Article
Electrophysiological predictors of hearing deterioration based on AEP-monitoring during posterior fossa approach for petroclival meningiomas
Search Medline for
Authors
Published: | June 9, 2017 |
---|
Outline
Text
Objective: To evaluate the risk of hearing deterioration as a function of amplitude and latency changes of the most stable components of the AEPs during surgery
Methods: We performed a retrospective study enrolling 40 consecutive patients harbouring petroclival meningiomas operated through retrosigmoid route at the University Clinic in Tübingen, Germany, from 2005 until 2011. clinical assessment was based on examination of the electronic medical files, image findings and electrophysiological data; tumour classification was reviewed through careful analysis of the preoperative MRIs. Radicality of the surgical resection was defined based on interpretation of post-operative MRIs. Neurological assessment immediate postoperatively, at the 5 th postoperative day, and late postoperatively, 3, 6 and 12 months after surgery, was documented. Amplitude and latency of the electrophysiological recordings were documented at many stages during surgery. For mathematical processing, we considered as initial values the waves before surgical incision and as final values those after suture of the dura-mater.
Results: In the 6 years time-window mentioned, we operated totally 78 cases of meningiomas in the cerebello-pontine angle. Among these, 40 cases of primary petroclival without previous treatment, excluding neurofibromatosis. The mean age was 59 ± 10 years, 31 female and only 9 male patients. As related to hearing, 12 patients out of 40 presented preoperatively with clinically detectable hearing impairment. Eight patients presented in the first assessment (prior to hospital discharge after surgery), deterioration of the hearing function; four of them reported subjective improvement. Among those 8 who reported initial deterioration, 4 recovered until the last assessment, and 4 remained stable. Next, we based on the constructed model to infere levels of wave parameters related to 10%, 20%, 50%, 80%, and 90% of hearing deterioration.
Conclusion: Our data indicate that latency measurements have high specificities (>92%, with advantage in favour of wave III), but relatively low sensitivities (71%). It means that latency is excelent in ruling out the chance of deficit (few false negatives, low type II error). Nevertheless, latency was more associated with false positives (false alarm, type I error). The amplitude of wave V, however, has a very high sensitivity (100%, no false positive), and low specificity (associated with some false negatives). Combining a very specific measurement like latency of the wave III, and a very sensitive measurement like amplitude of wave V seems to be the best strategy to predict precisely the chance of hearing deterioration in the immedate post-operative period.