British Heart Journal, 1977, 39, 974-981
Assessment of diagnostic value of technetium-99m pyrophosphate myocardial scintigraphy in 80 patients with possible acute myocardial infarction1 WARREN F. WALSH, HARISCHANDRA B. KARUNARATNE, LEON RESNEKOV, HEIKO R. FILL, AND PAUL V. HARPER From the Departments of Medicine, Radiology, and Surgery, and the Franklin McLean Memorial Research Institute (operated by the University of Chicago for the U.S. Energy Research and Development Administration), the University of Chicago Pritzker School of Medicine, Chicago, Illinois, U.S.A.
The diagnostic value of technetium-99m-pyrophosphate (Tc-pyrophosphate) myocardial scintigraphy was determined in 80 consecutive patients who had been admitted to the coronary care unit in order to rule out an acute myocardial infarction. Scintigraphic findings obtained within 5 days of admission were correlated with the final cardiac diagnosis determined for each patient. Significant myocardial uptake of Tc-pyrophosphate (positive scans) occurred in 13 of 22 patients (59%) had who enzyme and/or electrocardiographic proven acute myocardial infarct: 3 out of 5 with transmural myocardial infarct, 9 of 16 with nontransmural myocardial infarct, and 1 patient with left bundle-branch block. Of 58 patients who showed no evidence of acute myocardial infarction, positive scans occurred in 14 of 33 patients who had unstable angina pectoris (42%), 0 of 6 who had congestive heart failure, 6 of 9 who had other acute cardiac syndromes, and in 0 of 10 who had noncardiac chest pain. In the patients with unstable angina pectoris positive scans could not be predicted on the basis of the history, electrocardiographic findings, or the arteriographically determnined severity of the coronary artery disease. Blood levels of Tc-99m activity measured in 21 cardiac patients and in 6 volunteers did not correlate with the uptake intensity of Tc-pyrophosphate. These findings suggest caution in the use of this imaging method for the diagnosis of acute myocardial infarct in patients admitted with 'rule out myocardial infarction'. In 1973 technetium-99m pyrophosphate myocardial scintigraphy was introduced by Bonte and coworkers as a noninvasive method of visualising acute myocardial infarction (Bonte et al., 1974; Parkey et al., 1974). Since that time there have been a number of clinical and experimental studies which have confirmed the uptake of labelled pyrophosphate by acutely infarcted myocardium (Fink/Bennett et al., 1974; Bonte et al., 1975; Buja et al., 1975; Willerson et al., 1975a). These studies suggest that
myocardial imaging using Tc-pyrophosphate may be of value in the investigation of patients admitted 'Supported in part by a USPHS contract (Myocardial Infarction Research Unit), SCOR-Ischemic Heart Disease Grant, USPHS training grant, USPHS Center for Imaging Research Grant, U.S. Energy Research Development Administration under ERDA Contract, the Fogarty International Fellowship Program, and the Chicago Heart Association. Received for publication 21 December 1976
to the coronary care unit in whom the diagnosis of acute myocardial infarction is uncertain.
Subjects and methods PATIENT POPULATION
Eighty patients (mean ages: men 50 years, women 30 years, whole group 57 years), who had been admitted to the coronary care unit (CCU) in order to rule out an acute myocardial infarction, were investigated. Seventy-six were admitted with chest pain thought to be cardiac in origin, and 4 after either syncope or an episode of acute dyspnoea. In all cases a definite cardiac diagnosis had not been established at the time of the initial imaging study. IMAGING PROCEDURE
Myocardial scintigraphy was performed between
974
975
Tc-99m-pyrophosphate myocardial scintigraphy 12 hours and 5 days after admission to the CCU (mean 48 hours). After obtaining informed consent, patients were imaged at their bedside using a mobile gamma camera (Searle Radiographics Pho-Gamma HP or IV) equipped with a 16 000 parallel hole, high resolution collimator, and interfaced to a mobile digital acquisition unit (Ohio Nuclear 150 System) with tape storage facilities. Fifteen millicuries of Tc-pyrophosphate (Mallincrodt/Nuclear) were injected intravenously, and 60 to 180 minutes later (mean 102 minutes) analogue and digital images containing 800 000 counts were collected in both the anteroposterior and left anterior oblique 30° projections. Analogue images were recorded on Polaroid film and the digital images stored on magnetic tape for subsequent image processing. No complications resulting from the administration of the Tc-pyrophosphate occurred in any of the patients studied. SCAN INTERPRETATION
Myocardial scans were interpreted by one of the authors (H.F.) who was unaware of the clinical findings or the presumed diagnosis. The intensity of the myocardial uptake of Tc-pyrophosphate was visually graded on a 0 to III scale: 0, being no visible uptake of the agent in the region of the heart; grade I, faint diffuse uptake; grade II, definite but moderate uptake, either localised or diffuse in distribution; and grade III, intense localised uptake approximately equal to the intensity of the sternum (Fig. 1). Because of doubt about the sig0
nificance of grade I uptake, only grade II and III were regarded as abnormal and as representing positive scans. MEASUREMENT OF 99mTC BLOOD ACTIVITY
In 21 cardiac patients and in 5 normal volunteers, blood samples were taken both at the beginning and at the end of the imaging procedure. All 26 subjects were imaged 120 minutes or more after injection. The blood samples were counted for 99 mTc activity, and using injection standards the percentage of the injected dose in the blood was calculated, which was normalised to the estimated whole blood volume of the patient. CARDIAC CATHETERISATION
Of the 80 patients, 28 underwent cardiac catheterisation during the study admission as part of the investigation of their chest pain. After informed consent was obtained, bilateral selective coronary arteriography using the Judkins technique and biplane left ventricular cineangiography were performed. The coronary angiograms and left ventriculograms were interpreted by at least two observers. The degree of obstruction of lesions in the left anterior descending, left circumflex, and right coronary arteries, and in their branches, was subjectively assessed, and left ventricular function, ejection fraction, and segmental contractility were defined.
I
II
I
AP
I I
LA
--.-- ---' Doutbtfu..-.-------..............- - --------Positive ----. bt-----I Dou -----Fig. 1 Grading system (O to III) of the intensity of myocardial uptake of Tc-99m pyrophosphate used for interpreting the myocardial scintigrams 2 to 4 hours after intravenous injection of the agent. Grades II and III were regraded as definitely positive.
NeqtiVe
-
Walsh, Karunaratne, Resnekov, Fill, and Harper
976 Acute MI
myocardial infarction. Eight had both electrocardiographic and enzymatic evidence of infarction, 7 had enzymatic evidence, and 1 patient, who was admitted 4 days after his main episode of chest pain, had electrocardiographic evidence only. Of tnese 16 patients, 7 had positive scans (44%), of whom 5 showed a localised uptake pattern and in 2 a diffuse uptake pattern occurred. Nine patients (56%) had negative or doubtful scans (grade 0-I). One patient, who had chronic left bundle-branch block and an enzyme rise indicative of infarction, had a grade III positive scan. Peak enzymes, electrocardiographic abnormalities, and the time interval between admission and scanning did not distinguish between those patients with positive scans and those with negative or doubtful scans (Table).
Unstable angina
01
50]
lSCHF 0,
Non.cardiac
Other cardiac
4
50-
UNSTABLE ANGINA PECTORIS
Tc PYP uptake grade (0-11)
Fig. 2
Thirty-three patients were diagnosed as having had unstable angina. Criteria for diagnosis were: (1) a history of typical cardiac pain occurring within 24 hours of admission, either recent in onset or a change from a previously stable pattern, (2) normal levels of CK and LDH enzymes, and (3) no electrocardiographic evidence of recent myocardial infarction. The mean age of the 33 patients was 59 years. There were 22 men and 11 women. Twelve patients had transient ST and/or T-wave changes
Distribution of Tc-99m pyrophosphate uptake
grades (0 to III) in 22 patients with acute myocardial infarction, 33 patients with unstable angina pectoris, 6 patients with congestive heart failure, 9 patients with other acute cardiac syndromes, and 10 patients with
noncardiac
chest pain. TMI= transmural
myocardial
infarction (5 of the 22 patients).
Results
The available findings and the results of all laboratory and angiographic investigations were reviewed to determine the definitive diagnosis of each patient. Based on this
information, the 80 patients were
Table Time interval before admission and study, electrocardiographic findings, peak CK level, and scan findings in 22 patients with acute myocardial infarction
divided into 5 clinical groups: (1) acute myocardial infarction,
(3) con-
(2) unstable angina pectoris,
gestive heart failure, (4) other acute cardiac syndromes,
and (5) noncardiac chest pain.
The scan
findings were correlated with the final clinical diagnosis (Fig. 2).
ACUTE
MYOCARDIAL
INFARCTION
Twenty-two patients (mean age 58 years; 15 men, 7 women) were diagnosed as having had acute myocardial
infarction
were the triad of:
(Table).
Criteria
for
diagnosis
(1) prolonged chest pain,
(2)
evolutionary ST and T wave changes on the electrocardiogram, and/or (3) a transient rise and fall in the serum levels of the enzymes creatinine kinase
(GK) and lactic dehydrogenase (LDH). Development of new deep Q waves of greater than 0 04 s duration indicated a transmural infarction. Five patients met the diagnostic criteria for acute
transmural
infarction,
of whom
3
had
grade
II
positive scans, and 2 had grade I scans. Sixteen patients sustained acute nontransmural
Case No. 1 2 3 4
5 6 7 8 9 10 11 12 13 14
15 16 17 18 19 20 21 22
Time Electrocardiogram interval (h) 96
Old ant/inf (T)
24 48 72 96 60 12 24 24 72 24 48 24 24 96 48 24 48 96 36 72 12
Old ant (T) Ant (NT) Ant (T) LBBB Inf (T)
Nonspecific Ant (NT) Ant (NT) Ant (NT)
Inf (T) Ant (NT) Inf (T) Ant (NT) Ant (NT) Ant (NT) Inf (T) Ant (NT) RBBB, LAFB Old ant (T) Inf (T) Nomual
Peak CK (units)
Scan grade
3240 640 265 359
III III II II II II II II II II II I
1580 374 680 940
50 760 428 544 60 1974 6900 2800 960 312 808 374 480 1060
I I I I I I 0 0 0 0
Ant, anterior; Inf, inferior; LBBB and RBBB, left and right bundlebranch block; LAFB, left anterior fascicular block; (T), transmural; (NT) nontransmural.
977
Tc-99m-pyrophosphate myocardial scintigraphy LAO CB
MT
b
MB c
d
Fig. 3 Positive anteroposterior and left anterior oblique scintigrams occurring in patients with (a) acute nontransmural myocardial infarction, (b) unstable angina pectoris, (c) angina pectoris with normal coronary arteries, and (d) chest pain and mitral valve prolapse.
positive scans, of which 2 remained grade II positive 48 hours later, and 2 showed grade I uptake, 6 and 8 days, respectively, after the initial study. One of these latter 2 patients had undergone myocardial revascularisation surgery during the interval between scans. In view of the intriguing occurrence of positive scans in a high proportion of patients with unstable angina pectoris, the clinical data of the 33 patients were examined retrospectively to ascertain whether there were any specific features that might have predicted the occurrence of a positive scan. There were no features in the history and no specific electrocardiographic abnormalities which correlated with the presence of a positive scan. In order to determine if the significant uptake of Tc-pyrophosphate was related to the severity of the coronary artery disease, a study was made of the coronary arteriographic findings in the 22 patients (67%) who underwent cardiac catheterisation during the acute phase of unstable angina pectoris. Fig. 4 shows that while the majority of these patients had significant two or three vessel coronary artery disease, there was no relation between the number of vessels with obstructive disease of greater than 70 per cent, and the occurrence of a positive scan, assuming that grade 0 and grade I do not represent any significant myocardial uptake. CONGESTIVE HEART FAILURE
Six patients (mean age 69; 3 women, 3 men) were diagnosed as having had an exacerbation of cardiac failure, with no evidence of recent myocardial infarction. Five had underlying coronary artery disease, and one a primary congestive cardiomyopathy. Four of these patients had zero uptake of Tc-pyrophosphate and 2 had grade I uptake.
65, c
during pain, 11 had stable nonspecific electrocardiographic abnormalities, 7 had electrocardiographic evidence of old transmural myocardial infarction and 3 patients had normal electrocardio-
a
C.
4--1
EII-m LJ O-I
3-
I-
0
6
z
2
grams.
Despite the absence of evidence of recent infarcIL 0 3V tion, 14 patients (42%) had positive scans, of whom 2V Iv ov No. of coronary arteries > 700% stenosis 8 had a localised uptake pattern and 6 had a more diffuse pattern (Fig. 3). Fig. 4 Relation between the angiographically determined Seven patients with unstable angina pectoris had severity of the coronary artery disease and the uptake follow-up scans performed 1 to 8 days after the first grades (O to III) in 22 patients with unstable angina study. Four of these patients initially had grade II pectoris. V=coronary artery.
978
Walsh, Karunaratne, Resnekov, Fill, and Harper
OTHER CARDIAC SYNDROMES
NONCARDIAC CHEST PAIN
Nine patients comprised this clinical gr(oup (mean age 53 years; 5 men and 4 women). Thr ee patients had angiographically documented mitral valve prolapse with normal coronary arteries, of whLom 1 had a grade III positive scan, 1 a grade I, and the third a grade 0 scan. Three female patients who presented with typical angina pectoris had normal epicardial coronary arteries, as shown by selectivce coronary arteriography. Of these 3 patients, 2 ha(d grade II positive scans, including 1 patient who hsad a positive treadmill exercise test and whose m yocardium produced lactate when subjected to thee stress of atrial pacing. Two patients who presente a witn recurrent ventricular arrhythmias withoiut cardiac pain had grade II positive scans. One hlad undergone electroversion (250 joules) 24 hours before the imaging procedure. The sixth patient who had ulcerative colitis with a probable toxic nayocarditis had a negative scan.
Ten patients (mean age 44 years; 6 women, 4 men) were admitted with chest pain, whose symptoms upon careful review of the history and all investigations were not considered to be cardiac in origin. In 7 the pain was thought to be musculoskeletal. The other 3 patients were diagnosed as reflux oesophagitis, biliary colic, and viraemia, respectively. Six patients in this noncardiac group had zero uptake scans and 4 had grade I scans. Thus no patient with non-cardiac chest pain had a definitely positive scan.
I ___
0
22
_
o Normal
* Cardiac
20 E
la~ 8
0
0
b
0
0
0 -0
0
0
E 14 0^
12
uI 12-
E
._
0 o
0
0 0 0
0 I II mI Tc PYP uptake grade Fig. 5 Relation between the uptake g,rade of Tc-99m pyrophosphate and the blood levels of;Tc-99m activity at the time of imaging in 21 patients upith cardiac disease
and in 5 normal volunteers.
The relation between the visually graded uptake of Tc-pyrophosphate and the residual 99mTc blood pool activity at the time of the imaging study was investigated in 21 cardiac patients and in 5 normal control subjects. Fig. 5 shows the residual blood activity for each of the 26 subjects classified according to the scan uptake grade reading. These measurements were all made at least 120 minutes after administration of Tc-pyrophosphate in order to allow time for adequate clearance of the agent from the circulation. The mean percentage of the injected dose per whole blood volume for the patients with grade 0 scans was 12-2 per cent (±0 9 SE); grade I, 13-8 per cent ( +1-5); grade II, 16-4 per cent (±1-2), and grade III, 13-8 per cent (±2 4). Using a one-way analysis of variance no significant differences in blood pool activity were found between the two groups. Frequently subjects with 0 or grade I scans had higher blood pool levels of 99 mTc activity than patients with grade II activity. Thus, no qualitative or quantitative relation could be shown between the uptake grade of Tc-pyrophosphate and the blood pool activity.
Discussion
0
0 0 C IV
RELATION BETWEEN SCAN GRADE AND BLOOD ACTIVITY OF 99mTc
Both animal and clinical studies have shown that the bone scanning agent technetium-99m-pyrophosphate concentrates in acutely infarcted myocardium and can be imaged with a standard scintillation camera (Fink/Bennett et al., 1974; Bonte et al., 1975; Buja et al., 1975; Willerson et al., 1975a). Morphological animal studies have shown that uptake of labelled pyrophosphate, being in part flow dependent, tends to be confined to the more peripheral segments of an acute infarct where residual perfusion is greater (Buja et al., 1975; Zaret et al., 1975). Little uptake of Tcpyrophosphate can be detected in the adjacent histologically normal, but presumably ischaemic, tissue surrounding an infarct. The time course and
Tc-99m-pyrophosphate myocardial scintigraphy the distribution of pyrophosphate uptake has been observed to parallel the accumulation of ionised calcium by necrotic tissue (Buja et al., 1975, 1976). It has, therefore, been suggested that pyrophosphate is fixed intracellularly by binding with mitochondrial calcium deposits (Buja et al., 1975). The experimental data suggest that Tc-pyrophosphate myocardial scintigraphy may provide a specific diagnostic test for acute myocardial infarction and hence be a useful diagnostic aid in the coronary care unit. The initial clinical studies of Tc-pyrophosphate myocardial imaging concluded that this method had a high degree of diagnostic specificity and sensitivity for acute myocardial infarction (Fink/Bennett et al., 1974; Parkey et al., 1974; Willerson et al., 1975a). Willerson et al. (1975a), in the largest clinical series so far reported, studied 202 patients admitted to the CCU and observed that 96 of 101 patients who had proven acute myocardial infarction had positive scintigrams. In contrast, only 9 of 101 patients who failed to show evidence of acute infarction had positive scans. More recently, there have been reports of positive scans being obtained in patients with unstable angina pectoris (Abdulla et al., 1976; Donsky et al., 1976), ventricular aneurysms (Ahmad et al., 1976), cardiomyopathy (Perez et al., 1976), and even as an incidental finding in those having bone scans performed (Soin et al., 1975). These findings prompted us to examine the diagnostic specificity and sensitivity of this scintigraphic method in a group of patients admitted to the CCU with a diagnosis of possible acute myocardial infarction. This clinical category was selected because it frequently presents a diagnostic problem in the CCU and might be an area in which Tc-pyrophosphate myocardial imaging might be of great diagnostic value within the first few days in hospital. The findings in the present study indicate that both the diagnostic sensitivity and specificity of Tcpyrophosphate myocardial imaging for acute myocardial infarct are lower than have been reported previously (Willerson et al., 1975a, b). The detection rate in the 22 patients with proven acute myocardial infarct was only 55 per cent. The reason for the apparent discrepancy between this and other series is probably related to the method of patient selection. Our study, by focusing on patients admitted as 'rule out myocardial infarct' effectively excluded those patients who had typical electrocardiographic changes of acute transmural myocardial infarction on admission, thereby selecting a higher proportion of patients with the more subtle diagnostic features observed in nontransmural myocardial infarction. Since nontransmural in-
979 farcts are usually smaller in size than transmural infarcts they tend to contain lesser amounts of tracer activity which may be insufficient to be detected externally by the scintillation camera (Botvinick et al., 1975). The diffuse, moderate uptake pattern of Tcpyrophosphate has been proposed as a specific diagnostic feature of acute nontransmural myocardial infarct (Poliner et al., 1976). Though this pattern was seen in some of our patients with myocardial infarction it did not, however, prove to be specific for nontransmural myocardial infarct, since it also occurred in patients without evidence of myocardial infarction. These limitations of Tcpyrophosphate myocardial imaging in the detection of acute nontransmural myocardial infarction need to be recognised, particularly since recent studies have shown that the mortality rate for patients with nontransmural myocardial infarction may be- similar to that for transmural myocardial infarction (Madias et al., 1974; Rigo et al., 1975). Of great interest was the demonstration that a significant number of patients with unstable angina pectoris had positive scans. Most of our patients in this group had grade II positive scans, suggesting that it might not be possible to distinguish scintigraphically between patients with acute myocardial infarction, especially if nontransmural, and those with unstable angina pectoris. This distinction has great clinical and therapeutic importance, since the management of these two groups of patients may differ. Many centres now believe that patients who have unstable angina pectoris should have early coronary angiography followed by myocardial revascularisation surgery, provided the coronary anatomy is suitable (Miller et al., 1973; Bonte et al., 1974; Bonchek et al., 1974). This therapeutic course would be contraindicated in patients with proven acute myocardial infarction. The explanation for the occurrence of positive scintigrams in some patients with unstable angina pectoris is unclear. Two possibilities exist. First, reversibly injured myocardial cells may be capable of taking up pyrophosphate in significant amounts. This hypothesis is difficult to test in the experimental animal, since no satisfactory animal model of unstable angina pectoris exists. Experimental myocardial uptake and distribution studies of Tcpyrophosphate have, by necessity, been performed only in acute myocardial infarction. The absence of the usual electrocardiographic or enzymatic markers of recent infarction, together with the known severity of the myocardial ischaemia in unstable angina pectoris, might tend to favour this first explanation. A second possible explanation is that Tcpyrophosphate myocardial imaging is more sensi-
980 tive than the currently available methods for the diagnosis of acute myocardial infarction and is, therefore, capable of detecting an infarct missed by the usual diagnostic methods. Recent necropsy studies of unstable angina pectoris lend some support to this hypothesis. Guthrie and co-workers (1975) reported pathological studies on patients with clinical unstable angina pectoris who died after coronary arterial surgery. In 6 cases, recent myocardial infarction, histologically antedating the time of surgery, was shown. In addition, acute myocardial infarction may not be detected in the presence of left bundle-branch block or previous transmural myocardial infarction if the patient is seen a few days after the main symptomatic event when the serum CK and LDH enzymes have returned to normal levels. Thus, the pathogenesis of positive Tc-pyrophosphate scans in patients with clinically unstable angina pectoris must remain uncertain at the present time. Since this study focused on determining the diagnostic specificity and sensitivity of Tc-pyrophosphate myocardial imaging when used within the first few days of being admitted to hospital, serial scans were not usually obtained. In the small number of patients with unstable angina pectoris who had more than one scan performed, the changes seen were variable. The scan findings tended to be stable during the first week, but became less intense in uptake during the second week, as has been previously noted in patients with acute myocardial infarction (Willerson et al., 1975a). Positive scans occurring in patients with either mitral valve prolapse or recurrent ventricular arrhythmias have not been previously reported. It could be suggested that the metabolic or haemodynamic abnormalities, which have been shown to occur in these syndromes, result in myocardial cell injury (Samet, 1973; Natarajan et al., 1975). These damaged cells may then be capable of taking up pyrophosphate. Direct current shock has been shown in animals to result in positive technetium pyrophosphate scans, and this may have been a factor in the one patient with ventricular arrhythmias treated by electroversion (DiCola et al., 1976; Pugh et al., 1976). In the animal experiments reported, one to four DC shocks of 320 joules each produced a significant uptake of pyrophosphate in the absence of any demonstrable change in regional blood flow (DiCola et al., 1976). The diffuse grade I scan was a nonspecific finding in this series being observed in patients with and without cardiac disease. The diffuse grade II scan was, on the other hand, seen only in patients with acute cardiac disease, most commonly nontransmural infarction or unstable angina pectoris. The
Walsh, Karunaratne, Resnekov, Fill, and Harper nonspecific nature of the diffuse scan pattern has recently been confirmed by Prasquier et al. (1976). They noted this appearance in 3 patients with acute nontransmural infarction, in 7 of 17 (41%) patients with unstable angina pectoris, and in 2 of 39 (5%) patients with stable angina pectoris. The diffuse pattern was also seen in 11 per cent of 484 patients who had routine bone scans; however, 79 per cent of this group had had prior left mastectomy. Whether this diffuse pattern represents true myocardial uptake or a cardiac blood pool image is still uncertain. The lack of correlation between blood pool activity of 99 mTc and the scan uptake in our study suggests that myocardial uptake of 99 mTc is responsible for the diffuse pattern. These measurements were made at least 2 to 4 hours after injecting Tc-pyrophosphate intravenously in order to minimise the possibility of blood pool interference. Blood clearance studies of Tc-pyrophosphate show a biexponential pattern of clearance from the blood. The first exponent has a relatively rapid clearance half-time of 13-6 minutes, and the second exponent a slower half-time of 380 minutes. It is believed that the rapid first exponent represents decreasing radioactivity caused by bone uptake, while the second exponent mainly reflects renal excretion of the agent (Krishnamurthy et al., 1975). Hence, blood pool images are more likely to be obtained in patients with impaired renal function or those imaged within approximately the first hour after injection. In addition, since the degree of tissue attentuation will vary from patient to patient, those with thin chest walls are more likely to have blood pool Tc-pyrophosphate images. From the analysis of the scans of 80 patients admitted to the CCU with possible myocardial infarction, we have shown that Tc-pyrophosphate myocardial scintigraphy: (1) may give positive scans in patients with unstable angina pectoris and other acute cardiac syndromes, and (2) may fail to detect acute myocardial infarction especially if nontransmural. Our results, therefore, suggest caution in the diagnostic application of Tc-pyrophosphate myocardial scintigraphic findings in patients admitted to hospital with 'rule out myocardial infarction'. References
Abdulla, A. M., Canedo, M. I., Cortez, B. C., McGinnis, K. D., and Wilhelm, S. K. (1976). Detection of unstable angina by 99m technetium pyrophosphate myocardial scintigraphy. Chest, 69, 168-173. Ahmad, M., Dubiel, J. P., Verdon, T. A., Jr., and Martin, R. H. (1976). Technetium 99m stannous pyrophosphate myocardial imaging in patients with and without left
ventricular aneurysm. Circulation, 53, 833-838. Bonchek, L. I., Rahimtoola, S. H., Anderson, R. P., McAnulty, J. A., Rosch, J., Bristow, J. D., and Starr, A.
Tc-99m-pyrophosphate myocardial scintigraphy (1974). Late results following emergency saphenous vein bypass grafting for unstable angina. Circulation, 50, 972-977. Bonte, F. J., Parkey, R. W., Graham, K. D., and Moore, J. G. (1975). Distribution of several agents useful in imaging myocardial infarcts. J'ournal of Nuclear Medicine, 16, 132-135. Bonte, F. J., Parkey, R. W., Graham, K. D., Moore, J., and Stokely, E. M. (1974). A new method for radionuclide imaging of myocardial infarcts. Radiology, 110, 473-474. Botvinick, E. H., Shames, D., Lappin, H., Tyberg, J. V., Townsend, R., and Parmley, W. W. (1975). Noninvasive quantitation of myocardial infarction with technetium 99m pyrosphophate. Circulation, 52, 909-915. Buja, L. M., Parkey, R. W., Dees, J. H., Stokely, E. M., Harris, R. A., Jr., Bonte, F. J., and Willerson, J. T. (1975). Morphologic correlates of technetium-99m stannous pyrophosphate imaging of acute myocardial infarcts in dogs. Circulation, 52, 596-607. Buja, L. M., Tofe, A. J., Parkey, R. W., Dees, J. H., Mukherjee, A., Bonte, F. J., and Willerson, J. T. (1976). Technetium pyrophosphate and diphosphonate uptake and calcium levels in canine myocardial infarcts (abstract). Clinical Research, 24, 211A. DiCola, V. C., Freedman, G. S., and Zaret, B. L. (1976). Effect of direct current countershock upon myocardial technetium-99m stannous pyrophosphate uptake (abstract). American J'ournal of Cardiology, 37, 131. Donsky, M. S., Curry, G. C., Parkey, R. W., Meyer, S. L., Bonte, F. J., Platt, M. R., and Willerson, J. T. (1976). Unstable angina pectoris. Clinical, angiographic and myocardial scintigraphic observations. British Heart Journal, 38, 257-263. Fink/Bennett, D., Dworkin, H. J., and Lee, Y. H. (1974). Myocardial imaging of the acute infarct. Radiology, 113, 449-450. Guthrie, R. B., Vlodaver, Z., Nicoloff, D. M., and Edwards, J. E. (1975). Pathology of stable and unstable angina pectoris. Circulation, 51, 1059-1063. Krishnamurthy, G. T., Huebotter, R. J., Walsh, C. F., Taylor, J. R., Kehr, M. D., Tubis, M., and Blahd, W. H. (1975). Kinetics of 99m Tc-labelled pyrophosphate and polyphosphate in man. Journal of Nuclear Medicine, 16, 109-115. Madias, J. R., Chahine, R. A., Gorlin, R., and Blacklow, D. J. (1974). A comparison of transmural and nontransmural acute myocardial infarction. Circulation, 49, 498-507. Miller, D. C., Cannom, D. S., Fogarty, T. J., Schroeder, J. S., Dailey, P. O., and Harrison, D. C. (1973). Saphenous vein coronary artery bypass in patients with 'preinfarction angina'. Circulation, 47, 234-241. Natarajan, G., Nakhjavan, F. K., Kahn, D., Yazdanfar, S., Sahibzada, W., Khawaja, F., and Goldberg, H. (1975). Myocardial metabolic studies in prolapsing mitral leaflet syndrome. Circulation, 52, 1105-1110. Parkey, R. W., Bonte, F. J., Meyer, S. L., Atkins, J. M.,
981 Curry, G. L., Stokely, E. M., and Willerson, J. T. (1974). A new method for radionuclide imaging of acute myocardial infarction in humans. Circulation, 50, 540-546. Perez, L. A., Hoyt, D. B., and Freeman, L. M. (1976). Localization of myocardial disorders other than infarction with "mTc-labeled phosphate agents. Jrournal of Nuclear Medicine, 17, 241-246. Poliner, L., Parkey, R. W., Bonte, F. J., Stokely, E. M., Buja, L. M., and Willerson, J. T. (1976). Technetium-99m stannous pyrophosphate myocardial scintigrams to recognize acute subendocardial myocardial infarcts in patients (abstract). American Journal of Cardiology, 37, 162. Prasquier, R., Taradash, M., Botvinick, E., Shames, D., and Parmley, W. (1976). The specificity of the diffuse pattern of cardiac uptake in myocardial infarction imaging with Tc-99m pyrophosphate (abstract). American Journal of Cardiology, 37, 163. Pugh, B. R., Parkey, R. W., Bonte, F. J., Poliner, L., Buja, L. M., and Willerson, J. T. (1976). Cardioversion and its potential role in the production of 'false positive' technetium-99m stannous pyrophosphate myocardial scintigrams (abstract). American J'ournal of Cardiology, 37, 163. Rigo, P., Murray, M., Taylor, D. R., Weisfeldt, M. L., Strauss, H. W., and Pitt, B. (1975). Hemodynamic and prognostic findings in patients with transmural and nontransmural infarction. Circulation, 51, 1064-1070. Samet, P. (1973). Hemodynamic sequelae of cardiac arrhythmias. Circulation, 47, 399-407. Soin, J. S., Burdine, J. A., and Beal, W. (1975). Myocardial localization of 99mTc-pyrophosphate without evidence of acute myocardial infarction. Journal of Nuclear Medicine, 16, 944-946. Willerson, J. T., Parkey, R. W., Bonte, F. J., Meyer, S. L., Atkins, J. M., and Stokely, E. M. (1975a). Technetium stannous pyrophosphate myocardial scintigrams in patients with chest pain of varying etiology. Circulation, 51, 10461052. Willerson, J. T., Parkey, R. W., Bonte, F. J., Meyer, S. L., and Stokey, E. M. (1975b). Acute subendocardial myocardial infarction in patients: its detection by technetium 99m stannous pyrophosphate myocardial scintigrams. Circulation, 51, 436-441. Zaret, B. L., DiCola, V. C., Donabedian, R. K., Puri, S, Wolfson, S., Freedman, G. S., and Cohen, L. S. (1975). Dual radionuclide study of myocardial infarction: relationships between myocardial uptake of potassium 43, technetium-99m stannous pyrophosphate, regional myocardial blood flow, and creatine phosphokinase depletion. Circulation, 53, 422-428.
Requests for reprints to Professor Leon Resnekov, Deparanent of Medicine, Cardiology-Box 423, The University of Chicago, 950 East 59th Street, Chicago, Illinois 60637, U.S.A.
Assessment of diagnostic value of technetium-99m pyrophosphate myocardial scintigraphy in 80 patients with possible acute myocardial infarction1 WARREN F. WALSH, HARISCHANDRA B. KARUNARATNE, LEON RESNEKOV, HEIKO R. FILL, AND PAUL V. HARPER From the Departments of Medicine, Radiology, and Surgery, and the Franklin McLean Memorial Research Institute (operated by the University of Chicago for the U.S. Energy Research and Development Administration), the University of Chicago Pritzker School of Medicine, Chicago, Illinois, U.S.A.
The diagnostic value of technetium-99m-pyrophosphate (Tc-pyrophosphate) myocardial scintigraphy was determined in 80 consecutive patients who had been admitted to the coronary care unit in order to rule out an acute myocardial infarction. Scintigraphic findings obtained within 5 days of admission were correlated with the final cardiac diagnosis determined for each patient. Significant myocardial uptake of Tc-pyrophosphate (positive scans) occurred in 13 of 22 patients (59%) had who enzyme and/or electrocardiographic proven acute myocardial infarct: 3 out of 5 with transmural myocardial infarct, 9 of 16 with nontransmural myocardial infarct, and 1 patient with left bundle-branch block. Of 58 patients who showed no evidence of acute myocardial infarction, positive scans occurred in 14 of 33 patients who had unstable angina pectoris (42%), 0 of 6 who had congestive heart failure, 6 of 9 who had other acute cardiac syndromes, and in 0 of 10 who had noncardiac chest pain. In the patients with unstable angina pectoris positive scans could not be predicted on the basis of the history, electrocardiographic findings, or the arteriographically determnined severity of the coronary artery disease. Blood levels of Tc-99m activity measured in 21 cardiac patients and in 6 volunteers did not correlate with the uptake intensity of Tc-pyrophosphate. These findings suggest caution in the use of this imaging method for the diagnosis of acute myocardial infarct in patients admitted with 'rule out myocardial infarction'. In 1973 technetium-99m pyrophosphate myocardial scintigraphy was introduced by Bonte and coworkers as a noninvasive method of visualising acute myocardial infarction (Bonte et al., 1974; Parkey et al., 1974). Since that time there have been a number of clinical and experimental studies which have confirmed the uptake of labelled pyrophosphate by acutely infarcted myocardium (Fink/Bennett et al., 1974; Bonte et al., 1975; Buja et al., 1975; Willerson et al., 1975a). These studies suggest that
myocardial imaging using Tc-pyrophosphate may be of value in the investigation of patients admitted 'Supported in part by a USPHS contract (Myocardial Infarction Research Unit), SCOR-Ischemic Heart Disease Grant, USPHS training grant, USPHS Center for Imaging Research Grant, U.S. Energy Research Development Administration under ERDA Contract, the Fogarty International Fellowship Program, and the Chicago Heart Association. Received for publication 21 December 1976
to the coronary care unit in whom the diagnosis of acute myocardial infarction is uncertain.
Subjects and methods PATIENT POPULATION
Eighty patients (mean ages: men 50 years, women 30 years, whole group 57 years), who had been admitted to the coronary care unit (CCU) in order to rule out an acute myocardial infarction, were investigated. Seventy-six were admitted with chest pain thought to be cardiac in origin, and 4 after either syncope or an episode of acute dyspnoea. In all cases a definite cardiac diagnosis had not been established at the time of the initial imaging study. IMAGING PROCEDURE
Myocardial scintigraphy was performed between
974
975
Tc-99m-pyrophosphate myocardial scintigraphy 12 hours and 5 days after admission to the CCU (mean 48 hours). After obtaining informed consent, patients were imaged at their bedside using a mobile gamma camera (Searle Radiographics Pho-Gamma HP or IV) equipped with a 16 000 parallel hole, high resolution collimator, and interfaced to a mobile digital acquisition unit (Ohio Nuclear 150 System) with tape storage facilities. Fifteen millicuries of Tc-pyrophosphate (Mallincrodt/Nuclear) were injected intravenously, and 60 to 180 minutes later (mean 102 minutes) analogue and digital images containing 800 000 counts were collected in both the anteroposterior and left anterior oblique 30° projections. Analogue images were recorded on Polaroid film and the digital images stored on magnetic tape for subsequent image processing. No complications resulting from the administration of the Tc-pyrophosphate occurred in any of the patients studied. SCAN INTERPRETATION
Myocardial scans were interpreted by one of the authors (H.F.) who was unaware of the clinical findings or the presumed diagnosis. The intensity of the myocardial uptake of Tc-pyrophosphate was visually graded on a 0 to III scale: 0, being no visible uptake of the agent in the region of the heart; grade I, faint diffuse uptake; grade II, definite but moderate uptake, either localised or diffuse in distribution; and grade III, intense localised uptake approximately equal to the intensity of the sternum (Fig. 1). Because of doubt about the sig0
nificance of grade I uptake, only grade II and III were regarded as abnormal and as representing positive scans. MEASUREMENT OF 99mTC BLOOD ACTIVITY
In 21 cardiac patients and in 5 normal volunteers, blood samples were taken both at the beginning and at the end of the imaging procedure. All 26 subjects were imaged 120 minutes or more after injection. The blood samples were counted for 99 mTc activity, and using injection standards the percentage of the injected dose in the blood was calculated, which was normalised to the estimated whole blood volume of the patient. CARDIAC CATHETERISATION
Of the 80 patients, 28 underwent cardiac catheterisation during the study admission as part of the investigation of their chest pain. After informed consent was obtained, bilateral selective coronary arteriography using the Judkins technique and biplane left ventricular cineangiography were performed. The coronary angiograms and left ventriculograms were interpreted by at least two observers. The degree of obstruction of lesions in the left anterior descending, left circumflex, and right coronary arteries, and in their branches, was subjectively assessed, and left ventricular function, ejection fraction, and segmental contractility were defined.
I
II
I
AP
I I
LA
--.-- ---' Doutbtfu..-.-------..............- - --------Positive ----. bt-----I Dou -----Fig. 1 Grading system (O to III) of the intensity of myocardial uptake of Tc-99m pyrophosphate used for interpreting the myocardial scintigrams 2 to 4 hours after intravenous injection of the agent. Grades II and III were regraded as definitely positive.
NeqtiVe
-
Walsh, Karunaratne, Resnekov, Fill, and Harper
976 Acute MI
myocardial infarction. Eight had both electrocardiographic and enzymatic evidence of infarction, 7 had enzymatic evidence, and 1 patient, who was admitted 4 days after his main episode of chest pain, had electrocardiographic evidence only. Of tnese 16 patients, 7 had positive scans (44%), of whom 5 showed a localised uptake pattern and in 2 a diffuse uptake pattern occurred. Nine patients (56%) had negative or doubtful scans (grade 0-I). One patient, who had chronic left bundle-branch block and an enzyme rise indicative of infarction, had a grade III positive scan. Peak enzymes, electrocardiographic abnormalities, and the time interval between admission and scanning did not distinguish between those patients with positive scans and those with negative or doubtful scans (Table).
Unstable angina
01
50]
lSCHF 0,
Non.cardiac
Other cardiac
4
50-
UNSTABLE ANGINA PECTORIS
Tc PYP uptake grade (0-11)
Fig. 2
Thirty-three patients were diagnosed as having had unstable angina. Criteria for diagnosis were: (1) a history of typical cardiac pain occurring within 24 hours of admission, either recent in onset or a change from a previously stable pattern, (2) normal levels of CK and LDH enzymes, and (3) no electrocardiographic evidence of recent myocardial infarction. The mean age of the 33 patients was 59 years. There were 22 men and 11 women. Twelve patients had transient ST and/or T-wave changes
Distribution of Tc-99m pyrophosphate uptake
grades (0 to III) in 22 patients with acute myocardial infarction, 33 patients with unstable angina pectoris, 6 patients with congestive heart failure, 9 patients with other acute cardiac syndromes, and 10 patients with
noncardiac
chest pain. TMI= transmural
myocardial
infarction (5 of the 22 patients).
Results
The available findings and the results of all laboratory and angiographic investigations were reviewed to determine the definitive diagnosis of each patient. Based on this
information, the 80 patients were
Table Time interval before admission and study, electrocardiographic findings, peak CK level, and scan findings in 22 patients with acute myocardial infarction
divided into 5 clinical groups: (1) acute myocardial infarction,
(3) con-
(2) unstable angina pectoris,
gestive heart failure, (4) other acute cardiac syndromes,
and (5) noncardiac chest pain.
The scan
findings were correlated with the final clinical diagnosis (Fig. 2).
ACUTE
MYOCARDIAL
INFARCTION
Twenty-two patients (mean age 58 years; 15 men, 7 women) were diagnosed as having had acute myocardial
infarction
were the triad of:
(Table).
Criteria
for
diagnosis
(1) prolonged chest pain,
(2)
evolutionary ST and T wave changes on the electrocardiogram, and/or (3) a transient rise and fall in the serum levels of the enzymes creatinine kinase
(GK) and lactic dehydrogenase (LDH). Development of new deep Q waves of greater than 0 04 s duration indicated a transmural infarction. Five patients met the diagnostic criteria for acute
transmural
infarction,
of whom
3
had
grade
II
positive scans, and 2 had grade I scans. Sixteen patients sustained acute nontransmural
Case No. 1 2 3 4
5 6 7 8 9 10 11 12 13 14
15 16 17 18 19 20 21 22
Time Electrocardiogram interval (h) 96
Old ant/inf (T)
24 48 72 96 60 12 24 24 72 24 48 24 24 96 48 24 48 96 36 72 12
Old ant (T) Ant (NT) Ant (T) LBBB Inf (T)
Nonspecific Ant (NT) Ant (NT) Ant (NT)
Inf (T) Ant (NT) Inf (T) Ant (NT) Ant (NT) Ant (NT) Inf (T) Ant (NT) RBBB, LAFB Old ant (T) Inf (T) Nomual
Peak CK (units)
Scan grade
3240 640 265 359
III III II II II II II II II II II I
1580 374 680 940
50 760 428 544 60 1974 6900 2800 960 312 808 374 480 1060
I I I I I I 0 0 0 0
Ant, anterior; Inf, inferior; LBBB and RBBB, left and right bundlebranch block; LAFB, left anterior fascicular block; (T), transmural; (NT) nontransmural.
977
Tc-99m-pyrophosphate myocardial scintigraphy LAO CB
MT
b
MB c
d
Fig. 3 Positive anteroposterior and left anterior oblique scintigrams occurring in patients with (a) acute nontransmural myocardial infarction, (b) unstable angina pectoris, (c) angina pectoris with normal coronary arteries, and (d) chest pain and mitral valve prolapse.
positive scans, of which 2 remained grade II positive 48 hours later, and 2 showed grade I uptake, 6 and 8 days, respectively, after the initial study. One of these latter 2 patients had undergone myocardial revascularisation surgery during the interval between scans. In view of the intriguing occurrence of positive scans in a high proportion of patients with unstable angina pectoris, the clinical data of the 33 patients were examined retrospectively to ascertain whether there were any specific features that might have predicted the occurrence of a positive scan. There were no features in the history and no specific electrocardiographic abnormalities which correlated with the presence of a positive scan. In order to determine if the significant uptake of Tc-pyrophosphate was related to the severity of the coronary artery disease, a study was made of the coronary arteriographic findings in the 22 patients (67%) who underwent cardiac catheterisation during the acute phase of unstable angina pectoris. Fig. 4 shows that while the majority of these patients had significant two or three vessel coronary artery disease, there was no relation between the number of vessels with obstructive disease of greater than 70 per cent, and the occurrence of a positive scan, assuming that grade 0 and grade I do not represent any significant myocardial uptake. CONGESTIVE HEART FAILURE
Six patients (mean age 69; 3 women, 3 men) were diagnosed as having had an exacerbation of cardiac failure, with no evidence of recent myocardial infarction. Five had underlying coronary artery disease, and one a primary congestive cardiomyopathy. Four of these patients had zero uptake of Tc-pyrophosphate and 2 had grade I uptake.
65, c
during pain, 11 had stable nonspecific electrocardiographic abnormalities, 7 had electrocardiographic evidence of old transmural myocardial infarction and 3 patients had normal electrocardio-
a
C.
4--1
EII-m LJ O-I
3-
I-
0
6
z
2
grams.
Despite the absence of evidence of recent infarcIL 0 3V tion, 14 patients (42%) had positive scans, of whom 2V Iv ov No. of coronary arteries > 700% stenosis 8 had a localised uptake pattern and 6 had a more diffuse pattern (Fig. 3). Fig. 4 Relation between the angiographically determined Seven patients with unstable angina pectoris had severity of the coronary artery disease and the uptake follow-up scans performed 1 to 8 days after the first grades (O to III) in 22 patients with unstable angina study. Four of these patients initially had grade II pectoris. V=coronary artery.
978
Walsh, Karunaratne, Resnekov, Fill, and Harper
OTHER CARDIAC SYNDROMES
NONCARDIAC CHEST PAIN
Nine patients comprised this clinical gr(oup (mean age 53 years; 5 men and 4 women). Thr ee patients had angiographically documented mitral valve prolapse with normal coronary arteries, of whLom 1 had a grade III positive scan, 1 a grade I, and the third a grade 0 scan. Three female patients who presented with typical angina pectoris had normal epicardial coronary arteries, as shown by selectivce coronary arteriography. Of these 3 patients, 2 ha(d grade II positive scans, including 1 patient who hsad a positive treadmill exercise test and whose m yocardium produced lactate when subjected to thee stress of atrial pacing. Two patients who presente a witn recurrent ventricular arrhythmias withoiut cardiac pain had grade II positive scans. One hlad undergone electroversion (250 joules) 24 hours before the imaging procedure. The sixth patient who had ulcerative colitis with a probable toxic nayocarditis had a negative scan.
Ten patients (mean age 44 years; 6 women, 4 men) were admitted with chest pain, whose symptoms upon careful review of the history and all investigations were not considered to be cardiac in origin. In 7 the pain was thought to be musculoskeletal. The other 3 patients were diagnosed as reflux oesophagitis, biliary colic, and viraemia, respectively. Six patients in this noncardiac group had zero uptake scans and 4 had grade I scans. Thus no patient with non-cardiac chest pain had a definitely positive scan.
I ___
0
22
_
o Normal
* Cardiac
20 E
la~ 8
0
0
b
0
0
0 -0
0
0
E 14 0^
12
uI 12-
E
._
0 o
0
0 0 0
0 I II mI Tc PYP uptake grade Fig. 5 Relation between the uptake g,rade of Tc-99m pyrophosphate and the blood levels of;Tc-99m activity at the time of imaging in 21 patients upith cardiac disease
and in 5 normal volunteers.
The relation between the visually graded uptake of Tc-pyrophosphate and the residual 99mTc blood pool activity at the time of the imaging study was investigated in 21 cardiac patients and in 5 normal control subjects. Fig. 5 shows the residual blood activity for each of the 26 subjects classified according to the scan uptake grade reading. These measurements were all made at least 120 minutes after administration of Tc-pyrophosphate in order to allow time for adequate clearance of the agent from the circulation. The mean percentage of the injected dose per whole blood volume for the patients with grade 0 scans was 12-2 per cent (±0 9 SE); grade I, 13-8 per cent ( +1-5); grade II, 16-4 per cent (±1-2), and grade III, 13-8 per cent (±2 4). Using a one-way analysis of variance no significant differences in blood pool activity were found between the two groups. Frequently subjects with 0 or grade I scans had higher blood pool levels of 99 mTc activity than patients with grade II activity. Thus, no qualitative or quantitative relation could be shown between the uptake grade of Tc-pyrophosphate and the blood pool activity.
Discussion
0
0 0 C IV
RELATION BETWEEN SCAN GRADE AND BLOOD ACTIVITY OF 99mTc
Both animal and clinical studies have shown that the bone scanning agent technetium-99m-pyrophosphate concentrates in acutely infarcted myocardium and can be imaged with a standard scintillation camera (Fink/Bennett et al., 1974; Bonte et al., 1975; Buja et al., 1975; Willerson et al., 1975a). Morphological animal studies have shown that uptake of labelled pyrophosphate, being in part flow dependent, tends to be confined to the more peripheral segments of an acute infarct where residual perfusion is greater (Buja et al., 1975; Zaret et al., 1975). Little uptake of Tcpyrophosphate can be detected in the adjacent histologically normal, but presumably ischaemic, tissue surrounding an infarct. The time course and
Tc-99m-pyrophosphate myocardial scintigraphy the distribution of pyrophosphate uptake has been observed to parallel the accumulation of ionised calcium by necrotic tissue (Buja et al., 1975, 1976). It has, therefore, been suggested that pyrophosphate is fixed intracellularly by binding with mitochondrial calcium deposits (Buja et al., 1975). The experimental data suggest that Tc-pyrophosphate myocardial scintigraphy may provide a specific diagnostic test for acute myocardial infarction and hence be a useful diagnostic aid in the coronary care unit. The initial clinical studies of Tc-pyrophosphate myocardial imaging concluded that this method had a high degree of diagnostic specificity and sensitivity for acute myocardial infarction (Fink/Bennett et al., 1974; Parkey et al., 1974; Willerson et al., 1975a). Willerson et al. (1975a), in the largest clinical series so far reported, studied 202 patients admitted to the CCU and observed that 96 of 101 patients who had proven acute myocardial infarction had positive scintigrams. In contrast, only 9 of 101 patients who failed to show evidence of acute infarction had positive scans. More recently, there have been reports of positive scans being obtained in patients with unstable angina pectoris (Abdulla et al., 1976; Donsky et al., 1976), ventricular aneurysms (Ahmad et al., 1976), cardiomyopathy (Perez et al., 1976), and even as an incidental finding in those having bone scans performed (Soin et al., 1975). These findings prompted us to examine the diagnostic specificity and sensitivity of this scintigraphic method in a group of patients admitted to the CCU with a diagnosis of possible acute myocardial infarction. This clinical category was selected because it frequently presents a diagnostic problem in the CCU and might be an area in which Tc-pyrophosphate myocardial imaging might be of great diagnostic value within the first few days in hospital. The findings in the present study indicate that both the diagnostic sensitivity and specificity of Tcpyrophosphate myocardial imaging for acute myocardial infarct are lower than have been reported previously (Willerson et al., 1975a, b). The detection rate in the 22 patients with proven acute myocardial infarct was only 55 per cent. The reason for the apparent discrepancy between this and other series is probably related to the method of patient selection. Our study, by focusing on patients admitted as 'rule out myocardial infarct' effectively excluded those patients who had typical electrocardiographic changes of acute transmural myocardial infarction on admission, thereby selecting a higher proportion of patients with the more subtle diagnostic features observed in nontransmural myocardial infarction. Since nontransmural in-
979 farcts are usually smaller in size than transmural infarcts they tend to contain lesser amounts of tracer activity which may be insufficient to be detected externally by the scintillation camera (Botvinick et al., 1975). The diffuse, moderate uptake pattern of Tcpyrophosphate has been proposed as a specific diagnostic feature of acute nontransmural myocardial infarct (Poliner et al., 1976). Though this pattern was seen in some of our patients with myocardial infarction it did not, however, prove to be specific for nontransmural myocardial infarct, since it also occurred in patients without evidence of myocardial infarction. These limitations of Tcpyrophosphate myocardial imaging in the detection of acute nontransmural myocardial infarction need to be recognised, particularly since recent studies have shown that the mortality rate for patients with nontransmural myocardial infarction may be- similar to that for transmural myocardial infarction (Madias et al., 1974; Rigo et al., 1975). Of great interest was the demonstration that a significant number of patients with unstable angina pectoris had positive scans. Most of our patients in this group had grade II positive scans, suggesting that it might not be possible to distinguish scintigraphically between patients with acute myocardial infarction, especially if nontransmural, and those with unstable angina pectoris. This distinction has great clinical and therapeutic importance, since the management of these two groups of patients may differ. Many centres now believe that patients who have unstable angina pectoris should have early coronary angiography followed by myocardial revascularisation surgery, provided the coronary anatomy is suitable (Miller et al., 1973; Bonte et al., 1974; Bonchek et al., 1974). This therapeutic course would be contraindicated in patients with proven acute myocardial infarction. The explanation for the occurrence of positive scintigrams in some patients with unstable angina pectoris is unclear. Two possibilities exist. First, reversibly injured myocardial cells may be capable of taking up pyrophosphate in significant amounts. This hypothesis is difficult to test in the experimental animal, since no satisfactory animal model of unstable angina pectoris exists. Experimental myocardial uptake and distribution studies of Tcpyrophosphate have, by necessity, been performed only in acute myocardial infarction. The absence of the usual electrocardiographic or enzymatic markers of recent infarction, together with the known severity of the myocardial ischaemia in unstable angina pectoris, might tend to favour this first explanation. A second possible explanation is that Tcpyrophosphate myocardial imaging is more sensi-
980 tive than the currently available methods for the diagnosis of acute myocardial infarction and is, therefore, capable of detecting an infarct missed by the usual diagnostic methods. Recent necropsy studies of unstable angina pectoris lend some support to this hypothesis. Guthrie and co-workers (1975) reported pathological studies on patients with clinical unstable angina pectoris who died after coronary arterial surgery. In 6 cases, recent myocardial infarction, histologically antedating the time of surgery, was shown. In addition, acute myocardial infarction may not be detected in the presence of left bundle-branch block or previous transmural myocardial infarction if the patient is seen a few days after the main symptomatic event when the serum CK and LDH enzymes have returned to normal levels. Thus, the pathogenesis of positive Tc-pyrophosphate scans in patients with clinically unstable angina pectoris must remain uncertain at the present time. Since this study focused on determining the diagnostic specificity and sensitivity of Tc-pyrophosphate myocardial imaging when used within the first few days of being admitted to hospital, serial scans were not usually obtained. In the small number of patients with unstable angina pectoris who had more than one scan performed, the changes seen were variable. The scan findings tended to be stable during the first week, but became less intense in uptake during the second week, as has been previously noted in patients with acute myocardial infarction (Willerson et al., 1975a). Positive scans occurring in patients with either mitral valve prolapse or recurrent ventricular arrhythmias have not been previously reported. It could be suggested that the metabolic or haemodynamic abnormalities, which have been shown to occur in these syndromes, result in myocardial cell injury (Samet, 1973; Natarajan et al., 1975). These damaged cells may then be capable of taking up pyrophosphate. Direct current shock has been shown in animals to result in positive technetium pyrophosphate scans, and this may have been a factor in the one patient with ventricular arrhythmias treated by electroversion (DiCola et al., 1976; Pugh et al., 1976). In the animal experiments reported, one to four DC shocks of 320 joules each produced a significant uptake of pyrophosphate in the absence of any demonstrable change in regional blood flow (DiCola et al., 1976). The diffuse grade I scan was a nonspecific finding in this series being observed in patients with and without cardiac disease. The diffuse grade II scan was, on the other hand, seen only in patients with acute cardiac disease, most commonly nontransmural infarction or unstable angina pectoris. The
Walsh, Karunaratne, Resnekov, Fill, and Harper nonspecific nature of the diffuse scan pattern has recently been confirmed by Prasquier et al. (1976). They noted this appearance in 3 patients with acute nontransmural infarction, in 7 of 17 (41%) patients with unstable angina pectoris, and in 2 of 39 (5%) patients with stable angina pectoris. The diffuse pattern was also seen in 11 per cent of 484 patients who had routine bone scans; however, 79 per cent of this group had had prior left mastectomy. Whether this diffuse pattern represents true myocardial uptake or a cardiac blood pool image is still uncertain. The lack of correlation between blood pool activity of 99 mTc and the scan uptake in our study suggests that myocardial uptake of 99 mTc is responsible for the diffuse pattern. These measurements were made at least 2 to 4 hours after injecting Tc-pyrophosphate intravenously in order to minimise the possibility of blood pool interference. Blood clearance studies of Tc-pyrophosphate show a biexponential pattern of clearance from the blood. The first exponent has a relatively rapid clearance half-time of 13-6 minutes, and the second exponent a slower half-time of 380 minutes. It is believed that the rapid first exponent represents decreasing radioactivity caused by bone uptake, while the second exponent mainly reflects renal excretion of the agent (Krishnamurthy et al., 1975). Hence, blood pool images are more likely to be obtained in patients with impaired renal function or those imaged within approximately the first hour after injection. In addition, since the degree of tissue attentuation will vary from patient to patient, those with thin chest walls are more likely to have blood pool Tc-pyrophosphate images. From the analysis of the scans of 80 patients admitted to the CCU with possible myocardial infarction, we have shown that Tc-pyrophosphate myocardial scintigraphy: (1) may give positive scans in patients with unstable angina pectoris and other acute cardiac syndromes, and (2) may fail to detect acute myocardial infarction especially if nontransmural. Our results, therefore, suggest caution in the diagnostic application of Tc-pyrophosphate myocardial scintigraphic findings in patients admitted to hospital with 'rule out myocardial infarction'. References
Abdulla, A. M., Canedo, M. I., Cortez, B. C., McGinnis, K. D., and Wilhelm, S. K. (1976). Detection of unstable angina by 99m technetium pyrophosphate myocardial scintigraphy. Chest, 69, 168-173. Ahmad, M., Dubiel, J. P., Verdon, T. A., Jr., and Martin, R. H. (1976). Technetium 99m stannous pyrophosphate myocardial imaging in patients with and without left
ventricular aneurysm. Circulation, 53, 833-838. Bonchek, L. I., Rahimtoola, S. H., Anderson, R. P., McAnulty, J. A., Rosch, J., Bristow, J. D., and Starr, A.
Tc-99m-pyrophosphate myocardial scintigraphy (1974). Late results following emergency saphenous vein bypass grafting for unstable angina. Circulation, 50, 972-977. Bonte, F. J., Parkey, R. W., Graham, K. D., and Moore, J. G. (1975). Distribution of several agents useful in imaging myocardial infarcts. J'ournal of Nuclear Medicine, 16, 132-135. Bonte, F. J., Parkey, R. W., Graham, K. D., Moore, J., and Stokely, E. M. (1974). A new method for radionuclide imaging of myocardial infarcts. Radiology, 110, 473-474. Botvinick, E. H., Shames, D., Lappin, H., Tyberg, J. V., Townsend, R., and Parmley, W. W. (1975). Noninvasive quantitation of myocardial infarction with technetium 99m pyrosphophate. Circulation, 52, 909-915. Buja, L. M., Parkey, R. W., Dees, J. H., Stokely, E. M., Harris, R. A., Jr., Bonte, F. J., and Willerson, J. T. (1975). Morphologic correlates of technetium-99m stannous pyrophosphate imaging of acute myocardial infarcts in dogs. Circulation, 52, 596-607. Buja, L. M., Tofe, A. J., Parkey, R. W., Dees, J. H., Mukherjee, A., Bonte, F. J., and Willerson, J. T. (1976). Technetium pyrophosphate and diphosphonate uptake and calcium levels in canine myocardial infarcts (abstract). Clinical Research, 24, 211A. DiCola, V. C., Freedman, G. S., and Zaret, B. L. (1976). Effect of direct current countershock upon myocardial technetium-99m stannous pyrophosphate uptake (abstract). American J'ournal of Cardiology, 37, 131. Donsky, M. S., Curry, G. C., Parkey, R. W., Meyer, S. L., Bonte, F. J., Platt, M. R., and Willerson, J. T. (1976). Unstable angina pectoris. Clinical, angiographic and myocardial scintigraphic observations. British Heart Journal, 38, 257-263. Fink/Bennett, D., Dworkin, H. J., and Lee, Y. H. (1974). Myocardial imaging of the acute infarct. Radiology, 113, 449-450. Guthrie, R. B., Vlodaver, Z., Nicoloff, D. M., and Edwards, J. E. (1975). Pathology of stable and unstable angina pectoris. Circulation, 51, 1059-1063. Krishnamurthy, G. T., Huebotter, R. J., Walsh, C. F., Taylor, J. R., Kehr, M. D., Tubis, M., and Blahd, W. H. (1975). Kinetics of 99m Tc-labelled pyrophosphate and polyphosphate in man. Journal of Nuclear Medicine, 16, 109-115. Madias, J. R., Chahine, R. A., Gorlin, R., and Blacklow, D. J. (1974). A comparison of transmural and nontransmural acute myocardial infarction. Circulation, 49, 498-507. Miller, D. C., Cannom, D. S., Fogarty, T. J., Schroeder, J. S., Dailey, P. O., and Harrison, D. C. (1973). Saphenous vein coronary artery bypass in patients with 'preinfarction angina'. Circulation, 47, 234-241. Natarajan, G., Nakhjavan, F. K., Kahn, D., Yazdanfar, S., Sahibzada, W., Khawaja, F., and Goldberg, H. (1975). Myocardial metabolic studies in prolapsing mitral leaflet syndrome. Circulation, 52, 1105-1110. Parkey, R. W., Bonte, F. J., Meyer, S. L., Atkins, J. M.,
981 Curry, G. L., Stokely, E. M., and Willerson, J. T. (1974). A new method for radionuclide imaging of acute myocardial infarction in humans. Circulation, 50, 540-546. Perez, L. A., Hoyt, D. B., and Freeman, L. M. (1976). Localization of myocardial disorders other than infarction with "mTc-labeled phosphate agents. Jrournal of Nuclear Medicine, 17, 241-246. Poliner, L., Parkey, R. W., Bonte, F. J., Stokely, E. M., Buja, L. M., and Willerson, J. T. (1976). Technetium-99m stannous pyrophosphate myocardial scintigrams to recognize acute subendocardial myocardial infarcts in patients (abstract). American Journal of Cardiology, 37, 162. Prasquier, R., Taradash, M., Botvinick, E., Shames, D., and Parmley, W. (1976). The specificity of the diffuse pattern of cardiac uptake in myocardial infarction imaging with Tc-99m pyrophosphate (abstract). American Journal of Cardiology, 37, 163. Pugh, B. R., Parkey, R. W., Bonte, F. J., Poliner, L., Buja, L. M., and Willerson, J. T. (1976). Cardioversion and its potential role in the production of 'false positive' technetium-99m stannous pyrophosphate myocardial scintigrams (abstract). American J'ournal of Cardiology, 37, 163. Rigo, P., Murray, M., Taylor, D. R., Weisfeldt, M. L., Strauss, H. W., and Pitt, B. (1975). Hemodynamic and prognostic findings in patients with transmural and nontransmural infarction. Circulation, 51, 1064-1070. Samet, P. (1973). Hemodynamic sequelae of cardiac arrhythmias. Circulation, 47, 399-407. Soin, J. S., Burdine, J. A., and Beal, W. (1975). Myocardial localization of 99mTc-pyrophosphate without evidence of acute myocardial infarction. Journal of Nuclear Medicine, 16, 944-946. Willerson, J. T., Parkey, R. W., Bonte, F. J., Meyer, S. L., Atkins, J. M., and Stokely, E. M. (1975a). Technetium stannous pyrophosphate myocardial scintigrams in patients with chest pain of varying etiology. Circulation, 51, 10461052. Willerson, J. T., Parkey, R. W., Bonte, F. J., Meyer, S. L., and Stokey, E. M. (1975b). Acute subendocardial myocardial infarction in patients: its detection by technetium 99m stannous pyrophosphate myocardial scintigrams. Circulation, 51, 436-441. Zaret, B. L., DiCola, V. C., Donabedian, R. K., Puri, S, Wolfson, S., Freedman, G. S., and Cohen, L. S. (1975). Dual radionuclide study of myocardial infarction: relationships between myocardial uptake of potassium 43, technetium-99m stannous pyrophosphate, regional myocardial blood flow, and creatine phosphokinase depletion. Circulation, 53, 422-428.
Requests for reprints to Professor Leon Resnekov, Deparanent of Medicine, Cardiology-Box 423, The University of Chicago, 950 East 59th Street, Chicago, Illinois 60637, U.S.A.
