Autonomic modulation in heart failure patients by cardiopulmonary rehabilitation: who benefits?

Patients with heart failure and reduced ejection fraction (HFrEF) exhibit sympathetic activation and exercise intolerance,1,2 both of which independently predict foreshortened survival.3,4 We reported previously that their muscle sympathetic nerve activity (MSNA) is elevated at rest1 and, unlike healthy individuals, also during dynamic leg exercise.5 Both values related inversely to peak oxygen uptake (V̇O2peak),1,6 and fell after 6 months of exercise-based cardiac rehabilitation.7 However, sympatho-inhibition was not evident in all participants.7 Such autonomic variability suggests that conventional programmes benefit some with HFrEF more than others. Which patients derive the most autonomic benefit from exercise training is presently unknown, and identified in our recent review as an important HFrEF research question.8 We tested the hypothesis that those HFrEF patients with V̇O2peak less than the cohort median, and thus higher MSNA during exercise, would exhibit a greater training-induced reduction in sympathetic discharge during mild-to-moderate one-leg cycling than those whose V̇O2peak exceeded this median. We studied 22 stable HFrEF patients (6 women) referred to our Cardiovascular Prevention and Rehabilitation Programme; 21 in sinus rhythm and 1 in atrial fibrillation. Seventeen carried the diagnosis of ischaemic and 5 of dilated cardiomyopathy. Mean left ventricular ejection fraction (LVEF) was 29 ± 2% (range 12–40). For clinical relevance, participants continued with all prescribed therapy: angiotensin-converting enzyme inhibitors (n = 16; 73%); beta-adrenoceptor antagonists (n = 21; 95%); diuretics (n = 14; 64%); aspirin (n = 15; 68%), anticoagulants (n = 12; 55%); and statins (n = 14; 64%). This study complies with the Declaration of Helsinki, was approved by the Research Ethics Board of University Health Network, and informed written consent obtained from all participants. V̇O2peak was assessed on a cycle ergometer (17 watts/min) by open-circuit spirometry (Sensormedics Vmax Encore Metabolic Cart, Yorba Linda, CA, USA) and was expressed both as mL/kg/min and as per cent of the age, sex, body weight, and height-predicted V̇O2peak.9 Based on the median of 74%, patients were divided into 2 groups: 13 with V̇O2peak >74% (HIGH V̇O2peak) and 9 with <74% (LOW V̇O2peak). On a laboratory day soon after, heart rate (HR; electrocardiogram), blood pressure (Dinamap Pro 100; Critikon, Tampa, FL, USA), and multiunit post-ganglionic MSNA (microneurography; right fibular nerve) were recorded, as previously described.1 Signals were acquired at rest and during left-leg cycling for 4 min (2 min at zero load and 2 min at 50% of the work rate at V̇O2 peak, but halved, since only one leg exercised). Rating of perceived exertion (RPE; modified Borg scale 0–10) was assessed during exercise. Participant pre-training values were published recently as part of a larger cohort.10 Participants were studied before and after the standard 6 months Toronto Rehabilitation Institute cardiac programme. Training consisted of walking five times per week, at a heart rate of 60–70% V̇O2peak. This was adjusted weekly to maintain the same relative training intensity with an initial duration of 15 min increased to 1 h sessions. Upper and lower body resistance training was added twice a week as tolerated. Patient characteristics of the HIGH and LOW V̇O2peak HFrEF groups, pre- and post-training are summarized in Table 1. The two groups were similar with respect to age (63 ± 3 vs. 63 ± 3 years), LVEF (30 ± 3 vs. 27 ± 3%), body mass index, resting HR, blood pressure, and MSNA burst frequency (52 ± 3 vs. 50 ± 2 bursts/min), the principal neural correlate of norepinephrine release. A stable nerve site was maintained during exercise both before and after training in 11 of 13 patients in the HIGH V̇O2peak group and all 9 within the LOW V̇O2peak group. Representative MSNA recordings at rest and during exercise in the HIGH and LOW groups, pre- and post-training, are illustrated in Figure 1. Representative 15 s samples of multiunit MSNA recordings pre- and post-training in one heart failure and reduced ejection fraction patient in each of the HIGH (top) group and LOW V̇O2peak (bottom) group, recorded in the right leg at rest and during left-leg cycling. A stationary leg is required for microneurography. Patient characteristics of the HIGH and LOW V̇O2peak heart failure and reduced ejection fraction groups, pre- and post-training Values are given as mean ± standard error. Two-way analysis of variance indicated similar age, BMI, heart rate, blood pressures, and MSNA between groups pre-training. For V̇E/V̇CO2 slope and AT, reported data was in eight individuals within each group with no significant between-group or training effect for either variable. MSNA tended to be lower post-training in the HIGH group only (*P = 0.002 vs. pre-training). Training increased V˙O2peak in both groups (main effect P = 0.004); group differences in pre-training V˙O2peak were by design (**P = 0.001 vs. HIGH group). AT, anaerobic threshold as percentage of V˙O2peak⁠; BMI, body mass index; MSNA, muscle sympathetic nerve activity; V˙O2peak⁠, peak oxygen uptake; V˙O2peak slope, slope of ventilation and carbon dioxide output. Training increased V̇O2peak in both groups (main effect P = 0.004). As expected, HR increased during one-leg cycling proportionate to exercise intensity (P < 0.001); this response was not altered by training (Figure 2, upper panels). Similarly, RPE values during loaded cycling were comparable between groups and not significantly affected by training (pre-training HIGH 3 ± 0.4; LOW 4 ± 0.4; post-training HIGH 4 ± 0.3; LOW 4 ± 0.5; interaction P = 0.23). The MSNA burst frequency, at rest, decreased after training in the HIGH but not LOW group (P = 0.002; Table 1). During cycling, MSNA increased in both HIGH (P = 0.04) and LOW (P < 0.001) groups but was blunted post-training, specifically during zero load cycling, only in the HIGH group (P = 0.04 vs. 0.60 in LOW). Training-induced sympatho-inhibition during the recovery period immediately following exercise also was evident in the HIGH but not the LOW V̇O2peak subgroup (interaction P = 0.02; Figure 2, lower panels). Heart rate and muscle sympathetic nerve activity during exercise and recovery. Training did not alter heart rate at rest, during exercise or recovery in either group (upper panels). Muscle sympathetic nerve activity during exercise was significantly increased (main effect of exercise intensity) in both HIGH (P = 0.04) and LOW (P < 0.001) groups during cycling as expected. However, the muscle sympathetic nerve activity response during exercise and recovery was blunted post-training (main effect of training) in the HIGH (P = 0.04) but not the LOW group (P = 0.60; lower panels). The impact of exercise training varies considerably between individuals; for most previous outcomes evaluated, the greater benefit is generally observed in those with the lowest initial V̇O2peak.11 In HFrEF, V̇O2peak is inversely associated with MSNA during mild dynamic leg exercise, independently of resting MSNA,6 suggesting that those patients with the lowest exercise capacity would also exhibit the greatest autonomic benefit. However, the principal novel finding from this post-hoc analysis refutes this hypothesis: patients with HIGH and LOW V̇O2peak had similar pre-training resting MSNA, all trained at the same relative intensity within a conventional 6 months exercise-based cardiac rehabilitation programme. Training increased V̇O2peak significantly and similarly in both subgroups, yet only elicited sympathetic inhibition at rest and during and after one-leg cycling in those with relatively high or normal exercise capacity. This distinction between groups was clear despite the relatively low number of patients studied. This finding suggests the presence of a threshold V̇O2peak, below which the augmentation of exercise capacity by training is insufficient to induce concurrently beneficial autonomic modulation, or that a critical absolute exercise intensity is required before potential benefit, such as a reduction in exercise MSNA, is realisable, or both. This absolute intensity would be more difficult to reach if the initial V̇O2peak is low. In the present study, mean V̇O2peak in the LOW group was half that of the HIGH group (98 vs. 49% of predicted V̇O2peak). Translating ramped exercise test results into an effective constant work rate aerobic exercise intensity prescription remains a challenge in this population.12 Higher intensity training strategies, such as high intensity interval training 13 or alternate one-leg cycling training,14 if better tolerated and sustained by HFrEF patients with very low exercise capacity may permit those with low initial V̇O2peak also to secure autonomic benefit. C.F.N.: Conception and design; data acquisition, analysis, and interpretation; drafted and revised manuscript, prepared tables; approval and accountability. D.A.K.: Data acquisition and interpretation; critical review of manuscript; approval and accountability. M.B.B.: Data acquisition and interpretation; critical review of manuscript; approval and accountability. P.J.M.: Data acquisition and interpretation; critical review of manuscript; funding acquisition; approval and accountability. P.O.: Data acquisition and interpretation; supervision; critical review of manuscript; approval and accountability. J.S.F.: Conception and design; data acquisition, analysis, and interpretation; supervision; critical review of manuscript, funding acquisition; approval and accountability. The authors thank the technical and administrative support of Beverley Morris RN. This work was supported by operating grants from the Heart and Stroke Foundation of Canada [Grants T4938, NA6298 to J.S.F.], the Canadian Institutes of Health Research [Grant PJT148836 to J.S.F.], and the Natural Science and Engineering Research Council of Canada [Grant 06019 to P.J.M.]. D.A.K. and M.B.B. were both recipients of a post-doctoral fellowship from the Canadian Institutes of Health Research. J.S.F. held the Tier 1 Canada Research Chair in Integrative Cardiovascular Biology from the Canada Research Chairs. Data is available upon request.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.