- Email: [email protected]
Heart rate response during exercise testing and ambulatory ECG monitoring in patients with syndrome X
Heart rate response during exercise testing and ambulatory ECG monitoring in patients with syndrome X The response of the heart rate during exercise testlng and 24-hour ambulatory electrocardiographic (ECG) monitoring performed with patients not receiving antianginal treatment was assessed in 26 patients (9 men and 17 women; mean age 61 k 6 years) with syndrome X (angina pectoris with normal coronary arteries), in 27 patients with coronary artery disease (10 men and 17 women; mean age 65 f 9 years), and in 21 healthy subjects (8 men and 13 women; mean age 47 +- 11 years). In patients with syndrome X the slope of the regression line of heart rate versus time (heart rate/time slope) during exercise testing was similar to that of patients with coronary artery disease (3.3 + 0.8 versus 3.1 k 1.2 beatsjmin), but significantly lower than that in healthy subjects (4.2 + 1.1 beats/min; p < 0.003). In patients with syndrome X the intercept of the heart rate/time slope was significantly higher than that in coronary artery disease patients and healthy subjects (102 + 15, 86 + 18, and 90 k 18 beats/min, respectively; p < 0.015). Resting preexercise heart rate was also significantly higher In syndrome X, compared with coronary artery disease patients and healthy subjects (91 + 16, 79 k 16, and 80 rt 14 beats/min, respectively). During ambulatory ECG monitoring, mean diurnal heart rate (from 6 AM to 6 PM) was higher in patients with syndrome X (83 + 8 beats/min) than in patients with coronary artery disease (75 k 8 beats/min) and healthy subjects (74 + 11 beats/min) (p < 0.02). Mean nocturnal heart rate, however, was not significantly different In the three groups (72 + 11 versus 70 k 8 beats/min versus 69 + 10 beats/mln, respectively; p = NS). Resting circulating plasma catecholamine levels (measured in 10 of the 26 patients with syndrome X) were within normal limits. Thus the heart rate/time slope during exercise testing In patients with syndrome X is similar to that of patients with coronary artery disease and significantly different from that in healthy subjects. Compared with coronary artery disease patients and healthy subjects, patients with syndrome X exhibit a higher mean diurnal heart rate, which is not associated with increased resting catecholamine levels. (AM HEART J 1991;122:458.)
Alfred0 R. Galassi, MD, Juan Carlos Kaski, MD, Filippo Crea, MD, Giuseppe Pup&a, MD, Stavros Gavrielides, MD, Dimitris Tousoulis, MD, and Attilio Maseri, MD. London, England
The exercise response of patients with “syndrome X”l (angina pectoris, positive exercise test, no evidence of coronary spasm, and angiographically normal coronary arteries) appears to be indistinguishable from that of patients with coronary artery disease.2* 3 In some patients with syndrome X, systolic blood pressure and heart rate response during exercise are disproportionate to work load,4 which might suggest a role of an inappropriate sympathetic From the Cardiovascular Unit, Royal Postgraduate Medical School, Hammersmith Hospital. Received for publication Nov. 29, 1990; accepted Jan. 20, 1991. Reprint requests: Dr. Juan Carlos Kaski, MD, Cardiovascular Research Unit, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane Road, London W12, ONN, England. 411129562
458
response in the pathogenesis of this syndrome. A recent report? suggests that resting myocardial blood flow is increased in some of these patients. The heart rate, which is controlled predominantly by the autonomic nervous system, but also by central and peripheral reflexes and hormonal influences, reflects an integrated physiologic response.6 Previous investigations have demonstrated that a lower than normal heart rate response to exercise is a useful predictor of the presence and severity of coronary artery disease.7-g However, heart rate changes induced by standardized exercise, or the pattern of heart rate during daily life, has not been systematically studied in patients with syndrome X. The purpose of the present study was to ascertain whether the behavior of the heart rate during exercise testing,
Volume 122 Number 2
and during 24-hour ambulatory electrocardiographic monitoring, differs in patients with syndrome X, patients with coronary artery disease, and healthy volunteers. METtlODS Study population Syndrome X. The study group consisted of 26 patients
(9 men and 17 women) aged 40 to 66 years (mean 51), diagnosed as having syndrome X on the basis of typical exertional angina, positive exercise tests (2 1 mm horizontal or downsloping ST segment depression and angina), angiographically normal epicardial coronary arteries, and no evidence of coronary spasm.lO None of the patients had cardiomyopathy, congestive heart failure, valvular heart disease (including mitral valve prolapse), diabetes mellitus, systemic hypertension, left ventricular hypertrophy, or history of myocardial infarction. The chest wall syndrome and esophageal disturbances were also ruled out in all cases.None of the patients had a history typical of variant angina, and epicardial coronary spasm was ruled out in every patient on the basis of negative ergonovine tests. Selective coronary angiography documented completely normal coronary arteries in all patients. Coronary flow reserve was assessed by positron emission tomography (dipyridamole-l&oxygen-water) in 11 of the 26 patients, and was found to be below the value of 2.5, as reported recently by Geltman et a1.5 in patients with angina and normal coronary arteriograms, using a similar technique. Coronary artery disease. Twenty-seven patients (10 men and 17 women), aged 36 to 65 years (mean 55), with chronic stable angina pectoris (no change in symptoms during the previous 3 months) and without previous myocardial infarction, were studied. All patients had positive exercise tests and angiographically documented coronary artery disease (luminal diameter reduction 270% of at least one major branch). All patients were in sinus rhythm and none had evidence of heart failure, cardiomyopathy, or valvular disease. Healthy subjects. There were 21 healthy subjects (8 men and 13 women), aged 31 to 67 years (mean 47). In each subject, exercise testing and ambulatory electrocardiographic monitoring were performed in the absence of symptoms or abnormal signs suggestive of disease. All subjects were free of chest pain, had no history of cardiac disease, and were not taking medications. All had a normal resting 12-lead electrocardiogram, normal blood pressure, and normal cardiac physical examination. Study protocol. The study protocol was approved by the Hospital Ethics Committee and every patient gave written informed consent before study entry. After appropriate gradual wash-out of all antianginal medications, all subjects underwent treadmill exercise testing and 24-hour ambulatory electrocardiographic monitoring during unrestricted daily activities. Exercise testing. In patients with syndrome X and in patients with coronary artery disease exercise end points were fatigue, progressive angina, or ~3 mm ST segment depression. In healthy subjects end points were fatigue or
Heart rate response in s?;rtdrome X
459
maximal heart rate predicted for age and sex. Twelve-lead electrocardiograms and blood pressure measurements were recorded during the control preexercise period, at the end of each stage during exercise, and at each minute during recovery. Three leads were continuously monitored throughout the test and for 4 minutes after exercise. The level of the ST segment 0.06 second after the J point was measured in each lead every minute by a computer-assisted system (CASE Marquette 12, Marquette Electronics Inc., Milwaukee, Wise.). In patients with syndrome X and coronary artery disease, exercise time, heart rate, and heart rate-systolic blood pressure (rate-pressure product) were measured at 1 mm of ST segment depression and at peak exercise. The slope and intercept of the regression line of heart rate versus time (heart rate/time slope) were obtained by taking all values measured minute by minute up to 1 mm of ST segment depression in patients, and up to 85% of the maximal value of heart rate predicted for age and sex in healthy subjects; the slope and intercept of the regression line of systolic blood pressure versus time (systolic blood pressure/time slope) were obtained in similar fashion to that of heart rate/time slope. Ambulatory electrocardiographic monitoring. Twentyfour-hour ambulatory electrocardiographic monitoring was performed with Marquette 8500 ‘L-channel amplitude modulated tape recorders (Marquette Electronics Inc.). In each subject two leads corresponding to modified Vs and V5 were selected. Recordings were analyzed with a Marquette laser Holter system 8500 (Marquette Electronics Inc.). The heart rate was measured on an analog recording of the heart rate versus time obtained from the playback system (trend recording). Maximal, minimal, and average heart rates were calculated every hour during the 24 hours. Mean diurnal (from 6 AM to 6 PM) and nocturnal (from 6 PM to 6 AM) hourly heart rate, number of episodes of transient myocardial ischemia (defined as a horizontal or downsloping ST segment shift 20.1 mV from baseline occurring 60 msec after the J point and lasting at least 60 seconds), and the duration of the ST segment depression were also assessed. Plasma catecholamine levels. To assess t.he role of circulating catecholamines, samples were assayed for plasma catecholamines in 10 of the 26 patients with syndrome X. Blood was taken on a single occasion between 11:00 AM and 12:00 AM while patients were not receiving treatment and after they had rested in the recumbent position for 30 minutes. Subjects were not allowed to smoke or drink fluids containing caffein for 12 hours before the assay. Catecholamines were measured by A double isotope derivative technique in which the enzyme catechol-O-methy1 transferase, prepared from rat liver, was used to convert noradrenaline and adrenaline into their respective 3O-methyl analogues, normetadrenaline and metadrenaline. In the presence of 3H-labeled S-adenosyl methionine, a tritium label was incorporated into the 3-O-methyl derivatives. To correct for variable recovery, 14C-labeled S-adenosyl methionine was used to attach a carbon- 14 label to the added noradrenaline and adrenaline standards. The 3-O-methyl derivatives were separated from each other by thin layer chromatography, scraped from the glass
460
Galassi
et al.
American
August 1991 Heart Journsl
I. Hemodynamic parameters at rest and during exercise testing in patients with syndrome X, in patients with coronary artery disease, and in healthy subjects
Table
Rest HR beats/ min
BP (mm
of ST
1 mm
Hg)
RPP (beatslmin x mm Hg)
Time (min)
HR (beats/ min)
depression
BP (mm
Hg)
Peak
exercise
RPP (beatslmin x mm Hg)
Time (min)
HR (beats/ min)
BP (mm Hg)
RPP (beatslmin x mm Hg)
Syndrome
91 k 16
127 + 18
116 f 33
10.8 + 2.6 138 k 15
168 k 28
234 k 51
11.7 f 2.3
144 2 16
169 2 27
245 IL 54
X CAD
79 IL 16*
134 k 15
105 + 26
10.7 rt 2.3 122 t- 15*
168 k 32
206 k 53*
11.6 z!z 1.7
128 + 15*
174 k 27
223 + 50
13.6 f 2.4*
158 2 19*
170 f
270 + 54
of maximal
85% Time
Healthy 80 iz 14* 124 f subjects BP, Blood pressure;
sure product
(beat&in
18
CAD, coronary mm Hg
100 + 28
HR
11.3 k 2.5 142 -+ 13
artery disease; HR, heart rate (beats/min); 10-2j.
HR BP
163 f
RPP 18
231 k 29
8.5% of maximal
HR, predicted
22
for age and sex; RPP, heart rate-blood
pres-
*p < 0.015.
plates, and oxidized to the toluene-soluble 3H- and 14Cvanillin, which were quantitated simultaneously by channel ratio scintillation counting. This technique has been shown to provide high precision and sensitivity.ll Statistical analysis. Results are expressed as mean + standard error. Heart rate/time slope and systolic blood pressure/time slope for each individual subject were obtained by means of regression analysis. Comparisons between groups were performed using Student’s t test for unpaired observations. A value of p < 0.015, using the Bonferroni correction, was considered statistically significant. RESULTS Exercise
significantly
Preexercise resting heart rate was higher in patients with syndrome X
testing.
than in patients with coronary artery disease or healthy subjects (Table I). During exercise, all patients with syndrome X and patients with coronary artery disease had I 1 mm of ST segment depression. Although all patients with syndrome X had angina during exercise, in 12 of the 26 (46%) patients the exercise test was stopped because of fatigue and in the remaining 14 (54%) it was stopped because of progressive angina. In 6 of the 27 (22 % ) patients with coronary artery disease the test was stopped because 13 mm of ST segment depression developed in the absence of angina; in 10 patients (37% ) it was stopped because of fatigue; and in 11 patients (41% ) the test was stopped because of progressive angina. In 16 of the 21 healthy subjects (76%) the exercise test was stopped because the theoretical maximal heart rate had been achieved, and in the remaining five (24 % ) it was stopped because of fatigue. AII healthy subjects reached
adjusted for age.
~45%
of the maximal
heart rate
Patients with syndrome X and patients with coronary artery disease had similar peak exercise time (11.7 f 2.3 and 11.6 + 1.7 minutes, respectively), which was lower than that of healthy subjects (13.6 + 2.4 minutes) (p < 0.01). Heart rate at peak exercise and at 1 mm of ST segment depression, and rate-pressure product at 1 mm of ST segment depression were significantly higher in syndrome X patients than in coronary artery disease patients (Table I). However, rate-pressure product at peak exercise was not significantly different in syndrome X patients, in coronary artery disease patients, or in healthy subjects (Table I). Compared with healthy subjects, patients with syndrome X had significantly lower peak exercise time and heart rate (11.7 f 2.3 versus 13.6 + 2.4 minutes, p < 0.01; and 144 + 16 versus 158 + 19 beats/min, p < 0.015, respectively) (Table I). In patients with syndrome X, mean heart rate/time slope was similar to that of patients with coronary artery disease, but significantly lower than that observed in healthy subjects (3.3 f 0.8, 3.1 + 1.2, and 4.2 + 1.1 beats/min, respectively; p < 0.003) (Fig. 1). Mean systolic blood pressure/time slope was similar inaIlthreegroups(3.4 + 2.0,3.5 t- 2.2,and3.5 * 1.5 beats/min x mm Hg/min, respectively). No correlation was found between resting heart rate values and heart rate/time slope (Fig. 2). In patients with syndrome X the intercept of the heart rate/time slope was significantly higher than that of patients with coronary artery disease and healthy subjects (102 + 18 versus 86 + 18 beats/min and 90 f 16 beats/min, respectively, p < 0.015) (Fig. 1). The intercept of systolic blood pressure/time slope was similar in all three groups (13.200 + 2.900,
Volume
122
Number
2
Heart
rate response in s,*kwdrrme X
461
140 1
160
F a & 5 5
0
1201
l .
140 0 0
120 A:,
100
: AA
I" 80
A
,
40 12
60 i.,‘,‘,‘,‘,‘,.,.,.,‘I 0 2 4
6
8
10
1214
During exercise
1234
Recovery Time
(min)
1. Heart rate/time slope during exercise testing in patients with syndrome X (closed circles), in patients with coronary artery disease (triangles), and in healthy subjects (open circles). *l mm of ST segment depression in patients and 85 % of maximal heart rate in healthy subjects; tpeak exercise.
Fig.
13.100 + 2.500, and 12.700 -t 1.800 beats/min X mm Hg, respectively). In the recovery period, the pattern of heart rate was similar in patients with syndrome X, in those with coronary artery disease, and in healthy subjects. The heart rate of patients with syndrome X and coronary artery disease returned to the preexercise period value (94 + 15 versus 91 + 18 beats/min and 80 + 12 versus 79 +- 16 beats/min, respectively; p = NS) within 4 minutes into the recovery period. At the fourth minute of the recovery period, healthy subjects showed a significantly higher heart rate than that observed during the preexercise period (99 + 19 versus 80 + 14 beats/min, p < 0.015) (Fig. 1). Ambulatory electrocardiographic monitoring. Patients with syndrome X had a significantly higher mean hourly heart rate (from 6 AM to 6 PM) than patients with coronary artery disease and healthy subjects (84 + 8,75 +- 8, and 75 f 11 beats/min, respectively; p < 0.015). During the night, however, differences in mean hourly heart rate did not reach statistical significance (Fig. 3). The pattern of maximal and minimal heart rates observed during the 24 hours was similar in patients with syndrome X, patients with coronary artery disease, and healthy subjects. However, diurnal maximal and minimal hourly heart rate was significantly higher in syndrome X than in coronary artery disease patients and healthy subjects (Fig. 4). In patients with syndrome X and patients with coronary artery disease, 42 and 38 episodes of ST
,
,
,
3
4 Slope
0
,
, 5
,
, 6
,
( 7
2. The correlation between values of heart rate at rest during the preexercise period and heart rate/time slope was assessed in syndrome X patients Cclosed circles), in coronary artery disease patients (triangles). and in healthy subjects (open circles). Fig.
segment depression, respectively, were observed (mean duration 23 k 15 minutes; range: 2 to 48, and 21 I 18 minutes; range: 2 to 50, respectively; p = NS). The circadian distribution of the ischemic episodes was similar in the two groups: in patients with syndrome X, 32 episodes occurred between 6 PM and 6 AM and 10 episodes occurred between 6 PM and 6 AM. h patients with coronary artery disease, 29 episodes occurred between 6 PM and 6 AM, and nine episodes occurred from 6 PM to 6 AM. Plasma catecholamines. Mean plasma noradrenaline and adrenaline levels were 0.73 t 0.18 rig/ml (range from 0.52 to 1.14 rig/ml) and 0.12 +- 0.09 ng/ ml (range from 0.012 to 0.30 rig/ml), respectively. These values fell within the normal range of our laboratory (noradrenaline: 0.2 to 0.8 rig/ml and adrenaline: 0.02 to 0.2 rig/ml). In one patient noradrenaline and in another adrenaline levels were slightly above the normal range (1.14 rig/ml and 0.30 ngfml, respectively).l’ DISCUSSION
This study shows that during exercise testing patients with syndrome X have a heart rate/time slope similar to that of coronary artery diseme p&tents but significantly less steep in these two groups than in healthy subjects. Furthermore, during continuous 24-hour ambulatory electrocardiographic modtoring, patients with syndrome X have a higher mean diurnal heart rate than patients with coronary artery disease and healthy subjects. Numerous studies have shown that in padienta with typical angina pectoris and normal coronary arteries myocardial ischemia may develop due to an impaired vasodilatory capacity of the coronary microcircu1ation.l. la-l5 It has been suggested that a
Galassi et al.
462
E z
American
80-
F? 5 I"
70&IL; .I.I'I'I'I'I'I.I.I'I.I'I 0 2 4 6
60
Diurnal variation 8
10122
4
a.m.
1
, 6
8
1012
p.m.
Time
(hours)
Fig. 3. Circadian variation of the average heart rate during 24-hour ambulatory electrocardiographic monitoring in patients with syndrome X (closed circles), in patients with coronary artery disease (triangles), and in healthy subjects (open circles).
1minimal 50 0
2
4
6
81012
a.m.
2
p.m.
4
6
81012
Time(hours)
Fig. 4. Circadian variation of maximal and minimal values of heart rate observed during 24-hour ambulatory electrocardiographic monitoring in patients with syndrome X (closed circles), in patients with coronary artery disease (triangles), and in healthy subjects (open circles).
functional abnormality of the intramyocardial prearteriolar vessels is present in such patienta,i5-l7 as indicated by the provocation of myocardial ischemia during atria1 pacing and its worsening in response to ergonovine,13p la and the provocation of anginal pain and ST segment depression with coronary arteriolar vasodilatory agents such as dipyridamole.14 Indeed, in our study coronary flow reserve, assessed by dipyridamole-positron emission tomography, was reduced in all 11 patients tested, consistent with recent findings by Geltman et a1.5 Based mainly on clinical observations and on the response to treatment with &blocking agents,4 it has
August 1991 Heart Journal
been suggested that an inappropriate sympathetic drive plays a role in the pathogenesis of myocardial ischemia in patients with syndrome X. Increasing sympathetic output by stimulation of the nuclei of the sympathetic nervous system in the medulla oblongata has been shown to increase heart rate and systemic vascular resistance in experimental animals.lg In this study, 24-hour ambulatory electrocardiographic monitoring showed that during daytime hours patients with syndrome X had significantly higher heart rates than patients with coronary artery disease and healthy subjects. It is of interest that this finding was not associated with higher plasma catecholamine levels, as circulating noradrenaline and adrenaline measured during the daytime hours in about 40% of our syndrome X patients were within normal limits. However, although we are aware that a more accurate estimate of the role of circulating catecholamines would have been obtained through the assessment of the circadian variation of catecholamine levels during the 24 hours, in this study we aimed at sampling at a time of day when catecholamine levels were probably the highest, according to the known circadian variation in plasma catecholamine levels in the general population.20 Furthermore, the study of the specific role of the sympathetic nervous system in the pathogenesis of syndrome X was beyond the scope of this study, as our main aim was to provide objective evidence of the behavior of heart rate both during exercise and ordinary daily life, in a well-characterized group of patients with the socalled syndrome X. Although the population of patients with syndrome X in this study was rather limited, these patients were selected on the basis of strict inclusion criteria. Furthermore, as age, sex and physical fitness are known parameters accounting for differences in heart rate response during exercise and daily life among individuals:, 21,22patients with coronary artery disease and healthy subjects were screened so as to have gender and physical fitness closely matching those of patients with syndrome X. In our study, patients with syndrome X showed a reduced heart rate response to exercise throughout the course of progressive increments in work load despite higher heart rate values during the preexercise period, and this resulted in a heart rate/time slope similar to that of patients with coronary artery disease. A disproportionately slow progression of heart rate in response to a progressive increase of work load during stress testing can be frequently observed in patients with ischemic heart disease and has been referred to as “chronotropic incompetence.‘@ Its cause, however, is poorly understood at present. The reason why chronotropic incompetence
Volume 122 Number 2
is also present in syndrome X is not readily available. Ischemia, sinus node disease, and left ventricular dysfunction, which have been invoked as possible explanations for chronotropic incompetence in patients with coronary artery disease,61 8 do not seem to apply to syndrome X. Some form of autonomic dysfunction may perhaps be postulated in this syndrome. Although it has been shown that the pattern of ST segment depression in syndrome X patients during exercise and recovery closely resembles that of patients with coronary artery disease,2* 3 these similar findings in patients with syndrome X and patients with coronary artery disease do not necessarily imply that myocardial ischemia is always the cause of angina and ST segment depression in syndrome X (“syndrome X” is still an ill-defined entity, which probably encompasses heterogeneous groups of patients). However, it may indicate that myocardial ischemia could play a role in at least some patients with syndrome X, as suggested by the results of several studies.l, 1:1-15,I8 The reasons for this different pattern of heart rate during exercise testing and also during daily life in patients with angina and normal coronary arteriograms are not clear and remain a matter of speculation. We acknowledge the useful advice and kind assistance of Mr. M. Walter Weg, MPhil, Department of Chemical Pathology, Queen Charlotte’s and Chelsea Hospital. We also thank Dr. Peter Bogaty for helpful discussions concerning this manuscript, and Jean Powell for the illustrations. REFERENCES
1. Opherk D, Zebe H, Weihe E, Mall G, Durr C, Gravert B, Mehme1 HC, Schwarz F, Kubler W. Reduced coronary dilatory capacity and ultrastructural changes of the myocardium in patients with angina pectoris but normal coronary arteriograms. Circulation 1981;63:817-25. 2. Pupita G, Kaski JC, Gala& AR, Gavrielides S, Crea F, Maseri A. Similar time course of ST segment depression during and after exercise in patients with coronary artery disease and syndrome X. AM HEART J 1990;120:848154. 3. Gavrielidas S. Kaski JC. Tousoulis D. Galassi AR. Burton P. Hackett, DR. kecovery-dhase patterns of ST segment depres: sion in the heart rate domain cannot distinguish between angina patients with and without coronary artery disease [Abstract]. J Am Co11 Cardiol 1991;17:193A. 4. Romeo F, Gaspardone A, Ciavolella M, GioffrB P, Reale A. Verapamil vs acebutolol for syndrome X. Am J Cardiol 1988;62:312-:3. 5. Geltman EM, Henes CG, Senneff MJ, Sobel BE, Bergmann
Heart rate response in s,xwirnnw X
6. 7. 8. 9.
10. 11.
12.
13.
14.
15. 16. 17 AI. 18. 19. 20.
21. 22.
463
SR. Increased myocardial perfusion at rest and diminished perfusion reserve in patients with angina and normal coronary arteries. J Am Co11 Cardiol 1990;16:586-95, Hammond HK, Froelicher VF. Normal and abnormal heart rate responses to exercise. Prog Cardiovasc Res 1985;4:271-96. Bruce RA, Fischer LD, Cooper MN, Gey GO. Separation of effects of cardiovascular disease and age on ventricular function with maximal exercise. Am J Cardiol 1974;34:757-62. Ellestad MH, Wan MKC. Predictive implications of stress testing-follow-up of 2700 subjects after maximum treadmill stress testing. Circulation 1975;51:363-8. McNeer JF, Margolis JR, Lee KL, Kisslo JA. Peter RH, Kong Y, Behar VS, Wallace AG, McCants CB, Rgjaati RA. The role of exercise test in the evaluation of pat,ients for ischemic heart disease. Circulation 1978;57:64-71. Kaski JC, Crea F, Nihoyannopoulos P, Hackett I), Maseri A.Transient myocardial ischemia during daily life in patients with syndrome X. Am J Cardiol 1986;58:1242-7. Brown MJ, Jenner DA. Novel double-isotope t,echnique for enzymatic assay of catecholamines, permitting high precision. sensitivity and plasma sample capacity. Clin Sci 1981;61: 591-8. Causon R. High performance liquid chromatographic separation and determination of catecholamines. chapter VII. In: Marks N, Rodnight R, eds. Research method in neurochemistry. vol. 6. New York: Plenum Publishing Corp. 1985:229. Cannon RO. Bonow RO. Bacharach SL. Green MV. Rosine DR, Leon MB, Watson dM, Epstein SE. Left ventricular dy;I function in patients with angina pectoris, normal epicardial coronary arteries, and abnormal vasodilaror reserve. Circulation 198x71:218-26. Cannon RO III, Schenke WH, Leon MB, Rosing DR. Urquart J, Epstein SE. Limited coronary flow reserve after dipyridnmole in patients with ergonovine-induced coronarv vasoconstriction. Circulation 1987;75:163-74. Cannon RO III, Epstein SE. “Microvascular angina” as a cause of chest pain with angiographically normal coronary arteries. Am J Cardiol 1988;61:1338-43. Epstein SE, Cannon RO. Site of increased c~~ronarq resistance to flow in patients with angina pectoris and normal epicardial coronary arteries. J Am Co11 Cardiol 1986;?+:459-61. Maseri A. Crea F. Kaski JC. Crake T. Mechanisms of angina pectoris in syndrome X. J Am Co11 Cardioi 1991;17:499&6. Cannon RO, Watson RM, Rosing D, Epstein SC. Angina caused by reduced vasodilator reserve oi amall coronarl arteries. .J Am Co11 Cardiol 1983:1:1359-‘i8 Alexander RS. Tonic and reflex &n&ions rif medullary sympathetic cardiovascular centers. .! Neuropl~y&l 1946;9:2O:i17. Stene M, Panagiotis N, Tuck ML, Sowers JR, Mayes D, Berg G. Plasma norepinephrine levels are influenced by sodium intake, glucocorticoid administration, and circadian changes in normal man. J Clin Endocrinol Metab 1980;51:1340-5. Sheffield LT, Maloof JA, Sawyer JA. Roitman D. Maximal heart rate and treadmill performance of healthy women in relation to age. Circulation 1978;57:79-84. Kostis JB, Moreyra AE, Amendo MT, Di l’iefro .J, Cosgrove N, Kuo PT. The effect of age on heart. rate in subjects free of heart disease; studies by ambulators electrocardiography and maximal exercise stress tesr. -‘irr,ulat,ion 198%; 65:141-i?.
Alfred0 R. Galassi, MD, Juan Carlos Kaski, MD, Filippo Crea, MD, Giuseppe Pup&a, MD, Stavros Gavrielides, MD, Dimitris Tousoulis, MD, and Attilio Maseri, MD. London, England
The exercise response of patients with “syndrome X”l (angina pectoris, positive exercise test, no evidence of coronary spasm, and angiographically normal coronary arteries) appears to be indistinguishable from that of patients with coronary artery disease.2* 3 In some patients with syndrome X, systolic blood pressure and heart rate response during exercise are disproportionate to work load,4 which might suggest a role of an inappropriate sympathetic From the Cardiovascular Unit, Royal Postgraduate Medical School, Hammersmith Hospital. Received for publication Nov. 29, 1990; accepted Jan. 20, 1991. Reprint requests: Dr. Juan Carlos Kaski, MD, Cardiovascular Research Unit, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane Road, London W12, ONN, England. 411129562
458
response in the pathogenesis of this syndrome. A recent report? suggests that resting myocardial blood flow is increased in some of these patients. The heart rate, which is controlled predominantly by the autonomic nervous system, but also by central and peripheral reflexes and hormonal influences, reflects an integrated physiologic response.6 Previous investigations have demonstrated that a lower than normal heart rate response to exercise is a useful predictor of the presence and severity of coronary artery disease.7-g However, heart rate changes induced by standardized exercise, or the pattern of heart rate during daily life, has not been systematically studied in patients with syndrome X. The purpose of the present study was to ascertain whether the behavior of the heart rate during exercise testing,
Volume 122 Number 2
and during 24-hour ambulatory electrocardiographic monitoring, differs in patients with syndrome X, patients with coronary artery disease, and healthy volunteers. METtlODS Study population Syndrome X. The study group consisted of 26 patients
(9 men and 17 women) aged 40 to 66 years (mean 51), diagnosed as having syndrome X on the basis of typical exertional angina, positive exercise tests (2 1 mm horizontal or downsloping ST segment depression and angina), angiographically normal epicardial coronary arteries, and no evidence of coronary spasm.lO None of the patients had cardiomyopathy, congestive heart failure, valvular heart disease (including mitral valve prolapse), diabetes mellitus, systemic hypertension, left ventricular hypertrophy, or history of myocardial infarction. The chest wall syndrome and esophageal disturbances were also ruled out in all cases.None of the patients had a history typical of variant angina, and epicardial coronary spasm was ruled out in every patient on the basis of negative ergonovine tests. Selective coronary angiography documented completely normal coronary arteries in all patients. Coronary flow reserve was assessed by positron emission tomography (dipyridamole-l&oxygen-water) in 11 of the 26 patients, and was found to be below the value of 2.5, as reported recently by Geltman et a1.5 in patients with angina and normal coronary arteriograms, using a similar technique. Coronary artery disease. Twenty-seven patients (10 men and 17 women), aged 36 to 65 years (mean 55), with chronic stable angina pectoris (no change in symptoms during the previous 3 months) and without previous myocardial infarction, were studied. All patients had positive exercise tests and angiographically documented coronary artery disease (luminal diameter reduction 270% of at least one major branch). All patients were in sinus rhythm and none had evidence of heart failure, cardiomyopathy, or valvular disease. Healthy subjects. There were 21 healthy subjects (8 men and 13 women), aged 31 to 67 years (mean 47). In each subject, exercise testing and ambulatory electrocardiographic monitoring were performed in the absence of symptoms or abnormal signs suggestive of disease. All subjects were free of chest pain, had no history of cardiac disease, and were not taking medications. All had a normal resting 12-lead electrocardiogram, normal blood pressure, and normal cardiac physical examination. Study protocol. The study protocol was approved by the Hospital Ethics Committee and every patient gave written informed consent before study entry. After appropriate gradual wash-out of all antianginal medications, all subjects underwent treadmill exercise testing and 24-hour ambulatory electrocardiographic monitoring during unrestricted daily activities. Exercise testing. In patients with syndrome X and in patients with coronary artery disease exercise end points were fatigue, progressive angina, or ~3 mm ST segment depression. In healthy subjects end points were fatigue or
Heart rate response in s?;rtdrome X
459
maximal heart rate predicted for age and sex. Twelve-lead electrocardiograms and blood pressure measurements were recorded during the control preexercise period, at the end of each stage during exercise, and at each minute during recovery. Three leads were continuously monitored throughout the test and for 4 minutes after exercise. The level of the ST segment 0.06 second after the J point was measured in each lead every minute by a computer-assisted system (CASE Marquette 12, Marquette Electronics Inc., Milwaukee, Wise.). In patients with syndrome X and coronary artery disease, exercise time, heart rate, and heart rate-systolic blood pressure (rate-pressure product) were measured at 1 mm of ST segment depression and at peak exercise. The slope and intercept of the regression line of heart rate versus time (heart rate/time slope) were obtained by taking all values measured minute by minute up to 1 mm of ST segment depression in patients, and up to 85% of the maximal value of heart rate predicted for age and sex in healthy subjects; the slope and intercept of the regression line of systolic blood pressure versus time (systolic blood pressure/time slope) were obtained in similar fashion to that of heart rate/time slope. Ambulatory electrocardiographic monitoring. Twentyfour-hour ambulatory electrocardiographic monitoring was performed with Marquette 8500 ‘L-channel amplitude modulated tape recorders (Marquette Electronics Inc.). In each subject two leads corresponding to modified Vs and V5 were selected. Recordings were analyzed with a Marquette laser Holter system 8500 (Marquette Electronics Inc.). The heart rate was measured on an analog recording of the heart rate versus time obtained from the playback system (trend recording). Maximal, minimal, and average heart rates were calculated every hour during the 24 hours. Mean diurnal (from 6 AM to 6 PM) and nocturnal (from 6 PM to 6 AM) hourly heart rate, number of episodes of transient myocardial ischemia (defined as a horizontal or downsloping ST segment shift 20.1 mV from baseline occurring 60 msec after the J point and lasting at least 60 seconds), and the duration of the ST segment depression were also assessed. Plasma catecholamine levels. To assess t.he role of circulating catecholamines, samples were assayed for plasma catecholamines in 10 of the 26 patients with syndrome X. Blood was taken on a single occasion between 11:00 AM and 12:00 AM while patients were not receiving treatment and after they had rested in the recumbent position for 30 minutes. Subjects were not allowed to smoke or drink fluids containing caffein for 12 hours before the assay. Catecholamines were measured by A double isotope derivative technique in which the enzyme catechol-O-methy1 transferase, prepared from rat liver, was used to convert noradrenaline and adrenaline into their respective 3O-methyl analogues, normetadrenaline and metadrenaline. In the presence of 3H-labeled S-adenosyl methionine, a tritium label was incorporated into the 3-O-methyl derivatives. To correct for variable recovery, 14C-labeled S-adenosyl methionine was used to attach a carbon- 14 label to the added noradrenaline and adrenaline standards. The 3-O-methyl derivatives were separated from each other by thin layer chromatography, scraped from the glass
460
Galassi
et al.
American
August 1991 Heart Journsl
I. Hemodynamic parameters at rest and during exercise testing in patients with syndrome X, in patients with coronary artery disease, and in healthy subjects
Table
Rest HR beats/ min
BP (mm
of ST
1 mm
Hg)
RPP (beatslmin x mm Hg)
Time (min)
HR (beats/ min)
depression
BP (mm
Hg)
Peak
exercise
RPP (beatslmin x mm Hg)
Time (min)
HR (beats/ min)
BP (mm Hg)
RPP (beatslmin x mm Hg)
Syndrome
91 k 16
127 + 18
116 f 33
10.8 + 2.6 138 k 15
168 k 28
234 k 51
11.7 f 2.3
144 2 16
169 2 27
245 IL 54
X CAD
79 IL 16*
134 k 15
105 + 26
10.7 rt 2.3 122 t- 15*
168 k 32
206 k 53*
11.6 z!z 1.7
128 + 15*
174 k 27
223 + 50
13.6 f 2.4*
158 2 19*
170 f
270 + 54
of maximal
85% Time
Healthy 80 iz 14* 124 f subjects BP, Blood pressure;
sure product
(beat&in
18
CAD, coronary mm Hg
100 + 28
HR
11.3 k 2.5 142 -+ 13
artery disease; HR, heart rate (beats/min); 10-2j.
HR BP
163 f
RPP 18
231 k 29
8.5% of maximal
HR, predicted
22
for age and sex; RPP, heart rate-blood
pres-
*p < 0.015.
plates, and oxidized to the toluene-soluble 3H- and 14Cvanillin, which were quantitated simultaneously by channel ratio scintillation counting. This technique has been shown to provide high precision and sensitivity.ll Statistical analysis. Results are expressed as mean + standard error. Heart rate/time slope and systolic blood pressure/time slope for each individual subject were obtained by means of regression analysis. Comparisons between groups were performed using Student’s t test for unpaired observations. A value of p < 0.015, using the Bonferroni correction, was considered statistically significant. RESULTS Exercise
significantly
Preexercise resting heart rate was higher in patients with syndrome X
testing.
than in patients with coronary artery disease or healthy subjects (Table I). During exercise, all patients with syndrome X and patients with coronary artery disease had I 1 mm of ST segment depression. Although all patients with syndrome X had angina during exercise, in 12 of the 26 (46%) patients the exercise test was stopped because of fatigue and in the remaining 14 (54%) it was stopped because of progressive angina. In 6 of the 27 (22 % ) patients with coronary artery disease the test was stopped because 13 mm of ST segment depression developed in the absence of angina; in 10 patients (37% ) it was stopped because of fatigue; and in 11 patients (41% ) the test was stopped because of progressive angina. In 16 of the 21 healthy subjects (76%) the exercise test was stopped because the theoretical maximal heart rate had been achieved, and in the remaining five (24 % ) it was stopped because of fatigue. AII healthy subjects reached
adjusted for age.
~45%
of the maximal
heart rate
Patients with syndrome X and patients with coronary artery disease had similar peak exercise time (11.7 f 2.3 and 11.6 + 1.7 minutes, respectively), which was lower than that of healthy subjects (13.6 + 2.4 minutes) (p < 0.01). Heart rate at peak exercise and at 1 mm of ST segment depression, and rate-pressure product at 1 mm of ST segment depression were significantly higher in syndrome X patients than in coronary artery disease patients (Table I). However, rate-pressure product at peak exercise was not significantly different in syndrome X patients, in coronary artery disease patients, or in healthy subjects (Table I). Compared with healthy subjects, patients with syndrome X had significantly lower peak exercise time and heart rate (11.7 f 2.3 versus 13.6 + 2.4 minutes, p < 0.01; and 144 + 16 versus 158 + 19 beats/min, p < 0.015, respectively) (Table I). In patients with syndrome X, mean heart rate/time slope was similar to that of patients with coronary artery disease, but significantly lower than that observed in healthy subjects (3.3 f 0.8, 3.1 + 1.2, and 4.2 + 1.1 beats/min, respectively; p < 0.003) (Fig. 1). Mean systolic blood pressure/time slope was similar inaIlthreegroups(3.4 + 2.0,3.5 t- 2.2,and3.5 * 1.5 beats/min x mm Hg/min, respectively). No correlation was found between resting heart rate values and heart rate/time slope (Fig. 2). In patients with syndrome X the intercept of the heart rate/time slope was significantly higher than that of patients with coronary artery disease and healthy subjects (102 + 18 versus 86 + 18 beats/min and 90 f 16 beats/min, respectively, p < 0.015) (Fig. 1). The intercept of systolic blood pressure/time slope was similar in all three groups (13.200 + 2.900,
Volume
122
Number
2
Heart
rate response in s,*kwdrrme X
461
140 1
160
F a & 5 5
0
1201
l .
140 0 0
120 A:,
100
: AA
I" 80
A
,
40 12
60 i.,‘,‘,‘,‘,‘,.,.,.,‘I 0 2 4
6
8
10
1214
During exercise
1234
Recovery Time
(min)
1. Heart rate/time slope during exercise testing in patients with syndrome X (closed circles), in patients with coronary artery disease (triangles), and in healthy subjects (open circles). *l mm of ST segment depression in patients and 85 % of maximal heart rate in healthy subjects; tpeak exercise.
Fig.
13.100 + 2.500, and 12.700 -t 1.800 beats/min X mm Hg, respectively). In the recovery period, the pattern of heart rate was similar in patients with syndrome X, in those with coronary artery disease, and in healthy subjects. The heart rate of patients with syndrome X and coronary artery disease returned to the preexercise period value (94 + 15 versus 91 + 18 beats/min and 80 + 12 versus 79 +- 16 beats/min, respectively; p = NS) within 4 minutes into the recovery period. At the fourth minute of the recovery period, healthy subjects showed a significantly higher heart rate than that observed during the preexercise period (99 + 19 versus 80 + 14 beats/min, p < 0.015) (Fig. 1). Ambulatory electrocardiographic monitoring. Patients with syndrome X had a significantly higher mean hourly heart rate (from 6 AM to 6 PM) than patients with coronary artery disease and healthy subjects (84 + 8,75 +- 8, and 75 f 11 beats/min, respectively; p < 0.015). During the night, however, differences in mean hourly heart rate did not reach statistical significance (Fig. 3). The pattern of maximal and minimal heart rates observed during the 24 hours was similar in patients with syndrome X, patients with coronary artery disease, and healthy subjects. However, diurnal maximal and minimal hourly heart rate was significantly higher in syndrome X than in coronary artery disease patients and healthy subjects (Fig. 4). In patients with syndrome X and patients with coronary artery disease, 42 and 38 episodes of ST
,
,
,
3
4 Slope
0
,
, 5
,
, 6
,
( 7
2. The correlation between values of heart rate at rest during the preexercise period and heart rate/time slope was assessed in syndrome X patients Cclosed circles), in coronary artery disease patients (triangles). and in healthy subjects (open circles). Fig.
segment depression, respectively, were observed (mean duration 23 k 15 minutes; range: 2 to 48, and 21 I 18 minutes; range: 2 to 50, respectively; p = NS). The circadian distribution of the ischemic episodes was similar in the two groups: in patients with syndrome X, 32 episodes occurred between 6 PM and 6 AM and 10 episodes occurred between 6 PM and 6 AM. h patients with coronary artery disease, 29 episodes occurred between 6 PM and 6 AM, and nine episodes occurred from 6 PM to 6 AM. Plasma catecholamines. Mean plasma noradrenaline and adrenaline levels were 0.73 t 0.18 rig/ml (range from 0.52 to 1.14 rig/ml) and 0.12 +- 0.09 ng/ ml (range from 0.012 to 0.30 rig/ml), respectively. These values fell within the normal range of our laboratory (noradrenaline: 0.2 to 0.8 rig/ml and adrenaline: 0.02 to 0.2 rig/ml). In one patient noradrenaline and in another adrenaline levels were slightly above the normal range (1.14 rig/ml and 0.30 ngfml, respectively).l’ DISCUSSION
This study shows that during exercise testing patients with syndrome X have a heart rate/time slope similar to that of coronary artery diseme p&tents but significantly less steep in these two groups than in healthy subjects. Furthermore, during continuous 24-hour ambulatory electrocardiographic modtoring, patients with syndrome X have a higher mean diurnal heart rate than patients with coronary artery disease and healthy subjects. Numerous studies have shown that in padienta with typical angina pectoris and normal coronary arteries myocardial ischemia may develop due to an impaired vasodilatory capacity of the coronary microcircu1ation.l. la-l5 It has been suggested that a
Galassi et al.
462
E z
American
80-
F? 5 I"
70&IL; .I.I'I'I'I'I'I.I.I'I.I'I 0 2 4 6
60
Diurnal variation 8
10122
4
a.m.
1
, 6
8
1012
p.m.
Time
(hours)
Fig. 3. Circadian variation of the average heart rate during 24-hour ambulatory electrocardiographic monitoring in patients with syndrome X (closed circles), in patients with coronary artery disease (triangles), and in healthy subjects (open circles).
1minimal 50 0
2
4
6
81012
a.m.
2
p.m.
4
6
81012
Time(hours)
Fig. 4. Circadian variation of maximal and minimal values of heart rate observed during 24-hour ambulatory electrocardiographic monitoring in patients with syndrome X (closed circles), in patients with coronary artery disease (triangles), and in healthy subjects (open circles).
functional abnormality of the intramyocardial prearteriolar vessels is present in such patienta,i5-l7 as indicated by the provocation of myocardial ischemia during atria1 pacing and its worsening in response to ergonovine,13p la and the provocation of anginal pain and ST segment depression with coronary arteriolar vasodilatory agents such as dipyridamole.14 Indeed, in our study coronary flow reserve, assessed by dipyridamole-positron emission tomography, was reduced in all 11 patients tested, consistent with recent findings by Geltman et a1.5 Based mainly on clinical observations and on the response to treatment with &blocking agents,4 it has
August 1991 Heart Journal
been suggested that an inappropriate sympathetic drive plays a role in the pathogenesis of myocardial ischemia in patients with syndrome X. Increasing sympathetic output by stimulation of the nuclei of the sympathetic nervous system in the medulla oblongata has been shown to increase heart rate and systemic vascular resistance in experimental animals.lg In this study, 24-hour ambulatory electrocardiographic monitoring showed that during daytime hours patients with syndrome X had significantly higher heart rates than patients with coronary artery disease and healthy subjects. It is of interest that this finding was not associated with higher plasma catecholamine levels, as circulating noradrenaline and adrenaline measured during the daytime hours in about 40% of our syndrome X patients were within normal limits. However, although we are aware that a more accurate estimate of the role of circulating catecholamines would have been obtained through the assessment of the circadian variation of catecholamine levels during the 24 hours, in this study we aimed at sampling at a time of day when catecholamine levels were probably the highest, according to the known circadian variation in plasma catecholamine levels in the general population.20 Furthermore, the study of the specific role of the sympathetic nervous system in the pathogenesis of syndrome X was beyond the scope of this study, as our main aim was to provide objective evidence of the behavior of heart rate both during exercise and ordinary daily life, in a well-characterized group of patients with the socalled syndrome X. Although the population of patients with syndrome X in this study was rather limited, these patients were selected on the basis of strict inclusion criteria. Furthermore, as age, sex and physical fitness are known parameters accounting for differences in heart rate response during exercise and daily life among individuals:, 21,22patients with coronary artery disease and healthy subjects were screened so as to have gender and physical fitness closely matching those of patients with syndrome X. In our study, patients with syndrome X showed a reduced heart rate response to exercise throughout the course of progressive increments in work load despite higher heart rate values during the preexercise period, and this resulted in a heart rate/time slope similar to that of patients with coronary artery disease. A disproportionately slow progression of heart rate in response to a progressive increase of work load during stress testing can be frequently observed in patients with ischemic heart disease and has been referred to as “chronotropic incompetence.‘@ Its cause, however, is poorly understood at present. The reason why chronotropic incompetence
Volume 122 Number 2
is also present in syndrome X is not readily available. Ischemia, sinus node disease, and left ventricular dysfunction, which have been invoked as possible explanations for chronotropic incompetence in patients with coronary artery disease,61 8 do not seem to apply to syndrome X. Some form of autonomic dysfunction may perhaps be postulated in this syndrome. Although it has been shown that the pattern of ST segment depression in syndrome X patients during exercise and recovery closely resembles that of patients with coronary artery disease,2* 3 these similar findings in patients with syndrome X and patients with coronary artery disease do not necessarily imply that myocardial ischemia is always the cause of angina and ST segment depression in syndrome X (“syndrome X” is still an ill-defined entity, which probably encompasses heterogeneous groups of patients). However, it may indicate that myocardial ischemia could play a role in at least some patients with syndrome X, as suggested by the results of several studies.l, 1:1-15,I8 The reasons for this different pattern of heart rate during exercise testing and also during daily life in patients with angina and normal coronary arteriograms are not clear and remain a matter of speculation. We acknowledge the useful advice and kind assistance of Mr. M. Walter Weg, MPhil, Department of Chemical Pathology, Queen Charlotte’s and Chelsea Hospital. We also thank Dr. Peter Bogaty for helpful discussions concerning this manuscript, and Jean Powell for the illustrations. REFERENCES
1. Opherk D, Zebe H, Weihe E, Mall G, Durr C, Gravert B, Mehme1 HC, Schwarz F, Kubler W. Reduced coronary dilatory capacity and ultrastructural changes of the myocardium in patients with angina pectoris but normal coronary arteriograms. Circulation 1981;63:817-25. 2. Pupita G, Kaski JC, Gala& AR, Gavrielides S, Crea F, Maseri A. Similar time course of ST segment depression during and after exercise in patients with coronary artery disease and syndrome X. AM HEART J 1990;120:848154. 3. Gavrielidas S. Kaski JC. Tousoulis D. Galassi AR. Burton P. Hackett, DR. kecovery-dhase patterns of ST segment depres: sion in the heart rate domain cannot distinguish between angina patients with and without coronary artery disease [Abstract]. J Am Co11 Cardiol 1991;17:193A. 4. Romeo F, Gaspardone A, Ciavolella M, GioffrB P, Reale A. Verapamil vs acebutolol for syndrome X. Am J Cardiol 1988;62:312-:3. 5. Geltman EM, Henes CG, Senneff MJ, Sobel BE, Bergmann
Heart rate response in s,xwirnnw X
6. 7. 8. 9.
10. 11.
12.
13.
14.
15. 16. 17 AI. 18. 19. 20.
21. 22.
463
SR. Increased myocardial perfusion at rest and diminished perfusion reserve in patients with angina and normal coronary arteries. J Am Co11 Cardiol 1990;16:586-95, Hammond HK, Froelicher VF. Normal and abnormal heart rate responses to exercise. Prog Cardiovasc Res 1985;4:271-96. Bruce RA, Fischer LD, Cooper MN, Gey GO. Separation of effects of cardiovascular disease and age on ventricular function with maximal exercise. Am J Cardiol 1974;34:757-62. Ellestad MH, Wan MKC. Predictive implications of stress testing-follow-up of 2700 subjects after maximum treadmill stress testing. Circulation 1975;51:363-8. McNeer JF, Margolis JR, Lee KL, Kisslo JA. Peter RH, Kong Y, Behar VS, Wallace AG, McCants CB, Rgjaati RA. The role of exercise test in the evaluation of pat,ients for ischemic heart disease. Circulation 1978;57:64-71. Kaski JC, Crea F, Nihoyannopoulos P, Hackett I), Maseri A.Transient myocardial ischemia during daily life in patients with syndrome X. Am J Cardiol 1986;58:1242-7. Brown MJ, Jenner DA. Novel double-isotope t,echnique for enzymatic assay of catecholamines, permitting high precision. sensitivity and plasma sample capacity. Clin Sci 1981;61: 591-8. Causon R. High performance liquid chromatographic separation and determination of catecholamines. chapter VII. In: Marks N, Rodnight R, eds. Research method in neurochemistry. vol. 6. New York: Plenum Publishing Corp. 1985:229. Cannon RO. Bonow RO. Bacharach SL. Green MV. Rosine DR, Leon MB, Watson dM, Epstein SE. Left ventricular dy;I function in patients with angina pectoris, normal epicardial coronary arteries, and abnormal vasodilaror reserve. Circulation 198x71:218-26. Cannon RO III, Schenke WH, Leon MB, Rosing DR. Urquart J, Epstein SE. Limited coronary flow reserve after dipyridnmole in patients with ergonovine-induced coronarv vasoconstriction. Circulation 1987;75:163-74. Cannon RO III, Epstein SE. “Microvascular angina” as a cause of chest pain with angiographically normal coronary arteries. Am J Cardiol 1988;61:1338-43. Epstein SE, Cannon RO. Site of increased c~~ronarq resistance to flow in patients with angina pectoris and normal epicardial coronary arteries. J Am Co11 Cardiol 1986;?+:459-61. Maseri A. Crea F. Kaski JC. Crake T. Mechanisms of angina pectoris in syndrome X. J Am Co11 Cardioi 1991;17:499&6. Cannon RO, Watson RM, Rosing D, Epstein SC. Angina caused by reduced vasodilator reserve oi amall coronarl arteries. .J Am Co11 Cardiol 1983:1:1359-‘i8 Alexander RS. Tonic and reflex &n&ions rif medullary sympathetic cardiovascular centers. .! Neuropl~y&l 1946;9:2O:i17. Stene M, Panagiotis N, Tuck ML, Sowers JR, Mayes D, Berg G. Plasma norepinephrine levels are influenced by sodium intake, glucocorticoid administration, and circadian changes in normal man. J Clin Endocrinol Metab 1980;51:1340-5. Sheffield LT, Maloof JA, Sawyer JA. Roitman D. Maximal heart rate and treadmill performance of healthy women in relation to age. Circulation 1978;57:79-84. Kostis JB, Moreyra AE, Amendo MT, Di l’iefro .J, Cosgrove N, Kuo PT. The effect of age on heart. rate in subjects free of heart disease; studies by ambulators electrocardiography and maximal exercise stress tesr. -‘irr,ulat,ion 198%; 65:141-i?.