Supranormal lung function and incident atrial fibrillation.
Whether supranormal lung function is associated with incident atrial fibrillation (AF), and whether obstructive sleep apnoea (OSA) modifies this association, remain to be determined.
We analysed 2,408 participants from the Sleep Heart Health Study who were free of AF at baseline and had no spirometric airflow obstruction. The primary analysis classified forced expiratory volume in 1 second (FEV1) as below normal, normal, or supranormal using race-neutral Global Lung Function Initiative (GLI) Global (2022) reference equations. Categories defined by forced vital capacity (FVC) were examined in a secondary analysis. Associations with incident AF were evaluated using multivariable logistic regression to estimate odds ratios (ORs) and 95% confidence intervals (CIs). Firth penalised logistic regression and sensitivity analyses were used to assess the robustness of the findings. Interaction with OSA was explored.
Among 2,408 participants, 282 developed AF. Compared with normal FEV1, supranormal FEV1 was associated with lower odds of incident AF in the primary adjusted model (OR, 0.420; 95% CI, 0.198-0.888). The corresponding Firth estimate was 0.443 (95% CI, 0.199-0.869). Supranormal FVC showed a similar association in standard logistic regression (OR, 0.351; 95% CI, 0.167-0.736) and Firth penalised logistic regression (OR, 0.371; 95% CI, 0.168-0.721). The direction of the associations was generally consistent across sensitivity analyses. There was no statistically significant evidence of effect modification by OSA.
Supranormal lung function was associated with lower odds of incident AF. These findings support supranormal spirometry as a potential marker of favourable systemic and cardiopulmonary physiology.
We analysed 2,408 participants from the Sleep Heart Health Study who were free of AF at baseline and had no spirometric airflow obstruction. The primary analysis classified forced expiratory volume in 1 second (FEV1) as below normal, normal, or supranormal using race-neutral Global Lung Function Initiative (GLI) Global (2022) reference equations. Categories defined by forced vital capacity (FVC) were examined in a secondary analysis. Associations with incident AF were evaluated using multivariable logistic regression to estimate odds ratios (ORs) and 95% confidence intervals (CIs). Firth penalised logistic regression and sensitivity analyses were used to assess the robustness of the findings. Interaction with OSA was explored.
Among 2,408 participants, 282 developed AF. Compared with normal FEV1, supranormal FEV1 was associated with lower odds of incident AF in the primary adjusted model (OR, 0.420; 95% CI, 0.198-0.888). The corresponding Firth estimate was 0.443 (95% CI, 0.199-0.869). Supranormal FVC showed a similar association in standard logistic regression (OR, 0.351; 95% CI, 0.167-0.736) and Firth penalised logistic regression (OR, 0.371; 95% CI, 0.168-0.721). The direction of the associations was generally consistent across sensitivity analyses. There was no statistically significant evidence of effect modification by OSA.
Supranormal lung function was associated with lower odds of incident AF. These findings support supranormal spirometry as a potential marker of favourable systemic and cardiopulmonary physiology.