Cardiac Reserve and Pulmonary Gas Exchange Kinetics in Patients With Stroke
Background and Purpose— Cardiovascular and pulmonary factors contributing to impaired peak oxygen uptake (V̇O2) in patients with stroke (SP) are not well known. We assessed cardiovascular function, pulmonary gas exchange, and ventilation in SP and healthy age, gender, and activity-matched control subjects.
Methods— Ten hemiparetic SP and 10 control subjects were enrolled. Subjects completed cycle ergometry testing to assess peak and reserve V̇O2, carbon dioxide production, ventilation (tidal volume; breathing frequency; minute ventilation), and cardiac output. V̇O2, carbon dioxide production, and minute ventilation were measured throughout peak exercise recovery (off-kinetics) and at exercise onset (on-kinetics) along with heart rate during low-level exercise.
Results— Peak V̇O2 was 43% lower (P<0.001) in SP secondary to reduced peak and reserve cardiac output and minute ventilation. The impaired cardiac output reserve (P<0.001) was due to a 34% lower heart rate reserve (P=0.001). The impaired minute ventilation reserve (P=0.013) was due to a 41% lower tidal volume reserve (P=0.009). Stroke volume and breathing frequency reserve were preserved. V̇O2 off-kinetics were 29% slower in SP (P<0.001) and related to peak V̇O2 (R=−0.72, P<0.001) and peak cardiac output (R=−0.75, P<0.001) for the study group. Additionally, carbon dioxide production (P=0.016) and minute ventilation (P=0.023) off-kinetics were prolonged in SP. V̇O2 on-kinetics were 29% slower (P=0.031) during low-level exercise in SP.
Conclusions— The impaired peak V̇O2 in SP is secondary to a decline in peak and reserve cardiac output and ventilation. Prolonged V̇O2 kinetics in SP are associated in part with deconditioning and may be mediated by reduced O2 availability and/or the rate of muscle O2 use.
The cardiopulmonary factors that contribute to the abnormal oxygen uptake (V̇O2) in patients with stroke (SP) are not well known. Although left ventricular function may be compromised in acute SP,1 cardiac function during exercise has not been evaluated. Circulatory impairments adversely impact peak V̇O2 in addition to the recovery of V̇O2 after exercise,2 yet this relationship has not been established in SP. Additionally, it is not known whether impairments in the rate of V̇O2 readjustment are present during low-level exercise comparable to activities of daily living in SP.
We tested the hypotheses that (1) peak and reserve V̇O2 and cardiac output () would be lower in SP; (2) pulmonary gas exchange and ventilation off-kinetics would be slower in SP; (3) slower V̇O2 off-kinetics would be associated with a lower peak V̇O2 and ; and (4) V̇O2 on-kinetics would be prolonged during low-level exercise in SP compared with healthy control subjects.
Materials and Methods
Ten hemiparetic SP and 10 healthy control subjects were enrolled (Table 1). This investigation was approved by the research ethics board at the University of Alberta and participants provided written and informed consent.
Testing was completed on a custom-modified recumbent cycle ergometer to assess peak and reserve (peak−rest) V̇O2, carbon dioxide production (V̇CO2), ventilation (tidal volume, Vt; breathing frequency; minute ventilation, V̇E; Parvomedics, Salt Lake City, Utah), and (Minnesota Impedance Cardiograph, model 304B; Surcom, Minneapolis, Minn). V̇O2, V̇CO2, and V̇E off-kinetics were determined. On a separate day, subjects completed 3 square-wave exercise protocols separated by 25 minutes to assess V̇O2, V̇CO2, V̇E, and heart rate (HR) on-kinetics. A 5-minute resting baseline was followed by 5 minutes of cycling at a power output approximating 80% of the ventilatory threshold.
Breath-by-breath off-kinetics data were averaged into 10-s bins for curve fitting and on-kinetics data from the 3 repeats were interpolated to 1-s intervals, time-aligned, averaged to yield a single response, and averaged into 5-s bins. V̇O2, V̇CO2, and V̇E off-kinetics (Eq 1) and on-kinetics (Eq 2) were determined using the following equations:
where Y is the parameter at any time (t), peak is the greatest 30-s value of Y, and rest is the value of Y over 1 minute before exercise onset, A is the amplitude change in Y, τ is the time to reach a 63% change in Y, and TD is the time delay before the exponential change in Y.
Phase II V̇O2, V̇CO2, and V̇E on-kinetics were determined and HR on-kinetics were measured from exercise onset. The best curve fit was defined by minimization of the residual sum of squares (Origin 7.5; OriginLab Corp, Northampton, Mass).
Analysis of variance was used for between-group comparisons and correlation regression to determine variable relationships. Data are mean±SEM and P<0.05 was significant.
Peak V̇O2, V̇CO2, V̇E, HR, stroke volume (SV), , systemic vascular resistance, and power output were lower in SP (Table 2). V̇O2, V̇E, Vt, HR, and , but not SV reserve were lower in SP (Table 2). Power outputs were lower at the ventilatory threshold in SP (45±4 versus 96±5 W; P<0.001).
V̇CO2, and V̇E off-kinetics were prolonged in SP (Figure A). V̇CO2 and V̇E off-kinetics were correlated (R=0.80, P<0.001) and V̇O2 off-kinetics related to peak V̇O2 (R=−0.72, P<0.001) and peak (R=−0.75, P<0.001).
The power output was lower during low-level exercise (26±3 versus 59±5 W; P<0.001) and V̇O2 on-kinetics slower in SP (Figure B). V̇CO2 (66±10 versus 62±6 s; P=0.800), V̇E (63±11 versus 66±6 s; P=0.764), and HR (45±9 versus 32±8 s; P=0.307) on-kinetics were not different between SP and healthy control subjects, respectively. There was no relation between time poststroke and V̇O2 on-kinetics (R=−0.60, P=0.064).
The major new findings of this investigation are that (1) the reduction in peak and reserve V̇O2 in SP is secondary to an impaired peak and reserve and V̇E; (2) prolonged V̇O2 off-kinetics in SP are related to a reduced peak V̇O2 and ; and (3) SP have slower V̇O2 on-kinetics during low-level exercise.
We demonstrate that impaired V̇O2 in SP is secondary to a lower peak and reserve . The impaired reserve was secondary to an impaired HR reserve, because SV reserve was not different between groups (Table 2). Our finding of a similar SV reserve despite a higher systemic vascular resistance in SP (Table 2) is likely due to an enhanced preload or contractile reserve. This is plausible given that peak exercise end systolic volume likely increased secondary to an increased afterload. Additionally, systolic peak velocity is inversely related to afterload. Thus, the lower peak and reserve HR (Table 2) in our SP would lead to a greater diastolic filling time, preserving preload and SV reserve.
Peak and reserve V̇E was lower in SP secondary to an impaired peak and reserve Vt, because peak and reserve breathing frequency were not different between groups (Table 2). Electromyographic activity is reduced in ventilatory muscles contralateral to the brain lesion side in SP and may be exacerbated during increases in Vt.3 Lanini and colleagues4 reported no asymmetrical effect of resting breathing on Vt, which is consistent with our findings (Table 2). However, Vt of the paretic chest wall may be reduced during hyperventilation.4 Peak and reserve Vt was lower in our SP, suggesting the normal increase in breathing frequency throughout exercise exacerbated asymmetrical ventilation, thus contributing to an overall lower peak and reserve Vt.
V̇O2 remained elevated throughout recovery in SP (Figure A), reflecting a slower rate of skeletal muscle energy restoration. This is determined in part by the oxidative capacity and maximal rate of oxidative adenosine triphosphate synthesis of skeletal muscle, which may be mediated by aerobic fitness and the availability of O2 for phosphocreatine regeneration.5 Indeed, a lower peak V̇O2 was associated with prolonged V̇O2 off-kinetics in our SP. Consistent with the hypothesis that O2 availability may be rate-limiting for V̇O2 recovery, peak and V̇O2 off-kinetics were well correlated.
V̇CO2 and V̇E off-kinetics were prolonged in SP (Figure A). Prolonged V̇CO2 kinetics are explained by a greater CO2 tension in the skeletal muscle secondary to buffering of exercise-induced lactate. This is consistent with the ventilatory threshold occurring at a lower exercise intensity in SP, suggesting a more immediate reliance on anaerobic glycolysis. The lower may also have exacerbated the delay in recovery V̇CO2. Resulting CO2 levels likely contributed to the high V̇E, which is consistent with the well-correlated V̇CO2 and V̇E off-kinetics.
V̇O2 on-kinetics were slower in SP (Figure B), which indicates a slower rate of muscle O2 consumption. Unfavorable changes in skeletal muscle function related to stroke may account for this observation, because Type I and II muscle fibers may be atrophied in SP.6 A consequence of the alteration in the Type I muscle fibers is an associated decline in mitochondrial density and oxidative capacity, which may account for the slower V̇O2 on-kinetics.
Slower V̇O2 on-kinetics may have been due to a decline in vascular function. Given that endothelial function may be impaired in SP, coupled with findings that V̇O2 kinetics and blood flow become impaired with peripheral arterial disease,7 supports this hypothesis. Further investigation is required to establish the role of blood flow dynamics and mitochondrial function on muscle O2 consumption in SP to further clarify the mechanisms responsible for our findings.
A limitation of our investigation is that comorbidities such as diabetes, hypertension, and chronic obstructive pulmonary disease may also prolong V̇O2 kinetics. Additionally, the years (≈7.5) poststroke may account for the slower V̇O2 kinetics simply because of deconditioning secondary to inactivity. However, we matched SP and healthy control subjects on activity levels and the time poststroke and V̇O2 kinetics were not related. Future investigations are required to establish the impact of comorbid conditions and stroke type on exercise cardiopulmonary function and V̇O2 kinetics in SP.
We thank the participants of this study for their time. We also thank the staff at the Steadward Centre and the Community Rehabilitation Interdisciplinary Program (CRIS) for their assistance in identifying participants.
Sources of Funding
This study was funded by a grant to P.J.M. from the EFF Support for Advancement of Scholarship Fund. C.R.T. is a Canadian Institutes of Health Research (CIHR) Strategic Training Fellow in TORCH (Tomorrow’s Research Cardiovascular Health Professionals) and is supported by a doctoral Canada Graduate Scholarship from the Natural Sciences and Engineering Research Council of Canada. M.J.H. is a CIHR New Investigator.
- Received January 18, 2008.
- Accepted March 19, 2008.
Mayer SA, Lin J, Homma S, Solomon RA, Lennihan L, Sherman D, Fink ME, Beckford A, Klebanoff LM. Myocardial injury and left ventricular performance after subarachnoid hemorrhage. Stroke. 1999; 30: 780–786.
Cohen-Solal A, Laperche T, Morvan D, Geneves M, Caviezel B, Gourgon R. Prolonged kinetics of recovery of oxygen consumption after maximal graded exercise in patients with chronic heart failure. Analysis with gas exchange measurements and NMR spectroscopy. Circulation. 1995; 91: 2924–2932.
Haseler LJ, Lin AP, Richardson RS. Skeletal muscle oxidative metabolism in sedentary humans: 31P-MRS assessment of O2 supply and demand limitations. J Appl Physiol. 2004; 97: 1077–1081.
Bauer TA, Brass EP, Nehler M, Barstow TJ, Hiatt WR. Pulmonary VO2 dynamics during treadmill and arm exercise in peripheral arterial disease. J Appl Physiol. 2004; 97: 627–634.