J Vestib Res. 2009;19(3-4):127-135.
Author(s)
Abstract
OBJECTIVE: To determine normal limits and to analyse the test-retest reliability of the vestibular evoked myogenic potentials (VEMPs) parameters.
METHODS: The VEMP procedure was repeated on different test days to analyze test-retest differences. We calculated several reliability parameters: intraclass reliability coefficient (ICC), method error (ME), coefficient of variation of the method error (CV(ME)), standard error of measurement (SEM) and minimal difference (MD) for test-retest measurements. Normal values for left-right differences, based on the interaural ratio (IAR), were determined.
RESULTS: For each VEMP parameter, the ICC values indicated excellent reliability, except for p13 and corrected amplitude (fair to good reliability). The CV(ME)) values were less than 7% for p13, n23, threshold, MRV(females) and MRV(males). For the parameters corrected amplitude and raw amplitude, the CV(ME)) values exceeded 15%. The 95% IAR prediction intervals (PIs) were also largest for the parameters raw amplitude and corrected amplitude.
CONCLUSION: In order to evaluate a VEMP outcome in a patient, the VEMP parameters and IAR values can be compared with the 95% PI of the normal values. When successive measurements are performed within the same subjects, the minimal difference (MD) serves as a tool to decide whether these differences are clinically relevant or not.