双眼视知觉网络训练对弱视治疗短期视力提升效果的临床研究

目的探讨双眼视知觉网络训练中，影响最佳矫正视力(BCVA)值提升的相关因素以及各种训练内容的功效。 方法选取2018年10月至2019年2月于深圳市眼科医院斜视与小儿眼科门诊就诊的双眼弱视患者29例(58只眼)。其中，男性18例(36只眼)，女性11例(22只眼)；年龄3~15岁，平均年龄(6.1±2.6)岁。将所有患者按照屈光状态和弱视类型进行分组。全部患者均采用SJ-RS-WL2015型多媒体视觉功能训练治疗系统的网络平台进行训练3个月，检查并记录训练前和训练后，患者的最佳矫正视力、患眼屈光状态和患者的弱视类型。采用均数±标准差或中位数(四分位间距)描述患者的年龄、训练次数、BCVA值和等效球镜(SE)。采用配对t检验或Wilcoxon符号秩和检验，比较训练前后患者BCVA值和SE的差异。采用Spearman相关性系数法，分析训练前后患者BCVA的提高值分别与训练前BCVA值和SE的相关性。采用Mann-Whitney U秩和检验，比较训练前后不同弱视类型组患者的BCVA值与SE。采用单因素和多因素回归分析，寻找影响患者BCVA值的相关因素。 结果所有患者中，屈光不正性弱视患者有13例(26只眼)，占44.8%；屈光参差性弱视患者有11例(22只眼)，占37.9%；斜视性弱视患者有4例(8只眼)，占13.8%；形觉剥夺性弱视患者有1例(2只眼)，占3.5%。训练后3个月，患者右眼和左眼的平均BCVA提高值分别为(0.14±0.13)和(0.18±0.15)，经Spearman相关分析，右眼和左眼BCVA的提高值与训练前BCVA值均呈负相关，其相关性有统计学意义(r＝-0.753，-0.439；P<0.05)。患者右眼和左眼SE的提高值分别为(-0.15±0.78)D和(-0.25±0.79)D，经Spearman相关分析，右眼和左眼SE的提高值与训练前患眼SE不存在线性关系，其相关性无统计学意义(r＝-0.339，-0.270；P>0.05)。训练后3个月，SE≤3.00 D组、3.00 D≤SE≤6.00 D组和SE>6.00 D组患者右眼BCVA的提高值分别为0.00(0.10)、0.20(0.10)和0.20(0.10)；左眼BCVA的提高值分别为0.00(0.20)、0.30(0.15)和0.10(0.20)。三组患者右眼SE的提高值分别为0.00(0.00)D、0.00(1.19)D和-1.12(0.75)D；左眼SE的提高值分别为(-0.01±0.52)D、(-0.24±0.84)D和(-0.72±0.96)D。屈光不正性弱视组和屈光参差性弱视组患者，右眼和左眼SE提高值的比较，组间的差异无统计学意义(Z＝1.412，1.968；P>0.05)。精细刺激、视觉技巧、Gabor训练、对比敏感度和信息提取等不同训练内容对患者右眼BCVA提高值影响的比较，差异无统计学意义(t＝0.092，1.614，0.028，0.340，-1.016；P>0.05)；对患者左眼BCVA提高值影响的比较，差异无统计学意义(t＝-0.007，0.572，0.484，1.889，0.530；P>0.05)。训练后3个月，患者立体视的平均值为(-146.90±290.26)″。 结论短期双眼视知觉网络训练有助于弱视患者BCVA值的提升。患者的初始BCVA值越低，训练后BCVA的提高值越大。不同屈光状态与不同弱视类型组患者BCVA的提高值相当。然而，不同的训练内容对患者BCVA值的提升效果仍无法确认。

Analysis of the effect of binocular visual perception learning based on network training on short-term visual acuity improvement in amblyopia treatment

ObjectiveTo explore the related factors of amblyopia about the improvement of corrected visual acuity and the efficacy of various training contents after 3 months of binocular visual perception learning through network training. MethodsA total of 29 patients (58 eyes) with binocular amblyopia were selected from the outpatient Department of Strabismus and Pediatric Ophthalmology in Shenzhen Eye Hospital from October 2018 to February 2019. There were 18 males cases (36 eyes), 11 females cases (22 eyes); age 3 to 15 years-old, mean (6.1±2.6) years-old. All patients were divided into groupes according to different refractive status and different types of amblyopia. All patients were trained on the network platform of SJ-RS-WL2015 multimedia visual function training treatment system and the best corrected visual acuity (BCVA), refractive states, and amblyopia types of patients were examined and recorded before and after 3 months of training. Age, number of training, BCVA, spherical equivalent (SE) were expressed by mean±standard deviation or median (interquartile range). Paired t-test or Wilcoxon signed rank test were used to compare the differences of the BCVA and SE of patients before and after training. Spearman correlation was used to analyze the correlation among the increased value of BCVA and BCVA value and SE after binocular visual perception network training. Mann-Whitney U rank sum test analysis was used to compare the BCVA value and SE degree of patients with different types of amblyopia before and after training. The single factor and multiple factor regression analysis were used to find the relevant factors that affect the patient′s BCVA. ResultsAmong all patients, 13 cases (26 eyes) with ametropic amblyopia, were accounting for 44.8%; 11 cases (22 eyes) with anisometropic amblyopia, were accounting for 37.9%; 4 cases (8 eyes) with strabismic amblyopia, were accounting for 13.8% and 1 case (2 eyes) with form-deprivation amblyopia, were accounting for 3.5%. After training three months, the average increases of BCVA of the right eye and the left eye were (0.14±0.13) and (0.18±0.15), respectively; and the Spearman correlation analysis showed that the correlation were statistically significant (r=-0.753, -0.439; P<0.05). The average increases of SE of the right eye and the left eye were (-0.15±0.78)D and (-0.25±0.79)D respectively, and the Spearman correlation analysis showed that the correlation were non-statistically significant (r=-0.339, -0.270; P>0.05). After training three months, the increases of BCVA of the right eye in SE≤3.00 D group, 3.00 D≤SE≤6.00 D group and SE>6.00 D group were 0.00(0.10)D, 0.20(0.10)D and 0.20(0.10)D, respectively; the increases of BCVA of the left eye in three groups were 0.00(0.20), 0.30(0.15) and 0.10(0.20), respectively. The increases of SE of the right eye in there groups were 0.00(0.00)D, 0.00(1.19)D and -1.12(0.75)D, respectively; the increases of SE of the left eye were (-0.01±0.52)D, (-0.24±0.84)D and (-0.72±0.96)D, respectively. There was non-statistically significant in the increases of SE between the right and left eyes in the ametropic amblyopia group and the anisometropic amblyopia group (Z=1.412, 1.968; P>0.05). Different training contents such as fine stimulation, visual skills, Gabor, contrast sensitivity and information extraction had non-statistically significant with the increases of BCVA in the right eye (t=0.092, 1.614, 0.028, 0.340, -1.016; P>0.05); those of the increases of BCVA in the left eye had non-statistically significant (t=-0.007, 0.572, 0.484, 1.889, 0.530; P>0.05). After training three months, the average stereopsis were (-146.897±290.260)△. ConclusionsShort-term application of binocular visual perception network training is effective for improving BCVA in amblyopia. The lower the patient′s initial BCVA value, the greater the increase in BCVA after training. The improvement of BCVA of patients with different refractive status and different types of amblyopia is equivalent. Initial BCVA determines the improvement of BCVA after training, but the effect of different training contents on BCVA can not be confirmed for the time being.