Diagnosis of pulmonary embolism in patients with acute exacerbations of chronic obstructive pulmonary disease: A cross‐sectional study

INTRODUCTION: Pulmonary embolism (PE) remains a frequent complication in patients with chronic obstructive pulmonary disease (COPD). It is unclear that the extent to which the traditional risk stratifying scores for PE are accurate in this population. METHODS: Cross-sectional study of adult patients with COPD and suspected PE included in an Institutional Registry of Thromboembolic Disease at a tertiary teaching hospital in the city of Buenos Aires, Argentina. We estimated the area under the receiver operating characteristic curves (AU-ROC), sensitivity and specificity of the Wells and Geneva scores using a positive computed tomography angiography as the gold standard for PE. We also estimated the sensitivity and specificity for the presence of isolated worsening of dyspnea at presentation, without other cardinal symptoms of acute exacerbation of COPD. RESULTS: A total of 168 patients were included, of which 22% had confirmed PE. The AUC was 0.66 (95% CI 0.56-0.76) and 0.56 (95% CI 0.45-0.67) for the Wells and Geneva, respectively. Considering the most widely used cutoff points, the sensitivity and specificity were 24% and 90% for the Wells and 59% and 43% for the Geneva score, respectively. Isolated worsening of dyspnea on presentation had a sensitivity of 92% and specificity of 37%. CONCLUSIONS: Both Wells and Geneva scores exhibit poor diagnostic accuracy for the diagnosis of PE in patients with COPD. The presence of isolated worsening of dyspnea on presentation could be an easy to identify criteria for the initial triage in this population. Further validation of our findings remains warranted.