Journal of Science and Medicine in Sport

Transcription

1 Journal of Science and Medicine in Sport 16 (2013) 8 12 Contents lists available at SciVerse ScienceDirect Journal of Science and Medicine in Sport j o ur nal homep age: jsams Original research Contribution of autonomic dysfunction to abnormal exercise blood pressure in type 2 diabetes mellitus Kassia S. Weston a,b, Julian W. Sacre a,b, Christine L. Jellis b, Jeff S. Coombes a, a School of Human Movement Studies, The University of Queensland, Australia b School of Medicine, The University of Queensland, Princess Alexandra Hospital, Australia a r t i c l e i n f o Article history: Received 4 January 2012 Received in revised form 1 April 2012 Accepted 13 April 2012 Keywords: Peak exercise Exaggerated blood pressure Heart rate variability Diabetic neuropathies a b s t r a c t Objectives: The purpose of this study was to compare the presence and severity of autonomic dysfunction in type 2 diabetes mellitus patients, with and without exaggerated blood pressure responses to exercise. Design: We performed a cross-sectional analysis of 98 patients with type 2 diabetes mellitus (aged 59 ± 9). Methods: Both time (standard deviation of RR intervals, root-mean-square of successive RR interval differences) and frequency (total spectral power, high frequency, low frequency, very low frequency) domains of heart rate variability were analysed in a 5 min recording at rest and 20 min after a maximal treadmill test. An exaggerated blood pressure response to exercise was identified by peak blood pressure 190/105 mm Hg (women) or 210/105 mm Hg (men). Results: Each group of either exaggerated exercise blood pressure response or normal blood pressure response consisted of 49 patients. At rest there were no significant differences between groups for all time and frequency domain parameters of heart rate variability. Post-exercise, there was a significant (p < 0.05) reduction in the SDNN, RMSSD and TP in the exaggerated exercise blood pressure group. Independent correlates (p < 0.01) of exercise systolic blood pressure included post-exercise TP, resting systolic blood pressure, cardiac autonomic neuropathy and beta-blockers (beta = 0.28, adj. R 2 = 0.32, p < 0.001). Conclusions: Reduced post-exercise heart rate variability in patients with type 2 diabetes mellitus, with an exaggerated exercise blood pressure response suggests preclinical autonomic dysfunction characterized by impaired vagal modulation Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved. 1. Introduction An exaggerated blood pressure (BP) response to exercise predicts cardiovascular and all-cause mortality 1,2 and is common in patients with uncomplicated type 2 diabetes mellitus (T2DM). 3 In this context, abnormal exercise BP has been associated with vascular dysfunction 4 and insulin resistance. 5 Despite its wellrecognized influence on exercise hemodynamics, the potential contribution of autonomic nervous system dysfunction to exaggerated exercise BP remains equivocal. Reduced heart rate variability (HRV) during and after exercise in individuals without diabetes with exaggerated exercise BP may reflect impaired vagal activity. 6 However, the applicability of these findings to patients with T2DM is unclear, particularly since exercise BP is blunted in the setting of diabetic cardiac autonomic neuropathy (CAN). 7 Despite being strongly associated Corresponding author. address: jcoombes@uq.edu.au (J.S. Coombes). with adverse outcome and affecting up to one third of patients with T2DM, the pathophysiology and progression of CAN remains poorly understood. 8 Attenuated BP responses to exercise with fully evolved CAN may reflect sympathetic dysinnervation However, earlier diabetic disease may be characterized by specific vagal involvement with attendant relative sympathetic overactivity. 11 Whether these abnormalities predispose to exaggerated exercise BP typical of patients without CAN has not been studied. Since the recovery of heart rate and BP post-exercise is influenced by the function of the autonomic nervous system, 12 the association of exaggerated exercise BP and autonomic dysfunction was sought post-exercise. The purpose of the present study was to compare autonomic function using HRV measurements in T2DM patients with and without exaggerated exercise BP. Because preclinical autonomic dysfunction may be difficult to detect at rest, 12 we also evaluated post-exercise HRV during the recovery period following a maximal treadmill exercise stress test. It was hypothesized that T2DM patients with exaggerated exercise BP would have reduced postexercise HRV that was not detected at rest /$ see front matter 2012 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

2 K.S. Weston et al. / Journal of Science and Medicine in Sport 16 (2013) Methods Participants with T2DM (n = 107), aged 40 years and with no known cardiovascular disease, psychiatric or other severe illness, were recruited from the community. Patients with an abnormal electrocardiogram precluding HRV measurement or inability to attain 85% of age-predicted maximal heart rate during exercise testing were excluded. All patients provided written informed consent, and the study was approved by the University of Queensland and Princess Alexandra Hospital Ethics Committees, Brisbane Australia. Cardiac autonomic neuropathy (CAN) was defined by a score of 2 abnormal results from a battery of 7 autonomic function tests, including HR responses to deep breathing, Valsalva manoeuvre and standing, the systolic BP response to standing and three HRV frequency domain parameters. 8 Left ventricular mass index (LVMI) was measured by 2D targeted M-mode echocardiography using a Vivid 7 ultrasound machine (GE Vingmed, Horten, Norway). The formula used to calculate left ventricular mass was the Devereux formula and indexed to body surface area (g/m 2 ). 13 Biochemical analyses of fasting blood samples for lipids, glucose, and HbA 1c were measured according to standard hospital pathology laboratory protocols. Participants abstained from taking their medication for 24 h prior to testing, in order to limit any effect on exercise hemodynamics or HRV. Diabetic complications, besides the formally tested CAN, were documented and based on selfreport (i.e. mild retinopathy or peripheral neuropathy). Baseline measures of BP and HRV were performed after at least 15 min of supine rest. Participants then performed a maximal exercise stress test using a treadmill according to the Bruce protocol, as a multistage test using the standard progressive workloads. Simultaneous measurement of peak oxygen uptake (VO 2 peak) by expired gas analysis was also obtained (SensorMedics, Vmas Spectra 29c, California, USA). Upon cessation of exercise, patients resumed a supine position for 25 min during recovery. Due to the progressive reduction in heart rate during recovery from exercise and the resultant non-stationarity in the heart rate signal, 14 post-exercise HRV measurements were not commenced until 20 min following test termination, in accordance with previous studies. 12 Immediate heart rate recovery (HRR) in the first 3 min after exercise was also recorded. BP at rest and during exercise was recorded by manual mercury sphygmomanometry over the right brachial artery. An exaggerated BP response to exercise was identified by peak recorded BP of 190/105 mm Hg (women) or 210/105 mm Hg (men). 15 HRV at rest and post-exercise was measured from a continuous 5-min 3-lead ECG recording, acquired using a Powerlab 8SP data acquisition system linked to a commercially available software package with automatic R-R interval detection capabilities (Labchart v6.1, AD Instruments, Sydney, Australia). 16 We have previously shown favourable relative reliability of HRV parameters in patients with T2DM. 17 The R-R interval tachogram was verified by visual inspection and ectopic beats or artifacts were linearly interpolated. The following time domain parameters of HRV were assessed: (1) standard deviation of RR intervals (SDNN), a measurement of global HRV and (2) root-mean-square of differences in successive RR intervals (RMSSD), an estimate of parasympathetic modulation. 16 HRV parameters in the frequency domain included spectral power in very low frequency (<0.04 Hz; VLF), low frequency ( Hz; LF), and high frequency ( Hz; HF) bands and the total spectral power (TP). Frequency domain parameters were assessed by fast Fourier transform. HF power is thought to represent parasympathetic modulation, whereas the LF component reflects both parasympathetic and sympathetic activity, thereby facilitating an estimation of sympatho-vagal balance by the LF/HF ratio. A recent consensus statement by the Toronto Consensus Panel on Diabetic Neuropathy proposed that most HRV indices reflect predominantly parasympathetic modulation. 18,19 Since TP represents global HRV, 16 this marker was employed in regression analyses. Data were analysed using SPSS (version 16.0, Chicago, Ill) and presented as mean ± standard deviation (SD). Normality of distribution was assessed by the Kolmogorov Smirnov test and skewed variables were transformed using the natural logarithm. Variables with a skewed distribution were expressed by median (with interquartile range). Differences between groups were assessed by an independent Student s t-test and the chi-square test for continuous and categorical variables, respectively. Linear associations between variables were evaluated by a Pearson or Spearman rank correlation coefficient, where appropriate. Multiple linear regression was employed to determine independence of associations (of correlates with a significance of p < 0.1) and performed using the enter method. Statistical significance was defined by p < Results Clinical characteristics of the study population are displayed in Table 1. Nine participants with an abnormal electrocardiogram precluding HRV analysis at rest or post-exercise were excluded. Of the remaining 98 participants, an exaggerated exercise BP was identified in 49 (50%) participants. There were no significant differences between groups in clinical characteristics, including age, duration of diabetes, VO 2 peak, BMI, diabetes control or microvascular complications (i.e. retinopathy, peripheral neuropathy). However, there were significantly less females in the normal exercise BP group and their HDL cholesterol was lower. Patients with exaggerated exercise BP demonstrated higher prevalence of antihypertensive therapy, particularly the use of angiotensin II receptor blockers. Moreover, resting systolic and diastolic BP values in the exaggerated exercise BP group were higher than the normal BP response group. Rest and peak heart rates and HRR were not significantly different between the groups. CAN was identified in a small proportion of patients with a normal exercise BP but was absent in the exaggerated exercise BP group (p = 0.08). The prevalence of abnormal CAN tests in each group is displayed in Table 2. An abnormal HR response to deep breathing was significantly more common in patients with normal exercise BP, but no other group differences were observed. The change in peak systolic BP from rest to post exercise was significantly higher in the exaggerated exercise BP group (p < 0.001), indicating that exaggerated exercise BP in these patients was not solely attributable to higher resting BP. Exercise diastolic BP was also significantly higher in the exaggerated exercise BP group. There were no differences in LVMI, VO 2 peak or ejection fraction between groups. HRV parameters are displayed in Table 3. At rest there were no significant differences between groups in any time or frequency domain HRV parameters. However, post-exercise SDNN, RMSSD and TP were all significantly lower in patients with an exaggerated exercise BP. Post-exercise HF (p = 0.066) and VLF (p = 0.069) were trending towards being significantly lower in the exaggerated exercise BP group. There were no significant univariate associations between peak exercise systolic BP and post-exercise HRV measures in all patients. However, due to the possible confounding influence of numerous covariates, we examined the independence of these associations using multiple linear regression. In a model including factors demonstrating at least a borderline significant correlation (p < 0.1) with peak exercise systolic BP (triglycerides, resting systolic BP, left ventricular mass index, post-exercise TP and ACE inhibition)

3 10 K.S. Weston et al. / Journal of Science and Medicine in Sport 16 (2013) 8 12 Table 1 Clinical characteristics of T2DM patients with normal and exaggerated BP responses to exercise. Participants Normal exercise Exaggerated exercise SBP (n = 49) Age (years) 58.2 ± ± Females (n, %) 13, , BMI (kg/m 2 ) 30.2 ± ± Diabetes duration (yrs) 6.5 (7) 6 (7) 0.67 (median) [IQR] CAN (n, %) 3, , Retinopathy (n, %) 4, 8.2 6, Peripheral neuropathy (n, %) 9, , Blood biochemistry Fasting glucose (mmol/l) 8.8 ± ± HbA 1c (%) 7.7 ± ± Total cholesterol (mmol/l) 4.6 ± ± HDL cholesterol (mmol/l) 1.0 ± ± LDL cholesterol (mmol/l) 2.8 ± ± Triglycerides (mmol/l) 1.6 ± ± Hemodynamic values Peak VO 2 (ml/kg/min) 29.5 ± ± Resting SBP (mm Hg) 127 ± ± 19.7 <0.001 Resting DBP (mm Hg) 76 ± ± Exercise SBP (mm Hg) 183 ± ± 16.9 <0.001 Exercise DBP (mm Hg) 83 ± ± SBP (mm Hg) 62 ± ± 19.5 <0.001 Recovery (mm Hg) 55 ± ± 19.6 <0.001 Resting HR (bpm) 69 ± ± HRR 30 s (bpm) 13 ± ± HRR 1 min (bpm) 33 ± ± HRR 1:30 min (bpm) 43 ± ± HRR 2 min (bpm) 51 ± ± HRR 3 min (bpm) 59 ± ± Peak HR (bpm) 157 ± ± LVMI (g/m 2 ) 77.4 ± ± Ejection fraction (%) 65.2 ± ± Medications ACE inhibitor use (%) ARB use (%) Beta blocker use (%) Any antihypertensive medication (%) Metformin use (%) Statin use (%) Insulin use (%) DBP: diastolic BP; SBP: change from rest to peak; recovery: change from peak to 20 min post exercise; LVMI: left ventricular mass index; ACE: angiotensin converting enzyme; ARB: angiotensin II receptor blocker; HRR: heart rate recovery. Table 2 CAN test battery parameters in T2DM patients with normal and exaggerated BP responses to exercise. CAN test Normal exercise Exaggerated exercise SBP (n = 49) Cardiac reflex tests Valsalva HR ratio (n, %) 1, 2.0 0, Orthostatic max/min HR 3, 6.1 1, ratio (n, %) Orthostatic hypotension 0, 0 0, 0 (n, %) Deep breathing E/I HR 5, , ratio (n, %) Resting HR variability VLF (n, %) 4, 8.2 1, LF (n, %) 4, 8.2 3, HF (n, %) 3, 6.1 7, in addition to variables known to affect exercise haemodynamics (age, sex, BMI, pulse wave velocity, beta-blockade), independent correlates of exercise systolic BP included post-exercise TP, resting systolic BP, beta blockade and CAN (Table 4). 4. Discussion The exaggerated exercise BP of uncomplicated T2DM has prognostic significance, 3,20,21 although the etiology remains poorly understood. 5,20,21 The key findings of this study were that patients with T2DM and exaggerated exercise BP have reduced HRV during post-exercise recovery consistent with impaired cardiac parasympathetic modulation and notwithstanding normal autonomic function at rest. Moreover, independent associations were identified between post-exercise HRV and peak exercise systolic BP. It has been suggested that exaggerated exercise BP is significantly and strongly correlated with resting BP. 22 For this reason, it is not surprising that the exaggerated exercise BP group were more likely to have a higher resting BP and on antihypertensive therapy compared with the normal exercise BP group. However, the association of post-exercise HRV with exercise BP was independent of resting BP and antihypertensive medication use on multiple linear regression. Our findings showed evidence of cardiovascular autonomic dysfunction after exercise in patients with exaggerated exercise BP, despite apparently normal function at rest. Autonomic adaptations to an acute bout of exercise include a reduction in parasympathetic tone and concomitant sympathetic activation, thereby enabling sufficient chronotropic and inotropic responses. 23 In turn, postexercise recovery is characterized by parasympathetic reactivation independent of sympathetic withdrawal. 24 Lower post-exercise HRV in patients with exaggerated exercise BP is synonymous with diminished vagal function, since HRV is more closely linked to parasympathetic rather than sympathetic activity. 25,26 However it is difficult to determine whether these findings may be explained by either, (1) a greater withdrawal of parasympathetic activity during exercise or (2) impaired reactivation after exercise. Indeed, it would be expected that reduced vagal function would also be reflected in reduced HRR. The lack of group differences in HRR, which is known to be determined by vagal reactivation, 25 supports the former hypothesis. In the context of evidence that parasympathetic function may be affected earlier and to a greater extent in T2DM then sympathetic function, 11,27 the current results indicate that parasympathetic neuropathy may represent a mechanistic origin of the exaggerated exercise BP commonly observed in these patients. Indeed, we may have identified an early form of autonomic dysfunction undetectable at rest but unmasked by exercise. This has previously been described in a female T2DM cohort, where reduced post-exercise HRV was also reported despite normal resting values. 12 Baroreflex sensitivity is strongly associated with vagal function and buffers fluctuations in BP. 28 Therefore, the vagal withdrawal that occurs in autonomic dysfunction may result in impaired baroreflex sensitivity 7,29 and resultant vascular stiffening through endothelial dysfunction. Stewart et al. 4 suggested that impaired endothelium-dependent vasodilation contributed to exercise hypertension. Relevant to this study, endothelial dysfunction has been further demonstrated in patients with type 2 diabetes mellitus. 30 Without the homeostatic control in BP, insensitive baroreceptors may lead to exaggeration of the BP response to exercise. 31 Diabetic patients with fully-evolved CAN have reduced exercise tolerance due to an impairment of the autonomic

4 K.S. Weston et al. / Journal of Science and Medicine in Sport 16 (2013) Table 3 Rest and post-exercise heart rate variability parameters in T2DM patients with normal and exaggerated BP responses to exercise. Variable Resting HRV Post Exercise HRV Normal exercise SBP (n = 49 Exaggerated exercise Normal exercise Exaggerated exercise Time-domain parameters SDNN (ms) 32.9 (28.9) 25.4 (17.1) 16.8 (11.6) 14.3 (6.4) Log SDNN [log(ms)] 3.4 ± ± ± ± RMSSD (ms) 19.6 (15.2) 16.6 (13.9) 7.3 (6.6) 5.1 (3.2) Log RMSSD [log(ms)] 2.9 ± ± ± ± Frequency-domain parameters HF (ms 2 ) (203.3) (173.3) 14.9 (34.3) 7.8 (11.7) Log HF [log(ms 2 )] 4.7 ± ± ± ± LF (ms 2 ) (472.2) (236.8) 47.0 (136.1) 34.6 (40.4) Log LF [log(ms 2 )] 5.6 ± ± ± ± VLF (ms 2 ) (979.3) (539.9) (169.9) 101 (100.5) Log VLF [log(ms 2 )] 6.1 ± ± ± ± LF/HF ratio (ms 2 ) 2.4 (2.7) 2.0 (3.2) 3.8 (4.7) 4.3 (5.8) Log LF/HF ratio 0.85 ± ± ± ± TP (ms 2 ) 1092 (1848.8) (886.6) 235 (373.8) (140.6) Log TP [log(ms 2 )] 6.8 ± ± ± ± Median and interquartile range (IQR) in normal values is given above log transformed values. SDNN: standard deviation of R-R intervals; RMSSD: square root of mean of sum of the square of differences between R-R intervals; HF: high frequency power Hz; LF: low frequency power Hz; VLF: very low frequency power <0.04 Hz; TP: total spectral power. nervous system to increase BP and heart rate with exercise Indeed, we found an inverse relationship between CAN and peak exercise systolic BP, with a higher incidence of CAN in the normal, rather than the exaggerated, exercise BP group. This may reflect a pseudo-normalisation of exercise BP with the concomitant parasympathetic and sympathetic damage of overt CAN. In contrast, reduced post-exercise HRV in patients with exaggerated exercise BP may be consistent with predominant parasympathetic involvement in early autonomic dysfunction. Exaggerated exercise BP in T2DM due to parasympathetic dysfunction has many implications for cardiovascular risk. Reduced vagal function may be associated with less cardioprotection, 32 such that T2DM patients with exaggerated exercise BP may be at an increased risk of sudden cardiac death due to tachyarrhythmias. 32,33 Current HRV techniques are unable to separate parasympathetic and sympathetic innervations. 19 Since HRV parameters primarily reflect parasympathetic function, the potential contribution of relative sympathetic overactivity must be inferred. Moreover, HRV measurements were unable to be taken during or immediately after exercise in this study due to non-linear changes in HR violating the assumption of stationarity. 34 In line with previous studies, HRV measurements were therefore taken at 20 min post-exercise. 12 Including measurements of respiration with HRV may have been helpful in supporting the current findings. Future studies employing exercise protocols with submaximal and steady state workloads amenable to HRV measurement may be beneficial. Table 4 Factors associated with peak exercise SBP. Variable Univariate (r) Multivariate (ˇ) Age Male Body mass index T2DM duration Pulse wave velocity Glucose HbA1c Total cholesterol HDL LDL Triglycerides Rest SBP 0.43 < <0.001 Rest HR Peak HR Left ventricular mass index Ejection fraction VO 2 peak Cardiac autonomic neuropathy Post-exercise total power Angiotensin-converting enzyme Angiotensin receptor blocker Beta blockade Metformin Insulin use Statin Overall R 2 = 32%, p <

5 12 K.S. Weston et al. / Journal of Science and Medicine in Sport 16 (2013) Conclusion We have shown that patients with T2DM with an exaggerated exercise BP have reduced post-exercise HRV that is not detected at rest. These HRV changes suggest preclinical autonomic dysfunction characterized by decreased vagal activity and a relative sympathetic predominance. These findings are of clinical significance, as an exaggerated exercise BP is known to increase the risk of cardiovascular and all cause mortality. The results from this study contribute to the body of research exploring the mechanisms of exaggerated exercise BP in T2DM. 6. Practical implications Early autonomic dysfunction undetected at rest may be elicited with maximal exercise and identified in individuals with an abnormal exercise blood pressure response. Post-exercise recovery may be an under-utilised time period for autonomic testing. Exaggerated exercise BP by EST could be used as part of routine testing for type 2 diabetics as an indication for further autonomic function testing. Type 2 diabetics have a known predisposition to exaggerated exercise blood pressure, which increases the risk of mortality, however the mechanisms are unclear. This study identifies possible mechanisms of the exaggerated blood pressure response to exercise, thereby guiding treatment. Acknowledgements The authors wish to thank the participants for volunteering to be a part of this study. This study was supported in part by Centres for Clinical Research Excellence award (455832) from the National Health and Medical Research Council, Canberra, Australia. References 1. Williams SG, Jackson M, Ng LL, et al. Exercise duration and peak systolic blood pressure are predictive of mortality in ambulatory patients with mild moderate chronic heart failure. Cardiology 2005;104(4): Lund-Johansen P. Blood pressure response during exercise as a prognostic factor. J Hypertens 2002;20(8): Scott JA, Coombes JS, Prins JB, et al. Patients with type 2 diabetes have exaggerated brachial and central exercise blood pressure: relation to left ventricular relative wall thickness. Am J Hypertens 2008;21(6): Stewart KJ, Sung J, Silber HA, et al. Exaggerated exercise blood pressure is related to impaired endothelial vasodilator function. Am J Hypertens 2004;17(4): Brett SE, Ritter JM, Chowienczyk PJ. Diastolic blood pressure changes during exercise positively correlate with serum cholesterol and insulin resistance. Circulation 2000;101(6): Eryonucu B, Bilge M, Guler N, et al. The effect of autonomic nervous system activity on exaggerated blood pressure response to exercise: evaluation by heart rate variability. Acta Cardiol 2000;55(3): Hilsted J, Galbo H, Christensen NJ, et al. Haemodynamic changes during graded exercise in patients with diabetic autonomic neuropathy. Diabetologia 1982;22(5): Vinik AI, Ziegler D. Diabetic cardiovascular autonomic neuropathy. Circulation 2007;115(3): Vinik AI, Freeman R, Erbas T. Diabetic autonomic neuropathy. Semin Neurol 2003;23(4): Istenes I, Keresztes K, Hermányi Z, et al. Relationship between autonomic neuropathy and hypertension are we underestimating the problem? Diabet Med 2008;25(7): Freccero C, Svensson H, Bornmyr S, et al. Sympathetic and parasympathetic neuropathy are frequent in both type 1 and type 2 diabetic patients. Diabetes Care 2004;27(12): Figueroa A, Baynard T, Fernhall B, et al. Impaired postexercise cardiovascular autonomic modulation in middle-aged women with type 2 diabetes. Eur J Prev Cardiol 2007;14(2): Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in nonanatomic validation of the method. Circulation 1977;55: Terziotti P, Schena F, Gulli G, et al. Post-exercise recovery of autonomic cardiovascular control: a study by spectrum and cross-spectrum analysis in humans. Eur J Appl Physiol 2001;84(3): Oldershaw P. Stress testing: principles and practice: edition III. Int J Cardiol 1987;14(1): Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Eur Heart J 1996;17(3): Sacre JW, Jellis CL, Marwick TH, et al. Reliability of heart rate variability in patients with type 2 diabetes mellitus. Diabet Med Epub ahead of print. 18. Houle MS, Billman GE. Low-frequency component of the heart rate variability spectrum: a poor marker of sympathetic activity. Am J Physiol Heart Circ Physiol 1999;276(1):H215 H Bernardi L, Spallone V, Stevens M, et al. Methods of investigation for cardiac autonomic dysfunction in human research studies. Diabetes Metab Res Rev 2011;27(7): Kumagai S, Kai Y, Hanada H, et al. Relationships of the systolic blood pressure response during exercise with insulin resistance, obesity, and endurance fitness in men with type 2 diabetes mellitus. Metabolism 2002;51(10): Hypertension in Diabetes Study (HDS): I. Prevalence of hypertension in newly presenting type 2 diabetic patients and the association with risk factors for cardiovascular and diabetic complications. J Hypertens 1993;11(3): Bassett DR, Duey WJ, Walker AJ, et al. Exaggerated blood pressure response to exercise: importance of resting blood pressure. Clin Physiol 1998;18(5): Freeman JV, Dewey FE, Hadley DM, et al. Autonomic nervous system interaction with the cardiovascular system during exercise. Prog Cardiovasc Dis 2006;48(5): Imai K, Sato H, Hori M, et al. Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure. J Am Coll Cardiol 1994;24(6): Buchheit M, Laursen P, Al Haddad H, et al. Exercise-induced plasma volume expansion and post-exercise parasympathetic reactivation. Eur J Appl Physiol 2009;105(3): Buchheit M, Laursen PB, Ahmaidi S. Parasympathetic reactivation after repeated sprint exercise. Am J Physiol Heart Circ Physiol 2007;293(1):H133 H Subbalakshmi NK, Adhikari PMR, Rajeev A, et al. Independent predictors of cardiac parasympathetic dysfunction in type 2 diabetes mellitus. Singapore Med J 2008;49(2): Joyner MJ. Baroreceptor function during exercise: resetting the record. Exp Physiol 2006;91(1): Kougias P, Weakley SM, Yao QZ, et al. Arterial baroreceptors in the management of systemic hypertension. Med Sci Monit 2010;16(1):RA1 RA Taddei S, Salvetti A. Endothelial dysfunction in essential hypertension: clinical implications. J Hypertens 2002;20(9): Timmers HJLM, Wieling W, Karemaker JM, et al. Cardiovascular responses to stress after carotid baroreceptor denervation in humans. Ann NY Acad Sci 2004;1018(1): Volders PGA. Novel insights into the role of the sympathetic nervous system in cardiac arrhythmogenesis. Heart Rhythm 2010;7(12): Autonomic Dysfunction and Hypotension. London: Springer; Friedman BH, Allen MT, Christie IC, et al. Validity concerns of common heart-rate variability indices. IEEE Eng Med Biol Mag 2002;21(4):35 40.