Diagnostic utility of CRP to neopterin ratio in patients with acute respiratory tract infections

Abstract

Objectives: In this study we aimed to investigate the roles of neopterin, C-reactive protein (CRP) and the CRP to neopterin (C/N) ratio to differentiate bacterial from viral aetiology in patients with suspected acute respiratory tract infections (ARTIs) presenting to the emergency department (ED). Methods: Serum was taken from five hundred and sixty-one patients and used to measure neopterin and CRP levels. The primary outcome was bacterial or viral infection based on positive bacterial culture and positive viral serology. Patients were classified as either: group 1 with positive bacterial culture and mixed bacterial/viral growth; group 2 with virological aetiology, and group 3 with unknown microbiological aetiology. Results: The median of the C/N ratio was 10 times higher in patients with bacterial aetiology than with viral aetiology (12.5 vs 1.2 mg/nmol; P < 0.0001), and 42 times higher than those in healthy subjects (12.5 vs 0.3 mg/nmol; P < 0.0001). The area under the receiver-operator characteristic curve for the C/N ratio was 0.840 (0.783-0.898; P < 0.05). A cut-off value of "C/N ratio >3" for ruling in/out bacterial/viral infection yielded optimal sensitivity and specificity of 79.5% and 81.5% respectively. A sensitivity analysis performed on all patients (including unknown aetiology) with a cut-off value of "C/N ratio >3" yields a best-case scenario for ruling in/out bacterial/viral infection with sensitivity of 93.1% and specificity of 93.0%. Conclusion: This study shows that CRP and neopterin have a role in differentiating bacterial from viral causes of ARTI, and the C/N ratio yields optimal differentiation in the ED setting. © 2008 The British Infection Society.