Glober Nancy, Tainter Christopher R, Brennan Jesse, Darocki Mark, Klingfus Morgan, Choi Michelle, Derksen Brenna, Rudolf Frances, Wardi Gabriel, Castillo Edward, Chan Theodore
Department of Emergency Medicine, University of California at San Diego, San Diego, California.
J Emerg Med. 2018 May;54(5):585-592. doi: 10.1016/j.jemermed.2018.01.032. Epub 2018 Mar 2.
Assessment for pulmonary embolism (PE) in the emergency department (ED) remains complex, involving clinical decision tools, blood tests, and imaging.
Our objective was to examine the test characteristics of the high-sensitivity d-dimer for the diagnosis of PE at our institution and evaluate use of the d-dimer and factors associated with a falsely elevated d-dimer.
We retrospectively collected data on adult patients evaluated with a d-dimer and computed tomography (CT) pulmonary angiogram or ventilation perfusion scan at two EDs between June 4, 2012 and March 30, 2016. We collected symptoms (dyspnea, unilateral leg swelling, hemoptysis), vital signs, and medical and social history (cancer, recent surgery, medications, history of deep vein thrombosis or PE, chronic obstructive pulmonary disease, smoking). We calculated test characteristics, including sensitivity, specificity, and likelihood ratios for the assay using conventional threshold and with age adjustment, and performed a univariate analysis.
We found 3523 unique visits with d-dimer and imaging, detecting 198 PE. Imaging was pursued on 1270 patients with negative d-dimers, revealing 9 false negatives, and d-dimer was sent on 596 patients for whom negative Pulmonary Embolism Rule-Out Criteria (PERC) were documented with 2% subsequent radiographic detection of PE. The d-dimer showed a sensitivity of 95.7% (95% confidence interval [CI] 91-98%), specificity of 40.0% (95% CI 38-42%), negative likelihood ratio of 0.11 (95% CI 0.06-0.21), and positive likelihood ratio of 1.59 (95% CI 1.53-1.66) for the radiographic detection of PE. With age adjustment, 347 of the 2253 CT scans that were pursued in patients older than 50 years with an elevated d-dimer could have been avoided without missing any additional PE. Many risk factors, such as age, history of PE, recent surgery, shortness of breath, tachycardia and hypoxia, elevated the d-dimer, regardless of the presence of PE.
Many patients with negative d-dimer and PERC still received imaging. Our data support the use of age adjustment, and perhaps adjustment for other factors seen in patients evaluated for PE.
急诊科对肺栓塞（PE）的评估仍然很复杂，涉及临床决策工具、血液检查和影像学检查。
我们的目的是研究我院高敏D-二聚体诊断PE的检测特征，并评估D-二聚体的使用情况以及与D-二聚体假性升高相关的因素。
我们回顾性收集了2012年6月4日至2016年3月30日期间在两家急诊科接受D-二聚体和计算机断层扫描（CT）肺血管造影或通气灌注扫描评估的成年患者的数据。我们收集了症状（呼吸困难、单侧腿部肿胀、咯血）、生命体征以及医疗和社会史（癌症、近期手术、用药情况、深静脉血栓形成或PE病史、慢性阻塞性肺疾病、吸烟史）。我们计算了检测特征，包括使用传统阈值和年龄调整后的检测灵敏度、特异性和似然比，并进行了单变量分析。
我们发现了3523次进行D-二聚体和影像学检查的独立就诊，检测到198例PE。对1270例D-二聚体阴性的患者进行了影像学检查，发现9例假阴性，对596例记录有阴性肺栓塞排除标准（PERC）的患者进行了D-二聚体检测，随后影像学检测到PE的比例为2%。D-二聚体对PE影像学检测的灵敏度为95.7%（95%置信区间[CI]91-98%），特异性为40.0%（95%CI 38-42%），阴性似然比为0.11（95%CI 0.06-0.21），阳性似然比为1.59（95%CI 1.53-1.66）。经过年龄调整，在年龄大于50岁且D-二聚体升高的患者中进行的2253次CT扫描中，有347次可以避免，且不会遗漏任何额外的PE。许多危险因素，如年龄、PE病史、近期手术、呼吸急促、心动过速和缺氧，都会使D-二聚体升高，无论是否存在PE。
许多D-二聚体和PERC阴性的患者仍然接受了影像学检查。我们的数据支持使用年龄调整，也许还应根据评估PE患者中出现的其他因素进行调整。
