Methodological aspects of testing vestibular evoked myogenic potentials in infants at universal hearing screening program

Abstract

Motor development in infants is dependent upon the function of the inner ear balance organ (vestibular organ). Vestibular failure causes motor delays in early infancy and suboptimal motor skills later on. A vestibular test for newborns and infants that is applicable on a large scale, safe and cost effective is in demand in various contexts: in the differential diagnosis of early onset hearing loss to determine forms associated with vestibular failure; in early hearing habilitation with cochlear implant, indicating the vestibular predominant side; and in the habilitation of children affected by motor skill disorders, revealing the contribution of a vestibular failure. This work explored the feasibility of cervical vestibular evoked myogenic potentials (VEMP) in conjunction with newborn universal hearing screening program. VEMP was measured after the hearing tests and was evoked by bone-conducted stimuli. Moreover, stimulus delivery was regulated by neck muscle activity, with infants rested unconstrained in their parents´ arms and with the head supported by the operator´s hand. This VEMP protocol showed a high level of feasibility in terms of test viability and result reproducibility. VEMP integrated into the newborn hearing screening program may represent a practical method for large-scale assessment of balance function in infants.

Figure 1

Effect of the prestimulus EMG on the VEMP scaled amplitude: at EMG over 150 µVolts the amplitude is generally depressed and closer to the detection threshold (horizontal line: 0,18 value scaled amplitude).

Figure 2

Histogram showing the sample distribution (ears) related to the number of completed recordings. Vertical line points out the target level for test completion (120). Nearly half of the subjects reached the maximum default of 200 collected sweeps.

Figure 3

Box plot indicating the difference in prestimulus EMG between the completed and not completed tests. o: outlier; *extremes. Difference statistically significant at p < 0,01.

Figure 4

Distribution of identifiable vs not identifiable VEMP for subject (x axis) and amplitude (y-axis left) or scaled amplitude (y-axis right). The majority of the non-identifiable VEMP had an amplitude less than 24,7 µVolts or scaled amplitudes less than 0,18. The diagnostic precision of these two cut offs is provided.