Evaluating Arrhythmia Risk in Patients with Chronic Obstructive Pulmonary Disease Treated with Aclidinium/Formoterol Fumarate and Other Inhaled Bronchodilators: A Post-Authorization Safety Study.

The study included a large population-based cohort of COPD patients initiating inhaled bronchodilators in the UK.

• Total of 248,148 initiators were included in the study.
• Data sourced from the UK Clinical Practice Research Datalink Aurum database.
• Study period spanned January 2015 to March 2021.
• Adults with COPD initiating COPD medications were eligible for inclusion.

Aclidinium/formoterol showed the highest cardiac arrhythmia risk compared with current LABA use among comparator groups.

• Adjusted IRR for CA for aclidinium/formoterol versus current LABA use was 2.08 (95% CI: 1.36–3.18).
• Adjusted IRR for AF for aclidinium/formoterol versus current LABA use was 1.85 (95% CI: 1.15–3.00).
• These were the highest IRRs observed across comparator groups for both CA and AF endpoints.
• LAMA/LABA had the lowest IRRs versus current LABA use: 0.98 (95% CI: 0.69–1.41) for CA and 0.83 (95% CI: 0.55–1.24) for AF.

During the first episode of current single use, aclidinium showed numerically elevated but statistically non-significant risks of CA and AF compared with other LAMAs.

• Adjusted IRR for CA for aclidinium versus other LAMAs was 1.46 (95% CI: 0.93–2.29).
• Adjusted IRR for AF for aclidinium versus other LAMAs was 1.57 (95% CI: 0.94–2.62).
• Both confidence intervals crossed 1.0, indicating results were not statistically significant.
• Analysis was restricted to the first episode of current single use.

During the first episode of current single use, aclidinium/formoterol showed a statistically significant increased risk of CA but not AF compared with LAMA/LABA.

• Adjusted IRR for CA for aclidinium/formoterol versus LAMA/LABA was 2.15 (95% CI: 1.33–3.49), statistically significant.
• Adjusted IRR for AF for aclidinium/formoterol versus LAMA/LABA was 1.79 (95% CI: 0.96–3.33), confidence interval crossing 1.0.
• Analysis was conducted for the first episode of current single use.

Serious ventricular arrhythmia events were rare across all study groups.

• The abstract notes 'there were few SVA events,' limiting the ability to draw conclusions about SVA risk.
• SVA was assessed as one of three primary endpoints alongside CA and AF.
• Poisson regression models were used for estimation of SVA incidence rate ratios as with the other endpoints.

Increased risks of CA and AF for several medications relative to comparators may be driven by differences in baseline characteristics rather than drug effects.

• The authors noted that 'increased risks of CA and AF for several medications relative to LABA, LAMA, or LAMA/LABA may be driven by differences in baseline characteristics (eg, COPD severity).'
• Poisson regression models adjusted for clinically relevant covariables.
• LABA was used as a comparator for continuous current use analyses; LAMA and LAMA/LABA were used as comparators for single-use first-episode analyses.
• Residual confounding by COPD severity and other baseline factors is acknowledged as a potential explanation for the elevated IRRs.

CA and AF risks were increased for most study medications when compared with LABA as the reference group.

• Adjusted IRRs for CA versus current LABA use ranged from 0.98 (95% CI: 0.69–1.41) for LAMA/LABA to 2.08 (95% CI: 1.36–3.18) for aclidinium/formoterol.
• Adjusted IRRs for AF versus current LABA use ranged from 0.83 (95% CI: 0.55–1.24) for LAMA/LABA to 1.85 (95% CI: 1.15–3.00) for aclidinium/formoterol.
• LABA users may represent a lower-severity reference population, contributing to apparent elevated risks in other drug groups.
• This was a post-authorization safety study program examining cardiovascular safety of aclidinium bromide monotherapy and aclidinium bromide/formoterol fumarate.