Six-Month Evaluation of Spectrally Notched Hearing Aids in Tinnitus Treatment

Size: px

Start display at page:

Download "Six-Month Evaluation of Spectrally Notched Hearing Aids in Tinnitus Treatment"

Randolf Ferdinand Stone
5 years ago
Views:

1 Six-Month Evaluation of Spectrally Notched Hearing Aids in Tinnitus Treatment Dr. rer. nat. Lars Haab Akustika-Weiterbildungsseminar 2017 in Oberentfelden

19;107(3):1207-10, 2010 Stracke H, Okamoto H, Pantev C. Customized notched music training reduces tinnitus loudness. Commun Integr Biol.

2 Notched music approach (Prof. Pantev) Okamoto H, Stracke H, Stoll W, Pantev C. Listening to tailor-made notched music reduces tinnitus loudness and tinnitusrelated auditory cortex activity. Proc Natl Acad Sci USA. 19;107(3): , 2010 Stracke H, Okamoto H, Pantev C. Customized notched music training reduces tinnitus loudness. Commun Integr Biol. May;3(3):274-7, 2010 Teismann H, Okamoto H, Pantev C. Short and intense tailor-made notched music training against tinnitus: the tinnitus frequency matters. PLoS One. 6(9), ) 2) Matched tinnitus frequency Corresponding frequency notch 3)

Competitive advantage Progressive centralization Thalamocortical

3 Tinnitus as two-pronged process Tinnitus Genesis Tinnitus Chronification Homeostatic plasticity Hyperactivity Spike timing Competitive advantage Progressive centralization Thalamocortical dysrhythmia Hebbian plasticity Learned focal point of attention Limbic system involvement

Tinnitus and hearing loss Tinnitus is associated

these patients might have a hidden hearing loss,

4 Tinnitus and hearing loss Tinnitus is associated with hearing loss in the range of 85-95% of the cases. e.g., Norena et al., 2002 Yet around 10% of tinnitus patients do have a normal audiogram. Studies provided physiological evidence that these patients might have a hidden hearing loss, i.e. cochlear damage that is not detected by hearing threshold measurements. Schaette, McAlpine 2011

Surface of a rat cochlea (SEM) Normal cochlea Narrow-band dead

depression Dead regions cannot be reliably diagnosed from the

Moore, 2001 Neural response magnitude is renormalized within

5 Surface of a rat cochlea (SEM) Normal cochlea Narrow-band dead regions Severe noise trauma Confinded regions of sensory depression Dead regions cannot be reliably diagnosed from the audiogram. B.C.J. Moore, 2001 Neural response magnitude is renormalized within subsequent processing stages. Schaette, McAlpine 2011 e.g., B.C.J. Moore, 2001 Mechanism to compensate reduced input

6 The brain works differently, than you might think... John Searle (*1932)

Homeostatic plasticity homeostatic plasticity = "staying the same

Deviations from the target level engage homeostatic mechanisms

7 Homeostatic plasticity homeostatic plasticity = "staying the same through change" Target rate Hypoactivity STRESS Target rate Turrigiano et al., 2004 Synaptic drive increases due to sensory deprivation Deviations from the target level engage homeostatic mechanisms (arrows) that reduce or enhance the strength of synaptic inputs, thereby moving the neuronal activity back into the target zone. Hyperactivity HP

8 Numerical Model of Local Hyperactivity Digital Filter Cochlear Activation Allows for integration of frequency-specific attenuation and amplification effects due to hearing deficits and hearing aid technical properties Time-frequency representation of an arbitrary acoustic stimulus Auditory Nerve Model R. Schaette and R. Kempter, Dorsal Cochlear Nucleus Model R. MacGregor, 1987 X. Zheng and H. Voight, Inferior Colliculus Lateral Inhibition Model E. Izhikevich, O. Hoshino, K. Kuroiwa, 2002

9 Lateral Inhibition (in the DCN) [ ] Ball shaped synapses are excitatory, cone shaped synapses are inhibitory. Pearson Education, 2009

10 Neural analogues of intensity & contrast Intensity Intensity bright Loudest band: 139 db SPL loud dark Whisper: 30-40dB soft

11 Tinnitus Anschaulich dargestellt Unbalanced contrasts unbalanced contrast

12 Contrast restoration restored contrast

13 Effects on firing regularity: Simulation Y-Axis: ISI coefficient of variation (VC) for two simulations. X-Axis: Serial population of 800 neurons, representing one narrowbandwidth channel. The neurons have increased stability (reduced VC) due to increased spontaneous activity (pathologic). The grey line depicts VC for the untreated tinnitus Black line depicts VC during(!) notched sound stimulation. NOTE: The pathologic firing regularity is reduced during notched stimulation and the gradient between hyperactive and neighboring regions is smaller.

14 Measuring Tinnitus Chronification and Decompensation

15 Measures of attention & habituation Aversive binding Habituation

Strauss, 2014 Degree of tinnitus compensation is proportional to (long term) habituation (long term) habituation :=

16 High sensitive habituation analysis Circular von Mises analysis of Hardy space projected ERPs (theta-alpha border) Z. Mortezapouraghdam, L. Haab, F. I. Corona-Strauss, G. Steidl, D. J. Strauss, 2014 Degree of tinnitus compensation is proportional to (long term) habituation (long term) habituation := decrease of von Mises concentration in the instantaneous phase of bandlimited ERP components. objective treatment evaluation : increase in (long term) habituation

Experimental study Commercially available digital behind the ear hearing aids were modified by the producer in order to implement a notch filter.

17 Experimental study Commercially available digital behind the ear hearing aids were modified by the producer in order to implement a notch filter. The filter abruptly blocks out environmental sounds to suppress neural hyperactivity associated with tinnitus by lateral inhibition This notch filter is adjusted to the individual tinnitus percept and induces sharp spectral edges to the environmental sound spectra which might trigger lateral inhibition mechanisms Pure tone tinnitus: corresponding tonal frequency

18 Experimental study: Inclusion criteria* Patient inclusion criteria subjective chronic tinnitus (> 6 months) continuous distress (no temporal interruptions) tonal tinnitus (below 8kHz) no history of neurological diseases such as Alzheimer s disease, multiple sclerosis, Meniere syndrome, acusticusneurinom or psychopathologies such as attention deficit disorder and depression signed consent form *As approved by the local ethics committee (Ärztekammer des Saarlandes)

19 Experimental study: Subjects & Methods* A total of 34 patients (12 f., 24m. / mean age: 56.55y ± 10.13y / 19 with NEST, 15 controls) with tonal tinnitus from a hearing rehabilitation and tinnitus center entered the study. All these patients entered a well evaluated and established compact group based tinnitus retraining therapy concept which lasted for three weeks. The patients characterized their tinnitus distress before and after the therapy using the widely recognized TQ52 (Goebel/Hiller) questionnaire and the circular (long term) habituation analysis. Control group: mpta (tinnitus)=22.58±11.22, Tinnitus Freq.=5.13±3.24kHz, Post-therapy TF=4.89±2.96kHz mean age= 57±10.24y; 10m/5f Notch group: mpta (tinnitus)=25.26±14.83, Tinnitus Freq.=4.29±1.92kHz, Post-therapy TF=3.52±2.84kHz mean age= 56.21±10.31y; 14m/5f *As approved by the local ethics committee (Ärztekammer des Saarlandes)

20 18 60 16 14 12 10 8 6 4 control notch 0 M 3 M 6 M Duration [Months] 50 40 30 20 10 control

20 Scores Experimental study: Subjective Results Scores Control Notched environmental sounds 22 TQ12, notch: 19 subjects, control: 15 subjects 70 TQ52, notch: 19 subjects, control: 15 subjects control notch 0 M 3 M 6 M Duration [Months] control notch 0 M 3 M 6 M Duration [Months] Cohen s d effect size: 0.94 ± 0.15 (~0.8 Indicator for strong effects)

21 Experimental study: Statistical significance TQ52 score improvements as boxplots Notched hearing aids (orange) and control group (blue)

Experimental study: Objective results Exemplary subject [0 mo] [3 mo] [6 mo] Long-term habituation factor : First order polynomial fit over the phase stability of 25 consecutive segments of 24

22 Experimental study: Objective results Exemplary subject [0 mo] [3 mo] [6 mo] Long-term habituation factor : First order polynomial fit over the phase stability of 25 consecutive segments of 24 singe-sweeps each = habituation curve is inverse proportional to the slope of this habituation curve Lehser et al 2015 Large variances can be minimized when patients are grouped according to their tinnitus frequency. ALR distortion due to changes in stimulus frequency Lehser et al 2015

Experimental study: Effect size Cohen s d effect size: 3 months: 0.416 ± 0.15 (<0.

23 Experimental study: Effect size Cohen s d effect size: 3 months: ± 0.15 (<0.5 Indicator for weak effect) 6 months: 1.07 ± 0.13 (>0.8 Indicator for strong effect)

24 Summary I Scores Summary: In-silico the DCN lateral inhibition network is already sufficient to generate a narrow, well-defined band of hyperactive and precise principal neurons with a pronounced contrast to neighboring bands. Notched acoustic stimuli effective in normalizing neural activity and spike timing precision The effect confers to cortical potentials Plastic effect in long-term habituation performance during therapy Prolonged therapeutic effect greater sustainability Great patient acceptance TQ52, notch: 19 subjects, control: 15 subjects control notch 0 M 3 M 6 M Duration [Months]

25 Hearing loss and consequences Reduced sound perception Stress Increased sound sensitivity (Hyperacusis) Tinnitus Social integration

26 The Cocktail-Party phenomenon

27 Source separation Male voice Spatial information? CONTRAST Low pitch, small dynamics Cocktail party Spatial information Female voice Higher pitch, higher dynamics

Listening Effort sustained attention Studies on Temporal lobe hyperactivity demonstrate, that tinnitus-patients require increased effort to separete their tinnitus tone from their acoustic

28 Listening Effort sustained attention Studies on Temporal lobe hyperactivity demonstrate, that tinnitus-patients require increased effort to separete their tinnitus tone from their acoustic environment. Semantic completion: Missing sentence fragments must be completed by memory processes. This requires the use of ressource-limited processes, e.g., working memory and attention. The listener tries to increase the quota of correctly understood sentences by focused attention.

29 Always full speed ahead?

The PFC creates a model of a future action and predicts costs and

30 Action planing The prefrontal cortex (PFC): Planing and motivation Responsible for the conscious processing of reward and punishment. The PFC creates a model of a future action and predicts costs and benefits (reward prediction). Dopaminic neurons compare prediction and actual outcome (left)

31 Cost and benefit Serotonione Dopamin Cost / benefit balance action costs Reward for action taken

32 Motivation In the case of dopamine and serotonin imbalance, general motivation is reduced (animal studies). Reduced reward seeking Reduced drive for social interaction Reduction of appetitive learning (reward learning) Reduced use of coping strategies

33 Fatigue Fatigue occurs if the balance between invested effort and expected reward is disturbed Reduced Motivation to invest listening effort. Reversible: Rest Longterm cause of listening exhaustion

Fear of the unknown Tinnitus patients suffer from the fear, that their hearing loss increases, that their tinnitus tone will aggravate and is caused by a serious illness.

34 Fear of the unknown Tinnitus patients suffer from the fear, that their hearing loss increases, that their tinnitus tone will aggravate and is caused by a serious illness. In most cases these fears are ill-founded and based on a lack of substantiated Information. Those fears activate the amygdala in the limbic system. The amygdala in turn reduces the Dopaminrealease in the nucleus accumbens. Results often in fear avoidance behaviour.

35 Social withdrawal Fear avoidance Wir müssen den Menschen zuhören, sonst ist Schweigen das letzte was wir von ihnen hören.

36 Stress Social limitations, confusion, sleep disturbances, lack of concentration high blood pressure, emotional and bodily tension

Depression The sum of fears, stress and disturbances in the dopaminergic systems can result in a depressive episode. Depression is the most common comorbidity of tinnitus.

37 Depression The sum of fears, stress and disturbances in the dopaminergic systems can result in a depressive episode. Depression is the most common comorbidity of tinnitus. Both syndromes can reinforce each other. A tinnitus linked depression can be addressed by: drug treatment psychotherapeutic therapies van Gogh At Eternity's Gate Need of a holistic approach for the simultaneous treatment of the hearing loss, the tinnitus percept and psychological comorbidities.

Summary II Summary: Tinnitus and uncompensated hearing deficits lead to increased listening effort Increased listening effort in conjunction with fear, stress and social

38 Summary II Summary: Tinnitus and uncompensated hearing deficits lead to increased listening effort Increased listening effort in conjunction with fear, stress and social withdrawal can cause lasting imbalances in neurotransmitters of the reward system. Lack of motivation and reward: Downward-spiral of cognitive fatigue, exhaustion and depression

40 Addendum: Tricking the brain

Clinically proven: Spectral notching of amplification as a treatment for tinnitus

Clinically proven: Spectral notching of amplification as a treatment for tinnitus Jennifer Gehlen, AuD Sr. Clinical Education Specialist Signia GmbH 2016/RESTRICTED USE Signia GmbH is a trademark licensee