- Email: [email protected]
Comparison of dipyridamole-handgrip test and bicycle exercise test for thallium tomographic imaging
Comparisonof Dipyridamole-Handgrip Test and Bicycle ExerciseTest for ThalliumTomographicImaging HEIKKI V. HUIKURI, MD, ULLA R. KORHONEN, MD, K.E. JUHANI AIRAKSINEN, MD, MARKKU J. IKiiHEIMO, MD, JUHANI HEIKKlti, MSC, and JUHA T. TAKKUNEN, MD
Seventy-three patients with angina pectoris and 20 with atypical chest pain, who underwent coronary angiography, were examined by single-photon emission computed thallium tomography (li-SPECT) using a combined dipyridamoie-handgrip stress test. Perlusion defects were detected in 78 of 81 patients with angiographicaiiy significant coronary artery disease (CAD) (sensitivity 98 % ). in 9 of 12 patients without CAD, the thallium images were normal (specificity 75 % ). Thirty-five patients with CAD were reexamined by Ti-SPECT using a dynamk bicycle exercise stress test. The sensitivity of the dipyridamoie-handgrip test did not differ from the bicycle exercise test in diagnosing the CAD (97% vs
94 % ). Multiple thallium defects were seen in 19 of 22 (86%) patients with muitivessei CAD by the dipyridamoie-handgrip test but only in 14 of 22 (64 % ) by the bicycle exercise test. Noncardiac side-effects occurred in 17 of 93 (18 % ) patients after dipytidamoie infusion. Cardiac symptoms were less common during the dipyridamoie-handgrip test than during the bicycle exercise (15% vs 76%, p
M
tional dynamic exercise test in the evaluation of patients with CAD.
vocardial oerfusion scintiarauhv using thallium201 in conjunciion with sympt6m:limited eiercise is a useful technique in the noninvasive assessmentof coronary artery disease (CAD).*v2However, poor exercise performance caused, for example, by peripheral vascular disease, left ventricular dysfunction and ,& blocking therapy, may reduce the sensitivity of the test3 Myocardial thallium imaging with pharmacologic coronary vasodilation using dipyridamole is an alternative stress method.4-6 Combination of isometric handgrip exercise with intravenous dipyridamole produces further increase in coronary blood flow and gives an additional advantage in conjunction with perfusion imaging.7 We evaluated single-photon emission computed thallium tomography (Tl-SPECT) using this new stress method and compared it with the conven-
Methods Patients: Ninety-three patients undergoing coronary angiography either because of disabling angina pectoris (73) or atypical chest pain (20) were examined by Tl-SPECT after a combined intravenous dipyridamole and isometric handgrip test and after a d-hour rest. Thirty-five of the 81 patients with CAD were reexamined by Tl-SPECT using symptom-limited bicycle exercise test before either coronary artery bypass grafting or coronary angioplasty. The clinical and coronary angiographic data of the patients are listed in Table I. Informed consent was obtained from all patients. Dipyridamole handgrip test: Dipyridamole was infused through a peripheral venous cannula at a rate of 0.14 mg/kg/min for 4 minutes. One minute after the end of dipyridamole infusion, the patients performed a handgrip test at a force of 30% of mean maximal voluntary contraction for 4 minutes.* The thallium injection was given 3 minutes after the beginning of the handgrip test, after which the squeezing was continued for at least 1 minute. If severe symptoms developed or if blood pressure decreased by >20 mm Hg, aminophylline 100 to 200 mg, was administered, and in
From the Division of Cardiology, Department of Medicine, Oulu University Central Hospital, Oulu, Finland. This work was supported by a grant from the Paavo Nurmi Foundation for Cardiovascular Disease,Helsinki, Finland. Manuscript received April 10, 1987;revised manuscript received September 14, 1987,and accepted September 16. Address for reprints: Heikki V. Huikuri, MD, Division of Cardiology, Department of Medicine, Oulu University Central Hospital, 90 220 Oulu, Finland. 264
February 1, 1988
cases of angina1 chest pain sublingual nitroglycerin was given. Bicycle exercise test: An upright symptom-limited exercise test was performed on an electrically braked bicycle ergometer. The exercise was started at 30 watts and the workload was increased in l-minute steps by 15 watts for men and 10 watts for women. Standard electrocardiographic leads I, aVF and Vg were recorded, and cuff blood pressure was monitored.g The exercise was continued until symptoms of angina, dyspnea or fatigue occurred, or if >0.2 mV ST-segment depression or decrease of blood pressure by >20 mm Hg developed. Coronary angiography: Patients underwent leftsided cardiac catheterization with selective coronary angiography in multiple projections, including caudal and cranial views, performed by the Judkins technique. A coronary luminal diameter reduction of >70% was considered significant. Interpretation of coronary arteriograms was performed by 2 experienced observers blinded to the results of perfusion imaging. Concordance between the 2 observers in interpretation of individual stenoses being significant was 90%. Consensus was obtained by discussion. Thallium tomographic imaging: TI-SPECT using a combined dipyridamole-handgrip test was performed 1 day before coronary angiography, and a bicycle exercise test was done within 4 months afterwards. No changes in medication or severity of symptoms occurred between separate imaging periods. Thallium201 (2.5 mCi) was injected 30 to 60 seconds before completion of the exercise tests. Imaging was started within 5 minutes and was repeated after about 4 hours of rest. Data acquisition was carried out using the Siemens RotaTMZLC 75 gamma camera with a low-energy, all-purpose collimator. Thirty images every 6.0' were acquired for 35 seconds each into a matrix of 64 X 64 elements. Total acquisition time was 20 minutes. Data collection began from the left posterior oblique view and ended at the right anterior oblique view. The transaxial sections were reconstructed by filtered backprojection using the Gamma-11 computer system. No attenuation correction was made. Each reconstructed slice was l&mm thick. Coronal and sagittal tomograms were reorganized from the set of transaxial ‘tomograms, corresponding to transverse and longitudinal sections of the cardiac axis. Tomographic slices were stored on x-ray film for later analysis. The images from each patient were examined by 3 different observers without knowledge of the clinical and angiographic information. The observers interpreted the dipyridamole-handgrip studies separately from the bicycle exercise studies without knowing the type of stress test used. Five discrete regions of the left ventricular myocardium-anterior, septal, apical, inferior and posterolateral-were defined. The anterior or anteroseptal wall was assumed to represent the distribution of the left anterior descending coronary artery; the posterolateral wall the circumflex distribution and the inferior and inferoapical walls the distribution of the right coronary artery. The regional perfusion ‘was considered abnormal if a defect was visually pre-
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 81
265
TABLE I Clinical and Angiographlc Data
All Patients (n = 93) Age W Sex (males/females) Symptoms Atypical chest pain Angina* Grade 2 Grade 3 Grade 4 Previous Ml Non-Q-wave D-Wave Anterior Inferior Medication b-blocking agents Calcium antagonists Nitrates Coronary angiography Normal l-vessel disease P-vessel disease 3-vessel disease l New York Heart Association MI = myocardial infarction.
52f 74119
7
Patients studied by Two thallium stress tests (n = 35) 52% 8 3114
20 73 12 54 7
1 34 5 26 3
8 25 9 16
2 12 4 8
78 34 78
30 20 28
12 22 41 18
0 13 15 7
classification
of grade of angina.
sent in at least 2 tomographic sections with or without redistribution. If the interpretation of the scintigrams was not unanimous, the interpretation of the 2 observers in agreement was used. Complete agreement between the 3 observers concerning the presence or absence of CAD occurred in 80 of 93 (86%) patient analyses. The scintigraphic region of the left anterior descending coronary artery was uniformly interpreted in 90% of the patients, that of the left circumflex artery in 78% and that of the right coronary artery in 84%. Statistics: Chi-square or Fisher’s exact test was used to determine differences between proportions. The paired t test was used in determining the changes in hemodynamic data during the exercise tests.
Results Relation of TI-SPECT using the dipyridamolehandgrip test to the coronary angiographic data: Of 93 patients, 81 had angiographically significant CAD; 10 patients had normal coronary angiography and 2 had <70% luminal diameter reduction in 1 coronary artery. Thallium defects were detected in 150 vascular regions after dipyridamole-handgrip exercise. Twenty-eight were fixed defects and 122 were partly or totally reversible on delayed images. TI-SPECT using the dipyridamole handgrip test was 78/81(96%) sensitive and 9/X! (75%) specific in detecting angiographitally significant CAD. If 150% luminal reductions were considered significant, the specificity would have been 92% (11/12). The sensitivity, specificity and predictive accuracy of this method of identifying corresponding coronary artery involvement are summa-
DIPYRIDAMOLE-HANDGRIPTHALLIUM TOMOGRAPHY
TABLE II Accuracy of Thallium Tomographic lmaglng Using Two Different Relation to Locatlon of Individual Coronary Arterial Steno& Coronary stenosis (270%)
Stress Methods In
LAD
LC
RCA
Thallium defects/stenosed vessels Sensitivity Dipyridamole-handgrlp Dipyrldamole-handgrip Bicycle exercise test Specificity Dipyridamole-handgrip Dipyridamole-handgrip Bicycle exercise test Predlctlve accuracy Dlpyridamole-handgrip Dlpyrldamole-handgrip Bicycle exercise test
test test
(n = 93) (n = 35)’ (n = 35)’
55/58 (95%) 24/24 (100%) 22/24 (92%)
32/48 (67%) 11/16(69%) 8/16 (50%)
42148 (88 %) 20/22 (91%) 17122 (77%)
test test
(n = 93) (n = 35)’ (n = 35)’
28135 (74%) 9/11 (82%) 9/11 (82%)
39/45 (67 % ) 16119 (64%) 18/19 (95%)
39145 (87%) 10113 (77%) 10113 (77%)
test test
(n = 93) (n = 35)’ (n = 35)’
81193 (87%) 33/35 (94%) 31/35 (69%)
71/93 (76%) 27/35 (77%) 26/35 (74%)
El/93 (87%) 30/35 (66%) 27135 (77%)
Patients studied by thallium imaging using both dtpyridamole-handgrlp test and bicycle exercise test. LAD = left anterior descending coronary artery: LC = left circumflex coronary artery: RCA = right coronary artery. l
rized in Table II. A stenosed left circumflex coronary artery was less frequently detected by TI-SPECT than stenosedleft anterior descending or right coronary arteries (p
corresponding coronary artery involvement. Thallium perfusion defects in 12 vascular regions were seen in 19 of 22 (86(r0)patients with multivessel disease using the dipyridamole-handgrip test and in 14 of 22 (649’01 patients using the bicycle exercise test. Dipyridamolehandgrip exercise detected multivessel CAD in 5 patients in whom the bicycle exercise showed normal thallium perfusion or only single perfusion defect (Fig ures 1 and 2). None of these 5 patients reached 70% of their age-predicted maximal heart rate during dynamic exercise testing. All 10 patients with multivessel CAD who reached 70% of their age-predicted maximal heart rate also had multiple thallium defects after the bicycle exercise. The hemodynamic data and the side-effects of the stress tests are presented in Table IV. The heart rate and systolic blood pressure increased significantly during both stress tests, but the diastolic blood presOIPYRIDAMOLE
FIGURE 1. A thalllum tomographic Image (coronal sectlon) of a patlent wlth P-vessel dlsease In whom the dlpyrldamole-handgrlp test revealed reversible anteroseptal perluslon defect, which extends partly to the lnferlor region. The bicycle exercise test showed falsely normal anteroseptal perfuslon.
HANDGRIP
BICYCLE
EXERCISE
FIGURE 2. A thalllum tomographic Image (saglttal sectlon) of a patlent wlth 3-vessel dlsease In whom the dlpyrldamole-handgrlp test revealed reversible lnferoaplcal perfuslon defect, which extends to the anterlor reglon. The bicycle exercise showed falsely normal lnferoaplcal perfuslon.
February 1, 1988
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 61
287
TABLE III Assessment of Extent of Coronary Artery Disease by Thalllum Tomographic lmaglng Uslng Two Different Stress Methods No. of Coronary Arteries Anglographic Normal thallium scan Dipyridamole-handgrip Dipyrldamole-handgrip Bicycle exercise test One abnormal thallium vascular region Dipyrldamole-handgrip Dipyridamole-handgrip Bicycle exercise test 2 to 3 abnormal thallium vascular regions Dipyridamole-handgrip Dipyridamole-handgrip Bicycle exercise test l
Narrowed
1
findings
l/l4
>70%
in Diameter
2
3
(7%) 0 0
0 0 0
test test
(n = 93) (n = 35)’ (n = 35)’
z/22 (9%) l/13 (8%) Z/13 (15%)
test test
(n = 93) (n = 35)’ (n = 35)’
16122 (73%) 9/13 (69%) 9/13 (69%)
11141 (27%) 3/15 (20%) 5/15 (33%)
(6%) 0 317 (43%)
test test
(n = 93) (n = 35)’ (n = 35)’
4122 (18%) 3/13 (23%) Z/l3 (15%)
29/41(71%) 12115 (80%) 10115 (67%)
17/18 (94%) 7/7 (100%) 417 (47%)
Patients studied by thallium imaging ustng both dipyrldamole-handgrip
test and bicycle exercise
l/l8
test.
nized to be related to the sensitivity and specificity of the noninvasive tests.13J4Most of the patients in this study were middle-aged men with typical angina, perhaps making the noninvasive evaluation of the presence of CAD unnecessary for clinical purposes in these patients. The sensitivity was best in detecting stenosed left anterior descending coronary arteries and right coronary arteries. The ability to diagnose the stenosesof left circumflex arteries was poorer, however, and the specificity in the left anterior descending arterial region was not very high. Many factors may contribute to divergence of scintigraphic and angiographic findings. Myocardial scintigraphy reflects relative myocardial perfusion and interpretation of images is subjective.15-18The posterolateral segment is the only region that correlates to the angiographic diseaseof the left circumflex artery.lg The lower accuracy in detecting perfusion abnormalities in this region may be due to the great distance of this rather small myocardial segment from the chest wall, even if tomographic imaging improves the accuracy compared Discussion with planar imaging.20s21 Difficulties in the angiographic estimation of the Thallium perfusion studies using dynamic exercise or dipyridamole have indicated that CAD can be severity of coronary stenosesalong with other factors, detected by both techniques with equal accuracy.lOJ1 such as the length of the lesion, the effects of stenoses Intravenous dipyridamole combined with isometric in series,the influence of collateral blood flow and the handgrip results in a greater increase in coronary flow variability in dominance and regional distribution of than either intervention alone.7 Further, isometric the arteries, may also result in discrepancies between handgrip has been found to constrict the existing ste- angiography and scintigraphy.15-l*Despite these liminoses of epicardial coronary arteries, giving an addi- tations, the present data concur with previous studies tional advantage in the diagnosis of CAD with perfu- and show that stress Tl-SPECT imaging provides a sion imaging.12To our knowledge, the clinical useful- reasonably high accuracy in the identification of indiness of the combined stress method has not been vidual vessel involvement. largely studied up to now. In our previous pilot study it The overall sensitivity in CAD detection was simiappeared to be useful in the diagnosis of CAD in pa- lar by Tl-SPECT using either the bicycle exercise test or the dipyridamole handgrip test, but the latter methtients with aortic valve stenosis.* The present data showed that the combined dipy- od tended to be more sensitive in diagnosing multivesridamole-handgrip test is a sensitive method of diag- se1CAD. This finding may be explained by the fact nosing CAD. The high sensitivity is partly explained that the sensitivity of conventional exercise scintigraby the high pretest probability of CAD, which is recog- phy is reduced at low levels of stress3 More than half
sure increased only during the dipyridamole-handgrip test. Some noncardiac side-effects occurred in 17 patients (18%) during the dipyridamole-handgrip test and they were reversed by aminophylline in 4 patients. Angina pectoris developed in 11 patients (12%) after dipyridamole infusion. Four of these patients received aminophylline and 3 were treated with sublingual nitroglycerin. One patient with &vessel disease developed severe vasovagal reaction with hypotension and syncope, which was treated by atropin and aminophylline. In this patient the stresstest had to be stopped; all other patients completed the dipyridamole-handgrip test. The only noncardiac symptom was dizziness in 1 patient during the bicycle exercise test. Angina pectoris occurred more often during dynamic exercise than during the dipyridamole-handgrip test (p
268
DIPYRIDAMOLE-HANDGRIP
THALLIUM
TOMOGRAPHY
TABLE IV Hemodynamic Data and Symptoms During the Stress Tests Dipyridamole-handgrip (n = 93) Hemcdynamic data Heart rate (beatslmin) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Symptoms Noncardiac Nausea or vomiting Dizziness or hypotension Headache Cardiac Angina Dyspnea Total l
Exercise 88 f 19t 142 i 22’ 88 f 13’
Rest 68f 15 128 f 17 81 f 10
p
test
Bicycle exercise (n = 35) Rest 68f 10 126 f 16 8Oi 11
test
Exercise 110zk22t 155 f 307 82f 11
0
4 (4%) 8 (9%) 5 (5%)
1(3%) 0
11(12%) 3(3%) 31(33%)
18 (51%) 9 (26%) 28 (80%)
with data at rest.
of the patients stopped the bicycle exercise because of angina in this study without reaching the acceptable heart rate and most of the patients were receiving @blocker therapy, limiting their maximal exercise performance. Ischemia in the most severely compromised myocardial segment may terminate exercise before flow becomes inadequate in vesselswith less severe narrowing. Other factors, such as left ventricular dysfunction, lack of motivation, peripheral vascular disease or orthopedic problems, may also reduce the exercise performance. The increase in blood flow during dipyridamole-handgrip test is independent of such factors, perhaps explaining the greater accuracy in detection of multiple perfusion defects. The frequency and severity of side-effects did not prevent the completion of the stress test in most patients, Noncardiac side-effects occurred somewhat less often after dipyridamole infusion than previously reported.3~4*gJ0.22 Other investigators have used standing and walking after dipyridamole infusion in order to avoid symptoms related to hypotension. Isometric exercise produces an increase in both systolic and diastolic blood pressure, perhaps explaining the smaller frequency of side-effects. One patient had severe vasovagal reaction, emphasizing the need for alertness after dipyridamole infusion.z2
1. Beller GA, Watson DD, Gibson RS. Assessment of myocardial perfusion. In: Come PC, ed. Diagnostic Radiology. Philadelphia: J.S. Lippincott, 1985:125-
6. Homma S, Callahan RJ, Ameer B, McKusick KA. Strauss HW. Okada RD, Boucher CA. Usefulness of oral dipyridamole suspension for stress thallium imaging without exercise In the detection of coronary artery disease. Am J Cardiol 1988;57:503-508. 7. Brown BG, Josephson MA, Petersen RB. Pierce KD, Wong M, Hecht HS. Bolson E, Dodge HT. Intravenous dipyridamole combined with isometric handgrip for near maximal acute Increase In coronary flow in patients with coronary artery diseose. Am J Cardiol 1981;48:1077-1085. 6. Huikuri HV, Korhonen UR, Iktiheimo MJ. Heikkilg J, Takkunen JT. Detection of coronary artery disease by thallium imaging using a combined intravenous dipyridamole and isometric handgrip test in patients with aortic valve stenosis. Am J Card1011987;59:336-340. 9. Huikuri HV, Korhonen UR, Heikkilti J,Takkunen JT. Detectionofcoronary artery disease by thallium scintigraphy In patients with volvor heart disease. Br Heart J l986;56:148-151. 10. Josephson MA, Brown BG, Hecht AJ, Hopkins J, Pierce KD, Petersen RB. Noninvasive detection and localization of coronary stenosesin patients. Comparison of resting dipyridamole and exercise thallium-201 myocardial perfusion imaging. Am Heart J 1982;103:1008-1018. 11. Chamberlein BA. Comparison of dipyridamole and treadmill exercise for enhancing thaIIiumdOl perfusion defects In patients with coronary artery disease. Eur J Cardiol 1980;1:275-282. 12. Brown BG, Lee AB. Bolson EL, Dodge HT. Reflex constriction of significant coronary stenosis as a mechanism contributing to ischemic left ventricular dysfunction during isometric exercise. Circulation 1984;70:18-24. 13. Diamond GA. Monkey business. Am J Cardiol 1986;57:471-475. 14. Rozanski A, Diamond GA, Forrester JS,Berman DS, Morris D, Swan HJC. Alternative referent standards for cardiac normality: implication for diagnostic testing. Ann Intern Med 1984;101:164-171. 15. Massie BM, Botvinick EH, Brundage BH. Correlation of thaIIium-201 scintigrams with coronary anatomy: factors affecting region by region sensitivity. Am J Cardiol 1979;44:618-822. 16. McKillip JH. Murray RH, Turner JG, Bessent RG, Lorimer AR, Grelg WK. Can the extent of coronary artery disease be predicted from thallium-201 myocardial images? J Nucl Med 1979;20:715-719. 17. Gibson RS, Taylor GJ, Watson DD, Stebbins PT, Martin RP, Crampton RS, Beller GA. Predicting the extent and location of coronary artery disease during early postinfarction period by quantitative thalhum-201 scintigraphy. Am J Cardiol 1981;47:1010-1019. 18. Rigo P, Bailey IK, Griffith LSC. Pitt B, Burow RD, Wagner HV Jr, Becker LC. Value and limitations of segmental analysis of stress thallium myocardial Imaging for localization of coronary artery disease. Circulation 1980;61:973-
149.
981.
2. Ritchie JL. Trobaugh GB, Hamilton GW. Gould KL, Narahara KA, Murray IA, Williams DL. Myocardial imaging with thaIIium-201 at rest and during exercise. CircuIation 1977;56:66-71. 3. Massie BM, Wisneski, Kramer B. Comparison of myocardial thallium-201 clearance after maxim01 and submaximal exercise: implications for diagnosis of coronary disease: concise communication. J Nucl Med 1982;23:381-385. 4. Gould KL. Noninvasive assessment of coronary stenoses by myocardiol imaging during pharmacologic coronary vasodilation I. Physiologic basis and experimental validation. Am J Cordiol 1978;41:279-287. 5. Leppo J. Boucher CA, Okada RD. Newell JB. Strauss WH, Pohost GM. Serial thallium-201 myocardial Imaging after dipyridamole infusion: diagnostic utility in detecting coronary stenoses and relationship to regional ~011 motion. Circulation 1982X38:849-657.
19. Newman HN, Dunn RF, Harris PJ, Bantovich GJ, McLaughlin AF, Kelly DT. Differentiation between right and circumflex coronary artery disease on thallium myocardial perfusion scanning. Am J Card1011983:51:1052-1056. 20. Nohsra R. Kambara H, Suzuki Y, Tamaki S, Kadota K, Kawai C, Tamaka N, Torizuka K. Stress scintigraphy using single-photon emission computed
References
Seventy-three patients with angina pectoris and 20 with atypical chest pain, who underwent coronary angiography, were examined by single-photon emission computed thallium tomography (li-SPECT) using a combined dipyridamoie-handgrip stress test. Perlusion defects were detected in 78 of 81 patients with angiographicaiiy significant coronary artery disease (CAD) (sensitivity 98 % ). in 9 of 12 patients without CAD, the thallium images were normal (specificity 75 % ). Thirty-five patients with CAD were reexamined by Ti-SPECT using a dynamk bicycle exercise stress test. The sensitivity of the dipyridamoie-handgrip test did not differ from the bicycle exercise test in diagnosing the CAD (97% vs
94 % ). Multiple thallium defects were seen in 19 of 22 (86%) patients with muitivessei CAD by the dipyridamoie-handgrip test but only in 14 of 22 (64 % ) by the bicycle exercise test. Noncardiac side-effects occurred in 17 of 93 (18 % ) patients after dipytidamoie infusion. Cardiac symptoms were less common during the dipyridamoie-handgrip test than during the bicycle exercise (15% vs 76%, p
M
tional dynamic exercise test in the evaluation of patients with CAD.
vocardial oerfusion scintiarauhv using thallium201 in conjunciion with sympt6m:limited eiercise is a useful technique in the noninvasive assessmentof coronary artery disease (CAD).*v2However, poor exercise performance caused, for example, by peripheral vascular disease, left ventricular dysfunction and ,& blocking therapy, may reduce the sensitivity of the test3 Myocardial thallium imaging with pharmacologic coronary vasodilation using dipyridamole is an alternative stress method.4-6 Combination of isometric handgrip exercise with intravenous dipyridamole produces further increase in coronary blood flow and gives an additional advantage in conjunction with perfusion imaging.7 We evaluated single-photon emission computed thallium tomography (Tl-SPECT) using this new stress method and compared it with the conven-
Methods Patients: Ninety-three patients undergoing coronary angiography either because of disabling angina pectoris (73) or atypical chest pain (20) were examined by Tl-SPECT after a combined intravenous dipyridamole and isometric handgrip test and after a d-hour rest. Thirty-five of the 81 patients with CAD were reexamined by Tl-SPECT using symptom-limited bicycle exercise test before either coronary artery bypass grafting or coronary angioplasty. The clinical and coronary angiographic data of the patients are listed in Table I. Informed consent was obtained from all patients. Dipyridamole handgrip test: Dipyridamole was infused through a peripheral venous cannula at a rate of 0.14 mg/kg/min for 4 minutes. One minute after the end of dipyridamole infusion, the patients performed a handgrip test at a force of 30% of mean maximal voluntary contraction for 4 minutes.* The thallium injection was given 3 minutes after the beginning of the handgrip test, after which the squeezing was continued for at least 1 minute. If severe symptoms developed or if blood pressure decreased by >20 mm Hg, aminophylline 100 to 200 mg, was administered, and in
From the Division of Cardiology, Department of Medicine, Oulu University Central Hospital, Oulu, Finland. This work was supported by a grant from the Paavo Nurmi Foundation for Cardiovascular Disease,Helsinki, Finland. Manuscript received April 10, 1987;revised manuscript received September 14, 1987,and accepted September 16. Address for reprints: Heikki V. Huikuri, MD, Division of Cardiology, Department of Medicine, Oulu University Central Hospital, 90 220 Oulu, Finland. 264
February 1, 1988
cases of angina1 chest pain sublingual nitroglycerin was given. Bicycle exercise test: An upright symptom-limited exercise test was performed on an electrically braked bicycle ergometer. The exercise was started at 30 watts and the workload was increased in l-minute steps by 15 watts for men and 10 watts for women. Standard electrocardiographic leads I, aVF and Vg were recorded, and cuff blood pressure was monitored.g The exercise was continued until symptoms of angina, dyspnea or fatigue occurred, or if >0.2 mV ST-segment depression or decrease of blood pressure by >20 mm Hg developed. Coronary angiography: Patients underwent leftsided cardiac catheterization with selective coronary angiography in multiple projections, including caudal and cranial views, performed by the Judkins technique. A coronary luminal diameter reduction of >70% was considered significant. Interpretation of coronary arteriograms was performed by 2 experienced observers blinded to the results of perfusion imaging. Concordance between the 2 observers in interpretation of individual stenoses being significant was 90%. Consensus was obtained by discussion. Thallium tomographic imaging: TI-SPECT using a combined dipyridamole-handgrip test was performed 1 day before coronary angiography, and a bicycle exercise test was done within 4 months afterwards. No changes in medication or severity of symptoms occurred between separate imaging periods. Thallium201 (2.5 mCi) was injected 30 to 60 seconds before completion of the exercise tests. Imaging was started within 5 minutes and was repeated after about 4 hours of rest. Data acquisition was carried out using the Siemens RotaTMZLC 75 gamma camera with a low-energy, all-purpose collimator. Thirty images every 6.0' were acquired for 35 seconds each into a matrix of 64 X 64 elements. Total acquisition time was 20 minutes. Data collection began from the left posterior oblique view and ended at the right anterior oblique view. The transaxial sections were reconstructed by filtered backprojection using the Gamma-11 computer system. No attenuation correction was made. Each reconstructed slice was l&mm thick. Coronal and sagittal tomograms were reorganized from the set of transaxial ‘tomograms, corresponding to transverse and longitudinal sections of the cardiac axis. Tomographic slices were stored on x-ray film for later analysis. The images from each patient were examined by 3 different observers without knowledge of the clinical and angiographic information. The observers interpreted the dipyridamole-handgrip studies separately from the bicycle exercise studies without knowing the type of stress test used. Five discrete regions of the left ventricular myocardium-anterior, septal, apical, inferior and posterolateral-were defined. The anterior or anteroseptal wall was assumed to represent the distribution of the left anterior descending coronary artery; the posterolateral wall the circumflex distribution and the inferior and inferoapical walls the distribution of the right coronary artery. The regional perfusion ‘was considered abnormal if a defect was visually pre-
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 81
265
TABLE I Clinical and Angiographlc Data
All Patients (n = 93) Age W Sex (males/females) Symptoms Atypical chest pain Angina* Grade 2 Grade 3 Grade 4 Previous Ml Non-Q-wave D-Wave Anterior Inferior Medication b-blocking agents Calcium antagonists Nitrates Coronary angiography Normal l-vessel disease P-vessel disease 3-vessel disease l New York Heart Association MI = myocardial infarction.
52f 74119
7
Patients studied by Two thallium stress tests (n = 35) 52% 8 3114
20 73 12 54 7
1 34 5 26 3
8 25 9 16
2 12 4 8
78 34 78
30 20 28
12 22 41 18
0 13 15 7
classification
of grade of angina.
sent in at least 2 tomographic sections with or without redistribution. If the interpretation of the scintigrams was not unanimous, the interpretation of the 2 observers in agreement was used. Complete agreement between the 3 observers concerning the presence or absence of CAD occurred in 80 of 93 (86%) patient analyses. The scintigraphic region of the left anterior descending coronary artery was uniformly interpreted in 90% of the patients, that of the left circumflex artery in 78% and that of the right coronary artery in 84%. Statistics: Chi-square or Fisher’s exact test was used to determine differences between proportions. The paired t test was used in determining the changes in hemodynamic data during the exercise tests.
Results Relation of TI-SPECT using the dipyridamolehandgrip test to the coronary angiographic data: Of 93 patients, 81 had angiographically significant CAD; 10 patients had normal coronary angiography and 2 had <70% luminal diameter reduction in 1 coronary artery. Thallium defects were detected in 150 vascular regions after dipyridamole-handgrip exercise. Twenty-eight were fixed defects and 122 were partly or totally reversible on delayed images. TI-SPECT using the dipyridamole handgrip test was 78/81(96%) sensitive and 9/X! (75%) specific in detecting angiographitally significant CAD. If 150% luminal reductions were considered significant, the specificity would have been 92% (11/12). The sensitivity, specificity and predictive accuracy of this method of identifying corresponding coronary artery involvement are summa-
DIPYRIDAMOLE-HANDGRIPTHALLIUM TOMOGRAPHY
TABLE II Accuracy of Thallium Tomographic lmaglng Using Two Different Relation to Locatlon of Individual Coronary Arterial Steno& Coronary stenosis (270%)
Stress Methods In
LAD
LC
RCA
Thallium defects/stenosed vessels Sensitivity Dipyridamole-handgrlp Dipyrldamole-handgrip Bicycle exercise test Specificity Dipyridamole-handgrip Dipyridamole-handgrip Bicycle exercise test Predlctlve accuracy Dlpyridamole-handgrip Dlpyrldamole-handgrip Bicycle exercise test
test test
(n = 93) (n = 35)’ (n = 35)’
55/58 (95%) 24/24 (100%) 22/24 (92%)
32/48 (67%) 11/16(69%) 8/16 (50%)
42148 (88 %) 20/22 (91%) 17122 (77%)
test test
(n = 93) (n = 35)’ (n = 35)’
28135 (74%) 9/11 (82%) 9/11 (82%)
39/45 (67 % ) 16119 (64%) 18/19 (95%)
39145 (87%) 10113 (77%) 10113 (77%)
test test
(n = 93) (n = 35)’ (n = 35)’
81193 (87%) 33/35 (94%) 31/35 (69%)
71/93 (76%) 27/35 (77%) 26/35 (74%)
El/93 (87%) 30/35 (66%) 27135 (77%)
Patients studied by thallium imaging using both dtpyridamole-handgrlp test and bicycle exercise test. LAD = left anterior descending coronary artery: LC = left circumflex coronary artery: RCA = right coronary artery. l
rized in Table II. A stenosed left circumflex coronary artery was less frequently detected by TI-SPECT than stenosedleft anterior descending or right coronary arteries (p
corresponding coronary artery involvement. Thallium perfusion defects in 12 vascular regions were seen in 19 of 22 (86(r0)patients with multivessel disease using the dipyridamole-handgrip test and in 14 of 22 (649’01 patients using the bicycle exercise test. Dipyridamolehandgrip exercise detected multivessel CAD in 5 patients in whom the bicycle exercise showed normal thallium perfusion or only single perfusion defect (Fig ures 1 and 2). None of these 5 patients reached 70% of their age-predicted maximal heart rate during dynamic exercise testing. All 10 patients with multivessel CAD who reached 70% of their age-predicted maximal heart rate also had multiple thallium defects after the bicycle exercise. The hemodynamic data and the side-effects of the stress tests are presented in Table IV. The heart rate and systolic blood pressure increased significantly during both stress tests, but the diastolic blood presOIPYRIDAMOLE
FIGURE 1. A thalllum tomographic Image (coronal sectlon) of a patlent wlth P-vessel dlsease In whom the dlpyrldamole-handgrlp test revealed reversible anteroseptal perluslon defect, which extends partly to the lnferlor region. The bicycle exercise test showed falsely normal anteroseptal perfuslon.
HANDGRIP
BICYCLE
EXERCISE
FIGURE 2. A thalllum tomographic Image (saglttal sectlon) of a patlent wlth 3-vessel dlsease In whom the dlpyrldamole-handgrlp test revealed reversible lnferoaplcal perfuslon defect, which extends to the anterlor reglon. The bicycle exercise showed falsely normal lnferoaplcal perfuslon.
February 1, 1988
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 61
287
TABLE III Assessment of Extent of Coronary Artery Disease by Thalllum Tomographic lmaglng Uslng Two Different Stress Methods No. of Coronary Arteries Anglographic Normal thallium scan Dipyridamole-handgrip Dipyrldamole-handgrip Bicycle exercise test One abnormal thallium vascular region Dipyrldamole-handgrip Dipyridamole-handgrip Bicycle exercise test 2 to 3 abnormal thallium vascular regions Dipyridamole-handgrip Dipyridamole-handgrip Bicycle exercise test l
Narrowed
1
findings
l/l4
>70%
in Diameter
2
3
(7%) 0 0
0 0 0
test test
(n = 93) (n = 35)’ (n = 35)’
z/22 (9%) l/13 (8%) Z/13 (15%)
test test
(n = 93) (n = 35)’ (n = 35)’
16122 (73%) 9/13 (69%) 9/13 (69%)
11141 (27%) 3/15 (20%) 5/15 (33%)
(6%) 0 317 (43%)
test test
(n = 93) (n = 35)’ (n = 35)’
4122 (18%) 3/13 (23%) Z/l3 (15%)
29/41(71%) 12115 (80%) 10115 (67%)
17/18 (94%) 7/7 (100%) 417 (47%)
Patients studied by thallium imaging ustng both dipyrldamole-handgrip
test and bicycle exercise
l/l8
test.
nized to be related to the sensitivity and specificity of the noninvasive tests.13J4Most of the patients in this study were middle-aged men with typical angina, perhaps making the noninvasive evaluation of the presence of CAD unnecessary for clinical purposes in these patients. The sensitivity was best in detecting stenosed left anterior descending coronary arteries and right coronary arteries. The ability to diagnose the stenosesof left circumflex arteries was poorer, however, and the specificity in the left anterior descending arterial region was not very high. Many factors may contribute to divergence of scintigraphic and angiographic findings. Myocardial scintigraphy reflects relative myocardial perfusion and interpretation of images is subjective.15-18The posterolateral segment is the only region that correlates to the angiographic diseaseof the left circumflex artery.lg The lower accuracy in detecting perfusion abnormalities in this region may be due to the great distance of this rather small myocardial segment from the chest wall, even if tomographic imaging improves the accuracy compared Discussion with planar imaging.20s21 Difficulties in the angiographic estimation of the Thallium perfusion studies using dynamic exercise or dipyridamole have indicated that CAD can be severity of coronary stenosesalong with other factors, detected by both techniques with equal accuracy.lOJ1 such as the length of the lesion, the effects of stenoses Intravenous dipyridamole combined with isometric in series,the influence of collateral blood flow and the handgrip results in a greater increase in coronary flow variability in dominance and regional distribution of than either intervention alone.7 Further, isometric the arteries, may also result in discrepancies between handgrip has been found to constrict the existing ste- angiography and scintigraphy.15-l*Despite these liminoses of epicardial coronary arteries, giving an addi- tations, the present data concur with previous studies tional advantage in the diagnosis of CAD with perfu- and show that stress Tl-SPECT imaging provides a sion imaging.12To our knowledge, the clinical useful- reasonably high accuracy in the identification of indiness of the combined stress method has not been vidual vessel involvement. largely studied up to now. In our previous pilot study it The overall sensitivity in CAD detection was simiappeared to be useful in the diagnosis of CAD in pa- lar by Tl-SPECT using either the bicycle exercise test or the dipyridamole handgrip test, but the latter methtients with aortic valve stenosis.* The present data showed that the combined dipy- od tended to be more sensitive in diagnosing multivesridamole-handgrip test is a sensitive method of diag- se1CAD. This finding may be explained by the fact nosing CAD. The high sensitivity is partly explained that the sensitivity of conventional exercise scintigraby the high pretest probability of CAD, which is recog- phy is reduced at low levels of stress3 More than half
sure increased only during the dipyridamole-handgrip test. Some noncardiac side-effects occurred in 17 patients (18%) during the dipyridamole-handgrip test and they were reversed by aminophylline in 4 patients. Angina pectoris developed in 11 patients (12%) after dipyridamole infusion. Four of these patients received aminophylline and 3 were treated with sublingual nitroglycerin. One patient with &vessel disease developed severe vasovagal reaction with hypotension and syncope, which was treated by atropin and aminophylline. In this patient the stresstest had to be stopped; all other patients completed the dipyridamole-handgrip test. The only noncardiac symptom was dizziness in 1 patient during the bicycle exercise test. Angina pectoris occurred more often during dynamic exercise than during the dipyridamole-handgrip test (p
268
DIPYRIDAMOLE-HANDGRIP
THALLIUM
TOMOGRAPHY
TABLE IV Hemodynamic Data and Symptoms During the Stress Tests Dipyridamole-handgrip (n = 93) Hemcdynamic data Heart rate (beatslmin) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Symptoms Noncardiac Nausea or vomiting Dizziness or hypotension Headache Cardiac Angina Dyspnea Total l
Exercise 88 f 19t 142 i 22’ 88 f 13’
Rest 68f 15 128 f 17 81 f 10
p
test
Bicycle exercise (n = 35) Rest 68f 10 126 f 16 8Oi 11
test
Exercise 110zk22t 155 f 307 82f 11
0
4 (4%) 8 (9%) 5 (5%)
1(3%) 0
11(12%) 3(3%) 31(33%)
18 (51%) 9 (26%) 28 (80%)
with data at rest.
of the patients stopped the bicycle exercise because of angina in this study without reaching the acceptable heart rate and most of the patients were receiving @blocker therapy, limiting their maximal exercise performance. Ischemia in the most severely compromised myocardial segment may terminate exercise before flow becomes inadequate in vesselswith less severe narrowing. Other factors, such as left ventricular dysfunction, lack of motivation, peripheral vascular disease or orthopedic problems, may also reduce the exercise performance. The increase in blood flow during dipyridamole-handgrip test is independent of such factors, perhaps explaining the greater accuracy in detection of multiple perfusion defects. The frequency and severity of side-effects did not prevent the completion of the stress test in most patients, Noncardiac side-effects occurred somewhat less often after dipyridamole infusion than previously reported.3~4*gJ0.22 Other investigators have used standing and walking after dipyridamole infusion in order to avoid symptoms related to hypotension. Isometric exercise produces an increase in both systolic and diastolic blood pressure, perhaps explaining the smaller frequency of side-effects. One patient had severe vasovagal reaction, emphasizing the need for alertness after dipyridamole infusion.z2
1. Beller GA, Watson DD, Gibson RS. Assessment of myocardial perfusion. In: Come PC, ed. Diagnostic Radiology. Philadelphia: J.S. Lippincott, 1985:125-
6. Homma S, Callahan RJ, Ameer B, McKusick KA. Strauss HW. Okada RD, Boucher CA. Usefulness of oral dipyridamole suspension for stress thallium imaging without exercise In the detection of coronary artery disease. Am J Cardiol 1988;57:503-508. 7. Brown BG, Josephson MA, Petersen RB. Pierce KD, Wong M, Hecht HS. Bolson E, Dodge HT. Intravenous dipyridamole combined with isometric handgrip for near maximal acute Increase In coronary flow in patients with coronary artery diseose. Am J Cardiol 1981;48:1077-1085. 6. Huikuri HV, Korhonen UR, Iktiheimo MJ. Heikkilg J, Takkunen JT. Detection of coronary artery disease by thallium imaging using a combined intravenous dipyridamole and isometric handgrip test in patients with aortic valve stenosis. Am J Card1011987;59:336-340. 9. Huikuri HV, Korhonen UR, Heikkilti J,Takkunen JT. Detectionofcoronary artery disease by thallium scintigraphy In patients with volvor heart disease. Br Heart J l986;56:148-151. 10. Josephson MA, Brown BG, Hecht AJ, Hopkins J, Pierce KD, Petersen RB. Noninvasive detection and localization of coronary stenosesin patients. Comparison of resting dipyridamole and exercise thallium-201 myocardial perfusion imaging. Am Heart J 1982;103:1008-1018. 11. Chamberlein BA. Comparison of dipyridamole and treadmill exercise for enhancing thaIIiumdOl perfusion defects In patients with coronary artery disease. Eur J Cardiol 1980;1:275-282. 12. Brown BG, Lee AB. Bolson EL, Dodge HT. Reflex constriction of significant coronary stenosis as a mechanism contributing to ischemic left ventricular dysfunction during isometric exercise. Circulation 1984;70:18-24. 13. Diamond GA. Monkey business. Am J Cardiol 1986;57:471-475. 14. Rozanski A, Diamond GA, Forrester JS,Berman DS, Morris D, Swan HJC. Alternative referent standards for cardiac normality: implication for diagnostic testing. Ann Intern Med 1984;101:164-171. 15. Massie BM, Botvinick EH, Brundage BH. Correlation of thaIIium-201 scintigrams with coronary anatomy: factors affecting region by region sensitivity. Am J Cardiol 1979;44:618-822. 16. McKillip JH. Murray RH, Turner JG, Bessent RG, Lorimer AR, Grelg WK. Can the extent of coronary artery disease be predicted from thallium-201 myocardial images? J Nucl Med 1979;20:715-719. 17. Gibson RS, Taylor GJ, Watson DD, Stebbins PT, Martin RP, Crampton RS, Beller GA. Predicting the extent and location of coronary artery disease during early postinfarction period by quantitative thalhum-201 scintigraphy. Am J Cardiol 1981;47:1010-1019. 18. Rigo P, Bailey IK, Griffith LSC. Pitt B, Burow RD, Wagner HV Jr, Becker LC. Value and limitations of segmental analysis of stress thallium myocardial Imaging for localization of coronary artery disease. Circulation 1980;61:973-
149.
981.
2. Ritchie JL. Trobaugh GB, Hamilton GW. Gould KL, Narahara KA, Murray IA, Williams DL. Myocardial imaging with thaIIium-201 at rest and during exercise. CircuIation 1977;56:66-71. 3. Massie BM, Wisneski, Kramer B. Comparison of myocardial thallium-201 clearance after maxim01 and submaximal exercise: implications for diagnosis of coronary disease: concise communication. J Nucl Med 1982;23:381-385. 4. Gould KL. Noninvasive assessment of coronary stenoses by myocardiol imaging during pharmacologic coronary vasodilation I. Physiologic basis and experimental validation. Am J Cordiol 1978;41:279-287. 5. Leppo J. Boucher CA, Okada RD. Newell JB. Strauss WH, Pohost GM. Serial thallium-201 myocardial Imaging after dipyridamole infusion: diagnostic utility in detecting coronary stenoses and relationship to regional ~011 motion. Circulation 1982X38:849-657.
19. Newman HN, Dunn RF, Harris PJ, Bantovich GJ, McLaughlin AF, Kelly DT. Differentiation between right and circumflex coronary artery disease on thallium myocardial perfusion scanning. Am J Card1011983:51:1052-1056. 20. Nohsra R. Kambara H, Suzuki Y, Tamaki S, Kadota K, Kawai C, Tamaka N, Torizuka K. Stress scintigraphy using single-photon emission computed
References