Spirometric Volumes and Breathlessness across Levels of Airflow Limitation: The COPDGene Study

To the Editor:

Breathlessness is the cardinal symptom in people with chronic obstructive pulmonary disease (COPD) and is strongly associated with adverse health outcomes (1–3). Activity-related breathlessness, measured using the modified Medical Research Council (mMRC) breathlessness scale, is about twice as common among women as among men in the population for unclear reasons (1, 3).

Mechanistic and population-based data suggest that women are more prone to experience breathlessness in daily life owing to their lower absolute lung volumes and ventilatory capacity (3–6). No study has evaluated whether absolute lung function is related to breathlessness independent of the level of lung function impairment, and whether absolute volume may explain the higher breathlessness prevalence in women compared with men in COPD across severities of airflow limitation.

The aim of the present work was to evaluate whether the higher prevalence of breathlessness in women is related to their lower absolute spirometric lung volumes as measured by FEV₁, and whether this is seen both in people with normal lung function (FEV₁ ≥ 80% of predicted) and across severities of lung function impairment due to COPD.

Methods

This was a cross-sectional analysis of the COPDGene (Genetic Epidemiology of COPD) study (7). Inclusion criteria were as follows: age, 45–80 years; ethnic category, non-Hispanic white or African-American; and at least 10 years of smoking (COPD cohort) or no smoking (never-smoking control subjects), as detailed elsewhere (7). Exclusion criteria for the present analysis were as follows: missing data on spirometry (n = 63) or mMRC breathlessness score (n = 14).

Participants completed a modified American Thoracic Society Respiratory Questionnaire including smoking status, physician-diagnosed asthma, COPD, and emphysema. Activity-related breathlessness was self-rated on the mMRC scale (8). Post-bronchodilator spirometry was performed with an EasyOne spirometer (ndd Medizintechnik AG) according to American Thoracic Society standards (7). Predicted FEV₁ and FVC were calculated (9).

The sex difference in breathlessness was analyzed as the odds ratio (OR) of higher mMRC scores for women compared with men, using ordinal logistic regression. All models were adjusted for age, level of chronic airflow limitation (FEV₁/FVC), body mass index, pack-years of smoking, current smoking, and physician-diagnosed asthma, COPD, chronic bronchitis, emphysema, congestive heart failure, ischemic heart disease, and diabetes mellitus. No data were imputed. The impact of lung function on the sex difference in breathlessness was evaluated as the change in the sex estimate when adding each lung function measure to the fully adjusted model. Analyses were performed for the whole population, and in people with impaired lung function (FEV₁ < 80% of predicted) and normal lung function (FEV₁ ≥ 80% of predicted). Statistical analyses were conducted with SAS version 9.4 (SAS Institute, Inc.).

Results

In total, 10,223 participants were included, with a mean age of 59.6 years; 53.2% were men, and 4,935 (48.3%) had an FEV₁ less than 80% of predicted. Despite a similar FEV₁% predicted in men (76.2 ± 26.3%) and women (77.2 ± 24.8%), women had a mean 0.7-L lower absolute FEV₁ (1.9 ± 0.9 vs. 2.6 ± 1.0 L).

Higher absolute FEV₁ was associated with lower likelihood of breathlessness (Figure 1) (adjusted OR, 0.41; 95% confidence interval [95% CI], 0.37–0.44). This was consistent in both men (OR, 0.44; 95% CI, 0.40–0.49) and women (0.33; 95% CI, 0.28–0.38), and in people with normal lung function (OR, 0.53; 95% CI, 0.46–0.61) and people with impaired lung function (OR, 0.30; 95% CI, 0.26–0.35). Women were more likely to have breathlessness than men (46.1% vs. 37.4% had mMRC ≥ 2) (adjusted OR, 1.75; 95% CI, 1.62–1.90) (Table 1). Adding FEV₁% predicted to the model did not decrease the sex difference (adjusted OR, 1.72; 95% CI, 1.59–1.86). However, when instead adjusting for the absolute FEV₁, sex difference disappeared (adjusted OR, 0.93; 95% CI, 0.84–1.03) (P = 0.147). Adjusting for absolute FEV₁, breathlessness was similar in men and women with normal lung function, whereas breathlessness was even slightly lower among women in people with lung function impairment (Table 1). Using FVC instead of FEV₁, there was no sex difference even in people with impaired lung function (OR, 0.84; 95% CI, 0.71–1.01), and all other findings were similar when using FVC compared with FEV₁. Findings were robust when also adjusting for ethnicity, and when not including FEV₁/FVC in the models. Analysis using multinomial and dichotomous logistic regression had similar findings as reported with ordinal logistic regression.

Figure 1.

The relationship between absolute FEV₁ and breathlessness (mMRC ≥ 2) by sex in people with (A) normal FEV₁ (≥80% of predicted) and (B) impaired FEV₁ (<80% of predicted). Lower FEV₁ was associated with more breathlessness, both among men and women. When compared at a similar absolute FEV₁, breathlessness was similar between men and women. mMRC = modified Medical Research Council breathlessness scale.

Table 1.

FEV₁ and Sex Difference in Breathlessness

Models, All Adjusted for Confounders*	Sex Difference in mMRC Breathlessness Scale Score, for Women vs. Men [OR (95% CI)]
All participants (N = 10,223)
Without any FEV1 measure	1.75 (1.62–1.90)
With FEV1% predicted	1.72 (1.59–1.86)
With FEV1, L	0.93 (0.84–1.03)
Participants with FEV1 ≥ 80% predicted (n = 5,288)
Without any FEV1 measure	1.95 (1.73–2.19)
With FEV1, L	1.11 (0.94–1.31)
Participants with FEV1 < 80% predicted (n = 4,935)
Without any FEV1 measure	1.57 (1.41–1.75)
With FEV1, L	0.84 (0.73–0.95)

Definition of abbreviations: CI = confidence interval; mMRC = modified Medical Research Council; OR = odds ratio.

^*All models were adjusted for age, body mass index, pack-years of smoking, current smoking, FEV₁/FVC, asthma, chronic obstructive pulmonary disease or emphysema, chronic bronchitis, heart failure, cardiovascular disease, and diabetes mellitus.

Discussion: Main Findings

The key findings are that the markedly increased prevalence of breathlessness in women is related to their lower absolute lung function measured as FEV₁ or FVC. People with smaller spirometric lung volumes have a higher prevalence of breathlessness, both in men and in women. When matched on absolute spirometric lung volume, men and women have similar (or near similar) likelihoods of breathlessness both in people with normal lung function and in people with lung function impairment due to COPD.

This is the first study of breathlessness in relation to sex and spirometric lung volumes across severities of chronic airflow limitation, using data from the well-characterized COPDGene study. The likely mechanism underpinning the sex difference in breathlessness is that women have smaller airways and less respiratory musculature than men, even when matched for height and lung size, resulting in a lower ventilator capacity (4, 5). For a given level of work and ventilation, women experience more breathlessness as they use a greater fraction of their ventilatory capacity than men (4, 5). Despite similar lung function impairment (% of predicted) between the sexes, women may have markedly lower absolute FEV₁. This could explain the sex disparity in breathlessness seen in previous studies matching on the FEV₁% predicted (10). By matching on absolute spirometric volumes, this sex bias can be overcome.

Data were unavailable on breathlessness measured using a standardized exercise test, as well as on static lung volumes and diffusion capacity. The impact on the sex difference in breathlessness has been found to be similar for FEV₁ and FVC to that of static lung volumes (6).

The impact of a given lung function impairment on symptoms and function likely depends on the absolute lung volume and remaining ventilatory reserve. Relative and absolute lung volumes provide complementary information on the lung volume impairment and remaining ventilatory reserve and should be evaluated in both research and clinical care.