gms | German Medical Science

In most papers dealing with hearing preservation in cerebellopontine angle (CPA) surgery hearing level shortly after surgery is taken for outcome assessment. However there are some cases with delayed postoperative hearing loss described [Ref. 1], [Ref. 2], [Ref. 3], [Ref. 4]. On the other hand, in patients with preserved cochlear nerve and functioning cochlea improvement of hearing after surgery seems possible [Ref. 2], [Ref. 5], [Ref. 6]. Different types of pathology in CPA (vestibular schwannoma, other tumour, neurovascular conflict) can result in different damage mechanics of cochlear nerve, possibly influencing postoperative course of hearing disorders. We tried to analyse postoperative changes in auditory pathways using brainstem auditory evoked potentials (BAEP) and pure tone audiometry in groups with different pathology in CPA.

172 patients with pathology in CPA were operated in Neurosurgical Department of Medical University of Silesia in Katowice between February 1998 and November 2004. All patients were operated by the same team of surgeons via retrosigmoid approach. The patients had their neurophysiological functions, including BAEP, monitored using Nicolet Viking IV D unit with special intraoperative software. 104 of 172 monitored patients had at least some BAEP waves present at the beginning of operation. 51 of them showed no unfavorable BAEP changes at the end of surgery compared to initial recording. We have follow-up BAEP data in 39 patients of this group, and they are the subject of our analysis.

Patients were divided into 3 groups, based on the type of pathology in cerebellopontine angle: vestibular schwannoma (VS) – 12 patients; other tumours – 17 patients (mainly epidermoid cysts and meningiomas); neurovascular conflict (NVC) with trigeminal, facial or other cranial nerves – 10 patients. There were 23 women and 16 men in our group, aged from 12 to 69 years, with mean of 47.6 years. Follow-up BAEP examinations were performed using the same equipment as for intraoperative monitoring. Hearing deficit was assessed using pure tone audiometry and expressed as average of thresholds at 500, 1000, 2000 and 4000 Hz (PTA). Speech discrimination score was not available in some cases, so patients were classified according to modified Gardner-Robertson scale as used by Tonn et al. [Ref. 7], which is based only on PTA (PTA 0–30 dB is defined as class 1, 31–50 dB is class 2, 51–90 dB is class 3, 91 dB to maximum loss is class 4, and no response is class 5). BAEP recordings were graded according to our scale which is constructed with special attention to preservation or loss of conduction in cochlear nerve. The scale consists of : grade 0 – no reproducible BAEP waves, grade 1 – wave I visible, no other waves (preserved function of cochlea, no conduction through cochlear nerve), grade 2 – at least one of waves II, III and V is visible, showing preserved conduction in cochlear nerve, but some waves (I, III or V) are missing (marked disturbances), grade 3 – waves I, III and V are visible, with prolonged latency or inter-peak latency (minor disturbances), and grade 4 – normal BAEP. Mean time between surgery and last examination was 25.9 months, with range from 1 to 71 months.

Postoperative BAEP changes were seen in 49% (n=19) of examined group (Figure 1 [Fig. 1]). We noticed both improvement and deterioration, and even complete loss of BAEP responses. In group of patients with NVC changes in BAEP occurred in 3 cases (30%), but in 2 patients after temporary deterioration BAEP returned to the initial good level (grade 4), with worsening of responses from grade 4 to grade 2 in one remaining patient (10%). In groups of tumours, changes were more frequent, occurred in 67% of patients with VS and 47% of patients with other tumors. In 3 cases of VS direction of postoperative changes was not constant, resulting in fluctuations leading to improvement (1 case) or worsening of final response (2 cases). Mean PTA change expressed as dB difference between post- and pre-operative examination was 12.7 dB in VS group and 9.9 dB in group of other tumours, and the difference was not statistically significant. In 6 cases of 18 patients with available audiometry hearing level assessed according to modified Gardner-Robertson scale changed. In one patient, who had normal BAEP at the end of surgery, later tests revealed complete anacusis, and no BAEP waves could be recorded. One other patient had better hearing then before surgery, and 4 had worse.

Hearing assessment during follow-up after CPA surgery is mostly done by audiometry, and BAEP test are performed only in cases with some abnormalities noticed. We performed BAEP test routinely, independently of audiometric assessment. This showed relatively high incidence of postoperative changes (49% overall, 67% in VS). Neu et al. [Ref. 1] analysed hearing preservation and described only 1 patient (6%) of VS with stable intraoperative BAEP who suffered immediate postoperative hearing loss and another 1 case (6%) with delayed hearing loss. Moreover, we observed fluctuations in BAEP responses in 3 cases (25%) of VS and 2 cases (20%) of NVC. Grundy et al. [Ref. 8] performed BAEP tests in subgroup of analysed patients 20–36 weeks after surgery and noticed no improvement since recordings done 3–10 days after surgery. However, we believe that the possibility of improvement or transient deterioration should be considered when final outcome is to be discussed, because hearing level assessed shortly after surgery may change later-on [Ref. 1], [Ref. 2], [Ref. 3], [Ref. 6]. BAEP recovery as late as six years after operation was reported by Fahlbusch et al. [Ref. 9]

Comparing group of VS with other tumours we noticed that negative changes happened mostly in patients with vestibular schwannoma (33%, n=4),and only in 2 cases of other tumours (12%). Improvement of BAEP in postoperative period was found with similar frequency in groups of VS and other tumors (33% and 35% respectively). BAEP improvement was accompanied by better hearing level only in one case from group of non-VS tumours. This differs from ratios presented by Nakamura et al. [Ref. 5], who compared 7,9% of hearing level improvement after removal of CPA meningiomas to 1,4% of improvement after VS surgery. Analysing groups of tumours separately, we can see prevalence of BAEP improvement in group of other tumours (6 cases versus 2 cases of worsening), while in group of VS improvement and deterioration was equally frequent – 4 cases (33%) each.

Postoperative changes in audiometry were present in each group of patients, and similar cases were described by others [Ref. 1], [Ref. 2], [Ref. 3], [Ref. 4], [Ref. 5]. The changes included even complete hearing loss. Similarly, Neu et al. [Ref. 1], and Nakamura et al. [Ref. 5] noticed deafness despite no unfavorable BAEP changes at the end of surgery. These are important exceptions to generally accepted rule that stable BAEP predict hearing preservation [Ref. 1], [Ref. 8].

Change in PTA level was not dependent on BAEP change. In 2 cases worsening of hearing level could be explained by disturbances in conduction within auditory pathways seen in BAEP. But in 3 cases PTA worsened in spite of no noticeable changes in BAEP. At least in 2 of them deterioration was not caused by neural conduction disturbances, as BAEP remained normal in follow-up period. (In one remaining case BAEP were stable, but not normal, grade 2). In one case with preoperative deafness hearing returned to 61dB PTA while BAEP worsened. One must be aware of possibilities of hearing loss or preservation independently of BAEP as explained by Matthies and Samii [Ref. 10].

BAEP examination at the end of surgery cannot be treated as showing final state of auditory pathways. Postoperative changes can occur in every group of patients after surgery in CPA region. Only part of hearing impairment cases can be explained by sensorineural hearing loss assesed by BEAP.