Bagolini filter bars: an analysis of light transmittance and their use in a pediatric population

Purpose: The purpose of this study is to determine whether the Sbisa bar (Sbisa Ophthalmic Instruments) (referred to in this article as the BF(Sbisa) bar) and Bagolini Filter bar (C.O.I. Vision) (referred to in this article as the BF(new) bar) are repeatable and comparable, for measuring density of suppression or strength of abnormal binocular single vision (BSV), in a pediatric population. Methods: Using a spot photometer, percentage light transmittance of filters on two BF(Sbisa) bars and two BF(new) bars was measured. Thirty-three participants aged 5-12 years with unilateral strabismus were recruited. Density of suppression or strength of abnormal BSV was measured with two of each bar, in addition to a repeated measure with one of each bar. Results: Light transmittance did not reduce consistently with increasing filter number (the number assigned to each filter by the manufacturer). Using data as filter numbers, a BF(Sbisa) bar was not comparable to a BF(new) bar (ᵶ = -3.936, p < 0.001, r = -0.28); different BF(Sbisa) bars were not comparable (ᵶ = -3.103, p = 0.005, r = -0.22) but different BF(new) bars were comparable (ᵶ = -1.165, p = 1.285, r = -0.08). Using light transmittance values, a BF(Sbisa) bar was comparable to a BF(new) bar (ᵶ = -0.385, p = 3.535, r = -0.03); different BF(Sbisa) bars were comparable (ᵶ = -1.476, p = 0.720, r = -0.10) but different BF(new) bars were not comparable (ᵶ = -3.354, p < 0.001, r = -0.24). The BF(Sbisa) and BF(new) bars were not repeatable using filter numbers (ᵶ = -3.648, p < 0.001, r = -0.26 and ᵶ = -3.099, p = 0.005, r = -0.22, respectively) or light transmittance values (ᵶ = -3.743, p < 0.001, r = -0.27 and ᵶ = -2.727, p = 0.025, r = -0.19, respectively). An order effect existed using filter numbers (ᵶ = -4.473, p < 0.001, r = -0.32) and light transmittance values (ᵶ = -4.525, p < 0.001, r = -0.32). Conclusion: There is a need for the production of bars with consistently reducing and comparable light transmittance of filters if measurement of density of suppression or strength of abnormal BSV with filter bars is to continue, because the current bars show inconsistencies, particularly at lower filter numbers. In the meantime, the bars are limited by their lack of comparability or repeatability. Using only BF(new) bars is preferable to using BF(Sbisa) bars or a mixture of the two types of bar because two different BF(new) bars may be comparable using filter number readings, but filter number readings from different BF(Sbisa) bars or from a BF(Sbisa) and BF(new) bar are not comparable. As the BF(Sbisa) and BF(new) bars are currently labeled, a second measure of density of suppression or strength of abnormal BSV is not recommended within 2 min of an initial measurement, because the measurement procedure seems to cause a reduction in suppression or abnormal BSV that does not recover with repeated measurements 2 min apart.