Exercise oscillatory ventilation disturbances: finding order amongst chaos

Abstract

Abstract Background Exercise Oscillatory Ventilation (EOV) during Cardiopulmonary Exercise Test (CPET) predicts prognosis in patients with Heart Failure (HF). In these patients, O2 consumption (VO2) oscillations have also been described, possibly secondary to circulatory delay. We hypothesize that in clinically meaningful EOV, cardiac output variation is mirrored by VO2 oscillation, which is then chronologically followed by a similar oscillation in minute ventilation (VE) (Figure 1). Accordingly, we aimed to assess whether this new definition surpassed that of classical EOV. Methods This is a single-centre cohort study of consecutive patients undergoing CPET from 2016 to 2018. Patients with LVEF >50% were excluded. CPET was performed on a treadmill to the limit of tolerance. Data was collected as a rolling average of 20 seconds and a composite VE/time and VO2/time plot was created. Classical EOV was defined as three or more regular oscillations of the VE graph with a minimal average amplitude of five litters. The addition of exercise VO2-to-VE peak-to-peak ventilation asynchrony (EVA) to the previous criteria fulfilled the new definition. The primary endpoint was a composite of time to all-cause death, heart transplantation or HF hospitalization. Results Overall, 177 patients were enrolled (mean age 58±11 years, LVEF 34±9%), of whom 35 had EOV and 17 had EVA. Compared to those without EVA, patients with EVA had markers of more severe HF. During a median follow-up of 32 (21–42) months, 55 patients met the primary outcome (32 all-cause deaths, 15 heart transplants, 47 HF hospitalizations). In multivariate analysis, EVA was associated with a 2.5-fold increased risk of events (HR 2.489; 95% CI: 1.302–4.759; p=0.006), adjusted for peak VO2, VE to CO2 production ratio (VE/VCO2 slope) and LVEF. EVA outperformed EOV in predicting the primary endpoint at 1 year, with a similar sensitivity and higher specificity (96.2 vs. 83.2%). The rate of events between the subgroup of patients without EVA was similar regardless of presence of EOV, contrasting with a higher rate in the EVA subgroup (Figure 2). Conclusion EVA is a strong predictor of hard outcomes in a broad population with HF. The new definition may outperform that of classical EOV. The incidence and prognostic value of EVA in the management algorithm and risk stratification of patients with HF is worth being further explored. Funding Acknowledgement Type of funding sources: None.