• vHIT showed 42% sensitivity and 89% specificity for vestibular migraine. • Abnormal vHIT findings in 43% of patients with vestibular migraine. • Covert or overt saccades were reported in 36% of the included cases. • High specificity suggests vHIT may help rule in vestibular migraine. To estimate the proportion of patients with VM who have abnormal vHIT results and to determine the sensitivity and specificity of the test to detect VM. This study followed PRISMA guidelines to conduct a meta-analysis on the accuracy of vHIT in VM patients. A literature search was performed in ten indexes databases (PubMed, PubMed PMC, Embase, Scopus, Web of Science, Cochrane Library, BVS, ProQuest, Epistemonikos and MEDLINE) without time restrictions, using the descriptors “Vestibular Migraine” AND “Video Head Impulse Test”. Studies were selected based on the PICO strategy (Patient, Intervention, Comparison, and Outcome). Statistical analysis used the random-effects method to pool results, assessing heterogeneity among studies with Cochran’s Q test and the I 2 statistic. The accuracy of vHIT was estimated through sensitivity, specificity, and predictive values. A significance level was 5%. This review was not registered in PROSPERO (International Prospective Register of Systematic Reviews) because it evaluates previously published observational studies of diagnostic performance without patient-level data collection. Nine studies published between 2016 and 2023 were included in the meta-analysis. The prevalence of abnormal results was 43%. vHIT had a sensitivity of 42%, specificity of 89%, positive predictive value of 83%, and negative predictive value of 54% for VM diagnosis. Saccades was observed in 36% of cases. Normal VOR gain was found in 22% of cases, while low VOR gain was observed in 12%. The gains in the lateral, anterior, and posterior semicircular canals ranged from 0.87 to 0.90. Except for normal VOR gain, all other outcomes showed high heterogeneity across studies. 43% of VM patients exhibited abnormalities in vHIT. The test demonstrated low sensitivity and high specificity to detect VM. 1.
Neto et al. (Thu,) studied this question.