[Radiosurgery or microsurgery of vestibular schwannomas?]

Microsurgical excision is the reference treatment of vestibular schwannomas. Yet, morbidity and functional risk of this surgery is significant, as Pellet et al have demonstrated in their study. In order to define the role of gamma knife surgery we have designed a prospective study concerning effectiveness and tolerance of this treatment.

Between July 14, 1992 and August, 1997, 400 patients were included. All the treatments were carried out according to a homogeneous methodology, with multiisocentric planning. We use low marginal doses dependent mainly on the treatment volume: 14 Gy for small stade II tumors, 12 Gy or less for larger tumors and 16 Gy for intracanalicular tumors. The evaluation protocol included preoperative examination with clinical examination, House grading, Shirmer's test, tonal and vocal audiometry, brain stem electrical response audiometry (BERA), vestibular caloric and pendular tests, magnetic resonance imaging (MRI) or computerized tomography (CT) scan; control at 6 months, 1 year and 2 years with clinical examination, tonal audiometry and MRI and/or CT scan; at 3 years the preoperative examination was repeated and a questionnaire based on Pellet's concerning functional results was completed.

Among the 400 treatments, 80% were first intended treatment of unilateral vestibular schwannoma. At the time of the analysis, 100 consecutive patients with unilateral schwannoma (treated in first intention) had a 3 year follow-up (results concerned these 100 patients). Average age was 61 years (17-82 years). According to Portmann's classification, five patients presented a stade 1 tumor, 60 a stade 2 tumor, 33 a stade 3, and two a stade 4 tumor. According to House's grading at preoperative examination, there was 86 (86%) grade 1 tumors, 12 grade 2, two grade 3, and at 3 year follow-up: 77 (94%) grade 1 and five grade 2 (17 patients had no House grading). At preoperative examination, three patients presented an hemispasm; at 3 years this had disappeared for all patients. Two others patients presented a transient hemispasm at 8 and 11 months, respectively. At preoperative examination, four patients presented with facial numbness and 14 with hypoesthesia. At 3 year follow-up, trigeminal function was normal for all patients but one, for whom this had only improved. Seven others patients presented a transient numbness or hypoesthesia. At preoperative examination, five patients presented hydrocephalus without cerebro spinal fluid (CSF) shunting. At 3 year follow-up, seven patients presented CSF shunting: three presented a preoperative hydrocephalus, three a hydrocephalus after treatment and one a hydrocephalus secondary to tumor growth. Average age for these six patients was 71 years. Audiological classification was based on Gardner-Robertson's. Seventy percent of patients with normal hearing maintained useful hearing, and 50% of patients with useful hearing maintained serviceable hearing. Three (3%) patients (two with stade 2 tumors and one with a stade 3) had microsurgical excision at 16, 35 and 36 months, respectively. During microsurgical excision no unusual difficulty was encountered. Seventeen questionnaires investigating functional outcome and quality of life were completed: 100% (63% in Pellet's study) of the patients answered they had no facial motion disturbance, 49% (17% in Pellet's study) had no ocular problems, 20% (in Pellet's study 55%) had subjective trigeminal problems, 8% (in Pellet's study 13%) had deglutition problems, none (16% in Pellet's study) had other eating problems, 91% (61% in Pellet's study) had no change in their life. Mean hospitalization stay was 3 days (for 23 days in Pellet's study). All the patients who worked, except one, had the same professional activity (66% in Pellet's study). Mean work cessation was 7 days (130 days in Pellet's study). (ABSTRACT TRUNCATED)