Residual 201Tl Activity in Irreversible Defects as a Marker of Myocardial Viability
Background The objective of the present study was to characterize the relation between the residual 201Tl activity in irreversible perfusion defects and the extent of irreversible myocardial damage indicated by the volume fraction of myocardial interstitial fibrosis in patients with chronic coronary artery disease.
Methods and Results Stress planar 201Tl scintigraphy with tracer reinjection at rest was performed in 37 patients with ≥75% stenosis of the left anterior descending coronary artery, and anteroseptal 201TI activity was quantified by computer-assisted placement of regions of interest from the serial myocardial images. During coronary artery bypass grafting (performed within 6±3 weeks after scintigraphy), two transmural biopsy specimens were taken from the anterior wall of the left ventricle and the amount of interstitial fibrosis was assessed by use of light microscopic morphometry. A wide spectrum of interstitial fibrosis was obtained, ranging from 15 vol% to 60 vol%. Interstitial fibrosis was similar in patients with reversible (n=11) or irreversible (n=15) tracer defects in conventional stress-redistribution images. However, interstitial fibrosis was significantly lower in patients who had enhanced regional 201Tl activity after tracer reinjection compared with those who did not have enhancement of tracer activity after reinjection (28±8 vol%, n=7, versus 41±12 vol%, n=8; P=.031). The correlation between relative poststenotic 201Tl activity and interstitial fibrosis after tracer reinjection was significantly improved compared with conventional redistribution images (r=−.622 versus r=−.851, n=15; P<.01).
Conclusions The present data demonstrate that the level of regional 201Tl activity in redistribution and, in particular, reinjection images is significantly related to the mass of preserved viable myocytes in poststenotic left ventricular myocardium. Therefore, the residual 201Tl activity provides information about viability within irreversible perfusion defects and may itself serve as marker of myocardial viability.
During the past decade, myocardial scintigraphy with 201Tl has achieved a preeminent position for its use in the clinical assessment of myocardial viability in patients with coronary artery disease. By use of modified imaging protocols, such as the rest-reinjection technique,1 2 3 4 5 6 it has been shown that 201TI scintigraphy identifies quite the same myocardial regions as viable or nonviable as positron-emission tomography imaging with [18F]fluorodeoxyglucose,7 the current reference tracer for the assessment of myocardial viability.
In most of the previous studies, however, the severity of the reduction in 201Tl activity within irreversible defects was not investigated. Because the residual tracer activity itself might serve as evidence of viable myocardium,7 8 9 10 in the present study, the regional myocardial 201Tl activity was directly related to the extent of structural damage, as indicated by the amount of interstitial fibrosis in the poststenotic myocardium. Interstitial fibrosis was determined from transmural left ventricular biopsy specimens that were obtained during aortocoronary bypass grafting from patients with chronic coronary artery disease.
The study protocol was approved by the human study committee of the University of Heidelberg, and written informed consent was obtained from all patients.
The study group comprised 37 patients (n=3 women and 34 men; mean age, 57±10 years old) with chronic coronary artery disease and stable angina pectoris. Coronary arteriography demonstrated three-vessel disease in 15 patients and two-vessel disease in 17 patients; 5 patients had one-vessel disease of the left anterior descending coronary artery. All 37 patients had ≥75% proximal stenosis of the left anterior descending coronary artery. The site of the stenosis was proximal to the first septal perforator in 17 patients. In the other patients, the stenosis was distal to the origin of the first septal branch but proximal to the origin of the second diagonal branch. Fifteen of the 37 patients had a history of previous anterior non–Q-wave myocardial infarction, and 7 patients had a history of previous Q-wave anterior myocardial infarction. Patients with a left ventricular aneurysm and/or a global left ventricular ejection fraction <35% were excluded because of a potentially higher risk for biopsy-related complications.
Patients underwent 201Tl scintigraphy within 10 weeks after coronary angiography (mean, 7±4 weeks) without intercurrent changes in their clinical status.
Coronary arteriography and left ventricular angiography were performed by the Judkins technique with 35-mm cine filming by use of multiple projections, including angulated views. At least five projections were used for the left coronary artery, and at least two projections were used for the right coronary artery. Coronary artery stenoses were visually graded by two experienced cardiologists (R.Z. and B.R.), who were unaware of the patients’ clinical or scintigraphic data.
Additionally, global and regional left ventricular performance was determined with the centerline method11 by use of the left ventricular angiograms obtained in the right anterior oblique projection. As previously described,12 regional wall motion in the anterolateral segment was considered abnormal if wall motion was ≥2 SD below the mean anterolateral wall motion obtained from 64 normal patients.
Patients were investigated after an overnight fast and after withdrawal of all long-acting antianginal medications; only sublingual nitroglycerin was permitted as an antianginal treatment. Patients engaged in an exercise program that used an upright bicycle ergometer to implement a stepwise increase of workload (initial workload, 50 W, with 25-W increases every 2 minutes). A positive response to stress was defined as development of typical angina pectoris and/or ≥1.5-mm horizontal or downsloping ST-segment depression at 80 milliseconds after the J-point. One minute before the anticipated end of the symptom-inducing limited-stress test, 80 MBq IV (2.2 mCi IV) 201Tl chloride was injected into the patient.
Planar scintigraphy was performed 5 minutes (poststress images) and 4 hours (redistribution images) after tracer injection in the anterior and the 30° left anterior oblique view with a Siemens Orbiter ZLC 7500 gamma camera equipped with a high-resolution parallel hole collimator and interfaced to a Siemens MicroDELTA system. Data were collected for 5 minutes per view with a peak energy setting at 75 keV and a 20% window.
In case of persistent 201Tl defects on the 4-hour delayed images, 40 MBq (1.1 mCi) 201Tl was reinjected at rest, immediately after redistribution imaging, and a third set of images (reinjection images) was taken within 30 minutes after the second tracer injection under the same imaging protocol.1
Quantitative Image Analysis
Quantitative image analysis was performed off-line on standard 80386 IBM-AT personal computers, as previously described.13 In brief, after the MicroDELTA image data were converted to a PC/MS-DOS format, the original 128×128-pixel images were interactively centered, zoomed, and restored as 64×64-pixel images. After bilinear interpolative “background subtraction,”14 sequential scintigrams were spatially aligned with an automated image-superposition algorithm.15 The horizontal and vertical lags of sequential images are thereby identified by calculating the peak values of the (one-dimensional) cross-correlation function of the vertical and horizontal image-matrix projections.
After operator-interactive selection of the left ventricular cavity, for each projection, seven 9-pixel regions of interest were automatically drawn by the computer program (Fig 1⇓) from the center of the left ventricular cavity along each of 45 radii at the regional maximum of myocardial activity.16 These regions of interest were identically placed in the poststress and spatially aligned redistribution and reinjection scintigrams. Within each region of interest, the relative tracer activity, which was normalized to the peak activity obtained in the seven regions of interest within the respective scintigram (100%), was calculated. Regional 201Tl activity in the perfusion territory of the left anterior descending coronary artery was then calculated as the average tracer activity obtained in the three anterolateral regions of interest (anterior view) and the three septal regions of interest (30° left anterior oblique view).
In the poststress images, regional tracer activity <80% in the perfusion territory of the left anterior descending coronary artery was defined as a perfusion defect.17 Perfusion defects were classified as reversible if regional tracer activity achieved ≥80% in subsequent redistribution images or if the regional tracer activity increased by ≥10% on the redistribution images compared with the poststress images. Otherwise, the defects were classified as irreversible. Irreversible perfusion defects were further classified as moderate if regional tracer activity achieved ≥65% in the redistribution study and as severe if tracer activity remained <65%.
Patients with irreversible defects were considered to have enhanced 201Tl uptake after tracer reinjection (“fill-in”) if regional 201Tl activity was ≥80% in the reinjection images or increased by ≥10% compared with the activity in the redistribution images. Otherwise, defects were considered unchanged after reinjection (no “fill-in”).
Transmural Left Ventricular Biopsies
Patients underwent coronary artery bypass grafting within 12 weeks (mean, 6±3 weeks) after 201Tl scintigraphy. During cardiopulmonary bypass, two transmural biopsy specimens were taken from the anterior wall of the left ventricle before cardioplegia was begun. The site of the biopsy was selected by the surgeon (S.H.), who had knowledge of the angiographic status. Transmural biopsy specimens were obtained from the myocardial region surrounded by the left anterior descending coronary artery and the first or second diagonal branch. The site of the second biopsy was approximately 10 mm caudal to the first biopsy site.
The cylindrical samples were acquired by use of a Tru-Cut biopsy needle (Travenol Laboratories, Inc) with 1.5-mm luminal diameter. Biopsies were immediately fixed with glutaraldehyde, postfixed in osmium tetroxide, dehydrated in ethanol, and embedded in epoxy resin. As previously described,18 semithin sections (0.5 μm) were prepared and double-stained with paraphenylene-diamine and toluidine blue for the morphometric investigations.
Morphometric evaluations were performed by two investigators (G.Z., M.G.), who were unaware of the patients’ clinical or scintigraphic data, by use of standard light microscopy and a point-counting system, as previously described.12 18
Severe contraction bands, which are caused by mechanical disruption and are usually located at the border of the tissue samples, may occur artificially. Thus, only the central thirds of the sections were evaluated. Morphometry was performed with an eyepiece (Zeiss Integration Ocular, Zeiss) that provided 100 intersection points. At least 10 test areas were counted at ×160 magnification, and the average interstitial nonmuscular tissue was calculated independently for the two transmural biopsy samples. For the statistical analyses, the average volume fraction of the interstitial nonmuscular tissue obtained in the two biopsy specimens was used.
The normal range for the volume fraction of myocardial interstitial fibrosis was considered to be ≤12 vol%. This range was estimated from the structural data previously gathered in our laboratory from normal donor hearts before cardiac transplantation by applying identical morphometric methods.12 A normal range of ≤12 vol% is also in accordance with the data of other investigators, as assessed from autopsy or intraoperative myocardial samples.19 20 21 22
Continuous data are given as mean±1 SD. Correlations between the tracer deposition and the structural data were determined from least-squares linear regression analysis. The probability for a statistically significant correlation was calculated by the (nonparametric) Spearman rank statistics. To test for significant differences between means, Student’s t test for unpaired data or ANOVA followed by Bonferroni’s t test statistics was used as appropriate. Comparison between the regression lines was performed by Student’s t test.
Statistical analyses were performed with the SAS statistical software package (SAS Institute Inc). For all tests, a value of P<.05 was considered indicative of a statistically significant difference.
A pathological volume fraction of interstitial fibrosis (>12 vol%) was obtained in each of the 37 patients (range, 14.5 vol% to 59.6 vol%). Interstitial fibrosis was most pronounced in patients with previous anterior Q-wave infarction and was significantly higher in this subgroup (P=.020) compared with patients with non–Q-wave infarction or those without previous infarction (Fig 2⇓).
Thallium Stress-Redistribution Imaging
Regional 201Tl activity in the anterolateral and septal myocardial segments was normal in 9 patients; 11 patients had reversible and 15 patients had irreversible perfusion defects (Table 1⇓). The irreversible defects were further graded as moderate in 8 patients and severe in 7 patients. Two patients were not assigned to a particular subgroup because their scintigraphic results varied in the different views (ie, a reversible 201Tl defect in one view and an irreversible defect in the other).
As listed in Table 1⇑, interstitial fibrosis in the anterior left ventricular wall was not significantly different between the subgroups of patients with normal regional tracer uptake, redistribution, or persistent defects in conventional stress-redistribution images; fibrosis tended only to be increased (P=.070) in patients with severe irreversible defects.
Tracer reinjection was performed in the 15 study patients with persistent defects in their conventional stress-redistribution images. Enhanced regional 201Tl activity after thallium was reinjected was obtained in 7 patients (5 with moderate and 2 with severe apparently irreversible defects on stress-redistribution images), whereas tracer activity was unaffected by reinjection in 8 patients (3 with moderate and 5 with severe defects). In these 8 patients, interstitial fibrosis was significantly higher (40.8±12.2 vol%) compared with the 7 patients with enhanced regional 201Tl uptake after tracer reinjection (27.5±8.4 vol%, P=.031), whereas the clinical and angiographic data did not show significant differences (Table 2⇓).
Relation Between Regional Thallium Activity and Interstitial Fibrosis
Among the 15 patients with persistent defects in their stress-redistribution images, a significant inverse relation between the amount of interstitial fibrosis and the level of the regional 201Tl activity in the anterior wall was obtained in both redistribution and reinjection images (Figs 3⇓ and 4⇓). The correlation coefficient, however, was significantly higher (P<.01) for the reinjection images (r=−.851) compared with the redistribution images (r=−.622).
The level of regional 201Tl activity in the poststress images was not related to the structural data.
The data from the present study demonstrate that the level of regional 201Tl activity in redistribution and reinjection images is related to the mass of preserved viable myocytes in poststenotic myocardium, and therefore this level may itself serve as a marker of myocardial viability. Compared with 4-hour delayed-redistribution imaging, tracer reinjection significantly improves the inverse relation between regional 201Tl activity and the extent of irreversible structural damage.
Fibrous tissue, generally considered a structural correlate of irreversible myocardial damage, represents the predominant component of the myocardial “interstitial nonmuscular space.”23 24 25 Therefore, in agreement with other investigators,20 21 26 27 the term “interstitial fibrosis” was applied to describe the interstitial nonmuscular tissue.
Measurement of cardiac fibrosis allows for assessment of the extent of irreversible myocardial damage and estimation of the mass of viable myocytes, the latter being approximately equal to 100 vol% minus the volume fraction of fibrosis. In coronary artery disease, cardiac fibrosis arises from both interstitial and reparative (ie, replacement) fibrosis. Interstitial fibrosis is thereby characterized as intermyocardiocytic fibrosis in the absence of cell necrosis. Collagen synthesis may be stimulated as an underlying mechanism either directly, by increased wall stress, or indirectly, by neural or endocrine activation.28 A possible mechanism of myocardial repair by scar formation is that necrotic myocardium is removed and replaced by granulation tissue.29 30 In addition to these primary processes, fibrotic sheets between capillaries and myocytes may inhibit the transcapillary substrate exchange and thus induce secondary myocyte necrosis.12 31 32
Several lines of experimental evidence suggest that, after coronary occlusion, ischemic myocytes do not die instantaneously: mildly ischemic myocytes may survive indefinitely, and within the region that is infarcted, not all myocytes die simultaneously.33 The present clinical data are in accordance with this experimental observance: in patients with ≥75% stenoses of the left anterior descending coronary artery, interstitial fibrosis in the anterior left ventricular wall varies in a continuous manner ranging from 14.5 vol% to 59.6 vol%.
As expected, interstitial fibrosis was most pronounced in patients with previous Q-wave myocardial infarction (Fig 2⇑). However, interstitial fibrosis was also increased above the normal range in each of the patients without previous anterior myocardial infarction. Because all patients had high-grade stenosis of the left anterior descending coronary artery, it seems conceivable that recurrent ischemic episodes have led to focal irreversible injury in the poststenotic myocardium, despite the lack of clinical evidence for myocardial infarction.21 34 A further explanation derives from the fact that 21 of the 37 patients (57%) had a history of previous posterior myocardial infarction, and increased wall stress due to compensatory workload may have stimulated collagen synthesis in the noninfarcted anterior left ventricular wall.
The introduction of the rest-reinjection technique has been demonstrated to improve precision when assessing myocardial viability with 201Tl scintigraphy.1 2 3 4 5 6 7 Two confirming aspects can be derived from the present data. First, interstitial fibrosis was significantly lower in myocardial segments with enhanced regional 201Tl activity after reinjection compared with segments with unchanged activity after reinjection (27.5±8.4 vol% versus 40.8±12.2 vol%, P=.031). Second, compared with the conventional redistribution images, the correlation between interstitial fibrosis and the level of 201Tl activity in irreversible perfusion defects is significantly improved after tracer reinjection (P<.01).
To minimize artificial damage, intraoperative assessment of the two transmural biopsy specimens was performed during cardiopulmonary bypass before cardioplegia was begun. A low sampling error has been obtained with Travenol biopsy needles in clinical investigations and in autopsy hearts.21 35 36 37 The inhomogeneous distribution of structural alterations in coronary artery disease, however, may cause an elevated sampling error. Therefore, in the present study, two transmural biopsy specimens were obtained from the center of the perfusion territory of the left anterior descending coronary artery and the average value of the volume fraction of the interstitial fibrous tissue was used for the statistical analyses.
Also, assessment of two biopsy specimens per patient allowed us to compare the intraindividual variance (differences between the two biopsy specimens) with the interindividual variance (differences among the individual patients) of the structural data; the intraindividual variance was significantly lower (P<.01). Furthermore, since interstitial fibrosis was significantly higher in patients with impaired compared with normal anterolateral left ventricular performance (37.8±9.2 vol%, n=16, versus 25.8±7.4 vol%, n=21; P<.001), this finding might also indicate a tolerable sampling error in the morphometric data.
The volume fraction of interstitial fibrosis was assessed by standard light microscopy as previously described18 and validated.12 Light microscopy was used because this technique permits the analysis of larger test areas than electron microscopy.
To eliminate the intrinsic variability of visual image interpretations, quantitative criteria were applied for the classification of the scintigraphic studies as those with either reversible or irreversible 201Tl defects.
With planar imaging, individual coronary artery perfusion territories are superimposed. This has to be considered in the present study because the majority of the patients suffered from multiple-vessel coronary artery disease, as evidenced by the more frequent indication for surgical revascularization in these patients.
To balance the effect of superimposition of different perfusion territories, we calculated the average value for the relative anteroseptal tracer activity from the anterior and 30° left anterior oblique views. As illustrated in Fig 1⇑, this allows comparison of tracer activity in the perfusion territory of the left anterior descending coronary artery with that in the other two vascular regions: the perfusion territory of the right coronary artery (inferior wall) in the anterior view and the perfusion territory of the left circumflex coronary artery (lateral wall) in the 30° left anterior oblique view.
In addition to the methodological considerations mentioned above, when interpreting the present data, one should further consider that patients with a left ventricular aneurysm or a global left ventricular ejection fraction <35% were not included in the study because transmural biopsies in these patients might be associated with a higher risk for complications. Also, the study population included only three women.
Conclusions and Clinical Implications
With scintigraphic methods, the common definition of myocardial viability is based on the evidence of elementary aspects of cell function, such as cell membrane integrity or preserved intermediary metabolism. Lack of evidence for the existence of one or both of these components is considered a marker of cell necrosis. In view of the present data, however, this binary classification into viable or nonviable segments appears to be an oversimplification and does not reflect the continuous nature of structural damage in coronary artery disease.
The present results indicate that residual tracer activity in irreversible 201Tl defects (particularly after tracer reinjection) is proportional to the mass of preserved viable myocytes and therefore provides information about viability within perfusion defects. This suggests that quantification of residual 201Tl activity after rest-reinjection is a suitable tool to assess the possible value of therapeutic strategies.
This study was supported by a grant from the Deutsche Forschungsgemeinschaft within the SFB 320 “Herzfunktion und ihre Regulation,” Ruprecht-Karls-Universität, Heidelberg, Germany. We gratefully acknowledge the excellent technical assistance of Dorina Mack (Department of Nuclear Medicine) and Winfried Nottmeyer (Institute of Pathology). 201Tl was supplied by Du Pont Pharma GmbH, Bad Homburg, Germany. IBM, IBM-AT, and PC-DOS are trademarks of International Business Machines Corp; MicroDELTA, of Computer Design and Applications, Inc; and MS-DOS, of Microsoft Corp.
Presented in part at the 66th Scientific Sessions of the American Heart Association, Atlanta, Ga, November 8-11, 1993.
- Received August 4, 1994.
- Revision received September 19, 1994.
- Accepted October 2, 1994.
- Copyright © 1995 by American Heart Association
Kayden DS, Sigal S, Soufer R, Mattera J, Zaret BL, Wackers FJ. Thallium-201 for assessment of myocardial viability: quantitative comparison of 24-hour redistribution imaging with imaging after reinjection at rest. J Am Coll Cardiol. 1991;18:1480-1486. Erratum, J Am Coll Cardiol. 1991;19:1121.
Civelek AC, Ozguven M, Sostre S, Wagner HN Jr, Becker LC. Comparison of rest/stress and delayed reinjection Tl-201 SPECT imaging for detecting ischemic myocardium. J Am Coll Cardiol. 1993;21:283A. Abstract.
Bartenstein P, Hasfeld M, Schober O, Matheja P, Schäfers M, Scheld H, Breithard G. Tl-201 reinjection and improvement of left ventricular function following revascularization. J Nucl Med. 1993;34:45P. Abstract.
Bonow RO, Dilsizian V, Cuocolo A, Bacharach SL. Identification of viable myocardium in patients with chronic coronary artery disease and left ventricular dysfunction: comparison of thallium scintigraphy with reinjection and PET imaging with F-18-fluorodeoxyglucose. Circulation. 1991;83:26-37.
Gould KL. Myocardial viability: what does it mean and how do we measure it? Circulation. 1991;83:333-335.
Dilsizian V, Freedman NM, Bacharach SL, Perrone-Filardi P, Bonow RO. Regional thallium uptake in irreversible defects: magnitude of change in thallium activity after reinjection distinguishes viable from nonviable myocardium. Circulation. 1992;85:627-634.
Sabia PJ, Powers ER, Ragosta M, Smith WH, Watson DD, Kaul S. Role of quantitative planar thallium-201 imaging for determining viability in patients with acute myocardial infarction and a totally occluded infarct-related artery. J Nucl Med. 1993;34:728-736.
Sheehan FH, Dodge HT, Mathey DG, Brown BG, Bolson EL. Application of the centerline method: analysis of change in regional left ventricular wall motion in serial studies. IEEE Comput. 1982;9-12.
Zimmer G, Zimmermann R, Hess OM, Schneider J, Kübler W, Krayenbühl HP, Hagl S, Mall G. Decreased concentration of myofibrils and myofiber hypertrophy are structural determinants of impaired left ventricular function in patients with chronic heart diseases: a multiple logistic regression analysis. J Am Coll Cardiol. 1992;20:1135-1142.
Zimmermann R, Rauch B, Kapp M, Bubeck B, Neumann FJ, Seitz F, Stokstad P, Mall G, Tillmanns H, Kübler W. Myocardial scintigraphy with iodine-123 phenylpentadecanoic acid and thallium-201 in patients with coronary artery disease: a comparative dual-isotope study. Eur J Nucl Med. 1992;19:946-954.
Watson DD, Campbell NP, Read EK, Gibson RS, Teates CD, Beller GA. Spatial and temporal quantitation of plane thallium myocardial images. J Nucl Med. 1981;22:577-584.
Appledorn CR, Oppenheim BE, Wellman HN. An automated method for the alignment of image pairs. J Nucl Med. 1980;21:165-167.
Zimmermann R, Tillmanns H, Knapp WH, Helus F, Georgi P, Rauch B, Neumann FJ, Girgensohn S, Maier-Borst W, Kübler W. Regional myocardial nitrogen-13 glutamate uptake in patients with coronary artery disease: inverse poststress relation to thallium-201 uptake in ischemia. J Am Coll Cardiol. 1988;11:549-556.
Ragosta M, Beller GA, Watson DD, Kaul S, Gimple LW. Quantitative planar rest-redistribution Tl-201 imaging in detection of myocardial viability and prediction of improvement in left ventricular function after coronary bypass surgery in patients with severely depressed left ventricular function. Circulation. 1993;87:1630-1641.
Mall G, Schwarz F, Derks H. Clinicopathologic correlations in congestive cardiomyopathy. Virchows Arch Pathol Anat. 1982; 397:67-82.
Schwarz F, Mall G, Zebe H, Schmitzer E, Manthey J, Scheurlen H, Kübler W. Determinants of survival in patients with congestive cardiomyopathy: quantitative morphologic findings and left ventricular hemodynamics. Circulation. 1984;70:923-928.
Hess OM, Schneider J, Nonogi H, Carroll JD, Schneider K, Turina M, Krayenbühl HP. Myocardial structure in patients with exercise-induced ischemia. Circulation. 1988;77:967-977.
Schaper J. Effects of multiple ischaemic events on human myocardium: an ultrastructural study. Eur Heart J. 1988;9:A141-A149.
Weber KT, Brilla CG. Pathological hypertrophy and cardiac interstitium: fibrosis and renin-angiotensin-aldosterone system. Circulation. 1991;83:1849-1865.
Berry JJ, Hoffman JM, Steenbergen C, Baker JA, Floyd JA, van Trigt P, Hanson MW, Coleman RE. Human pathologic correlation with PET in ischemic and nonischemic cardiomyopathy. J Nucl Med. 1993;34:39-47.
Kunkel B, Schneider M. Myocardial structure and left ventricular function in hypertrophic and dilative cardiomyopathy and aortic valve disease. Z Kardiol. 1987;76(suppl 3):9-13.
Krayenbühl HP, Hess OM, Monrad ES, Schneider J, Mall G, Turina M. Left ventricular myocardial structure in aortic valve disease before, intermediate, and late after aortic valve replacement. Circulation. 1989;79:744-755.
Rosengurt E. Early signals in the mitogenic response. Science. 1986;234:161-166.
Reimer KA, Jennings RB. Myocardial ischemia, hypoxia, and infarction. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, eds. The Heart and the Cardiovascular System. New York, NY: Raven Press; 1986:1133-1201.
Reimer KA, Lowe JE, Rasmussen MM, Jennings RB. The wavefront phenomenon of ischemic cell death: myocardial infarct size vs duration of coronary occlusion in dogs. Circulation. 1977;56:786-794.
Schwarz F, Schaper J, Becker V, Kübler W, Flameng W. Coronary collateral vessels: their significance for left ventricular function. Circulation. 1977;56:786-794.