Introduction Patients with chronic obstructive pulmonary disease (COPD) may not recover swiftly from exacerbations. In 23% of patients, symptoms do not recover within 35 days.1 It would be valuable for clinicians to have a systemic biomarker that taken at initial consultation for an exacerbation could predict exacerbation length. We hypothesised that inflammation, measured by serum C-reactive protein (CRP) predicts the time to symptomatic resolution.
Methods We analysed daily diary cards prospectively collected from the London COPD cohort between October 1995 and January 2010. All patients had an FEV1≤80% predicted and an FEV1/FVC ratio ≤70%. An exacerbation was defined as an increase for 2 consecutive days in respiratory symptoms, with at least one major symptom (dyspnoea, sputum purulence or volume) plus another major or minor symptom (wheeze, cold sore throat and cough). Exacerbations were separated by five symptom free days. Patients were assessed and sampled at clinic visits within a median of 2 days of exacerbation onset. Exacerbation length was defined as the number of days from onset that increased airway symptoms were still recorded, and resolution time was defined as the time from clinic assessment to symptom cessation.
Results 268 patients with ≥365 days of diary card data experienced at least one exacerbation. Median baseline CRP was 4 mg/l. CRP at onset was available for 172 recovered exacerbations (Abstract P114 Figure 1). Mean age was 68.0 years (SD 8.3), with mean FEV1 1.11 l (0.45), FEV1 predicted 44.6% (16.2). Patients with a CRP<4 mg/l at initial clinic review had a significantly shorter resolution time than patients with CRP≥4 mg/l, median 6 (IQR 3-11; n=46) vs 10 (5–23; n=126) days, p=0.0017. Cox proportional hazard risk models with adjustment for repeated measures showed that CRP predicted both exacerbation length (p=0.039) and resolution time (p=0.029).
Conclusions COPD exacerbations can occur without an elevated systemic inflammatory response. These events are associated with a short resolution time. Thus, using CRP as a biomarker, it may be possible to identify patients that have a short recovery period and potentially may not require systemic treatment for their exacerbation.