M Y O C A R D I A L I N F A R C T SIZE: CLINICOPATHOLOGIC AGREEMENT AND DISCORDANCE* PauI J. Boo~, M.D.,t and Edward S. Reynolds, M.D."
Abstract An accurate postmorteln method of planimetrically estimating tim extent of myocardial infarction was employed in 16 cases. Delineation of necrotic myocardiuln was enlmnced by a macroscopic stailfing technique, wlfich utilizes a tetrazoliuln dye. Comparison of infarct size with peak serum creatine phosphokinase levels showed a general correlation between the two that was not statistically significant. Two lnarkedly disparate cases serve to emphasize the need for clinical awareness of the temporal relationship between myocardial infarction and creatinine phosphokinase analysis as well as the possibility of other anatomic sources of elevation of serum enzyme levels. Comparison of infarct sizes in cardiogenic shock and nonshock patients confirms the existence of a significant relationslfip between a larger myocardial infarct and shock. However, the data from several patients in the group again emplmsize the possibility of maintaining a reasonable blood pressure in the face of a massive myocardial.infarction or, more importantly, of manifesting "cardiogenic" shock when only a small amount of left ventricular damage lms been sustained. The latter possibility may be related to other anatomic events, e,g., bowel infarction, hemorrhage, or possibly right ventricular ischemia, infarction, or dysfunction.
Attempts have recently been made to plmrlnacologically aher tim dimensions of myocardial infarcts on the assumption that linfiting infarct size will favorably alter the natural course of imman myocardial infarction) '2 Such clinical attempts make it necessary for the pathologist to assess myocardial damage lnore accurately at postmortem examination.
We have developed and tested a method to quantitate myocardial nmss and infarct mass by planimetric lneasurement of successive myocardial slices of known tlfickness. T h e delineation of necrotic tissue in these heart slices is enhanced by staining with a tetrazolium dye. The mathematical model upon which tim method is based and the overall accuracy
*Study supported by grants ttL-07066 and t!1.-06370 fronl the National Institutesof lleahh and carried out at the Peter Bent Brigham Hospital,Boston,Massachusetts. tAssistant l'rofessor, Departnmnt of l'athology,The Universityof Texas MedicalBranch, GalvestOil, Texas. w of Pathology and Chairperson, Department of Pathology, The Univi~rsity of Texas Medical Branch, Galveston, Texas.
685
HUMAN PATHOLOGY--VOLUME 8, NUMBER 6 November 1977 of these techniques have recently been described? Tiffs planimetric method as well as similar morphometric analyses has been applied to experimental cardiac material.4 Tlfis report describes the relationslfip between myocardial infarct size measured at postmortem examin'ation and common clinical premortem parameters of myocardial injury in 16 cases of myocardial infarction.
Hearts from 16 patients with the clinical diagnosis of acute myocardial inFarction at the time of death were randomly selected for study. Clinical data were obtained by a review of hospital records. Each patient had a clinical history of acute severe chest pain or sudden onset of shortness of breath before admission to the hospital. T h e patients in this group survived these initial clinical symptoms for three to 14 days. All patients were admitted to the cardiac intensive care ward of our hospital, although two patients were first admitted to an outside hospital and then transferred to our hospital with the diagnosis of acute myocardial infarction and cardiogenic shock. Nine Of the 16 patients were considered to be in "cardiogenic shock" clinically. Four of these patients were treated by intra-aortic balloon counterpulsation. T h e mean systolic blood pressure of patients in shock was 86 __+ 5 ram. Hg. Multiple serum enzyme levels, including creatine phosphokinase, lactate dehydrogenase, glutamic oxaloacetic (aspartate) transanfi!mse, and glutanfic pyruvic (alanine) transaminase, were deternfined in 15 patients in the series b), the routine clinical laboratory techniques. Each death occurred in the hospital, and complete postmortem examinations were performed on all patients within 18 hours,of death.
First, tile complete coronary arterial tree is dissected free from the heart and is fixed for later examination by standard microscopic techniques. T h e n the fresh whole heart is marked at 1 cm. intervals from the apex to the base along the entire posteroinferior (diaplmgmatic) surface and is cut carefidly in parallel sections, or "breadloafed." Sections are incubated for 20 to 30 minutes at room temperature in a I per cent triphenyl tetrazolium chloride solution buffered in 0.2 M tris-buffer to pH 7.8 at 25 ~ C. This macroscopic staining technique is a modification of the method originally described by Jestadt and Sandtitter '~ and has been utilized for sometime as an aid in vistmlizatiou of infarcted myocardium. G The triphenyhetrazolium chloride dye is reduced in the presence of tissue dehydrogenase enzymes, imparting a dark red color to noninfarcted myocardium (Figs. 2, 3, 6). For tiffs stud)', all slices of myocardium were totally immersed in the dye solution and were separated by screens to prevent nonstaining of myocardium due to either the tissue floating to the top of the container or contact between cut surfaces. Tlfis apparatus was employed because right ventricular nonstaining was noted in the preliminary trials of the stain at this institution,6 and at tlmt time floating of tissue slices above the fluid level or prolonged contact between slices during staining could not be ruled out as causes of nonstaining. Immediately after staining, transparencies (Kodacllrome-EHB) of the basal surface of each slice were made. Muhiple microscopic sections were taken to substantiate evidence of cellular necrosis in areas of nonstaining. Th.e transparencies of stained heart slices were projected in a rear projection theater and cross sectional areas were planimetrically measured (Fig. 1). Area measnrements were converted with the assistance of a Wang 600 programmable calculator to total myocardial weight and infarct weight based on a mathematical model previously described, a
Pathology Procedures
Statistical Considerations
Cases of clinically suspected myocardial infarction that come to autopsy at our hospital undergo special dissection and macroscopic staining procedures.
Tile total weight of tile left ventricle is given in grams tllroughout this presentation whereas myocardial infarct weight is reported as grams per kilogram of body
METHODS Patient Selection and Clinical Data
686
MYOCARDIAL
INFARCT
SIZE--BooR,
REYNOLDS
Figure 1. Planimetric apparatns used to measure tile extent o f myocardial infarction. T h e data display is immediately below the rear projection theater. T r a n s p a r e n c i e s o f successive cardiac slices are projected on tile screen and areas are traced with the planimeter a r m (below).
687
HUMAN PATHOLOGY--VOLUME
8, N U M B E R 6
November 1977
Figure 2. Example of a small posterior myocardial infarction limited to tile inner half of tile left ventricular wall. Top: A slice of m)ocardium before triphenyl tetrazolium cldoride staining (patient 6). Bottom:The same slice with the small pale area of nonstaining representing an area of myocardium deficient in dehydrogenase enzymes (arrow). Visualization of necrotic myocardium is greatl)' enhanced by this staining techtfique, thus facilitating planimetric measurement.
688
MYOCARDIAL INFARCT SIZE-BooR, REV,~OLDS
Figure 3. Massivemyocardial infarction involving the entire circumference of the left ventricle. The infarct is focallytransmural. A small m-ea of endocardial scarring (arrow) represents old and healed myocardial damage.
weight. In this manner, infarct size was adjusted for body weight, affording.a more accurate comparison of infarct s~ze with serum enzynte concentrations, which vary with the vascular distribution space. Mean values are expressed as the mean ___ SEM. Tests of statistical significance were also p e r f o r m e d utilizing a desktop calculator and include Student's t-test and linear regression analysis.
RESULTS
Infarct sizes d e t e r m i n e d by o u r planimetric m e t h o d varied from small focal infarctions o f 0.12 and 0.26 nag. pei" kg. of body weight (patients 9 and 6) to m~assive infarctions involving the majority of the left ventricle, as in patients 1 , 4 , and 10 (Table 1). T h e mean infarct size o f the group in shock was 1.17___ 0.13 gm. per kg. of body weight or 53 _+ 8 per cent of the measured left ventricular mass. This infarct size is statistically different ( p < 0.05) from the infarct size in the seven patients
not in cardiogenic shock. In that group tile mean infarct size was 0.65 + 0.18 gm. per kg. of body weight or 26 + 7 per cent o f t h e left ventricular myocardium. In both groups there were several patients who exhibited markedly discrepant infarct sizes (Fig. 4). All serum enzyme levels were elevated in the majority of patients, but linear regression analysis failed to show a significant relationship between the degree of elevation of any serum enzyme level and infarct size. Creatlne phosphokinase values, however, appeared to retlect the extent of myocardial infarction most accurately. In Figure 5 the highest creatine phosphokinase value for each patient is plotted against infarct size. Linear regression of these values gives a relatively poor correlation coefficient (r = 0,42), indicating no statistically significant linear relationship between infarct size and highest sermn level o f creatine phosphokinase recorded, ttowever, if the two highly disparate values in patients 6 and 12 (open circles) in Figure 5 are omitted (see Dis-
689
HUMAN PATHOLOGY--VOLUME T A B L E 1.
Series No.
8, N U M B E R 6
November 1977
CLINICAL DATA AND CALCULATED EXTENT OF MYOCARDIAL INFARCTION
Left Ventrieular Weight (gin.)
Infarct Gm./kg. % Cardioget, ic Body Weight LV Myocardium Shock
1
118
2 3 4 5
142 261 174 188
1.13 1.11 0.41 1,43 1.35
48 41 8 61 54
+ +
6
145
0.26
12
+
7 8 9 10
116 145 101 186 196
0.28 0.48 0.12 1.56 0.19 (LV) 0.94 (LV+ RV) 1.35 0.76 1.38 1.09 1,44
14 13 8 59
+
6
+
11
12 13 14 15 16
85 130 150 179 209
IABP*
Right Ventricular Nonstainlng?
+
+ + +
+
+ +
30 95 38 76 43 55
+
+ +
+
*Intra-aortic balloon counter pulsation. ~Staining by triphenyl tetrazolium chloride technique. See Methods.
1.5 9 p ! - 16 9
t'-
,.o
_z.. 0.5-
ept-6
Nonshock
Group
690
Cardiogenic
Shock
Figure 4. l'atients, with a clinical diagnosis of cardiogenic shock (mean systolic blood pressure: 86 • 5 ram. Hg) had statistically larger infarct sizes (expressed as gm./kg, of body weight) than the group without shock (p
Abstract An accurate postmorteln method of planimetrically estimating tim extent of myocardial infarction was employed in 16 cases. Delineation of necrotic myocardiuln was enlmnced by a macroscopic stailfing technique, wlfich utilizes a tetrazoliuln dye. Comparison of infarct size with peak serum creatine phosphokinase levels showed a general correlation between the two that was not statistically significant. Two lnarkedly disparate cases serve to emphasize the need for clinical awareness of the temporal relationship between myocardial infarction and creatinine phosphokinase analysis as well as the possibility of other anatomic sources of elevation of serum enzyme levels. Comparison of infarct sizes in cardiogenic shock and nonshock patients confirms the existence of a significant relationslfip between a larger myocardial infarct and shock. However, the data from several patients in the group again emplmsize the possibility of maintaining a reasonable blood pressure in the face of a massive myocardial.infarction or, more importantly, of manifesting "cardiogenic" shock when only a small amount of left ventricular damage lms been sustained. The latter possibility may be related to other anatomic events, e,g., bowel infarction, hemorrhage, or possibly right ventricular ischemia, infarction, or dysfunction.
Attempts have recently been made to plmrlnacologically aher tim dimensions of myocardial infarcts on the assumption that linfiting infarct size will favorably alter the natural course of imman myocardial infarction) '2 Such clinical attempts make it necessary for the pathologist to assess myocardial damage lnore accurately at postmortem examination.
We have developed and tested a method to quantitate myocardial nmss and infarct mass by planimetric lneasurement of successive myocardial slices of known tlfickness. T h e delineation of necrotic tissue in these heart slices is enhanced by staining with a tetrazolium dye. The mathematical model upon which tim method is based and the overall accuracy
*Study supported by grants ttL-07066 and t!1.-06370 fronl the National Institutesof lleahh and carried out at the Peter Bent Brigham Hospital,Boston,Massachusetts. tAssistant l'rofessor, Departnmnt of l'athology,The Universityof Texas MedicalBranch, GalvestOil, Texas. w of Pathology and Chairperson, Department of Pathology, The Univi~rsity of Texas Medical Branch, Galveston, Texas.
685
HUMAN PATHOLOGY--VOLUME 8, NUMBER 6 November 1977 of these techniques have recently been described? Tiffs planimetric method as well as similar morphometric analyses has been applied to experimental cardiac material.4 Tlfis report describes the relationslfip between myocardial infarct size measured at postmortem examin'ation and common clinical premortem parameters of myocardial injury in 16 cases of myocardial infarction.
Hearts from 16 patients with the clinical diagnosis of acute myocardial inFarction at the time of death were randomly selected for study. Clinical data were obtained by a review of hospital records. Each patient had a clinical history of acute severe chest pain or sudden onset of shortness of breath before admission to the hospital. T h e patients in this group survived these initial clinical symptoms for three to 14 days. All patients were admitted to the cardiac intensive care ward of our hospital, although two patients were first admitted to an outside hospital and then transferred to our hospital with the diagnosis of acute myocardial infarction and cardiogenic shock. Nine Of the 16 patients were considered to be in "cardiogenic shock" clinically. Four of these patients were treated by intra-aortic balloon counterpulsation. T h e mean systolic blood pressure of patients in shock was 86 __+ 5 ram. Hg. Multiple serum enzyme levels, including creatine phosphokinase, lactate dehydrogenase, glutamic oxaloacetic (aspartate) transanfi!mse, and glutanfic pyruvic (alanine) transaminase, were deternfined in 15 patients in the series b), the routine clinical laboratory techniques. Each death occurred in the hospital, and complete postmortem examinations were performed on all patients within 18 hours,of death.
First, tile complete coronary arterial tree is dissected free from the heart and is fixed for later examination by standard microscopic techniques. T h e n the fresh whole heart is marked at 1 cm. intervals from the apex to the base along the entire posteroinferior (diaplmgmatic) surface and is cut carefidly in parallel sections, or "breadloafed." Sections are incubated for 20 to 30 minutes at room temperature in a I per cent triphenyl tetrazolium chloride solution buffered in 0.2 M tris-buffer to pH 7.8 at 25 ~ C. This macroscopic staining technique is a modification of the method originally described by Jestadt and Sandtitter '~ and has been utilized for sometime as an aid in vistmlizatiou of infarcted myocardium. G The triphenyhetrazolium chloride dye is reduced in the presence of tissue dehydrogenase enzymes, imparting a dark red color to noninfarcted myocardium (Figs. 2, 3, 6). For tiffs stud)', all slices of myocardium were totally immersed in the dye solution and were separated by screens to prevent nonstaining of myocardium due to either the tissue floating to the top of the container or contact between cut surfaces. Tlfis apparatus was employed because right ventricular nonstaining was noted in the preliminary trials of the stain at this institution,6 and at tlmt time floating of tissue slices above the fluid level or prolonged contact between slices during staining could not be ruled out as causes of nonstaining. Immediately after staining, transparencies (Kodacllrome-EHB) of the basal surface of each slice were made. Muhiple microscopic sections were taken to substantiate evidence of cellular necrosis in areas of nonstaining. Th.e transparencies of stained heart slices were projected in a rear projection theater and cross sectional areas were planimetrically measured (Fig. 1). Area measnrements were converted with the assistance of a Wang 600 programmable calculator to total myocardial weight and infarct weight based on a mathematical model previously described, a
Pathology Procedures
Statistical Considerations
Cases of clinically suspected myocardial infarction that come to autopsy at our hospital undergo special dissection and macroscopic staining procedures.
Tile total weight of tile left ventricle is given in grams tllroughout this presentation whereas myocardial infarct weight is reported as grams per kilogram of body
METHODS Patient Selection and Clinical Data
686
MYOCARDIAL
INFARCT
SIZE--BooR,
REYNOLDS
Figure 1. Planimetric apparatns used to measure tile extent o f myocardial infarction. T h e data display is immediately below the rear projection theater. T r a n s p a r e n c i e s o f successive cardiac slices are projected on tile screen and areas are traced with the planimeter a r m (below).
687
HUMAN PATHOLOGY--VOLUME
8, N U M B E R 6
November 1977
Figure 2. Example of a small posterior myocardial infarction limited to tile inner half of tile left ventricular wall. Top: A slice of m)ocardium before triphenyl tetrazolium cldoride staining (patient 6). Bottom:The same slice with the small pale area of nonstaining representing an area of myocardium deficient in dehydrogenase enzymes (arrow). Visualization of necrotic myocardium is greatl)' enhanced by this staining techtfique, thus facilitating planimetric measurement.
688
MYOCARDIAL INFARCT SIZE-BooR, REV,~OLDS
Figure 3. Massivemyocardial infarction involving the entire circumference of the left ventricle. The infarct is focallytransmural. A small m-ea of endocardial scarring (arrow) represents old and healed myocardial damage.
weight. In this manner, infarct size was adjusted for body weight, affording.a more accurate comparison of infarct s~ze with serum enzynte concentrations, which vary with the vascular distribution space. Mean values are expressed as the mean ___ SEM. Tests of statistical significance were also p e r f o r m e d utilizing a desktop calculator and include Student's t-test and linear regression analysis.
RESULTS
Infarct sizes d e t e r m i n e d by o u r planimetric m e t h o d varied from small focal infarctions o f 0.12 and 0.26 nag. pei" kg. of body weight (patients 9 and 6) to m~assive infarctions involving the majority of the left ventricle, as in patients 1 , 4 , and 10 (Table 1). T h e mean infarct size o f the group in shock was 1.17___ 0.13 gm. per kg. of body weight or 53 _+ 8 per cent of the measured left ventricular mass. This infarct size is statistically different ( p < 0.05) from the infarct size in the seven patients
not in cardiogenic shock. In that group tile mean infarct size was 0.65 + 0.18 gm. per kg. of body weight or 26 + 7 per cent o f t h e left ventricular myocardium. In both groups there were several patients who exhibited markedly discrepant infarct sizes (Fig. 4). All serum enzyme levels were elevated in the majority of patients, but linear regression analysis failed to show a significant relationship between the degree of elevation of any serum enzyme level and infarct size. Creatlne phosphokinase values, however, appeared to retlect the extent of myocardial infarction most accurately. In Figure 5 the highest creatine phosphokinase value for each patient is plotted against infarct size. Linear regression of these values gives a relatively poor correlation coefficient (r = 0,42), indicating no statistically significant linear relationship between infarct size and highest sermn level o f creatine phosphokinase recorded, ttowever, if the two highly disparate values in patients 6 and 12 (open circles) in Figure 5 are omitted (see Dis-
689
HUMAN PATHOLOGY--VOLUME T A B L E 1.
Series No.
8, N U M B E R 6
November 1977
CLINICAL DATA AND CALCULATED EXTENT OF MYOCARDIAL INFARCTION
Left Ventrieular Weight (gin.)
Infarct Gm./kg. % Cardioget, ic Body Weight LV Myocardium Shock
1
118
2 3 4 5
142 261 174 188
1.13 1.11 0.41 1,43 1.35
48 41 8 61 54
+ +
6
145
0.26
12
+
7 8 9 10
116 145 101 186 196
0.28 0.48 0.12 1.56 0.19 (LV) 0.94 (LV+ RV) 1.35 0.76 1.38 1.09 1,44
14 13 8 59
+
6
+
11
12 13 14 15 16
85 130 150 179 209
IABP*
Right Ventricular Nonstainlng?
+
+ + +
+
+ +
30 95 38 76 43 55
+
+ +
+
*Intra-aortic balloon counter pulsation. ~Staining by triphenyl tetrazolium chloride technique. See Methods.
1.5 9 p ! - 16 9
t'-
,.o
_z.. 0.5-
ept-6
Nonshock
Group
690
Cardiogenic
Shock
Figure 4. l'atients, with a clinical diagnosis of cardiogenic shock (mean systolic blood pressure: 86 • 5 ram. Hg) had statistically larger infarct sizes (expressed as gm./kg, of body weight) than the group without shock (p
