Gerald R. Popelka

A New Standardized Format for Reporting Hearing Outcome in Clinical Trials

The lack of an adequate standardized method for reporting level of hearing function in clinical trials has hampered the ability of investigators to draw comparisons across studies. Variability in data reported and presentation format inhibits meta-analysis and makes it impossible to accumulate the large patient cohorts needed for statistically significant inference. Recognizing its importance to the field and after a widely inclusive discussion, the Hearing Committee of the American Academy of Otolaryngology–Head and Neck Surgery endorsed a new minimal standard for reporting hearing results in clinical trials, consisting of a scattergram relating average pure-tone threshold to word recognition score. Investigators remain free to publish their hearing data in any format they believe is interesting and informative, as long as they include the minimal data set to facilitate interstudy comparability.

Effect of activating signal bandwidth on acoustic‐reflex thresholds

Acoustic‐reflex thresholds were determined as a function of activating‐signal bandwidth in 10 subjects with normal hearing sensitivity and eight subjects with noise‐induced sensorineural hearing loss. Acoustic‐reflex thresholds were measured by monitoring the acoustic impedance at the tympanic membrane for multicomponent activating stimuli centered logarithmically around 250, 500, 1000, 2000, and 4000 Hz. Normal listeners displayed monotonically decreasing acoustic‐reflex thresholds as bandwidth was increased by adding additional components to the complex activating signals. A knee in the functions occurred at values of bandwidth similar to previously reported critical bandwidths for the acoustic reflex. Sensorineural hearing loss altered the functions such that acoustic‐reflex thresholds tended to be elevated when the spectral content of the activating signal extended into the hearing‐loss region. This resulted in flatter functions relating acoustic‐reflex thresholds to activating‐signal bandwidth.

Auditory Impairment in Infants at Risk for Bilirubin-Induced Neurologic Dysfunction

Classical and subtypes of kernicterus associated with bilirubin toxicity can be differentiated in part with physiological auditory measures that include auditory-evoked potentials and measures of cochlear integrity. The combination of these auditory measures suggests that bilirubin exposure results in auditory system damage initially at the level of the brainstem, progressing to the level of the VIII cranial nerve and then to greater neural centers. There is no evidence of neural damage at the level of the cochlea. Auditory neural damage from bilirubin toxicity ranges from neural timing deficits, including neural firing delays and dyssynchrony, to neural response reduction and even elimination of auditory neural responses. This condition is comprehensively described as auditory neuropathy spectrum disorder. Independent measures of cochlear function and auditory neural function up to the level of the brainstem can effectively diagnose auditory neural damage resulting from bilirubin neurotoxicity. Intervention, including cochlear implants can be effective.

In vivo imaging of mammalian cochlear blood flow using fluorescence microendoscopy.

Aims: We sought to develop techniques for visualizing cochlear blood flow in live mammalian subjects using fluorescence microendoscopy. Background: Inner ear microcirculation appears to be intimately involved in cochlear function. Blood velocity measurements suggest that intense sounds can alter cochlear blood flow. Disruption of cochlear blood flow may be a significant cause of hearing impairment, including sudden sensorineural hearing loss. However, inability to image cochlear blood flow in a nondestructive manner has limited investigation of the role of inner ear microcirculation in hearing function. Present techniques for imaging cochlear microcirculation using intravital light microscopy involve extensive perturbations to cochlear structure, precluding application in human patients. The few previous endoscopy studies of the cochlea have suffered from optical resolution insufficient for visualizing cochlear microvasculature. Fluorescence microendoscopy is an emerging minimally invasive imaging modality that provides micron-scale resolution in tissues inaccessible to light microscopy. In this article, we describe the use of fluorescence microendoscopy in live guinea pigs to image capillary blood flow and movements of individual red blood cells within the basal turn of the cochlea. Methods: We anesthetized eight adult guinea pigs and accessed the inner ear through the mastoid bulla. After intravenous injection of fluorescein dye, we made a limited cochleostomy and introduced a compound doublet gradient refractive index endoscope probe 1 mm in diameter into the inner ear. We then imaged cochlear blood flow within individual vessels in an epifluorescence configuration using one-photon fluorescence microendoscopy. Results: We observed single red blood cells passing through individual capillaries in several cochlear structures, including the round window membrane, spiral ligament, osseous spiral lamina, and basilar membrane. Blood flow velocities within inner ear capillaries varied widely, with observed speeds reaching up to approximately 500 μm/s. Conclusion: Fluorescence microendoscopy permits visualization of cochlear microcirculation with micron-scale optical resolution and determination of blood flow velocities through analysis of video sequences.

Effect of sensorineural hearing loss on acoustic stapedius reflex growth functions.

The growth function of the acoustic stapedius reflex was measured in subjects with normal hearing and sensorineural hearing loss of cochlear origin. The effects of age and magnitude of hearing loss were controlled. Activating stimuli were 500, 1000, and 2000 Hz tones and broadband noise. Stapedius muscle activity was inferred from acoustic impedance measures in the contralateral ear. The mean growth functions for tones were essentially linear in log–log plots with the rate of growth equal for the two groups. The mean growth function for the noise signal was curvilinear for the normal hearing group and linear for the hearing loss group. Comparison of slope functions derived from the fitted data indicated that the rate of reflex growth for the noise signal, over a limited range above reflex threshold, is greater in ears with cochlear lesions than normal ears. For higher level noise signals, however, the rate of reflex growth is similar for normal and pathological ears. The effect of a cochlear lesion on the...

Extra-facial gestures in relation to speechreading

Abstract In a pilot study ( N = 32) it was found that extra-facial gestures, by themselves, elicited responses ranging from identical to similar but non-identical. It was hypothesized that the use of extra-facial gestures would influence the proficiency of speechreading. Twenty sentences based upon corresponding concepts of gestures used in the pilot study were presented to two groups of 16 subjects each. Each of the subject groups received alternate test items without gestures and with gestures. To minimize possible facial gestures a face-mask was worn by the speaker which allowed only his lips to be seen. Results indicated that for the 20 gestures employed in this study, there was a statistically significant increase in the proficiency of speechreading.

Loudness and the acoustic reflex

The hypothesis that loudness is constant at acoustic‐reflex threshold is examined. Acoustic‐reflex thresholds were measured in five normal‐hearing adult subjects for octave frequencies from 250 to 4000 Hz, for a narrow‐band noise (BW=50 Hz) centered at 500 Hz, and for a wide‐band noise. After acoustic‐reflex threshold was determined, the loudness of the reflex‐threshold stimulus was measured utilizing a monaural loudness‐balance procedure employing a 1000‐Hz comparison tone and a Method of Constant Stimuli. Results indicate that for the activating stimuli used in this experiment, loudness at acoustic‐reflex threshold varies over a 17‐dB range. The range would probably increase if lower frequency activating stimuli were used. The data suggest that the acoustic reflex is not dependent on some critical loudness of the activating stimulus.Subject Classification: 65.50, 65.22, 65.75.

Is It Valid to Calculate the 3-Kilohertz Threshold by Averaging 2 and 4 Kilohertz?

Many guidelines for reporting hearing results use the threshold at 3 kilohertz (kHz), a frequency not measured routinely. This study assessed the validity of estimating the missing 3-kHz threshold by averaging the measured thresholds at 2 and 4 kHz. The estimated threshold was compared to the measured threshold at 3 kHz individually and when used in the pure-tone average (PTA) of 0.5, 1, 2, and 3 kHz in audiometric data from 2170 patients. The difference between the estimated and measured thresholds for 3 kHz was within ±5 dB in 72% of audiograms, ±10 dB in 91%, and within ±20 dB in 99% (correlation coefficient r = 0.965). The difference between the PTA threshold using the estimated threshold compared with using the measured threshold at 3 kHz was within ±5 dB in 99% of audiograms (r = 0.997). The estimated threshold accurately approximates the measured threshold at 3 kHz, especially when incorporated into the PTA.

Identification of Noise Sources That Influence Distortion Product Otoacoustic Emission Measurements in Human Neonates

Objective: The objective of this study was to identify individual sources of noise and their contribution to the overall noise that influences valid measurement of otoacoustic emissions in neonates. The hypothesis was that careful selection of eliciting signals and signal processing parameters, unique analysis of measured results, and control of certain subject characteristics would allow isolation of these individual noise sources and determine their relative influence. Design: Eliciting signal parameters were optimized and held constant to minimize equipment noise. Analysis of noise floors in relation to signal level was used to identify equipment‐related noise associated with changes in signal parameters. Analysis of noise floor distributions was used to determine whether environmental noise entered the measurements via inadequate coupling of the probe to the ear. The acoustic characteristics of the middle ear were varied via subject selection to determine the influence of middle‐ear characteristics on noise floor levels. Results: The two sources of noise associated with the measurement equipment need not contribute to the noise floor for biologically relevant otoacoustic emissions measurements (eliciting signal levels between 30 and 75 dB SPL). Of the two pathways identified for environmental noise, the pathway resulting from an inadequate seal between the probe and the ear canal can be eliminated. One of the two sources of noise related to the subject, noise resulting from biologic activity unrelated to the ear can be minimized. However, the remaining factor, the status of the middle ear, has been shown to contribute as much as 6 dB to the overall noise floor. Conclusions: Careful selection of signal parameters and additional data analyses and procedural variables can isolate or control several sources of noise that influence distortion product otoacoustic emission measurements in neonates. Tight coupling between the probe unit and the external ear canal should be maintained for all measurements. Middle ear abnormalities can increase noise floors up to 6 dB.

Clinical Significance of Relative Probe-Tone Levels on Distortion Product Otoacoustic Emissions

The effect of systematic variations in the relative levels (L1, L2) of two primary tones (f1, f2) on the amplitude of the distortion product otoacoustic emission (DPOAE) at 2f1-f2 and f1 < f2 was investigated in 14 ears from 7 normally hearing human subjects. The primary tones (f2:f1 = 1.23) were geometrically centred at the standard clinical frequencies of 0.5, 1, 1.5, 2, 3, 4, 6 and 8 kHz. The quantity L1-L2 was varied systematically from -10 dB through + 10 dB with L2 held constant at 75 dB SPL for negative values, L1 held constant at 75 dB SPL for positive values, and L1 = L2 = 75 dB SPL at 0 dB relative difference. The maximum amplitudes of the distortion products were generated when L1 = L2 at all geometric centre frequencies except 8 kHz. The reduction of the DPOAE with reduction of L1 was linear at a rate that gradually increased as a function of geometric mean frequency. To a lesser extent, the reduction of the DPOAE with reduction of L2 also was linear but at a rate that systematically decreased as a function of geometric mean frequency. The results suggest, that to maximize the level of the distortion product for clinical purposes, the relative levels of the primary tones should be equal to each other, at least when overall stimulus levels are around 75 dB SPL and f2:f1 = 1.23.