NIHR Signal C-reactive protein testing in general practice safely reduces antibiotic use for COPD flare-ups
Published on 24 September 2019
Use of a rapid C-reactive protein (CRP) blood test in general practice for people with a flare-up of COPD reduces the proportion who take antibiotics over the next month by about 20 percentage points compared with usual care alone. The reduction in antibiotic use does not lead to worse health, more visits to the doctor or greater need for antibiotics later on.
Flare-ups of COPD can be caused by infections of the airways or environmental triggers, and cause about 115,000 admissions to hospital every year. Determining whether a flare-up is due to infection is difficult in primary care, and antibiotics are often prescribed to treat a presumed underlying bacterial infection.
This NIHR-funded trial provides strong evidence that a rapid test for raised CRP levels, which occur with serious infection, could help prescribers in primary care make better decisions about who needs antibiotics to treat a flare-up of COPD and who doesn’t.
- Health management, Medicines, Respiratory disorders, Primary care
Why was this study needed?
COPD is a progressive lung disease affecting at least 1.2 million people in the UK and causing around 30,000 deaths annually. Each year around half of the people living with COPD will experience a flare-up of their symptoms (called an exacerbation).
Flare-ups can be caused by bacterial or viral infections of the airways or environmental triggers such as smoking. As it is not easy to rapidly identify the cause of a flare-up, antibiotics are often prescribed in case a bacterial infection of the airways is the cause. However, this may not be the case. Such unnecessary prescriptions can contribute to increases in drug-resistant bacteria.
A previous study has suggested that patients who have raised levels of CRP in their blood are most likely to need antibiotics. This well-conducted trial tested whether rapid CRP testing in general practice can safely reduce unnecessary prescribing of antibiotics to patients with COPD flare-ups.
What did this study do?
This randomised controlled trial recruited 649 adults with a flare-up of COPD attending 86 general practices in the UK. Symptoms could include becoming more breathless, more coughing, or producing more discoloured sputum. Patients with severe illness (such as suspected pneumonia) were not included.
One group of patients was assigned to have CRP tests alongside usual care whilst the other group received usual care only. In the CRP group, clinicians were given advice on interpreting test results as part of a comprehensive assessment of likely benefits and risks of antibiotics for each patient. For example, they were advised that antibiotics usually should not be prescribed when CRP levels were lower than 20mg/L.
All clinicians were provided with a summary of NICE and international guidance on managing COPD.
What did it find?
- CRP testing reduced the proportion of patients who reported taking antibiotics in the first four weeks of follow-up, compared with those receiving usual care only (57.0% with CRP testing vs 77.4% with usual care alone; adjusted odds ratio [aOR] 0.31, 95% confidence interval [CI] 0.20 to 0.47). When looking at patient subgroups, the reduction in antibiotic use was only seen in those with at least two symptoms of a flare-up.
- The group who received CRP-guided treatment were no worse off in their COPD-related health after two weeks’ follow-up than those who received usual care alone (adjusted mean difference in Clinical COPD Questionnaire score: -0.19 points, 90% CI -0.33 to -0.05; score ranges from 0 to 6, with a lower score indicating better health; a difference of 0.4 points was considered clinically important).
- Prescribing patterns suggested that those who didn’t get antibiotics at the first consultation weren’t more likely to need them later on or to need more of other COPD treatments. Using the CRP tests reduced the proportion of patients who received an antibiotic prescription at their initial consultation and during the first four weeks of follow-up (prescription at initial consultation: 47.7% with CRP testing vs 69.7% with usual care alone; aOR 0.31, 95% CI 0.21 to 0.45; prescription during follow-up: 59.1% with CRP testing vs 79.7% with usual care alone; aOR 0.30, 95% CI 0.20 to 0.46). There was no difference between the groups in the use of other COPD treatments during the first four weeks of follow-up (aOR 0.79, 95% CI 0.43 to 1.46).
- In the CRP-guided group, 76% of those tested at their first consultation had CRP levels lower than 20mg/L. Antibiotics were prescribed for about one-third of these patients, but around 90% of those with CRP levels above this.
- There was no difference between the groups in primary care or secondary care consultations during six months of follow-up. There was also no difference between the groups in pneumonia or adverse effects of antibiotics in this period.
What does current guidance say on this issue?
The NICE guidelines on COPD management state that for patients with acute exacerbations sending sputum samples for culture is not recommended as routine practice in primary care. They do not include recommendations on the use of CRP testing to guide antibiotic prescribing.
NICE has developed Medtech Innovation Briefings on two rapid CRP tests that can be used during primary care consultations. These briefings are not guidance, but describe the technologies and appraise the existing studies assessing their use. In both cases, the briefings looked at suspected lower respiratory tract infections as a whole rather than COPD exacerbations specifically.
What are the implications?
Overuse of antibiotics is a major concern globally, due to the increase in drug-resistant bacteria. There is also a drive to reduce overtreatment and unnecessary prescribing. To do this in patients with COPD exacerbations, in whom there is a high level of antibiotic use, clinicians ideally need point-of-care tools that can quickly identify bacterial infections in general practice.
This study provides strong evidence to suggest that rapid point-of-care CRP testing for these patients may help with antibiotic stewardship, without compromising patient outcomes or safety. The findings seem likely to influence future guidance and practice.
Citation and Funding
Butler CC, Gillespie D, White P et al. C-reactive protein testing to guide antibiotic prescribing for COPD exacerbations. New Engl J Med. 2019: 381: 111-120
This research was funded by the NIHR Technology Assessment Programme (project number 12/33/12). The testing machines used in the study were loaned to researchers by the manufacturer, who also provided training on their use. The manufacturer had no other role in any part of the trial.
British Lung Foundation. Chronic obstructive pulmonary disease (COPD) statistics. London: British Lung Foundation; 2019
GOLD. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease 2019 report. Wisconsin: Global Initiative for Chronic Obstructive Lung Disease; 2019.
Miravitlles M, Moragas A, Hernandez S et al. Is it possible to identify exacerbations of mild to moderate COPD that do not require antibiotic treatment? Chest. 2013;144(5):1571-7
NICE. Alere Afinion CRP for C-reactive protein testing in primary care. MIB81. London: National Institute for Health and Care Excellence; 2016.
NICE. Antibiotics should be restricted for COPD. London: National Institute for Health and Care Excellence; 2018.
NICE. Chronic obstructive pulmonary disease (acute exacerbation): antimicrobial prescribing. NG114. London: National Institute for Health and Care Excellence; 2018.
NICE. Chronic obstructive pulmonary disease in over 16s: diagnosis and management. NG115. London: National Institute for Health and Care Excellence; 2018.
NICE. QuikRead go for C-reactive protein testing in primary care. MIB78. London: National Institute for Health and Care Excellence; 2016.
O’Neill J. Rapid diagnostics: stopping unnecessary use of antibiotics. London: Review on Antimicrobial Resistance; 2015.
Currently, around 80% of those presenting in primary care with moderate exacerbations of COPD are treated with antibiotics. Antibiotics overall result in improved clinical outcomes but can they be better targeted to those in need?
The best predictors of adverse outcomes in the placebo arm of an earlier randomised controlled trial were purulent sputum and CRP over 40mg/L. In the current study, the addition of CRP in the clinical assessment of exacerbations in general practice reduced antibiotic exposure without any evidence of short-term harm (at four weeks) or longer-term harms (at six months). There is currently no evidence regarding longer-term outcomes and rare adverse events.
This study might tip the balance for commissioners in the decision to invest in point-of-care CRP testing since it can be used as part of clinical assessment in both uncomplicated lower respiratory tract infection and now COPD exacerbations to better target antibiotic therapy.
Professor Michael Moore, Professor of Primary Care Research, Primary Care and Population Sciences, University of Southampton
The commentator declares no conflicting interests