An audiologist friend of mine frequently emails me asking, “Do you know the answer yet?” Sadly, I always reply no. The question she always asks? Should this implant patient continue with a hearing aid in the other ear or receive a second implant?
My friend likes to mention that I spent years studying this for my doctorate, so how can I still not know the answer? I think most clinicians continue to struggle with this. My current thinking is that it is almost always better to wear something on the ear contralateral to the implant, but it remains to be seen if that should be a hearing aid or implant. The difficulty we clinicians face is that we cannot try both options to see which works best for the patient.
A contralateral cochlear implant is a major procedure, and cannot be reversed. We know that cochlear implants often provide excellent results. We know that bilateral cochlear implants are an improvement over one implant. But we also know that beneficial complementary information can be obtained from an implant used with a contralateral hearing aid (bimodal stimulation).
Many studies have demonstrated the benefits of bilateral and bimodal stimulation, and it is difficult to synthesize all the research to reach a conclusion. The individual studies often have small numbers of subjects, and can provide conflicting results. This is where meta-analysis may help. Statistically analyzing and integrating the results of several independent studies can improve the reliability of any conclusions.
Now an article from the International Journal of Audiology sheds a little light on this subject with another meta-analysis from Schafer and colleagues at the University of North Texas. They previously published a meta-analytic comparison of the binaural benefits between bilateral implants and bimodal stimulation in 2007.
A Meta-Analysis to Compare Speech Recognition in Noise with Bilateral Implants and Bimodal Stimulation
Schafer EC, Amlani AM, et al Int J Audiol 2011;50:871
Schafer et al combined results in 2007 from 16 peer-reviewed publications related to speech-recognition in noise at fixed signal-to-noise ratios for bimodal or bilateral stimulation. They found that binaural stimulation (bimodal or bilateral) provided clear benefit over monaural stimulation. No significant differences between the bimodal and bilateral groups were found for any of the binaural phenomena. Performance in noise was equivalent between the two groups, so they recommended first attempting a bimodal arrangement, only proceeding to bilateral implantation if the contralateral hearing aid offered no benefit.
In the 2011 study, Schafer and her team aimed to improve on the previous meta-analysis in two ways. The new study tried to account for repeated measures designs; this means considering the advantage that the same subjects were used for different conditions in some studies. A repeated measures design reduces the effect of natural variation between subjects. The researchers also included fixed and adaptive speech recognition test methods.
The meta-analysis looked at three binaural benefits: squelch, summation, and head shadow effect. When speech and noise come from different places, the brain can compare the signals arriving at each ear and help separate the speech from the noise. This is squelch. Binaural summation is helpful even when the speech and noise come from the same direction; the brain processes the repetitive information received at each ear and hears sounds a little better. The head shadow effect simply takes advantage of having ears on each side of the head by allowing the brain to focus more on the ear nearest the sound we want to hear.
The study included 42 peer-reviewed bimodal or bilateral articles that they considered to be well-designed studies. Of course, the studies involved different speech stimuli and different noise stimuli, so it was not a simple job to combine all the results. Instead they used effect size, which provides a number to show how big the difference is between two treatments, where 0 signifies no difference, 0.2 is a small effect, 0.5 is medium, and 0.8 is large. Schafer's analysis considered squelch, summation, and head shadow effect separately, and looked at the effect size of bimodal or bilateral stimulation compared with just one cochlear implant. The analyses for squelch, summation, and head shadow effect were based on more than 300 subjects.
Overall, Schafer and her team found that bilateral cochlear implant users performed significantly better with two implants than with one on all three binaural-listening effects. The average bimodal listener, however, only showed the advantages of summation and head shadow effect, not squelch. The authors mentioned that some bimodal individuals did achieve substantial squelch, but it is difficult to treat an individual patient based on average data. Not surprisingly, the binaural phenomenon that showed the largest effect size was head shadow, not fancy binaural processing in the brain, but simply a result of being able to pick up sounds arriving on both sides of the head.
The study found that fixed noise testing gave slightly larger effect sizes for all three binaural phenomena than the adaptive noise method. They suggested that this may be because the adaptive test method is more challenging because the test occurs at threshold levels in noise. A fixed noise test generally occurs at suprathreshold levels.
What does this mean for us clinically? Can I definitively tell my friend that bilateral is best? Well, no. For one thing, this study just examines the three speech-in-noise binaural effects, but does not take into account any other measures, such as localisation, how patients felt about their hearing, and music perception. Of course, the real world is not a sound booth, and in real life squelch, summation, and head shadow work together in many situations. An individual patient may choose a hearing aid over a second implant because of surgical risk, reimbursement, or cosmesis.
Conversely, some patients with no residual hearing in the contralateral ear may have an easier decision in some respects. The authors remind us that many bimodal listeners do indeed show all three binaural phenomena; it is only on average that squelch is not shown in the bimodal listener. The conclusion of this paper is similar to the researchers' 2007 meta-analysis: a bimodal arrangement is an appropriate first choice treatment, and a second implant should be considered only if no measurable functional or perceptual bimodal benefit is seen.
• Read an abstract of the featured article at http://bit.ly/Schafer.
• Click and Connect! Access the links in The Hearing Journal by reading this issue on thehearingjournal.com.
• Comments about this article? Write to HJ at HJ@wolterskluwer.com.
• Follow us on Twitter at twitter.com/hearingjournal.