A meta-analysis of clinical electro-oculography values.

BACKGROUND: The aim of the meta-analysis was to derive a range of mean normal clinical electrooculogram (EOG) values from a systematic review of published EOG studies that followed the guidelines of the ISCEV standard for clinical electro-oculography.

METHODS: A systematic literature review was performed using four relevant databases limited to peer-reviewed articles in English between 1967 and February 2017. Studies reporting clinical EOG or FO normal values were included when the report used a standard 30° horizontal saccade, a retinal luminance of between 100 and 250 cd m-2 , and had > 10 subjects in their normative values. The search identified 1145 articles after duplicates were removed with subsequent screening of the abstracts excluding a further 1098, resulting in 47 full-text articles that were then assessed by the author (PC) with a final nine articles meeting the inclusion criteria. An overall effect estimate using inverse variance-weighted meta-analysis was performed to estimate the mean values for the light peak/dark trough ratio (LP:DT ratio) (dilated and undilated), the time to the LP, the amplitude of the LP, dark trough (DT) and the fast oscillation (FO) peak-to-trough ratio from the included studies.

RESULTS: The mean dilated LP:DT ratio was 2.35 (95% CI 2.28-2.42); undilated LP:DT ratio was 2.37 (95% CI 2.28-2.45); LP amplitude was 835 (95% CI 631-1039) µV and the mean time to the LP being 8.2 (95% CI 7.7-8.7) min. The mean DT amplitude was 358 (95% CI 292-424) µV, and the mean FO peak-to-trough ratio was 1.13 (95% CI 1.11-1.16). The results of the LP/DT ratio are drawn from studies with a mean ± standard deviation (SD) age of 34.08 ± 12.93 years for dilated and 33.65 ± 12.28 years for undilated LP/DT ratios.

CONCLUSIONS: The meta-analysis of EOG studies has generated a reference range of normal mean values for clinicians to refer to when using the ISCEV clinical EOG. It provides a potential method to generate similar data sets from published normal values in related visual electrophysiology tests.