Pulmonary embolism in younger adults: Clinical presentation and comparison of two scoring systems used to estimate pretest probability of disease in the emergency department

Abstract

Study objectives: Previous reports suggested that clinical presentation of pulmonary embolism could be different in younger compared with older patients. We test this hypothesis and compare 2 scoring systems used to estimate pretest probability of pulmonary embolism in younger and older patients admitted to the emergency department (ED). Methods: This is an observational retrospective study of 80 consecutive patients (40 younger adults, age ≤45 years; 40 older adults, age >45 years) admitted to the ED of an urban hospital from January 1996 to June 2003 for suspected pulmonary embolism, confirmed by high-probability lung scan or spiral computed tomography. We excluded patients with cardiopulmonary diseases. We collected data of history, age, sex, symptoms, signs and parameters, laboratory tests, chest radiograph, ECG, and arterial blood gas and estimated pretest probability of pulmonary embolism using Wells system according to 7 variables: suspected deep venous thrombosis, an alternative diagnosis is less likely than pulmonary embolism, pulse rate greater than 100 beats/min, recent immobilization or surgery, previous deep venous thrombosis/pulmonary embolism, hemoptysis, and malignancy; and Wicki system, based on 8 variables: recent surgery, previous thromboembolic event, older age, hypocapnia, hypoxemia, tachycardia, band atelectasis, or elevation hemidiaphragm on chest radiograph. A probability score was calculated by adding points assigned to these variables. We used Student's <i>t</i> or Fisher exact test to compare 2 groups. Spearman rank correlation coefficient was calculated for 2 score models and between each model and number of unperfused lung segments. Results: Mean ages were 39 years (range 25 to 45 years) for younger adults and 81 years (range 70 to 93 years) for older adults. After comparison with older patients, younger patients had less typical pulmonary embolism risk factors (in particular recent immobilization: 20% versus 35%, <i>P</i>=.03; malignancy: 6% versus 36%, <i>P</i><.001); less abnormal ECG (40% versus 70%, <i>P</i><.001) (right bundle-branch block was the more frequent abnormality in younger adults, 30% versus 17%, <i>P</i><.001); less abnormal radiograph (20% versus 40%, <i>P</i><.001); lower d-dimer levels (mean of 875, 8 μg/L versus 1,889, 7 μg/L in older adults [<i>P</i><.05]). Among younger adults, the more frequent symptoms were dyspnea (50%), chest pain (50% versus 30% in older adults, <i>P</i>=.006), cough (20% versus 5% in older adults, <i>P</i>=.003), and hemoptysis (15% versus 5%, <i>P</i><.03). At presentation, younger patients showed better parameters and better air room arterial blood gas (PaO₂, oxygen alveolar-arterial difference). The 2 predictive systems showed a different distribution of scores in 2 groups. Among older adults, with both systems, there was a predominance of high scores: 85% with Wells score greater than 2 (high pulmonary embolism probability) and 85% with Wicki score greater than 5 (high pulmonary embolism probability). Instead, among younger adults there was a high predominance of high score with Wells score: 90% with score greater than 2, but few younger adults had a Wicki score greater than 5 (45%). Between the 2 systems, there is a weak positive correlation (<i>r</i>=0.31, <i>P</i>>.05). The Wells system showed a moderate positive correlation with the number of unperfused lung segments (<i>r</i>=0.68, <i>P</i>=.004). Conclusion: In this study, among younger adults with pulmonary embolism, there is a more insidious clinical presentation; the Wells system has a predominance of high scores and seems to have a positive correlation with pulmonary embolism severity. The low number of patients is the main limitation of the study.