Abstract 2: Prevalence of PE Among Patients Hospitalized for COPD

Prevalence of pulmonary embolism among patients with COPD hospitalized with acutely worsening respiratory symptoms Couturaud F, Bertoletti L, Pastre J, et al. JAMA. 2021;325(1):59-68.

SUMMARY:

• Pulmonary embolism (PE) continues to present some of the most difficult diagnostic questions for clinicians in the ED. The issue addressed in this article is whether to suspect PE in patients with chronic obstructive pulmonary disease (COPD). This idea has a long and sordid history, which the authors summarize by citing a recent meta-analysis finding a 16% PE rate for patients admitted with acute COPD exacerbation. Understanding the prevalence estimates in the articles underlying that meta-analysis has been a challenge, because some of the studies enrolled only patients with COPD who had unexplained worsened symptoms, whereas others enrolled all comers with COPD exacerbation and worsened symptoms. Most of the studies had a small N (<200).
• The point of this study was to provide a more precise estimate of the prevalence of PE in patients admitted with moderate to severe COPD exacerbation. This was a prospective, observational study conducted in 7 hospitals across France from 2014 to 2017. The study included adult patients with COPD exacerbation and with a generally known history of COPD. The exclusion criteria were basic and included aspects such as renal insufficiency and other reasons for worsening shortness of breath.
• All patients underwent Geneva Score Risk stratification on admission. Patients with high probability of PE were directly sent to receive computed tomographic pulmonary angiography, whereas patients with low-to-moderate probability of PE underwent D-dimer testing and imaging according to the results (a D-dimer cutoff of 500 µg/L rather than an age-adjusted cutoff was used). Importantly, the senior physician on duty was asked to indicate whether PE was suspected, as a single-item-type question, before any testing was performed. Bilateral duplex ultrasounds were obtained if computed tomographic pulmonary angiography was indicated, but the findings were negative. 
• The primary outcome was diagnosis of PE within 48 hours after admission. Over the 3-year study period across the 7 hospitals, 740/2,268 patients were enrolled. Half were excluded because they were already taking anticoagulants, and most of the rest were excluded because of a lack of signed informed consent. The mean age was 69 years, and the patients were quite ill: 20% had baseline immobility, and 10% had cancer. Only 2% of the cohort had a high pretest probability of PE; 19% had low probability; and a surprisingly large number had intermediate probability of PE (79%). PE was suspected by the senior ED physician in 40% of cases. 
• Ultimately, PE was diagnosed in 44/740 patients within 48 hours after admission (5.9%). The prevalence was 10% for patients in whom the senior physician suspected a PE (30 of 299) and 3.2% for patients in whom the senior physician did not suspect a PE (14 of 441). The consequences of the diagnosed PEs were unclear: 50/750 patients died within 3 months. Only a few of the deaths were due to PE, and the patients were not necessarily in the group without suspected PE. 
• Although it is plausible that 2-3% of patients with COPD and respiratory decompensation severe enough to be admitted might have PE, many questions persist. What is the PE rate for all hospital admissions of ill patients? Is COPD special, or does it reflect more ill, older people? Does treating these people affect anything? This article will probably bolster anyone’s initial position. If you thought all these people should be screened, you will think 6% is a lot. If you are a skeptic, you will say that this prevalence is one-third that reported in previous studies, and it decreases to 3% (one-sixth that reported in previous studies) when a simple physician screen is taken into account. So ask yourself if you have a serious suspicion for something other than simple COPD exacerbation; if the answer is yes, pursue a PE diagnosis, but there is no reason for everyone to receive this screening. 

PMID: 33399840

EDITOR’S COMMENTARY: This study reports that 5.9% of patients admitted for COPD exacerbation to hospitals in France had a concurrent diagnosis of PE. However, the prevalence of PE among people with COPD exacerbation in whom the treating physician did not have clinical suspicion of PE was only 3%. On the basis of this and previous studies, I do not think that screening patients with COPD for PE is necessary unless there is significant clinical suspicion.

Copyright 2021 by Emergency Medical Abstracts – All Rights Reserved 04/21 - #02