目的: 回顾性分析了解血脂异常与类风湿关节炎肺间质病变(rheumatoid arthritis associated interstitial lung disease, RA-ILD)的相关性。方法: 收集2015年1月至2020年7月期间在北京大学深圳医院风湿免疫科住院符合《2010年美国风湿病学会/欧洲风湿病防治联合会类风湿关节炎(rheumatoid arthritis, RA)分类标准》的患者病例资料，按照有无合并肺间质病变(interstitial lung disease, ILD)分组，对一般资料、临床特征及检验检查等数据单因素检验后将差异有意义的因素纳入Logistics多因素回归分析。根据有无血脂异常分组，分析血脂异常与RA-ILD的发生率、发生的中位时间及临床表现等相关性。结果: 共纳入737例RA患者，其中282例(38.26%)发生ILD，从开始出现RA相关的临床症状至发生ILD的中位时间为13(95%CI 11.33~14.67)年。多因素Logistic回归分析结果显示：低密度脂蛋白胆固醇(low-density lipoprotein cholesterol, LDL-C)是RA-ILD可能的危险因素(OR 1.452，95%CI 1.099~1.918，P=0.009)，高密度脂蛋白胆固醇(high-density lipoprotein cholesterol，HDL-C)是RA-ILD可能的保护因素(OR 0.056，95%CI 0.025~0.125，P < 0.001)。在737例RA患者中，LDL-C升高及HDL-C降低的RA患者ILD的发生率均分别高于LDL-C正常及HDL-C正常的患者(分别为57.45% vs. 36.96%，P < 0.001；47.33% vs. 33.81%，P < 0.001)。HDL-C降低的RA患者出现ILD的中位时间明显短于HDL-C正常的RA患者[10.0(95%CI 9.33~10.67)年vs. 17.0(95%CI 14.58~19.42)年，P < 0.001]。RA患者的HDL-C水平与疾病活动呈负相关。RA-ILD患者中，HDL-C降低的患者胸部HRCT出现寻常型间质性肺炎(usual interstitial pneumonia, UIP)的比例高于HDL-C正常的患者(60.00% vs. 53.29%，P=0.002)，LDL-C升高的RA-ILD患者出现用力肺活量(forced vital capacity，FVC)下降的发生率高于LDL-C正常的RA-ILD患者(50.00% vs.21.52%，P=0.015)，HDL-C降低的RA-ILD患者出现FVC及一氧化碳弥散量(diffusing capacity of the lung for carbon monoxide, DLCO)下降的发生率高于HDL-C正常的RA-ILD患者(分别为26.92% vs.16.18%，P=0.003；80.76% vs. 50.00%，P=0.010)。结论: LDL-C可能是RA-ILD的潜在危险因素；HDL-C可能是RA-ILD的潜在保护因素；HDL-C水平与RA病情活动呈负相关；HDL-C降低的RA患者出现ILD的时间更早。

Objective: To study the correlation between dyslipidemia and rheumatoid arthritis associa-ted interstitial lung disease (RA-ILD) by retrospective analysis of the clinical data. Methods: The clinical data of patients with rheumatoid arthritis (RA), who were hospitalized in the Department of Rheumatism and Immunology of Peking University Shenzhen Hospital from January 2015 to July 2020 and fulfilled the criteria of the 2010 Rheumatoid Arthritis Classification Criteria established by American College of Rheumatology/European League Against Rheumatism collaborative initiative, were collected and analyzed. Results: There were 737 RA patients included, of whom 282(38.26%)were with interstitial lung disease (ILD). The median time from the onset of the first RA-related clinical symptoms to the onset of ILD was 13 years (95%CI 11.33-14.67). By multivariate Logistic regression analysis, we found that low-density lipoprotein cholesterol (LDL-C) was an independent risk factor for RA-ILD (OR 1.452, 95%CI 1.099-1.918, P=0.009), whereas high-density lipoprotein cholesterol (HDL-C) was a protective factor for RA-ILD (OR 0.056, 95%CI 0.025-0.125, P < 0.001). The RA patients with high LDL-C or low HDL-C had higher incidence of ILD than that of the RA patients with normal LDL-C or HDL-C(57.45% vs. 36.96%, P < 0.001; 47.33% vs. 33.81%, P < 0.001, respectively). The median time of ILD onset in the RA patients with low HDL-C was shorter than that of the RA patients with normal HDL-C [10.0(95%CI 9.33-10.67)years vs.17.0 (95%CI 14.58-19.42) years, P < 0.001]. HDL-C level was negatively correlated with disease activity. Among the RA-ILD patients, the patients with low HDL-C had higher percentage of usual interstitial pneumonia (UIP) then that of the patients with normal HDL-C (60.00% vs. 53.29%, P=0.002). The RA-ILD patients with high LDL-C had higher incidence rate of decrease in forced vital capacity (FVC) than that of the RA-ILD patients with normal LDL-C (50.00% vs. 21.52%, P=0.015). The RA-ILD patients with low HDL-C had higher incidence rate of decrease in FVC (26.92% vs. 16.18%, P=0.003) and carbon monoxide diffusion (80.76% vs. 50.00%, P=0.010) than that of RA-ILD patients with normal HDL-C. Conclusion: LDL-C was possibly a potential independent risk factor for RA-ILD. HDL-C was possibly a potential protective factor for RA-ILD. HDL-C level was negatively correlated with disease activity of RA. The median time of ILD onset in the RA patients with low HDL-C was significantly shorter than that of the RA patients with normal HDL-C.