类风湿关节炎患者并发骨质疏松的列线图预测模型的构建

Construction of a column chart prediction model for osteoporosis in patients with rheumatoid arthritis

目的分析类风湿关节炎(rheumatoid arthritis,RA)患者并发骨质疏松(osteoporosis,OP)的危险因素并构建Nomogram预测模型。方法选取2022年7月~至2023年12月青海省中医院收治的RA患者312例作为研究对象,并根据患者是否并发OP分为OP组(n=143)和非OP组(n=169)。比较两组患者一般临床资料,多因素Logistic回归分析筛选影响RA患者并发OP的独立风险因素,R软件“rms”包构建Nomogram预测模型。结果两组间患者年龄、体质量指数(body mass index,BMI)、钙摄入>600 mg/d、使用糖皮质激素史、25-羟维生素D[25-hydroxyvitamin D,25(OH)D]、骨碱性磷酸酶、甲状旁腺激素、白细胞介素(interleukin,IL)-6、类风湿因子、Sharp评分、28个关节疾病活动度评分(disease activity score-28,DAS28)等差异均有统计学意义(P<0.05)。年龄>63岁、使用糖皮质激素史、25(OH)D≤15.85 ng/mL、甲状旁腺激素>36.54 pg/mL、IL-6>22.15 pg/mL、类风湿因子>97.87 IU/mL、Sharp评分>79分、DAS28评分>6分是RA患者并发OP的独立危险因素(P<0.05),基于此构建的预测模型C指数为0.949(0.924~0.974),且校正曲线趋近于理想曲线。决策曲线结果显示,风险阈值>0.06时,此预测模型所提供的临床净收益较高。结论年龄较大、有使用糖皮质激素史、25(OH)D较低、甲状旁腺激素较高、IL-6较高、类风湿因子较高、Sharp评分较高、DAS28评分较高是RA患者并发OP的独立危险因素,基于此构建的Nomogram预测模型能有效预测RA患者并发OP的高风险群体。

Objective To analyze the risk factors of osteoporosis(OP)in patients with rheumatoid arthritis(RA)and construct a Nomogram prediction model.Methods A total of 312 RA patients admitted to Qinghai Provincial Traditional Chinese Medicine Hospital from July 2022 to December 2023 were induded,and they were divided into the OP group(n=143)and non-OP group(n=169)based on whether they had concurrent OP.The general clinical data of two groups of patients were compared,and multivariate logistic regression analysis was conducted to screen independent risk factors affecting the concurrent OP in RA patients.A Nomogram prediction model was constructed using the“rms”package of R software.Results The difference in age,body mass index(BMI),calcium intake>600 mg/d,history of glucocorticoid use,25 hydroxyvitamin D[25(OH)D],bone alkaline phosphatase,parathyroid hormone,interleukin-6(IL-6),rheumatoid factor,Sharp score,and 28 joint disease activity scores(DAS28 score)between the two groups of patients were statistically significant(P<0.05).More than 63 years old,history of glucocorticoid use,25(OH)D≤15.85 ng/mL,parathyroid hormone>36.54 pg/mL,IL-6>22.15 pg/mL,rheumatoid factor>97.87 IU/mL,Sharp score>79 points,and DAS28 score>6 points were independent risk factors for OP in RA patients(P<0.05).The C-index of the predictive model constructed based on above data was 0.949(0.924-0.974),and the correction curve tended to the ideal curve.The decision curve results show that when the risk threshold was greater than 0.06,the clinical net benefit provided by this prediction model was higher.Conclusion Older age,a history of using glucocorticoids,vitamin D deficiency,higher levels of parathyroid hormone,IL-6 and rheumatoid factor,higher Sharp score,and higher DAS28 score are independent risk factors for OP in RA patients.The Nomogram prediction model constructed based on the results can effectively predict the high-risk population of OP in RA patients.