糖尿病和心脏病是ANCA相关性血管炎中利妥昔单抗诱导急性血小板减少的关键预测因素
该研究基于日本全国多中心注册队列,纳入175例接受利妥昔单抗诱导治疗的新发显微镜下多血管炎或肉芽肿性多血管炎患者,评估利妥昔单抗诱导急性血小板减少(RIAT)的风险及危险因素。结果显示RIAT发生率为35.4%,糖尿病(OR=4.96)、心脏病(OR=3.57)、低血小板计数(每降低10^4/μl,OR=1.04)、血清肌酐升高(每升高1 mg/dl,OR=1.44)、低白蛋白(每降低1 g/dl,OR=2.33)和高KL-6(每升高100 U/ml,OR=1.09)为显著预测因素,预测模型C-statistic为0.817。
要点
- 175例接受利妥昔单抗诱导的AAV患者中,RIAT发生率为35.4%。
- 糖尿病(OR=4.96)和心脏病(OR=3.57)是RIAT的强危险因素。
- 包含上述因素的预测模型C-statistic为0.817。
核心结论
- RIAT是AAV患者利妥昔单抗诱导治疗的常见并发症,糖尿病和心脏病是强危险因素,预测模型可用于风险评估以优化治疗策略。
研究设计与样本
- 日本全国多中心注册队列研究,纳入175例接受利妥昔单抗诱导治疗的新发显微镜下多血管炎或肉芽肿性多血管炎患者。
- RIAT定义为利妥昔单抗治疗后28天内血小板计数降低。
- 采用逐步选择法的逻辑回归识别RIAT危险因素并开发预测模型,模型性能用C-statistic评估。
主要发现
- 175例接受利妥昔单抗的AAV患者中,RIAT发生率为35.4%。
- 糖尿病(OR=4.96)、心脏病(OR=3.57)、低血小板计数(每降低10^4/μl,OR=1.04)、血清肌酐升高(每升高1 mg/dl,OR=1.44)、低白蛋白(每降低1 g/dl,OR=2.33)和高KL-6(每升高100 U/ml,OR=1.09)为显著预测因素。
- 包含上述因素的预测模型C-statistic为0.817。
关键数值与推荐
- RIAT定义为利妥昔单抗治疗后28天内血小板计数降低。
- 糖尿病OR=4.96,心脏病OR=3.57,低血小板计数每降低10^4/μl OR=1.04,血清肌酐每升高1 mg/dl OR=1.44,白蛋白每降低1 g/dl OR=2.33,KL-6每升高100 U/ml OR=1.09。
- 预测模型C-statistic为0.817。
临床怎么用
- 预测模型可对接受利妥昔单抗诱导治疗的AAV患者进行RIAT风险评估,帮助临床医生优化治疗策略。
局限性
- 原文未述局限性
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: Rituximab (RTX) is a key treatment for ANCA-associated vasculitis (AAV), but RTX-induced acute thrombocytopenia (RIAT) remains a concern. We aimed to evaluate the risk and risk factors of RIAT in patients with AAV undergoing RTX induction therapy. METHODS: Patients with new-onset microscopic polyangiitis or granulomatosis with polyangiitis who received RTX in a nationwide multicentre registry in Japan were included. RIAT was defined by platelet count reductions within 28 days post-RTX. Risk factors for RIAT were identified by using logistic regression with stepwise selection, and prediction models were developed. Model performance was assessed using C-statistics. RESULTS: Among 175 patients with AAV receiving RTX, RIAT occurred in 35.4% of them. Diabetes mellitus (odds ratio [OR] = 4.96), cardiac disease (OR = 3.57), low platelet count (OR = 1.04 per 104/μl decrease), increased serum creatinine (OR = 1.44 per 1 mg/dl increase), low albumin (OR = 2.33 per 1 g/dl decrease), and high KL-6 (OR = 1.09 per 100 U/ml increase) were identified as significant predictors. A predictive model incorporating these factors achieved a C-statistic of 0.817. CONCLUSION: RIAT is a frequent complication of RTX induction in patients with AAV, with diabetes mellitus and cardiac disease being strong risk factors. Our predictive model enables risk assessment for RIAT, allowing clinicians to optimize treatment strategies.
引用
- Akao S, Shimizu M, Maruo K, et al. Diabetes mellitus and cardiac disease as key predictors of rituximab-induced acute thrombocytopenia in ANCA-associated vasculitis. Rheumatology (Oxford). 2026. PMID: 41071071. doi:10.1093/rheumatology/keaf529
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。