The aim of this study was to investigate the effect of visual perception training on binocular visual function recovery and the changes of perceptual eye position for postoperative patients with intermittent exotropia.

70 patients(70 eyes)with intermittent exotropia who underwent strabismus surgery from January 2019 to May 2021 in Department of Strabismus and Pediatric Ophthalmology, Shanxi Eye Hospital were recruited, including 37 males (37 eyes) and 33 females (33 eyes) with an average age of (9.9±4.6) years-old (ranged from 4 to 27 years-old). According to the choice of their parents and patients, they were divided into training group (38 cases, 38 eyes) and control group (32 cases, 32 eyes). All patients were received visual perception examination; the training group were trained for 2 months. Before and after training, visual function, near stereopsisand perceptual eye position were examined. Gender, the number of people with simultaneous visual function were countable data, and expressed by case and percentage, and compared by Chi-square test and Fisher exact probability method between groups. The age, range of fusion, and degree of perceptual eye position deviation were measurable data, and expressed as ±s after normality test and compared by independent sample t test between groups; otherwise, described by the median and quartile and compared by non-parametric rank sum test. The stereo acuity was graded data, expressed by case and percentage, and compared by non-parametric rank sum test.

Before training, there were 25 cases (78.1%) and 28 cases (73.7%) with simultaneous visual function in the control group and the training group. After training for 2 months, there were 27 patients (84.4%) and 37 patients (97.4%) with simultaneous visual function in two groups, respectively. There was no statistically significant difference between them using Fisher exact probability method (P>0.05). Before training, the fusion range of control group and training group were 12.0(18.75)° and 6.5(17.25)°; after training for 2 months, those were 13.0(20.75)° and 18.0(12.25)°, respectively. The training group was significantly enlarged between before and after training using non-parametric rank sum test (Z=-4.54, P<0.05); there was statistically significant difference between groups (Z=-2.49, P<0.05). After training for 2 months, the number of the distance stereo vision function of patients in the training group was 22 cases (57.9%) with the significant difference between after and before training (χ²=-4.46, P<0.05); there was statistically significant difference between groups (χ²=-2.08, P<0.05). After training for 2 months, the number of the near-stereoscopic vision function of patients in the control group and the training group was 20 cases (62.5%) and 30 cases (78.9%) with significant difference between before and after training using Chi-square test (χ²=-2.65, -3.26; P<0.05). Before training, the degree of horizontal perceptual eye position of control group and training group were 148.5 (232.25) pixels and 78.5(234.25) pixels, respectively; the degree of vertical perceptual eye position were 14.0 (23.75) pixels and 10.0 (28.00) pixels, respectively. After training for 2 months, the degree of horizontal perceptual eye position were 43.0 (155.25) pixels and 13.0 (27.25) pixels; the degree of vertical perceptual eye position were 9.0 (14.25) pixels and 4.0 (6.00) pixels. There were statistically significant difference in the degree of perceptual eye between after and before training (Z=-4.50, -3.43; P<0.05); there were statistically significant difference between groups after training using non-parametric rank sum test (Z=-3.34, -3.28; P<0.05).

The perceptual eye position could more accurately and quantitatively reflect the damage degree of central visual perception function. Visual perception training could be helpful to repair central visual perception function and reconstruct binocular vision function.