Many patient, device, and surgical factors contribute to patient success with auditory brainstem implants. This chapter examines device positioning, duration of deafness, choice of device, intelligence or motivation of patient, etiology of deafness, tumor size, age of patient, surgical technique, programming, and rehabilitation. Significant variability in techniques are discussed however the ideal approach to implantation is not fully understood.
16 Variability in Performance of Auditory Brainstem Implants
All auditory implants are subject to a multitude of variables that affect their performance outcomes. Understanding this variability allows for continuous adjustments and improvements to achieve optimal audiometric results. Cochlear implants (CIs), for example, are subject to some variability in positioning and from the types of electrodes used. However, in the absence of a malformation, the CI will generally occupy the same position due to the anatomic confines of the structure. Techniques such as “soft surgery” and hearing preservation, designed to prevent trauma in the cochlea with subsequent postoperative hydrops and performance variabilities, have been developed. Preservation of low frequency hearing is possible and patients may use a combination of electric and acoustic hearing (EAS) to take advantage of low frequency pitch cues along with the speech signal reproduced by the CI. In part due to the tonotopic layout of frequency information in the cochlea, the information provided by a CI will result in a signal more readily decipherable by the higher auditory system, when compared with an auditory brainstem implant (ABI).
The higher auditory pathways and their variabilities certainly affect CI performance. Auditory neuropathy, caliber variations in the VIIIth cranial nerve in certain types of malformations, and cochlear malformations may result in decreased information from the periphery entering the auditory system, but the higher auditory pathways, including the brainstem nuclei, ascending pathways, and ultimately the auditory cortex, have inherent variability and play a role in outcomes. Older patients, for example, often take longer to gain benefit from CI due to temporal processing issues in the auditory cortex.
In ABI patients, the information coming from the peripheral auditory system is not as tonotopically organized as information coming from the cochlea. This information is degraded when compared to stimuli from a CI. However, outcomes in adults and children implanted with ABI show a wide variability, from no benefit, to benefits with lipreading, to closed-set speech perception, to open-set speech perception with relatively normal development of the auditory system. 1 What factors enter into such variability? In this chapter, we examine multiple factors that play a role in the wide range of outcomes seen with ABI. 2 Known variables in the CI literature are explored as well as additional factors specific to the ABI. Device positioning, duration of deafness, choice of device, intelligence/motivation of patient, etiology of deafness, tumor size, age of patient, surgical technique, programming, and rehabilitation will be discussed in detail.
16.2 Potential Factors Involved in Variability
16.2.1 Surgical Factors
Device positioning is an important variable impacting ABI audiometric outcomes. Due to the anatomy of the lateral recess there is significantly more variability allowed in the positioning of devices when compared to CIs. Ideally, the device is to be positioned within the lateral recess of the foramen of Luschka with the paddle in contact with the cochlear nucleus. The surgical landmarks for identification of the foramen of Luschka include the origin of the lower cranial nerves and the choroid plexus, a tuft of which reliably projects into the cerebellopontine angle from within the lateral recess. However, these landmarks may be absent or difficult to identify in some cases (e.g., if the lower cranial nerves are involved with tumor). Large schwannomas may dilate the lateral recess such that the electrode may not maintain its ideal position within a patulous foramen of Luschka. In other cases, the lateral recess may be obscured by the presence of tumor involving the lower cranial nerves; these are typically not removed when removing a vestibular schwannoma due to fear of lower cranial nerve complications. Rarely, the lateral recess may be imperforate, or its entrance blocked by arachnoid septations.
The cochlear nucleus has a three-dimensional tonotopic organization that runs orthogonal to the surface. 3 Therefore, flat paddle electrode arrays are suboptimal for taking advantage of this anatomic orientation. Some authors have sought to optimize implants to better correlate with the anatomy of the region. Otto et al investigated the use of a penetrating implant to capitalize on the three-dimensional organization of the cochlear nucleus. In their study, they were not able to show improved audiometric outcomes with the penetrating electrode; however, they did report consistently lower thresholds in the penetrating ABI recipients. 4 Although some patients showed promising results, eventually penetrating electrodes were abandoned in favor of devices with more straightforward placement strategies due to the technical difficulty of implantation.
Other authors have investigated the variability that stems from device location and angle of placement. Barber et al reported on the existence of a surface “sweet spot” that resulted in lower thresholds and less nonaudiometric side effects by achieving an optimal angle of the device. 5 Additionally, they reported a wide range of angle variability among their study population with a weak correlation between angle and audiometric outcomes. See Fig. 16.1 for details on ABI placement and the associated angle placement variability.
Anatomic variability between patients contributes to the difficulty of electrode placement. Intraoperative testing can facilitate device placement. Matthies et al discussed the use of a quadripolar test electrode and evoked auditory brainstem response (EABR) mapping to precisely localize the cochlear nucleus or placement “sweetspot” prior to device implantation. 6 Mandalà et al found a significant improvement in audiometric outcomes when they used near-field compound action potentials (CAP) to position the device compared to EABRs (p=0.0051). 7 In their study of 18 patients they report 78.9% correct on open-set testing in the CAP group compared with 56.7% speech perception in the traditional EABR cohort. 7 These studies show that significant variability in ABI performance can be explained by device placement making it a critical factor for success.
Brainstem trauma from tumor removal or anatomic distortion may impact audiometric outcomes; however, there is limited date available on this topic. See Fig. 16.2 for an example of brainstem distortion noted on preoperative ABI scan. This hypothesis is supported by the fact that audiometric outcomes in nontumor patients generally are superior to patients with tumors. 1 Behr et al specifically investigated the impact of anatomic distortion by looking at the correlation between speech discrimination and tumor stage. 8 Their study focused on stage 3 (touching brainstem) and stage 4 (brainstem compression) vestibular schwannoma patients. Interestingly, their data found that there was no correlation between the tumor stage and word recognition, arguing against brainstem distortion as a significant factor impacting outcomes. 8 Goyal et al investigated the impact of anatomic variations in nontumor patients. 9 They studied cerebellar flocculus size in pediatric patients and corresponding audiometric outcomes. This study reported more difficulty with electrode placement in patients with high grade flocculus size. More data is needed to fully understand the impact of anatomic distortion.
It is also theorized that trauma from excitotoxicity or cautery injury may negatively impacton outcomes. Iseli et al conducted a study in gerbils where they compared cold-steel cutting of the cochlear nerve versus cauterizing and cutting the cochlear nerve. 10 They found no difference between the two approaches when the site was sufficiently far from the cochlear nucleus. However, more proximal bipolar cautery caused significant changes in the cochlear nucleus and likely impacts outcomes in ABI recipients. The House Clinic group investigated this parameter and reports improved outcomes with modified surgical techniques using conservative cautery, thus minimizing vascular damage. 11 Surgical technique and cautery use are important contributors to audiometric outcomes. Open-set speech outcomes are typically found at high volume centers only, underscoring the importance of surgeon’s experience and surgical volume. 12
Surgical positioning may potentially contribute to ABI outcomes. Patient positioning affects brain relaxation, bleeding, and the amount of cautery needed for hemostasis. Two main surgical positions are employed for ABI surgery. In Europe, a semi-sitting position is used whereas in the United States a supine position is more popular. When comparing the European and US literature, there is statically significant better results with the semi-sitting position (p=0.041). 8 However, there are several other surgical differences including the surgeon, the device used, and the impact on hydrostatic that contribute to the figures in this study. Many surgeons in the United States avoid a semi-sitting position due to the perceived risk of air embolism. There are no controlled studies to specifically compare this variable. 13 It is therefore difficult to determine if surgical positioning has any significant impact on audiometric outcomes.
16.2.2 Patient Factors
Duration of Deafness
The patient’s duration of deafness is another crucial variable impacting ABI performance. From the CI literature, we know this is an important factor for implantable audiometric outcomes. Multiple authors have demonstrated this axiom likely holds true with ABI users. Behr et al reviewed 26 patients with open-set scores greater than 30% and found an inverse relationship between duration of deafness and word recognition scores. Patients who were deafened for less than a year had better speech recognition outcomes than those who were deaf for longer period. Matthies et al demonstrated similar findings in a prospective study of 18 patients receiving ABIs. 14 They looked at the cumulative duration of deafness between both ears and found it was one of the strongest predictors of sentence scores among their patients. These findings suggest duration of deafness is an important factor and advise implantation of surgical candidates as early as possible.
When compared to CIs, ABIs require significantly more auditory rehabilitation and motivation to achieve optimal outcomes. Patients with multiple disabilities and other comorbid conditions have worse performance outcomes. 15
Noij et al conducted a systematic review of 162 nontumor children who underwent ABI placement. 16 They found nonauditory disabilities correlated with worse audiometric outcomes. Additionally, they demonstrated that audiometric outcomes correlated with duration of use. Patients continued to improve for about 24 months before reaching a plateau. Similarly, Otto et al reported initial disappointment at activation but significant improvement after a period of adaptation and learning. 17 Their study showed continued gains for up to 8 years with device use and some patients with initial poor performance were able to achieve open-set after a few years of use. 17
Although there is still a relatively small body of ABI literature available, thus far age of implantation has not been shown to be a significant factor in predicting audiometric outcomes. Several studies have examined this parameter but did not find an age trend that reaches statistical significance. 16 , 18 , 19 A systematic review by Noij et al of 162 children found age was not correlated with audiometric outcomes. 16 Jung et al reviewed nontumor ABI patients and found postlingually deafened adults had higher CAP scores than the pediatric population; however, this finding was not statistically significant. 19 In the current literature, age does not appear to play a significant role in audiometric outcomes.
Etiology of Deafness
The etiology of deafness significantly contributes to audiometric outcomes. Originally the only indication for an ABI was neurofibromatosis type 2 (NF2) and therefore this patient population comprises the majority of the literature. Increasingly indications for implantation have expanded and new performance outcomes in non-NF2 patients are now available. Overall, patients without cerebellopontine angle tumors have better outcomes than patients with tumors. 1 , 20 Additionally, within the tumor group, size of the tumor does not predict audiometric outcomes. 4 , 8 , 14 , 17 , 18 , 20
Indications for ABI now include trauma, altered cochlear patency, auditory neuropathy, and cochlear nerve aplasia. Some of the variability in patient outcomes may stem from disruption of central auditory pattern recognition. Colletti et al reviewed a large series of nontumor patients and found patients with more distal lesions had better audiometric outcomes than patients with proximal auditory pathway lesions. 1 Patients with neuropathy and other neurologic disorders were poor performers, while cochlear malformations were typically intermediate performers, and the best results were reported in patients with cochlear trauma or severely altered cochlear patency. Similar results of poor performance with absent cochlear nerves have been reported by other authors. 11 , 19 , 20 , 21 , 22 However, this is still an area of active research. Sennaroglu and others have reported on some rare patients with open-set speech understanding with an absent cochlear nerve. 21 , 22 They hypothesize there is a degree of cochlear nucleus development independent of the cochlear nerve but overall patients with cochlear nerves fibers tend to do better than those with complete nerve agenesis.