# Characterization of dysfunctional breathing using cardiopulmonary exercise testing

## TL;DR

This study shows that dysfunctional breathing is a common cause of unexplained shortness of breath and can be identified using cardiopulmonary exercise testing.

## Contribution

The study characterizes the prevalence and impact of different dysfunctional breathing patterns in patients with unexplained dyspnoea using CPET.

## Key findings

- Dysfunctional breathing was identified in 42.5% of patients with unexplained dyspnoea.
- Over half of patients with dysfunctional breathing had isolated breathing pattern disorder.
- Patients with dysfunctional breathing had significantly reduced peak VO2 compared to those with normal CPETs.

## Abstract

Cardiopulmonary exercise testing (CPET) is emerging as a useful tool in the identification of dysfunctional breathing (DB). We aimed to evaluate the prevalence and functional impact of different patterns of DB in 628 adult patients referred for CPET due to unexplained dyspnoea (August 2019–December 2023). Patients were assigned to four groups following CPET interpretation: normal, breathing pattern disorder (BPD), hyperventilation (HV), and combined BPD with HV (BPDHV). Demographic and CPET performance data were analyzed using non‐parametric tests as appropriate. 94 (15.0%) patients had normal CPETs and 267 (42.5%) were identified as having DB. The remaining 267 were excluded as having alternative diagnoses. Of those with DB, 145 (54.3%) had BPD, 41 (15.4%) had HV, and 81 (30.3%) had BPDHV. VE/VCO2 was significantly increased in HV or BPDHV only (p < 0.001). Patients in all three DB groups exhibited significantly impaired peak VO2 compared to those with normal CPETs (p < 0.001). These CPET findings highlight DB as a common driver of symptoms in unexplained dyspnoea. Over half of patients with DB had isolated BPD, which requires visual inspection of relevant CPET plots to diagnose. Those identified with DB had significantly reduced peak VO2, which may be a useful classifier of functional severity in DB.