Author + information
- Catherine F. Notarius, PhD∗ (, )
- Philip J. Millar, PhD,
- Hisayoshi Murai, MD, PhD,
- Beverley L. Morris, RN and
- John S. Floras, MD, DPhil
- University Health Network and Mount Sinai Hospital Division of Cardiology, University of Toronto, Toronto, Ontario, Canada
- ↵∗Toronto General Hospital, 6ES-414, 200 Elizabeth Street, Toronto, Ontario, M5G 2C4, Canada
To the Editor:
Muscle sympathetic nerve activity (MSNA), as measured during supine rest, is similar between patients with chronic heart failure (HF) due to left ventricular systolic function with relatively preserved capacity for exercise (percentage of peak oxygen uptake [VO2peak] predicted by age, sex, and weight >56%) and healthy control subjects but is augmented in those patients whose peak VO2 is <56% of predicted (1). We previously identified an inverse relationship between resting calf MSNA and VO2peak in patients with HF but not in age-matched healthy untrained control subjects (2). This relationship was specific to skeletal muscle sympathetic nerve traffic; there was no correlation between cardiac norepinephrine spillover and VO2peak (3). In patients with HF performing dynamic handgrip exercise, fibular MSNA increased. Sympathetic activation was elicited at a lower-intensity threshold than in age-matched control subjects, was greatest in those with low VO2peak, and bore no relationship to left ventricular ejection fraction (LVEF) (1).
To date, there have been no published reports of sympathetic recordings from the fibular nerve of patients with HF during leg exercise. Thus, whether peripheral sympathetic vasoconstriction elicited by exercise could limit the capacity to exercise by restricting skeletal muscle blood flow or normal blood flow redistribution is as yet unknown. We hypothesized that VO2peak is a function of MSNA elicited by moderate dynamic leg exercise. To test this hypothesis, we recruited a cohort comprising subjects both with and without HF.
We studied 11 patients with HF (61 ± 3 years of age [mean ± SE]; 2 women) with a mean LVEF of 32 ± 2% and 11 healthy control subjects (55 ± 2 years of age; 2 women) on 2 separate days. Inclusion criteria were sinus rhythm, diagnosis of HF due to left ventricular systolic dysfunction and LVEF <40%. Those with diabetes were excluded. Patients were maintained on stable optimum therapy for HF. All received beta-adrenoceptor antagonists. No subject was participating in an exercise training program. On the first day, VO2 was assessed on a cycle ergometer during a 15 W/min ramped protocol to peak effort and expressed as a percentage of that predicted based on age, sex, and weight (3). On the second day, we recorded MSNA by microneurography (2) (left fibular nerve) at rest and during 1-legged cycling (right leg) for 4 min (2 at 0 load and 2 at 50% VO2peak) and determined MSNA frequency (bursts/min) and incidence (bursts/100 heart beats). Heart rate, blood pressure, and rating of perceived exertion (RPE; Borg scale 0 to 10) were also assessed. Multiple linear regression was performed with exercise MSNA burst frequency (minute 4 of exercise at 50% VO2peak) and resting MSNA as the independent variables and percent of predicted VO2peak achieved as the dependent variable (SigmaStat, version 3.5, Systat Software Inc., Chicago, Illinois).
For patients with HF and control subjects, respectively, mean age, weight (76.6 ± 3.1 kg vs. 79.3 ± 2.8 kg), body mass index (27.0 ± 0.8 kg/m2 vs. 26.2 ± 0.9 kg/m2), resting heart rate (60.0 ± 2.4 beats/min vs. 66.0 ± 3.2 beats/min), blood pressure (111 ± 4/66 ± 2 mm Hg vs. 114 ± 4/69 ± 2 mm Hg) were not significantly different between groups. Also, there was no significant difference in MSNA burst frequency between groups (50.6 ± 2.7 bursts/min vs 44.3 ± 2.6 bursts/min; p = 0.12). VO2peak was significantly lower in patients with HF, whether adjusted for weight (19.0 ± 2.3 vs. 32.7 ± 3.2 ml/kg·min; p = 0.004) or normalized as percent of predicted VO2 achieved (71 ± 8% vs. 117 ± 9%; p = 0.002). The mean heart rate response during the second minute of cycling at 50% VO2peak was similar in both groups (HF +13.5 ± 2.1 beats/min vs. control +17.4 ± 2.7 beats/min; p = 0.27), and mean RPE, equivalent to a moderate work rate, was comparable (HF 4.2 ± 0.5 vs. control 4.0 ± 0.5).
Multiple linear regression analysis demonstrated a significant inverse relationship between exercise MSNA (minute 2 at a work rate of 50% VO2peak) and VO2peak (% predicted) across a broad range (31% to 173%) of the latter variable that was not influenced by resting values for MSNA: VO2peak predicted (%) = 160.229 − (0.0816 × resting MSNA burst frequency) − (1.303 × exercise MSNA burst frequency) (r = −0.59; p = 0.02) (Fig. 1). VO2peak percent predicted also correlated with the absolute change in MSNA burst frequency elicited by exercise (r = −0.59; p = 0.02) (not shown).
The novel finding presented in this correspondence is that in middle-aged subjects, peak exercise capacity relates inversely (and independently of resting MSNA) with the magnitude of MSNA elicited by moderate-intensity leg cycling exercise: approximately one-third of the predicted VO2peak could be attributed to exercise-induced MSNA. This also represents the first report of fibular MSNA recorded during contralateral dynamic leg exercise in patients with HF. An augmented neurogenic vasoconstrictor response to dynamic exercise in patients with HF, as has been demonstrated for handgrip (1), could impair exercise capacity by limiting muscle blood flow or altering its distribution. Whether exercise training attenuates exercise MSNA of patients with HF merits investigation.
Please note: This study was supported by Grants-in-Aid from the Heart and Stroke Foundation of Ontario (T4938, NA6298). The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Notarius C.F.,
- Atchison D.J.,
- Floras J.S.
- Notarius C.F.,
- Ando S.,
- Rongen G.A.,
- Floras J.S.
- Notarius C.F.,
- Azevedo E.R.,
- Parker J.D.,
- Floras J.S.