目的：探讨视网膜静脉阻塞继发黄斑水肿(RVO-ME)患者房水中单核细胞趋化蛋白-1(MCP-1)水平与黄斑区微循环的关系。

方法：选择2022-07/2024-07在本院就诊的RVO-ME患者327例327眼为研究对象，根据是否复发分为未复发组291例291眼，复发组36例36眼。收集临床资料、黄斑区微循环指标、房水MCP-1水平。采用无序多分类Logistic回归分析在排除其他因素的影响后，分析MCP-1对RVO-ME复发的影响。利用立方样条分析黄斑区微循环指标及MCP-1表达水平与RVO-ME复发的剂量-反应关系。采用多元线性回归分析黄斑区微循环指标与MCP-1表达水平的关系，分析不同MCP-1表达水平在不同黄斑区微循环指标下复发的差异。Bootstrap法检验黄斑区微循环指标在MCP-1表达水平和RVO-ME复发中的中介效应。

结果：复发组患者的ME病程、玻璃体黄斑牵拉(VMT)发生率、纤维膜黄斑前膜(ERM)发生率、最佳矫正视力(BCVA)、视网膜内外层点状强反射(HRF)数量、注射抗VEGF次数及剂量显著高于未复发组(均P<0.05)，视网膜外界膜(ELM)完整性、视网膜椭圆体带(EZ)完整性显著低于未复发组(均P<0.05)。复发组患者的黄斑中心厚度(CFT)、黄斑中心凹厚度(CMT)、视网膜浅层毛细血管丛(SCP)血管密度、深层毛细血管丛(DCP)血管密度和MCP-1均高于未复发组(均P<0.05)，黄斑中心凹无血管区面积(FAZ)低于未复发组(P<0.05)。Logistic分析结果显示，调整混杂因素前后，MCP-1是RVO-ME复发的危险因素。黄斑区微循环指标及MCP-1表达水平与RVO-ME复发风险均呈明显的非线性剂量-反应关系(非线性检验，均P<0.001)。CFT、CMT、SCP和DCP血管密度与MCP-1表达水平均呈正相关(均P<0.05)，FAZ与MCP-1表达水平呈负相关(P<0.05)。随着CFT、CMT、SCP和DCP血管密度的增加以及FAZ的减小，MCP-1表达水平随之上升，并且RVO-ME复发风险呈现上升趋势，Q3(>28.47 pg/mL)组MCP-1的比例最高(P<0.05)。黄斑区微循环指标在MCP-1表达水平对RVO-ME复发情况之间发挥中介效应。

结论：房水中MCP-1水平是RVO-ME复发的危险因素，并且与黄斑区微循环关系密切，黄斑区微循环指标在MCP-1水平与RVO-ME复发之间存在中介效应。

METHODS:A total of 327 patients(327 eyes)with RVO-ME treated in our hospital from July 2022 to July 2024 were selected as the research objects. According to the recurrence or not, they were divided into non-recurrence group(291 cases, 291 eyes)and recurrence group(36 cases, 36 eyes). The clinical data, macular microcirculation index and MCP-1 level were collected. Unordered multinomial Logistic regression was used to analyze the effect of MCP-1 on RVO-ME recurrence after excluding the influence of other factors. Cubic spline was used to analyze the dose-response relationship between macular microcirculation indexes and MCP-1 expression level and RVO-ME recurrence. Multiple linear regression was used to analyze the relationship between macular microcirculation parameters and MCP-1 expression level, and the difference of MCP-1 expression level in recurrence under different macular microcirculation parameters was analyzed. Bootstrap method was used to test the mediating effect of macular microcirculation indexes on MCP-1 expression level and RVO-ME recurrence.

RESULTS: The course of ME, the incidence of vitreomacular traction(VMT), the incidence of fibrous membrane epiretinal membrane(ERM), best corrected visual acuity(BCVA), hyperreflective foci in the inner and outer retinal layers(HRF), the frequency and dose of anti-vascular endothelial growth factor(VEGF)injection in the recurrence group were significantly higher than those in the non-recurrence group(all P<0.05). The external limiting membrane(ELM)integrity and ellipsoid zone(EZ)integrity in the recurrence group were significantly worse than those in the non-recurrence group(all P<0.05). The central foveal thickness(CFT), central macular thickness(CMT), superficial capillary plexus(SCP)vascular density, deep capillary plexus(DCP)vascular density and MCP-1 in the recurrence group were higher than those in the non-recurrence group(all P<0.05), and the foveal avascular zone(FAZ)area was lower than that in the non-recurrence group(P<0.05). Logistic analysis showed that MCP-1 was a risk factor for RVO-ME recurrence before and after adjusting for confounding factors. There was a significant non-linear dose-response relationship between macular microcirculation indexes and MCP-1 expression and the risk of RVO-ME recurrence(non-linear test, all P<0.001). The vascular density of CFT, CMT, SCP and DCP was positively correlated with the expression level of MCP-1(all P<0.05), while the FAZ was negatively correlated with MCP-1 expression level(P<0.05). With the increase of vessel density in CFT, CMT, SCP and DCP, and the decrease of FAZ, the expression level of MCP-1 increased, and the risk of RVO-ME recurrence showed an upward trend. The proportion of MCP-1 in Q3(>28.47 pg/mL)group was the highest(P<0.05). Macular microcirculation parameters play a mediating effect between MCP-1 expression level and RVO-ME recurrence.

CONCLUSION: The level of MCP-1 in aqueous humor is positively correlated with RVO-ME recurrence, and it is closely related to macular microcirculation. Macular microcirculation has a mediating effect between MCP-1 level and RVO-ME recurrence.