Use of intraoperative auditory evoked potentials to preserve hearing in unilateral acoustic neuroma removal

Electrophysiologic identification of the cochlear nerve fibers during cerebello-pontine angle surgery

To facilitate identification and preservation of the auditory nerve during cerebello-pontine angle surgery, bipolar recording of cochlear nerve compound action potentials (CNAPs) was performed. Two silver wires insulated with teflon up to the exposed ends were utilized as electrodes. They were twisted together, the distance between the two tips being 1 mm or less. Rarefaction polarity clicks (31/s) ranging from the psychoacoustical threshold to 120 dB pe SPL were used as stimuli. The investigation was performed in three groups of patients. The first group consisted of 9 patients submitted to vestibular neurectomy and 4 patients operated on by microvascular decompression of the eighth nerve. The second group comprised 8 patients with acoustic tumors smaller than 2 mm and serviceable hearing. Postoperative audiometric results in the subjects in the second group were compared with those obtained in well-matched homogeneous controls consisting of patients with acoustic neuroma operated on without the aid of CNAP recording. Bipolar recording from the eighth nerve was extremely selective, a good response being obtained only when positioning the electrode on the cochlear portion of the eighth nerve. During removal of the acoustic neuroma, repeated bipolar probing of the tumor and eighth nerve facilitated the task of distinguishing the cochlear nerve from other nervous structures and from the tumor, and contributed to preserving hearing in most patients.

Preservation of hearing in surgery for acoustic neuromas

Preservation of hearing was attempted in 161 cases of histologically confirmed acoustic neuroma removed by the senior author between January 1, 1970, and September 30, 1991. There were 136 patients with unilateral tumors; 22 patients had bilateral tumors (neurofibromatosis 2) and underwent a total of 25 procedures. Hearing was initially preserved in 35% of patients with unilateral tumors and in 44% of those with bilateral tumors. Results are reported in terms of pre- and postoperative pure tone average and speech discrimination scores. Surgical access to the tumor was obtained via middle cranial fossa and suboccipital approaches. The latter has been used more often over the past 5 years because of a lower associated incidence of transient facial paresis. Persistent postoperative headaches have been the most common complication following the suboccipital approach. The results of preoperative brain-stem auditory evoked response (BAER) studies were useful in predicting the outcome of hearing preservation attempts. Patients with intact BAER waveform morphology and normal or delayed latencies had a higher probability of hearing preservation in comparison to those with abnormal preoperative BAER morphology.

Preservation of cochlear nerve function in acoustic neurinoma surgery

A total of 55 cases with unilateral acoustic neurinoma which were operated on by the lateral suboccipital approach was studied to elucidate factors which influence postoperative hearing acuity. We analyzed several factors: preoperative hearing level, tumour size, tumour consistency (cystic or solid), and anatomical location of the cochlear nerve. The size of the tumours ranged from 1.2 to 5.8 cm in diameter. Thirty of 55 cases (55%) preoperatively had remaining cochlear function. The smaller the size of tumour, the higher was the preoperative hearing level excepting those tumours with a diameter of 5 cm or greater, which had relatively good hearing and often contained large cysts. As to the consistency of the tumours, 41 were solid and 14 were cystic, where 19 (46%) and 11 (79%) cases had had preoperative hearing, respectively. Anatomical continuity of the cochlear nerve was maintained at surgery in 15 of 30 cases with preoperatively remaining hearing; cochlear function was preserved after surgery in 9 of the 15 cases. It was located counter-clockwise (caudally) to the facial nerve at an angle of 50 degrees on average when they were projected on the right side. The distance or interrelation between the two nerves had no bearing on postoperative hearing preservation. Postoperatively, hearing acuity was improved in 6 cases (20%) with a mean value of 5.6 dB, unchanged in 3 (10%), and deteriorated in 21 (70%) among the 30 cases with remaining preoperative-hearing. When the tumour was less than 2 cm or cystic, better hearing preservation was expected.(ABSTRACT TRUNCATED AT 250 WORDS)

Posterior fossa craniotomy for lesions of the cerebellopontine angle. Technical note

A safe technique is described for performing a lateral posterior fossa craniotomy to gain access to the cerebellopontine angle. The method makes use of currently available high-speed air drills. Thus, it is possible to replace the removed bone at the conclusion of the procedure and to re-establish normal tissue planes while providing rigid protection to the posterior fossa.

Laser-Doppler measurements and electrocochleography during ischemia of the guinea pig cochlea: implications for hearing preservation in acoustic neuroma surgery

Interruption of cochlear blood flow has been implicated as one of the causes of the sensorineural hearing loss that may occur during acoustic neuroma surgery. With the guinea pig as an animal model for cerebellopontine angle surgery, laser-Doppler measurements were used to estimate the cochlear blood flow changes caused by compression of the eighth nerve complex. With compression, the laser-Doppler measurements decreased abruptly; somewhat later, the electrocochleographic potentials declined. When compression was released, laser-Doppler measurements usually returned immediately, followed later by return of the electrical potentials. Some of these potentials, including the compound action potential of the auditory nerve, often became transiently larger than their precompression values. Interposing bone between the laser-Doppler probe and the otic capsule, so that the total bone thickness approximated the thickness of the human otic capsule, decreased the laser-Doppler measurement, but changes caused by compression were still apparent. Thus, although the human otic capsule is much thicker than the guinea pig capsule, it may still be possible to make laser-Doppler estimates of human cochlear blood flow. Laser-Doppler monitoring during acoustic neuroma surgery may be beneficial, because it could give earlier warning of ischemia than is currently available from electrocochleographic monitoring, thereby enabling earlier corrective action. Electrocochleography complements laser-Doppler measurements by indicating the physiologic state of the cochlea.

Tumor control after stereotactic radiosurgery in neurofibromatosis patients with bilateral acoustic tumors

During a 4-year interval, 17 patients with bilateral acoustic tumors (vestibular schwannomas) underwent unilateral stereotactic radiosurgery using a multisource gamma unit; 2 patients underwent radiosurgery of both tumors in separate sessions. Eleven patients with unoperated contralateral tumors served as concurrent controls to compare the effects of radiosurgery with the natural history of acoustic tumors. After radiosurgery, the tumor control and regression rates were 89.5 and 21.1%, respectively (median neuroimaging follow-up, 1.4 years; range, 0.3-3.9). The tumor regression rate increased to 40% for patients evaluated at least 12 months after radiosurgery. In comparison to the unoperated contralateral tumors, stereotactic radiosurgery achieved tumor control, as assessed by the ultimate change in tumor size at follow-up (P, 0.012), the change in tumor size over time (P, 0.006), and tumor growth rates (P, 0.003). This study provided convincing evidence that tumor stabilization after radiosurgery (as assessed by neuroimaging) truly represented tumor control. The incidence of delayed facial neuropathy after radiosurgery compared favorably with the incidence reported after microsurgical removal. Some hearing was preserved in one-third of the patients who had preoperative hearing, including three patients who were contralaterally deaf. Stereotactic radiosurgery should be considered as a primary surgical modality for many patients with neurofibromatosis Type II.

Contralateral effects of cerebello-pontine angle exposure on human auditory brain-stem evoked potentials

Auditory brain-stem evoked potentials (ABEPs) were recorded during surgical procedures which exposed the cerebello-pontine angle (CPA) in humans. Recordings made with the CPA contralateral to stimulus exposed were compared with those obtained with the skin sutured at the end of surgery. Single-channel as well as 3-channel Lissajous' trajectory (3-CLT) analyses were used to evaluate the effect of the surgical exposure on ABEP. The results suggest that exposure of the CPA contralateral to the stimulated ear did not affect dipole equivalent orientation nor magnitude, but did affect timing of the recorded activity being more pronounced for segments 'd'-'e' (corresponding to waves IV-V) than for 'a'-'b' (waves I-II). The results imply that the effects of disrupting the volume conductor may have been overwhelmed by other effects, such as local temperature changes. These changes, although not associated with clinical sequella, should be accounted for when analyzing subtle quantitative changes involving surgical exposures.

Inconsistencies in the correlation between loss of brain stem auditory evoked response waves and postoperative deafness

This case underscores the difficulty of predicting postoperative hearing status from brain stem auditory evoked response (BAER) monitoring when wave I is preserved and all later waves are lost. During an operation involving the base of the skull, sudden and irreversible loss of all BAER waves beyond wave I occurred unilaterally. Wave I was preserved, with reduced amplitude and minimal latency shift. There was no permanent postoperative hearing sensitivity loss or speech discrimination loss.

Hearing preservation in acoustic neurinoma surgery

The authors have reviewed hearing results obtained in 99 patients operated on via the suboccipital approach for acoustic neurinoma, who were not deaf prior to surgery (pure tone average less than 70 dB). Tumor size was less than 10 mm in four cases, 10 to 19 mm in 26 cases, 20 to 29 mm in 39 cases, and 30 mm or greater in 30 cases. Removal was macroscopically complete in 92 cases and incomplete in seven, including four cases with bilateral acoustic neurofibromatosis. Hearing was preserved in 29 patients (29.3%), of whom 23 had neurinomas smaller than 30 mm and six had tumors exceeding 30 mm in size. Postoperative hearing was good in eight cases (four with neurinomas less than 20 mm and four with neurinomas greater than 20 mm), serviceable in four cases (three with neurinomas less than 20 mm and one with a tumor greater than 30 mm), and poor in 17 cases (eight with neurinomas less than 20 mm and nine with tumors greater than 20 mm). Fifty-seven patients underwent intraoperative brain-stem auditory evoked potential monitoring: the rate of hearing preservation was found to be higher in this group than in the 42 without monitoring (p less than 0.05). A statistical study using stepwise regression analysis showed that the two preoperative factors most significantly associated with postoperative hearing preservation are a good auditory level for low frequencies measured by pure tone audiometry and a small-sized tumor. Overall results indicate that, even if hearing is more easily preserved when the neurinoma is small and the preoperative auditory condition is good, the surgeon should try to save hearing in all patients who have preserved hearing before surgery.

Current results of the retrosigmoid approach to acoustic neurinoma

Since 1984 when cranial nerve monitoring became routinely performed at the Mayo Clinic, 255 patients have undergone 256 procedures using the retrosigmoid approach for the removal of acoustic neurinomas. Of these, 221 patients had some hearing before surgery and 52 maintained hearing following surgery. The anatomical continuity of the facial nerve was preserved in 237 of these 256 procedures. It was possible to perform a primary end-to-end anastomosis in seven of the remaining 19 patients, and one patient had a cable graft inserted. Thus, 95.7% of these patients were believed to have potential for spontaneous facial nerve function. Of the 11 patients in whom this was not possible, seven underwent early spinal accessory facial anastomosis, in two hypoglossal-facial anastomosis was performed, and two had no facial nerve procedures and have paralysis of the facial nerve. There were two deaths from a pulmonary embolus in the early postoperative period, both 4 days following otherwise uneventful surgery. The most common postoperative complication was cerebrospinal fluid leakage, which has not resulted in significant permanent morbidity although early repair for this problem is now routinely recommended. Other complications were quite rare and have generally not resulted in any major change in patient lifestyle or activity level. This review reconfirms that the retrosigmoid surgical treatment of acoustic tumors continues to be an acceptable treatment option.

Eighth cranial nerve dysfunction in hyperostosis cranialis interna

Hyperostosis cranialis interna is a recently described autosomal dominant bone disorder characterised by hyperostosis and osteosclerosis confined to the skull, especially the calvarium and the skull base. In the affected family members, we found variable simultaneous involvement of cranial nerves I, II, VII and VIII from late childhood onwards, most likely due to nerve entrapment. Auditory and vestibular functions were followed in 3 young family members for 8 years. At the first examination, pure tone audiograms were normal in all 3 cases and case 1 showed no caloric response in the right ear. During follow-up, this ear developed severe hearing loss progressing to deafness. The left ear showed transient sensorineural hearing loss and a temporarily diminished caloric response. Similar observations were made in case 2. Both cases showed abnormal brain stem auditory-evoked responses during and after the sudden hearing loss, in which initially only wave I was preserved and later on wave V returned with significantly prolonged I-V interval. The latter phenomenon was also observed in case 3 on both sides in the presence of normal audiograms during and after transient unilateral facial nerve paralysis, which was accompanied by bilateral diminished caloric responses.

The fate of tumour rests following removal of acoustic neuromas: an MRI Gd-DTPA study

The fate of capsular fragments left attached to vital structures at the time of otherwise total tumour removal was studied in 14 of 21 such patients who underwent acoustic neuroma surgery. Imaging using magnetic resonance Gd-DTPA at post-operative intervals of 6 months-12 years (mean 70 months) showed evidence of persistent tumour in half the patients. None of the patients had developed new symptoms and computed tomography had failed to demonstrate tumour recurrence. Persistence of the tumour was more likely if the residual fragments were not cauterized at the time of operation. Four of the seven persisting tumour rests showed evidence of gradual enlargement. The implications for patient management, particularly if an attempt is made to preserve hearing, are discussed.

The feasibility of using oto-acoustic emissions to monitor cochlear function during acoustic neuroma surgery

The feasibility of using evoked oto-acoustic emission (EOAE) measurement for intra-operative monitoring of cochlear function was assessed during removal of an acoustic neuroma in a 53-year-old woman with normal hearing on the operated side prior to surgery. The high level of noise in the operating theatre was the only material problem encountered and this was not sufficient to prevent recording of identifiable waveforms. During manipulation of the brainstem, damage to the cochlea was indicated by an increase in EOAE latency and its eventual disappearance. A total hearing loss in the operated ear was revealed after surgery. Monitoring cochlear function with EOAEs is probably best considered at present as an adjunct to auditory brainstem response monitoring of the composite cochlea and eighth nerve, thus providing differential information.

Intraoperative monitoring by transtympanic electrocochleography and brainstem electrical response audiometry in acoustic neuroma surgery

The preservation of hearing is a major aim of contemporary temporal bone surgery. Our present findings demonstrate that intraoperative monitoring is a key method for attaining serviceable postoperative hearing after the removal of an acoustic neuroma. Both electrocochleography (ECoG) and brainstem electrical response audiometry were performed in 96 patients operated on for acoustic neuromas. The specificity of the different monitoring methods was affected by surgical manipulations in addition to such non-specific influences as CSF drainage, core body temperature and anesthesia. In the present study ECoG was found to be more reliable in assessing the intra- and postoperative course with respect to the preservation of cochlear function.

Microvascular decompression for hemifacial spasm: analyses of operative findings and results in 310 patients

The operative findings and results of microvascular decompression (MVD) on 310 Chinese patients with hemifacial spasm are analyzed in this report. The operations were performed at the Neurological Institute of the Veterans General Hospital-Taipei between January 1983 and June 1990. The length of follow-up ranged from 6 months to 8 years (mean, 4.3 years); 273 patients (88%) had complete relief of spasm within 3 days after one MVD, and the remaining 37 patients (12%) showed no immediate postoperative improvement. Sixteen (5.2%) of these 37 initially unresponsive patients subsequently experienced complete relief, which occurred from 4 days to 22 months (median, 21 days) after one MVD; 13 others (4.2%) had complete relief immediately after the second MVD; another 3 (1%) had delayed complete relief 6, 9, and 11 months after the second MVD, respectively; and the remaining 5 (1.6%) only had delayed partial relief, which occurred 2 to 9 weeks after one MVD. Late recurrence occurred in three patients (1%). These immediate and long-term results lend support to the conclusion that the timing of reoperation can be postponed for a period of 3 to 4 weeks in the event of an initial failure to get improvement, and that a second MVD may be of value.

Technical improvements in the lateral suboccipital approach for neurovascular decompression of the trigeminal and facial nerves

Some refinements of the surgical procedure are described in 12 patients (five females, seven males) operated on for neurovascular decompression (ND) in the presence of trigeminal neuralgia and hemifacial spasm. The patients were operated upon in the supine position, in which the use of a lumbar drainage facilitates the dissection from the beginning. The lateral suboccipital approach was performed by means of a semi-osteoplastic craniotomy with the use of bone chips to close the bone defects so that a good cosmetic effect could be achieved.

Occurrence of vestibular and facial nerve injury following cerebellopontine angle operations

To elucidate how surgery in the cerebellopontine (CP) angle may cause vestibular and facial nerve injury, the 7th and 8th cranial nerves of dogs were manipulated as in human surgery along with monitoring of auditory evoked brain stem responses. Postoperatively, histological examinations were performed to investigate the effect of the surgical manipulations. We found that the occurrence of vestibular, facial and cochlear nerve injury was dependent on the direction of the excessive movement of the nerves in the cerebellopontine (CP) angle. Caudal-to-rostral shift of the nerve trunk most effectively avulsed the vestibular nerve. Haemorrhages were revealed between the vestibular ganglion and the fundus of the internal auditory canal. This caudal-to-rostral retraction could also damage the facial nerve in its intrapetrous labyrinthine portion. This was likely to be one of the pathophysiological mechanisms responsible for postoperative facial nerve palsy occasionally observed in human cases. Rostral-to-caudal retraction of the cerebellum damaged the cochlear nerve selectively. Although caudal-to-rostral retraction, instead of lateral-to-medial one, has been recommended to protect the cochlear nerve, this retraction was shown to be dangerous to the vestibular nerve if excessive. The clinical significance of the fragility of the vestibular nerve was discussed and the importance of preserving the vestibular nerve function is stressed.

Surgery for acoustic neurinoma. An analysis of 100 translabyrinthine operations

A consecutive series of 100 translabyrinthine operations for removal of acoustic neurinoma is reported. Complete tumor removal was achieved in 97% of cases. There were three perioperative deaths, and the postoperative morbidity rate was low. The facial nerve was preserved anatomically in 82% of patients. Preoperative facial weakness and tumor size larger than 2.5 cm were predictive of poor facial recovery. In patients with an intact nerve but complete facial palsy 1 week after surgery, electroneuronography was a good predictor of the final facial outcome and may influence the timing of rehabilitative procedures. With the exception of patients who have small tumors and good speech discrimination (where a suboccipital approach may preserve hearing acuity), the translabyrinthine operation is the procedure of choice for removal of acoustic neurinomas in patients who have normal hearing in the contralateral ear.

[Auditory evoked potentials in the monitoring of 15 laser microsurgeries of the brain stem]

Based upon the results observed in 15 patients submitted to tumor resection by the CO2 microlaser technique and submitted to brain stem auditive evoked potential (BAEP) intraoperative determination, the authors emphasize the importance of the technique. BAEP during the intraoperative period and the ultrasonic control of removal of tumors proved useful in the 15 cases of brain stem tumors or tumor adherent to it. Morbidity and mortality are described, and the important changes in BAEP that precede bradycardia are detailed. The importance of advances obtained through BAEP in peroperative control in such surgeries is stressed.

The small acoustic tumour; a chance to preserve hearing

The authors report their experience in the surgical management of small (less than 1.5 cm diameter) acoustic neuromas. Twenty-nine patients were treated. They had suffered symptoms of unilateral eighth nerve dysfunction on average for 37 months. High resolution CT with iv contrast was the primary investigation. In equivocal cases, air meatography was performed. All operations were performed by the senior author (LS). All tumours were totally removed, there were no deaths and all patients had an excellent result. There were 6 complications (CSF leak 3; meningitis 2; deep venous thrombosis 1). The facial nerve was preserved in 26 cases (and functioned adequately in all) and the cochlear nerve in 62%. In 4 cases with hearing pre-operatively, the decibel loss was the same or less post operatively. In 8 cases speech discrimination remained above 35%.

Neurophysiologic monitoring in posterior fossa surgery. II. BAEP-waves I and V and preservation of hearing

Of 135 cases operated upon for posterior fossa lesions 103 showed preoperative hearing. In 34 acoustic neurinomas 14 had postoperative initially preserved hearing, in 20 microvascular decompressions 19 had preserved postoperative hearing and in 49 other lesions 5 lost hearing. The relationship between preservation of hearing and the preservation or loss of brainstem auditory evoked potentials (BAEP) waves I and V in the three groups of namely: acoustic neurinomas, microvascular decompressions and other lesions are presented. It is noteworthy that only patients with preserved waves I or V are suitable candidates for intraoperative monitoring. The loss of wave V is usually associated with hearing loss (10 out of 13 cases). But hearing loss is also possible despite preservation of wave I (3 out of 60) or despite preservation of wave V (2 out of 68). The predictive value of the preservation of waves I and V is not an absolute one, but it strongly suggests preserved hearing postoperatively. The dilemma remains that once waves I or V are lost during surgery there is no certainty as to postoperative hearing. If wave V recovers after an initial loss, hearing is usually preserved but not in all cases. In wave I amplitude changes alone were more frequent than in wave V, where latency changes alone were more frequently observed. Particular surgical manoeuvres could be found to be often associated with a wave deterioration. Hearing preservation could never be achieved in patients who already preoperatively had no BAEP. It is concluded that BAEP monitoring is of great value in surgery for microvascular decompression and cerebello-pontine-angle (CPA) tumours with preserved hearing.

Do evoked potentials measure depth of anaesthesia?

Somatosensory evoked potentials have been suggested as a possible neurophysiological in of anaesthetic depth. In order to test the usefulness of this technique of monitoring, 8 patients were monitored during balanced anaesthesia. Following induction with a sleep dose of thiopentone, a neuromuscular blocking agent was used and the patients' lungs ventilated with N2O/O2(F1O2 = 0.3). This was supplemented with halothane or fentanyl. Somatosensory evoked potentials were recorded every 10 minutes during surgery, blood pressure and heart rate being recorded at the end of each evoked potential measurement. Induction of anaesthesia was associated with the reduction in SEP amplitude from 4.2 +/- 1.04 micro volts to 1.6 +/- 0.9 micro volts. There was an increase in latency from 19.2 +/- 1.3 msec to 20.0 +/- 1.3 msec to 20.0 +/- 1.5 msec. Recovery from anaesthesia was accompanied by an increase in SEP amplitude pre-induction values. There were no apparent correlations between the changes in evoked potential measures and changes in heart rate or blood pressure. Consistent change in somatosensory evoked potentials (prolonged latency, decreased amplitude) occurred with both narcotic and volatile anaesthetics. There was a recovery of evoked potentials at the end of anaesthesia. This is a promising technique, worthy of further evaluation.

Damage of the peripheral auditory system after operations in the cerebellopontine angle. A scanning electron-microscopic observation in dogs

Cochlear nerve injuries caused by manipulations in the cerebellopontine angle were electrophysiologically and morphologically investigated in dogs. Operative procedures similar to those performed in the cerebellopontine angle of humans were applied in dogs. Auditory-evoked potentials were recorded throughout the experiments. Postoperatively, the temporal bones were studied with a scanning electron microscope. The portions of the cochlear nerve fibers and branches of the internal auditory artery that exit from the bony cochlea into the internal auditory canal were susceptible to traction force derived from manipulations in the cerebellopontine angle. The cochlear nerve fibers from the basal turn of the cochlea were most easily pulled out from the fundus of the internal auditory canal. In some cases, massive hemorrhages and exudation of plasma were observed in the entire modiolus; these compressed the cochlear nerve trunk. The Schwann-glial junctions of the cochlear nerves were separated in some dogs, indicating this junction was one of the most vulnerable sites to operative manipulations in the cerebellopontine angle.

Effect of click-polarity on abnormality of intraoperatively monitored brainstem acoustic evoked potentials

The configuration of brainstem acoustic evoked potentials (BAEP) is influenced by the type of click stimuli used and may thus affect detectability of abnormalities. In a group of 19 patients with lesions in the posterior fossa BAEP were recorded pre- and intraoperatively. Repeat recordings were performed in each patient in two alternating series with rarefaction and condensation click stimuli. The findings demonstrated that intraoperative potential changes in latency and amplitude were different between the two stimulation modes, but did not vary significantly in their incidence. It was also not possible to predict from the preoperative BAEP which click polarity would demonstrate intraoperative changes more markedly, taking latency and amplitude as parameters. Two conclusions are drawn from this study: None of the two stimulation modes is superior in detecting intraoperative changes and therefore no recommendation can be made which click polarity to use. When working with only one click polarity it is recommended to use occasional control recordings with the other click polarity.

A modified extended middle cranial fossa approach for acoustic nerve tumors. Results of 125 operations

During the past 10 years, 125 operations for acoustic nerve tumors were performed on 114 patients at the authors' institution using a modified extended middle cranial fossa approach. This approach is based on a combination of King and Morrison's translabyrinthine-transtentorial approach and on the extended approach through the middle cranial fossa described by Bochenek and Kukwa. There were two hospital deaths (operative mortality 1.6%). In 102 operations on the initial tumor, total removal was performed in 89 cases (87%), and in 71 (80%) of these the facial nerve was anatomically preserved. Intracranial end-to-end anastomosis was performed on five of the 18 sacrificed facial nerves; a facial-hypoglossal anastomosis was carried out in the remaining 13 patients and in five (7%) of the 71 patients whose anatomically preserved facial nerve functioned poorly. In seven (39%) of the 18 patients in whom an attempt to preserve hearing was made, postoperative hearing was saved. In 23 operations on 17 patients for recurrent tumors, most of which had previously been removed subtotally via the suboccipital approach, total removal was accomplished in 13 (57% of the 23 reoperations and 76% of the 17 patients). At reoperation, the facial nerve was preserved in six (55%) of the 11 patients in whom the facial nerve had not been sacrificed. Postoperative leakage of cerebrospinal fluid occurred in 11 cases (8.8%), with rhinorrhea in 10 cases and otorrhea in one. Five of the fistulas were corrected by surgery and the rest healed spontaneously. Other complications were not significant.

Surgical management of acoustic neuromas during the last five years. Part I

A consecutive series of 94 patients with acoustic neuromas, who were treated in the years 1980-1984, was reviewed. Hearing loss was the most common presenting symptom. During 1980-1984, all acoustic neuromas were treated surgically; 93 cases had a total excision and only one case had a subtotal excision. Mortality and morbidity were low and patients who had a fair to poor outcome rating had large tumors. Overall, cranial nerve VII was preserved in 64.1% of cases, while cranial nerves VII and VIII were anatomically preserved in 27% of cases.

Effects of cerebellar retractions on the cochlear nerve: an experimental study on rhesus monkeys

Cerebellar retractions were performed in the cerebellopontine (CP) angle of rhesus monkeys to study the relationship between changes in brainstem auditory evoked potentials (BAEP) and/or compound action potentials (CAP) from the cochlear nerve, and morphological changes in the cochlear nerve in order to investigate the nature of cochlear nerve injuries caused by this surgical procedure. The changes in the BAEP and/or CAP from the cochlear nerve that were seen during cerebellar retraction were similar to those seen in human patients during operation. After completion of each experiment the temporal bones were prepared for histological examination. Hemorrhages often developed at the fundus of the internal auditory canal as a direct result of cerebellar retraction. These hemorrhages were located at the area cribrosa, where the branches of the internal auditory artery penetrate the cribriform plate. As the hemorrhages appeared to be due to avulsion rupture of these branches, they were considered to be the probable cause of the sudden loss of all the components of the auditory responses. The cochlear nerve fibers, especially those at the basal turn of the cochlea, avulsed simultaneously at the area cribrosa in most cases.

Surgical management of acoustic neuromas during the last five years. Part II: Results for facial and cochlear nerve function

Postoperative facial and cochlear nerve function in 83 consecutive patients with acoustic neuromas, who were undergoing their initial surgical procedure during 1980-1984, have been examined. The facial nerve was preserved in anatomic continuity in 71% of cases. Various nerve grafting procedures were used when the facial nerve was divided; the most common of these was a faciohypoglossal anastomosis, which was performed in 20 cases. The facial and cochlear nerves were anatomically preserved in 30.1% of all patients having their initial surgical procedure. Good speech discrimination was preserved in four patients, whereas more crude hearing was preserved in six other patients.

Electrophysiological monitoring during acoustic neuroma and other posterior fossa surgery

Techniques used to monitor the function of the seventh and eighth cranial nerves during acoustic neuroma and other posterior fossa surgery are reviewed. The auditory brainstem response (ABR), electrocochleogram (ECochG) and direct recording from the auditory nerve (CNAP) were compared. The best technique is the ECochG, although in many cases, the CNAP should be used as a back-up technique. The CNAP is especially useful for the identification of the auditory nerve. Both can provide real-time feedback on the physiological integrity of the auditory nerve. The ABR may be helpful in monitoring brainstem function. For some procedures, optimal monitoring requires the combined recording of all three techniques. Monopolar constant-voltage intracranial stimulation of the facial nerve is helpful for the identification and preservation of the facial nerve. Audio monitoring of spontaneous electromyographic activity provides real-time feedback on the effect of surgical manipulation of the nerve. Monitoring of ephaptic transmission in the facial nerve during microvascular decompression for hemifacial spasm aids in the identification of the offending vessel.

Cochlear nerve monitoring during cerebellopontine angle operations

The authors present their experience with intraoperative monitoring of cochlear nerve action potentials (AP) in 30 adult patients. Operative procedures were acoustic neuroma excision with attempted hearing preservation and selective vestibular neurectomy in patients with incapacitating Meniere's disease and serviceable hearing (SRT less than 50 db, discrimination greater than 60%). Loss of AP is detected rapidly and has been demonstrated after manipulation of the cochlear nerve and after coagulation of small arteries on the tumour capsule. Presence of an AP at the end of the procedure usually correlates with postoperative preservation of hearing. AP monitoring appears to be a reliable means of detecting potentially reversible changes in cochlear nerve function intraoperatively.

Hearing preservation in unilateral acoustic neuroma surgery

This paper reviews the principal English literature on hearing preservation in unilateral acoustic neuroma surgery. Seventeen case reports and 13 surgical series are included. In addition, we report ten cases of our own, two with successful hearing preservation. The purpose of this report is to study feasibility, success rate, and associated problems. Previous reports have been compared in terms of criteria that we have selected. A classification system similar to Silverstein's is used. The total number of cases under review is 621, with 221 reported successes. Cases limited to those having a unilateral acoustic neuroma, with valid supportive audiometry, were 394, with 131 successes. The approximate overall rate of success is 33%. There are five cases of hearing preservation with unilateral acoustic neuromas 3 cm or larger when supporting audiometric data are available, the largest being "4-5 centimeters." Problems included mixing of unilateral acoustic neuromas with other types of tumors and failure to include comprehensive data, particularly audiometry. We conclude 1) that hearing preservation is a reasonable goal in unilateral acoustic neuroma surgery, although the number of available candidates is relatively small and 2) that intelligent selection of patients and high quality surgical technique are the keys to success.

Surgical approaches to acoustic nerve tumours

A critical account is given of the orthodox neurosurgical approach to acoustic nerve tumours and alternatives through the petrous bone or tentorium.

Intraoperative monitoring of the electrocochleogram and the preservation of hearing during acoustic neuroma excision

We have monitored the electrocochleogram (ECochG) of 24 patients, using a transtympanic electrode, during acoustic neuroma excision. All patients had unilateral tumours with good preoperative hearing and complete excision was achieved in each case. Of the 24 patients, seven retained some hearing, however, a further two patients had normal ECochG waveforms at the end of operation but were nevertheless deaf. Thus, there is not an invariable correlation between immediate preservation of the ECochG and hearing. As expected, tumour size was important in hearing preservation. Five of seven patients with tumours less than 1.5 cm in diameter retained some hearing after operation, whereas 15 of 17 patients with tumours greater than 1.5 cm in diameter were deaf.

Preservation of hearing in surgical removal of acoustic neuromas of the internal auditory canal and cerebellar pontine angle

The surgical results in 69 patients with unilateral tumors of the cerebellopontine angle or internal auditory canal in whom total tumor removal was accomplished, and in whom an attempt was made to preserve hearing, are presented. The success rate of preservation of hearing and facial nerve function was correlated with the size of the tumor. Useful hearing, as defined by speech reception threshold no poorer than 70 dB and a discrimination score of at least 15%, was preserved in 73% of cases in which the tumor extension to the posterior fossa was no greater than 0.5 cm. In contrast, useful hearing was preserved in 22% of cases in which posterior fossa extension was greater than 2.5 cm. No significant correlation was found between preoperative evoked responses and success in preservation of hearing. The techniques and value of intraoperative monitoring of electrocochleogram (ECoG) and brain stem evoked responses are discussed. A theory of pathogenesis of intraoperative hearing loss, based on correlation of changes in evoked responses and simultaneous surgical events, is presented.

Transtragal, transtympanic electrode placement for intraoperative electrocochleographic monitoring

Transtragal, transtympanic electrode placement was performed for the purpose of intraoperative electrocochleographic (ECoG) monitoring during seven suboccipital acoustic neuroma resections. The promontory, the tragus, and two external sutures stabilized the electrode during surgery. The only noted otologic sequallae of electrode placement were small circular lesions at tympanic membrane puncture sites. Satisfactory ECoG recordings were obtained in five of seven cases. Peak-to-peak N1 amplitudes were (on average) 13.4-fold larger than the corresponding amplitudes of peak I of the surface-recorded brainstem auditory evoked potential (BAEP) recordings. During acoustic neuroma resection, ECoG and BAEP recordings changed relatively independently, which suggests multiple mechanisms and/or sites of injury to the cochlea or cochlear nerve. However, once they had become manifest, none of the observed changes exhibited a tendency to return to preoperative patterns. Hearing was preserved postoperatively in only two of seven patients, one of whom could not be monitored due to technical difficulty. Transtragal, transtympanic electrode placement provided a rapid, stable, and safe method of obtaining intraoperative ECoG recordings. Although combined intraoperative monitoring of ECoG and BAEP responses appeared to provide more precise documentation of injury to the cochlea and/or cochlear nerve, it was probably not influential in preservation of hearing in this series.

Intra-operative electrocochleography to monitor cochlear potentials during acoustic neuroma excision

Intra-operative electrocochleography (ECochG) was used in an attempt to monitor the action potential of the cochlear nerve during acoustic neuroma surgery in 14 patients with useful pre-operative hearing. Five patients had ECochG potentials preserved and yet only three could hear when tested audiometrically later. Of those losing the potentials intra-operatively all were subsequently deaf and the pattern of waveform loss allowed determination of the probable cause of hearing loss. Complete excision of the tumour was achieved in each case regardless of the ultimate effect on the AP as it was not felt justified to risk subsequent recurrence. In common with other studies one of the best prognostic factors for these preservation of hearing was found to be the size of the tumour.