Search

The prevalence of a patent foramen ovale was assessed by simultaneously performing transesophageal contrast echocardiography and transcranial contrast Doppler sonography (TCD) in 137 subjects (mean age 36 years) with stroke of unclarified etiology (n = 41), clarified otiology (n = 33), and in normal subjects (n = 63; mean age 32 years). Patent foramen ovale was found significantly more often in patients with unclarified than clarified strokes or in normal subjects (66% vs 33%, or 43%). Massive paradoxical embolism through a patent foramen ovale, identified by TCD, occurred significantly (p

In 66 patients with suspected coronary artery disease (CAD), exercise electrocardiography (ECG), exercise echocardiography, dobutamine stress echocardiography (dosage, 5 to 40 [mu]g/kg/min), single-photon emission computed tomography (SPECT) using methoxy-isobutyl-isonitrile (MIBI) and coronary angiography were performed prospectively to compare methods for detecting CAD. CAD was defined as 70% luminal area stenosis in at least 1 coronary artery at coronary angiography. Significant CAD was present in 50 patients. Compared with exercise ECG, exercise echocardiography, dobutamine stress echocardiography and MIBI-SPECT had a significantly higher sensitivity (52% vs 80, 79 and 89%; p Coronary artery disease (CAD) is the leading cause of death in western countries. Detection of CAD for adequate treatment is therefore of primary importance. Exercise electrocardiography (ECG) has found wide acceptance and application for CAD identification. However, it is known that exercise ECG has only a limited sensitivity, especially in patients with 1-vessel disease. To detect stress-inducible changes in regional myocardial perfusion or contraction, several stress modalities such as exercise, pacing and pharmacologic agents have been combined with imaging techniques such as thallium scintigraphy, technetium radio-nuclide ventriculography and echocardiography. Exercise echocardiography is a valuable method for detecting stress-induced wall motion abnormalities,[1-5] but exercise testing is hampered by exertional hyperpnea and motion artifacts. Dobutamine stress echocardiography partially circumvents these problems and has been shown to be a useful diagnostic tool.[6,7] Myocardial single-photon emission computed tomography (SPECT) is nowadays a well-established method for detecting CAD with a high sensitivity.[8] This study prospectively compares the accuracy of exercise ECG, exercise echocardiography, dobutamine stress echocardiography and technetium-99m methoxy-isobutyl-isonitrile (MIBI)-SPECT for detecting CAD in each of 66 patients undergoing subsequent coronary angiography. METHODS Patients: The study group consisted of patients without prior Q-wave myocardial infarction who were referred for evaluation of suspected CAD. Sixty-six patients (51 men and 15 women, mean age 57 [+ or -] 10 years) were examined prospectively. All patients underwent supine bicycle exercise echocardiography, dobutamine stress echocardiography, MIBI-SPECT and subsequently coronary angiography. Medication was discontinued 24 hours before examination. All patients gave written informed consent. Exercise electrocardiography: A modified Bruce protocol was followed with evaluation according to standard criteria.[9] Exercise testing was performed beginning with a work load of 50 W and increased by steps of 25 W every 2 minutes. Twelve-lead electrocardiographic monitoring was recorded continuously during the exercise test and up to 6 minutes after cessation of exercise. Exercise was continued until 85% of the expected maximal heart rate was achieved, but stopped in case of exhaustion, development of severe angina, significant electrocardiographic changes, serious arrhythmia or hypotension. An abnormal test was defined as >0.1 mV of horizontal or downsloping ST-segment depression 80 ms after the J point in [greater than or equal to]2 leads. Blood pressure recordings were obtained from an automatic cuff sphygmomanometer. Bicycle exercise echocardiography: Exercise echocardiography was performed simultaneous with exercise ECG. Before exercise, resting sequences were acquired in the parasternal short- and long-axis and apical 4- and 2-chamber views (Siemens SL, 3.5 MHz) with the patient in the left lateral decubitus position. Images were digitized and stored on floppy disk using a Freeland Computer system. The system acquires and digitizes 8 serial echocardiographic frames at 50 ms intervals during systole of a single cardiac cycle, triggered by the electrocardiogram. The images can be displayed in cine-loop format and side-by-side with the postexercise images. Patients then performed symptom-limited bicycle exercise with electrocardiographic monitoring according to the criteria previously described. Immediately after cessation of exercise, patients resumed their initial left lateral decubitus position for repeated imaging of the 4 described views. Recording was completed within 60 seconds of exercise termination. Dobutamine stress echocardiography: Resting sequences in the lateral decubitus position of the described 4 views were acquired before infusion of dobutamine. Dobutamine infusion was begun at a rate of 5 [mu]g/kg/min, increasing every 2 minutes to 10, 20, 30 and 40 [mu]g/kg/min. End points were maximal dosage, a heart rate of 85% of age-predicted maximal heart rate, horizontal or downsloping ST-segment depression of >0.2 mV 0.08 second after the J point in [greater than or equal to]2 leads, or angina. If this was not achieved by maximal dobutamine infusion alone, intravenous atropine in a dosage of 0.5 to 1.5 mg was given along with the dobutamine infusion. Echo images were acquired again in the described manner during infusion. All exercise echocardiograms were interpreted by 2 experienced independent observers unaware of all other data. A scheme modified from that proposed by Bourdillon et al,[10] dividing the left ventricle into 16 segments, was used for grading wall motion (Figure 1). Using the side-by-side imaging technique to compare resting and exercise images, new wall motion abnormalities described as hypokinetic, akinetic or dyskinetic could be detected more easily. Patients in whom none of the 4 views showed sufficient image quality for evaluation of all left ventricular segments were excluded from the study. Technetium-99m MIBI-SPECT: Fifty-five patients underwent technetium-99m MIBI-SPECT simultaneously with bicycle exercise testing. Four hundred MBq of technetium 99m-MIBI were injected intravenously during maximal stress load, 1.5 minutes before termination of stress. Data acquisition with a rotating gamma camera (Siemens Gammasonics Rota-Dual) was performed 2 hours later. In case a perfusion defect occurred during exercise, MIBI-SPECT was repeated under resting conditions within a period of maximal 2 weeks. Transversal, long- and short-axis cuts through the left ventricle were obtained by means of a dedicated computer system and evaluated quantitatively as described elsewhere.[11] Coronary angiography: All patients underwent coronary angiography within 2 weeks of exercise testing using the Judkins technique. The angiogram was interpreted by angiographers unaware of other clinical data. CAD was defined as luminal area stenosis of >70% of at least 1 major coronary artery branch. Two orthogonal planes were used to measure the degree of luminal area narrowing. The measurements were performed manually with calipers. Subsequently, sensitivity, specificity and overall accuracy for exercise ECG, exercise echocardiography, dobutamine stress echocardiography and MIBI-SPECT were evaluated using the coronary angiogram as the gold standard. Statistical comparisons were made using the chi-square test. Differences were significant at p >0.05. RESULTS Angiography: Significant stenosis was angiographically detected in 50 patients, 1-vessel disease was seen in 29 patients (11 of them had a right-sided CAD), stenosis of the left anterior descending artery was seen in 13, and left circumflex disease in 5. Ten patients had 2-vessel and 11 had 3-vessel disease. Sixteen patients had no significant stenosis. Exercise electrocardiography: The rate-pressure product reached at maximal stress load was 24,851 [+ or -] 4,230 mm Hg [min.sup.-1]. Seven patients did not reach their target heart rate because of dyspnea or leg fatigue. Significant segment depression was recorded in 26 of the 50 patients with CAD, resulting in a sensitivity of 52%. Thirteen patients with 1-vessel disease had an abnormal exercise ECG result (sensitivity 45%). One of 16 patients without CAD had a positive exercise ECG result (specificity 93%). The overall accuracy was 62%. Exercise echocardiography: Postexercise echocardiography showed insufficient endocardial border definition in 6 of 66 patients (9%). Sensitivity, specificity and accuracy for detection of CAD in all 66 patients were 80, 87 and 82%, respectively. Comparison with exercise ECG showed a better sensitivity and accuracy (p Dobutamine stress echocardiography: Dobutamine stress echocardiography was performed in 64 patients. Two patients were not examined for safety reasons because they developed severe arrhythmias during the previous exercise test. In 4 patients endocardial border definition with maximal dobutamine infusion was insufficient. In 24 patients predefined maximal heart rate was reached with the maximal dobutamine dosage (40 [mu]g/kg/min), 26 patients needed additional atropine, and in 10 patients a dobutamine dosage of Technetium-99m MIBI-SPECT: The sensitivity of MIBI-SPECT for detecting CAD was 89%; however, specificity was only 71%. No significant difference was found comparing MIBI-SPECT with exercise and dobutamine echocardiography (Table I). Because MIBI was injected at peak exercise, the rate-pressure product was identical to that of exercise ECG and exercise echocardiography. There was agreement in detecting hemodynamically significant stenosis in 41 of 50 patients, in whom exercise echocardiography and MIBI-SPECT were performed and evaluable. In 2 patients MIBI-SPECT results were negative, whereas results of exercise echocardiography were positive. One patient had a left anterior descending artery stenosis and 1 had no significant coronary artery stenosis. Of the 7 patients with positive MIBI-SPECT but negative exercise echocardiography, 2 had 3-vessel disease, 1 had left anterior descending and 1 left circumflex artery stenosis. In 3 patients significant stenosis could be excluded. Concordance between MIBI-SPECT and dobutamine stress echocardiography was found in 38 of 50 patients, in whom both methods were performed. Sensitivity with regard to number of diseased vessels: Sensitivity of exercise ECG for detecting CAD was only 45% in patients with 1-vessel disease compared with 62% in patients with 2- and 3-vessel disease. Sensitivity of exercise and dobutamine stress echocardiography in patients with 1-vessel disease was higher (79 and 78% respectively). Similar results were found for multivessel disease (Table II). MIBI-SPECT resulted in a sensitivity of 84% for patients with 1-vessel disease and 94% for those with 2- and 3-vessel disease. There was a highly significant difference in the sensitivity of exercise ECG and the other stress tests in detecting 1-vessel disease. Comparison of sensitivities for detecting 1-vessel disease versus multivessel disease was not significant. [TABULAR DATA II OMITTED] Sensitivity with regard to the location of the diseased vessel: With both echocardiography stress tests, sensitivity in patients with 1-vessel disease was lowest for detection of left circumflex artery stenosis, whereas sensitivity for detection of right coronary artery stenosis was highest. With exercise ECG, highest sensitivity was found for the left anterior descending artery (Table III). However, none of the evaluated exercise tests showed a significant difference in sensitivity between the diseased vessels. [TABULAR DATA III OMITTED] Incremental value of exercise and dobutamine echocardiography in patients with normal exercise electrocardiogram: Of the 39 patients with normal exercise ECG, 24 had CAD on coronary angiography. Sixteen of the 24 patients (67%) with false-negative results on exercise ECG had a positive exercise echocardiogram and 17 (71%) had a positive dobutamine stress echocardiogram. MIBI-SPECT yielded positive results in 84% of patients with negative results on exercise ECG. In the 29 patients with 1-vessel disease there were 16 with false-negative results on exercise ECG. In these 16 patients exercise and dobutamine echocardiography yielded additional 10 (63%) and 11 (69%) positive results, respectively. MIBI-SPECT yielded positive results in 84% of 1-vessel CAD patients with negative results on exercise ECG. DISCUSSION Background: Recognition of CAD is important for the treatment of morbidity and the prevention of mortality. Screening methods are therefore of primary importance. Exercise ECG, though widely applied, has only a limited sensitivity. Froelicher[12] reported a sensitivity of 64% (range 33 to 82) for exercise ECG in a review of 8 different studies. Myocardial SPECT is a well recognized method with high sensitivity. Pooled data from exercise thallium-201 SPECT studies indicate a 90% (range 82 to 98) overall sensitivity for this method of detecting CAD.8 However this technique has the disadvantage of being expensive and having restricted availability. Exercise echocardiography was first studied as a screening test for CAD in 1979 by Wann et al.[13] Subsequently, promising results were reported.[3,5,14-16] However, this technique is technically difficult, especially because of insufficient image quality caused by hyperpnea. With the development and application of digital imaging and storing techniques and the acquisition of images immediately after[17] rather than during peak exercise, the percentage of patients with technically adequate studies has increased substantially.[11,19] Exercise echocardiography is therefore increasingly accepted as a screening test for CAD.[20] Dobutamine stress echocardiography, in contrast, is still investigational for CAD screening, although favorable reports have been published.[6,7,21-25] Several advantages make this method attractive: (1) less motion artifacts than with exercise echocardiography, (2) patients with physical inabilities to perform physical exercise can be examined, (3) low costs, and (4) equipment is widely available.[6] Different dosages of dobutamine have been applied in the existing studies, reaching from 20 to 50 [mu]g/kg/min,[6,7,21] and additional atropine is sometimes given to increase heart rate. The reported sensitivity was high, ranging from 78 to 95%.[6,7,21-25] Significance of the study: This study for the first time systematically evaluates the sensitivity, specificity and accuracy of 4 noninvasive stress tests compared with coronary angiography in the same collective of patients. Our results confirm the low sensitivity of exercise ECG for detecting CAD, especially in patients with 1-vessel disease. The overall sensitivity of 52% as well as the sensitivity of 45% for those with 1-vessel disease are in the range reported by Froelicher.[12] The overall sensitivity in our group is always determined by the high percentage of patients with 1-vessel disease in this study (29 of 50). Exercise echocardiography had a significantly higher sensitivity than exercise ECG (80%), and a specificity of 87%, similar to exercise ECG. In this study only patients without prior Q-wave infarction were investigated. No patient in this study had wall motion abnormalities at rest. Therefore, sensitivity of studies was only determined by new transient wall motion abnormalities. The selection of patients influences the sensitivity of a study.[26] Studies including patients with prior Q-wave infarction and resulting wall motion abnormalities at rest result in a higher sensitivity if no clear distinction between preexisting wall motion abnormalities and transient exercise-induced dyssynergy is made. This has not always been done. In this study population, sensitivity in patients with 1-vessel disease was similar to patients with 2- and 3-vessel disease. Lower sensitivity in patients with 1-vessel disease has been reported.[4,26] However, these studies were not significant, similar to this investigation. In the group with 1-vessel disease, the sensitivity was lowest in patients with left circumflex artery, which might be due to the smaller perfusion bed of the circumflex artery and the fact that its endocardium is defined by the lateral rather than the axial resolution. No reports exist indicating a significantly higher sensitivity for detecting left anterior descending coronary artery stenosis compared with left circumflex artery stenosis, although similar trends, as in this study, have been reported.[4,27] In 9% of patients, endocardial border definition immediately after exercise was deemed to be inadequate of analysis. Dobutamine stress echocardiography had a sensitivity and specificity similar to those in the small number of studies published so far.[5,6,22,23,25] We did not reach the high sensitivity of 96% reported by Marcovitz and Armstrong.[23] This may be in part due to a different proportion of patients with normal resting wall motion. In their subgroup of patients with normal resting wall motion, a sensitivity of 87% was reported. In this study, high-dose dobutamine in combination with additional atropine was used to achieve a high rate-pressure product. Our maximal rate-pressure product was higher than that reported by other groups.[23,24] Theoretically this should induce more ischemia; on the other hand, hyperkinesia the heart deteriorates image quality, and we had to exclude 4 patients (6%) from analysis because of insufficient image quality during administration of high-dose dobutamine. However, the number of patients excluded from further analysis of dobutamine stress echocardiography was lower than that for the exercise echocardiography. Note that since we did not perform continuous imaging throughout the test, the rate-pressure product at the onset of ischemia was not determined. Although a lower sensitivity for identifying patients with 1-vessel disease in the left circumflex coronary artery distribution was identified compared with the right and left coronary artery, this difference was not statistically significant. Segar et al[24] also did not detect a significant difference in the percentage of positive studies regarding the 3 coronary artery distributions. Marcovitz and Armstrong[23] found similar sensitivities, distinguishing only between anterior and posterior circulation. In the same study no significant difference was reported between sensitivity of 1-vessel disease and multivessel disease, similar to our results. SPECT scintigraphy had the highest sensitivity but a markedly lower specificity than the other exercise tests, as has also been reported.[8,27] Clinical implications: This study clearly supports the use of exercise echocardiography as a test for detecting CAD. Dobutamine stress echocardiography had the advantage over exercise echocardiography of having a slightly higher image acquisition success rate due to less motion artifacts under stress conditions. The sensitivity of dobutamine stress echocardiography was similar to exercise echocardiography with 79% instead of 80%, although the maximal rate-pressure product reached was significantly lower. This is probably due to the higher image quality as a result of less motion artifacts and the fact that we acquired exercise echocardiographic images after exercise, whereas dobutamine stress echocardiography images were acquired during dobutamine infusion. The sensitivity of dobutamine stress echocardiography was in a range similar to exercise MIBI-SPECT, which is in accordance with a recent study by Savas et al.[22] All but 1 patient did not develop serious side effects requiring the termination of dobutamine infusion before reaching an end point. However, we did not examine 2 patients in whom the preceding exercise echocardiography had led to ventricular tachycardia. One patient with severe 3-vessel disease developed a sudden decrease in systolic blood pressure, but recovered after stopping the dobutamine infusion. In patients with suspected CAD but negative exercise ECG, exercise and dobutamine echocardiography yield substantial incremental information supporting its use in a stepwise diagnostic approach to CAD. The main limitation of this study is clearly the limited number of patients. Further studies with larger study populations are needed to support the described findings. [1.] Armstrong W, O'Donnell J, Dillon J, McHenry P, Mortis S, Feigenbaum H. Complementary value of two-dimensional exercise echocardiography to routine treadmill exercise testing. Ann Intern Med 1986;105:829-835. [2.] Visser C, Van Der Wieken R, Kan G, Lie KI, Busemann-Sokele E, Meltzer RS. Durrer D. Comparison of two-dimensional echocardiography with radionuclide angiography during dynamic exercise for the detection of coronary artery disease. Am Heart J 1983; 106:528-534. [3.] Maurer G, Nanda NC. Two dimensional echocardiographic evaluation of exercise-induced left and right ventricular asynergy: correlation with thallium scanning. Am J Cardiol 1981;48:720-727. [4.] Ryan T, Vasey CG, Presti CF, O'Donnell JA, Feigenbaum H, Armstrong WF. Exercise echocardiography: detection of coronary artery disease in patients with normal left ventricular wall motion at rest. J Am Coll Cardiol 1988; 11:993-999. [5.] Robertson WS, Feigenbaum H, Armstrong WF, Dillon JC, O'Donnell J, McHenry PW. Exercise echocardiography: a clinically practical addition in the evaluation of coronary artery disease. J Am Coll Cardiol 1983;2:1085-1091. [6.] Berthe C, Pierard LA, Hiernaux M, Trotteur G, Lempereur P, Carlier J, Kulbertus HE. Predicting the extent and location of coronary artery disease in acute myocardial infarction by echocardiography during dobutamine infusion. Am J Cardiol 1986;58:1167-1172. [7.] Sawada SG, Segar DS, Ryan T, Brown SE, Dohan AM, Williams R, Fineberg NS, Armstrong WF, Feigenbaum J. Echocardiographic detection of coronary artery disease during dobutamine infusion. Circulation 1991;83:1605-1614. [8.] Mahmarian JJ, Verani MS. Exercise thallium-201 perfusion scintigraphy in the assessment of coronary artery disease (abstr). Am J Cardiol 1991;67:2d-11d. [9.] Braunwald: Heart Disease. A Textbook of Cardiovascular Medicine. Saunders. 1992; 163-166. [10.] Boardillon PDV, Broderick TW, Sawada SG, Armstrong WF, Ryan T, Dillon JC, Fineberg NS, Feigenbaum H. Regional wall motion index for infarct and noninfarct regions after perfusion in acute myocardial infarction: comparison with global wall motion index. J Am Soc Echo 1989;2:398-407. [11.] Buell U, Dupont F, Uebis R, Kaiser HJ, Kleinhans E, Reske SN, Hanrath P. [sup.99]TCm-methoxy-isobuthyl-isonitrile SPECT to evaluate index from regional myocardial uptake after exercise and at rest. Results of a four hour protocol in patients with coronary heart disease and in controls. Nucl Med Commun 1990;11:77-94. [12.] Froelicher VF. Use of the exercise electrocardiogram to identify latent coronary artherosclerotic heart disease. In: Amsterdam EA, Wilmore JH, DeMaria AN, eds. Exercise in Cardiovascular Health and Disease. New York: York Medical Books, 1977:189-208. [13.] Wann LS, Faris JV, Childress RH, Dillon JC, Weyman AE, Feigenbaum H. Exercise cross-sectional echocardiography in ischemic hearl disease. Circulation 1979;60:1300-1308. [14.] Morgenroth J, Chen CC, David D. Exercise cross-sectional echocardiographic diagnosis of coronary artery disease. Am J Cardiol 1981;47:20-26. [15.] Limacher MC, Quiones MA, Poliner LR, Nelson JC, Winters WL, Waggoner AD. Detection of coronary artery disease with exercise two-dimensional echocardiography. Circulation 1983;67:1211-1218. [16.] Crawford MH, Amon KW, Vance WS. Exercise 2-dimensional echocardiography. Am, J Cardiol 1983;51:1-6. [17.] Berberich SN, Zager JRS, Plotnick GD, Fisher ML. A practical approach to exercise echocardiography: immediate postexercise echocardiography. J Am Coll Cardiol 1984;3:284-290. [18.] Bairey CN, Rozanski A, Berman DS. Exercise echocardiography: ready or not? J Am Coll Cardiol 1988;11:1355-1358. [19.] Crouse LJ, Harbrecht JJ, Vacek JL, Rosamond TL, Kramer PH. Exercise echocardiography as a screening test for coronary artery disease and correlation with coronary arteriography. Am J Cardiol 1991;67:1213-1218. [20.] Armstrong WF. Exercise echocardiography: ready, willing and able. J Am Coll Cardiol 1988;11:1359-1361. [21.] Mannering D, Cripps T, Leech G, Mehta N, Valantine H, Gilmour S, Bennett ED. The dobutamine stress test as an alternative to exercise testing after acute myocardial infarction. Br Heart J 1988;59:521-526. [22.] Savas V, Ajluni SC, Juni JE, Ostascewski T, Hauser AJ. Dobutamine stress echocardiography: an alternative to thallium scintigraphy (abstr). Circulation 1990;82 (suppl III):III-744. [23.] Marcovitz PA, Armstrong WF. Accuracy of dobutamine stress echocardiography in detecting coronary artery disease. Am J Cardiol 1992;69:1269-1273. [24.] Segar DS, Brown SE, Sawada SG, Ryan T, Feigenbaum H. Dobutamine stress echocardiography: correlation with coronary lesion severity as determined by quantitative angiography. J Am Coll Cardiol 1992; 19:1197-1202. [25.] Mazeika PK, Nadazdin A, Oakaey CM. Dobutamine stress echocardiography for detection and assessment of coronary artery disease. J Am Coll Cardiol 1992; 19:1203-1211. [26.] Armstrong WF, O'donnell J, Ryan T, Feigenbaum H. Effect of prior myocardial infarction and extent and location of coronary disease on accuracy of exercise echocardiography. J Am Coll Cardiol 1987; 10:531-538. [27.] Kiat H, Maddahi J, Roy LT, Train KV, Friedman J, Resser K, Berman DS. Comparison of technetium 99m methoxy-isobutyl isonitrile and thallium 201 for evaluation of coronary artery disease by planar and tomographic methods. Am Heart J 1989; 117:1-11. From the Medical Clinic I and the Department of Nuclear Medicine, Aachen, Germany. Manuscript received December 7, 1992; revised manuscript received April 27, 1993, and accepted April 28. Address for reprints: Rainer Hoffmann, MD, Medical Clinic I, Klinikum der RWTH Aachen, Pauwelsstrasse, D-5100 Aachen, Germany.

In 20 consecutive patients (18 men and 2 women, aged 42 to 72 years) undergoing repeat coronary angiography because of new onset of angina pectoris 4 months to 11 years (mean 53 months) after aortocoronary saphenous venous bypass operation, the graft to the left anterior descending (n = 12), left circumflex (n = 4) or right coronary (n = 2) artery, or a diagonal branch (n = 2) was studied by both intravascular ultrasound and angiography. Sonographic images were obtained using a 4.8Fr catheter with a crystal mechanically rotated at 900 rpm; quantitative coronary angiograms were recorded in biplane projections. In 18 patients, qualitatively as well as quantitatively evaluable images could be recorded; no complications occurred. The venous wall in general appeared to be homogenous; there were no separate layers identifiable. Simultaneous ultrasound and angiographic measurements were performed at a total of 75 sites (2 to 6 per bypass). In 4 of these patients (10 of 75 sites), neither intravascular ultrasound nor angiography revealed any pathologic changes; these bypasses were classified as normal. At the remaining 65 sites, arteriosclerotic lesions were detected in each case by ultrasound, but at only 33 sites by angiography. Median wall thickness was 0.59 mm (95% confidence interval 0.54 to 0.63) in normal grafts and 1.02 mm (0.99 to 1.07; p (Am J Cardiol 1992;70:455-458)