Resumen:
Objective: To estimate how much could intratympanic gentamicin (ITG) interfere with the vestibular-ocular reflex (VOR) parameters on instrumental head impulse test (HIT), either with scleral search coil or video head impulse test and, eventually, foresee the control of vertigo crisis in unilateral intractable Meniere's disease (MD). Methods: A literature search was conducted in PubMed, Scopus, Web of Science and Cochrane search engines. The search terms used were "vestibular ocular reflex", "head impulse test", "gentamicin," and "Meniere's disease". Limitations included text availability to be full text, species to be humans and language to be English. All study types were included. 89 articles were screened identifying four eligible studies were identified. Studies were included after consensus of the authors. Meta-analysis was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Data was analysed using Review Manager software. Results: Instrumental HIT, after ITG for MD, demonstrated, in the treated ear, a decreased gain in the horizontal, posterior and superior semicircular canals (SCC), of 0.36 (0.26; 0.47; 95% CI), 0.35 (0.22; 0.48; 95% CI) and 0.28 (0.21; 0.35; 95% CI), respectively. Gain asymmetry increases between the treated and non-treated ear of 23.78 (7.22; 40.35; 95% CI), 32.01 (12.27; 51.76; 95% CI) and 17.49 (9.99; 24.99; 95% CI), were similarly detected in the horizontal, posterior and superior SCC, respectively. Significantly smaller gain values after the first treatment were observed for a single injection group versus multiple injection group in the horizontal (p = 0.002) and superior SCCs (p = 0.016). Conclusions: Instrumental HIT is effective in evaluating the SCC function after ITG for intractable unilateral MD. VOR gain changes in the direction of the treated ear in the three SCC have been clearly registered. An increased reduction of the VOR gain in the horizontal and anterior SCC also seemed to foresee the control of vertigo crisis. Still, after meta-analysis, the small number of patients' data available did not allow to define a treatment end-point value. This review also indicated that further and better-designed studies are warranted. (C) 2018 Elsevier B.V. All rights reserved.