探讨紫外线辐照地区年龄相关性白内障(ARC)患者临床特征及其房水谷胱甘肽过氧化物酶(GPX)含量的作用及机制。

选取2024年7月至2025年7月在北京大学第三医院、西藏自治区人民医院、赤峰松山医院、科左中旗医院及锡林郭勒盟中心医院就诊的ARC患者140例(140只眼)。其中，女性83例(83只眼)，男性57例(57只眼)；年龄51~89岁，平均年龄为(69.7±9.3)岁。按照2011—2014年年紫外辐照累积量平均值分为低紫外线组、中紫外线组及高紫外线组；按照年龄将患者划分为50~<65组、≤65~<75组、≤75~<85组及≥85组。所有ARC患者在入组前均接受白内障术前常规检查，包括最佳矫正视力(BCVA)、眼压(IOP)、裂隙灯显微镜检查及眼底检查。问询并记录其高度近视眼、糖尿病、高血压及心脑血管病史情况。对年龄、最佳矫正视力(BCVA)、眼压(IOP)及谷胱甘肽过氧化物酶(GPX)等经检验符合正态分布的连续变量采用±s描述，组间比较采用单因素方差ANOVA检验，当差异有统计学意义时，进一步采用Bonferroni法两两比较。两分类变量的两组间比较采用Mann-Whitney U检验，多分量变量的多组间比较采用Kruskal-Wallis H检验。性别和眼别等计量资料采用例数和百分比描述，组间比较采用卡方检验。连续变量与有序等级变量的相关性分析采用Spearman相关系数检验，连续变量与连续变量的相关性分析采用Pearson相关系数检验。采用多元线性回归分析GPX的影响因素。

低、中及高紫外线组中皮质混浊评分大于3分的患者占比依次为23.2%、29.2%及44.4%，后囊膜下混浊评分大于3分的患者占比依次为32.1%、58.3%及75.0%，皮质混浊程度分级和后囊膜下混浊程度分级分布的比较差异有统计学意义(H＝6.45，15.74；P<0.05)，但核性混浊程度差异无统计学意义(H＝3.78，P>0.05)。低紫外线组、中紫外线组及高紫外线组患者的BCVA依次为(0.84±0.78)最小分辨角对数视力(logMAR)、(1.53±0.85)logMAR及(1.81±0.86)logMAR，其组间分布比较差异有统计学意义(F＝17.46，P<0.05)；低紫外线组与中紫外线组，低紫外线组与高紫外线组两两比较，差异均有统计学意义(P_校正<0.05)。低紫外线组、中紫外线组及高紫外线组患者房水GPX含量依次为为(332.19±187.74)U/ml、(292.98±177.55)U/ml及(183.76±144.06)U/ml，组间比较差异有统计学意义(F＝8.16，P<0.05)。低紫外线组与高紫外线组，中紫外线组与高紫外线组两两比较差异有统计学意义(P_校正<0.05)。低紫外线组、中紫外线组及高紫外线组间IOP差异无统计学意义(F＝1.89，P>0.05)。50~<65组、≤65~<75组、≤75~<85组及≥85组评分≤3分的患者占比分别为74.4%、72.5%、54.5%及30.8%；50~<65组、≤65~<75组、≤75~<85组及≥85组评分3~5分的患者占比分别为20.9%、17.6%、33.3%及38.4%；50~<65组、≤65~<75组、≤75~<85组及≥85组评分≥5分的患者占比分别为4.7%、9.8%、12.1%及30.8%；不同年龄段ARC患者核性混浊程度分级分布比较差异有统计学意义(H＝12.11，P<0.05)。GPX含量与紫外线分级、皮质混浊程度分级、核性混浊程度分级、后囊膜下混浊程度分级及BCVA呈负相关且有统计学意义(r＝-0.33，-0.36，-0.18，-0.27，-0.27；P<0.05)，与年龄及IOP不相关且无统计学意义(r＝0.007，-0.005；P>0.05)。多元线性回归分析显示紫外线分级和皮质混浊程度分级是房水GPX含量的独立负向影响因素(β＝-50.45，-62.87，95%CI：-90.60~-10.31，-109.14~-16.60；t＝-2.49，-2.69；P<0.05)，而核性混浊程度分级和后囊膜下混浊程度分级不影响房水GPX含量(β＝16.56，-18.40，95%CI：-35.20~68.32，-59.03~20.02；t＝-0.63，-0.98；P>0.05)。

长期紫外线辐照会降低ARC患者房水的GPX含量，降低抗氧化应激能力；房水GPX含量与皮质混浊、后囊膜下混浊程度加重及视力下降相关；紫外线分级和皮质混浊程度是影响房水GPX含量的关键因素。

The aim of this study is to investigate the clinical characteristics of age-related cataract (ARC) patients in ultraviolet-irradiated areas and the role and mechanism of glutathione peroxidase (GPX) content in aqueous humor.

From July 2024 to July 2025, 140 ARC patients (140 eyes) who visited Peking University Third Hospital (Beijing), Tibet Autonomous Region People′s Hospital (Lhasa), Chifeng Songshan Hospital (Chifeng), Kezuo Banner Hospital (Tongliao) and Xilinhot Xilingol League Central Hospital were enrolled. Among them, there were 83 patients (83 eyes) female and 57 male (57 eyes) with a mean age of (69.7±9.3) years (ranging from 51 to 89 years ). According to the mean annual accumulated ultraviolet radiation from 2011—2014, subjects were classified into low-ultraviolet groups, medium-ultraviolet groups, and high-ultraviolet groups. According to age, patients were divided into the 50 to <65, ≥65 to <75, ≥75 to <85, and ≥85 groups. All ARC patients underwent routine preoperative cataract examinations before enrollment, including best-corrected visual acuity (BCVA), intraocular pressure (IOP), slit-lamp microscopy, and fundus examination. History of high myopia, diabetes, hypertension, and cardiovascular/cerebrovascular disease was queried and recorded. Continuous variables that passed tests for normality (age, BCVA, IOP, GPX, etc.) are described as ±s, between-group comparisons were performed using one-way ANOVA, and when differences were statistically significant, pairwise comparisons were further performed using the Bonferroni method. Comparisons between two groups for binary variables used the Mann-Whitney U test; comparisons among multiple groups for ordinal variables used the Kruskal-Wallis H test. Categorical data such as sex and laterality are presented as counts and percentages and compared using the chi-square test. Correlations between continuous variables and ordered categorical variables were assessed with Spearman′s correlation coefficient; correlations between continuous variables were assessed with Pearson′s correlation coefficient. Multivariate linear regression was used to explore factors influencing GPX.

The proportions of patients with cortical opacity >3 scores in the low, medium, and high UV groups were 23.2%, 29.2%, and 44.4%, respectively; the proportions with posterior subcapsular opacity scores >3 were 32.1%, 58.3%, and 75.0%, respectively. The distributions of cortical opacity grading and posterior subcapsular opacity grading differed significantly (H= 6.45, 15.74; P<0.05), whereas nuclear opacity grading showed no significant difference (H=3.78, P> 0.05). The BCVA in the low, medium, and high UV groups were (0.84±0.78) logarithmic minimum angle of resolution (logMAR), (1.53±0.85) logMAR, and (1.81±0.86) logMAR, respectively, with significant differences among groups (F=17.46, P<0.05); pairwise comparisons between low vs. medium and low vs. high UV groups were both statistically significant (adjusted P<0.05). Aqueous humor GPX levels in the low, medium, and high UV groups were (332.19±187.74) U/ml, (292.98±177.55) U/ml, and (183.76±144.06) U/ml, respectively, with significant differences among groups (F=8.16, P<0.05). Pairwise comparisons of low vs. high and medium vs. high UV groups were both significant (adjusted P< 0.05). There was no significant difference in IOP among the low, medium, and high UV groups (F=1.89, P>0.05). The proportion of patients with ≤ 3 scores in 50 to <65 group, ≤ 65 to <75 group, ≤ 75 to <85 group and ≥ 85 group were 74.4%, 72.5%, 54.5% and 30.8%, respectively; The proportion of patients with of 3 to 5 scores in 50 to <65 group, ≤ 65 to <75 group, ≤ 75 to <85 group and ≥ 85 group were 20.9%, 17.6%, 33.3% and 38.4%, respectively; The proportion of patients with ≥ 5 scores in 50 to <65 group, ≤ 65 to <75 group, ≤ 75 to <85 group and ≥ 85 group was > 4.7%, 9.8%, 12.1% and 30.8%, respectively. There were statistically significant differences in the grading distribution of nuclear opacity in arc patients at different ages ( H=12.11, P<0.05). GPX level was significantly negatively correlated with UV grading, cortical opacity grading, nuclear opacity grading, posterior subcapsular opacity grading, and BCVA (r= -0.33, -0.36, -0.18, -0.27, -0.27; P<0.05), and showed no significant correlation with age or IOP (r= 0.007, -0.005; P>0.05). Multiple linear regression showed that UV grading and cortical opacity grading were independent negative predictors of aqueous humor GPX level (β= -50.45, -62.87, 95%CI: -90.60 to -10.31, -109.14 to -16.60; t=-2.49, -2.69; P<0.05), whereas nuclear opacity grading and posterior subcapsular opacity grading did not significantly affect aqueous humor GPX level (β=16.56, -18.40, 95%CI: -35.20 to 68.32, -59.03 to 20.02; t=-0.63, -0.98; P>0.05).

Long-term exposure to UV radiation reduces aqueous humor GPX levels in ARC patients, impairs ocular antioxidant stress capacity, and is closely associated with aggravated cortical and posterior subcapsular opacities and visual acuity decline. UV radiation grade and cortical opacity grading are key factors affecting aqueous humor GPX levels.