As we all know, many of the exacerbations we see in the ED are precipitated by either infection, cold weather, pulmonary embolus, pneumothorax, CHF or trauma. We have an algorithm for treating exacerbations that uses a combination of the GOLD criteria, which stratifies exacerbations based on severity, and the Anthonisen criteria which is a subjective means of using our clinical gestalt to see who needs antibiotics. 

With there being a strong push for limiting the use of antibiotics and preventing resistance, I wanted to take a look at this study which questions whether testing C-Reactive Protein (CRP) changes our management and limits our antibiotic use for COPD exacerbations.

The authors of this trial wanted to test point of care CRP in patients with acute COPD exacerbations.  They hypothesized that the results of Point-of-care (POC) CRP may help inform prescribing decisions for acute COPD exacerbations.  It was a multicenter, open-label, randomized controlled trial where patients were recruited from 86 general medical practices in the UK.

•  Patients with a diagnosis of COPD seen by a physician for an acute COPD exacerbation were randomized into two groups: usual care guided by CRP POC testing (CRP-guided group) and usual care alone (the control group). 

• Clinicians were then advised that antibiotics are unlikely to be beneficial when the CRP level is lower than 20 mg per liter, likely to be beneficial when the CRP level is higher than 40 mg per liter, and might be beneficial when the CRP level falls between these extremes and purulent sputum is present. 

• Clinicians were asked to base prescribing decisions on “a comprehensive assessment of likely risks and benefits” and not solely on the CRP level. 

1. The percentage of patients who reported using antibiotics within 4 weeks after randomization was significantly lower in the CRP-guided group than in the control group (57% vs. 77%), yet clinical outcomes were similar in the two groups.

2. In the CRP-guided group, antibiotics were prescribed for one-third of the patients with CRP levels lower than 20 mg per liter, but approximately 90% of the patients with CRP levels of at least 20 mg per liter or higher received an antibiotic prescription. 

What were the strengths?

1. 1. It was a multicenter randomized controlled trial

   2. The CRP point of care testing software was sponsored by Abbott Labs at no cost, and they had no influence in the testing, data interpretation of the trial. 

What were the limitations? 

1. 1. It was an unblinded trial for patients, meaning that the CRP tested group was aware that they received POC testing. This may have in fact allowed them to over-estimate their health status on the questionnaires. 

   2. These were non-ED settings, meaning these patients may not be as critical or sick as patients presenting to the emergency department. Therefore, it’s difficult to generalize these results to a sicker population.