Differentiation Between Reperfusion and Occlusion Myocardial

Necrosis With Technetium-99m Fyrophosphate Scans

RICHARD LONG, MD JIM SYMES, MD JEAN ALLARD, MD TOM BURDON ROBERT LISBONA, MD ISTVAN HUTTNER, MD ALLAN SNIDERMAN, MD, FACC

Montreal, Quebec, Canada

From the Cardiovascular Research Unit, Royal Victoria Hospital, Montreal, Quebec, Canada. This study was supported by Grant USPH 1 ROI HL 22098-01 from the National Heart, Luna. and Blood Institute, National Institutes of Health, Be- thesda, Maryland. Manuscript received July 19, 1979; revised manuscript received Apri! 14. 1980; accepted April 18. 1980.

Address for reprints: Allan Sniderman, MD, Cardiovascular Research Unit, Royal Victoria Hospital, 887 Pine Avenue West, Montreal, Que- bec, Canada H3A 1A 1.

Although infarct-avid scanning is used to diagnose myocardial infarction, no distinction is made between two very different mechanisms of myo- cardial cell death. These are occlusion necrosis and reperfusion necrosis. The former is the major form of cell injury after a sustained occlusion of a coronary artery; the latter occurs immediately after reperfusion of ischemic myocardium. The present study examines whether the tech- netium pyrophosphate scan becomes positive more rapidly with one form of injury rather than the other. Myocardial damage was produced in dogs by either sustained coronary occlusion (Group I, 6 dogs) or reperfusion (Group II, 16 dogs). Infarct-avid myocardial scans obtained with tech- netium-99m pyrophosphate were positive in only one of six animals in Group I when isotope was injected 4.5 hours after sustained coronary occlusion. In contrast, 12 of 13 animals (Group Ila) with reperfusion ne- crosis had a positive scan when isotope was injected 3.5 hours after re- perfusion. When isotope was injected only 30 minutes after the onset of reperfusion (Group Mb), only two of five scans were clearly positive. The results indicate that if the intervat between myocardial injury and a positive scan is known, inferences can be made concerning the predominant type of injury.

Although the distinction is seldom made in the clinical setting, it is clear from pathologic studies that inadequate coronary flow may cause two entirely distinct forms of myocardial necrosis.’ The first, which results from sustained interruption of coronary flow, is characterized histo- logically by decreased staining of myofilaments, with prominence of their cross-striations as well as nuclear pyknosis and karyolysis. Cell death does not start until at least 1 hour after interruption of flow and proceeds thereafter.2b We refer to this form of cell death as “occlusion necrosis” or coagulation necrosis. In contrast, if a portion of myocardium is ren- dered severely ischemic but coronary flow is restored at the end of 1 hour, cell death is much more rapid and the histologic appearance is entirely different. It is characterized by the occurrence of swelling, contraction bands and an increased cell content of calcium.4J’ We refer to this form of cell death as “reperfusion necrosis.”

There are clinical situations in which one or the other form of cell death may predominate. For example, perioperative infarction after coronary bypass surgery could be caused either by occlusion of the graft or native circulation or by reperfusion after completion of the graft construction. Bulkley and Hutchins6 demonstrated that the majority of perioperative infarcts occurred in the territories supplied by patent vessels and the histologic appearance in these regions is the same as that seen in reperfusion necrosis.

Restoration of myocardial flow after an ischemic period is an integral feature of reperfusion damage. The recognition of infarction by in- farct-avid isotopes depends on the accumulation of these markers within the damaged region, but such accumulation requires adequate flow to deliver the isotope to the damaged cells. With reperfusion necrosis, flow is present; with occlusion necrosis, particularly early after occlusion, flow

September 1980 The American Journal of CARDIOLOGY Volume 46 413

PYROPHOSPHATE SCANS AND TYPES OF MYOCARDIAL INFARCTION-LONG ET AL.

to the central region is virtually absent. Such a clear difference should produce a positive pyrophosphate scan very early after the production of a reperfusion injury, whereas with occlusion necrosis the scan should become positive much more slowly. Therefore, the fol- lowing study was designed to test the hypothesis that in the dog the infarct-avid scans will become positive earlier with reperfusion necrosis than with occlusion necrosis.

of the coronary occlusion and late scans were made at 48 or 96 hours after a second injection. In the remaining five dogs (Group IIb) a single scan wae performed after an injection of isotope made only 30 minutes after release of the coronary ligature.

Methods

Experimental procedure: Twenty-four fasting mongrel dogs (18 to 25 kg), anesthetized with sodium pentobarbital, underwent a left thoracotomy. The adequacy of ventilation was monitored by serial observations of arterial partial pres- sures of oxygen and carbon dioxide. Blood pressure was re- corded continuously through a can&la inserted through the femoral artery. After the pericardium was opened, a large diagonal branch of the left anterior descending artery was dissected at its origin and encircled with a ligature and all visible epicardial, collateral vessels to the zone to be made ischemic were ligated. Ten to twelve epicardial electrocar- diographic leads (30 gauge Teflon@-coated electrical wire) were then inserted over the anterolateral surface of the left ven- tricle. Recordings were made on an’Electronics for Medicine recorder at paper speeds of 25 and 100 mm/s. The experi- mental preparation has been described in greater detail e!sewhere.2

Myocardial biopsy: After the initial scan was completed (Groups I and IIa), the incisions were closed and the animals were allowed to recover from the anesthetic. Before the fol- low-up scan, the heart was again exposed, and final epicardial tracings were taken using the original leads. The heart was then arrested with potassium chloride and biopsy specimens were taken from representative sites. Tissues were fixed for 3 to 6 hours in 10 percent paraformaldehyde buffered with S-collidine. Specimens were processed for light microscopy by embedding in paraffin and were then cut and stained with hematoxylin-eosin and phosphotungstic acid-hematoxylin.

Experimental protocol: The 24 animals were separated into two groups. In six dogs, Group I, occlusion necrosis was produced by permanent ligation of the previously isolated branch of the left anterior descending coronary artery. In 18 dogs, Group II, reperfusion necrosis was produced by allowing reflow after a 1 hour occlusion of this branch. In Group I, ep- icardial electrocardiograms were recorded before and every 15 minutes for the first 2 hours after coronary occlusion, and then every 30 minutes for an additional 2 hours. In Group II, tracings were obtained before occlusion and every 15 minutes for the next 2 hours and then every 30 minutes for the subse- quent 2 hours after occlusion. A Q wave was defined as an initial negative deflection with an amplitude greater than 1.5 ’ mV and a duration of more than 20 ms.

Permanent coronary occlusion (Group I): As anticipated from our previous study,2 myocardial in- farction in the six dogs in Group I with permanent coronary occlusion was characterized by late onset of Q waves in the epicardial electrocardiogram (between 3 and 48 hours) and by the presence of histologically typical coagulation necrosis in the center of the in- farcted area (Fig. 1). At the periphery of the infarcted zones the histologic features resembled those of reper- fusion necrosis, showing necrotic cells with contraction bands (Fig. 2).

In five of the six dogs in Group I (Table I), early myocardial scans were negative (Fig. 3, left) (isotopes injected at 4.5 hours after occlusion and scan obtained at 6 hours). However, each of the three dogs undergoing scanning at 48 hours had a markedly positive scan (Fig. 3, center). The one dog with an early positive scan died suddenly and adequate postmortem histologic data were not obtained.

Myocardial nuclear scans: Technetium-99m pyrophos- phate scans were performed 1.5 hours after intravenous in- jection of 15 mCi of the isotope, utilizing a Searle Phogamma V camera with converging low energy, high resolution colli- mator. Images were recorded in anterior, 45’ left oblique and left lateral projections. Scans were graded 0 to 4+ according to Willerson7; only those scans graded 2+ or greater were ac- cepted as positive.

All myocardial scans were read only at the completion of the study by an observer who did not know the timing of the scans (early or late) or the mechanism of infarction (occlusion versus reperfusion). Intraobserver variation was tested by having the scans read a second time with the observer unaware of his previous reading. In only two cases, was there dis- agreement; in one a 2+ scan was read as l+, whereas in the other, the reverse occurred. Both dogs were in Group IIb. In both cases, the final reading was taken as 1+ (that is, a nega- tive result) to protect against falsely promoting the hypothesis because of any variation in reading.

Coronary reperfusion (Group II): In 13 dogs (Group IIa), isotope was injected 3.5 hours after coro- nary perfusion had been restored and the myocardial scan was made at 5 hours. In 12 of these dogs, a positive early scan was obtained (Fig. 3, right, Table II). In the one animal with a negative early scan, a second scan obtained at 48 hours was positive. Except in this last animal, the infarcts were histologically typical of re- perfusion necrosis with prominent contraction bands throughout the region of infarction (Fig. 4). In contrast, in the dog with a negative early scan pathologic studies revealed predominant coagulation necrosis with only a rim of reperfusion necrosis.

In the six dogs with permanent coronary occlusion (Group I) an initial myocardial scan was performed by injecting the radiotracer 4.5 hours after occlusion, and in three of these animals, scans were repeated at 48 hours. In the 18 dogs in which perfusion was restored after 1 hour (Group II), the early scan was made by injecting isotope at 3.5 hours after release

In five dogs (Group IIb), we tested whether scans might be positive even earlier after restoration of per- fusion. Within 2 hours after release of the snare, Q waves appeared in ischemic sites, thus confirming that necrosis was produced by reperfusion. A single injection of iso- tope 30 minutes after release of the snare with a scan performed 90 minutes later, resulted in unequivocably positive scans in two dogs with doubtful or negative scans in the remaining three (Table I).

This study was designed to test the hypothesis that infarct-avid myocardial scans using technetium-99m

414 September 1990 The American Journal of CARDIOLOGY Volume 46

Results

Discussion

PYROPHOSPHATE SCANS AND TYPES OF MYOCARDIAL INFARCTION-LONG ET AL.

FIGURE 1. Lett, normal dog myocar- dium. Cardiac muscle cells exhibit cross-striation and well defined nuclei. Right, coagulation necrosis after SUS-

tained coronary arterial ligation. Cross-striation in cardiac muscle cells is indistinct and nuclei are poorly de- fined. Note also margination of poly- morphonuclear leukocytes. (Hema- toxylin-eosin X660, reduced by 25 percent.)

pyrophosphate become positive much earlier with re- perfusion than with occlusion necrosis. Our results confirmed this hypothesis: With necrosis due to per- manent occlusion, scans were usually negative 6 hours after the onset of ischemia, whereas with reperfusion necrosis they were positive at this time. Our results

FIGURE 2. Low power view of myocardial infarct after sustained cor- onary arterial ligation. At the periphery of the infarct (center) reperfu- sion-type contraction band necrosis is evident; the rest of the infarct (bottom) shows coagulation-type necrosis. Note normal myocardium at the top. (Phosphotungstic acid-hematoyxlin X250, reduced by 25 percent.)

clearly indicate in the animal model used that these two forms of myocardial necrosis can be distinguished by the time at which the infarct-avid myocardial scan becomes positive.

Diagnostic implications: Our data are important for several reasons. First, they may reconcile differing reports as to the time at which infarct-avid scans be- come positive after the onset of a myocardial infarction. Although most clinical and experimental reports indi- cate that scans do hot become positive for at least 12 hours after coronary occlusion, individual exceptions have been noted.slo It is likely that the latter-as one report notedlo-are examples of necrosis due to reper- fusion rather than to occlusion.

Infarct-avid myocardial scanning in human beings has been delayed for 24 to 48 hours after a possible ep- isode of myocardial necrosis. Our findings indicate that this policy should be reassessed. Spasm-induced infarcts that are followed by reperfusion would be recognized by an early positive scan. Not only might the patho- genesis be more evident, but it should be possible, in a proportion of suspected cases of subendocardial in-

TABLE I

Results of Myocardlal Scanning in 24 Dogs

Time of Scan* Dogs ECG (Q wave

Group (n) within 2 hours) 1.5h 4.5 h 48 h 96h

I 6 2144 (4.5%) . l/6 313 Ila 13 65/103(63.1%) Itb 5 23151 (45.1%) ii5

12113 677 4i4 . . .

l Time at which isotope was injected after vessel occlusion; data indicate proportion of animals with positive scan.

ECG = electrocardiogram; . . . = scan not performed.

September 1980 The American Journal of CARDIOLOGY Volume 46 415

PYROPHOSPHATE SCANS AND TYPES OF MYOCARDIAL INFARCTION-LONG ET AL.

fKiuRE 3. L&l and center, Group I. Technetium99m pyrophcsphate myocardial scans recorded 6 (I( mft) and 48 (center) hours after co&nary occlusion. RI@, Group Ila. Technetium-99m pyrophosphate myocardial scan recorded 5 hours after reperf ‘usion.

TABLE II

Results of Myocardlal Scanning In Group Ila

Time of Scan l

4.5 h 48 h 96 h

4+ 1+ 2+

x:

t : 3+ 2+

;: 2+ 3+

5% o+

it

4-k 4+ 2+ . . . . .

2”: 3+ 2+

l Time at which isotope was injected after vessel occlusion; data indicate grade of scan (grade of 2-k or more indicates a poiitive scan).

= scan not performed.

416 September 1980 The American Journal of CARDIOLOGY

farction, to advance the time at which a positive diag- nosis can be made.

The time at which the scan becomes positive after myocardial necrosis may also throw light on the nature of the infarction process. For example, in perioperative infarction after coronary bypass surgery, myocardial damage may occur due to occlusion of the graft or native circulation, or to reperfusion after ischemia following aortic cross-clamping. Our results suggest that an early positive scan would favor reperfusion necrosis as the underlying mechanism in a particular case. This dis- tinction would be important because necrosis due to occlusion would imply an error in graft construction, whereas necrosis due to reperfusion would imply inad- equacy of myocardial protection during surgery.

Therapeutic implications: These results further suggest the possibility that the myocardial infarction associated with human atherosclerosis may consist of

Volume 46

PYROPHOSPHATE SCANS AND TYPES OF MYOCARDIAL INFARCTION-LONG ET AL.

two distinct components-occlusion coagulation ne- crosis occurring in a central area that is then surrounded by a marginal area of reperfusion necrosis. Because there is good evidence that the latter form of injury is preventable by pharmacologic treatment,3J1-13 the possibility of quantifying its extent raises considerable therapeutic possibilities.

The evidence for such a suggestion is at present no more than circumstantial. The infarct that follows coronary ligation in the dog does indeed consist of these two components.l”ls. The delayed appearance of pos- itive scans in human infarction may well be related to the return of blood flow to the predominantly epicardial areas, a phenomenon that is well reported in experi- mental infarction in the dog.1g-22 Could the zone with contraction bands and mitochondrial calcification at the periphery of such an infarct be tissue damaged by the reperfusion process? If this zone indeed reflects a rela- tively delayed reperfusion injury, then interventions to reduce infarct size might be helpful even if applied more than 2 hours after the initial result.

Conclusions: Our results indicate that in this dog model occlusion necrosis and reperfusion necrosis differ not only in histologic appearance and electrocardio- graphic evolution, but also in the length of the latent period before the pyrophosphate myocardial scan be- comes positive. Although it remains to be proved whether these findings can be extrapolated to human subjects, this observation may explain differences previously reported and may provide a basis for probing the extent to which human myocardial infarction may be attributed to occlusion or reperfusion type of cell death. The exact basis for the difference now reported is not established by our study. A recent preliminary report by Parkey et al. 23 does confirm the existence of a difference between reperfusion and occlusion infarc- tion in time to positive scan. Clearly, sufficient flow to deliver an adequate amount of isotope to the infarcted

FIGURE 4. Group Ila. Reperfusion necrdsis aher temporary coronary arterial ligation. Cardiac muscle cells exhibii marked contraction bands. (Hematoxylin_e9sin X660, reduced by 25 percent.)

region is the key requirement for development of a positive scan. But the increased tissue calcium content in reperfusion injury415 would be expected to bind a greater proportion of the delivered pyrophosphate isotope. That this factor might be an additional inde- pendent reason for our results is again supported by the report of Parkey et al. 23 Whatever the precise expla- nations, our data indicate that the infarct-avid scan can be used not only to recognize myocardial injury but also to define its type.

References

Baroldi G. Human myocardial infarction: coronarogenic or non- coronarogenic coagulation necrosis? Myocardiology 1972; 1: 399-413. Ylckleborough L, Huitner I, Symes J, Polrler N, Snlderman A. Significance of epicardial Q waves as an acute marker of myo- cardial necrosis in dogs. Cardiovasc Res 1978; 12:376-88. Symes JF, Snld&man AD, Polrler N, Mlckleborough L, Dobell ARC. Perioperative infarction: pathogenesis and prevention. Surg Forum 1978;29:282-3. Kloner RA, Ganote CE, Whalen DA Jr, Jennings RB. Effect of a transient period of ischemia on myocardial cells. II. Fine structure during the first few minutes of reflow. Am J Pathol 1974;74: 399-422.

9.

10.

11.

Shen AC, Jennlngs RB. Myocardial calcium and magnesium in acute ischemic injury. Am J Pathol 1972;67:417-40. Bulkley BH, Hutchlns GM. Myocardial consequences of coronary artery bypass graft surgery. The paradox of necrosis in areas of revascularization. Circulation 1977;58:906-13. Wllknon JT, Parkey RW, Bonte FJ, Meyer SL, Atkins JM, Stokely EM. Technetium stannous pyrophosphate myocardial scintigrams in patients with chest pain of varying etiology. Circulation 1975; 51:104&5_2. Wynne J, Hoiman BL, Lesch M. Myocardial scintigraphy by in- farct-avid radiotracers. In: Holman BL, Sonnenbllck EH, Lesch M, eds. Principles of Cardiovascular Nuclear Medicine. New York:

12.

13.

14.

15.

Grune & Stratton, 1978;219-42. Roberts AJ, Combes JR, Jacobsteln JO, Alonso DR, Post MR. Perioperative myocardial infarction associated with coronary artery bypass graft surgery: improved sensitivity in the diagnosis within 6 hours after operation with 99m Tc-glucoheptonate myocardial imaging and myocardial-specific isoenzyme. Ann Thorac Surg 1979;27:42-8. Bruno FP, Cobb FR, Rlvas F, Goodrlch JK. Evaluation of 99m technetium stannous pyrophosphate as an imaging agent in acute myocardial Infarction. Circulation 1976;54:71-7. Relmer KA, Rasmussen MM, Jennings RB. On the nature of pro- tection by propranolol against myocardial necrosis after temporary coronary occlusion in dogs. Am J Cardiol 1976;37:520-7. Relmer KA, Lowe JE, Jennings RB. Effect of the calcium antag- onist verapamil on necrosis following temporary coronary artery occlusion in dogs. Circulation 1977;55:581-7. Powell WJ, DlBona DR, Flares J, Leaf A. The protective effect of hyperosmotic mannitol in myocardial ischemia and necrosis. Circulation 1976;54:603-14. Buja LM, Parkey RW, Deer JH, el al. Morphologic correlates of technetium 99m stannous pyrophosphate imaging of acute myo- cardial infarcts In dogs. Circulation 1975;52:596-607. Buja LM, Tote AJ, Kulkarnl PV, et al. Sites and mechanisms of localization of technetium-99m phoaphaous radiopharmaceuticals in acute myocardial infarcts and other tissues. J Clin Invest

September 1990 The American Journal of CARDIOLOGY Volume 46 417

PYROPHOSPHATE SCANS AND TYPES OF MYOCARDIAL INFARCTION-LONG ET AL.

1977;60:724-40. 16. Buja LM, Parkey RW, Stokely EM, Bonte FJ, Wilierson JT.

Pathophysiology of technetium99m stannous pyrophosphate and thallium-102 scintigraphy of acute anterior myocardial infarcts in dogs. J Clfn Invest 1976;57:1508-22.

17. Zaret BL, DlCola VC, Donabedlan RD, et al. Dual radionuclide study of myocardial infarction. Relationships between myocardial uptake of potassium43, technetium99m stanncus pyrophosphate. regional myocardial blood flow and creatine phosphokinase de- pletion. Circulation 1976;53:422-8.

18. Coleman RE, Klein MS, Ahmed SA, Welas ES, Buchhols WM, Sobel BE. Mechanisms contributing to myocardial accumulation of technetium99m stannous pyrophosphate after coronary arterial occlusion. Am J Cardiol 1977;39:55-9.

19. Cox JL, Paaa HL, Wechaler AS, Oldham HN Jr, Sablston DC. Coronary collateral blood flow in acute myocardial infarction. J

Thorac Cardiovasc Surg 1975;69:117-25. 20. Hlrrel HO, Neleon GR, Sonnenbllck EH, Klrk ES. Redistribution

of collateral blood flow from necrotic to surviving myocardium following coronary occlusion in the dog. Circ Res 1976:39: 214-22.

21. Rlvas F, Cobb FR, Bacha RJ, Qeenflekl JC. Relationship between blood flow to ischemic regions and extent of myocardial infarction. Series measurement of blood flow to ischemic regions in dogs. Circ Res 1976;38:439-47.

22. Malaky PM, Vokonar PS, Paul SJ, Robblns SL, Hood WB. Auto- radiographic measurement of regional blood flow in normal and ischemic myocardium. Am J Physiol 1977;232:H576-83.

23. Parkey RW, Kulkarnl PV, Lewis JF, et al. Effect of coronary flow and site of injection on 99M Tc-Pyp detection of early canine myocardial infarcts (abstr). Am J Cardiol 1980;45:464.

416 September 1980 The American Journal of CARDIOLOGY Volume 46