Spirometric pattern and cardiovascular risk: a prospective study of 0.3 million Chinese never-smokers

Abstract

Background: Existing studies have not provided robust evidence about the CVD risk of non-smoking patients with restrictive spirometric pattern (RSP) or airflow obstruction (AFO), and how the risk is modified by body shape. We aimed to bridge the gap.

Methods: We used never-smokers' data from the China Kadoorie Biobank (CKB) and performed Cox models by sex (278,953 females and 50,845 males). Spirometry was used to assess the baseline spirometric pattern. CVD outcomes were captured through multiple sources.

Findings: Females' results were presented primarily, though males' results were similar. During a median 12-year (maximum 14.5 years) follow-up, both RSP and AFO patients had increased risks of acute myocardial infarction (AMI), other ischaemic heart disease (other IHD), heart failure, pulmonary heart disease, arrhythmia, and intracerebral haemorrhage (ICH). RSP was also associated with ischaemic stroke (IS). The HRs (95% CIs) for AFO in females ranged from 1.29 (1.15-1.45) for ICH to 8.84 (7.79-10.03) for pulmonary heart disease, while those for RSP ranged from 1.11 (1.08-1.15) for IS to 3.17 (2.80-3.59) for pulmonary heart disease. These risks increased with the severity of AFO and reduced FVC. RSP/AFO was more strongly associated with other IHD, heart failure, and pulmonary heart disease in underweight females than in normal and obese counterparts, respectively.

Interpretation: With the confounding of smoking fully controlled, both RSP and AFO were associated with higher risks of various CVD outcomes, which further increased with the severity of AFO and reduced FVC. These associations were even stronger in underweight individuals.

Funding: National Natural Science Foundation of China, National Key Research and Development Program of China, Ministry of Science and Technology of the People's Republic of China, Kadoorie Charitable Foundation, UK Wellcome Trust, UK Medical Research Council, Cancer Research UK, and British Heart Foundation.

Keywords: Cardiovascular disease; Chronic obstructive pulmonary disease; Forced expiratory volume in one second; Obesity; Restrictive lung function; Underweight.

Fig. 1

Associations between spirometric pattern and CVD outcomes among females. AFO = airflow obstruction; CI = confidence interval; CVD = cardiovascular disease; FEV1 = forced expiratory volume in 1 s; FEV1%P = FEV1 percent predicted; FVC = forced vital capacity; FVC%P = FVC percent predicted; HR = hazard ratio; RSP = restrictive spirometric pattern. “Normal” refers to FEV1/FVC ≥ 0.7 and FVC ≥ 80% predicted; “With RSP” refers to FEV1/FVC ≥ 0.7 and FVC < 80% predicted; “With AFO” refers to FEV1/FVC < 0.7. Analyses of FVC%P and FEV1%P were conducted among participants with RSP or AFO, respectively. The HRs (95% CIs) in red were statistically significant after Bonferroni correction (<0.0062). Multivariable models were adjusted for age, level of education, occupation, marital status, household income, alcohol consumption, physical activity, intake frequencies of red meat, fruits, and vegetables, body mass index, waist circumference, fuel types currently used in cooking, fuel types currently used in heating, years of cooking with solid fuels, years of heating with solid fuels, stove ventilation in the baseline house, passive smoking, and family histories of heart disease and stroke.

Fig. 2

Associations between severity of reduced FVC and CVD outcomes among females without AFO. AFO = airflow obstruction; CI = confidence interval; CVD = cardiovascular disease; FEV1 = forced expiratory volume in 1 s; FVC = forced vital capacity; HR = hazard ratio. “Normal” refers to FEV1/FVC ≥ 0.7 and FVC z-score > −1.65; “Mild” refers to FEV1/FVC ≥ 0.7 and −2.5 ≤ FVC z-score ≤ −1.65; “Moderate” refers to FEV1/FVC ≥ 0.7 and −4 ≤ FVC z-score < −2.5; “Severe” refers to FEV1/FVC ≥ 0.7 and FVC z-score < −4. The HRs (95% CIs) in red were statistically significant after Bonferroni correction (<0.0062). Multivariable models were adjusted for age, level of education, occupation, marital status, household income, alcohol consumption, physical activity, intake frequencies of red meat, fruits, and vegetables, body mass index, waist circumference, fuel types currently used in cooking, fuel types currently used in heating, years of cooking with solid fuels, years of heating with solid fuels, stove ventilation in the baseline house, passive smoking, and family histories of heart disease and stroke.

Fig. 3

Associations between GOLD grades of AFO and CVD outcomes among females without RSP. AFO = airflow obstruction; CI = confidence interval; CVD = cardiovascular disease; FEV1 = forced expiratory volume in 1 s; FEV1%P = FEV1 percent predicted; FVC = forced vital capacity; GOLD = Global Initiative for Chronic Obstructive Lung Disease; HR = hazard ratio; RSP = restrictive spirometric pattern. “GOLD 1” refers to FEV1/FVC < 0.7 and FEV1%P ≥ 80%; “GOLD 2” refers to FEV1/FVC < 0.7 and 50% ≤ FEV1%P < 80%; “GOLD 3” refers to FEV1/FVC < 0.7 and 30% ≤ FEV1%P < 50%; “GOLD 4” refers to FEV1/FVC < 0.7 and FEV1%P < 30%. The HRs (95% CIs) in red were statistically significant after Bonferroni correction (<0.0062). Multivariable models took those with FEV1/FVC ≥ 0.7 and FVC ≥ 80% predicted (the “normal” group) as the reference, and were adjusted for age, level of education, occupation, marital status, household income, alcohol consumption, physical activity, intake frequencies of red meat, fruits, and vegetables, body mass index, waist circumference, fuel types currently used in cooking, fuel types currently used in heating, years of cooking with solid fuels, years of heating with solid fuels, stove ventilation in the baseline house, passive smoking, and family histories of heart disease and stroke.

Fig. 4

Associations between spirometric pattern and CVD outcomes stratified by body shape among females. AFO = airflow obstruction; BMI = body mass index; CI = confidence interval; CVD = cardiovascular disease; FEV1 = forced expiratory volume in 1 s; FVC = forced vital capacity; HR = hazard ratio; RSP = restrictive spirometric pattern; WC = waist circumference. In the spirometric pattern, “With RSP” refers to FEV1/FVC ≥ 0.7 and FVC < 80% predicted; “With AFO” refers to FEV1/FVC < 0.7. In the body shape, “Underweight” refers to BMI < 18.5 kg/m²; “Normal” refers to BMI 18.5–23.9 kg/m² and WC < 80/85 cm; “General or central obesity” refers to BMI ≥ 24 kg/m², or BMI 18.5–23.9 kg/m² and WC ≥ 80/85 cm. Multivariable models took those with FEV1/FVC ≥ 0.7 and FVC ≥ 80% predicted (the “normal” group) as the reference, and were adjusted for age, level of education, occupation, marital status, household income, alcohol consumption, physical activity, intake frequencies of red meat, fruits, and vegetables, fuel types currently used in cooking, fuel types currently used in heating, years of cooking with solid fuels, years of heating with solid fuels, stove ventilation in the baseline house, passive smoking, and family histories of heart disease and stroke.