Exercise training effects on metabolic and ventilatory changes in heart failure patients with exercise oscillatory ventilation: systematic review and meta-analysis

Abstract

Exercise oscillatory ventilation (EOV) is a phenomenon characterized by cyclic oscillations in the ventilatory pattern observed during the cardiopulmonary exercise test (EOV characteristics: Supplementary material online, Figure 1S). It is considered an independent predictor for death and adverse cardiovascular events in chronic heart failure (CHF) patients. 1 A recent meta-analysis showed that EOV-positive patients exhibited a worse peak oxygen uptake (VO 2PEAK) and ventila-tory efficiency (VE/VCO 2 slope). 2 Although the EOV pathophysiology is not fully understood, current evidence suggests that circulatory delay, pulmonary congestion, high chemosensitivity, and exacerbated ergoreflex signalling are the main triggering factors. 1,3 Several treatments were investigated to alleviate EOV symptom-atology and thereby improve the patient's prognosis (e.g. adaptive servo-ventilation, phosphodiesterase 5 inhibition, exercise training and levosimendan infusion). Although the evidence regarding exercise training as EOV treatment is limited, there is also no effective pharmacological treatment for EOV. Thus, targeting EOV by exercise intervention is highly relevant. This systematic review aims to verify the exercise training effect on EOV reversal, and other prognostic factors in CHF patients and EOV coexistence. This study followed the recommendations of the PRISMA statement and was registered at PROSPERO: CRD42021254587. The search was performed in several databases from inception to 30 th July 2021, with no language restriction. Inclusion criteria followed the PICOT question: (P) EOV patients; (I) exercise training; (C) pre-training values in EOV group-single arm; (O) EOV reversal (pri-mary), VO 2PEAK and VE/VCO 2 slope (secondary); and (T) clinical trial: randomized, non-randomized, or uncontrolled trials. Full methods' description is available as Supplementary material online. Nine potentially eligible studies were identified, and three trials met the eligibility criteria (PRISMA flowchart; Supplementary material on-line, Figure 2S). Zurek et al. 4 assessed two non-randomized groups composed of EOV-positive patients (intervention and control group). Panagopoulou et al. 5 analysed two groups who performed the exercise training (EOV-positive and EOV-negative; uncontrolled trial), and Yamauchi et al. 6 evaluated a single group (uncontrolled trial). Only data referring to the EOV-positive intervention groups were analysed. The characteristics of the included studies and the methodological quality are shown in Supplementary material online, Tables S1 and S2, respectively. All studies (Table 1) included aerobic exercise three times per week: 30 or 40 min of cycling at the anaerobic threshold, 5,6 or 45-min cycling at 60-80% of VO 2PEAK (twice daily). 4 One study analysed two protocols, 5 40 min of high-intensity interval training (HIIT) on a cycle ergometer or 20-min of HIIT plus four resistance exercises (three sets of 10-12 repetitions at 55-65% of 2-repetition maximum). Besides, resistance training was proposed in the other two studies (three to four exercises). 4,6 These protocols agree with current guidelines for cardiovascular rehabilitation of CHF patients. 7 No study reported adverse events, major complications, or sample loss during the follow-up. VO 2PEAK and VE/VCO 2 slope values were available in three trials 4-6 (98 patients), and EOV reversal in two trials (72 patients). 4,5 A random-effects model with a 95% confidence interval was applied to assess the relative likelihood for EOV reversal, as well as to measure the exercise effect on VO 2PEAK and VE/VCO 2 slope [standardized mean difference (SMD)]. The mean difference analysis was also performed. The meta-analysis (Figure 1) showed a moderate effect on