Comparison of a new, filter-free stereopsis test (BEST) with the RANDOt stereotest in a pediatric cohort

Purpose: To evaluate the Bernell Evaluation of Stereopsis Test (BEST) (Bernell Corporation, Mishawaka, IN), a new lenticular technology and filter-free test to measure stereopsis in children, and compare it to the Randot Stereotest (Randot) (Stereo Optical, Inc., Chicago, IL). Methods: This was a retrospective review of the medical records of children examined at the Center for Pediatric Ophthalmology at Hadassah-Hebrew University Medical Center from July to November 2018. All children were evaluated with the Randot and BEST in a random order. Results: The study included 100 children (53% female, 64% orthophoric) with a mean age of 8.52 ± 3.18 years (range: 3.3 to 17.8 years) and mean best corrected visual acuity of 0.178 ± 0.16 logMAR (range: 0 to 0.7 logMAR). The mean BEST stereoacuity was 1.772 ± 0.27 log seconds of arc (arcsec), whereas the mean Randot stereoacuity was 1.778 ± 0.39 log arcsec (P = .835). The Bland–Altman analysis revealed an overall bias of 0.0073 log arcsec (95% confidence interval: 0.04219 to 0.05679 log arcsec), with limits of agreement of -0.4816 to 0.4962 log arcsec (0.3299 to 3.1347 arcsec). A significant proportional bias was noted because the difference between the tests was significantly larger in higher log arcsec values (t = 5.566, P < .001). Age, gender, strabismus, and visual acuity did not affect the differences between the stereoacuity tests (P > .05). Both tests were influenced by strabismus but not monocular amblyopia. Conclusions: BEST stereoacuity measurements were comparable to those of the Randot, with no significant difference in crude values, minimal bias, and narrow limits of agreement. A larger, clinically insignificant variability between tests was noted with lower stereoacuity values, because children achieved better scores with the BEST. The BEST may be a valuable tool in the arsenal of the ophthalmologist.