Introduction Spirometry is required to make a clinical diagnosis of COPD, confirming persistent airflow limitation. However, studies suggest that making the correct diagnosis of COPD in primary care can be challenging. We sought to evaluate the clinical characteristics and management in primary care of patients registered with COPD but with incompatible spirometry, compared to patients with spirometry compatible with COPD.

Methods This analysis used data from the Welsh national COPD primary care audit, which prospectively collected data of patients, registered with COPD from 61% of GP practices in Wales covering January 2014-March 2015. Patients registered with COPD but with incompatible spirometry (post-bronchodilator forced expiratory lung volume in 1 s (FEV₁)/forced vital capacity (FVC) ≥0.70) were compared to COPD patients with compatible spirometry (FEV₁/FVC <0.70).

Results In total, 8957 patients registered with COPD were evaluated, 2255 (25%) had incompatible spirometry, 6702 had compatible spirometry. There were no differences in age (71±10 vs 70±11 years) or asthma co-diagnosis (13.1% vs 13.1%), between these respective groups, but patients with incompatible spirometry had a higher body mass index (29.41±6.69 vs 27.04±5.94 kg/m², p<0.001) and were more likely female (48.8% vs 45.6%, p=0.009). There were also differences in smoking status (14.3% vs 8.6% never-smokers, p<0.001), and spirometry [(FEV₁/FVC: 0.78 vs 0.55, p<0.001), (FEV₁% predicted: 72±18% vs 58±18%, p<0.001)], but similar levels of breathlessness (MRC: 2.52±0.94 vs 2.55±0.98, p=0.17) and exacerbation frequency (6.86±7.5 vs 7.3±8.8, p=0.05). Despite incompatible spirometry, 30.4% of these patients received long acting beta agonist (LABA) therapy, 51.7% inhaled corticosteroids (ICS), 73% long acting muscarinic antagonist (LAMA), and 74.6% combined inhaled LABA/ICS therapy. Furthermore, there were no differences for LAMA (73% vs 73%, p=0.99) or LABA/ICS (76.7% vs 74.3%, p=0.37) use in patients with incompatible spirometry with or without a diagnosis of asthma respectively, although more patients with asthma received LABA (38.9% vs 29.1%, p=0.001) and ICS (57.1% vs 50.8%, p=0.05) therapies.

Conclusions Patients without persistent airflow limitation diagnosed incorrectly with COPD, are symptomatic and receiving inappropriate pharmacological therapies. These data suggest that a breathlessness pathway may be helpful to aid diagnosis and management of such patients seen in primary care.