Abstract: Objective： To evaluate the value of conventional and age-adjusted D-dimer cut-off value combined with 2-level Wells score for diagnosis of suspected pulmonary embolism. Methods: In the stu-dy, 335 patients with suspected pulmonary embolism who visited Peking University People’s Hospital were enrolled retrospectively, then 274 patients with age over fifty years were chosen. The 2-level Wells score was applied to evaluate the clinical probability of pulmonary embolism, the diagnostic value of traditional D-dimer cut-off value (500 μg/L) and age adjusted D-dimer cut-off value (age×10 μg/L above 50 years) combined with Wells score no greater than 4 were compared. Computed tomography pulmonary arteriography (CTPA) was considered as the gold standard for diagnosis of pulmonary embolism. Results: (1)The area under a receiver operating characteristic (ROC) curve (AUC) in analysis of the combination of Wells score no greater than 4 and traditional D-dimer cut-off value was 0.764(95%CI：0.703-0.818). On the other hand, the AUC in a ROC analysisof the combination of Wells Score no greater than 4 and age-adjusted D-dimer cut-off value was 0.814(95%CI：0.756-0.863).These two results did not differ statistically(Z=0.05，P =0.121). (2) The sensitivity, specificity, positive predictive value, negative predictive value and Youden index of the diagnosis of pulmonary embolism of the combination of traditional D-dimer cut-off value and 2-level Wells Score were 100%, 48.9%, 28.8%, 100%, and 0.49, respectively. Meanwhile, the sensitivity, specificity, positive predictive value, negative predictive value and Youden index of the diagnosis of pulmonary embolism of the combination of age-adjusted Ddimer cut-off value and 2-level Wells Score were 97.4%, 62.3%, 35.5% 99.1%, and 0.60, respectively. Compared with using traditional D-dimer cut-off value, using age-adjusted D-dimer cut-off value could improve the diagnosis specificity (traditional D-dimer cut-off value group: 48.9%, age-adjusted D-dimer cut-off value group: 62.3%) of pulmonary embolism without reducing the sensitivity (traditional D-dimer cut-off value group: 100%, age-adjusted D-dimer cut-off value group: 99.1%). (3) Among the 222 patients with Wells Score no greater than 4, 90 patients were with D-dimer less than traditional cut-off value (500 μg/L), and 25 patients (account for 11.3% of all 222 patients) were with D-dimer between traditional cut-off value and age-adjusted cut-off value. Conclusion: The application of age-adjusted D-dimer cut-off value can improve the diagnostic specificity of pulmonary embolism in patients over 50 years, without reducing the sensitivity. It can be used for ruling out suspected pulmonary embolism safely.

Key words: Pulmonary embolism, D-dimer cut-off , Wells score, Computed tomography pulmonary arteriography