- Email: [email protected]
Directional variability in the R wave response during serial exercise testing in patients with coronary artery disease
Directional variability in the R wave response during serial exercise testing in patients with coronary artery disease The reproducibility of the direction of R wave amplitude response to exercise was analyzed in patients with coronary artery disease. Forty-three serial exercise tests were performed by 10 patients with exertional angina pectoris and documented coronary artery disease (CAD). Seventeen tests (37%) resulted in no change or an increase in R wave amplitude (abnormal response). Twenty-six tests (63%) resulted in a decrease in R wave amplitude. The direction of the R wave amplitude response was variable in at least one exercise test in 7 of 10 patients with CAD, all of whom had reproducible ischemic ST segment responses during serial testing. The inconsistent R wave response in these patients was unrelated to heart rate, workload, or duration of exercise. Because of the variability in the directional R wave response during serial exercise testing in CAD patients, we conclude that the R wave response during exercise is unreliable for the detection of CAD or ischemia-related myocardial dysfunction. (AM HEART J 108:983, 1984.)
Wyatt F. Voyles, M.D., Neale D. Smith, Albuquerque,
M.D., and Jonathan
Exercise testing is a frequently used technique for the evaluation of cardiac disease. ST segment changes which occur during exercise are of value in detecting coronary artery disease (CAD) in individual patients, but because of relatively poor speciflcity and sensitivity they are less helpful in screening ECG parameters large patient groups. lz2 Additional have been evaluated in efforts to improve the diagnostic sensitivity and specificity of exercise testing.3-5 Changes in R wave amplitude occur during exercise6e8 and an increase in R wave amplitude or a failure to decrease during exercise has been reported to be of value in detecting CAD in symptomatic4~5~!‘-11 and asymptomatic patients.12, l3 The etiology of the R wave response in patients with CAD has been related to ischemia-induced alterations in mechanical or electrical properties of the left ventricle.4* 1ri-1s If the direction of the R wave amplitude response to exercise is a reliable indicator of ventricular ischemia, it would be an additional exercise ECG parameter useful for detecting myocardial ischemia and From the Division of Cardiology, New Mexico School of Medicine. Received Reprint Medical
for publication
Abrams, M.D.
N.M.
March
Department 13, 1984;
requests: Wyatt F. Voyles, Center, 5400 Gibson Blvd.,
of Medicine, accepted
M.D., Research S.E., Albuquerque,
Apr.
University 29, 1984.
Division, Lovel~sce NM 87108.
of
would be helpful in determining the efficacy of drug therapy designed to prevent or alleviate exerciseinduced ischemia.20 In order to correctly interpret R wave changes during serial exercise testing, the directional R wave response to exercise must be shown to be reproducible in individual subjects. The purpose of this study was to examine the reproducibility of R wave changes which occur during exercise testing in patients with CAD. METHODS Patients. Forty-three treadmill exercise tests were performed in 10 patients with the clinical diagnosis of CAD and angina pectoris. These patients were participating in a serial treadmill study designed to evaluate the efficacy of a long-acting beta-blocking agent. All patients received placebos during the initial 4 weeks of their participation. The results of serial exercise tests performed weekly during the placebo testing period were analyzed and are the subject of this report. Coronary atherosclerosis was diagnosed by a positive treadmill test with at least 1% mm horizontal or downsloping ST segment depression 80 msec after the J point, a prior myocardial infarction documented by serial cardiac enzymes or ECG changes, or greater than 85% narrowing of the cross-sectional area of at least one coronary artery demonstrated by angiography. The clinical characteristics of these patients upon study entry are shown in Table I. All patients were males, with an average age of 58 years
983
October,
984
Voyles, Smith
Table
I. Clinical characteristics of coronary artery diseasepatients upon entry to study
Patient 1 2 3
4 5 6 7 8 9 10
Age 69
62 56 65 56 40 63 59 52 58
and Abrams
American
Duration of angina (mos) 30 48 60 3 30 9 96 192 12
24
NYHA class
Stress
I1 II I1 II 11 III II II I1 II
AS = anteroseptal; Ant = anterior; Inf = inferior; Inf-Lat = inferior-lateral; circumflex coronary artery; NYHA = New York Heart Association. Positive stress test is defined as at least 1 ‘,‘.I mm of horizontal or downsloping defined as greater than 85”( narrowing of vessel cross-sectional area.
(range = 40 to 69 years) and all had stable exertional angina for an average of 50.4 months (3 to 192 months). Exercise testing. Exercise testing was done on a motor-driven treadmill. No food or medications were taken within 2 hours of testing. Exercise beganwith a 10% grade at 1 mph with 0.5 mph increasesin speed each 3 minutes of exercise.Basedon the initial exercisetest, the treadmill speedwas adjusted to limit total exerciseduration to lessthan 10 minutes. Exercise wascontinued until the patient experienced mild (l+ to 2+) chest pain (scale of 4+) or was unable to continue because of fatigue. Standard 12-lead ECGs were done in the supine and standing positions before exercise.Continuous ECG monitoring wascarried out.during exerciseand recovery, and a 12-lead tracing was recorded during each minute of exerciseand recovery. The recovery phasewascontinued until heart rate, blood pressure, and ST segmentchanges (if any) returned to baseline. R wave amplitude wasdefined as the distance from the isoelectric line to the peak of the R wave,” and was measuredusingcalipers to the nearesttenth of a millimeter. All measurementswere taken from lead V, with the patient upright at rest and at the end of exercise. Four consecutive beats during rest and six consecutive beats at the end of exercise were averaged to determine the rest and exercise R wave response.Any decreasein R wave amplitude wascalled a negative response;no changeor an increase in R wave amplitude was called a positive response.”Downsloping or horizontal ST depression1% mm or more below baselineand persisting 80 msecafter the J point for at least five consecutivebeats wasconsidered a positive ST segment response. Records were reviewed independently by two authors (WV and NS) to determine R wave amplitude and ST segment,responses. RESULTS
A total of 43 exercise tests was performed by 10 CAD patients. Seven patients performed four exercise tests and three patients performed three exer-
Diagnostic MI
test
Positive
ST segment
Angiographg
LAD RCA,LAD,CIRC LAD LAD
Lat Inf-Lat Ant AS
Positive anterior
criteria
AS
Positive Positive
= left
1984 Journal
AS Ant Inf
Positive Positive Positive
LAD
Heart
descending;
depression
RCA
80 msec after
= right
coronary
the J point.
artery: Positive
CIRC angiography
= left is
cise tests. Two patients were retested and completed an additional series of three exercise tests each. The average duration of exercise was 6.5 + 4.8 minutes (mean ? SD) at an average maximal treadmill speed of 2.1 f 0.7 mph. The mean peak heart rate for the group was 134 +- 11 bpm. Mean maximal blood pressure responses were 162 t 4 mm Hg and 90 -t 13 mm Hg for systolic and diastolic pressures, respectively. The calculated double product (divided by 100) was 216.1 f 40.1. R wave response. There was complete agreement by two authors after independent analysis of the R wave amplitude and ST segment responses during exercise. The R wave amplitude response was positive in 17 tests (37 % ) and negative in 26 tests (63 % ). The mean positive R wave response was +1.3 mm (range = 0.0 to 5.2 mm) and the mean negative R wave response was -1.4 mm (0.2 to 4.9 mm). Three CAD patients (patients No. 4, 6, and 10) had consistently negative R wave amplitude responses in 10 tests. Directional R wave amplitude responses to exercise testing were inconsistent in at least one test in the remaining patients. No CAD patient had a consistently positive R wave amplitude response. The results of serial exercise testing in the seven CAD patients with inconsistent directional R wave responses are shown in Table II. Duration of exercise, heart rate, and work load achieved (double product) were examined to determine their effect on
R wave amplitude response in the seven patients with a variable R wave response. Four of seven patients had at least one negative R wave amplitude response after exercising a period of time equal to or longer than the duration of exercise that produced a positive R wave amplitude response. All seven patients had at least one negative R wave amplitude
Volume
108
Number
4, Part 1
Exercise R wave reaponse in CAD
985
Table Il. Results of serial exercise testing in the seven coronary artery disease patients with variable R wave
amplitude responses
Patient 1
2
Duration (min)
R waue amplitude response (mm) *
Heart rate (bpm)
Z)ouble product/l00
4.5
115
182.8
-(l.O)
7.0 9.7
115 130
200.1
+(l.o)
+ t +
120
226.2 177.6
-(0.8)
7.1
-(0.5)
+
5.3
125
185.0
HO.0)
+
5.3 3.3
130 150
197.6
-(1.5)
+
150 150
+(0.5) t(o.7)
t
4.9 8.0
261.0 267.0
145
+(0.6) 42.9)
+
9.3
276.0 263.9
9.0 4.4
140 150
231.0
t (0.0)
4.3
160
258.0 243.2
+(2.5) +(l.o)
5.4 5.0
135 125
195.8 225.0
t(1.0) +(0.5)
4.2
125
245.0
t(o.5)
4.7 8.0
125 135
167.5 179.5
-(l.O) -(4.9)
5.3 7.0
115 130
186.3
t(2.5)
10.2 3.0
140 125
215.9 249.2
41.9) -(5.2)
181.2
+(4.1)
+ +
3.0
135
3.2 3.5
130 140
170.0 208.0
+(3.0) -(0.2)
t +
130
182.0 192.4
-(1.4) +(0.4)
t
9.8 6.4 9.0
140 145
238.0
t(o.1)
9.0
145
266.8 217.5
-(0.8) -(0.4)
7.7 11.9
140 140
215.6
-(2.5)
8.1
120
193.2 158.4
t(o.0) -(0.8)
10.6
130
192.4
-(0.9)
*There was complete agreement on the direction of R wave amplitude change when were compared, In this table, the values for absolute changes in R wave amplitude
response after exercising to a heart rate equal to or greater than that occurring with a positive R wave response. Six of seven patients had at least one negative R wave response after achieving a workload equal to or greater than that achieved with a positive R wave amplitude response. ST response. There were 34 positive and 9 negative ST segment responses to exercise (Table II). Nine of 10 patients had consistent ST segment responses to exercise. Two patients with a consistently negative R wave amplitude response also had a consistently negative ST segment response to exercise. The third patient with a consistently negative R wave amplitude response had a negative ST segment response in three of four tests. In the seven patients with variable R wave amplitude response,
ST segment response
+ + + t + t + + + + + +
+ + + t + + t +
the results of independent measurements by two authors (NS and WV) as determined by one author (NS) are shown in parenthesis.
ST segment responses were consistently (see Table II).
positive
DISCUSSION Reproducibility. Wolthuis et a12’ have reported within-subject estimates of the variability of the magnitude of R wave amplitude changes in normal subjects. No prior studies have examined the reproducibility of the directional R wave amplitude response to exercise in CAD patients. In our study, records of serial exercise tests performed by patients with CAD were reviewed to determine the reproducibility of the R wave amplitude response to exercise. We found that, in contrast to the ST segment responses, the R wave amplitude responses to exercise were variable in 7 of 10 CAD patients (Table II).
October,
966
Voyles, Smith and Abrams
Furthermore, the directional variability in R wave responses did not appear to be a function of the duration of exercise, heart rate, or workload achieved during serial exercise testing. Mechanisms. Changes in several electromechanical properties of the heart have been postulated to be responsible for R wave changes during exercise. Alterations in electrical properties include changes in vector orientation,‘” augmented atria1 depolarization forces,14 prior myocardial damage that results in a change in the activation sequence during ischeconduction mia,14-‘7 and increased intramyocardial time in ischemic areas.lg Postulated changes in cardiac mechanical properties include changes in intraventricular blood mass or volume (Brody effect),4z lx ischemia-related ventricular dysfunction (e.g., segmental or global wall motion abnormalities) ,I. li changes in wall thickness,25-‘7 and increased stress on dyskinetic myocardial segments.“6 Other factors which appear to influence the R wave response to exercise include the extent of CAD,’ body position,‘” heart rate,‘,” duration of exercise,“” respiratory motion,4,2” and lead location.14 We have examined these factors in efforts to explain our observations. Indices of intraventricular blood mass or volume, ventricular function, and the precise extent of CAD were not determined during the time of exercise testing. However, serial tests were completed in a 4-week period during which there were no overt changes in the patients’ clinical status, suggesting stable left ventricular function during the study period. Other clinical studies2gr30 and recent experimental data26 fail to support the hypothesis that the R wave response during exercise in normals and in CAD patients is different because of exercise-related alterations in intraventricular volumes. Measurement. Variability resulting from errors in measurements of R wave amplitude was minimized by independent data evaluation. All R wave amplitude responses were determined with the patients standing, and changes resulting from respiratory motion were minimized by averaging beats. Using body surface mapping techniques, Mirvis14 demonstrated that lead location influences the R wave amplitude response during exercise in normals. In our study, all electrodes were placed by the same technician, but the torsos of the subjects tested were not marked. Variations in lead placement should have influenced responses in both normals and CAD patients, and this was not the case. Heart rate. Heart rate achieved during exercise has been noted to be particularly important in the interpretation of the R wave response.1,2’ Studies in
American
Heart
1984 Journal
normal subjects have shown an increase in R wave amplitude at lower heart rates, and a decrease at higher rates.‘, ‘j. “I Because of these observations, it has been suggested that diagnostic changes in R wave amplitude should be assessed only at heart rates of 120 bpm or greater.’ Furthermore, it has been postulated that the differences in R wave responses in normals and in CAD patients may be a result of lower heart rates usually achieved by CAD patients.“’ In the seven CAD patients with variable R wave responses, 30 of 33 exercise tests resulted in maximal heart rates greater than 120 bpm (Table II). CAD and R wave response. The mechanism of exercise-induced R wave amplitude changes in patients with CAD is unknown. Abdulmassih et a1.3’ have noted the variable results of studies examining the sensitivity and specificity of R wave amplitude changes during exercise to detect CAD. These authors suggest that changes in ejection fraction and ventricular volumes may independently affect the R wave response during exercise and account for the different results obtained in clinical studies. However, the work of Battler et a1.Z5 and David et al.?” provide important data which do not support volume changes during exercise as the mechanism of R wave amplitude changes. Ellestad3” has suggested that the increase in R wave amplitude which accompanies infarction, spasm, exercise, or injection of contrast material into coronary arteries is the result of ischemia-related changes in electrical properties of the ventricle. This suggestion is supported by experimental studies in dogs which have shown that R wave changes during ischemia are related to intramyocardial conduction delays.lg It seems unlikely that alterations in either the electrical or mechanical properties of the left ventricle would explain the variable results in studies using R wave amplitude changes to detect CAD. As previously suggested, both mechanical and electrical alterations may occur during exercise and may result in variable R wave response.17 The results of our study would indicate that either mechanical or electrical properties which result in R wave changes during exercise must vary within individual patients with CAD. If R wave changes are related to the extent of myocardial ischemia,3” one would have to postulate variable degrees of myocardial ischemia during serial exercise tests. We did not try to quantitate the extent of myocardial ischemia during our study, but all CAD patients with variable R wave responses had reproducible ischemic ST segment changes during each exercise test (Table II). The mechanism(s) of R wave amplitude changes
Volume Number
108 4, Part 1
during exercise and an explanation of the directional variability observed in these CAD patients requires further investigation. Conclusions. In this study, the direction of R wave amplitude response was inconsistent in 7 of 10 CAD patients. These seven patients had reproducible ST segment changes indicating ischemia during symptom-limited serial exercise testing. The variability of R wave amplitude response in these patients did not appear to be a function of the heart rate, duration, or workload of exercise. The mechanism of the R wave response to exercise remains unknown. Because of the directional variability in R wave amplitude response to exercise demonstrated in this study, we believe that interpretation of the R wave response in serial exercise testing should be done with caution, whether to determine efficacy of therapy used to reduce ischemia,20 or as a measure of ventricular dysfunction during exercise.34 Since all CAD patients had at least one exercise test which resulted in a decrease in R wave’amplitude, this study supports the conclusion of others,21~2s’2gthat the routine determination of the R wave response to exercise testing appears to have little diagnostic value. The authors gratefully acknowledge the technical support of Ms. Mona Berg and the secretarial support of Ms. Carolyn Johnson, Ms. Loretta McSweeney, Ms. Ruth Alexander Witt, and Ms. Linda Bernhardt. We also thank Ernest R. Greene, Ph.D., and Gregory S. Uhl, M.D., for reviewing this manuscript. REFERENCES
4.
5.
6.
7. 8.
9.
10.
Ellestad M: Stress testing, Principles and practice. Philadelphia, 1980, F.A. Davis Co, pp. 97, 193, and 303. Fortuin NJ, Weiss JL: Exercise stress testing. Circulation 56:699, 1977. Edwards H, Katzeff IE: Relation between S-T segment depression and S wave amplitude (Letter to editor). Am J Cardiol 36:270, 1975. Bonoris PE, Greenberg PS, Christison GW, Castellanet MJ, Ellestad MH: Evaluation of R-wave amplitude changes versus ST segment depression in stress testing. Am J Cardiol 57:904, 1977. Greenberg PS, Friscia DA, Ellestad MH: Predictive accurascy of Q-X/Q-T ratio. Q-Tc interval. St dearession and R-wave amplitude during’stiess testing. Am J Cardiol 44:18, 1979. Simonson E: Effect of moderate exercise on the electrocardiograph in healthy young and middle-aged men. J Appl Physiol 5:584, 1953. Lloyd-Thomas HG: The effect of exercise on the electrocardiogram in healthy subjects. Br Heart J 23:260, 1961. Bruce RA, Mazzareha JA, Jordan JW, Green E: Quantitation of QRS and ST segment responses to exercise. AM HEART J 7 1:455, 1966. Bonoris PE, Greenberg PS, Castellanet MJ, Ellestad MH: Significance of changes in R-wave amplitude during treadmill stress testing: Angiographic correlation. Am J Cardiol41:846, 1978. Christison GW, Bonoris PE, Greenberg PS, Castellanet MJ, Ellestad MH: Predicting coronary artery disease with treadmill stress testing: Changes in R-wave amplitude compared with ST segment depression. J Electrocardiol 12:179, 197!).
Exercise
R wave response
in CAD
987
11. Baron DW, Ilsley C, Sheiban I, Poole-Wilson PA, Rickards AF: R wave amplitude during exercise-relation to left ventricular function and coronary artery disease. Br Heart J 44~512, 12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
1980.
Yiannikas J, Marcomichelakis J, Taggart P, Kelly BH, Emanuel R: Analvsis of exercise-induced changes in R-wave amplitude in asymptomatic men with electrocardiographic ST-T changes at rest. Am J Cardiol 47:2X$ 1981. Uhl GS, Hopkirk JAC: Analysis of exercise-induced R-wave amplitude changes in detection of coronary artery disease in asymptomatic men with left bundle branch block. Am .J Cardiol 44:1247, 1979. Mirvis DM: Body surface distribution of exercise-induced QRS changes in normal subjects. Am -7 Cardiol 46:988, 1980. Ruffy R, Lovelace DE, Mueller TM, Knoebell SB, Zipes DP: Relationship between changes in left ventricular bipolar electrograms and regional myocardial blood flow during acute coronary occlusion &r the dog. Circ Res 45:764, 1979.Holland RP. Brooks H: The QRS comnlex durine mvocardial ischemia. J &in Invest 57:541, 1976. a Simoons ML: QRS changes in coronary artery disease (Letter to Editor). Circulation 59:841, 1979. Brody DA: A theoretical analysis of intracavitary blood mass influence on the heart-lead relationship. Circ Res 4:731, 1956. David D, Naito M, Michelson E, Watanabe E’, Chen CC, Morganroth J, Shaffenburg M, Blenko T: Intramyocardial conduction: A major determinant of R wave amplitude during acute myocardial ischemia. Circulation 65:161, 1982. Lerman J, Mele E, Chiozza M, Svetlize H, Perosio AM: Effects of nitrates on R-wave variations after exercise in coronary heart disease. Chest 80:137, 1981. Wolthuis RA, Froelicher VG, Hopkirk A, Fischer JR, Keiser N: Normal electrocardiographic waveform characteristics during treadmill exercise testing. Circulation 60:1028, 1979. Fabian J, Stolz I, Janota M, Rohac J: Reproducibility of exercise tests in patients with symptomatic ischaemic heart disease. Br Heart J 37:785, 1975. Lassvik C: Reproducibility of work performance at serial exercises in patients with angina pectoris. Stand J Clin Lab Invest 38~747, 1978. Fox KM, Selwyn AP, Shillingford JP: A method for precordial surface mapping of the exercise electrocardiogram. Br Heart J 40:1339, 1978. Battler A, Froelicher VF, Gallagher KP, Kumada T, McKown D. , Kemuer. WS. Ross J: Effects of changes in ventricular size on regional and surface QRS amplitudesin the conscious dog. Circulation 62:174, 1980. David D, Naito M, Chen CC, Michelson EL, Morganroth J, Schatfenburg M: R-wave amplitude variations during acute experimental myocardial ischemia: An inadequate index for changes in intracardiac volume. Circulation 63:1364, 1981. Lekven J, Chatterjee K, Tyberg JV, Parmley WW: Reduction in ventricular endocardial and epicardial potentials during acute increments in left ventricular dimensions. AM HEART J 98:200, 1979. Wagner S, Cohn K, Selzer A: Unreliability of exerciseinduced R-wave changes as indexes of coronary artery disease. Am J Cardiol 44:1241, 1979. Battler A, Froelicher V, Slutskv R, Ashburn W: Relationship of QRS amplitude changes during exercise to left ventricular function and volumes and the diagnosis of coronary artery disease. Circulation 60:1004, 1979. Ishikawa K, Berson AS, Pipberger HV: Electrocardiographic changes due to cardiac enlargement. AM HEAKT J 81:635, 1971. Simoons ML, Block P: Toward the optimal lead system and optimal criteria for exercise electrocardiography. Am J Cardiol 47:1366, 1981. Abdulmassih S, Iskandrian AS, Hakk AH, Horowitz L, Mintz r.
i
October,
Voyles, Smith and Abrams
American
GJ, Anderson GJ, Kane S, Segal BL: Changes m R wave during exercise: Correlation with left ventricular function and volumes. J Electrocardiol l&199, 1982. 33. Ellestad HM: The mechanism of exercise-induced R-wave amplitude changes in coronary heart disease. Arch Intern Med 142:963. 1982.
Heart
1984 Journal
34. Brown JB, Wynne J, Holman BL, Atkins HL, Cohn PI? Relationship between regional ejection fraction during exercise and ST and R wave changes on the ECG (abstr). J Am Co11 Cardiol 1:649, 1983.
Evaluation of habitual excessive alcohol consumption on myocardial infarction risk in coronary disease patients Overt coronary heart disease does occur at times in a setting of alcoholism. In an attempt to test the hypothesis that habitual excessive drinking may have an aggravating effect upon coexisting ischemic heart disease and may help precipitate new coronary events, we compared myocardial infarct prevalence among heavy drinkers and non-heavy drinkers with angiographically documented coronary artery disease. Infarct prevalence was found to be higher for heavy drinkers than for non-heavy drinkers under age 60 years, after controlling for differences in smoking habits and underlying atherosclerosis severity. A reversal in trend which may be due to the operation of selective factors causing premature coronary death among alcoholics was observed for individuals above age 60 years. These results, although open to differing interpretations, are consistent with the notion that heavy drinking has a destabilizing effect upon preexisting ischemic heart disease and may increase acute coronary event risk. (AM HEART J 108:988,
1984.)
Sigismond Deutscher, M.D., M.P.H., Howard E. Rockette, Venkatraman Krishnaswami, M.D. Pittsburgh, Pa.
While recent epidemiologic studies,‘-4 point to the existence of a negative correlation between ischemic heart disease and mild-to-moderate consumption of alcohol, little is known about the relationship of the former with heavy drinking. More particularly, the question arises whether habitual excessive drinking might not aggravate the manifestations of preexisting coronary disease and even increase infarct risk through the various changes it is known to impart upon the cardiovascular system.s If the answer to this question is affirmative, one might expect heavy drinkers to develop severe ischemic manifestations From the Veterans Administration Medical Center and the Department Medicine, University of Pittsburgh School of Medicine; and from Department of Biostatistics, Graduate‘ School of Public Health. Received accepted
for publication March 14, 1984.
Reprint requests: Sigismond cal Center, University Drive
988
Oct.
1, 1983;
revision
received
Deutscher, M.D., Medical C, Pittsburgh, PA 15240.
March Service,
of the
5, 1984; VA Medi-
Ph.D., and
and incur new acute coronary events more frequently than non-heavy drinkers with comparable coronary artery involvement. Taken one step further, this line of thought implies that heavy drinkers with relatively little coronary artery involvement who would otherwise have had only mild symptoms, might instead present with more severe disease manifestations. In an attempt to test the above ideas, we assessed the drinking habits of patients who underwent coronary angiography because they had developed severe ischemic manifestations; heavy drinkers were then compared to non-heavy drinkers in regard to the severity of their underlying atherosclerotic disease and, given a similar degree of atherosclerosis severity, to the frequency with which they had sustained a prior infarction. Since several report@+’ indicate that cigarette smoking and hypertension may be associated with both heavy drinking and
Wyatt F. Voyles, M.D., Neale D. Smith, Albuquerque,
M.D., and Jonathan
Exercise testing is a frequently used technique for the evaluation of cardiac disease. ST segment changes which occur during exercise are of value in detecting coronary artery disease (CAD) in individual patients, but because of relatively poor speciflcity and sensitivity they are less helpful in screening ECG parameters large patient groups. lz2 Additional have been evaluated in efforts to improve the diagnostic sensitivity and specificity of exercise testing.3-5 Changes in R wave amplitude occur during exercise6e8 and an increase in R wave amplitude or a failure to decrease during exercise has been reported to be of value in detecting CAD in symptomatic4~5~!‘-11 and asymptomatic patients.12, l3 The etiology of the R wave response in patients with CAD has been related to ischemia-induced alterations in mechanical or electrical properties of the left ventricle.4* 1ri-1s If the direction of the R wave amplitude response to exercise is a reliable indicator of ventricular ischemia, it would be an additional exercise ECG parameter useful for detecting myocardial ischemia and From the Division of Cardiology, New Mexico School of Medicine. Received Reprint Medical
for publication
Abrams, M.D.
N.M.
March
Department 13, 1984;
requests: Wyatt F. Voyles, Center, 5400 Gibson Blvd.,
of Medicine, accepted
M.D., Research S.E., Albuquerque,
Apr.
University 29, 1984.
Division, Lovel~sce NM 87108.
of
would be helpful in determining the efficacy of drug therapy designed to prevent or alleviate exerciseinduced ischemia.20 In order to correctly interpret R wave changes during serial exercise testing, the directional R wave response to exercise must be shown to be reproducible in individual subjects. The purpose of this study was to examine the reproducibility of R wave changes which occur during exercise testing in patients with CAD. METHODS Patients. Forty-three treadmill exercise tests were performed in 10 patients with the clinical diagnosis of CAD and angina pectoris. These patients were participating in a serial treadmill study designed to evaluate the efficacy of a long-acting beta-blocking agent. All patients received placebos during the initial 4 weeks of their participation. The results of serial exercise tests performed weekly during the placebo testing period were analyzed and are the subject of this report. Coronary atherosclerosis was diagnosed by a positive treadmill test with at least 1% mm horizontal or downsloping ST segment depression 80 msec after the J point, a prior myocardial infarction documented by serial cardiac enzymes or ECG changes, or greater than 85% narrowing of the cross-sectional area of at least one coronary artery demonstrated by angiography. The clinical characteristics of these patients upon study entry are shown in Table I. All patients were males, with an average age of 58 years
983
October,
984
Voyles, Smith
Table
I. Clinical characteristics of coronary artery diseasepatients upon entry to study
Patient 1 2 3
4 5 6 7 8 9 10
Age 69
62 56 65 56 40 63 59 52 58
and Abrams
American
Duration of angina (mos) 30 48 60 3 30 9 96 192 12
24
NYHA class
Stress
I1 II I1 II 11 III II II I1 II
AS = anteroseptal; Ant = anterior; Inf = inferior; Inf-Lat = inferior-lateral; circumflex coronary artery; NYHA = New York Heart Association. Positive stress test is defined as at least 1 ‘,‘.I mm of horizontal or downsloping defined as greater than 85”( narrowing of vessel cross-sectional area.
(range = 40 to 69 years) and all had stable exertional angina for an average of 50.4 months (3 to 192 months). Exercise testing. Exercise testing was done on a motor-driven treadmill. No food or medications were taken within 2 hours of testing. Exercise beganwith a 10% grade at 1 mph with 0.5 mph increasesin speed each 3 minutes of exercise.Basedon the initial exercisetest, the treadmill speedwas adjusted to limit total exerciseduration to lessthan 10 minutes. Exercise wascontinued until the patient experienced mild (l+ to 2+) chest pain (scale of 4+) or was unable to continue because of fatigue. Standard 12-lead ECGs were done in the supine and standing positions before exercise.Continuous ECG monitoring wascarried out.during exerciseand recovery, and a 12-lead tracing was recorded during each minute of exerciseand recovery. The recovery phasewascontinued until heart rate, blood pressure, and ST segmentchanges (if any) returned to baseline. R wave amplitude wasdefined as the distance from the isoelectric line to the peak of the R wave,” and was measuredusingcalipers to the nearesttenth of a millimeter. All measurementswere taken from lead V, with the patient upright at rest and at the end of exercise. Four consecutive beats during rest and six consecutive beats at the end of exercise were averaged to determine the rest and exercise R wave response.Any decreasein R wave amplitude wascalled a negative response;no changeor an increase in R wave amplitude was called a positive response.”Downsloping or horizontal ST depression1% mm or more below baselineand persisting 80 msecafter the J point for at least five consecutivebeats wasconsidered a positive ST segment response. Records were reviewed independently by two authors (WV and NS) to determine R wave amplitude and ST segment,responses. RESULTS
A total of 43 exercise tests was performed by 10 CAD patients. Seven patients performed four exercise tests and three patients performed three exer-
Diagnostic MI
test
Positive
ST segment
Angiographg
LAD RCA,LAD,CIRC LAD LAD
Lat Inf-Lat Ant AS
Positive anterior
criteria
AS
Positive Positive
= left
1984 Journal
AS Ant Inf
Positive Positive Positive
LAD
Heart
descending;
depression
RCA
80 msec after
= right
coronary
the J point.
artery: Positive
CIRC angiography
= left is
cise tests. Two patients were retested and completed an additional series of three exercise tests each. The average duration of exercise was 6.5 + 4.8 minutes (mean ? SD) at an average maximal treadmill speed of 2.1 f 0.7 mph. The mean peak heart rate for the group was 134 +- 11 bpm. Mean maximal blood pressure responses were 162 t 4 mm Hg and 90 -t 13 mm Hg for systolic and diastolic pressures, respectively. The calculated double product (divided by 100) was 216.1 f 40.1. R wave response. There was complete agreement by two authors after independent analysis of the R wave amplitude and ST segment responses during exercise. The R wave amplitude response was positive in 17 tests (37 % ) and negative in 26 tests (63 % ). The mean positive R wave response was +1.3 mm (range = 0.0 to 5.2 mm) and the mean negative R wave response was -1.4 mm (0.2 to 4.9 mm). Three CAD patients (patients No. 4, 6, and 10) had consistently negative R wave amplitude responses in 10 tests. Directional R wave amplitude responses to exercise testing were inconsistent in at least one test in the remaining patients. No CAD patient had a consistently positive R wave amplitude response. The results of serial exercise testing in the seven CAD patients with inconsistent directional R wave responses are shown in Table II. Duration of exercise, heart rate, and work load achieved (double product) were examined to determine their effect on
R wave amplitude response in the seven patients with a variable R wave response. Four of seven patients had at least one negative R wave amplitude response after exercising a period of time equal to or longer than the duration of exercise that produced a positive R wave amplitude response. All seven patients had at least one negative R wave amplitude
Volume
108
Number
4, Part 1
Exercise R wave reaponse in CAD
985
Table Il. Results of serial exercise testing in the seven coronary artery disease patients with variable R wave
amplitude responses
Patient 1
2
Duration (min)
R waue amplitude response (mm) *
Heart rate (bpm)
Z)ouble product/l00
4.5
115
182.8
-(l.O)
7.0 9.7
115 130
200.1
+(l.o)
+ t +
120
226.2 177.6
-(0.8)
7.1
-(0.5)
+
5.3
125
185.0
HO.0)
+
5.3 3.3
130 150
197.6
-(1.5)
+
150 150
+(0.5) t(o.7)
t
4.9 8.0
261.0 267.0
145
+(0.6) 42.9)
+
9.3
276.0 263.9
9.0 4.4
140 150
231.0
t (0.0)
4.3
160
258.0 243.2
+(2.5) +(l.o)
5.4 5.0
135 125
195.8 225.0
t(1.0) +(0.5)
4.2
125
245.0
t(o.5)
4.7 8.0
125 135
167.5 179.5
-(l.O) -(4.9)
5.3 7.0
115 130
186.3
t(2.5)
10.2 3.0
140 125
215.9 249.2
41.9) -(5.2)
181.2
+(4.1)
+ +
3.0
135
3.2 3.5
130 140
170.0 208.0
+(3.0) -(0.2)
t +
130
182.0 192.4
-(1.4) +(0.4)
t
9.8 6.4 9.0
140 145
238.0
t(o.1)
9.0
145
266.8 217.5
-(0.8) -(0.4)
7.7 11.9
140 140
215.6
-(2.5)
8.1
120
193.2 158.4
t(o.0) -(0.8)
10.6
130
192.4
-(0.9)
*There was complete agreement on the direction of R wave amplitude change when were compared, In this table, the values for absolute changes in R wave amplitude
response after exercising to a heart rate equal to or greater than that occurring with a positive R wave response. Six of seven patients had at least one negative R wave response after achieving a workload equal to or greater than that achieved with a positive R wave amplitude response. ST response. There were 34 positive and 9 negative ST segment responses to exercise (Table II). Nine of 10 patients had consistent ST segment responses to exercise. Two patients with a consistently negative R wave amplitude response also had a consistently negative ST segment response to exercise. The third patient with a consistently negative R wave amplitude response had a negative ST segment response in three of four tests. In the seven patients with variable R wave amplitude response,
ST segment response
+ + + t + t + + + + + +
+ + + t + + t +
the results of independent measurements by two authors (NS and WV) as determined by one author (NS) are shown in parenthesis.
ST segment responses were consistently (see Table II).
positive
DISCUSSION Reproducibility. Wolthuis et a12’ have reported within-subject estimates of the variability of the magnitude of R wave amplitude changes in normal subjects. No prior studies have examined the reproducibility of the directional R wave amplitude response to exercise in CAD patients. In our study, records of serial exercise tests performed by patients with CAD were reviewed to determine the reproducibility of the R wave amplitude response to exercise. We found that, in contrast to the ST segment responses, the R wave amplitude responses to exercise were variable in 7 of 10 CAD patients (Table II).
October,
966
Voyles, Smith and Abrams
Furthermore, the directional variability in R wave responses did not appear to be a function of the duration of exercise, heart rate, or workload achieved during serial exercise testing. Mechanisms. Changes in several electromechanical properties of the heart have been postulated to be responsible for R wave changes during exercise. Alterations in electrical properties include changes in vector orientation,‘” augmented atria1 depolarization forces,14 prior myocardial damage that results in a change in the activation sequence during ischeconduction mia,14-‘7 and increased intramyocardial time in ischemic areas.lg Postulated changes in cardiac mechanical properties include changes in intraventricular blood mass or volume (Brody effect),4z lx ischemia-related ventricular dysfunction (e.g., segmental or global wall motion abnormalities) ,I. li changes in wall thickness,25-‘7 and increased stress on dyskinetic myocardial segments.“6 Other factors which appear to influence the R wave response to exercise include the extent of CAD,’ body position,‘” heart rate,‘,” duration of exercise,“” respiratory motion,4,2” and lead location.14 We have examined these factors in efforts to explain our observations. Indices of intraventricular blood mass or volume, ventricular function, and the precise extent of CAD were not determined during the time of exercise testing. However, serial tests were completed in a 4-week period during which there were no overt changes in the patients’ clinical status, suggesting stable left ventricular function during the study period. Other clinical studies2gr30 and recent experimental data26 fail to support the hypothesis that the R wave response during exercise in normals and in CAD patients is different because of exercise-related alterations in intraventricular volumes. Measurement. Variability resulting from errors in measurements of R wave amplitude was minimized by independent data evaluation. All R wave amplitude responses were determined with the patients standing, and changes resulting from respiratory motion were minimized by averaging beats. Using body surface mapping techniques, Mirvis14 demonstrated that lead location influences the R wave amplitude response during exercise in normals. In our study, all electrodes were placed by the same technician, but the torsos of the subjects tested were not marked. Variations in lead placement should have influenced responses in both normals and CAD patients, and this was not the case. Heart rate. Heart rate achieved during exercise has been noted to be particularly important in the interpretation of the R wave response.1,2’ Studies in
American
Heart
1984 Journal
normal subjects have shown an increase in R wave amplitude at lower heart rates, and a decrease at higher rates.‘, ‘j. “I Because of these observations, it has been suggested that diagnostic changes in R wave amplitude should be assessed only at heart rates of 120 bpm or greater.’ Furthermore, it has been postulated that the differences in R wave responses in normals and in CAD patients may be a result of lower heart rates usually achieved by CAD patients.“’ In the seven CAD patients with variable R wave responses, 30 of 33 exercise tests resulted in maximal heart rates greater than 120 bpm (Table II). CAD and R wave response. The mechanism of exercise-induced R wave amplitude changes in patients with CAD is unknown. Abdulmassih et a1.3’ have noted the variable results of studies examining the sensitivity and specificity of R wave amplitude changes during exercise to detect CAD. These authors suggest that changes in ejection fraction and ventricular volumes may independently affect the R wave response during exercise and account for the different results obtained in clinical studies. However, the work of Battler et a1.Z5 and David et al.?” provide important data which do not support volume changes during exercise as the mechanism of R wave amplitude changes. Ellestad3” has suggested that the increase in R wave amplitude which accompanies infarction, spasm, exercise, or injection of contrast material into coronary arteries is the result of ischemia-related changes in electrical properties of the ventricle. This suggestion is supported by experimental studies in dogs which have shown that R wave changes during ischemia are related to intramyocardial conduction delays.lg It seems unlikely that alterations in either the electrical or mechanical properties of the left ventricle would explain the variable results in studies using R wave amplitude changes to detect CAD. As previously suggested, both mechanical and electrical alterations may occur during exercise and may result in variable R wave response.17 The results of our study would indicate that either mechanical or electrical properties which result in R wave changes during exercise must vary within individual patients with CAD. If R wave changes are related to the extent of myocardial ischemia,3” one would have to postulate variable degrees of myocardial ischemia during serial exercise tests. We did not try to quantitate the extent of myocardial ischemia during our study, but all CAD patients with variable R wave responses had reproducible ischemic ST segment changes during each exercise test (Table II). The mechanism(s) of R wave amplitude changes
Volume Number
108 4, Part 1
during exercise and an explanation of the directional variability observed in these CAD patients requires further investigation. Conclusions. In this study, the direction of R wave amplitude response was inconsistent in 7 of 10 CAD patients. These seven patients had reproducible ST segment changes indicating ischemia during symptom-limited serial exercise testing. The variability of R wave amplitude response in these patients did not appear to be a function of the heart rate, duration, or workload of exercise. The mechanism of the R wave response to exercise remains unknown. Because of the directional variability in R wave amplitude response to exercise demonstrated in this study, we believe that interpretation of the R wave response in serial exercise testing should be done with caution, whether to determine efficacy of therapy used to reduce ischemia,20 or as a measure of ventricular dysfunction during exercise.34 Since all CAD patients had at least one exercise test which resulted in a decrease in R wave’amplitude, this study supports the conclusion of others,21~2s’2gthat the routine determination of the R wave response to exercise testing appears to have little diagnostic value. The authors gratefully acknowledge the technical support of Ms. Mona Berg and the secretarial support of Ms. Carolyn Johnson, Ms. Loretta McSweeney, Ms. Ruth Alexander Witt, and Ms. Linda Bernhardt. We also thank Ernest R. Greene, Ph.D., and Gregory S. Uhl, M.D., for reviewing this manuscript. REFERENCES
4.
5.
6.
7. 8.
9.
10.
Ellestad M: Stress testing, Principles and practice. Philadelphia, 1980, F.A. Davis Co, pp. 97, 193, and 303. Fortuin NJ, Weiss JL: Exercise stress testing. Circulation 56:699, 1977. Edwards H, Katzeff IE: Relation between S-T segment depression and S wave amplitude (Letter to editor). Am J Cardiol 36:270, 1975. Bonoris PE, Greenberg PS, Christison GW, Castellanet MJ, Ellestad MH: Evaluation of R-wave amplitude changes versus ST segment depression in stress testing. Am J Cardiol 57:904, 1977. Greenberg PS, Friscia DA, Ellestad MH: Predictive accurascy of Q-X/Q-T ratio. Q-Tc interval. St dearession and R-wave amplitude during’stiess testing. Am J Cardiol 44:18, 1979. Simonson E: Effect of moderate exercise on the electrocardiograph in healthy young and middle-aged men. J Appl Physiol 5:584, 1953. Lloyd-Thomas HG: The effect of exercise on the electrocardiogram in healthy subjects. Br Heart J 23:260, 1961. Bruce RA, Mazzareha JA, Jordan JW, Green E: Quantitation of QRS and ST segment responses to exercise. AM HEART J 7 1:455, 1966. Bonoris PE, Greenberg PS, Castellanet MJ, Ellestad MH: Significance of changes in R-wave amplitude during treadmill stress testing: Angiographic correlation. Am J Cardiol41:846, 1978. Christison GW, Bonoris PE, Greenberg PS, Castellanet MJ, Ellestad MH: Predicting coronary artery disease with treadmill stress testing: Changes in R-wave amplitude compared with ST segment depression. J Electrocardiol 12:179, 197!).
Exercise
R wave response
in CAD
987
11. Baron DW, Ilsley C, Sheiban I, Poole-Wilson PA, Rickards AF: R wave amplitude during exercise-relation to left ventricular function and coronary artery disease. Br Heart J 44~512, 12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
1980.
Yiannikas J, Marcomichelakis J, Taggart P, Kelly BH, Emanuel R: Analvsis of exercise-induced changes in R-wave amplitude in asymptomatic men with electrocardiographic ST-T changes at rest. Am J Cardiol 47:2X$ 1981. Uhl GS, Hopkirk JAC: Analysis of exercise-induced R-wave amplitude changes in detection of coronary artery disease in asymptomatic men with left bundle branch block. Am .J Cardiol 44:1247, 1979. Mirvis DM: Body surface distribution of exercise-induced QRS changes in normal subjects. Am -7 Cardiol 46:988, 1980. Ruffy R, Lovelace DE, Mueller TM, Knoebell SB, Zipes DP: Relationship between changes in left ventricular bipolar electrograms and regional myocardial blood flow during acute coronary occlusion &r the dog. Circ Res 45:764, 1979.Holland RP. Brooks H: The QRS comnlex durine mvocardial ischemia. J &in Invest 57:541, 1976. a Simoons ML: QRS changes in coronary artery disease (Letter to Editor). Circulation 59:841, 1979. Brody DA: A theoretical analysis of intracavitary blood mass influence on the heart-lead relationship. Circ Res 4:731, 1956. David D, Naito M, Michelson E, Watanabe E’, Chen CC, Morganroth J, Shaffenburg M, Blenko T: Intramyocardial conduction: A major determinant of R wave amplitude during acute myocardial ischemia. Circulation 65:161, 1982. Lerman J, Mele E, Chiozza M, Svetlize H, Perosio AM: Effects of nitrates on R-wave variations after exercise in coronary heart disease. Chest 80:137, 1981. Wolthuis RA, Froelicher VG, Hopkirk A, Fischer JR, Keiser N: Normal electrocardiographic waveform characteristics during treadmill exercise testing. Circulation 60:1028, 1979. Fabian J, Stolz I, Janota M, Rohac J: Reproducibility of exercise tests in patients with symptomatic ischaemic heart disease. Br Heart J 37:785, 1975. Lassvik C: Reproducibility of work performance at serial exercises in patients with angina pectoris. Stand J Clin Lab Invest 38~747, 1978. Fox KM, Selwyn AP, Shillingford JP: A method for precordial surface mapping of the exercise electrocardiogram. Br Heart J 40:1339, 1978. Battler A, Froelicher VF, Gallagher KP, Kumada T, McKown D. , Kemuer. WS. Ross J: Effects of changes in ventricular size on regional and surface QRS amplitudesin the conscious dog. Circulation 62:174, 1980. David D, Naito M, Chen CC, Michelson EL, Morganroth J, Schatfenburg M: R-wave amplitude variations during acute experimental myocardial ischemia: An inadequate index for changes in intracardiac volume. Circulation 63:1364, 1981. Lekven J, Chatterjee K, Tyberg JV, Parmley WW: Reduction in ventricular endocardial and epicardial potentials during acute increments in left ventricular dimensions. AM HEART J 98:200, 1979. Wagner S, Cohn K, Selzer A: Unreliability of exerciseinduced R-wave changes as indexes of coronary artery disease. Am J Cardiol 44:1241, 1979. Battler A, Froelicher V, Slutskv R, Ashburn W: Relationship of QRS amplitude changes during exercise to left ventricular function and volumes and the diagnosis of coronary artery disease. Circulation 60:1004, 1979. Ishikawa K, Berson AS, Pipberger HV: Electrocardiographic changes due to cardiac enlargement. AM HEAKT J 81:635, 1971. Simoons ML, Block P: Toward the optimal lead system and optimal criteria for exercise electrocardiography. Am J Cardiol 47:1366, 1981. Abdulmassih S, Iskandrian AS, Hakk AH, Horowitz L, Mintz r.
i
October,
Voyles, Smith and Abrams
American
GJ, Anderson GJ, Kane S, Segal BL: Changes m R wave during exercise: Correlation with left ventricular function and volumes. J Electrocardiol l&199, 1982. 33. Ellestad HM: The mechanism of exercise-induced R-wave amplitude changes in coronary heart disease. Arch Intern Med 142:963. 1982.
Heart
1984 Journal
34. Brown JB, Wynne J, Holman BL, Atkins HL, Cohn PI? Relationship between regional ejection fraction during exercise and ST and R wave changes on the ECG (abstr). J Am Co11 Cardiol 1:649, 1983.
Evaluation of habitual excessive alcohol consumption on myocardial infarction risk in coronary disease patients Overt coronary heart disease does occur at times in a setting of alcoholism. In an attempt to test the hypothesis that habitual excessive drinking may have an aggravating effect upon coexisting ischemic heart disease and may help precipitate new coronary events, we compared myocardial infarct prevalence among heavy drinkers and non-heavy drinkers with angiographically documented coronary artery disease. Infarct prevalence was found to be higher for heavy drinkers than for non-heavy drinkers under age 60 years, after controlling for differences in smoking habits and underlying atherosclerosis severity. A reversal in trend which may be due to the operation of selective factors causing premature coronary death among alcoholics was observed for individuals above age 60 years. These results, although open to differing interpretations, are consistent with the notion that heavy drinking has a destabilizing effect upon preexisting ischemic heart disease and may increase acute coronary event risk. (AM HEART J 108:988,
1984.)
Sigismond Deutscher, M.D., M.P.H., Howard E. Rockette, Venkatraman Krishnaswami, M.D. Pittsburgh, Pa.
While recent epidemiologic studies,‘-4 point to the existence of a negative correlation between ischemic heart disease and mild-to-moderate consumption of alcohol, little is known about the relationship of the former with heavy drinking. More particularly, the question arises whether habitual excessive drinking might not aggravate the manifestations of preexisting coronary disease and even increase infarct risk through the various changes it is known to impart upon the cardiovascular system.s If the answer to this question is affirmative, one might expect heavy drinkers to develop severe ischemic manifestations From the Veterans Administration Medical Center and the Department Medicine, University of Pittsburgh School of Medicine; and from Department of Biostatistics, Graduate‘ School of Public Health. Received accepted
for publication March 14, 1984.
Reprint requests: Sigismond cal Center, University Drive
988
Oct.
1, 1983;
revision
received
Deutscher, M.D., Medical C, Pittsburgh, PA 15240.
March Service,
of the
5, 1984; VA Medi-
Ph.D., and
and incur new acute coronary events more frequently than non-heavy drinkers with comparable coronary artery involvement. Taken one step further, this line of thought implies that heavy drinkers with relatively little coronary artery involvement who would otherwise have had only mild symptoms, might instead present with more severe disease manifestations. In an attempt to test the above ideas, we assessed the drinking habits of patients who underwent coronary angiography because they had developed severe ischemic manifestations; heavy drinkers were then compared to non-heavy drinkers in regard to the severity of their underlying atherosclerotic disease and, given a similar degree of atherosclerosis severity, to the frequency with which they had sustained a prior infarction. Since several report@+’ indicate that cigarette smoking and hypertension may be associated with both heavy drinking and