HsCRP in Acute Respiratory Infection
Use point-of-care CRP testing with established thresholds (20 mg/L and 100 mg/L) to guide antibiotic decisions in adults with acute respiratory infections when there is clinical uncertainty about bacterial etiology.
Recommended CRP-Guided Antibiotic Algorithm
The most recent NICE guideline 1 provides the clearest framework for using CRP in acute respiratory infections:
- CRP > 100 mg/L: Offer immediate antibiotics
- CRP 20-100 mg/L: Consider a back-up (delayed) antibiotic prescription
- CRP < 20 mg/L: Do not routinely offer antibiotics
This stratification should be combined with clinical judgment and severity assessment (such as CRB65 scoring for pneumonia).
Evidence Supporting CRP-Guided Prescribing
CRP testing significantly reduces antibiotic prescribing without compromising patient outcomes. A 2022 meta-analysis demonstrated that CRP-guided protocols reduced antibiotic prescribing by 24% at index consultation (RR=0.76) and by 23% during 28-day follow-up (RR=0.77), with no significant differences in clinical recovery or adverse events 2. A 2023 meta-analysis in hospitalized patients showed CRP guidance reduced antibiotic duration by 1.82 days without increasing mortality or infection relapse 3.
Important Caveats and Limitations
Diagnostic Accuracy Issues
The evidence reveals significant limitations in CRP's diagnostic performance:
- For pneumonia detection: CRP alone has poor diagnostic accuracy (sensitivity 83%, specificity 55% at 20 mg/L threshold) 1. Many patients without lower respiratory tract infection will have CRP >20 mg/L.
- For bacterial vs. viral differentiation: A 2005 systematic review found CRP is "neither sufficiently sensitive to rule out nor sufficiently specific to rule in" bacterial etiology 4. Sensitivities ranged from 8-99% and specificities from 27-95% across studies.
Critical Clinical Pitfalls
Viral testing does not exclude bacterial co-infection: The presence of influenza or other viral pathogens should not guide antibiotic decisions alone 1. Bacterial superinfection remains possible.
CRP thresholds have different performance characteristics: While most patients without LRTI have CRP <100 mg/L (high specificity), the 20 mg/L threshold has poor specificity, leading to potential overtreatment 1.
Single CRP measurements in low-range values have limited utility: When initial CRP is <60 mg/L, a single measurement has poor discriminatory ability (AUC 0.57). However, CRP kinetics (trend over time) can improve diagnostic accuracy to AUC 0.83 5. A CRP rise >3.47 mg/L/hour discriminates bacterial from viral infection with 93.8% specificity.
When CRP Testing Adds Most Value
CRP testing is most useful when:
- Clinical uncertainty exists about bacterial etiology
- Symptoms persist >10 days without improvement
- Symptoms are severe (fever >39°C, purulent discharge, facial pain >3 days)
- "Double sickening" pattern (worsening after initial improvement)
The American College of Physicians guideline 6 emphasizes using clinical criteria first to identify bacterial likelihood before considering adjunctive testing.
Practical Implementation Considerations
Resource limitations matter: Many primary care settings lack point-of-care CRP testing 1. The test adds financial and time costs that may be prohibitive in resource-limited settings. However, where available, CRP testing provides value by reducing unnecessary antibiotic prescriptions 2.
Integration with Clinical Assessment
CRP should never replace clinical judgment. The NICE guideline explicitly states that "clinical judgment is the most effective way to assess a patient's risk of severe disease" 1. CRP serves as an adjunct when clinical assessment alone leaves uncertainty about antibiotic necessity.
For patients with clinical pneumonia diagnosis, combine CRP results with CRB65 scoring (confusion, respiratory rate ≥30, blood pressure <90/60, age ≥65) to guide both antibiotic decisions and disposition (home vs. hospital) 1.