SHORT COMMUNICATION Annals of Nuclear Medicine Vol.13, No.2, 121-125, 1999 Simultaneous assessment of function and perfusion during dipyridamole-handgrip Tc-99m sestamibi imaging in chronic coronary artery disease Teresa MASSARDO, Patricio GONZALEZ, Pamela HUMERES, Hernan CHAMORRO, Francisco AYALA, Jorge YOVANOVICH and Patricio AVENDANO Nuclear Medicine Center, University of Chile Clinical Hospital, Santiago, Chile The main goal of this work was to know the value of ventricular function in addition to perfusion Tc-99m sestamibi images in the assessment of coronary artery disease (CAD) when using dipyridamole (DIP) associated to isometric exercise. We analyzed 52 patients with suspected CAD; 40 of them had coronary lesions >= 50% and 12 patients without CAD, conforming study and control groups, respectively. Twenty-eight patients had prior myocardial infarction. A two-day sestamibi protocol was employed with i.v. DIP-handgrip and rest injections, acquiring ECG-gated first pass and planar perfusion images. Sensitivity for perfusion images was 85% and specificity was 91.7%. There was no change between rest and DIP ejection fraction (EF) in controls. CAD patients presented a significant EF decrease with DIP (p: 0.0015). Patients with ischemia in perfusion images had larger EF decrease (p: 0.0001). For the analysis, an EF drop >= 5% and any wall motion abnormality (WMA) were considered as having an abnormal response to DIP. CAD sensitivity improved significantly to 92.5% when adding EF drop and to 90% when adding WMA parameters, but specificity decreased to 75% with EF drop, and to 58.3% with WMA. In conclusion, first pass parameters from DIP-isometric exercise in addition to perfusion images are not a significant help in the assessment of CAD. Key words: dipyridamole, Tc-99m sestamibi, ventricular function, isometric exercise INTRODUCTION RADIONUCLIDE PERFUSION STUDIES are used widely for assessing coronary artery disease (CAD) and vasodilators such as dipyridamole (DIP) or adenosine may replace exercise with sensitivity and specificity around 70-95% for both Thallium-201 and Technetium-99m sestamibi. This latter also offers the alternative of measuring ventricular function simultaneously by the first pass technique and more recently by gated SPECT. Its value with Received August 17, 1998, revision accepted January 18, 1999. Supported by Grant 0972/90 Fondecyt Chile. For reprint contact: Teresa Massardo, M.D., Hospital Clinico U. de Chile, Santos Dumont 999-1E, Independencia, Santiago, CHILE. Area Code: 6531063 E-mail: tmassardo@ns.hospital.uchile.cl tmassard@entelchile.net exercise stress has been well demonstrated1 but combined with DIP or adenosine is controversial.2-5 The goal of this report was to evaluate the additive value of sestamibi functional parameters (EF and regional wall motion) over perfusion images when using DIP in association with handgrip in CAD patients. The use of pharmacological stress, especially vasodilators is very common in clinical practice and some groups, including echocardiographic ones, have combined it with isometric exercise in order to improve ischemia detection.6 We wanted to include isometric exercise in our protocol because it could increase its sensitivity, and also diminish collateral effects due to the increase in sympathetic tone; even though it has been demonstrated that left ventricular function changes with isometric handgrip do not allow sufficient clear discrimination between normal and CAD subjects.7 The value of DIP first pass angiography appears also as a good predictor of future events in CAD patients.8 MATERIALS AND METHODS We prospectively studied 52 patients with suspected CAD (37 males and 15 females, age 58 +- 9 years old, ranging from 39 to 76) who submitted to contrast coronary angiography as a screening procedure, or for CAD evaluation (12 with stable angina, one with coronary bypass grafting and 28 with prior myocardial infarction); 40 of them had significant coronary stenosis of >= 50%, 14 with single, 16 with double, and 10 patients with triple-vessel disease. Twelve other patients (8/12 females) without significant angiographic lesions (< 35%) were the control group; 5 were hypertensive, one had right ventricle arrhythmogenic dysplasia, one had mild aortic stenosis, but the other 5 did not have any demonstrable disease detected with other methods. Their age was 55 +- 8 y.o. (p: ns from the CAD group). Beta-blockers and xantines were withdrawn 24 hours prior to the radionuclide study, performed within 3 months of contrast angiography ( 16 +- 18 days). Sestamibi Protocol: First day: DIP was administered and immediately a first pass angiography was acquired. One hour later a perfusion image set was done. Second day: the same study was performed at rest. Quality control of Tc-99m sestamibi demonstrated labeling efficiency of 97.8 +- 1.3%. Dipyridamole Test: A standard 4 minute DIP continuous infusion was used (0.56 mg/kg). At minute eight, 925 MBq of Tc-99m MIBI was injected as a bolus in an external jugular or right antecubital vein. Isometric Exercise with a handgrip was performed during minutes 6-10 of DIP infusion (approximately at 25%-50% of maximum). The model was a Harpenden British Indicators Ltd. AM006-003. Aminophylline was administered in some cases, after the radionuclide injection. First pass angiography.' ECG-gated list mode acquisition in RAO 30 degrees, were acquired with a Digital PDP 11-34 computer, and a 32 x 32 matrix. The first 20 studies were acquired with a high resolution collimator; and for the other 32 we had available a high sensitivity one. For processing we included in the analysis only those studies with good quality bolus, at least 5 similar beats and enough statistical information (2.000 counts at end diastole), and the computer generated a representative cycle. The studies were read by one observer blinded to the patient's condition information. Segmental wall motion was analyzed visually. Manual end diastolic and end systolic after lung background correction regions of interest were applied for EF measurement. Linear correlation between rest sestamibi EF and contrast EF was 0.52, but the slopes were significantly different (p < 0.0001 ). Perfilsion images; All patients received a caloric snack 20 minutes after radionuclide injection, intending to diminish gallbladder activity. Three view gated images with 10 frames were recorded in a 64 x 64 matrix selecting the best lateral anterior oblique, anterior or right anterior oblique and left lateral views. Summed original, background subtracted diastolic and systolic from gated images were analyzed by 2 blinded observers (including summed radial profiles comparing rest with DIP). Five myocardial segments were compared with contrast angiography : antero-lateral, septal, inferior, postero-lateral and infero-apical. All perfusion studies were technically adequate and only one had bowel activity very close to left ventricle. Statistics: Diagnostic accuracy and predictive values were calculated for radionuclide perfusion and function studies, considering both the decrease in EF and wall motion abnormality at rest or detection of new defects. The additive value of function over perfusion was also calculated. EF variation was analyzed by t test (p at 0.05). The DIP EF normal limit was considered to be at 5% under the rest value. This figure was obtained after analysis of our control group behavior. RESULTS Baseline and DIP changes in hemodynamic parameters in both control and CAD groups are shown in Table 1; there was no difference between controls and CAD patients in heart rate or blood pressure values, but there was a difference (p < 0.04) after DIP in both groups; 57% of CAD patients had an abnormal electrocardiogram with DIP but none of controls. New perfusion defects appeared with DIP in 23 CAD patients and in one control. The sensitivity for perfusion images was 85% and the specificity 92%; CAD prevalence was 77%. Excluding infarcted patients, sensitivity was 83%; accuracy 88%; positive and negative predictive values were 91% and 85%, respectively. No change was observed between rest and DIP EF in controls mean values were 46.9 +- 9.3 and 45.5 +- 8.6%, respectively; CAD patients presented a significant EF decrease with DIP (p: 0.015). The analysis of CAD patients according to the presence of ischemia on perfusion images is shown in Table 2; those with ischemia had a significant EF decrease with DIP (p < 0.0001 ) and also a lower rest EF than non ischemic patients (p: 0.034). CAD patients without ischemia did not have a significant EF change. On the other hand, basal EF was similar in the control group compared to patients with a different number of diseased vessel. However, DIP EF was significantly lower in patients with single or double vessel disease (p < 0.05); triple-vessel patients had the same trend. New wall motion abnormalities with DIP were observed in 11 CAD patients and also in 4 controls. Sensitivity and specificity with EF drop >= 5% or the occurrence of new wall motion abnormalities were lower than perfusion values (Table 3). Combined function and perfusion CAD sensitivity improved from 85% to 92.5% if adding EF drop and to 90% if adding new wall motion abnormalities. Specificity decreased from 91.7% to 75% with EF drop and to 58.3% when using new motion abnormalities, but none of the differences were significant (Table 4). The agreement between perfusion and functional abnormalities was only 54% (kappa 0.13). DISCUSSION In the results the value of DIP-handgrip perfusion images for CAD assessment was good and comparable with that reported in the current literature for exercise and pharmacological stress,1,9 and our specificity was somehow higher considering we did not include true normal subjects. When left ventricular functional parameters are added, rest images may help to detect CAD only if they are abnormal. A decrease in left ventricular EF or WMA with stress indicates ischemia in CAD patients. In exercise and DIP, the normal response of hemodynamic parameters differs: isotonic exercise induces a clear increase in EF, blood pressure and heart rate. Isometric exercise has a different behavior with only a weak EF effect in normals (it could even decrease and some patients develop motion abnormalities); compared to bicycle exercise, heart rate and blood pressure have a greater increase.7,10,11 The response to DIP infusion in normals is a mild decrease in systolic, diastolic and mean blood pressure and an increase in heart rate and EF. It has been reported that DIP induced coronary hyperemia produces mild hemodynamical changes in patients with and without CAD,12 Moreover, aging does not affect EF DIP response when healthy young are compared to an elderly population.13 In CAD, EF behavior was similar for DIP and exercise for Indolfi14; DIP EF changes sensitivity and specificity were 75% and 76%, respectively. Cates2 reported that DIP was very specific but a moderately sensitive test in severe CAD. Konishi studied a group of CAD patients finding a low correlation between exercise and DIP in EF and WMA, only 8% of the group DIP EF decrease versus 77% in exercise. The discrepancy was lower for WMA and also less abnormalities were observed with DIP.15 The lack of EF improvement in our controls may be explained by the combined effects of DIP and handgrip. These could be supported by the observations made with isometric exercise when compared with isotonic exercise, with opposite effects in left ventricular EF.16 Another study in young healthy volunteers compared 4 different stresses: dobutamine exercise, adenosine tri-phosphate (ATP) and DIP17; dobutamine obtained the greater left ventricular EF increase and DIP the minor (4 +- 3%; with a doses of 0.84 ug/kg/min), they used equilibrium ventriculography; furthermore, another group using standard doses had reported an EF decrease of more than 3% with DIP as a normal limit18 for CAD detection (with first pass ventriculography); 77% of their patients with reversible thallium-201 defects were under their normal EF limit. Sciagra4 evaluated the value of DIP first pass added to perfusion sestamibi images and his results demonstrated no helpful clinical information over perfusion images alone, his EF sensitivity was 63% and 70% for WMA, and 77% for any parameters. These exercise values are higher than our results. On the other hand, Bodenheimer,19 considered handgrip EF to be more sensitive than WMA for detecting CAD. Ischemic patients had a significantly diminished EF with DIP-handgrip in the protocol, as expected. Nevertheless, there are some reports that indicates little value in adding DIP or even exercise functional data to perfusion in CAD evaluation5,20 and our experience with DIP-handgrip also support this. On the other hand, triple vessel patients did not reach a significant decrease in DIP-handgrip EF versus basal EF, whereas single and double vessel patients did, probably due to the smaller size of that subgroup. Takeishi21 compared DIP EF in different patients also, and significant changes were observed only in multi-vessel disease. Limitations of the study Our control group had a rest EF value lower than usual considering the exercise and DIP values; this may be explained by problems with the measurement, but the analysis of EF changes was performed by comparing the same group under 2 sets of conditions, and they have statistical value. Another point is that the low correlation between DIP-handgrip perfusion and function (54%) may also be related to difficulties in acquiring first pass data with a single crystal camera. We only included studies with good technical characteristics in counts and bolus for the analysis, in which adequate processing was done.22,23 Other reports with DIP gave similar results,15,24 j.e. Zafrir et al. reported a 65% agreement between perfusion and function. We are aware that normal catheterized patients are not true normals but, after all, they represent clinical practice. It is well known that patients with low probability of CAD have better specificity for perfusion studies than patients with normal angiograms submitted to the test to evaluate a possible CAD. As reported by Rozansky25 this is true also for radionuclide ventriculography exercise. In that work angiographically normal patients included about a third of patients with abnormal EF and abnormal wall motion response, but in this paper we did not anaiyze exercise stress. DIP has a different mechanism of action; even reflex vasoconstriction stenoses in isometric exercise plays a role,26 and is able to modify the response. In conclusion, in these patients, simultaneous assessment of perfusion and function during DIP and handgrip stress has a limited incremental diagnostic value for the detection of CAD. Considering the importance of cost versus benefit concepts, this protocol does not appear a good choice, but for risk stratification assessment it needs to be established. ACKNOWLEDGMENTS Boehringer Ingelheim Chile for supporting with Persantine and to Ms. Andrea Rojas, Rita Alay and Pedro Quispe, Medical Technologists for their important help and also to Dr. Steven Port and Daniel Berman for their comments. REFERENCES 1. Palmas W, Friedman J, Diamond G, Silber H, Kiat H, Berman D. Incremental value of simultaneous assessment of myocardial function and perfusion with Technetium-99m Sestamibi for prediction of extent of coronary artery disease. J Am Coll Cardiol 25: 1024-1031, 1995. 2. Cates C, Kronenberg M, Collins W, Sandler M. Dipyridamole radionuclide ventriculography: A test with high specificity for severe coronary artery disease. J Am Coll Cardiol 13: 841-851, 1989. 3. Klein H. Ninio R, Eliyahu S, Bakst A, Levi A, Dean H, et al. Effects of the Dipyridamole Test on Left Ventricular Function in Coronary Artery Disease. Am J Cardiol 69: 482-488, 1992. 4. Sciagra R, Bisi G, Santoro G, Briganti V, Leontini M, Fazzini P. Evaluation of Coronary Artery Disease using Tc99m SESTAMIBI first pass and perfusion imaging with Dipyridamole infusion. J Nucl Med 35: 1251-1264, 1994. 5. Koutelou M. Chin T, Guidry G, Mahmarian J, Verani M. Simultaneous assessment of myocardial perfusion and ventricular function by adenosine-Sestamibi cardiac imaging. J Am Coll Cardiol 21 : 221A, 1993. 6. Afridi I, Main M, Parrish D, Kisilbash A, Levine B. Graybum P. Usefulness of isometric hand grip exercise in detection coronary artery disease during dobutamine atropine stress echocardiography in patients with either stable angina pectoris or another type or positive stress test. Am J Cardiol 82: 564-568, 1998. 7. Jones Rl, Lahiri A, Cashman PM, Dore C, Raftery EB. Left ventricular function during isometric hand-grip and cold stress in normal subjects. Br Heart J 55: 246-252, 1986. 8. Bassevich R. Zafrir N, Sulkes J, Lubin E. Dipyridamole First-Pass radionuclide Ventriculography: Prediction of future cardiac events. Am J Cardiol 74: 1229-1232, 1994. 9. Verani MS. Myocardial perfusion imaging versus two-dimensional echocardiography: comparative value in the diagnosis of coronary artery disease. J Nucl Cardiol 1 : 399-414, 1994. 10. Peter C, Jones R. Effects of isometric handgrip and dynamic exercise on left-ventricular function. J Nucl Med 21 : 1131-1138, 1980. 11. Loodbrook P, Byrne J, Reed F, Mc Knight R. Modification of left ventricular diastolic behavior by isometric handgrip exercise. Circulation 62: 357-370, 1980. 12. Ogilby JD, Kegel JG, Heo J, Iskandrian AE. Correlation between hemodynamic changes and tomographic sestamibi imaging during dipyridamole-induced coronary hyperemia. J Am Coll Cardiol 31: 75-82, 1998. 13. Chikamori T, Doi Y, Seo H. Kawamoto A, Akagi N, Maeda T, et al. Effects of Dipyridamole left ventricular systolic and diastolic function in healthy young and elderly subjects as assessed by radionuclide angiography. Am J Cardiol 73: 1024-1029, 1994. 14. Indolfi C, Betocci S, Piscione F, Perrone-Filardi P, et al. Assessment of left ventricular function using radionuclide angiography after Dipyridamole infusion. Chest 96: 1026-1030, 1989. 15. Konishi T, Koyama T, Aoki T, Futagami Y, Nakano T. Dipyridamole radionuclide ventriculography in patients with ergometer exercise. Angiologv 41 : 518-524, 1990. 16. Heng MK, Bai JX, Marin J. Changes in left ventricular wall stress during isometric and isotonic exercise in healthy male. Am J Cardiol 62: 794-798, 1988. 17. Coma-Canella I, Garcia Velloso MJ, Maceira A, Cabrera A, Villas A, Albadelejo V, et al. Radionuclide angiography in young healthy volunteers. Changes with different stress tests. Rev Esp Cardiol 50: 709-714, 1997. 18. Tono-oka Y, Meguro M, Takeishi Y, Maskane Y, Yamaguchi Y, Hoshi H, et al. Relationship of thallium-201 defect and left ventricular function after dipyridamole infusion. Jpn Circ J 53: 707-715, 1989. 19. Bodenheimer MM, Banka VS, Fooshe CM. Herma GA, Helfant RH. Comparison of wall motion and regional ejection fraction at rest and during isometric exercise: concise communication. J Nucl Med 20: 721-732, 1979. 20. Hambye AS, Vervaet A, Lieber S, Ranquin R. Diagnostic value and incremental contribution of bicycle exercise, first pass radionuclide angiography and 99mTc-labeled Sestamibi single photon emission computed tomography in the identification of coronary artery disease in patients without infarction. J Nucl Cardiol 3: 464-474, 1996. 21. Takeishi Y, Suekawa H, Saito H, Nishimura S, Shibu T, Sasaki Y, et al. Left ventricular function and myocardial perfusion during Dipyridamole infusion assessed by a single injection of 99mTc sestamibi in patients unable to exercise. Nucl Med Commun 15: 697-703, 1994. 22. Wackers J, Sinusas A, Saari M, Mattera J. Is list mode ECG-gated first pass LVEF accurate using single crystal gamma camera? Implications for Tc99m labeled perfusion imaging agents. J Nucl Med 32 (5): 939, 1991. 23. Nichols K, DePuey G, Gooneratne N, Salensky H, Friedman M, Cochoff S. First pass ventricular ejection fraction using a single-crystal nuclear camera. J Nucl Med 35 (8): 1292-1230, 1994. 24. Zafrir N, Bassevitch R, Shimoni A, Teplitsky I, Lubin E. Effect of Dipyridamole on myocardial perfusion and function using Technetium-99m MIBI. Int J Cardiol 49: 25-31, 1995. 25. Rozanski A, Diamond GA. Forrester JS, Berman DS, Morris D, Swan HJ. Alternative referent standards for cardiac normality. Implications for diagnosting testing. Ann Int Med 101: 164-171, 1984. 26. Brown BG, Lee AB, Bolson EL, Dodge HT. Refiex constriction of significant coronary stenosis as a mechanism contributing to ischemic left ventricular dysfunction during isometric exercise. Circulation 70: 18-24, 1984.