Stephen L. Bacharach

Partial-Volume Effect in PET Tumor Imaging

PET has the invaluable advantage of being intrinsically quantitative, enabling accurate measurements of tracer concentrations in vivo. In PET tumor imaging, indices characterizing tumor uptake, such as standardized uptake values, are becoming increasingly important, especially in the context of monitoring the response to therapy. However, when tracer uptake in small tumors is measured, large biases can be introduced by the partial-volume effect (PVE). The purposes of this article are to explain what PVE is and to describe its consequences in PET tumor imaging. The parameters on which PVE depends are reviewed. Actions that can be taken to reduce the errors attributable to PVE are described. Various PVE correction schemes are presented, and their applicability to PET tumor imaging is discussed.

Real-time radionuclide cineangiography in the noninvasive evaluation of global and regional left ventricular function at rest and during exercise in patients with coronary-artery disease.

Although coronary angiography defines regions of potential ischemia in patients with coronary-artery disease, accurate assessment of the presence and functional importance of ischemia requires appraisal of regional and global left ventricular function during stress. To perform such assessment, we developed a noninvasive real-time radionuclide cineangiographic procedure permitting continuous monitoring and analysis of left ventricular function during exercise. In 11 patients with coronary disease who had normal regional and global ventricular function at rest, new regions of dysfunction developed during exercise (P less than 0.001), and in 10, global ejection fraction dropped 7 to 47 per cent. Fourteen age-matched normal subjects were studied; during exercise none had regional dysfunction, and each increased global ejection fraction (average increase, 23 +/- 3 per cent [+/-S.E.], P less than 0.001 as compared with patients with coronary disease). Radionuclide cineangiography during exercise permits accurate assessment of the presence and functional severity of ischemic heart disease.

Impaired left ventricular diastolic filling in patients with coronary artery disease: assessment with radionuclide angiography.

To assess left ventricular (LV) diastolic filling at rest in patients with coronary artery disease (CAD), we analyzed high-resolution time-activity curves (10-20 msec/frame) obtained from gated radionuclide angiograms in 231 patients. Peak LV filling rate (PFR), expressed in end-diastolic volumes per second (EDV/sec), was subnormal in CAD patients (1.8 +/- 0.6 [+/- SD] vs normal mean of 3.3 +/- 0.6, p les than 0.001) and time to PFR (TPFR), measured from end-systole to PFR, was prolonged (171 +/- 41 msec vs normal mean of 136 +/- 23 msec, p less than 0.001). These indexes were also abnormal in the 141 patients with normal resting LV ejection fraction (PFR = 2.1 +/- 0.5 EDV/sec; TPFR = 175 +/- 36 msec) and in 123 patients without Q waves on the ECG (PFR = 2.1 +/- 0.5 EDV/sec; TPFR = 168 +/- 38 msec). Abnormal LV filling at rest (PFR less than 2.5 EDV/sec or TPFR greater than 180 msec) was found in 91% of all patients with CAD, 86% of patients with normal resting LV ejection fractions, 85% of patients without Q waves, and 82% of patients with normal resting LV ejection fraction, no resting regional wall motion abnormalities and no Q waves. Thus, LV diastolic filling, evaluated noninvasively by radionuclide angiography, is abnormal in a high percentage of patients with CAD at rest independent of LV systolic function or previous myocardial infarction.

Effects of verapamil on left ventricular systolic function and diastolic filling in patients with hypertrophic cardiomyopathy.

Verapamil improves exercise capacity in patients with hypertrophic cardiomyopathy (HCM), but its mechanisms of action are unknown. We examined the effects of oral verapamil (320–480 mg/day) on resting left ventricular (LV) systolic and diastolic function in patients with HCM. High‐temporal‐resolution time‐activity curves from gated technetium‐99m radionuclide angiograms were analyzed before and after verapamil therapy in 40 patients, of whom 16 were also studied during propranolol therapy (80–960 mg/day). All but one patient had normal or supranormal systolic function, but 70% had evidence of diastolic dysfunction, defined as peak LV filling rate (PFR) < 2.5 end‐diastolic volumes (EDV)/sec or time to PFR > 180 msec. Verapamil did not change LV ejection fraction, peak ejection rate or ejection time, but did increase PFR (control 3.3 ± 1.0 EDV/sec, verapamil 4.1 ± 1.1 EDV/sec; p < 0.001) and reduce time to PFR (control 187 ± 56 msec, verapamil 159 ± 34 msec; p < 0.001). Only 30% of patients had evidence of diastolic dysfunction during verapamil. In contrast, propranolol did not change LV ejection fraction, PFR or time to PFR, but did prolong ejection time and reduce peak ejection rate. Thus, LV diastolic filling is abnormal in a high percentage of patients with HCM, and verapamil normalizes or improves these abnormalities without altering systolic function. This mechanism may contribute to the clinical improvement of many HCM patients during verapamil therapy.

Sensitivity, specificity and predictive accuracy of radionuclide cineangiography during exercise in patients with coronary artery disease. Comparison with exercise electrocardiography.

Noninvasive radionuclide cineangiography permits the assessment of global and regional left ventricular function during intense exercise. To assess the sensitivity of the technique in detecting coronary artery disease, we studied 63 consecutive patients with ≥ 50% stenosis of at least one coronary artery. Fiftynine (94%) had regional dysfunction with exercise; 56 (89%) developed lower-than-normal ejection fractions during exercise. When both regional dysfunction and subnormal ejection fractions are considered together, the sensitivity was 95%. Each patient also underwent exercise electrocardiography to either angina or 85% of predicted maximal heart rate. Of the 42 patients who developed angina during exercise electrocardiography, 26 (62%) developed ≥1 mm ST-segment depression; four additional patients (10%) had Q waves diagnostic of previous myocardial infarction. In contrast, 39 (93%, p < 0.001) developed regional dysfunction during radionuclide study, and one additional patient developed a subnormal ejection fraction without regional dysfunction. To assess specificity, we studied 21 consecutive patients with chest pain who had normal coronary arteries. None developed regional dysfunction; ejection fraction increased in all to levels within the range previously defined as normal. The predictive accuracy in this symptomatic population was 100%. We conclude that radionuclide cineangiography is highly sensitive (more so than exercise electrocardiography), predictive and specific in detecting patients with coronary artery disease.

Myocardial viability in patients with chronic coronary artery disease. Comparison of 99mTc-sestamibi with thallium reinjection and [18F]fluorodeoxyglucose.

Background99mTc-sestamibi and thallium imaging have similar accuracy when used for diagnostic purposes, but whether sestamibi provides accurate information regarding myocardial viability in patients with chronic coronary artery disease has not been established. Since there is minimal redistribution of sestamibi over time, it may overestimate nonviable myocardium in patients with left ventricular dysfunction, in whom blood flow may be reduced at rest. Methods and ResultsWe studied 54 patients with chronic coronary artery disease with a mean ejection fraction of 34±14%. Patients underwent stress/redistribution/reinjection thallium tomography and, within a mean of 5 days, same-day rest/stress sestamibi imaging using the same exercise protocol and with patients achieving the same exercise duration. Of the 111 reversible thallium defects on either the redistribution or reinjection study, 40 (36%) were determined to be irreversible on the rest/stress sestamibi study, whereas only 3 of 63 irreversible thallium defects despite reinjection (5%) were classified to be reversible by sestamibi imaging. The concordance regarding reversibility of myocardial defects between thallium stress/redistribution/reinjection and same day rest/ stress sestamibi studies was 75%. A subgroup of 25 patients also underwent positron emission tomography (PET) studies with 15O-labeled water and [18F]fluorodeoxyglucose (FDG) at rest after an oral glucose load. As in the overall group of 54 patients, there was concordance between thallium and sestamibi imaging regarding defect reversibility in 51 of 73 regions (70%). In the remaining 22 discordant regions (30%), 18 (82%) appeared irreversible by sestamibi imaging but were reversible by thallium imaging. Myocardial viability was confirmed in 17 of 18 regions, as evidenced by normal FDG uptake (10 regions) or FDG/blood flow mismatch (7 regions) on PET. These regions were present in 16 of the 25 patients studied (64%). We then explored methods to improve the sestamibi results. First, when the 18 discordant regions with irreversible sestamibi defects were further analyzed according to the severity of defects, 14 (78%) demonstrated only mild-tomoderate reduction in sestamibi activity (51% to 85% of normal activity), suggestive of predominantly viable myocardium, and the overall concordance between thallium and sestamibi studies increased to 93%. Second, when an additional 4-hour redistribution image was acquired in 18 patients after the injection of sestamibi at rest, 6 of 16 discordant irreversible regions (38%) on the rest/stress sestamibi study became reversible, thereby increasing the concordance between thallium and sestamibi studies to 82%. ConclusionsThese data indicate that same-day rest/stress sestamibi imaging will incorrectly identify 36% of myocardial regions as being irreversibly impaired and nonviable compared with both thallium redistribution/reinjection and PET. However, the identification of reversible and viable myocardium can be greatly enhanced with sestamibi if an additional redistribution image is acquired after the rest sestamibi injection or if the severity of reduction in sestamibi activity within irreversible defects is considered.

Verapamil-induced improvement in left ventricular diastolic filling and increased exercise tolerance in patients with hypertrophic cardiomyopathy: short- and long-term effects.

Verapamil improves exercise tolerance and decreases symptoms in many patients with hypertrophic cardiomyopathy. The mechanisms responsible for these effects are not completely understood, although previous studies indicate that verapamil enhances left ventricular relaxation and diastolic filling in such patients. To investigate the association between changes in left ventricular filling and exercise tolerance after verapamil, we studied 55 patients with hypertrophic cardiomyopathy by radionuclide angiography and graded treadmill testing before and after 1 to 4 weeks of therapy with orally administered verapamil, 320 to 640 mg/d. The verapamil-induced increase in peak left ventricular filling rate at rest (from 3.1 +/- 1.3 to 3.7 +/- 1.3 end-diastolic volumes/sec; p less than .001) was associated with an increase in exercise tolerance (from 5.9 +/- 3.6 to 8.7 +/- 4.7 min; p less than .001); exercise capacity increased in 34 of 43 patients (79%) manifesting an increase in peak filling rate but only one of 12 patients (8%) with unchanged or decreased peak filling rate (p less than .001). This initial trend persisted in 25 patients studied after 1 year of therapy; 11 of 16 patients (69%) with a persistent increase in peak filling rate had persistent improvement in exercise tolerance relative to preverapamil values, compared with only one of nine patients (11%) in whom peak filling rate was unchanged or decreased relative to preverapamil levels (p less than .02). Verapamil withdrawal after 1 to 2 years in 24 patients resulted in reduction in peak filling rate (p less than .001) and was associated with deterioration in exercise tolerance in 17 patients (71%). Hence, verapamil-induced changes in left ventricular peak filling rate were associated significantly with objective symptomatic improvement. These data support the concept that enhanced left ventricular diastolic filling is an important mechanism contributing to the clinical improvement experienced by many patients with hypertrophic cardiomyopathy during therapy with verapamil.

Myocardial perfusion abnormalities in patients with hypertrophic cardiomyopathy: assessment with thallium-201 emission computed tomography.

Myocardial ischemia may play a critical role in the symptomatic presentation and natural history of hypertrophic cardiomyopathy (HCM). To assess the relative prevalence and functional significance of myocardial perfusion abnormalities in patients comprising the broad clinical spectrum of HCM, we studied 72 patients (ages 12 to 69 years, mean 40) using thallium-201 emission computed tomography. Imaging was performed immediately after maximal exercise and again after a 3 hr delay. Regional perfusion defects were identified in 41 of the 72 patients (57%). Fixed or only partially reversible defects were evident in 17 patients, 14 of whom (82%) had left ventricular ejection fractions of less than 50% at rest. Twenty-four patients demonstrated perfusion defects during exercise that completely reversed at rest; all had normal or hyperdynamic left ventricular systolic function (ejection fraction greater than or equal to 50%). Perfusion abnormalities were present in all regions of the left ventricle. However, the fixed defects were observed predominantly in segments of the left ventricular wall that were of normal or only mildly increased (15 to 20 mm) thickness; in contrast, a substantial proportion (41%) of the completely reversible defects occurred in areas of moderate-to-marked wall thickness (greater than or equal to 20 mm, p less than .001). Neither a history of chest pain nor its provocation with treadmill exercise was predictive of an abnormal thallium study, since regional perfusion defects were present in 10 of 18 (56%) completely asymptomatic patients, compared with 31 of 54 (58%) symptomatic patients. These data indicate that myocardial perfusion abnormalities occur commonly among patients with HCM. Fixed or only partially reversible defects suggestive of myocardial scar and/or severe ischemia occur primarily in patients with impaired systolic performance. Completely reversible perfusion abnormalities occur predominantly in patients with normal or supranormal left ventricular systolic function. Such dynamic changes in regional thallium activity may reflect an ischemic process that contributes importantly to the clinical manifestations and natural history of HCM.

Asynchronous left ventricular regional function and impaired global diastolic filling in patients with coronary artery disease: reversal after coronary angioplasty.

Left ventricular diastolic filling is impaired in many patients with coronary artery disease and normal left ventricular systolic function, and is improved in many patients after coronary angioplasty (PTCA). To investigate the mechanisms for this improvement, we studied regional asynchrony by radionuclide angiography in 26 patients with single-vessel coronary artery disease before and after successful PTCA. Before PTCA, all patients had normal ejection fractions at rest and normal qualitative left ventricular regional wall motion, as determined by radionuclide and contrast angiography. Quantitative left ventricular regional function was assessed by dividing the left ventricular region of interest into 20 sectors. Phase analysis was performed on each sectors time-activity curve, and the average intersector phase difference was used as an index of left ventricular regional synchrony. Before PTCA, average intersector phase difference was increased compared with normal (6.0 +/- 2.2 vs 4.0 +/- 1.7 degrees, p less than .005), indicating asynchronous regional function. After PTCA, ejection fraction at rest was unchanged, but peak left ventricular filling rate at rest increased from 2.5 +/- 0.6 to 3.0 +/- 0.6 end-diastolic volume/sec (p less than .001) and was associated with a decrease in average intersector phase difference from 6.0 +/- 2.2 to 5.1 +/- 2.3 degrees (p less than .05). Average intersector phase difference decreased in 16 of 21 patients in whom peak filling rate increased after PTCA (p less than .005), compared with one of five patients in whom peak filling rate was unchanged or decreased. Hence, improved global left ventricular filling after PTCA was associated with more synchronous left ventricular regional behavior. To identify the cause of regional asynchrony before PTCA, we then generated time-activity curves from each of four left ventricular quadrants. These data indicated that the asynchrony was caused by regional variation in timing of diastolic rather than systolic events and that PTCA resulted in reduction in regional diastolic asynchrony. These data suggest that in many patients with coronary artery disease and normal left ventricular systolic function, impaired global diastolic filling may result from asynchronous left ventricular regional diastolic function, which is a reversible manifestation of myocardial ischemia or reduced coronary flow.

Exercise-induced left ventricular dysfunction in symptomatic and asymptomatic patients with aortic regurgitation: assessment with radionuclide cineangiography.

In patients with aortic regurgitation,, left ventricular dysfunction at rest, which is associated with a poor long-term prognosis, often develops before severe symptoms. To determine whether evidence of left ventricular dysfunction could be detected before it appeared at rest, 43 patients with severe aortic regurgitation were studied using radionuclide cineangiography during exercise. In 30 normal subjects, left ventricular ejection fraction increased during exercise (57 +/- 1 percent [mean +/- standard error] at rest, 71 +/- 2 percent during exercise, P less than 0.001). In contrast, among 21 symptomatic patients, ejection fraction was normal at rest in 14 patients (average 47 +/- 2 percent) but normal during exercise in only one patient (average 38 +/- 2 percent, P less than 0.001). Ejection fraction was normal at rest in 21 of 22 asymptomatic patients (average 62 +/- 2 percent) but was normal during exercise in only 13 (average 57 +/- 3 percent, P less than 0.001). Thus, exericse-induced left ventricular dysfunction can precede symptoms and dysfunction at rest. Radionuclide assessment of left ventricular function during exercise may prove valuable in sequentially following the state of left ventricular function in patients before the onset of symptoms or of irreversible left ventricular failure.