中华急诊医学杂志  2022, Vol. 31 Issue (8): 1056-1060   DOI: 10.3760/cma.j.issn.1671-0282.2022.08.007
动脉血二氧化碳分压联合Wells评分有助于预测急性肺栓塞
左冬晶 , 曹玉丹 , 张艳慧 , 赵丽新 , 腾飞 , 郭树彬 , 何新华     
首都医科大学附属北京朝阳医院急诊医学中心,心肺脑复苏北京市重点实验室,北京 100020
摘要: 目的 探索动脉血二氧化碳分压(partial pressure of carbon dioxide, PaCO2)联合Wells评分预测急性肺栓塞(acute pulmonary embolism, APE)的临床价值。方法 回顾性收集2016年1月1日至2021年8月31日期间就诊于首都医科大学附属北京朝阳医院急诊科疑似APE的患者,以计算机断层扫描肺血管造影术(computed tomography pulmonary angiography, CTPA)结果阳性者为肺栓塞(pulmonary embolism, PE)组,阴性者为非肺栓塞(non-pulmonary embolism, Non-PE)组,对两组人口统计学特征、症状、生命体征、基础疾病、静脉血栓危险因素、动脉血气分析、Wells评分等进行统计学分析比较,并对PaCO2联合Wells评分预测APE的临床效度进行评价。结果 共筛选疑似APE患者1 869例,最终入选1 492例,PE组537例,Non-PE组955例。胸痛、呼吸困难、单侧下肢水肿、肺栓塞或静脉血栓史、3月内手术或制动史、3月内骨折史、活动期恶性肿瘤、Wells评分升高及PaCO2降低在PE组出现的频率显著高于Non-PE组(均P < 0.05)。Wells评分的ROC曲线下面积(AUC)为0.784(95%CI: 0.758~0.810),预测APE的敏感度为61.64%,特异度为88.48%;PaCO2降低的AUC为0.679(95%CI: 0.651~0.707),预测APE的敏感度为79.89%,特异度为55.92%;PaCO2降低联合Wells评分的AUC为0.837(95%CI: 0.816~0.858),预测APE的敏感度为74.12%,特异度为77.07%。PaCO2降低联合Wells评分的AUC显著大于Wells评分的AUC(P < 0.001)和PaCO2降低的AUC(P < 0.001)。结论 PaCO2降低联合Wells评分预测APE的效能优于单独应用其中一种,是对APE患者筛查的有益补充,同时也将对急诊减少CTPA的盲目使用起到一定帮助。
关键词: 二氧化碳分压    Wells评分    急性肺栓塞    急诊    
Arterial partial pressure of carbon dioxide combined with Wells score helps predict acute pulmonary embolism
Zuo Dongjing , Cao Yudan , Zhang Yanhui , Zhao Lixin , Teng Fei , Guo Shubin , He Xinhua     
Emergency Medicine Clinical Research Center, Beijing Chao-Yang Hospital, Capital Medical University, Beijing Key Laboratory of Cardiopulmonary Cerebral Resuscitation, Beijing 100020, China
Abstract: Objective To explore the clinical value of arterial partial pressure of carbon dioxide (PaCO2) combined with Wells score in predicting acute pulmonary embolism (PE). Methods Patients with suspected acute PE admitted to Emergency Department of Beijing Chaoyang Hospital, Capital Medical University from January 1, 2016 to August 31, 2021 were screened. Patients with positive computed tomography pulmonary angiography (CTPA) results were classified as the PE group, and those with negative CTPA results were classified as the non-PE group. Demographic characteristics, symptoms, vital signs, underlying diseases, risk factors for venous thrombosis, arterial blood gas analysis and Wells scores were statistically analyzed and compared between the two groups, and the clinical efficacy of PaCO2 combined with Wells score in predicting acute PE was evaluated. Results A total of 1 869 patients with suspected acute PE were screened, and 1 492 patients were finally selected. There were 537 cases in the PE group and 955 cases in the non-PE group. The frequency of chest pain, dyspnea, unilateral lower limb edema, history of PE or deep venous thrombosis, history of surgery or immobilization within 3 months, history of fracture within 3 months, active malignant tumor, elevated Wells score and reduced PaCO2 in the PE group was significantly higher than that in the non-PE group (all P < 0.05). The area under receiver operating characteristic (ROC) curve (AUC) of Wells score was 0.784 (95% CI: 0.758-0.810), and the sensitivity and specificity of predicting acute pulmonary embolism were 61.64% and 88.48%, respectively. The AUC of reduced PaCO2 was 0.679 (95% CI: 0.651-0.707), and the sensitivity and specificity of predicting acute pulmonary embolism were 79.89% and 55.92%, respectively. The AUC of reduced PaCO2 combined with Wells score was 0.837 (95% CI: 0.816-0.858), and the sensitivity and specificity of predicting acute pulmonary embolism were 74.12% and 77.07%, respectively. The AUC of reduced PaCO2 combined with Wells score was significantly greater than the AUC of Wells score (P < 0.001) and the AUC of reduced PaCO2 (P < 0.001). Conclusions The efficacy of PaCO2 reduction combined with Wells score in predicting acute PE was superior to that of either of them alone. This was a beneficial supplement to the screening of patients with acute PE, and would also help reduce the abuse of CTPA in the emergency department.
Key words: Partial pressure of carbon dioxide    Wells score    Acute pulmonary embolism    Emergency    

急性肺动脉血栓栓塞症,简称急性肺栓塞(acute pulmonary embolism, APE),由栓子阻塞肺动脉及其分支所致,是目前继急性心肌梗死、卒中之后的第三位致死性血管性疾病[1],只有7% APE致死的患者能够在死亡前明确诊断[2]。超过85%的APE患者是通过计算机断层扫描肺血管造影术(computed tomography pulmonary angiography, CTPA)确诊的[3],且阳性检出率多在8%左右[4-5]。较低的检出率反映出CTPA的过度使用,由此带来造影剂肾病、射线相关疾病等诸多弊端[6]。为减少这种情况,需要尽可能提高APE的诊断策略。

Wells评分是目前最为权威的APE筛查手段[7],但评分中缺乏快速有效的实验室依据,降低了其筛查效度。随着科学技术的进步,D-dimer、血气分析等快速床旁检测都可以对APE的诊断做出提示。已有研究证实D-dimer可以提高Wells评分诊断APE的效能[8],但高龄、肿瘤、感染、肾功能不全等可以使非APE患者D-dimer水平升高,导致假阳性[9]。鉴于此,本团队继续探索临床上快速可获得的床旁检测结果对Wells评分提供有效的补充。临床中发现很多APE患者的动脉血气提示二氧化碳分压(partial pressure of carbon dioxide, PaCO2)降低,因此本研究探讨PaCO2降低联合Wells评分对APE的诊断效能。

1 资料与方法 1.1 一般资料

本研究为回顾性研究,收集2016年1月1日至2021年8月31日期间就诊于首都医科大学附属北京朝阳医院急诊科疑似APE的患者资料。急诊科对疑似APE患者的诊断临床路径均依据《肺血栓栓塞症诊治与预防指南》[10]中诊断流程进行筛查,直至最终完善CTPA或通气/灌注扫描。

纳入标准:(1)首诊医师或呼吸科二线会诊医师根据患者咯血、胸痛、呼吸困难、晕厥、单侧下肢水肿等症状体征及D-dimer升高等实验室结果初步诊断疑似APE; (2)CTPA检查正式报告证实存在APE。排除标准:(1)因CTPA相对禁忌证,如碘剂过敏、肾功能不全、妊娠等,未行CTPA检查者; (2)通过其他方式(肺通气/灌注扫描、超声心动图)确诊者; (3)病史记录不详缺乏重要研究信息者,或检查不全遗漏关键研究结果者。

本研究经首都医科大学附属北京朝阳医院伦理委员会批准(批件号:2018-科-281)。

1.2 Wells评分

根据Wells评分量表计算每一位患者的Wells评分。

1.3 数据收集

收集患者就诊时的以下数据:人口统计学特征(年龄、性别)、症状(发作时间、咯血、胸痛、呼吸困难、晕厥、单侧下肢水肿)、生命体征、基础疾病、静脉血栓危险因素(PE或深静脉血栓史、久坐、3个月内手术或制动史、3个月内骨折史、活动期恶性肿瘤)、D-dimer、动脉血气分析、Wells评分。根据CTPA检查结果把患者分成PE组和非PE(Non-PE)组,对数据进行统计学分析。

1.4 统计学方法

使用SPSS 23.0软件统计分析结果。符合正态分布的计量资料采用均数±标准差(x±s)表示,组间采用独立样本t检验进行比较; 二分类变量使用频数(构成比)描述,采用卡方检验或Fisher' s精确检验进行比较。采用多变量logistic回归模型分析独立预测因素。通过受试者工作特征(receiver operating characteristic, ROC)曲线下的面积(area under ROC curve, AUC)来评估测试效率。然后,计算预测模型和评分的敏感度、特异度、阳性预测值、阴性预测值、阳性似然比和阴性似然比。以P < 0.05为差异有统计学意义。

2 结果

共筛选疑似APE患者1 869例,排除因肾功能不全、碘过敏、妊娠、血液动力学不稳定等因素未经过CTPA确诊的患者139例,以及缺失重要研究资料的患者238例,最终共纳入研究1 492例,根据CTPA结果分为PE组(537例)和Non-PE组(955例)。见图 1

图 1 患者入组流程图 Fig 1 Flowchart of patient enrollment
2.1 人口学特征、症状体征、基础疾病情况

PE组的胸痛、呼吸困难、单侧下肢水肿、PE或静脉血栓史、3个月内手术或制动史、3个月内骨折史、活动期恶性肿瘤显著高于Non-PE组(P < 0.05),其他指标在两组中差异无统计学意义。见表 1

表 1 PE组和Non-PE组人口学特征、症状体征、基础疾病的比较 Table 1 Comparison of demographic characteristics, symptoms, signs, and underlying diseases between the PE and the non-PE groups
变量 PE组
n=537)
Non-PE组
n=955)
t/χ2 P
年龄(岁,x±s 64±15 66±16 4.828 0.074
性别,女,(n, %) 281(52.3) 538(56.3) 2.229 0.135
症状,(n, %)
    咯血 24(4.5) 39(4.1) 0.126 0.722
    胸痛 152(28.3) 199(20.8) 10.654 0.001
    呼吸困难 167(31.1) 217(22.7) 12.617 < 0.001
    晕厥 43(8.0) 88(9.2) 0.625 0.429
    单侧下肢水肿 32(6.0) 12(1.3) 26.557 < 0.001
静脉血栓危险因素,(n, %)
    肺栓塞或静脉血栓史 73(13.6) 61(6.4) 21.838 < 0.001
    3月内手术或制动史 50(9.3) 58(6.1) 5.366 0.021
    3月内骨折史 9(2.6) 4(0.5) 7.150 0.007
    动期恶性肿瘤 34(6.3) 24(2.5) 13.413 < 0.001
基础疾病,(n, %)
    高血压 194(36.1) 368(38.5) 0.848 0.357
    糖尿病 57(10.6) 127(13.3) 2.290 0.130
    冠心病 65(12.1) 144(15.1) 2.524 0.112
    心房颤动 18(3.4) 26(2.7) 0.476 0.490
    心功能不全 12(3.5) 35(4.6) 0.740 0.390
    慢性肺病 26(4.8) 65(6.8) 2.316 0.128
    血液病 1(0.3) 2(0.3) - 1.000
注:PE为肺栓塞
2.2 Wells评分、PaCO2分析

两组患者Wells评分和PaCO2的比较见表 2。Wells评分越高,发生PE的风险越高,PE组中Wells评分≥2分且 < 7分占比最高,而Non-PE组88.5%的患者Wells评分 < 2分。PaCO2降低在PE组的占比显著高于在Non-PE组的占比(P < 0.001)。

表 2 PE组和Non-PE组PaCO2和Wells评分的比较 Table 2 Comparison of PaCO2 and Wells score between the PE and non-PE groups
变量 PE组
n,%)
Non-PE组
n,%)
χ2 P
Wells评分
    低度可能,Wells评分(<2) 206(38.4) 845(88.5) 414.700 < 0.001
    中度可能,Wells评分(≥2,<7) 316(58.8) 109(11.4) 379.606 < 0.001
    高度可能,Wells评分(≥7) 15(2.8) 1(0.1) 23.420 < 0.001
PaCO2
    PaCO2降低(<35 mmHg) 429(79.9) 421(44.1) 179.750 < 0.001
    PaCO2未降低(≥35 mmHg) 108(20.1) 534(55.9) 179.750 < 0.001
注:1 mmHg=0.133 kPa
2.3 Wells评分、PaCO2降低及二者联合的ROC曲线分析

表 3所示,Wells评分的AUC为0.784(95%CI: 0.758~0.810),预测APE的敏感度为61.64%,特异度为88.48%;PaCO2降低的AUC为0.679(95%CI: 0.651~0.707),预测APE的敏感度为79.89%,特异度为55.92%;PaCO2降低联合Wells评分的AUC为0.837(95%CI: 0.816~0.858),预测APE的敏感度为74.12%,特异度为77.07%。图 2所示为Wells评分、PaCO2降低及二者联合的ROC曲线,Wells评分联合PaCO2降低的AUC显著大于Wells评分的AUC(P < 0.001); Wells评分联合PaCO2降低的AUC显著大于PaCO2降低的AUC(P < 0.001)。

表 3 Wells评分、PaCO2降低及二者联合的比较 Table 3 Comparison of Wells score, PaCO2 reduction and their combination
变量 AUC 95%CI 敏感度(%) 特异度(%) 阳性预测值(%) 阴性预测值(%) 阳性似然比 阴性似然比
Wells评分 0.784 0.758~0.810 61.64 88.48 75.05 80.40 5.35 0.43
PaCO2降低 0.679 0.651~0.707 79.89 55.92 50.47 83.18 1.81 0.36
Wells评分+PaCO2降低 0.837 0.816~0.858 74.12 77.07 64.51 84.12 3.23 0.34

图 2 Wells评分、PaCO2降低及二者联合的ROC曲线比较 Fig 2 Comparison of ROC curves of Wells score, PaCO2 reduction and their combination
3 讨论

Wells评分作为最早提出的PE评分量表,较其他评分量表更成熟,且应用广泛[11-12]。该方法通过结合患者症状、体征、PE高危因素,综合评估APE发生的可能性,分值越高,诊断为APE的可能性越大,被美国内科医师学会和美国家庭医师学会作为预测PE可能性的评估方法。

本研究结果显示Wells评分预测APE的灵敏度为61.64%,特异度为88.48%,ROC曲线下面积为0.784(95%CI: 0.758~0.810),提示其对APE初步诊断筛查的有效性。作为PE的经典评分,既往已有众多学者对Wells评分预测PE的能力做出评估。一项回顾性研究发现,相较于Geneva评分,Wells评分针对继发于深静脉血栓的PE更具有预测价值[13]。刘剑等[14]研究发现Wells评分诊断APE的灵敏度为88.72%,特异度为35.37%,AUC为(0.622±0.019)。本研究结果与这些既往报道的结论大致相似,可见这一经典评分仍保留着较高的临床应用价值,很多学者围绕其临床应用不断探索更新,其诊断效力得到了必要的补充。

研究发现D-dimer联合Wells评分能够提高Wells评分的效能[8, 15],也有研究进一步说明Wells联合经过年龄校正的D-dimer能够降低假阳性的发生率[16-17]。还有研究表明心电图异常联合Wells评分,D-dimer、超敏C反应蛋白联合Wells评分,肝细胞因子HGF、D-dimer联合Wells评分,对APE诊断的灵敏度与特异度均有所提升。但目前尚无Wells评分与PaCO2联合应用的相关研究,为补充这一空白,本研究对二者联合预测APE的效能进行探索。本研究结果显示Wells评分联合PaCO2降低对评估APE的灵敏度为74.12%,特异度为77.07%,阳性预测值为70.05%,阴性预测值为80.40%,AUC为0.837(95%CI: 0.816~0.858),显著大于单独应用Wells评分的AUC,以及单独应用PaCO2降低的AUC,提高了Wells评分诊断PE的效能。

有研究发现高危组PE患者溶栓前PaCO2明显低于溶栓后[18]。PaCO2降低与APE的病理生理学改变密切相关。PE发生时,通气/血流(V/Q)比例失调。栓塞部位后血流明显减少,造成高通气低血流,可使V/Q显著 > 0.8,甚至可高达10以上,此时血液中氧分压明显上升,但因氧及二氧化碳解离曲线的特性,此时氧含量增加很少,而二氧化碳分压与含量均明显降低; 健康的肺区因PE而血流增加,形成功能性分流,这些部位通气正常,血流量增加导致V/Q低于正常值,血液无法充分动脉化,导致氧含量减少,但二氧化碳分压与含量可增加。在肺不张的区域,无通气造成静脉血直接汇入动脉导致真性分流。严重的低氧血症发生时,刺激呼吸中枢,机体代偿性发生过度通气,进一步导致PaCO2降低。因此,PaCO2降低对APE的诊断具有一定提示意义。本研究结果也证明的这一推断,PaCO2降低预测APE的AUC达到了0.679(95%CI: 0.651~0.707),灵敏度为79.89%,特异度为55.92%,阳性预测值为50.47%,阴性预测值为83.18%,其灵敏度与阴性预测值甚至超过了单独应用Wells评分,显示出良好的临床应用潜力。

利益冲突  所有作者声明无利益冲突

作者贡献声明  何新华、郭树彬、腾飞:研究设计; 左冬晶、张艳慧、赵丽新:数据收集与整理; 腾飞、曹玉丹:统计学分析; 左冬晶、曹玉丹:初稿撰写; 左冬晶、曹玉丹、腾飞、何新华:论文修改; 何新华:批评性审阅

参考文献
[1] Jamil A, Johnston-Cox H, Pugliese S, et al. Current interventional therapies in acute pulmonary embolism[J]. Prog Cardiovasc Dis, 2021, 69: 54-61. DOI:10.1016/j.pcad.2021.11.009
[2] Cohen AT, Agnelli G, Anderson FA, et al. Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality[J]. Thromb Haemost, 2007, 98(4): 756-764.
[3] Zhai ZG, Wang DY, Lei JP, et al. Trends in risk stratification, in-hospital management and mortality of patients with acute pulmonary embolism: an analysis from the China pUlmonary thromboembolism REgistry Study (CURES)[J]. Eur Respir J, 2021, 58(4): 2002963. DOI:10.1183/13993003.02963-2020
[4] Alshumrani G, Al Bshabshe A, Mousa WF. Diagnostic yield of CT pulmonary angiography for pulmonary embolism in clinically suspected patients[J]. Medicine (Baltimore), 2021, 100(22): e26213. DOI:10.1097/MD.0000000000026213
[5] Higashiya K, Ford J, Yoon HC. Variation in positivity rates of computed tomography pulmonary angiograms for the evaluation of acute pulmonary embolism among emergency department physicians[J]. Perm J, 2022, 26(1): 58-63. DOI:10.7812/TPP/21.019
[6] Zhang NJ, Rameau P, Julemis M, et al. Automated pulmonary embolism risk assessment using the wells criteria: validation study[J]. JMIR Form Res, 2022, 6(2): e32230. DOI:10.2196/32230
[7] Wells PS, Ginsberg JS, Anderson DR, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism[J]. Ann Intern Med, 1998, 129(12): 997-1005. DOI:10.7326/0003-4819-129-12-199812150-00002
[8] 腾飞, 何新华, 李彦媚, 等. Wells评分联合D-dimer在急性肺栓塞诊断的临床预测价值[J]. 中华急诊医学杂志, 2015, 24(4): 422-426. DOI:10.3760/cma.j.issn.1671-0282.2015.04.019
[9] 石岩, 黄中伟. 基于D-二聚体分层法联合肺栓塞临床可能性评估对急诊肺栓塞预测评估效果的比较[J]. 中华急诊医学杂志, 2021, 30(12): 1501-1505. DOI:10.3760/cma.j.issn.1671-0282.2021.12.017
[10] 中华医学会呼吸病学分会肺栓塞与肺血管病学组, 中国医师协会呼吸医师分会肺栓塞与肺血管病工作委员会, 全国肺栓塞与肺血管病防治协作组. 肺血栓栓塞症诊治与预防指南[J]. 中华医学杂志, 2018, 98(14): 1060-1087. DOI:10.3760/cma.j.issn.0376-2491.2018.14.007
[11] Hepburn-Brown M, Darvall J, Hammerschlag G. Acute pulmonary embolism: a concise review of diagnosis and management[J]. Intern Med J, 2019, 49(1): 15-27. DOI:10.1111/imj.14145
[12] Doherty S. Pulmonary embolism: an update[J]. Aust Fam Physician, 2017, 46(11): 816-820.
[13] Chen XY, Liu X, Liu JL, et al. Pulmonary embolism secondary to deep venous thrombosis: a retrospective and observational study for clinical characteristics and risk stratification[J]. Phlebology, 2021, 36(8): 627-635. DOI:10.1177/0268355521990964
[14] 刘剑, 孙林, 向华, 等. Wells评分、Geneva评分和YEARS法对疑似肺栓塞住院患者诊断价值的对比研究[J]. 介入放射学杂志, 2021, 30(6): 552-556. DOI:10.3969/j.issn.1008-794X.2021.06.005
[15] 田小雨, 赵景成, 郭树彬, 等. 急诊科Wells评分中低危老年患者深静脉血栓发生情况分析[J]. 中华急诊医学杂志, 2020, 29(7): 981-983. DOI:10.3760/cma.j.issn.1671-0282.2020.07.016
[16] van Es N, Kraaijpoel N, Klok FA, et al. The original and simplified Wells rules and age-adjusted D-dimer testing to rule out pulmonary embolism: an individual patient data meta-analysis[J]. J Thromb Haemost, 2017, 15(4): 678-684. DOI:10.1111/jth.13630
[17] Geersing GJ, Takada T, Klok FA, et al. Ruling out pulmonary embolism across different healthcare settings: a systematic review and individual patient data meta-analysis[J]. PLoS Med, 2022, 19(1): e1003905. DOI:10.1371/journal.pmed.1003905
[18] 高伟波, 石茂静, 窦丽稳, 等. 110例急性肺栓塞溶栓患者临床分析[J]. 中华急诊医学杂志, 2020, 29(3): 398-403. DOI:10.3760/cma.j.issn.1671-0282.2020.03.005