Christopher S. Graffeo, Michael J. Link, and Michael T. Lawton
Michael J. Link, Douglas Kondziolka, and Madjid Samii
Michael J. Link, Robert J. Coffey, Douglas A. Nichols, and Deborah A. Gorman
✓ Over the past 5 years 29 patients with dural arteriovenous fistulas (AVFs) were treated by the authors using the Leksell radiosurgical gamma knife unit. Within 2 days after radiosurgery, 17 patients with AVFs that exhibited retrograde pial or cortical venous drainage (12 patients) and/or produced intractable bruit (eight patients) underwent particulate embolization of external carotid feeding vessels. The rationale for this treatment strategy was that radiosurgery was expected to cause obliteration of most fistulas after 12 to 36 months. In patients with bruit, ocular symptoms, or in those at risk for hemorrhage, treatment with embolization after radiosurgery kept the fistulas angiographically visible for radiosurgical targeting yet offered palliation of symptoms and temporary, partial protection from hemorrhage during the latency period. In 12 patients, preobliteration embolization immediately reduced (10 patients) or eliminated (two patients) retrograde pial venous drainage. To date, no lesion has hemorrhaged after treatment. Angiography 1 to 3 years posttreatment in 18 patients showed total obliteration of 13 fistulas (72%) and partial obliteration of five (28%). Radiosurgery, followed by embolization when retrograde pial venous drainage, intractable bruit, and/or major external carotid artery supply is present, appears to be a promising treatment for selected patients with symptomatic dural AVFs.
Brian D. Milligan, Bruce E. Pollock, Robert L. Foote, and Michael J. Link
Gamma Knife surgery (GKS) for vestibular schwannoma (VS) is an accepted treatment for small- to medium-sized tumors, generally smaller than 2.5 cm in the maximum posterior fossa dimension. The purpose of this study was to evaluate the efficacy and toxicity of GKS for larger tumors.
Prospectively collected data were analyzed for 22 patients who had undergone GKS for VSs larger than 2.5 cm in the posterior fossa diameter between 1997 and 2006. No patient had symptomatic brainstem compression at the time of GKS. The median treated tumor volume was 9.4 cm3 (range 5.3–19.1 cm3). The median maximum posterior fossa diameter was 2.8 cm (range 2.5–3.8 cm). The median tumor margin dose was 12 Gy (range 12–14 Gy). Serial imaging, audiometry (10 patients with serviceable hearing pre-GKS), and clinical follow-up were available for a median of 66 months (range 26–121 months). Tumor control failure was defined as either a progressive increase in tumor diameter of at least 2 mm in any dimension or a later resection.
Four patients met the criteria for GKS failure, including 1 patient who demonstrated sarcomatous degeneration more than 7 years after GKS and died 3 months after microsurgical debulking. An enlarging cystic component was the surgical indication in 1 of the 2 patients who required resection, although 27% of tumors (6 lesions) were cystic before GKS. The 3-year actuarial rate of tumor control, freedom from new facial neuropathy, and preservation of functional hearing were 86%, 92%, and 47%, respectively. At 5 years post-GKS, these rates decreased to 82%, 85%, and 28%, respectively. At the most recent follow-up, 91% of tumors were smaller than at the time of GKS and the median maximum posterior fossa diameter reduction was 26%. On multivariate analysis, none of the following factors was associated with GKS failure, new facial weakness, new trigeminal neuropathy, or loss of serviceable hearing: patient age, tumor volume, tumor margin dose, and preoperative cranial nerve dysfunction.
Single-session radiosurgery is a successful treatment for the majority of patients with larger VSs. Although tumor control rates are lower than those for smaller VSs managed with GKS, the cranial nerve morbidity of GKS is significantly lower than that typically achieved via resection of larger VSs.
Cormac O. Maher, Robert E. Anderson, Robyn L. McClelland, and Michael J. Link
Object. The authors evaluated a new non—cross-linked, propylene oxide—treated, acellular collagen matrix for use as a dural substitute in rabbits. They then compared this material to a commonly used dural substitute as well as to native dura mater used during primary closure.
Methods. Forty-six rabbits were randomly assigned to eight groups of five or six rabbits each. These groups differed according to the type of closure material that was used during surgery (native dura, control dural substitute, or experimental dural substitute) and the duration of convalescence. At the end of the experiment, the tightness of the duraplasty was assessed in each live rabbit by continuous infusion of fluid into the cisterna magna until leakage was detected. The animals were killed and each specimen was sectioned and studied histologically. The authors found that the experimental dural substitute was safe in animals for this application, that it held sutures well, and that a watertight closure was usually achieved. There were fewer adhesions between the experimental material and neural tissue was less likely to adhere to the cranium than the control graft. Histological examination showed that the experimental material had slightly more spindle cells and vascularity than the control graft.
Conclusions. The experimental graft material has several features that make it an attractive candidate for use as a dural substitute.
Theresa M. Cheng, Michael J. Link, and Burton M. Onofrio
✓ Extreme lateral disc herniations are increasingly recognized as a cause of lumbar nerve root compression syndromes. This disorder often presents major diagnostic and therapeutic challenges, especially in the presence of multiple degenerative changes and chronic back pain in elderly patients. The authors describe two patients with presentations and findings that have not been previously described in the literature. Both patients had histories of upper lumbar back and leg pain. Degenerative spine disease, gaseous degeneration of the intervertebral discs, and epidural gas in the lateral recesses were noted on imaging studies. However, because both patients had undergone prior epidural diagnostic and therapeutic procedures, the epidural gas in the lateral recesses could be attributed either to gaseous disc degeneration or to the previous intraspinal procedures. One patient was found to have a large, far lateral extruded disc fragment that contained air. The nerve root in the second patient was impaled by an unusual combination of a small extruded disc fragment as well as an air-filled sac that was surrounded by the walled-off fragment's capsule and which freely communicated with the gaseous degenerated disc space. The suspected mechanism of root compression is illustrated and discussed. The possibility of disc herniation should be seriously considered in cases of nerve root compression in which epidural gas is present, especially those associated with gaseous degenerated discs.
Mark P. Piedra, Paul D. Brown, Paul C. Carpenter, and Michael J. Link
✓ The authors present the case of a 58-year-old woman who presented with symptoms of diabetes insipidus (DI) 1 year after she was found to have a Stage 3 (of 4) estrogen receptor—positive infiltrating ductal adenocarcinoma of the left breast with pulmonary and bone metastases. Magnetic resonance images demonstrated a solitary site of metastasis in the patient's pituitary stalk, and gamma knife surgery (GKS) was performed to treat the lesion. Three months after GKS the patient was able to reduce the medication she required for the DI. There was no evidence of pituitary failure and no negative effect on her vision.
Bruce E. Pollock, Michael J. Link, and Robert L. Foote
The decline in cranial nerve morbidity after radiosurgery for vestibular schwannoma (VS) correlates with dose reduction and other technical changes to this procedure. The effect these changes have had on tumor control has not been well documented.
The authors performed a retrospective review of 293 patients with VSs who underwent radiosurgery between 1990 and 2004 and had a minimum of 24 months of imaging follow-up (90% of the entire series). The median radiation dose to the tumor margin was 13 Gy. Treatment failure was defined as progressive tumor enlargement noted on 2 or more imaging studies. The mean postradiosurgical follow-up was 60.9 ± 32.5 months.
Tumor growth was noted in 15 patients (5%) at a median of 32 months after radiosurgery. Radiographically demonstrated tumor control was 96% at 3 years and 94% at 7 years after radiosurgery. Univariate analysis revealed 2 factors that correlated with failed radiosurgery for VS: an increasing number of isocenters (p = 0.03) and tumor margin radiation doses ≤ 13 Gy (p = 0.02). Multivariate analysis showed that only an increasing number of isocenters correlated with failed VS radiosurgery (hazard ratio 1.1, 95% CI 1.02–1.32, p < 0.05). The tumor margin radiation dose (p = 0.22) was not associated with tumor growth after radiosurgery.
Distortion of stereotactic MR imaging coupled with increased radiosurgical conformality and progressive dose reduction likely caused some VSs to receive less than the prescribed radiation dose to the entire tumor volume.
Case report and review of the literature
Brian D. Milligan, Caterina Giannini, and Michael J. Link
✓The authors report a case of a posterior fossa ganglioglioma centered in the cerebellopontine angle occurring in a child. As with cortically based gangliogliomas, the primary therapy is resection. When the tumor presents in the posterior fossa, often only partial resection can be accomplished without significant neurological deficit. The role of adjuvant chemotherapy and radiation therapy remains controversial, although these are usually reserved for high-grade lesions or progressive growth. The literature regarding the natural history, surgical outcomes, and indications for adjuvant therapy is reviewed. Although it occurs rarely, ganglioglioma should be included in the differential diagnosis of a posterior fossa mass in a child or young adult.