Plain chest x-ray (CXR) in the diagnosis of chronic obstructive pulmonary disease (COPD)

Abstract

Background: The diagnosis of COPD is based on spirometry1. A CXR can be reported as suggestive of COPD but is not diagnostic. This could inadvertently cause confusion or worry. Aim: We examined the frequency of COPD on CXR reports and whether this correlates with a spirometric diagnosis of COPD and the severity of COPD. Method: Patients were referred from general practice to a direct access spirometry service. All patients had a CXR within 3 months of spirometry. Patients were classified as not COPD, mild, moderate or severe COPD. X-ray reports were then reviewed. Results: 183 patients all over the age of 40 (mean 65.7yrs) were found. Spirometry classified them as not COPD (101), mild COPD (35), moderate COPD (37) or severe COPD (10). The COPD patients were more likely to be male 48/82 vs 42/101 and older 68 years vs 63.8 years. Average FEV1 of the COPD and non COPD groups were 1.94L (75% predicted) and 2.7L (104.6% predicted) respectively. Patients with COPD were more likely to have COPD mentioned on their CXR report, 29/82 vs 13/101. Chi2 12.95, p= 0.0003. There was no difference in CXR reporting between the mild (12/35), moderate (14/37) and severe (3/10) COPD subcategories. Chi2 0.24, p= 0.88. The sensitivity and specificity for CXR report as diagnostic for COPD is 35% and 87% respectively. Conclusions: Patients who have COPD were more likely to get a radiological diagnosis of COPD on CXR reporting. One third of patients with positive x-ray reports did not have COPD on spirometry. This has the potential to cause misdiagnosis and errors in care for a sizeable number of patients.