PDF(Pubmed)
Abstract：
OBJECTIVE: To investigate the impact of disease activity and treatment with disease-modifying antirheumatic drugs (DMARDs) on all-cause mortality in patients with rheumatoid arthritis and prevalent interstitial lung disease (RA-ILD).
METHODS: Patients with RA-ILD were selected from the biologics register Rheumatoid Arthritis: Observation of Biologic Therapy (RABBIT). Using time-varying Cox regression, the association between clinical measures and mortality was investigated. The impact of DMARDs was analysed by (1) Cox regression considering cumulative exposure (ie, treatment months divided by total months) and (2) time-varying Cox regression as main approach (treatment exposures at monthly level).
RESULTS: Out of 15 566 participants, 381 were identified as RA-ILD cases with 1258 person-years of observation and 2.6 years median length of follow-up. Ninety-seven patients (25.5%) died and 34 (35.1%) of these were not receiving DMARD therapy at the time of death. Higher inflammatory biomarkers but not swollen and tender joint count were significantly associated with mortality. Compared with tumour necrosis factor inhibitors (TNFi), non-TNFi biologic DMARDs (bDMARDs) exhibited adjusted HRs (aHRs) for mortality below 1, lacking statistical significance. This finding was stable in various sensitivity analyses. Joint aHR for non-TNFi biologics and JAKi versus TNFi was 0.56 (95% CI 0.33 to 0.97). Receiving no DMARD treatment was associated with a twofold higher mortality risk compared with receiving any DMARD treatment, aHR 2.03 (95% CI 1.23 to 3.35).
CONCLUSIONS: Inflammatory biomarkers and absence of DMARD treatment were associated with increased risk of mortality in patients with RA-ILD. Non-TNFi bDMARDs may confer enhanced therapeutic benefits in patients with RA-ILD.
METHODS: Patients with RA-ILD were selected from the biologics register Rheumatoid Arthritis: Observation of Biologic Therapy (RABBIT). Using time-varying Cox regression, the association between clinical measures and mortality was investigated. The impact of DMARDs was analysed by (1) Cox regression considering cumulative exposure (ie, treatment months divided by total months) and (2) time-varying Cox regression as main approach (treatment exposures at monthly level).
RESULTS: Out of 15 566 participants, 381 were identified as RA-ILD cases with 1258 person-years of observation and 2.6 years median length of follow-up. Ninety-seven patients (25.5%) died and 34 (35.1%) of these were not receiving DMARD therapy at the time of death. Higher inflammatory biomarkers but not swollen and tender joint count were significantly associated with mortality. Compared with tumour necrosis factor inhibitors (TNFi), non-TNFi biologic DMARDs (bDMARDs) exhibited adjusted HRs (aHRs) for mortality below 1, lacking statistical significance. This finding was stable in various sensitivity analyses. Joint aHR for non-TNFi biologics and JAKi versus TNFi was 0.56 (95% CI 0.33 to 0.97). Receiving no DMARD treatment was associated with a twofold higher mortality risk compared with receiving any DMARD treatment, aHR 2.03 (95% CI 1.23 to 3.35).
CONCLUSIONS: Inflammatory biomarkers and absence of DMARD treatment were associated with increased risk of mortality in patients with RA-ILD. Non-TNFi bDMARDs may confer enhanced therapeutic benefits in patients with RA-ILD.
摘要：
目的：研究疾病活动性和抗风湿药（DMARDs）治疗对类风湿性关节炎和间质性肺病（RA-ILD）患者全因死亡率的影响。
方法：RA-ILD患者选自生物制剂注册类风湿性关节炎：生物治疗观察（RABBIT）。使用时变Cox回归，研究了临床指标与死亡率之间的关系.通过（1）考虑累积暴露的Cox回归分析了DMARDs的影响（即，治疗月除以总月）和（2）时变Cox回归作为主要方法（每月水平的治疗暴露）。
结果：在15566名参与者中，381例被确定为RA-ILD病例，观察人数为1258人-年，中位随访时间为2.6年。97例患者（25.5%）死亡，其中34例（35.1%）在死亡时未接受DMARD治疗。较高的炎症生物标志物，但不肿胀和压痛关节计数与死亡率显着相关。与肿瘤坏死因子抑制剂（TNFi）相比，非TNFi生物DMARDs(bDMARDs)在死亡率低于1时显示校正HR(aHRs),无统计学意义.这一发现在各种灵敏度分析中是稳定的。非TNFi生物制剂和JAKI与TNFi的联合aHR为0.56（95％CI0.33至0.97）。与接受任何DMARD治疗相比，未接受DMARD治疗的死亡风险高两倍。AHR2.03(95%CI1.23至3.35)。
结论：炎症生物标志物和缺乏DMARD治疗与RA-ILD患者死亡风险增加相关。非TNFibDMARDs可在RA-ILD患者中赋予增强的治疗益处。
方法：RA-ILD患者选自生物制剂注册类风湿性关节炎：生物治疗观察（RABBIT）。使用时变Cox回归，研究了临床指标与死亡率之间的关系.通过（1）考虑累积暴露的Cox回归分析了DMARDs的影响（即，治疗月除以总月）和（2）时变Cox回归作为主要方法（每月水平的治疗暴露）。
结果：在15566名参与者中，381例被确定为RA-ILD病例，观察人数为1258人-年，中位随访时间为2.6年。97例患者（25.5%）死亡，其中34例（35.1%）在死亡时未接受DMARD治疗。较高的炎症生物标志物，但不肿胀和压痛关节计数与死亡率显着相关。与肿瘤坏死因子抑制剂（TNFi）相比，非TNFi生物DMARDs(bDMARDs)在死亡率低于1时显示校正HR(aHRs),无统计学意义.这一发现在各种灵敏度分析中是稳定的。非TNFi生物制剂和JAKI与TNFi的联合aHR为0.56（95％CI0.33至0.97）。与接受任何DMARD治疗相比，未接受DMARD治疗的死亡风险高两倍。AHR2.03(95%CI1.23至3.35)。
结论：炎症生物标志物和缺乏DMARD治疗与RA-ILD患者死亡风险增加相关。非TNFibDMARDs可在RA-ILD患者中赋予增强的治疗益处。
