Long Cochlear Implant Electrodes: Incomplete insertion, loss of residual hearing and balance

By Thomas J. Balkany, MD



Preservation of residual hearing has been a goal of CI surgery since it was first reported in 1989.1 It is an indication of good surgical technique and may result in enhanced speech perception in conventional2 and hybrid devices.3 Very-long electrodes (> 28 mm) have been used successfully for over two decades, but continue to be debated due to anecdotal inferences of loss of residual hearing and vestibular function, as well as a high incidence of insertion failure.

New Information

A recent study by Nordfalk et al4 of Oslo University provides new data to address these issues.

Loss of Residual Hearing

  • In this study, 35 adult subjects received very-long electrodes (31.5 mm and 28 mm.) In addition, four received more typical, 24 mm electrodes of similar design.
  • A statistically significant relationship between loss of residual hearing and very-long electrodes occurred at 250 Hz (p < .05) but not for the low frequency pure tone average (lf-PTA.) A possible ceiling effect is difficult to avoid due to greater pre-op hearing loss at 500 and 1000 Hz.

Loss of Vestibular Function

Vestibular Symptoms


24 mm electrode



28 mm electrode



31.5 mm electrode



The table suggests that very-long electrodes are associated with subjective and objective indications of vestibular system damage proportional to their length. No vestibular symptoms or VEMP abnormalities were seen with the standard length electrodes

Failure to completely insert long electrodes-

Nearly 1/5 (18%) of very-long electrodes could not be fully inserted by this group of highly respected surgeons. These findings confirm anecdotal reports and clinical experience of a high rate of full-insertion failure with very-long electrodes. A typical insertion failure rate is in the range of 3 to 4%.5 The goal of surgery is full insertion in every case.

Take Home

Nordfalk and colleagues at the University of Oslo have shown that very-long CI electrodes (> 28 mm) are associated with a higher incidence of cochlear and vestibular dysfunction when compared to more standard length electrodes (24 mm) of similar design. The authors also demonstrated that failure to completely insert this type of electrode occurred in 18% of recipients, even in highly competent and experienced hands. These data appear to confirm previous anecdotal clinical experience regarding negative aspects of very-long electrodes, but should not become canon unless replicable.


    1. Boggess WJ, Baker JE, Balkany TJ. Loss of residual hearing after cochlear implantation. Laryngoscope. 1989; 99:1002-5

    2. Sheffield SW, Jahn K, Gifford RH. Preserved acoustic hearing in cochlear implantation improves speech perception. J Am Acad Audiol. 2015 Feb;26 (2):145-54.

    3. Roland JT Jr, Gantz BJ, Waltzman SB, et al. United States multicenter clinical trial of the cochlear nucleus hybrid implant system. Laryngoscope. 2015 [Epub ahead of print]

    4. Nordfalk K F, Rasmussen KH, Bunne, M et al. Insertion Depth in Cochlear Implantation and Outcome in Residual Hearing and Vestibular Function. Ear and Hear 2015. Epub ahead of Print.

    5. Brito R, Alves T…Bento RF. Surgical complications in 550 consecutive cochlear implantations. Braz. J. Otorhinolaryngol. 78; 3: May/June 2012.

Leave a Comment

Iain Butler 2 years ago

Ceiling effects of current post-implant tests make it difficult to assess the quality of recipients listening experience. Standard speech perception outcomes are similar. A question I have is whether recipients experience the same sound 'quality' with a 24mm electrode array as with a 28mm array. The other important factor relates to long term hearing preservation after EAS surgery which Erixon & Raske-Anderson recently highlighted in Ear & Hearing. They found that unaided thresholds dropped by 30dB over 3 years in the operated ear. Should we implant a shorter array in order to preserve low frequency hearing for a couple of years to the detriment of sound quality? It is gratifying to see that both companies whose electrodes I implant seem to be moving slowly towards a happy mean regarding length and electrode design - maybe around 26mm!

Prof. Dr. Rolf-Dieter Battmer 2 years ago

Comment By: Prof. Dr. Rolf-Dieter Battmer, Director, Center for Clinical Technology Research, Unfallkrankenhaus Berlin----- Clinical results with very-long electrode arrays consistently show that speech perception results and patient acceptance were better when the most apical electrode contacts are not activated [10, 12, 15, 17]. In addition, stimulation with apical electrodes has been associated with poor pitch discrimination and even pitch reversals [5, 10, 12, 17]. Temporal bone insertion studies of very-long electrodes similarly demonstrate an increased risk of trauma due to buckling of deep inserted electrode arrays. Trauma includes dislocation and buckling of the electrode tip leading to ruptures of the basilar membrane and fracture of the spiral lamina [1, 2, 13, 17]. Given the lack of benefit and an increased risk of insertion trauma, the use of very-long electrodes has to be put into question; especially in the light of recent developments in atraumatic electrode insertion and hearing conservation. REFERENCES 1)Finley, Skinner, et al, “Role of Electrode Placement as a Contributor to Variability in Cochlear Implant Outcomes”, O&N 2008. 2)Gani, et al, “Implications of Deep Electrode Insertion on Cochlear Implant Fitting”, JARO 8: 69-83 2007. 3)Kos, et al, “Partial withdrawal of deeply inserted cochlear electrodes: observations of two patients”, EurArchORL, 2007. 4)Pesch J, Büchner A, Böhm M, Lenarz T, “Does More Apical Stimulation Provide Increased CI Benefits?“, CIAP, Lake Tahoe 2007 5)Baumann, et al, “The cochlear implant electrode-pitch function”, Hearing Research 213[1-2], 34- 42 2006. 6)Adunka O, Kiefer J, Unkelbach MH, Radeloff A, Gstöttner W: Evaluating cochlear implant trauma to the scala vestibule. Clin Otolaryngol 30(2): 121-7 2005. 7)Adunka, Kiefer, “Impact of electrode insertion depth on intracochlear trauma”, Otolaryngology – Head and Neck Surgery 135, 374-382 2006 . 8)Gstoettner, et al, “Cochlear implant deep electrode insertion: extent of insertional trauma”, Acta Otolaryngol (Stockh) 117: 274-7 1997.

Dan Schwartz 2 years ago

The research should be framed not in terms of electrode length, but as angular depth of insertion, as a 25mm perimodiolar placement will be almost as deep as a 31mm lateral placement. As for electrode length for hybrid (EAS) patients, I believe it's better to use a full electrode and shut off channels as needed, rather than a short one, as if residual hearing is indeed lost either through insertion trauma, debris from the cochleostomy in the perilymph, cytokines, & such, the patient still has a full electrode to fall back on as an insurance policy. Of course, EAS should only be used for patients who have a stable hearing loss: Please see The Curious Hybrid (EAS) Cochlear Implant Recipient Dan Schwartz, Editor, The Hearing Blog


Long electrodes can result in damage to basilar membrane and spiral lamina in the apical region as demonstrated in temporal bone studies. I have a question- Can this cause problems during Reimplantation, as damage to spiral lamina can lead to increased fibrosis?

Thomas J. Balkany, MD 2 years ago

Thank you for your comment. Any electrode re-insertion may cause additional damage to neural elements and it seems logical that extraordinary fibrosis or ossification caused by the initial insertion may result in additional damage. However, I am not aware of data to support these notions. Reinsertion into the original mesothelial electrode envelope within the scala is a good way to perform revisions and may minimize additional damage.

Karl F Nordfalk 2 years ago

I am the first author of the paper referred to in the blog post. I appreciate your analysis of our study and your new overview of the vestibular losses. It is always beneficial when scientific data is re-analysed and evaluated with new eyes, but I feel I have to add some comments: 1) About the vestibular losses: The differences on vestibular loss and vertigo symptoms between the three electrode array types are not statistically significant. This means that we cannot rule out that they are just coincidental. 2) It is stated in the conclusion that electrodes (> 28 mm) are associated with a higher incidence of cochlear and vestibular dysfunction when compared to more standard length electrodes (24 mm) of similar design. - Statistically we were only able to prove that the insertion depth and actually not the electrode type correlated with loss of residual hearing at 250 Hz. The cochlear insertion depth varies according to both electrode type, but also strongly with cochlear size. 3) Lastly, but most important, I must emphasize that the high ratio of non-complete insertions were only seen with the +FlexSoft (31, 5 mm) electrode array and not the +Flex28 (28 mm). The +Flex28 (28 mm) electrode array actually comes out very positive in our study since 0% had incomplete insertion and loss of residual hearing was not higher compared to the shorter +Flex24 electrode array. Best regards Karl F Nordfalk, First author

Anandhan Dhanasingh 2 years ago

Seems that the information given in this blog is too exaggerated. In-fact if one reads the article by Nordfalk and colleagues, it will be clear that the info given in the article does support deep insertion with flexible electrodes. One of the important thing to note in Nordfalk article is that in none of the cases, the electrode deviated from scala-tympani to scala vestibuli. Such deviation is considered the most traumatic situation which usually happens with Contour Advance electrode, which is a pre-shaped in its design. In fact literature shows that 35-40% of patients who were implanted with the pre-shaped modiolar hugging electrode had the electrode deviation from ST to SV which is an irrepairable trauma even in the hands of experienced surgeons. Coming back to this blog, it is well known that vertigo is one of the several acute side effects of cochlear implantation which will be vanished in a month time. Moreover the method by which those vestibular evoked myogenic potential is measure is so sensitive that the chances of measuring the error is much more. There are too many problems associated with the pre-shaped electrodes even if it is made thin. May be the comment section is not enough and not the correct way to mention it. Anand