Noninvasive identification of myocardium at risk in patients with acute myocardial infarction and nondiagnostic electrocardiograms with technetium-99m-Sestamibi

BACKGROUND

Patients who have chest pain without electrocardiographic ST elevation are not candidates for thrombolytic therapy in most clinical trials. This study examined the value of technetium-99m-Sestamibi tomographic imaging to assess myocardial perfusion in patients during chest pain without ST elevation.

METHODS AND RESULTS

Tc-99m-Sestamibi was injected in 14 patients who had chest pain without ST elevation, who subsequently developed enzymatic evidence of myocardial infarction within 24 hours. Tomographic imaging was performed 1-6 hours after injection. Thirteen of 14 patients showed significant perfusion defects indicative of acute myocardial infarction consistent with absent perfusion (20 +/- 15% of the left ventricle; range, 2-53%); one patient had normal images. Because of the absence of definitive electrocardiographic changes, only five patients received reperfusion therapy within 6 hours of the onset of chest pain. Regional wall motion abnormalities were present in nine of nine patients undergoing contrast ventriculography and correlated with the location of the Tc-99m-Sestamibi perfusion defect. At the time of subsequent coronary angiography, total arterial occlusion was present in 11 of the 14 patients. The infarct-related artery could be identified in 13 of the 14 patients. In six of these 13 patients, the left circumflex was the infarct-related artery.

CONCLUSIONS

Patients who have chest pain without electrocardiographic ST elevation may have arterial occlusion and significant myocardium at risk. Tc-99m-Sestamibi imaging may be of benefit in identifying these patients early so that they can be considered for acute reperfusion therapy.

Limitations of the electrocardiogram in estimating infarction size after acute reperfusion therapy for myocardial infarction

OBJECTIVE

To assess the ability of the 12-lead electrocardiogram to estimate infarction size after reperfusion therapy for acute myocardial infarction.

DESIGN

The presence or absence of Q waves and the Selvester QRS score obtained before and after hospital discharge were compared with radionuclide estimates of infarction size and ejection fraction at discharge and 6 weeks later, regional wall motion at discharge and 6 weeks later, and myocardial perfusion defect size quantitated with Tc-99m-sestamibi at discharge.

SETTING

A tertiary referral center.

PATIENTS

A consecutive series of 43 patients with acute myocardial infarction who received acute reperfusion therapy and were assessed using 12-lead electrocardiography, radionuclide angiography, and Tc-99m-sestamibi tomographic imaging before discharge.

INTERVENTIONS

All 43 patients received acute reperfusion therapy: 21 patients received intravenous tissue plasminogen activator, and 22 patients underwent primary percutaneous transluminal coronary angioplasty.

MAIN OUTCOME MEASURE

The correlation of QRS score and Q waves with three radionuclide estimates of infarction size.

RESULTS

A significant correlation was found between myocardial perfusion defect size at discharge and both left ventricular ejection fraction and regional wall motion at discharge and 6 weeks later (r = -0.71 to -0.81; all comparisons, P less than 0.001). Little correlation was found between electrocardiographic findings and radionuclide measurements of left ventricular function and perfusion. Presence or absence of Q waves at discharge was not associated with any difference in ejection fraction, regional wall motion, or perfusion defect at discharge. No correlation was found between QRS score and ejection fraction or myocardial perfusion defect size at discharge. The QRS score at discharge correlated only weakly with regional wall motion at discharge and 6 weeks later. This lack of correlation was unchanged when electrocardiograms obtained after hospital discharge were analyzed.

CONCLUSION

Although inexpensive and readily available, the 12-lead electrocardiogram does not appear to provide a reliable estimate of infarction size after reperfusion therapy for acute myocardial infarction.

Comparison of exercise performance in left main and three-vessel coronary artery disease

From a consecutive series of patients who underwent rest and exercise radionuclide angiography over several years, we retrospectively identified 34 patients with left main coronary artery disease and 103 patients with three-vessel coronary artery disease who did not have significant left main disease. The results of gated equilibrium radionuclide angiography were compared in these 2 groups. Multiple exercise hemodynamic, exercise electrocardiographic, and exercise radionuclide angiographic parameters were considered in an attempt to separate the 2 groups. The only parameter that was significantly different between the 2 groups was exercise heart rate. However, no value of the exercise heart rate could meaningfully separate the 2 groups. Despite their known difference in prognosis, patients with left main and three-vessel disease had very similar exercise performance and could not be distinguished from one another by exercise electrocardiography or exercise radionuclide angiography. The inability to distinguish these two groups is a clear limitation of noninvasive exercise modalities.

Determination of diastolic function by radionuclide ventriculography

Diastolic filling can be measured by radionuclide ventriculography with use of several techniques including those based on gated and list-mode acquisitions, the first-pass method, and the nuclear probe. Radionuclide ventriculography specifically assesses volumes, rates of volume change, and intervals during ventricular filling. Normal values for diastolic filling measurement vary depending on the individual radionuclide methods used and the age of the patient. Comparative studies of the radionuclide method with contrast angiographic and Doppler echocardiographic techniques for measuring diastole are discussed, and the advantages and disadvantages of the radionuclide techniques are explored. The role of radionuclide assessment of diastolic function in specific clinical examples of hypertrophic cardiomyopathy, hypertension, anthracycline-induced cardiomyopathy, and coronary artery disease is reviewed. Radionuclide ventriculography is an accurate and easily applicable procedure for studying left ventricular volume changes in diastole.

Prevention of late ventricular dilatation after acute myocardial infarction by successful thrombolytic reperfusion

To examine the sequential changes in left ventricular volume after thrombolytic therapy for acute myocardial infarction, gated radionuclide ventriculography was performed within 12 hours of thrombolysis and at 1 and 6 weeks in 34 consecutive patients who received intravenous thrombolytic therapy in the Thrombolysis in Myocardial Infarction Trial. Angiographic confirmation of immediate reperfusion (mean 5.6 hours after onset of symptoms) that persisted at 24 hours was noted in 24 patients; 10 patients were not reperfused. A small (9.5%), but significant (p = 0.05), increase in end-diastolic volume index was noted in the reperfused group between 1 and 6 weeks; however, a marked degree of dilatation (35%) was noted in the non-reperfused group (p = 0.01). The change in left ventricular volume between 1 and 6 weeks differed in the 2 groups for both end-diastolic volume index and end-systolic volume index (p = 0.01 and p = 0.02, respectively). By 6 weeks, both end-diastolic volume index and end-systolic volume index were greater in the nonreperfused group (p less than 0.05). Between the acute and 6-week studies, definite increases in end-diastolic volume index (p less than 0.05) and end-systolic volume index (p less than 0.01) occurred commonly in the nonreperfused group but rarely in the reperfused group. Compared to the nonreperfused group, the reperfused group also had significantly higher ejection fractions at both 1 and 6 weeks (p less than 0.05). The change in end-diastolic volume index between 1 and 6 weeks correlated significantly and inversely with the ejection fraction at 1 week (r = -0.60, p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

Supine exercise electrocardiography compared with exercise radionuclide angiography in noninvasive identification of severe coronary artery disease

STUDY OBJECTIVE

To determine the incremental value of exercise radionuclide angiography for identification of severe coronary artery disease.

DESIGN

Retrospective analysis comparing logistic regression models.

SETTING

A tertiary care referral center.

PATIENTS

Three hundred and ninety-one consecutive patients who had normal resting electrocardiograms (ECGs) and no digoxin therapy within the previous week.

MEASUREMENTS AND MAIN RESULTS

The exercise ECG model, consisting of magnitude of ST depression, exercise heart rate, and patient gender, was highly predictive of three vessel or left main coronary artery disease (chi 2 = 100, P less than 0.0001). The model correctly classified 60% of the study group which included 56 patients with and 179 without severe disease. The addition of radionuclide angiographic variables improved the predictive power of the model (chi 2 = 124, P less than 0.0001). However, the exercise radionuclide angiographic variables increased the number of patients who were correctly classified by only 11 and the percentage by 3% (to a total of 63% of the study group).

CONCLUSIONS

The modest additional advantage provided by exercise radionuclide angiography for identification of three vessel or left main coronary artery disease in patients with normal resting ECGs would not appear to justify its routine use for this purpose. Before this conclusion is used as a guide for clinical practice, our results should be prospectively confirmed in a separate sample of patients in another institution.

Left ventricular systolic response to exercise in patients with systemic hypertension without left ventricular hypertrophy

Supine exercise radionuclide angiography was performed in 367 men to assess left ventricular (LV) systolic response to exercise; 58 had systemic hypertension without LV hypertrophy on a resting electrocardiogram and 309 were normotensive. All patients met the following criteria defining a low pretest likelihood of coronary artery disease: age less than 50 years; normal electrocardiographic response to exercise; absence of typical or atypical chest pain; and exercise heart rate greater than 120 beats/min. Patients taking beta-receptor blockers were excluded. There were no significant differences between hypertensive and normotensive groups in peak exercise heart rate, workload or exercise duration. However, hypertensive patients had significantly higher peak exercise systolic blood pressures and peak exercise rate-pressure products. There were no differences between patients with and without hypertension in resting ejection fraction, peak exercise ejection fraction (hypertensive patients 0.71 +/- 0.01, normotensive patients 0.70 +/- 0.05) or change in ejection fraction at peak exercise (hypertensive patients 0.07 +/- 0.01, normotensive patients 0.07 +/- 0.04). Diastolic and systolic ventricular volumes tended to be smaller in the hypertensive patients, but the difference was not statistically significant. The change in systolic volume with exercise was similar in the 2 groups (hypertensive -10 +/- 3 ml/m2, normotensive -10 +/- 1 ml/m2). In the absence of electrocardiographic evidence of LV hypertrophy, systemic hypertension does not influence LV systolic response to exercise.

Influence of age and sex on left ventricular filling at rest in subjects without clinical cardiac disease

Left ventricular (LV) filling at rest was studied by radionuclide ventriculography using alternate R-wave gating in 42 patients (29 men, 13 women) who had a low likelihood of cardiac disease. LV filling measurements differed little between men and women. Age was correlated positively with atrial filling duration (r = 0.55), atrial filling duration fraction (r = 0.52) and atrial filling fraction (r = 0.56) and negatively with rapid filling fraction (r = -0.58). Age was not correlated with peak filling rate, time to peak filling rate and first-half filling fraction. The heart rate at rest was significantly negatively correlated with rapid (r = -0.62), slow (r = -0.81) and atrial (r = -0.72) filling durations, but not with isovolumic duration. The heart rate at rest was weakly positively correlated with peak filling rate in end-diastolic volume per second (r = 0.36) and negatively correlated with first-half filling fraction (r = -0.35). Systolic pressure at rest influenced atrial filling duration. LV ejection fraction and end-diastolic volume index were not correlated significantly with LV filling in relatively normal subjects.

Myocardial stunning in hypertrophic cardiomyopathy: recovery predicted by single photon emission computed tomographic thallium-201 scintigraphy

A young woman with hypertrophic cardiomyopathy confirmed by echocardiography and cardiac catheterization presented with chest pain and features of a large left ventricular aneurysm. The initial diagnosis was myocardial ischemia with either an evolving or an ancient myocardial infarction. Subsequently, verapamil therapy was associated with complete resolution of the extensive left ventricular wall motion abnormalities, normalization of left ventricular ejection fraction and a minimal myocardial infarction. Normal thallium uptake on single photon emission computed tomographic scintigraphy early in the hospital course predicted myocardial viability in the region of the aneurysm. Thus, orally administered verapamil may reverse spontaneous extensive myocardial ischemia in hypertrophic cardiomyopathy and possibly limit the extent of myocardial infarction in such circumstances.

Gender-related differences in cardiac response to supine exercise assessed by radionuclide angiography

This study examines the recently reported gender differences in cardiac responses to exercise. The study group consisted of 192 men and 67 women with a low probability of coronary artery disease who underwent supine exercise radionuclide angiography. Men had a lower rest ejection fraction than that of women (0.63 versus 0.66, p = 0.02) and greater increases in ejection fraction with exercise (0.08 versus 0.02, p = 0.0001). The slope relating ejection fraction to metabolic equivalents of exercise (METs) was greater (p = 0.004) for men, even after adjustment for differences in rest ejection fraction and end-diastolic volume index. Compared with men, women had a smaller rest end-diastolic volume index (87 versus 97 ml/m2, p = 0.003) and a greater increase in end-diastolic volume index with exercise (6 versus -2 ml/m2, p = 0.002). The slope relating end-diastolic volume to METs was greater for women, even after adjustment for differences in rest end-diastolic volume index and peak work load. There are clear gender differences in the supine exercise response of ejection fraction and end-diastolic volume that are not explained by differences in exercise capacity.

Significance of T-wave pseudonormalization during exercise. A radionuclide angiographic study

In 84 consecutive patients with resting T-wave inversion, radionuclide angiography revealed significant new wall motion abnormalities in 13 (28 percent) of the 47 patients with persistent T-wave inversion and in 23 (62 percent) of the 37 patients with T-wave pseudonormalization during exercise (p less than 0.01). The response of the ejection fraction to exercise was better in patients with persistent T-wave inversion than in those with pseudonormalization (p less than 0.04). Mechanical evidence of ischemia was seen in 14 (61 percent) of the 23 patients with T-wave pseudonormalization but without ST-segment depression. In patients with resting T-wave inversion, pseudonormalization was slightly more sensitive but less specific than a positive exercise test for predicting significant new wall motion abnormalities or decreases in the ejection fraction with exercise. Although pseudonormalization is not extremely useful alone, the presence or absence of this finding can increase the diagnostic accuracy of exercise electrocardiography in patients with resting T-wave inversion and suspected ischemic heart disease.

Diastolic measurements from alternate R-wave gating of radionuclide angiograms

Left ventricular diastolic filling measurements were determined by means of standard consecutive R-wave gating, list mode acquisition, and alternate R-wave gating. Time-activity curves obtained by the latter two methods were equally accurate in quantifying rapid, slow, and atrial left ventricular filling, whereas curves obtained by means of standard gating were inadequate for this purpose.

Prognostic value and limitations of exercise radionuclide angiography in medically treated coronary artery disease

We investigated whether exercise radionuclide angiography provides prognostic information in addition to that identified by resting left ventricular function and coronary anatomy in patients with medically treated coronary artery disease. Clinical follow-up (median, 21.7 months) was obtained in 424 medically treated patients who underwent exercise radionuclide angiography and coronary angiography. The mean age of the study population was 58 years, and 67% were men. Cardiac death occurred in 16 patients, nonfatal myocardial infarction in 16, and nonfatal out-of-hospital cardiac arrest in 1. Univariate analysis showed that multiple variables were associated with future cardiac events, including number of diseased vessels, exercise and rest radionuclide ejection fraction, history of myocardial infarction, exercise and rest left ventricular end-systolic and end-diastolic volume indices, peak exercise workload, age, abnormal resting electrocardiogram, and peak exercise ST-segment depression. Only three variables were independently associated with cardiac events on follow-up: number of diseased vessels, radionuclide ejection fraction at rest, and age. In patients with three-vessel disease and a resting radionuclide ejection fraction of more than 40%, a subgroup with higher risk could not be identified on the basis of exercise radionuclide response.

Stability of radionuclide left ventricular volume measurements

Left ventricular volume measurements are useful in the evaluation of cardiac function and are important in the long-term management of patients with various cardiac diseases. Although there are many methods of measuring left ventricular volumes, a non invasive and reproducible method relies on radionuclide techniques. The errors in estimation of left ventricular volumes have previously been well studied. To date there is little information on the reproducibility of left ventricular volume measurements made by this technique at different points in time. This study evaluated 61 patients with stable coronary artery disease over a period of approximately 1 year. All patients had two resting radionuclide gated blood pool studies. Patients had no changes in symptoms, electrocardiographic findings or medication between studies. Using +/- 2 SD as 95% confidence limits for a true change, an end diastolic volume index change greater than -34 ml m-2 and +38 ml m-2 or an end systolic volume index change greater than -24 ml m-2 and +26 ml m-2 are required to state with confidence that a change has occurred between two examinations. These data provide guidelines to assess whether interval changes in left ventricular volumes are real or are due to variations within the technique.

Noninvasive identification of severe coronary artery disease using exercise radionuclide angiography

The ability of exercise radionuclide angiography to predict the risk of having significant left main or three vessel coronary artery disease was examined in 681 patients who underwent both radionuclide and coronary angiography. There were significant differences in multiple variables between patients with or without such disease. Logistic regression analysis identified seven variables as independently predictive of the presence of left main or three vessel disease. Using these variables, low, intermediate and high probability groups could be identified. The four most important variables--the magnitude of exercise ST segment depression, peak exercise ejection fraction, peak exercise rate-pressure product and sex of the patient--can provide practical estimates of the risk of having left main or three vessel disease. Exercise radionuclide angiography can provide a clinically useful noninvasive estimate of the risk of having significant left main or three vessel disease.

Guidelines for the interpretation of the exercise radionuclide ventriculogram for diagnosing coronary artery disease

In 622 patients with known coronary artery anatomy, heart rate (HR).blood pressure (BP) product and left ventricular (LV) ejection fraction (EF) at maximal supine exercise measured by radionuclide ventriculography were used to estimate, by logistic regression analysis, the probabilities of absence of significant coronary artery disease (CAD), presence of significant CAD, presence of multivessel CAD and presence of 3-vessel CAD. Thus, for example, estimated probabilities of each of the aforementioned 4 categories of CAD are 0.39, 0.61, 0.32 and 0.12, respectively, for HR.BP product of 26,000 beats.mm Hg/min and LVEF of 0.6 at maximal exercise and 0.08, 0.92, 0.77 and 0.48, respectively, for HR.BP of 15,000 and LVEF of 0.4. The graphic presentations of these estimated probabilities form useful guidelines for interpreting the results of exercise radionuclide ventriculography. In addition, specific cutoff values at maximal exercise defined 2 groups: (HR.BP product greater than or equal to 21,000 beats.mm Hg/min and LVEF greater than or equal to 0.55) with a high (70%) likelihood of absence of significant CAD or 1-vessel CAD and a low (7%) likelihood of 3-vessel CAD, and (HR.BP product less than 21,000 and LVEF less than 0.55) with a high (72%) likelihood of multivessel CAD and a low (8%) likelihood of absence of CAD.

Exercise response of the systolic pressure to end-systolic volume ratio in patients with coronary artery disease

The exercise response of the ratio of systolic blood pressure to end-systolic volume was studied in 243 patients with chest pain and coronary artery disease who underwent supine rest and exercise equilibrium radionuclide angiography. There was a wide variation in both rest and exercise variables in this group. The exercise response of the systolic pressure/volume ratio also varied greatly, ranging from a decrease of 59% to an increase of 136%. Twenty-one clinical, catheterization and radionuclide angiographic variables were examined to determine their relation to the exercise response of the systolic pressure/volume ratio; nine variables were individually correlated with this ratio. Multiple regression analysis identified the change in end-diastolic volume index with exercise, rest systolic blood pressure, coronary artery Gensini score and peak work load as significant independent predictors of the exercise response of the systolic pressure/volume ratio; the latter correlated significantly with the change in ejection fraction with exercise (r = 0.73, p less than 0.0001). Its sensitivity for the detection of coronary artery disease in the study group (84%) and its "normalcy rate" in a group of 120 patients with a low likelihood of coronary artery disease (81%) were similar to those of the peak exercise ejection fraction (75 and 82%, respectively). These results demonstrate that the exercise response of the systolic pressure/end-systolic volume ratio is a complex response that is influenced by several pathophysiologic variables in the presence of coronary artery disease. It does not offer any advantage over ejection fraction measurements for the detection of exercise-induced ischemia.

Anatomic and functional significance of a hypotensive response during supine exercise radionuclide ventriculography

The significance of a decline in systolic blood pressure (BP) during supine exercise was examined in 820 patients who underwent both supine exercise gated equilibrium radionuclide ventriculography and coronary angiography. Twenty-seven patients, 3% of the study population, had a decrease in systolic BP at peak exercise of more than 10 mm Hg from the systolic BP at rest. Other indicators of ischemia--angina, ST-segment depression, a decrease in ejection fraction and wall motion abnormality during exercise--were present frequently but not uniformly in these patients. Although most patients had a decline in ejection fraction and a new wall motion abnormality with exercise, 4 patients had an increase in ejection fraction with exercise without any regional wall motion abnormalities. Coronary angiography in the 27 patients with systolic hypotension demonstrated severe coronary artery disease (CAD). Twenty-two patients (81%) had 3-vessel or left main CAD. Twenty of these 22 patients with 3-vessel CAD had at least 2 arteries with 90% or more diameter stenoses. Systolic hypotension during supine exercise radionuclide angiography is infrequent, usually associated with evidence of global and regional left ventricular dysfunction, and a marker of very severe CAD.

Cardiac involvement in Lyme disease: manifestations and management

Cardiac involvement in Lyme disease may manifest as atrioventricular block, myopericarditis, and left ventricular dysfunction. Diagnosis depends on recognition of the systemic nature of Lyme disease, including cardiac involvement, and its natural history. Serologic tests that are both sensitive and specific may aid in diagnosis. Although current recommendations for the treatment of Lyme disease with carditis include antibiotics and salicylates or corticosteroids, these types of therapy have not been unequivocally demonstrated to alter the natural history of cardiac involvement. Supportive therapy may necessitate temporary transvenous cardiac pacing in symptomatic patients.

Exercise radionuclide ventriculography in evaluation of coronary artery disease

The ability of radionuclide variables obtained at rest and at peak exercise to discriminate the number of stenosed (greater than or equal to 70% luminal diameter narrowing) major coronary arteries was evaluated in 296 patients undergoing supine exercise radionuclide ventriculography. Stepwise linear discriminant analysis of the data from the first 200 patients identified a significant (p less than 0.001) discriminatory combination. Application of this function to the remaining 96 patients provided correct classification of arteriographically determined zero, one, two, and three stenosed arteries in 59%, 18%, 14%, and 60% of cases, respectively. The discriminant function classified minimal stenoses (zero or one artery) and multivessel stenoses (two or three arteries) correctly by arteriography in two thirds of cases in each group. Arteriographic presence of three stenoses was unlikely in those classified as having no stenosis, and absence of stenosis was rare in those classified as having three stenoses. Exercise radionuclide ventriculography is most helpful in identifying minimal and multivessel coronary disease rather than number of stenosed major coronary arteries.

Reproducibility of noninvasive tests of peripheral occlusive arterial disease

We studied the reproducibility of four tests of peripheral occlusive arterial disease in 54 subjects, 32 of whom had this disease. We found that the reproducibility of systolic blood pressures obtained at rest from the thighs, calves, and ankles approximated that of arm systolic and diastolic blood pressures, as did the ankle-to-arm systolic blood pressure ratios. The average of the tenth and ninetieth percentile ranges of the resting systolic blood pressure ankle-to-arm ratios was +/- 0.10. Systolic blood pressures from the fingers were somewhat less reproducible, and those from the toes were even more variable. Systolic blood pressure ankle-to-arm ratios measured after the patient had exercised were less reproducible than resting ratios. The average of the tenth and ninetieth percentile ranges of the 1-, 3-, 5-, and 10-minute ratios after exercise was -0.13 to +0.16. Skin temperatures from the fingers and toes were approximately as reproducible as systolic blood pressures from the arms and legs and as the resting ankle-to-arm blood pressure ratios. Pulse-volume recordings from the thighs, calves, ankles, feet, toes, and fingers were very poorly reproducible. We conclude that information on the reproducibility of these measurements is essential in the evaluation of noninvasive arterial tests that are used to determine the course of peripheral occlusive arterial disease.

Cardiovascular hemodynamics of bicycle and handgrip exercise in normal subjects before and after administration of propranolol

Radionuclide angiography was used to study the effects of supine and upright bicycle exercise and handgrip exercise in 17 (12 well-trained) normal subjects before (control) and immediately after the administration of propranolol (160 mg/day for 4 days). Cardiac hemodynamic values were related to position in that control left ventricular volumes and the cardiac index were greater in the supine position than in the upright at rest but resting left ventricular ejection fraction was similar in both positions. The pressure volume index was greater in the upright position than in the supine. At maximal exercise before treatment, however, similar cardiovascular hemodynamic measurements were recorded in both positions. Propranolol increased left ventricular end-diastolic volume at rest and at maximal exercise. Left ventricular end-systolic volume, however, was substantially greater only in the upright position both at rest and at maximal exercise when compared with control values. Heart rate, systolic arterial pressure, cardiac index, and pressure volume index were decreased at rest and maximal exercise after treatment with propranolol. Ejection fraction was decreased in the upright position after propranolol administration but was unchanged in the supine position. Handgrip exercise primarily increased heart rate and arterial pressure and did not affect cardiac volume, and this response was unaffected by propranolol.

Effect of dipyridamole and aspirin on late vein-graft patency after coronary bypass operations

To study the prevention of occlusion of aortocoronary-artery bypass grafts, we concluded a prospective, randomized, double-blind trial comparing long-term administration of dipyridamole (begun two days before operation) plus aspirin (begun seven hours after operation) with placebo in 407 patients. Results at one month showed a reduction in the rate of graft occlusion in patients receiving dipyridamole and aspirin. At vein-graft angiography performed in 343 patients (84 per cent) 11 to 18 months (median, 12 months) after operation, 11 per cent of 478 vein-graft distal anastomoses were occluded in the treated group, and 25 per cent of 486 were occluded in the placebo group. The proportion of patients with one or more distal anastomoses occluded was 22 per cent of 171 patients in the treated group and 47 per cent of 172 in the placebo group. All grafts were patent within a month of operation in 94 patients in the placebo group and 116 patients in the treated group; late development of occlusions was reduced from 27 per cent in the placebo group to 16 per cent in the treatment group. The results show that dipyridamole and aspirin continue to be effective in preventing vein-graft occlusion late after operation, and we believe that such treatment should be continued for at least one year.