Thallium-201 myocardial perfusion scintigraphy: Results of standard and multi-pinhole tomographic techniques

247

SUMMARY Visual interpretation of stress-redistribution thallium-201 (201T1) scintigrams is subject to observer variability and is suboptimal for evaluation of extent of coronary artery disease (CAD). An objective, computerized technique has been developed that quantitatively expresses the relative space-time myocardial distribution of 201TI. Multiple-view, maximum-count circumferential profiles for stress myocardial distribution of 201T1 and segmental percent washout were analyzed in a pilot group of 31 normal subjects and 20 patients with CAD to develop quantitative criteria for abnormality. Subsequently, quantitative analysis was applied prospectively to a group of 22 normal subjects and 45 CAD patients and compared with visual interpretation of scintigrams for detection and evaluation of CAD. The sensitivity and specificity of the quantitative technique (93% and 91%, respectively) were not significantly different from those of the visual method (91% and 86%).The quantitative analysis significantly (p < 0.05) increased the sensitivity of 20'Tl imaging over the visual method in the left anterior descending artery (from 56% to 80%), left circumflex artery (from 34% to 63%) and right coronary artery (from 65% to 94%) without significant loss of specificity. Using quantitative analysis, sensitivity for detection of diseased vessels did not diminish as the number of vessels involved increased, as it did with visual interpretation. In patients with one-vessel disease, 86% of the lesions were detected by both techniques; however, in patients with three-vessel disease, quantitative analysis detected 83% of the lesions, while the sensitivity was only 53% for the visual method. Seventy percent of the coronary arteries with moderate (50-75%) stenosis were detected quantitatively, compared with 35% by the visual method.We conclude that this quantitative technique for analysis of stress-redistribution 201T1 scintigrams is objective and more sensitive than the visual method, especially in patients with multiple-vessel disease and those with moderate coronary artery stenosis.EXERCISE THALLIUM-20 1 (201T1) myocardial perfusion scintigraphy is a useful noninvasive method for detecting and evaluating patients with significant coronary artery disease (CAD). However, interpretation of 201T1 images is usually visual, and even when analyzed by experienced readers, there is substantial observer variability.' Furthermore, although the method is highly specific for localization of CAD, it has major limitations in sensitivity for detection of individual coronary stenoses.2 " The visual method is particularly limited in detection of individual coronary lesions in patients with multiple-vessel CAD2-6 and when the degree of stenosis is not severe.2 ' Finally, although the regional washout characteristics of the myocardium for 201T1 contain important diagnostic information, washout abnormalities are difficult to detect by visual inspection.We have developed a computerized technique that objectively expresses the relative distribution of 20...

Section: Methods Patient Population and Study Designmentioning

confidence: 99%

Improved noninvasive assessment of coronary artery disease by quantitative analysis of regional stress myocardial distribution and washout of thallium-201.

Maddahi¹,

Garcia²,

Berman³

et al. 1981

247

“…Following a 5 min monitored recovery period, patients were promptly taken to the imaging room. Scintigraphy was performed by the seven-pinhole tomographic technique described by Vogel et al15 16 All studies were performed by one experienced technologist. Acquisition was terminated after 750,000 counts were obtained, and delayed scintigraphy was performed 3 hr later.…”

Section: Methodsmentioning

confidence: 99%

“…Twelve sections, beginning 7.5 cm from the collimator and at nominal intervals of 1.25 cm, were reconstructed with the algorithms developed by Vogel et tivity at 6 degree intervals were generated for each radian, as described by Vogel et al13 16 The area of redistribution was computed automatically as the algebraic sum of the areas for the initial and 3 hr profiles (in arbitrary units of percent times degrees), as illustrated in figure 1. The 3 hr clearance rates were calculated by our previously published methods, which were modified for tomographic scintigrams from those of Garcia et al 8 This rate was calculated radian by radian by subtracting the 3 hr activity (in counts/min of acquisition time) from the initial activity (same units) and dividing the difference by the initial activity (same units).…”

Section: Methodsmentioning

confidence: 99%

Scintigraphic quantification of myocardial ischemia: a new approach.

Massie¹,

Hollenberg²,

Wisneski³

et al. 1983

This study was undertaken to develop a quantitative scintigraphic measurement of ischemia. We recorded 201TI scintigrams by the seven-pinhole tomographic technique immediately after exercise and 3 hr later in 15 normal subjects with a low likelihood of coronary disease and in 55 catheterized patients with chest pain. Circumferential profiles of the initial and 3 hr tracer distribution and of the 3 hr clearance rate were generated for each of three left ventricular sections. A circumferential profile of the 3 hr clearance rate (initial counts minus 3 hr counts divided by initial counts, expressed as percent) was also generated for each of these sections. A scintigraphic ischemic score (SIS) was then derived by summing for the three sections the area (in arbitrary units) between the exercise and 3 hr profiles and the area by which the clearance profile fell below the lower limits of normal for clearance derived from the normal subj'ects. This summed area was then normalized for the level of stress by dividing by the product of the exercise duration (in minutes) and the fraction of agepredicted maximum heart rate achieved. This SIS was above the 95% confidence limits derived from the normal subjects in 44 of 46 (96%) patients with the significant coronary disease and in only one of nine with less than a 50% obstruction. The SIS was 52 + 58, 233 + 220, 427 325, and 826 551 U (mean SD) for patients without coronary disease and for those with one-, two-, and' three-vessel disease, respectively. The intergroup differences were statistically significant, but there was considerable overlap among individual patients. More importantly, the SIS correlated significantly with a coronary arteriography score designed to reflect the potential for ischemia based on the coronary anatomy (r = .78, p < .001) and with an index of ischemia generated from the exercise electrocardiogram (r = .72, p < .001). These findings suggest that a continuous and quantitative scintigraphic index of myocardial ischemia can be derived from analysis of the postexercise distribution and clearance of 201T1. Such an index should be valuable in determining prognosis and choosing therapy for patients with coronary artery disease and in assessing their response to therapeutic interventions. Circulation 68, No. 4, 747-755, 1983. THALLIUM-201 (207T1) scintigraphy is widely used to detect coronary disease, usually'in conjunction with exercise testing. In addition, many groups have evaluated its ability to localize obstructed vessels and to estimate the extent of anatomic involvement.24 Unfortunately, the scintigraphic results with these latter applications have often been disappointing, at least in part because of the relative nature of the conventional interpretation criteria, which occasionally make it dif-

“…51 When this technique is applied to circumferential profile analysis, the lower limit of normal varies with anatomic location. Using this approach with "minor editing," Vogel et al 18 found that sevenpinhole 2''Tl tomography was 94% sensitive and 96% specific in detecting CAD. Using 2 standard deviations for the lower limit of normal, we found a sensitivity of 86% and a specificity of 50%.…”

Section: Quantitation Of Perfusion Scintigramsmentioning

confidence: 99%

Tomographic thallium-201 myocardial perfusion scintigrams after maximal coronary artery vasodilation with intravenous dipyridamole. Comparison of qualitative and quantitative approaches.

Francisco¹,

Collins²,

Go³

et al. 1982

121

SUMMARY Eighty-six patients had thallium-201 (20'Tl) myocardial perfusion scintigrams after intense coronary artery dilation with i.v. dipyridamole. The dipyridamole was infused at 0.15 mg/kg/min for 4 minutes. Significant side effects were infrequent. Coronary angiography showed that 24 of our 86 patients had normal coronary vessels or -30% luminal obstruction in one coronary artery; 11 patients had a moderate coronary obstruction (30-55% diameter narrowing or systolic compression) in one major vessel; and 51 patients had significant coronary artery disease (CAD) (G 70% luminal obstruction in one or more major vessels). Tomographic and planar 201T1 scintigrams were obtained in each patient. Tomographic scintigrams were interpreted using quantitative or visual criteria; planar scintigrams were assessed using visual criteria only. From the tomographic images, "lower limit of normal" curves were derived from the normal group by computing a "mean minus N standard deviations" curve and by computing a range curve. Using the mean minus standard deviations curve, we could not obtain a test with high (> 80%) sensitivity and specificity. The range approach produced better results. Inclusion of patients with moderate obstructions in the normal group significantly distorted the resulting lower limit of normal curve. When visual criteria were used, interobserver variability was 40% for tomographic scintigrams and 44% for planar scintigrams. In the 24 patients with normal or nonsignificant CAD, quantitative analysis of the tomograms (range approach) indicated that one of 24 (4%) had a positive image (specificity 96%); in contrast, when visual criteria were used to interpret the tomographic or planar 201TI scintigrams, eight of 24 (33%) had positive scintigrams (specificity 67%). In the 51 abnormal patients, the sensitivity of detecting CAD was 46 of 51 (90%) for tomographic scintigrams interpreted quantitatively, 39 of 51 (76%) for tomographic scintigrams interpreted visually and 41 of 51 (80%) for planar scintigrams assessed visually. The tomographic imaging procedure (quantitative interpretation) also demonstrated a high sensitivity (89%) and specificity (100%) in 28 patients (10 normal and 18 CAD), with a clinical diagnosis of unstable angina pectoris. Overall, the predictive accuracy of an abnormal scintigram with quantitative tomographic imaging (98%) was significantly better (p < 0.05) than either qualitative planar or pinhole imaging.THALLIUM-201 (201T1) perfusion scintigraphy is widely used in the diagnosis of coronary artery disease (CAD).'-8 Although it enhances our ability to detect CAD, three major problems are associated with conventional 201Tl perfusion images: interpretation of the scintigrams has usually involved qualitative visual criteria; exercise is required to induce coronary vasodilation;'-' and sensitivity (usually 65-90%) is not optimal.' 8 When 201T1 myocardial perfusion scintigrams are interpreted using qualitative criteria, many problems arise, including considerable interobserver variability,9 obs...