CROS, BiCROS & Severe-to-Profound Fitting Strategies

CROS, BiCROS & Severe-to-Profound Fitting Strategies

Asymmetric sensorineural hearing loss, Single-Sided Deafness (SSD), and severe-to-profound cochlear impairment present unique rehabilitative challenges. Standard bilateral hearing aid fittings are ineffective when one ear cannot benefit from direct amplification due to poor speech discrimination, or when cochlear dead regions render amplification counterproductive.

---

Single-Sided Deafness (SSD) & Psychoacoustic Deficits

Single-Sided Deafness (SSD) refers to a clinical presentation where one ear exhibits profound sensorineural hearing loss or an unaidable word recognition score (WRS $< 20-30\%$), while the contralateral ear has normal or aidable hearing. SSD causes three profound auditory processing deficits:

1. The Head Shadow Effect

The human head acts as an acoustic barrier for sound waves whose wavelengths are shorter than the diameter of the head ($> 1.5-2.0\,\text{kHz}$). Sounds arriving from the side of the deafened ear are physically blocked by the skull. While low-frequency sounds ($< 1\,\text{kHz}$) bend around the head via diffraction, high-frequency speech sounds (critical for consonant clarity) are attenuated by $6$ to $12\,\text{dB}$ before reaching the hearing ear. In reverberant, noisy environments where speech originates from the impaired side, speech understanding is severely compromised.

2. Loss of Binaural Squelch & Binaural Summation

• Binaural Squelch: When listening with two functional ears, the central auditory nervous system compares the disparate phase and amplitude cues arriving at both ears, effectively suppressing background noise and improving the internal signal-to-noise ratio by $+2$ to $3\,\text{dB}$. SSD abolishes binaural squelch.
• Binaural Summation: Central integration of acoustic energy from two ears improves pure-tone thresholds by approximately $3\,\text{dB}$ and suprathreshold loudness perception by $5-6\,\text{dB}$. In SSD, the patient must rely entirely on monaural acoustic input.

3. Loss of Horizontal Sound Localization

Horizontal plane sound localization requires comparison of acoustic arrival differences between the two ears:

• Interaural Time Differences (ITD): Low frequencies ($< 1.5\,\text{kHz}$) provide sub-millisecond arrival time cues processed by the medial superior olive (MSO).
• Interaural Level Differences (ILD): High frequencies ($> 1.5\,\text{kHz}$) create intensity differences processed by the lateral superior olive (LSO). In SSD, all sound localization in the horizontal plane is lost. Patients cannot determine the direction of approaching vehicles, speakers in group settings, or warning sirens.

---

CROS vs. BiCROS Systems

Contralateral routing technology overcomes the head shadow effect by capturing sound on the unaidable side and wirelessly streaming it to the better hearing ear.

[UN-AIDABLE EAR]                             [BETTER HEARING EAR]
Transmitter Mic                               Receiver / Hearing Aid
       |                                                |
       +=====( Wireless 2.4 GHz / NFMI Stream )========>|
                                                        v
                                         [CROS]: Open coupling, 0 dB Gain
                                         [BiCROS]: Amplifies local + streamed sound

1. CROS (Contralateral Routing of Offsite Signals)

• Candidacy: One ear has unaidable profound hearing loss (or zero usable speech discrimination); the contralateral ear has normal hearing across all frequencies (thresholds $\le 20-25\,\text{dB HL}$ from $250$ to $8000\,\text{Hz}$).
• Mechanics:
  • The unaidable ear wears a CROS transmitter, consisting of a microphone array, DSP encoder, and wireless radio.
  • The normal ear wears a CROS receiver fitted with an open-fit silicone dome or widely vented hollow canal mold.
  • Audio picked up on the unaidable side is transmitted via Near-Field Magnetic Induction (NFMI) or $2.4\,\text{GHz}$ digital radio to the normal ear.
  • Acoustic Gain Prescribed: The CROS receiver applies zero net acoustic amplification ($0\,\text{dB}$ insertion gain). Its sole purpose is to reproduce the contralateral sound in the normal ear canal without occluding natural sound.
• Crucial Patient Expectation: CROS does not restore horizontal sound localization or true binaural hearing because all sound arrives at a single cochlea. It resolves the head shadow effect, allowing the patient to hear individuals speaking into their deaf side without physically turning their head.

2. BiCROS (Bilateral Contralateral Routing of Offsite Signals)

• Candidacy: One ear is unaidable (profound loss or non-functional discrimination); the contralateral (better) ear has aidable sensorineural hearing loss (mild to moderately severe loss requiring amplification).
• Mechanics:
  • The unaidable ear wears a wireless transmitter.
  • The better ear wears a fully functional hearing aid coupled with an appropriately vented earmold or dome.
  • The hearing aid on the better ear contains an internal mixer that combines inputs from its own local microphone with the wireless stream from the contralateral transmitter.
  • The processor applies prescriptive WDRC amplification to the combined signal to meet target audibility for the better ear's hearing loss.
  • Transmitter-to-Receiver Ratio: The balance between the local hearing aid microphone and the transmitted contralateral microphone is clinically adjustable (typically $1:1$, or adjusted to prioritize local conversation).

Severe-to-Profound SNHL: Cochlear Dead Regions

When sensorineural hearing loss exceeds $70-80\,\text{dB HL}$, the underlying pathology frequently involves complete destruction of both outer hair cells and inner hair cells (IHCs), along with degeneration of corresponding spiral ganglion neurons along a discrete segment of the basilar membrane. This acoustic void is termed a cochlear dead region.

Off-Frequency Listening

When a high-level pure tone is presented within a dead region, no functional IHCs respond at that characteristic frequency. However, as the stimulus intensity increases ($> 70-85\,\text{dB HL}$), the basilar membrane vibration envelope spreads to adjacent, healthier cochlear segments that still possess functional IHCs. The patient perceives the tone, but via off-frequency listening (detecting high frequencies through lower-frequency hair cells). This yields an elevated audiometric threshold, but the brain receives distorted, smeared tonotopic information.

The TEN (Threshold Equalizing Noise) Test

Developed by Brian C.J. Moore and colleagues, the TEN test provides an objective, psychophysically validated protocol to diagnose cochlear dead regions:

• Stimulus: A calibrated broadband masking noise that produces equal masked pure-tone thresholds across all test frequencies ($500$ to $4000\,\text{Hz}$) in ears with functioning IHCs.
• Protocol:
  1. Measure unmasked pure-tone thresholds at the test frequencies.
  2. Introduce TEN(HL) noise, specified in dB HL per ERB, at a level at or above the unmasked threshold (often about 10 dB above it). Re-measure pure-tone thresholds in the presence of the continuous TEN noise.
• Diagnostic Criteria for a Dead Region: A cochlear dead region is present at the test frequency if BOTH criteria are met: $\text{Threshold}_{\text{Tone+TEN}} - \text{Level}_{\text{TEN}} \ge 10\,\text{dB}$ $\text{AND} \quad \text{Threshold}_{\text{Tone+TEN}} - \text{Threshold}_{\text{Unmasked}} \ge 10\,\text{dB}$

Clinical Fitting Implications of Dead Regions

• Extensive Dead Regions ($> 1.5-2.0\,\text{octaves}$): Providing acoustic amplification within an extensive high-frequency dead region causes speech degradation. Amplified energy stimulates off-frequency low-frequency channels, creating harmonic smearing and worsening word recognition scores compared to leaving the high frequencies unamplified.
• Restricted Dead Regions ($< 1\,\text{octave}$): Modest amplification up to $1.5 \times f_e$ (edge frequency) can yield modest speech benefits. Above this cut-off, gain must be rolled off, and the audiologist should implement frequency lowering technology to shift high-frequency speech cues into surviving lower-frequency cochlear zones.

---

High-Frequency Lowering Technologies

When high-frequency hearing loss is severe or dead regions are present, conventional amplification cannot restore audibility due to acoustic feedback limits and cochlear desensitization. Frequency lowering translocates inaudible high-frequency consonant cues (e.g., voiceless fricatives /s/, /z/, /f/, /θ/, and affricates /tʃ/) into lower, audible frequency regions.

Frequency Lowering Paradigms:

1. Frequency Compression (NFC):
   [ 0 kHz -------- Cutoff (2.5 kHz) ==(Compress 2:1)==> 6 kHz ]
   (Squeezes all frequencies above cutoff into a compressed band)

2. Frequency Transposition:
   [ Source: 4-6 kHz ] ==(Shift Down 1 Octave)==> [ Destination: 2-3 kHz ]
   (Superimposes shifted source band on top of uncompressed audio)

3. Frequency Translation / Composition:
   [ Detect High /s/ Energy ] ==(Synthesize / Copy)==> [ Inject into 1.5-2.5 kHz ]
   (Injects spectral peak into lower band only when consonant is active)

1. Nonlinear Frequency Compression (NFC)

• Mechanics: All frequencies below a chosen start frequency (kneepoint/cutoff, e.g., $1.5-2.5\,\text{kHz}$) remain unprocessed and linear. Frequencies above the cutoff are compressed into a narrower high-frequency band at a selectable compression ratio ($1.5:1$ to $4:1$).
• Trade-off: Preserves all input spectral components, but warps harmonic relationships. Over-compression can cause the /s/ sound to be perceived as /ʃ/ ("sh"), causing patients to confuse "sail" and "shail."

2. Frequency Transposition

• Mechanics: A designated high-frequency source band (e.g., $3-6\,\text{kHz}$) is shifted down by a fixed interval (e.g., one octave) and mixed directly on top of the lower destination frequency band without compressing the bandwidth.
• Trade-off: Can cause spectral masking and intermodulation dissonance if transposed consonant energy collides with lower vowel formants.

3. Frequency Translation / Spectral Copying

• Mechanics: High-frequency spectral peaks are detected in real time. The algorithm copies the consonant envelope and synthesizes a replica in an audible lower-frequency band, active only when high-frequency consonants are present.

Verification via Probe-Microphone Speech Mapping

Frequency lowering cannot be verified using standard pure-tone aided thresholds. The clinician must use calibrated real-ear speech mapping:

1. Present calibrated speech stimuli containing the phoneme /s/ (e.g., the recorded sentence "The standard is sixty-six").
2. Observe the spectral energy peak of the lowered /s/ on the real-ear display screen.
3. The /s/ peak must sit within the patient's residual audible dynamic range (above hearing threshold, below UCL), while maintaining acoustic separation from the lower-frequency /ʃ/ formant peak (typically centered at $2-2.5\,\text{kHz}$). If /s/ and /ʃ/ overlap, the start frequency must be raised or the compression ratio lowered.

---

Power BTE Fittings & Acoustic Feedback Management

Fitting severe-to-profound hearing loss requires instruments delivering high Maximum Power Output ($> 135-144\,\text{dB SPL}$) and high acoustic gain ($> 70-85\,\text{dB}$). At these intensities, acoustic feedback oscillation is the primary clinical obstacle.

Acoustic Feedback Physics

Acoustic feedback occurs when amplified sound escaping from the ear canal reaches the hearing aid microphone at an amplitude equal to or greater than the original input signal, with an in-phase relationship ($0^\circ$ or integer multiples of $360^\circ$). This initiates a runaway oscillatory feedback loop, perceived as a high-pitched squeal.

Mechanical & Plumbing Requirements for Power Fittings

1. Earmold Material: Thick, soft medical-grade silicone or soft vinyl is mandatory. Silicone flexes with mandibular movement during chewing and talking, maintaining an airtight acoustic seal against the canal walls.
2. Canal Length & Bore: Long canal bore extending past the second canal bend into the bony canal aperture. This places the receiver discharge closer to the tympanic membrane, reducing the trapped ear canal volume ($V_c$). Reducing canal volume from $2.0\,\text{cc}$ to $0.5\,\text{cc}$ increases sound pressure by $+6-12\,\text{dB}$ for the same acoustic output, reducing the electronic gain required from the amplifier.
3. Sound Tubing: Thick-wall (#13 Thick) or double-wall (#13 Super Thick) vinyl tubing. Thin-walled tubing allows acoustic energy to radiate through the tubing walls into the microphone, causing feedback.
4. Venting: Unvented or pressure-vented ($0.6-0.8\,\text{mm}$) molds only. Any vent $\ge 1.0\,\text{mm}$ leaks high-level acoustic energy, triggering feedback in power fittings.
5. Acoustic Dampers: Sintered metal or cloth dampers ($1500-2200\,\Omega$) inserted into the earhook smooth out sharp mechanical receiver resonance peaks ($1.5-3\,\text{kHz}$), which are prime triggers for acoustic feedback.