Acoustic Neuromas (Vestibular Schwannoma)

Historically and incorrectly referred to as an acoustic neuroma, we prefer the designation of vestibular schwannoma which recognizes the anatomic and histologic origins of these tumors.  There may be no other intracranial neuropathology about which the proper treatment arouses as much controversy as the vestibular schwannoma. Neurosurgeons cite the series of surgeons with enormous experience removing these tumors to justify suboccipital removal, while otolaryngologists sacrifice the inner ear during the translabarynthine approach in an attempt to better expose and preserve the facial nerve. Radiosurgery's proponents cite excellent tumor control and low morbidity but must acknowledge that although the tumor often shrinks, it is still there. Therefore, it is in the best interest of our patients that long term outcome in patients treated with these three modalities be thoroughly evaluated, and that small series, anecdotal evidence and personal beliefs not weigh too heavily in our minds.

The first vestibular schwannomas treated with the gamma knife were by Leksell and Steiner in 1969. Since then more than 21,272 have been treated around the world through June 2003. The indications for gamma knife surgery for this tumor vary.  Some physicians advocate gamma knife surgery in medically high risk patients, patients who refuse microsurgery, and  in patients with post-operative residual tumor.  However, others advocate gamma knife surgery as the treatment of choice in nearly all cases of vestibular schwannomas.  The usefulness of irradiation in the post-operative period was shown by Wallner in 1987 where external beam irradiation lowered the recurrence rate from 46 to 6 percent in Boldrey's surgical series at the University of California at San Francisco. By then, gamma surgery was already being widely applied to this disease under many circumstances.  The fact is that for a number of reasons few neurosurgeons acquire the necessary competency to satisfactorily extirpate these tumors.  This situation may change if a method to improve the acquisition of skills required for extirpation of these lesions is found.

The advent of MRI has made planning for this procedure much more exact. With a high quality MRI scan and a relatively small tumor, the seventh cranial nerve can occasionally be visualized and carefully excluded from the treatment field. The trigeminal nerve can nearly always be identified except with the largest tumors, which in most cases should not be treated primarily with gamma surgery.

Small collimators are used to better match the isodose configuration to the size and shape of the tumor. We have had no brainstem related complications. Previously we used minimum periphery doses up to 20Gy and maximum doses up to 70Gy. Presently, we use a margin dose of 11 to 15Gy at the 30 to 50 percent isodose curve. The incidence of cranial nerve palsies rose considerably at the higher doses without significant improvement in the degree of tumor control.

At the University of Virginia, we have treated 400 patients with vestibular schwannomas. One-hundred fifty-three of these patients with greater than twelve months follow-up have been reported. Of these radiosurgery was the primary treatment for 96 and was adjutant (following microsurgery) in 57. The volume of the treated tumors ranged from 0.02 to 18.3 cm³.

Of the patients treated primarily with Gamma surgery, a decrease in tumor size was seen in 81 percent (78 patients), no change in 12 percent, and an increase in size in 6 percent. Among those 78 patients with a decrease in the size of their tumors, the decrease was greater than 50 percent in 20 patients. It is our policy to not consider decreases in volume of less than 15 percent as significant. This is true of all tumors and vascular malformations that we treat. Radiological follow-up for these patients ranged from 1 to 10 years.

Of the 57 patients treated with Gamma surgery after microsurgery, a decrease in tumor size was seen in 65 percent, no change in 25 percent, and an increase in size in 10 percent. Among the 37 patients with a decrease in the size of their tumors the decrease was greater than 50 percent in 12 patients.  The outcome in terms of post-radiosurgical volume reduction in patients who had prior microsurgery is not as favorable as those who were primarily treated with Gamma surgery.  This difference is likely a result of the increased difficulty with accurate targeting in those who have undergone prior microsurgery.  Of note, although our experience with treating large vestibular schwannomas is small (n=19), we have observed a 95% tumor control rate in these following gamma surgery.

In our patients, there were five with transient changes in trigeminal sensation and three with facial paresis. One of the patients with facial weakness was operated upon shortly after Gamma surgery and was lost to follow-up. Another patient recovered completely in six weeks, and the third has nearly completely recovered at ten months.   Of the patients with useful hearing prior to gamma surgery, 58% retained their hearing following radiosurgery, 42% experienced some degree of deterioration, and 31% lost useful hearing. The majority of hearing changes were observed at the 2-year checkup, and additional auditory changes were observed as late as 8 years post-radiosurgery.

Other centers report similar rates of tumor control (i.e. with no change or decrease in the size of the tumor) seen in 89 to 100 percent of patients.

Evaluation of the material from the Karolinska group included evaluation of radiographic changes besides size.   The most common change was loss of central enhancement within the tumor on either contrasted MRI or CT studies. This occurred in 70 percent of patients and typically was observed within 6 to 12 months of treatment.  However, these changes were reversible. Another change that was observed and that we have often seen is a transient increase in the size of the tumor during the first 6 months after Gamma surgery. This is commonly seen in tumors that then regress to their original size or smaller. 

Previously published incidence of cranial neuropathies at other centers was 17 percent at Karolinska and 29 percent at Pittsburgh for facial paresis which in the vast majority of cases was transient or mild. The tri

geminal nerve was affected in a variety of ways in 33 percent of the time at Pittsburgh, most commonly a mild hypoesthesia. Recent complication rates at these institutions are comparable to those at our center.

We have not seen an instance of cerebellar edema or hydrocephalus requiring spinal fluid diversion following Gamma surgery for vestibular schwannomas, but both of these have been reported elsewhere.

TABLE 5

OUTCOME OF RADIOSURGERY FOR ACCOUSTIC NEUROMAS