探讨基线房水细胞因子水平在糖尿病性黄斑水肿(DME)患者预测抗血管内皮生长因子(VEGF)治疗疗效中的价值。

收集2021年10月至2023年7月在沈阳爱尔卓越眼科医院就诊DME患者65例(65只眼)的病例资料。其中，男性36例(36只眼)，女性29例(29只眼)；年龄29～79岁，平均年龄(54.5±11.4)岁。所有患者均接受3次阿柏西普玻璃体腔注药治疗。根据治疗3个月后中央黄斑厚度(CMT)下降水平分为应答组34例(34只眼)和无应答组31例(31只眼)。比较纳入患者在注药前和三次注药后的最佳矫正视力(BCVA)、眼压(IOP)、CMT及房水细胞因子水平变化情况。BCVA、年龄、糖尿病病程、IOP等计量资料若服从正态分布，以±s表示，组间比较采用独立样本t检验，组内前后比较采用配对t检验；不服从正态分布的计量资料以中位数和四分位数表示，采用Mann－Whitney U检验比较；性别以例数和百分比表示计数资料，组间比较采用χ²检验。采用多因素Logistic回归分析影响抗VEGF药物无应答的危险因素。不同细胞因子与CMT降幅的相关性采用Spearman相关性分析。

治疗前后患眼BCVA分别为(1.10±0.27)最小分辨视角的对数(logMAR)和(0.61±0.30)logMAR；治疗前后患眼CMT分别为(583.54±108.05)μm和(419.95±80.01)μm，其差异均有统计学意义(t＝12.166，11.447；P<0.05)。治疗前后患眼IOP分别为(15.98±2.68)mmHg(1 mmHg＝0.133 kPa)和(15.88±2.48)mmHg，其差异无统计学意义(t＝0.736，P>0.05)。无应答组患眼的基线房水VEGF、细胞间黏附因子(ICAM)及白细胞介素(interleukin，IL)－8分别为114.00(88.75，156.85)pg/ml、1891.30(1124.60，2673.00)pg/ml及57.70(49.40，76.15)pg/ml；应答组患眼分别为190.40(147.50，333.70)pg/ml、6821.80(4387.15，9079.75)pg/ml及26.50(19.70，40.90)pg/ml，其差异均有统计学意义(Z＝3.612，6.291，5.700；P<0.05)。无应答组患眼的房水碱性成纤维细胞生长因子(bFGF)、IL－6及血管细胞黏附因子(VCAM)水平分别为8.90(4.80，22.05)pg/ml、61.90(19.35，134.60)pg/ml及2528.60(919.65，3353.30)pg/ml；应答组分别为13.80(4.20，31.95)pg/ml、76.00(31.70，235.9)pg/ml及2480.60(1086.55，3656.00)pg/ml，差异无统计学意义(Z＝0.585，1.596，0.525；P>0.05)。多因素Logistic回归分析患者的年龄、病程、糖化血红蛋白、基线视力、基线IOP、基线CMT、房水VEGF、bFGF、IL－6、VCAM、IL－8及ICAM对抗VEGF治疗应答的影响，但未发现以上任何因素是其独立危险因素(β＝－0.466，－2.012，6.723，－8.634，3.664，－0.009，0.049，－0.109，－0.032，0.007，－1.068，0.015；P>0.05)。经Spearman相关性分析，VEGF和ICAM水平与CMT降幅呈正相关性具有统计学意义(r＝0.262、0.750，P<0.05)；IL－8水平与CMT降幅呈负相关性具且有统计学意义(r＝－0.628，P<0.05)。bFGF、IL－6以及VCAM水平与CMT降幅无相关性不具有统计学意义(r＝0.127，0.098，－0.241；P>0.05)。

房水VEGF、ICAM－1以及IL－8水平与治疗应答相关性，可用于预测DME患者抗VEGF的治疗效果。

The aim of this study is to analyze the correlation between baseline cytokine levels in the aqueous humor and the response to anti-vascular endothelial growth factor (VEGF) treatment in DME patients.

The patients with diabetic macular edema (DME) who received intravitreal injection of Aflibercept for 3 times in Shenyang Aier Excellence Eye Hospital from October 2021 to July 2023 were collected. Among them, there were 36 males (36 eyes) and 29 females (29 eyes)with an average age of (54.5±11.4) years (ranged from 29 to 79) years old. According to the decrease of central macular thickness (CMT) after three months of treatment, all patients were divided into response group and non-response group. Best corrected visual acuity (BCVA), intraocular pressure (IOP), CMT, and the aqueous cytokine levels of patients before and after treatment for 3 months were detected. BCVA, age, course of diabetes, IOP and other measurement data, after the normal test, if conformed to the normal distribution, were expressed as ±s, and compared by by independent sample t test for the inter group , and paired t test for the intra group. Quantitative data that did not follow a normal distribution were represented as median and interquartile (non interquartile range) and compared using Mann-Whitney U test. Gender was represented by cases and percentages, and compared using χ². Using multiple logistic regression analysis was to explore potential risk factors for non response to anti VEGF drugs. The correlation between different cytokines and the decrease in CMT was used Spearman correlation analysis.

The BCVA of the affected eye before and after treatment were (1.10±0.27) logarithm of the minimum angle of resolution (logMAR) and (0.61±0.30) logMAR, respectively. The CMT of the affected eye before and after treatment were (583.54±108.05)μm, and (419.95±80.01) μm, respectively . The differences were statistically significant (t=12.166, 11.447; P<0.05). The IOP of the affected eye before and after treatment were (15.98±2.68) mmHg (1 mmHg=0.133 kPa) and (15.88±2.48) mmHg, respectively, with no statistically significant difference (t=0.736, P>0.465). The baseline aqueous humor VEGF, intercellular adhesion molecule (ICAM), and interleukin (IL)-8 levels in the non-response group were 114.00(88.75, 156.85)pg/ml, 1891.30(1124.60, 2673.00)pg/ml and 57.70(49.40, 76.15)pg/ml, respectively, while those in the responsive group were 190.40(147.50, 333.70)pg/ml, 6821.80(4387.15, 9079.75)pg/ml and 26.50(19.70, 40.90)pg/ml, with statistical significance between them (Z=3.612, 6.291, 5.700; P<0.05). The levels of basic fibroblast growth factor (bFGF), IL-6, and vascular cell adhesion molecule (VCAM) in aqueous humor of the non-response group were 8.90 (4.80, 22.05)pg/ml, 61.90 (19.35, 134.60)pg/ml, and 2528.60 (919.65, 3353.30)pg/ml, respectively, while those of the responsive group were 13.80 (4.20, 31.95)pg/ml, 76.00 (31.70, 235.9)pg/ml, and 2480.60 (1086.55, 3656.00)pg/ml, with no statistically significant differences (Z=0.585, 1.596, 0.525; P>0.05). The age, course of disease, glycated hemoglobin, baseline visual acuity, baseline IOP, baseline CMT, as well as aqueous humor VEGF, bFGF, IL-6, VCAM, IL-8, and ICAM were used for multivariate logistic regression analysis to explore impact factors for anti VEGF treatment response, but no independent risk factors were found (β=-0.466, -2.012, 6.723, -8.634, 3.664, -0.009, 0.049, -0.109, -0.032, 0.007, -1.068, 0.015; P>0.05). According to Spearman correlation analysis, the levels of VEGF and ICAM were positively correlated with the decrease in CMT, and there was statistical significance (r=0.262, 0.750, P<0.05). The level of IL-8 was negatively correlated with the decrease in CMT with a statistical significance (r=-0.628, P<0.05). The levels of bFGF, IL-6, and VCAM were not correlated with CMT decrease without a statistically significance (r=0.127, 0.098, -0.241; P>0.05).

The levels of VEGF, ICAM-1, and IL-8 in aqueous humor are correlated with treatment response, which can be used to predict the therapeutic effect of anti VEGF in DME patients.