Altered cardiovascular vagal responses in nonelderly female patients with subclinical hyperthyroidism and no apparent cardiovascular disease

Transcription

1 Clinical Endocrinology (2007) 67, doi: /j x ORIGINAL ARTICLE Blackwell Publishing Ltd Altered cardiovascular vagal responses in nonelderly female patients with subclinical hyperthyroidism and no apparent cardiovascular disease Renata Boschi Portella, Roberto Coury Pedrosa, Claudia Medina Coeli, Alexandru Buescu and Mario Vaisman Endocrine Service and Cardiology Service of the Federal University of Rio de Janeiro, Rio de Janeiro, Brazil Summary Objective Subclinical hyperthyroidism (SH) has been associated with exercise intolerance, changes in cardiac morphology, atrial arrhythmias and sympathovagal imbalance. The aim of this study was to evaluate the vagal reserve and modulation by a sympathetic stimulus in nonelderly patients with SH without cardiovascular problems. Design We carried out a cross-sectional study, comparing data of the heart rate variability (HRV) of SH patients and healthy controls at rest and after vagal and sympathetic stimulation. Patients We studied 16 female patients with at least 6 months of SH and 16 healthy female controls with the same median age (40 vs years). Measurements We used the tilt test, with electrocardiographic record at rest, during the respiratory sinus arrhythmia (RSA) manoeuvre and after tilting, in order to analyse HRV in the frequency domain (%high frequency (HF) and low/high frequency ratio (LF/HF) using Biopotentials Captation System software. Results The median TSH level was 0 03 mu/l in patients and 1 37 mui/l in controls. The median free T4 was 1 37 ng/dl in patients and 1 20 ng/dl in controls. Patients demonstrated a significantly smaller difference between %HF during the RSA and %HF at rest than controls (median 7 5 vs. 36 6, P < 0 001). There was a lower difference between LF/HF ratio after tilting and LF/HF ratio at rest in patients than in controls (1 5 vs. 5 3, P = 0 005). Conclusion Subclinical hyperthyroidism affects cardiovascular autonomic balance in otherwise apparently healthy nonelderly females by blunting vagal responses. (Received 31 October 2006; returned for revision 6 December 2006; finally revised 5 March 2007; accepted 8 March 2007) Correspondence: Renata Boschi Portella, Av. Oswaldo Cruz 73/2201 Flamengo, Rio de Janeiro, Brazil, cep Tel.: ; Fax: ; portella@novanet.com.br Introduction Subclinical hyperthyroidism can be defined as a serum thyrotropin (TSH) level below the lower limit of the reference range with normal thyroid hormone levels (free thyroxine (free T4) and triiodothyronine (T3). Euthyroid sick syndrome, central hypothyroidism, pregnancy, advanced age and the use of drugs that could suppress TSH must be excluded. 1 4 There is not enough evidence to support the need for immediate treatment in all patients who present with these laboratory changes, although there are many studies suggesting an increased cardiovascular morbidity and mortality in some patients. In patients over 60 years old there is increased risk of atrial fibrillation, 5,6 and increased mortality from cardiovascular causes within 5 years (not sustained in the 10-year analysis). 7 In middle-aged and elderly patients with suppressed postradioiodine TSH levels after treatment of hyperthyroidism, mortality does not seem to be increased. 8 Nevertheless, when symptomatic young, middle-aged and elderly patients have been grouped together, some morphological and functional cardiac alterations have been described, such as increased left ventricular mass index (without hypertrophy), 9 increased interventricular septum and left ventricular posterior wall thickness and decreased diastolic function; increased mean heart rate (HR) and increased frequency of atrial extra-systoles; decreased exercise performance; 13,15 increased intimal carotid thickness 16 and sympathovagal imbalance A study with asymptomatic patients showed only increased left ventricular mass and a higher mean HR, 21 and another one with nonelderly patients showed no exercise performance impairment. 22 Autonomic nervous system dysfunction plays an important role in the susceptibility to cardiac arrhythmias. 23 Classically, the adrenergic tone is enhanced in hyperthyroidism, 24 and catecholamine levels may be normal or increased. 25,26 There is a sympathovagal imbalance at rest, with a decreased vagal tone. 26 Cardiovascular autonomic disturbance is a mortality predictor in postacute myocardial infarction patients 27,28 and in diabetes mellitus. 28 To assess its role in the mortality rate of subclinical hyperthyroid patients, it would be necessary to perform long-term studies. There is no information on the influence of this dysfunction in nonelderly patients with no cardiac disease. 290 Journal compilation 2007 Blackwell Publishing Ltd

2 Dysautonomy in subclinical hyperthyroidism 291 Heart rate variability (HRV) has been established as a reliable and reproducible marker of cardiovascular autonomic function. 29 We chose HRV to assess cardiovascular autonomic system function (vagal reserve and sympathovagal balance after a sympathetic stimulus) in a sample of subclinical hyperthyroid patients without previous cardiovascular disease. Materials and methods A cross-sectional study was carried out. Files of adult patients followed at our endocrine clinic for subclinical hyperthyroidism were reviewed (n = 70). We excluded those without subclinical hyperthyroidism for at least 6 months (TSH serum levels below the lower reference limit and normal free T4 and T3 serum levels), patients with any chronic illness (especially cardiovascular, including hypertensive patients, even if treated), patients taking drugs that could interfere with the thyroid status or with the cardiovascular system, and patients 60 years. The remaining 16 patients had a 24-h Holter register free from arrhythmias, a normal echocardiogram and absence of myocardial ischaemia and a normal exercise performance for age, gender and body mass index (BMI) 30 in the cardiopulmonary test with measurement of gas exchange. The control group comprised 16 healthy female subjects from an outpatient population with similar socioeconomic status. Patients and controls signed an informed consent, approved by the local ethics committee. Both groups were submitted to the tilt test. The HRV was analysed in the electrocardiographic register obtained at rest and after simple manoeuvres designed to elicit sympathovagal responses. They were instructed to come in on the morning of the test after a 6-h fast and were led to an acclimatized and sound-proof room, where they laid on a hospital bed and were monitored with disposable electrodes in Frank s orthogonal derivations. The signals were collected and stored in a personal computer. The software used to process the data was the Biopotentials Captation System, developed by Barbosa in 1997, 31 which allows the examiner to eliminate QRS complexes with artefacts, to measure the HRV in the time and frequency domains. The method was validated for reproducibility in the same publication 31 and has been used in other studies. 32,33 After 5 min of rest, the patients underwent the respiratory sinus arrhythmia (RSA) manoeuvre, with a controlled respiratory rate of 12 incursions per min (over 2 min), to provoke a vagal response. 34,35 They remained supine until 20th min, when the bed was tilted to 70 degrees, and a sympathetic response was expected. 36 The frequency domain was analysed: the parameters used were the high frequency percentage (%HF), which represents the vagal component of the cardiovascular autonomic nervous system related to total power, and the low frequency: high frequency ratio (LF/HF ratio), which represents the sympathovagal balance. The HF and LF were measured in normalized units, which represent the relative value of each power component in proportion to the total power. The vagal reserve was measured by the increment in %HF during the RSA manoeuvre compared to rest. The vagal modulation of the sympathetic component was calculated by the increment in LF/HF ratio in the first 5 min after tilt compared to rest, where the greater differences reflect the best responses. We chose not to analyse timedomain parameters because they would be less reliable and less robust if any patient developed an atrial arrhythmia during the exam (which did not happen). 37 Serum TSH was measured by an immunometric assay (Immulite- DPC, Los Angeles, CA), with a reference range of mui/l and a sensitivity of mui/l. The serum free T4 and the serum total T3 were measured by chemoluminescence (Immulite-DPC), with a reference range of ng/dl and ng/dl and a sensitivity of 0 15 ng/dl and 7 ng/dl, respectively. For statistical analysis we used the Stata software (version 8 0, 2003) and CIA software. 38 Median values with interquartile range (IQR) were reported because of the skewed distribution of data. Variables were compared by the Mann Whitney U-test. We also presented 95% confidence intervals for the median differences. 39 We calculated Spearman s rank correlation coefficients to evaluate the correlation between %HF (at rest, during the RSA manoeuvre and after orthostatism) and the following explanatory variables: TSH level, free T4 level and time since diagnosis. We also calculated Spearman s rank correlation coefficients to evaluate the correlation between the formerly mentioned explanatory variables and the variation ( = RSA manoeuvre rest) of %HF. Finally, we evaluated the correlation between LF/HF ratio, after orthostatism and at rest, and the same explanatory variables. A P-value < 0 05 was accepted as significant for all analyses in the study. Results Patients had a median age of 40 years (IQR, 34 46) and a median BMI of 26 5 kg/m 2 (IQR, 25 27). The median time since diagnosis was 4 years (IQR, years). There was one smoker (who smoked 30 packs/year) and all patients had a sedentary lifestyle. There were four patients with exogenous subclinical hyperthyroidism (treated papillary thyroid carcinoma) and 12 patients with endogenous subclinical hyperthyroidism (three patients with multinodular goitres, two patients with hyperfunctioning adenomas and seven patients with diffuse goitres). The median serum TSH was 0 03 mu/l (IQR, ), the median serum free T4 was 1 37 ng/dl (IQR, ) and the median serum T3 was ng/dl (IQR, ) (Table 1). The control group had a median age of 34 5 years (IQR, ) and a median BMI of 24 8 kg/m 2 (IQR, ). There were two smokers (who smoked 20 and 27 5 packets/year, respectively) and all were sedentary. The median serum TSH was 1 37 mui/l (IQR, ) and the median serum free T4 was 1 20 ng/dl (IQR, ) (Table 1). There was no statistically significant difference between median age, BMI and serum free T4 between groups. The patients had a lower median serum TSH (P < 0 001). At rest, the median %HF in patients was 16 3% (IQR, ) and in controls, 17 8% (IQR, ), P = During the RSA manoeuvre, the median %HF in patients was 7 5% (IQR, ) and in controls, 51 1% (IQR, ), P < After orthostatism, the median %HF in patients was 13 2% (IQR, ) and in controls it was 4 7% (IQR, ), P = (Table 2). There was a decrease in %HF in patients after the RSA manoeuvre (median 7 5, IQR, 25 4 to 4 3) and an increase in controls (median 36 6, IQR, ), P < (Table 2). At rest, the median LF/HF ratio was 1 3 (IQR, ) in patients and 2 3 (IQR, ) in controls (P = 0 1). During the RSA

3 292 R. B. Portella et al. Table 1. Clinical and hormonal characteristics of patients and controls Variables Patients (n = 16) Median (IQR) Controls (n = 16) Median (IQR) Median difference (CI 95%) P-value Age (years) 40 (34 46) 34 5 ( ) 4 ( 5 to 10) 0 42 BMI (kg/m 2 ) 26 5 (25 27) 24 8 ( ) 1 3 ( 1 9 to 4 2) 0 27 TSH levels (mu/l) 0 03 ( ) 1 37 ( ) 1 31 ( 2 29 to 1 05) < FT4 levels (ng/dl) 1 37 ( ) 1 20 ( ) 0 15 ( 0 14 to 0 29) 0 31 BMI, body mass index; TSH, thyrotropin; FT4, free thyroxine; IQR, interquartile range; CI, confidence interval. Table 2. Frequency domain results obtained during the tilt test in patients and controls Variables Patients (n = 16) Median (IQR) Controls (n = 16) Median (IQR) Median difference (CI 95%) P-value %HF (%) At rest 16 3 ( ) 17 8 ( ) 3 9 ( 5 8 to 2 7) 0 42 During RSA manoeuvre 7 5 ( ) 51 1 ( ) 36 9 ( 48 6 to 16 3) < Orthostatism 13 2 ( ) 4 7 ( ) 6 1 ( ) RSA at rest 7 5 ( 25 4 to 4 3) 36 6 ( ) 46 9 ( 62 3 to 28 1) < LF/HF ratio At rest 1 3 ( ) 2 3 ( ) 0 9 ( 2 2 to 0 3) 0 10 During RSA manoeuvre 7 3 ( ) 0 2 ( ) 6 3 ( ) < Orthostatism 2 4 ( ) 7 7 ( ) 4 9 ( 8 5 to 1 7) Orthostatism at rest 1 5 ( ) 5 3 ( ) 3 8 ( 6 3 to 1 1) %HF, high frequency domain percentage; RSA manoeuvre, respiratory sinus arrhythmia manoeuvre; RSA at rest, difference between the %HF during the RSA manoeuvre and at rest/lf/hf ratio, ratio between low frequency and high frequency; Orthostatism at rest, difference between the LF/HF ratio after orthostatism and at rest; IQR, interquartile range; CI, confidence interval. manoeuvre, the median LF/HF ratio was 7 3 (IQR, ) in patients and 0 2 (IQR, ) in controls (P < 0 001). After orthostatism, it changed to 2 4 (IQR, ) in patients, and to 7 7 (IQR, ) in controls (P = 0 004) (Table 2). There was a higher increase in the median LF/HF ratio after orthostatism in controls (5 3, IQR, ) than in patients (1 5, IQR, ), P = (Table 2). There was no correlation between serum TSH, serum free T4 or time since diagnosis and the %HF (at rest, during the RSA manoeuvre or after orthostatism) or the LF/HF ratio (at rest, during the RSA manoeuvre or after orthostatism). The same applied when we tried to find a correlation between the former variables and the differences between the %HF during the RSA manoeuvre and at rest or the LF/HF ratio after orthostatism and at rest. Discussion This study presents strong evidence that, even in nonelderly female patients with subclinical hyperthyroidism and no apparent heart disease or cardiac changes attributable to the subclinical hyperthyroid state, the cardiovascular vagal modulation is affected. This was demonstrated by the reduced vagal reserve after the RSA manoeuvre (paradoxical decrement in %HF in patients compared to an increment in controls) and by the blunted vagal response to a sympathetic stimulus (lower increase in the LF/HF ratio after passive orthostatism compared to the healthy controls). Patients and controls had the same vagal status at rest. Three other studies used heart rate variability to assess cardiac autonomic function in endogenous subclinical hyperthyroid patients and found some alterations. Petretta et al. 18 analysed the 24-h Holter register of 30 patients at the National Research Council cybernetics laboratory and found reduced time-domain parameters compared to controls and a small low frequency power in frequency domain, without difference in the LF/HF ratio. Osman et al. 20 subjected 110 patients to a 24-h Holter register, with time-domain analysis using the Pathfinder 700 series analyser (Reynolds Medical), and found a reduction in some time-domain parameters and an altered heart rate turbulence. Only one study used approximately the same methodology as ours: Goichot et al. 19 performed a 20-min electrocardiogram in the supine position followed by a 15-min register after standing in 12 patients and found reduced time-domain parameters and no differences between patients and controls in the frequency domain at rest (LF, HF and LF/(LF+HF)); after standing, patients had a decrease in the time-domain parameters and a predominance of the sympathetic component of the autonomic nervous system in the frequency domain caused by a reduced vagal influence (increase in LF/ (LF+HF) and diminished (HF). The last response was different from our result, showing that those patients still had a physiological response after a sympathetic stimulus. However, their patients were submitted to active orthostatism and the response to RSA (vagal reserve) was not evaluated. The software used for the analysis was not specified. The three studies pointed towards vagal tone reduction at rest. TSH, free T4 serum levels and time since diagnosis had no influence on the test response in our study. Duration of disease could not

4 Dysautonomy in subclinical hyperthyroidism 293 be accurately given in the endogenous subclinical hyperthyroidism subgroup, although it is an important parameter. Only three of our patients had responses similar to controls in the tests performed. These three patients presented no distinguishing features that might explain this normal behaviour: two had a diffuse goitre, diagnosed 4 years ago and one had a treated papillary thyroid carcinoma, diagnosed 11 years ago. This could represent milder disease, as two of them (endogenous cases) had TSH serum levels > 0 1 mui/l (of a total of six) and the other patient had a TSH serum level < 0 1 mui/l (as did nine nonresponsive patients). Cardiovascular sympathovagal imbalance renders a worse prognosis to diabetic and postacute myocardial infarction patients. Atrial arrhythmias are more frequent in subclinical hyperthyroidism 5,6 and could be caused by sympathovagal imbalance. Mortality in elderly patients with a low TSH could be increased, 7 but this is still controversial. 6,8 Long-term studies are needed to confirm whether it is increased in nonelderly patients and which subgroups are most affected. It is not known whether treatment of subclinical hyperthyroidism or the use of β-blocking agents could improve this imbalance or affect morbidity and mortality. The morphological cardiac changes in overt hyperthyroid patients seem to be easily reversible after 1 month of treatment, 40 but there are no studies about the reversibility of autonomic changes. We have demonstrated that nonelderly female patients with subclinical hyperthyroidism are subjected to a cardiovascular autonomic modulation defect. The altered cholinergic reserve provides a logical explanation for the enhanced sympathetic activity found clinically in hyperthyroidism. If reproducible, these findings could contribute to the understanding of susceptibility to arrhythmias in subclinical hyperthyroidism and point towards a possible early therapeutic intervention. References 1 Col, N.F., Surks, M.I. & Daniels, G.H. (2004) Subclinical thyroid disease: clinical applications. Journal of the American Medical Association, 291, Helfand, M. (2004) Screening for subclinical thyroid dysfunction in nonpregnant adults: a summary of the evidence for the U.S. Preventive Services Task Force. Annals of Internal Medicine, 140, Braverman, L.E. & Pearce, E.N. (2002) Hipertireoidismo subclínico. Thyroid International, 2, McDermott, M.T., Woodmansee, W.W., Haugen, B.R., Smart, A. & Ridgway, E.C. (2003) The management of subclinical hyperthyroidism by thyroid specialists. Thyroid, 13, Sawin, C.T., Geller, A., Wolf, P.A., Belanger, A.J., Baker, E., Bacharat, P., Wilson, P.W.F., Benjamin, E.J. & D Agostino, R.B. (1994) Low serum thyrotropin concentrations as a risk factor for atrial fibrillation in older persons. New England Journal of Medicine, 331, Cappola, A.R., Fried, L.P., Arnold, A.M., Danese, M.D., Kuller, L.H., Burke, G.L., Tracy, R.P. & Ladenson, P.W. (2006) Thyroid status, cardiovascular risk, and mortality in older adults. Journal of the American Medical Association, 295, Parle, J.V., Maisonneuve, P., Sheppard, M.C., Boyle, P. & Franklyn, J.A. (2001) Prediction of all-cause mortality in elderly people from one low serum thyrotropin result: a 10-year cohort study. The Lancet, 358, Franklyn, J.A., Sheppard, M.C. & Maisonneuve, P. (2005) Thyroid function and mortality in patients treated for hyperthyroidism. Journal of the American Medical Association, 294, Dörr, M., Wolff, B., Robinson, D.M., John, U., Lüdemann, J., Meng, W., Felix, S.B. & Völzke, H. (2005) The association of thyroid function with cardiac mass and left ventricular hypertrophy. Journal of Clinical Endocrinology and Metabolism, 90, Biondi, B., Fazio, S., Carella, C., Amato, G., Cittadini, A., Lupoli, G., Saccà, L., Bellastella, A. & Lombardi, G. (1993) Cardiac effects of long term thyrotropin-suppressive therapy with levothyroxine. Journal of Clinical Endocrinology and Metabolism, 77, Fazio, S., Biondi, B., Carella, C., Sabatini, D., Cittadini, A., Panza, N., Lombardi, G. & Saccà, L. (1995) Diastolic dysfunction in patients on thyroid-stimulating hormone suppressive therapy with levothyroxine: beneficial effect of β-blockade. Journal of Clinical Endocrinology and Metabolism, 80, Biondi, B., Palmieri, E.A., Fazio, S., Cosco, C., Nocera, M., Saccà, L., Filetti, S., Lombardi, G. & Perticone, F. (2000) Endogenous subclinical hyperthyroidism affects quality of life and cardiac morphology and function in young and middle-aged patients. Journal of Clinical Endocrinology and Metabolism, 85, Mercuro, G., Panzuto, M.G., Bina, A., Leo, M., Cabula, R., Petrini, L., Pigliaru, F. & Mariotti, S. (2000) Cardiac function, physical exercise capacity and quality of life during long-term thyrotropin-suppressive therapy with levothyroxine: effect of individual dose tailoring. Journal of Clinical Endocrinology and Metabolism, 85, Sgarbi, J.A., Villaça, F.G., Garbeline, B., Villar, H.E. & Romaldini, J.H. (2003) The effects of early antithyroid therapy for endogenous subclinical hyperthyroidism in clinical and heart abnormalities. Journal of Clinical Endocrinology and Metabolism, 88, Biondi, B., Fazio, S., Cuocolo, A., Sabatini, D., Nicolai, E., Lombardi, G., Salvatore, M. & Saccà, L. (1996) Impaired cardiac reserve and exercise capacity in patients receiving long-term thyrotropin suppressive therapy with levothyroxine. Journal of Clinical Endocrinology and Metabolism, 81, Völzke, H., Robinson, D.M., Schminke, U., Lüdemann, J., Rettig, R., Felix, S.B., Kessler, C., John, U. & Meng, W. (2004) Thyroid function and carotid wall thickness. Journal of Clinical Endocrinology and Metabolism, 89, Faber, J., Wiinberg, N., Schifter, S. & Mehlsen, J. (2001) Haemodynamic changes following treatment of subclinical and overt hyperthyroidism. European Journal of Endocrinology, 145, Petretta, M., Bonaduce, D., Spinelli, L., Vicario, M.L.E., Nuzzo, V., Marciano, F., Camuso, P., De Sanctis, V. & Lupoli, G. (2001) Cardiovascular haemodynamics and cardiac autonomic control in patients with subclinical and overt hyperthyroidism. European Journal of Endocrinology, 145, Goichot, B., Brandenberger, G., Vinzio, S., Perrin, A.E., Geny, B., Schlienger, J.L. & Simon, C. (2004) Sympathovagal response to orthostatism in overt and in subclinical hyperthyroidism. Journal of Endocrinological Investigations, 27, Osman, F., Franklyn, J.A., Daykin, J., Chowdhary, S., Holder, R.L., Sheppard, M.C. & Gammage, M.D. (2004) Heart rate variability and turbulence in hyperthyroidism before, during and after treatment. The American Journal of Cardiology, 94, Shapiro, L.E., Sievert, R., Ong, L., Ocampo, E.L., Chance, R.A., Lee, M., Nanna, M., Ferrick, K. & Surks, M.L. (1997) Minimal cardiac effects in asymptomatic athyreotic patients chronically treated with thyrotropin-suppressive doses of 1-thyroxine. Journal of Clinical Endocrinology and Metabolism, 82, Portella, R.B., Silva, J.L., Wagman, M.B., Oliveira, F.P., Buescu, A. &

5 294 R. B. Portella et al. Vaisman, M. (2006) Exercise performance in young and middle-aged female patients with subclinical hyperthyroidism. Thyroid, 16, Olshansky, B. (2005) Interrelationships between the autonomic nervous system and atrial fibrillation. Progress in Cardiovascular Diseases, 48, Levey, G.S. (1990) Catecholamine thyroid hormone interactions and the cardiovascular manifestations of hyperthyroidism. The American Journal of Medicine, 88, Coulombe, P., Dussault, J.H., Letarte, J. & Simard, S.J. (1974) Catecholamines metabolism in thyroid diseases. I. Epinefrine secretion rate in hyperthyroidism and hypothyroidism. Journal of Clinical Endocrinology and Metabolism, 42, Burggraaf, J., Tulen, J.H.M., Lalezari, S., Schoemaker, R.C., De Meyer, P.H.E.M., Meinders, A.E., Cohen, A.F. & Pijl, H. (2001) Sympathovagal imbalance in hyperthyroidism. American Journal of Physiology Endocrinology and Metabolism, 281, E190 E Bigger, J.T., Fleiss, J.L. & Steinman, R.C. (1992) Correlations among time and frequency domain measures of heart period variability two weeks after myocardial infarction. The American Journal of Cardiology, 69, Nolan, J., Flapan, A.D., Goodfield, N.E., Prescott, R.J., Bloomfield, P., Neilson, J.M. & Ewing, D.J. (1996) Measurement of parasympathetic activity from 24-hours ambulatory electrocardiograms and its reproducibility and sensitivity in normal subjects, patients with symptomatic myocardial ischemia, and patients with diabetes mellitus. The American Journal of Cardiology, 77, Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996) Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation, 93, American College of Sport Medicine (2000) ACSM s guidelines for exercise testing and prescription, 6th edn. Lippincott Williams & Wilkins, Philadelphia, PA. 31 Barbosa, P.R.B., Barbosa Filho, J., Cordovil, I., Medeiros, A.B. & Nadal, J. (2000) Phase response of the spectral coherence between heart rate variability and ventricular repolarization duration in normal subjects. Computers in Cardiology, 27, Barbosa Filho, J., Barbosa, P.R. & Cordovil, I. (2002) Autonomic modulation of the heart in systemic arterial hypertension. Arquivos Brasileiros de Cardiologia, 78, Bittencourt, M.I., Barbosa, P.R., Drumond Neto, C., Bedirian, R., Barbosa, E.C., Brasil, F., Bonfim Ade, S. & Albanesi Filho, F.M. (2005) Assessing autonomic function in hypertrophic cardiomyopathy. Arquivos Brasileiros de Cardiologia, 85, Bainbridge, F.A. (1920) The relation between respiration and the pulse-rate. Journal of Physiology, London, 54, Baldwa, V.S. & Ewing, D.J. (1977) Heart rate response to Valsalva maneuver: reproducibility in normals and relation to variation in resting heart rate in diabetics. British Heart Journal, 39, Wieling, W., Borst, C., Karemaker, J.M. & Dunning, A.J. (1985) Testing for autonomic neuropathy: initial heart rate response to active and passive changes of posture. Clinical Physiology, 5 (Suppl.), Kleiger, R.E. & Bornes, M.S. (1996) Correlation of time and frequencydomain measures of heart rate variability. In: A. Moss, S. Stern eds. Noninvasive Electrocardiology Clinical Aspects of Holter Monitoring. W.B. Saunders Company Ltd, London, Bryant, T.N. (2000) CIA confidence interval analysis for Windows, version Available at: (accessed 1 May 2007). 39 Campbell, M.J. & Gardner, M.J. (2000) Medians and their differences. In: D.G. Altman, D. Machin, T.N. Bryant, M.J. Gardner eds. Statistics with Confidence. BMJ Books, London, Ching, G.W., Franklyn, J.A., Stallard, T.J., Daykin, J., Sheppard, M.C. & Gammage, M.D. (1996) Cardiac hypertrophy as a result of longterm thyroxine therapy and thyrotoxicosis. Heart, 75,