Measures of low lung function and the prediction of incident COPD events and acute coronary events

Abstract

Background: Although reduced lung function is associated with both COPD and coronary events (CE), the pattern of lung function impairment could be different for the two outcomes. We examined different measures of lung function in relation to incident COPD events and CE in a population-based cohort.

Methods: Baseline spirometry and lung clearance index (LCI) were assessed in 672 men aged 55 years. Outcomes included incident COPD events and CE (hospitalisation or mortality). Cox regression was used to obtain HRs per 1-standard deviation (SD) decrement in baseline lung function. The Lunn-McNeil competing risks approach was used to assess if differences in risks for incident COPD events and CE were significant.

Results: Over 44 years follow-up there were 85 incident COPD events and 266 incident CE. Low FEV₁ and FEV₁/VC and high LCI showed significantly stronger relationships with COPD events than CE (adjusted HRs per 1SD decrement and p-value for equal associations: FEV₁; HR_COPD: 2.11 (1.66-2.68), HR_CE: 1.30 (1.13-1.49) p < 0.001, FEV₁/VC; HR_COPD 1.95 (1.60-2.36), HR_CE 1.11 (0.98-1.26) p < 0.0001, LCI; HR_COPD: 1.58 (1.26-1.98), HR_CE: 1.14 (1.00-1.31) p = 0.015. Low VC was significantly associated with both COPD and CE, but HRs were not significantly different between the outcomes (p-value for equal associations = 0.706).

Conclusions: Low FEV₁ and FEV₁/VC and high LCI at baseline show significantly stronger relationships with future COPD events than CE. Low VC at baseline is similarly associated with future COPD events and CE. This indicates differences but also an important similarity in the "lung function profile" for developing incident COPD events or incident CE later in life.

Keywords: Chronic obstructive pulmonary disease; Coronary heart disease; Incidence; Lung clearance index; Mortality; Spirometry.