal cancer: treatment results and late toxicity after extended follow-up. Int J Radiat Oncol Biol Phys 78: , Sheehan JP, Niranjan A, Sheehan JM, Jane JA Jr, Laws ER, Kondziolka D, et al: Stereotactic radiosurgery for pituitary adenomas: an intermediate review of its safety, efficacy, and role in the neurosurgical treatment armamentarium. J Neurosurg 102: , Sheehan JP, Pouratian N, Steiner L, Laws ER, Vance ML: Gamma Knife surgery for pituitary adenomas: factors related to radiological and endocrine outcomes. Clinical article. J Neurosurg 114: , Stafford SL, Pollock BE, Leavitt JA, Foote RL, Brown PD, Link MJ, et al: A study on the radiation tolerance of the optic nerves and chiasm after stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 55: , Tishler RB, Loeffler JS, Lunsford LD, Duma C, Alexander E III, Kooy HM, et al: Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys 27: , Wagenmakers MA, Netea-Maier RT, van Lindert EJ, Timmers HJ, Grotenhuis JA, Hermus AR: Repeated transsphenoidal pituitary surgery (TS) via the endoscopic technique: a good therapeutic option for recurrent or persistent Cushing s disease (CD). Clin Endocrinol (Oxf) 70: , 2009 Manuscript submitted September 15, Accepted February 16, Please include this information when citing this paper: published online March 16, 2012; DOI: / JNS Address correspondence to: Jason P. Sheehan, M.D., Ph.D., De partment of Neurological Surgery, University of Virginia Health Sys tem, Box , Charlottesville, Virginia jsheehan@virginia.edu J Neurosurg / Volume 116 / June 2012