Post-Pulmonary Embolism Phenotypes Described by Invasive Cardiopulmonary Exercise Testing

Background: Post-pulmonary embolism (PE) dyspnea is common. Existing noninvasive studies have demonstrated that post-PE dyspnea is associated with elevations in right ventricular (RV) afterload, dead space ventilation, and deconditioning. We aimed to use invasive cardiopulmonary exercise testing (iCPET) parameters in patients with post-PE dyspnea to identify unique physiologic phenotypes.

Research question: Do distinct post-PE dyspnea physiologic phenotypes exist that are described with iCPET?

Study design and methods: Patients were enrolled at the time of acute PE and through our pulmonary hypertension (PH) and dyspnea clinic. iCPET was performed if high suspicion existed for PH or if unexplained dyspnea was present. A hierarchical cluster analysis was performed to identify dyspnea phenotypes. iCPET parameters assessing pulmonary hemodynamics, ventilation, and peripheral oxygen use then were compared within and across each cluster and with iCPET control participants against peak oxygen consumed per minute (VO₂).

Results: One hundred seventy-three patients were enrolled. Sixty-seven patients underwent iCPET. All patients showed reductions in peak VO₂ and peak cardiac index relative to control participants. Three clusters were identified. Cluster 1 was defined by having elevated RV afterload and impaired ventilatory efficiency. Cluster 2 was defined by elevated RV afterload with reductions in respiratory mechanics. Cluster 3 was defined by mild derangement in RV afterload with mild reductions in peak cardiac output.

Interpretation: iCPET revealed significant heterogeneity in post-PE dyspnea. Three phenotypes are characterized by differences in RV afterload: ventilatory efficiency, respiratory mechanics, and peripheral oxygen use.