The first two authors contributed equally to this work. Disclosure: A.A. owns the patent rights for the FHR procedure (US 64,449,945 B1) and also owns 100% of SyGeSa, which owns the rights for the proprietary software used in the calculation of RUR.
The impact of metabolic syndrome and endothelial dysfunction on exercise-induced cardiovascular changes†
Article first published online: 16 MAR 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 1, pages E143–E148, January 2013
How to Cite
Rossi, A. M., Davies, E., Lavoie, K. L., Arsenault, A., Gordon, J. L., Meloche, B. and Bacon, S. L. (2013), The impact of metabolic syndrome and endothelial dysfunction on exercise-induced cardiovascular changes. Obesity, 21: E143–E148. doi: 10.1002/oby.20258
See the online ICMJE Conflict of Interest Forms for this article.
- Issue published online: 16 MAR 2013
- Article first published online: 16 MAR 2013
- Manuscript Accepted: 27 APR 2012
- Manuscript Received: 17 SEP 2011
There is limited information regarding the synergistic or additive effects of metabolic syndrome (MS) and endothelial dysfunction (ED) on cardiovascular disease (CVD). Altered cardiovascular responses to exercise have been shown to predict future cardiovascular events as well as assess autonomic function. The present study evaluated the impact of MS and brachial artery reactivity (a proxy of ED) on peak exercise-induced cardiovascular changes.
Design and Methods:
Individuals (n = 303) undergoing a standard nuclear medicine exercise stress test were assessed for MS. Participants underwent a Forearm Hyperaemic Reactivity test and were considered to have dysfunctional reactivity if their rate of uptake ratio (RUR) was <3.55. Resting and peak blood pressure (BP) and heart rate (HR) were measured. Reactivity was calculated as the difference between peak and resting measures.
Analyses, adjusting for age, sex, resting HR, total metabolic equivalents (METs), and a history of major CVD, revealed a main effect of MS (F = 5.51, η2 = 0.02, P = 0.02) and RUR (F = 6.69, η2 = 0.02, P = 0.01) on HR reactivity, such that patients with MS and/or poor RUR had reduced HR reactivity. There were no interactive effects of RUR and MS. There were no effects of RUR or MS on systolic BP (SBP) or diastolic BP (DBP) reactivity or rate pressure product (RPP) reactivity.
The presence of decreased HR reactivity among participants with MS or poor brachial artery reactivity, combined with the lack of difference in other exercise-induced cardiovascular changes, indicates that these patients may have some degree of parasympathetic dysregulation. Further longitudinal studies are needed to understand the long-term implications of MS and endothelial abnormalities in this context.
Cardiovascular disease (CVD) is responsible for approximately a third of all deaths in Canada each year, and incurs upwards of 18 billion dollars in annual health care costs (1,2). Metabolic syndrome (MS) and endothelial dysfunction (ED) are recognized as significant risk factors for CVD (3,4). However, the precise mechanisms by which MS and/or ED contribute to the development of CVD are still largely unknown. Exercise testing can be employed to evaluate cardiovascular abnormalities in patients with MS and/or ED and is often used as a diagnostic tool for latent CVD (5,6). The autonomic nervous system (ANS) is largely responsible for maintaining normal physiological functions, including regulating heart rate (HR) and blood pressure (BP) (7,8). Malfunctioning of this system has been found to have negative repercussions on cardiovascular health (9). Additionally, the ANS is crucial for readjusting and adapting these parameters when the cardiovascular system is perturbed, such as during exercise (10,11). Thus, various forms of exercise (e.g., aerobic and isometric) can be used to assess the performance of the ANS (12). Moreover, exercise has advantages over other methods of ANS measurement, such as microneurography, which are more involved, sometimes invasive, and costly (13).
The purpose of this study was to identify and better understand cardiovascular abnormalities in patients with MS and/or ED by examining cardiovascular changes during exercise stress testing. Previous studies have indicated a degree of ANS dysregulation in people with both ED and MS (4,14). However, to our knowledge, no studies have considered the interactive effects of these two conditions on cardiovascular responses to strenuous exercise.
Methods and Procedures
The data presented here is a subanalysis of the cross-sectional Mechanisms and Longitudinal Outcomes of Silent Myocardial Ischemia (MOSMI) study. The project recruited 904 patients that were referred for exercise stress testing using single photon emission computed tomography imaging between May 2005 and December 2006. All testing was performed in the Nuclear Medicine Department at the Montreal Heart Institute. There were no exclusion criteria for age, sex, or race; however, only patients fluent in French or English were eligible. Participants were also excluded if they were pregnant or nursing, had serious non-cardiovascular related disease (e.g., cancer or chronic obstructive pulmonary disease), suffered from chronic pain (other than angina), had used non-steroidal inflammatory drugs in the past week or used an analgesic on the testing day. Ethical approval for this study was obtained through the Human Ethics Committee of the Montreal Heart Institute, and written informed consent was obtained from each participant. A subsample of the patients participating in the MOSMI study were randomly selected to undergo forearm hyperaemic reactivity (FHR) testing, and it was with this cohort of 303 men and women (for whom we had complete data sets) that this analysis was performed. The participants in this cohort were referred to the Nuclear Medicine Department at the Montreal Heart Institute for cardiovascular testing, which suggests that participants possibly already had CVD or were at risk for CVD. No significant differences for age, sex, or history of CVD were observed between the subsample and those who did not undergo FHR testing.
All participants underwent a standard, medically required, single photon emission computed tomography exercise stress test according to standard procedures (15). This protocol consists of 2 days of testing. On day 1, the participants underwent an exercise stress test and scan. Before this test, participants were approached for the study. Following the exercise stress test, patients completed a series of questionnaires assessing psychological (e.g., Beck Depression Inventory, Anxiety Sensitivity Index, and Toronto Alexithymia Scale), socioeconomic (e.g., household composition, years of education, and family income), and medical (including information on medication usage) histories. On the second day of testing, participants completed a resting single photon emission computed tomography scan. Prior to the resting scan participants had a fasting blood sample drawn. Anthropometric measures including height, weight, and waist circumference were taken. Waist circumference was measured at the top of the iliac crest. Additionally, participants completed the FHR test.
Exercise stress testing
The stress test was performed on the treadmill following the standard modified Bruce protocol (16) in order to accommodate elderly and sedentary individuals. HR was continuously measured using a standard 12-lead ECG (Marquette Medical Systems, Milwaukee, WI). BP was measured every other minute with a manual sphygmomanometer (Welch Allyn Tycos-767 series, Skaneateles Falls, NY). All readings, for BP and HR, were taken by experienced technicians.
Brachial artery reactivity
All of the patients included in this analysis underwent FHR testing to assess brachial artery reactivity, which was used as a proxy of endothelial function in the present study. The FHR test protocol has been described previously (17). Participants were seated in front of a large field-of-view gamma camera and an occlusion cuff was inflated proximal to the right elbow to 50 mm Hg higher than their resting systolic BP (SBP) for 5 min. Once the cuff was released, 30 s were allowed to elapse before injecting the patient with technetium-99m-tetrofosmin (Myoview), through a venous catheter placed in the median antebrachial vein of the left arm at the level of the cubital fossa. Upon scanning the activity time curves of the tracer, the peak slopes of the right (hyperaemic) arm were divided by the left (control), thus calculating the rate of uptake ratio (RUR). This measure of brachial artery reactivity has been shown to predict the presence of CVD (18), has a high test-retest reliability (r = 0.89) (19), excellent inter- and intra-rater reliability (r = 0.98) (20), and is consistent with similar nuclear medicine based techniques (21). Participants were classified as having poor function if they had an RUR <3.55; this cutoff value has been previously reported to be highly correlated to CVD (17) with a lower RUR score is indicative of ED.
Prior to the FHR test a trained technician measured the participants' waist circumference and drew a blood sample (from the indwelling catheter inserted for the FHR test). Blood samples were analyzed by the Haematology Department at the Montreal Heart Institute, using standard protocols. Participants were categorized as having MS if they met any three of the following criteria: BP ≥130 mm Hg SBP or ≥85 mm Hg diastolic BP or antihypertensive drug treatment in patients with a history of hypertension, waist circumference ≥102 cm in men and ≥88 cm in women, fasting glucose ≥5.5 mmol/l (100 mg/dl) or drug treatment for elevated glucose, triglycerides ≥1.7 mmol/l (150 mg/dl) or drug treatment for elevated triglycerides, and HDL <1.03 mmol/l (40 mg/dl) for men, and <1.3 mmol/l (50 mg/dl) for women. All values and measurements followed American Heart Association (AHA) Guidelines for the diagnosis of MS (22).
Rate pressure product (RPP) was calculated as the product of SBP and HR/100, for any given measurement. Reactivity was measured as the difference between peak and rest measurements taken before and during testing for HR, SBP, diastolic BP (DBP), and RPP. Separate General Linear Models (GLM, using SAS's proc glm function) were performed to assess the main and interaction effects of RUR and MS on HR reactivity (ΔHR), SBP reactivity (ΔSBP), DBP reactivity (ΔDBP), and RPP reactivity (ΔRPP), adjusting for age, sex, resting cardiovascular measure (e.g., for ΔHR, resting HR was included as a covariate), total metabolic equivalents (METs) achieved during the exercise stress test, and cardiac history (previous myocardial infarction, coronary artery bypass graft surgery, percutaneous coronary intervention) measured by self-report. These covariates were a priori defined due to their previous associations with the physiological processes under examination. As part of the review process we were requested to add smoking status as an additional covariate for which a separate series of analyses were conducted. Demographic and sample data is reported as mean ± SD. A value of P < 0.05 was considered statistically significant.
Please refer to Table 1. The mean (SD) age of the participants was 60 (10) years and the majority were males (75%). As detailed in Table 1, 34% (n = 103) of the population were free of either poor RUR or MS, 23% (n = 71) had both poor RUR and MS, 21% (n = 62) had poor RUR only, and 22% (n = 67) had MS only. Overall, the participants had relatively average body mass index measurements (mean (SD): 27.8 ± 4.5 kg/m2), waist circumference (mean (SD): 99.5 ± 12.1 cm), triglycerides (mean (SD): 1.5 ± 0.9 mmol/l), and HDL (mean (SD): 1.3 ± 0.4 mmol/l) levels. No between-group differences were noted. However, on average this sample had abnormally high (according to AHA guidelines (22)) fasting blood glucose levels of 5.7 ± 1.3 mmol/l indicating that a great proportion of the population studied likely had impaired glucose tolerance or insulin resistance. No differences were noted between groups for baseline, resting HR, SBP, DBP, and RPP measures (see Table 1).
Exercise-induced cardiovascular changes
As seen in Figure 1a, there were main effects of MS (F = 5.51, η2 = 0.02, P = 0.02) and poor RUR (F = 6.69, η2 = 0.02, P = 0.01) on ΔHR. Patients with MS or low RUR in this study had decreased ΔHR in comparison to those without MS or ED. There was no interaction effect between low RUR and MS (F = 0.60, P = 0.44) on ΔHR. As detailed in Figure 1b and c, there were no main or interaction effects of MS or poor RUR on ΔSBP or ΔDBP. There also appears to be a trend of decreased ΔRPP among the participants with both low RUR and MS, however this result was not statistically significant (see Figure 1d). When smoking was added to these analyses there was no substantive change in the results found.
Although no differences in baseline HR were observed between groups, the study participants with poor RUR and/or MS were found to have independent effects of decreasing ΔHR to exercise compared to those without either condition. It was also observed that MS and poor RUR did not have any multiplicative effects on the measured cardiovascular parameters during exercise. According to the collected data, the presence of poor RUR and/or MS did not influence the ΔSBP, ΔDBP, or ΔRPP. Given that both MS (23) and ED (4) have been linked to ANS dysregulation, this finding is likely reflective of autonomic dysfunction, and perhaps specific to the parasympathetic branch.
The individual components of MS have been linked to altered autonomic activity measured both systemically and regionally. A review by Tentolouris and colleagues (23) clearly outlines these associations. Acute infusion of insulin, for instance, can reduce the parasympathetic impact on cardiac function and concomitantly stimulate sympathetic nervous system (SNS) activity, as measured by HR variability, in healthy women (24). Comparison of autonomic activity in type II diabetics with and without MS displayed greater cardiac sympathetic predominance in those with MS (25). Additionally, it has been suggested that increased SNS activity may result from insulin resistance (26). Similarly, sympathetic overactivity has been observed in obese women (23,27). Autonomic dysfunction also varies according to fat distribution, whereby those with higher visceral fat had lower baroreceptor sensitivity (28) and higher basal muscle sympathetic nerve activity compared to matched controls (29). Autonomic dysregulation has also been widely documented in those with hypertension (23,26). Overactivity of the SNS is involved in the pathogenesis of hypertension (30) and has been shown to be related to hypertension severity (31). Increased HR, greater noradrenaline spillover, baroreflex dysfunction, and elevated sympathetic activity measured by microneurography have been observed in individuals with high BP (26). Of note, baroreflex impairment and SNS overactivity are amplified when patients present with both hypertension and obesity (32). Lastly, autonomic impairment has also been associated with dyslipidemia. Acute infusion of nonesterified fatty acids and triglycerides decreased baroreceptor sensitivity in obese hypertensive individuals to a greater extent than in the healthy control group (33). These previous findings illustrate a clear relationship between autonomic dysfunction and the MS.
Similarly, ANS impairments have been observed with ED. Under normal circumstances vascular endothelial cells and the ANS operate in an antagonistic manner to maintain appropriate vessel tone; the endothelium works to produce a vasodilatory effect and the SNS is chiefly responsible for vasoconstriction. A review by Harris and Matthews (4) outlines both direct and indirect associations between ED and ANS dysregulation. Which impairment comes first in the development of CVD remains unknown; however there is evidence to suggest this relationship may be mediated by sex hormones, oxidative stress, platelet activation, the renin–angiotensin system, the hypothalamic–pituitary–adrenal axis, insulin resistance, and aging (4).
The dampened exercise HR response in participants with poor RUR and/or MS observed here may be reflective of autonomic dysregulation in these individuals because the control of HR during exercise is achieved by a fine balance between the deactivation of the parasympathetic nervous system (PNS) and concomitant stimulation of the SNS, in addition to various hormonal influences (34,35). Typically HR increases during exercise; the extent to which depends on the dose of physical activity (i.e., the intensity and duration of exercise) as well as the individual's cardiovascular fitness (36,37). The rapid augmentation of HR at the very onset of aerobic exercise results from a decrease in cardiovagal modulation of HR which continuously decreases to the point that the signal is extremely low, even undetectable, as exercise workload is increased (37). Simultaneously, there is an increase in sympathetic activation (37). Thus because PNS inhibition has been found to be largely responsible for the exercise-induced HR increases, it could be that patients with ED and/or MS have some level of dysregulation of the parasympathetic system. It is also important to acknowledge the contributing role of the SNS in regulating HR and be aware that there may also be malfunction in this division of the ANS. However, the lack of statistical support for differences in ΔSBP, ΔDBP, and ΔRPP between participants with poor RUR and/or MS and those without one or the other condition may indicate adequate functioning of the sympathetic system. It should be noted that the data presented here cannot directly confirm this statement and further investigation would be necessary to establish the relative contributions of PNS and SNS dysfunction in these populations. Whilst further research is necessary, the findings of this study suggest that there is partial dysfunction of the PNS. It still remains unknown whether it is the disease state (i.e., having ED or MS) which brings about autonomic dysregulation or vice versa.
The results of this study should be considered in light of several limitations. The participants in this study were all referred to the Nuclear Medicine Department at the Montreal Heart Institute for cardiovascular testing, meaning that they all had CVD or were likely at risk for CVD in some capacity. As the cardiovascular reactivity levels measured in this study are all relative to those without the conditions, the results may be underestimated due to the fact that they were not measured against a healthy control group. However, history of CVD was adjusted for in the statistical analysis and therefore any effect of this should be reduced. Also, the guidelines for the diagnosis of MS vary between organizations, thus if different classification guidelines, such as those of the World Health Organization, were employed within the same population, then it is likely that some participants would have been categorized differently consequently altering the results. However, because the AHA guidelines are those employed most frequently for classification of MS and the various series of diagnostic parameters differ marginally from one another, the use of the AHA guidelines is justified. Additionally, analysis of HR variability or baroreflex sensitivity could provide insight into the question of sympathetic vs. parasympathetic dysregulation. However, this analysis was not possible with the data collected and should be considered in future studies. In spite of these limitations, there are several strengths of the present study that should be noted, specifically that it sampled a large number of patients, and the testing was performed in a controlled environment. Also the measurement technique used to determine the presence of brachial artery reactivity is a reliable and validated, though not widely used, method (17). Important confounding factors such as age, sex, cardiovascular history, and total METs were adjusted for in the statistical analysis, which further strengthens the significance of the results.
Studies assessing PNS function by measuring HR variability and HR recovery have shown links with CVD outcomes/endpoints, e.g., MI, sudden cardiac death (38,39) and other comorbidities, e.g., depression (40) and thus it is plausible that PNS dysregulation is observed in pre-CVD conditions, e.g., MS and ED, and is manifested through changes in exercise HR. Though there is still debate regarding the interpretations of HR variability and HR recovery, understanding how PNS function may be altered by CVD precursors can help contribute to the prognostic value of this parameter. Future studies should consider using HR variability, HR recovery, HR spectral-analysis and/or microneurography to address questions related to the function of the ANS components, specifically PNS dysregulation in this patient population.
Given the information linking ED and MS with ANS dysregulation, it is reasonable to hypothesize that patients affected with both conditions would have worse ANS functioning. However, the data presented here does not encourage the notion of the existence of any multiplicative effects of MS and ED as would have been expected based on the frequently observed pairing of MS risk factors and various indicators of ED. Overall, the findings of this study support further research in the area of ANS function, and specifically the PNS, in regards to how malfunctioning contributes towards the origins of CVD. Additionally, it would be of significant value to explore whether or not a relationship exists between the level of ED and the extent of ANS dysregulation. A better understanding of the aetiology of CVD, with specific attention to ANS dysregulation, would potentially lead to the formulation of new therapeutic intervention strategies, including both pharmacological aids and lifestyle regimes, intended to effectively treat CVD.
The results reported here indicate that people with MS and/or ED have some degree of ANS dysregulation. Given that only a significant decrease was observed for ΔHR it is likely that the dysregulation specifically involves PNS activity. The lack of significance for other cardiovascular parameters indicates that patients with poor RUR and/or MS may have relatively normal SNS regulation. The exact mechanisms by which ANS dysregulation occurs is still largely unknown; however, building a better understanding of the characteristics of pre-CVD conditions, such as ED and MS, research is closer to uncovering the pathogenesis of CVD, which could lead to improved treatment and early diagnosis strategies.
The authors would like to acknowledge operating funding support from the Canadian Institutes of Health Research (CIHR: MOP) and the Heart and Stroke Foundation of Canada, and new investigator funding from Fonds de la Recherche en Santé du Québec (FRSQ: for S.L.B. and K.L.L.), CIHR (for S.L.B.), and the Vanier Canada Graduate Scholarship (CIHR: for A.M.R. and J.L.G.).
- 1Heart and Stroke Foundation. Statistics. (Cited 7 September 2011). <http://www.heartandstroke.com/site/c.ikIQLcMWJtE/b.3483991/k.34A8/Statistics.htm>.
- 2Statistics Canada. Mortality, Summary List of Causes 2005. Ottawa, Division HS; 2009 Contract No.: 84F0209X.
- 3The metabolic syndrome, insulin resistance and cardiovascular disease. Cardiovasc J S Afr 2002; 13: 187–188..
- 18RUR and EWRU, new markers of endothelial function. Psychosom Med 2005; 67: A38., , , .
- 19Test-retest reliability of a new method to measure endothelial function. Psychosom Med 2005; 67: A54., , , .
- 20Inter-observer reliability of a new method to measure endothelial function. Psychosom Med 2006; 68: A36., , , , , .