Clinical utility of C-reactive protein to predict treatment response during cystic fibrosis pulmonary exacerbations.

Ashutosh Sharma,Kate Skolnik,Gordon Kirkpatrick,Bradley S Quon,Virginia Chen,Zsuzsanna Hollander,Pearce Wilcox,Wei Shi

doi:10.1371/journal.pone.0171229

Abstract

RationaleC-reactive protein (CRP) is a systemic marker of inflammation that correlates with disease status in cystic fibrosis (CF). The clinical utility of CRP measurement to guide pulmonary exacerbation (PEx) treatment decisions remains uncertain.ObjectivesTo determine whether monitoring CRP during PEx treatment can be used to predict treatment response. We hypothesized that early changes in CRP can be used to predict treatment response.MethodsWe reviewed all PEx events requiring hospitalization for intravenous (IV) antibiotics over 2 years at our institution. 83 PEx events met our eligibility criteria. CRP levels from admission to day 5 were evaluated to predict treatment non-response, using a modified version of a prior published composite definition. CRP was also evaluated to predict time until next exacerbation (TUNE).Measurements and main results53% of 83 PEx events were classified as treatment non-response. Paradoxically, 24% of PEx events were characterized by a ≥ 50% increase in CRP levels within the first five days of treatment. Absolute change in CRP from admission to day 5 was not associated with treatment non-response (p = 0.58). Adjusted for FEV1% predicted, admission log10 CRP was associated with treatment non-response (OR: 2.39; 95% CI: 1.14 to 5.91; p = 0.03) and shorter TUNE (HR: 1.60; 95% CI: 1.13 to 2.27; p = 0.008). The area under the receiver operating characteristics (ROC) curve of admission CRP to predict treatment non-response was 0.72 (95% CI 0.61–0.83; p<0.001). 23% of PEx events were characterized by an admission CRP of > 75 mg/L with a specificity of 90% for treatment non-response.ConclusionsAdmission CRP predicts treatment non-response and time until next exacerbation. A very elevated admission CRP (>75mg/L) is highly specific for treatment non-response and might be used to target high-risk patients for future interventional studies aimed at improving exacerbation outcomes.

Highlights

Individuals with cystic fibrosis (CF) are prone to pulmonary exacerbations (PEx) with 45% of adults and 25% of children experiencing at least one event per year requiring IV antibiotics [1]
Adjusted for forced expiratory volume in 1 second (FEV1)% predicted, admission log10 C-reactive protein (CRP) was associated with treatment non-response (OR: 2.39; 95% confidence interval (CI): 1.14 to 5.91; p = 0.03) and shorter time until next exacerbation (TUNE) (HR: 1.60; 95% CI: 1.13 to 2.27; p = 0.008)
A very elevated admission CRP (>75mg/L) is highly specific for treatment non-response and might be used to target high-risk patients for future interventional studies aimed at improving exacerbation outcomes

Summary

Introduction

Individuals with cystic fibrosis (CF) are prone to pulmonary exacerbations (PEx) with 45% of adults and 25% of children experiencing at least one event per year requiring IV antibiotics [1]. Regardless of the underlying trigger, uncontrolled infection and inflammation play important roles in the pathophysiology of PEx in CF [8]. Both airway and systemic inflammatory responses are exuberant upon clinical presentation with a PEx and decrease significantly in response to antibiotic therapy [9,10,11,12]. Numerous studies have demonstrated that CRP is elevated at the onset of a PEx requiring IV antibiotics and decreases significantly by the end of treatment, corresponding to clinical improvement based on physician assessment [9, 10, 12,13,14,15,16,17,18,19,20,21,22,23]

Objectives

Methods

Results

Conclusion