血清高敏C反应蛋白水平对肺栓塞预测价值的研究

Relationship Between Baseline Levels of High Sensitivity C-reactive Protein With the Morbidity of Pulmonary Embolism

目的:探讨基线血清高敏C反应蛋白(hs-CRP)水平对肺栓塞的预测价值。方法:本研究采用前瞻性队列研究方法,以参加2006-07至2007-10健康查体的l0l 510名开滦集团职工作为观察队列,符合入选标准者94 314例(男75 252例,女19 062例),按基线hs-CRP水平四分位数分为第一四分位数组(<0.3l mg/L,n=25 592)、第二四分位数组0.3l^0.77 mg/L,n=21 746)、第三四分位数组(0.78~1.9 mg/L,n=23 504)和第四四分位数组(≥2.0 mg/L,n=23 472),随访(81.5±4.76)个月,收集肺栓塞的发病信息,用多变量Cox比例风险回归分析基线血清hs-CRP水平对肺栓塞的预测价值。肺栓塞参照2001年中华医学会呼吸病学分会制定的肺栓塞诊断与治疗指南,采用计算机断层摄影肺动脉造影成像法进行诊断。结果:肺栓塞累计总发病率为0.15%,女性高于男性,但差异无统计学意义(P>0.05)。随着基线hs-CRP水平升高,肺栓塞发生率逐渐增高,差异有统计学意义(P<0.05);多变量Cox比例风险回归分析表明,校正年龄、性别、吸烟、体重指数、收缩压、舒张压、甘油三酯、空腹血糖、中性粒细胞、白细胞及心率后,基线hs-CRP最高四分位数组发生肺栓塞的相对危险为最低四分位数组的2.84倍。随着基线hs-CRP水平升高,研究对象的平均年龄、体重指数、收缩压、空腹血糖、中性粒细胞水平逐渐增高,差异均有统计学意义(P均<0.05);舒张压和甘油三酯水平逐渐升高、心率逐渐增快,但这些指标在第三四分位组与第四四分位数组相比差异无统计意义(P均>0.05)。白细胞计数在前三组呈递增趋势,差异有统计学意义,但在第四四分位数组呈下降趋势,与第三四分位数组相比差异无统计学意义(P>0.05)。结论:基线hs-CRP可预测肺栓塞的发病风险,hs-CRP水平较高提示肺栓塞发生风险增加,是肺栓塞的强预测因子。

Objective： To evaluate the predictive value for baseline levels of high sensitivity C-reactive protein （hs-CRP） in morbidity of pulmonary embolism （PE）. Methods： We conducted a prospective cohort study of 101510 subjects in Kailuan Group by regular physical examination from 2006-07 to 2007-10, and 94314 subjects were enrolled by relevant criteria including 75252 male and 19062 female. The baseline levels of hs-CRP were divided by quartile levels as （1） hs-CRP〈0.31 mg/L, n=25592, （2） hs-CRP （0.31-0.77） mg/L, n=21746, （3） hs-CRP （0.78-1.9） mg/L, n=23504 and （4） hs-CRP≥2.0 mg/L, n=23472. The subjects were followed-up for （81.5 ± 4.76） months, the morbidity of PE was collected and the predictive value of baseline levels of hs-CRP for PE occurrence was evaluate by multivariable Cox proportional hazard regression analysis. Results： The total PE morbidity was 0.15%, the female subjects were similar with male subjects, P〉0.05. As the increased baseline level of hs-CRP, PE occurrence was elevated accordingly, P〈0.05. Multivariable Cox proportional hazard regression analysis presented that with adjusted age, gender, smoking, body mass index （BMI）, systolic blood pressure （SBP）, diastolic blood pressure （DBP）, triglyceride （TG）, fasting blood glucose （FBG）, neutrophile granulocyte （NEU）, white blood cells （WBC）and heart rate （HR）, the subjects in the highest quartile group had 2.84 times higher risk for PE occurrence than the subjects in the lowest quartile group. With the elevated baseline level of hs-CRP, the subjects＇ mean age, BMI, SBP, FBG and NEU levels increased accordingly, all P〈0.05; while DBP, TG and HR were similar between quartile （3） and quartile （4） groups, all P〉0.05. WBC counts had the increasing trend in quartile （1）, （2）, （3） groups and had the decreasing trend in quartile （4） group, while it was similar between quartile （3） and quartile （4） groups, P〉0.05. Conclusion： Baseline hs-CRP level may predict the risk of PE morbidity; the increased hs-CRP level could be used as one of the predictors for PE occurrence.