Clinical Question

Are inflammatory biomarkers beneficial or harmful when used to guide antibiotic treatment for acute respiratory infections?

Evidence-Based Answer

C-reactive protein (CRP) point-of-care tests can reduce unnecessary antibiotic prescribing without delaying clinical recovery when used to inform treatment decisions for acute respiratory infections in primary care (number needed to test = 9; 95% CI, 6 to 14). (Strength of Recommendation [SOR]: A, consistent, good-quality patient-oriented evidence.) It remains unclear whether procalcitonin point-of-care tests affect antibiotic use or patient recovery.¹

Practice Pointers

A 2016 study using the 2010–2011 National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey found that nearly one in three antibiotic prescriptions in U.S. ambulatory care settings was inappropriate.² Optimizing antibiotic prescribing practices in outpatient settings reduces the risk of antibiotic resistance, potential adverse drug effects, and harmful bacterial over-growth in patients and aligns with national and international public health goals.^3,4

The Cochrane review included 13 randomized controlled trials (RCTs) and cluster RCTs in a meta-analysis of 8,200 adults and 2,335 children.¹ Studies were conducted in primary health care settings in Europe and Asia. One study included participants from nursing homes. In 12 studies, moderate-certainty evidence showed that using CRP point-of-care tests decreased the number of participants prescribed an antibiotic (number needed to test = 9; 95% CI, 6 to 14), although considerable heterogeneity existed. High-certainty evidence showed that using CRP point-of-care tests also reduced the number of participants prescribed an antibiotic within 28 days of follow-up (number needed to test = 8; 95% CI, 6 to 11).

Moderate-certainty evidence showed that there was no difference in clinical recovery rates at seven or 28 days among those who did or did not undergo a CRP test. The use of these tests did not change mortality or hospital admission rates, although low event numbers regarding these outcomes limited the evidence. A conclusion about what effect procalcitonin point-of-care tests might have on antibiotic prescribing and patient recovery could not be made due to the limited sample size of 317 adults from one trial.

CRP point-of-care tests use drops of blood from a finger prick and return a result in about three minutes. There is no standardized CRP cutoff to guide antibiotic prescriptions. Considerable differences in CRP cutoffs and application to antibiotic prescriptions existed across studies in the Cochrane review. However, a sensitivity analysis of nine included studies that used a cutoff of 2 mg per dL (20 mg per L) was consistent with the overall conclusion that CRP tests can reduce antibiotic prescriptions. In the United Kingdom, the National Institute for Health and Care Excellence (NICE) provides CRP cutoffs to guide antibiotic prescribing in conjunction with clinical assessment for possible pneumonia in adults presenting to primary care. NICE recommends avoiding antibiotic prescription if CRP is less than 2 mg per dL, delaying prescription if CRP is between 2 and 10 mg per dL (100 mg per L), and offering antibiotics if CRP is greater than 10 mg per dL.⁵ Clinicians should acknowledge the lack of test validation in children and older adults with comorbidities and the possible opportunity cost of suboptimal time use, patient dissatisfaction, and false-positive or false-negative results when integrating CRP tests into primary care practice. Although the use of CRP tests might increase health care costs, European studies suggest it is likely a cost-effective intervention in reducing antibiotic prescriptions.¹ Future research should be aimed at evaluating procalcitonin and new biomarkers.

Editor's Note: The numbers needed to test and related CIs reported in this Cochrane for Clinicians were calculated by the authors based on raw data provided in the original Cochrane review.

The practice recommendations in this activity are available at https://www.cochrane.org/CD010130.