- Email: [email protected]
Early detection of successful coronary reperfusion based on serum concentration of human heart-type cytoplasmic fatty acid-binding protein
ELSEVIER
Clinica Chimica Acta 262 (1997) 13-27
Early detection of successful coronary reperfusion based on serum concentration of human heart-type cytoplasmic fatty acid-binding protein Junnichi Ishii a'*, Youichi Nagamura b, Masanori Nomura a, Jian-hua Wang a, Shinn Taga a, Masatomo Kinoshita a, Hiroshi Kurokawa a, Masatsugu Iwase a, Takeshi Kondo a, Yoshihiko Watanabe ~, Hitoshi Hishida ~, Takao Tanaka ~, Keishiro Kawamura c aDepartment of Internal Medicine, Fujita Health University School of Medicine, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake, Aichi 470-11, Japan bDepartment of Clinical Chemistry, Fujita Health University, School of Health Sciences, Toyoake, Aichi 470-11, Japan CThird Division, Department of Internal Medicine, Osaka Medical, College, Takatsuki, Osaka 569, Japan Received 9 August 1996; revised 27 December 1996; accepted 30 December 1996
Abstract
Both human heart-type cytoplasmic fatty acid-binding protein (H-FABPc) and myoglobin are low molecular weight proteins that are abundant in the cytoplasm of myocardial cells. Unlike myoglobin, H-FABPc content in the skeletal muscle is less than in cardiac muscle. To investigate the usefulness of the serum concentration of H-FABPc in the early detection of successful coronary reperfusion, we measured serum concentrations of H-FABPc and myoglobin in 45 patients with acute myocardial infarction treated with intracoronary thrombolysis or direct percutaneous transluminal coronary angioplasty. Coronary angiography was performed every 5 rain for reperfusion therapy to identify the onset of reperfusion. Reperfusion, defined as a TIMI grade 2 or 3, was achieved within 60 rain of the initiation of reperfusion therapy in 30 patients (the reperfused group), but was not achieved in 15 patients (the non-reperfused group). Blood samples were obtained before initiation of treatment and 15, 30 and 60 min after initiation of treatment in the non-reperfused group. In the reperfused group, samples were obtained before reperfusion and 15, 30 and 60 min after reperfusion. The H-FABPc ratio (the ratio of value after to value before the initiation of treatment or reperfusion) increased sharply after the onset of Corresponding author. Tel.: + 81 562 932312; fax: + 81 562 932315. 0009-8981/97/$17.00 Copyright © 1997 Elsevier Science B.V. All rights reserved Pll S0009-8981 (97)06547-9
14
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
reperfusion, peaking at 41 _-+_18 re.in, and decreased rapidly thereafter. The predictive accuracy of an H-FABPc ratio of > 1.8 for the detection of reperfusion within 60 min of the initiation of treatment was 93% at 15 min after reperfusion, 98% at 30 min, and 100% at 60 rnin. Similar rates of predictive accuracy were observed for a myoglobin ratio > 2.4. The H-FABPc ratio detected successful reperfusion as early as 15 rain after the onset of reperfusion and was highly accurate in detecting reperfusion within 60 min of the onset of reperfusion. The predictive accuracy of the H-FABPc ratio was similar to that of the myoglobin ratio for the early detection of successful coronary reperfusion. © 1997 Elsevier Science B.V.
Keywords: Human heart-type cytoplasmic fatty acid-binding protein; Myoglobin; Coronary reperfusion
1. Introduction
The usefulness of methods based on the washout phenomenon of biochemical markers, such as creatine kinase-2 (CK; EC 2.7.3.2) [1-4], myoglobin [5-9], troponin T [4], and CK-2 and CK-3 isoforms [6,10-13], for early evaluation of the results of reperfusion therapy has been investigated. Myoglobin may be an especially useful biochemical marker for early detection of myocardial cell damage and coronary reperfusion because of its low molecular weight, cytosolic localization and rapid assay method. We previously found that the serum concentration of myoglobin increased sharply after the onset of reperfusion, peaking at 44+--18 rain, and that measurements of serum myoglobin concentration accurately predicted the success of coronary reperfusion as early as 15 min after the onset of reperfusion [8]. Heart-type cytoplasmic fatty acid-binding protein (H-FABPc) has been proposed as an early marker of acute myocardial infarction [14-16]. This small (15 kDa) cytoplasmic protein is abundant in myocardial cells and is believed to be involved in the uptake, transport and metabolism of fatty acids, but its precise role has not been determined [17-19]. Unlike myoglobin, H-FABPc content in the skeletal muscle is less than in cardiac muscle [20-23]. Therefore, H-FABPc would be expected to be a more suitable marker than myoglobin for early detection of myocardial cell damage and coronary reperfusion. H-FABPc has been found to provide a higher level of accuracy than CK, CK-2 [14-16] and myoglobin [24] for early diagnosis of acute myocardial infarction. In the present study, we compared the predictive accuracy of serum concentrations of HFABPc and myoglobin for early diagnosis of coronary reperfusion to determine the diagnostic usefulness of the serum concentration of H-FABPc. We attempted to exactly determine the onset of reperfusion and to confirm the perfusion status of the infarct-related coronary artery by obtaining coronary angiograms every 5 min for 60 min after the initiation of reperfusion therapy.
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
15
We excluded from analysis those patients in whom angiography showed spontaneous reperfusion and those who demonstrated re-occlusion within 60 min of the initiation of treatment. In the present study, to observe the washout phenomenon of H-FABPc after reperfusion precisely and to evaluate how early H-FABPc can detect the success of reperfusion after the onset of reperfusion, we compared changes in H-FABPc after reperfusion in the reperfused group with those after the start of treatment in the non-reperfused group. Most previous studies observed the changes in biochemical markers after the start of reperfusion therapy in reperfused and non-reperfused patients. Therefore, to examine the accuracy of H-FABPc for the detection of successful reperfusion at 60 rain after the initiation of treatment, we also analyzed changes in H-FABPc at 60 min after the start of treatment in the reperfused and non-reperfused groups.
2. Methods
2.1. Study patients We studied 61 Japanese patients with an acute myocardial infarction and no history of previous myocardial infarction who received emergency coronary angiography within 6 h of the onset of chest pain. The diagnosis of acute myocardial infarction was based on the following criteria: (1) typical chest pain that lasted more than 30 min and which was not relieved by sublingual nitroglycerin, and (2) ST segment elevations of 0.1 mV or higher in at least two adjacent electrocardiographic leads. We excluded two patients with persistent shock from the present study because it was difficult to perform the study protocol completely. In addition, patients who showed spontaneous reperfusion of the infarct-related coronary artery before treatment (10 patients) and intermittent re-occlusion within 60 min of the initiation of reperfusion therapy (four patients) were excluded by means of serial coronary angiography. The remaining 45 patients (36 men and nine women, aged 41-76 years, mean (_S.D.) age, 60.0---9.8 years) had total occlusion of the infarct-related coronary artery that corresponded to electrocardiographic changes. No patients had frequent intramuscular injections, repeated courses of electric defibrillation, trauma, muscle disorders or renal failure. We explained the purpose of the study to the patients and the members of their families and obtained informed consent before beginning the study.
2.2. Reperfusion therapy Total occlusion of the infarct-related coronary artery was confirmed by coronary angiography, using Judkins' technique [25], before initiation of
16
J. Ishii et al. I Clinica Chimica Acta 262 (1997) 13-27
reperfusion therapy. Direct percutaneous transluminal coronary angioplasty (PTCA) was performed in seven patients. Intracoronary thrombolysis was administered to 38 patients, in whom tissue-type plasminogen activator (t-PA; tisokinase, in a total dose of 320-640 x 104 units) was infused into the coronary artery over 20-40 min. Rescue PTCA was performed according to clinical indications if reperfusion was not achieved within 60 min of initiation of thrombolytic therapy. All patients received an intravenous infusion of heparin for at least 48 h after completion of reperfusion therapy. The activated partial thromboplastin time was maintained at 1.5-2.0 times the upper limit of normal.
2.3. Coronary angiography All patients underwent coronary angiography every 5 min for 60 min after the start of reperfusion therapy to identify the onset of reperfusion and to confirm the perfusion status of the infarct-related artery. Coronary reperfusion was defined as an increase in the Thrombolysis in Myocardial Infarction Trial (TIMI) flow grade from 0 to 2 or 3 [26]. The time at which reperfusion was achieved was defined as the point at which reperfusion of the infarct-related coronary artery was initially confirmed by coronary angiography. Patients who achieved reperfusion within 60 min of the initiation of intracoronary thrombolysis or direct PTCA were assigned to the reperfused group (30 patients); patients in whom reperfusion was not achieved within 60 min were assigned to the non-reperfused group (15 patients).
2.4. Serum markers To determine the serum concentrations of H-FABPc and myoglobin, 6-ml blood samples were obtained just before initiation of treatment and 15, 30, 60, 120, 180 and 240 min after the initiation of treatment in the non-reperfused group. In the reperfused group, samples were obtained before the onset of reperfusion and 15, 30, 60, 120, 180 and 240 min after the onset of reperfusion. Blood samples were drawn via a sheath placed in the femoral vein, centrifuged at 1000 × g for 15 min, and stored at - 8 0 ° C until assayed. The serum concentration of H-FABPc was determined by a recently developed competitive enzyme immunoassay using an anti-human H-FABPc monospecific polyclonal antibody, the accuracy and reproducibility of which have been described previously [14]. The normal H-FABPc concentration is below 3.0/zg/1 [16]. The serum concentration of myoglobin was measured by the turbidimetric latex agglutination method (Mb-latex Seiken, Denka Seiken Co., Ltd., Tokyo, Japan) using an automatic chemical analyzer (30R, Toshiba Medical Corp., Tokyo, Japan) [8,9,27]. The time required to determine the
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
17
myoglobin concentration was approximately 10 min [8,9]. The upper limit of normal for myoglobin is 96/zg/1. The H-FABPc and myoglobin ratios were calculated as indices of coronary reperfusion. In the reperfused group, the H-FABPc and myoglobin ratios were obtained by dividing the H-FABPc and myoglobin concentrations at 15, 30, 60, 120, 180 and 240 min after the onset of reperfusion by the respective values obtained before the onset of reperfusion. In the non-reperfused group, the H-FABPc and myoglobin concentrations obtained at 15, 30, 60, 120, 180 and 240 min after initiation of treatment were divided by the values obtained just before initiation of treatment. In addition, in the reperfused group, the H-FABPc and myoglobin concentrations obtained at 60 min after initiation of treatment were also divided by the values obtained just before initiation of treatment. 2.5. Statistical analysis Data are the mean_S.D. The results are expressed in terms of sensitivity (number of true positive test results in all patients of the reperfused group), specificity (number of true negative test results in all patients of the nonreperfused group), and predictive accuracy (number of true positive and negative test results of all positive and negative test results observed). Differences in clinical characteristics between the reperfused and non-reperfused groups were analyzed by the nonpaired t-test, the X2-test and Fisher's exact test. Differences in the H-FABPc and myoglobin ratios between groups were compared by the nonpaired t-test or the Welch test. Simple regression analysis was used to determine the correlation between the H-FABPc and myoglobin ratios. The differences in sensitivity, specificity and predictive accuracy between H-FABPc and myoglobin ratios were evaluated by the sign test. A P level < 0.05 was accepted as statistically significant.
3. Results
3.1. Clinical characteristics There were no significant differences in age, gender, the location of the infarct-related coronary artery, or the time from the onset of symptoms to initiation of treatment between the reperfused (30 patients) and the nonreperfused (15 patients) groups (Table 1). The TIMI perfusion grade 60 min after the initiation of treatment was 0 in 14 patients, 1 in one patient, 2 in seven patients, and 3 in 23 patients. In the reperfused group, coronary reperfusion was achieved 21-+-13 min after the initiation of treatment. Reperfusion was observed at 25_-_12 min after the start of
18
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
Table 1 Patient characteristics
Number Age (years) Gender (M/F) Time to therapy after onset of chest pain (h) Time to reperfusion after onset of chest pain (h) Infarct-related artery LAD LCX LMT RCA Treatment direct PTCA t-PA
Reperfused group
Non-reperfusedgroup
30 61.0___9.9 23/7 3.5+_1.6 3.9 +_1.5
15 58.0+_9.2 13/2 3.6+_1.3
13 4 1 12
7 2 0 6
6 24
1 14
Values are mean+__S.D. LAD, left anterior descending coronary artery; LCX, left circumflex coronary artery; LMT, left main trunk; RCA, right coronary artery; PTCA, percutaneous transluminal coronary angioplasty; t-PA, tissue-type plasminogen activator. thrombolysis in 24 patients: seven patients had TIMI grade 2 and 17 patients had TIMI grade 3 at 60 min after the initiation of treatment. In the six patients in whom direct PTCA was successful, a TIMI grade 3 flow was obtained 5.3+-2.5 min after insertion of the guidewire. In the reperfused group, blood samples were obtained 4 5 - 7 2 min (mean, 60+-7 min) after initiation of therapy because blood sampling was performed 15, 30 and 60 min after the onset of reperfusion. In the non-reperfused group, 13 of 15 patients received rescue PTCA after the 60-min blood sample was obtained. Reperfusion was subsequently achieved in 11 of these 13 patients. There was no significant difference between the time from the onset of acute myocardial infarction to reperfusion in the reperfused group and the time from the onset of acute myocardial infarction to treatment in the non-reperfused group. There were no significant differences in the H-FABPc and myoglobin concentrations before reperfusion in the reperfused group (25.8___ 15.9 and 300+__260/zg/1, respectively) and before initiation of treatment in the non-reperfused group (20.5+-12.4 and 256+-200 /zg/1, respectively). 3.2. H - F A B P c and myoglobin ratios in the reperfused and the non-reperfused groups
The H-FABPc ratio was significantly correlated with the myoglobin ratio (H-FABPc ratio = 0.74 X myoglobin ratio + 0.44, r = 0.88, n = 300; P < 0.01). In the reperfused group, the H-FABPc ratio was 3.9+--2.2 at 15 min after reperfusion, 4.6+-2.8 at 30 min and 4.6+-2.3 at 60 min; these values were
19
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
significantly (P < 0.01) higher than those obtained after treatment in the nonreperfused group: 1.2___0.2, 1.3__+0.2 and 1.4___0.2, respectively (Fig. 1). The myoglobin ratio 15 min after reperfusion in the reperfused group was 4.9__-3.1, that 30 min after reperfusion was 6.1 __+4.2, and that 60 min after reperfusion was 5.9-+-3.9, which was also significantly (P < 0.01) higher than the corresponding values in the non-reperfused group (1.4+0.4, 1.4-+-0.4 and 1.7-+-0.4, respective]y). The H-FABPc ratio at 60_+7 min after initiation of treatment in the reperfused group (4.7_+2.8) was significantly (P < 0.01) higher than the H-FABPc ratio at
H-FABPc ratio
I
I
"i
.
I
I
I
15
10 •
8
|
e
! "
•
T
18
.
|
' • = 1
,.T ../
v/
.I.I ~,.i
..
5 o
1 . 8 =~
. . . . . .
,,,I ~I
. . . . . . . . .
I
15min
.
"T
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I
30min
I
60rain
L reperfused group - I time after onset 1 of reperfusion J
. . . . . . .
I
15min
. . . . . . . .
I
30min
I
60min
L non reperfusedgroupJ rtime after initiation] L of treatment ,J
Fig. 1. H-FABPc ratio in the reperfused and non-reperfused groups. H-FABPc ratio at 15, 30 and 60 rain after reperfusion in the reperfused group, and the H-FABPc ratio at 15, 30 and 60 min after the initiation of treatment in the non-reperfused group. Values are mean_S.D.; *P < 0.01.
20
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
60 min after initiation of treatment in the non-reperfused group (1.4±0.2). The myoglobin ratio at 60±7 min after initiation of treatment in the reperfused group (6.0±3.9) was also significantly (P < 0.01) higher than the ratio at 60 min after initiation of treatment in the non-reperfused group (1.7±0.4).
3.3. Detection of coronary reperfusion There were few overlaps in H-FABPc ratio at all evaluation periods between the reperfused and non-reperfused groups, indicating that the H-FABPc ratio could accurately detect successful reperfusion 15, 30 and 60 min after reperfusion (Fig. 1). When an H-FABPc ratio > 1.8, which represented the mean + 2S.D. of the H-FABPc ratio at 60 min in the non-reperfused group, was used as the criterion for coronary reperfusion within 60 min of initiating treatment, the sensitivity, specificity and predictive accuracy were 90, 100 and 93%, respectively, at 15 min after reperfusion; 97, 100 and 98%, respectively, at 30 min; and 100, 100 and 100%, respectively, at 60 min. The sensitivity, specificity and predictive accuracy of an H-FABPc ratio > 1.8 for detection of reperfusion at 60+7 rain after the start of treatment were 97, 100 and 98%, respectively. Like H-FABPc ratio, few overlaps were found in myoglobin ratio at all evaluation periods between the reperfused and non-reperfused groups, indicating that the myoglobin ratio also could accurately detect successful reperfusion 15, 30 and 60 min after reperfusion. A myoglobin ratio > 2 . 4 had similar sensitivity, specificity and predictive accuracy as a marker of successful coronary reperfusion within 60 min of initiation of treatment (Table 2).
3.4. H-FABPc and myoglobin ratios in relation to TIMI perfusion grades 2 and 3 The H-FABPc ratios in patients with a TIMI grade 3 at 15, 30 and 60 min after reperfusion were significantly higher than in patients with a TIMI grade 2 Table 2 Sensitivity, specificity and predictive accuracy in detection of coronary reperfusion within 60 min of start of treatment Sensitivity Time after reperfusion (min): H-FABPc ratio (%) Myoglobin ratio (%)
15 90 90
Time after start of treatment (min): H-FABPc ratio (%) Myoglobin ratio (%)
60---7 97 93
30 97 93
Specificity 60 100 100
15 100 100 60+--7 100 100
30 100 100
Predictive accuracy 60 100 100
15 93 93 60-+7 98 96
30 98 96
60 100 100
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
21
flow: 4.5_+2.3 vs. 2.2_+0.3 at 15 min, 5.2_+2.9 vs. 2.7_+0.8 at 30 min, and 5.1---2.4 vs. 3.1 _+0.6 at 60 min (P < 0.01 ). The myoglobin ratios in patients with a TIMI grade 3 at 15, 30 and 60 min after reperfusion were also significantly higher than in patients with a TIMI grade 2 flow: 5.4___2.5 vs. 2.4_+0.5 at 15 min (P < 0.01), 6.3_+3.3 vs. 3.6_+0.6 at 30 min (P < 0.05), and 6.0_+2.5 vs. 3.8_+0.8 at 60 rain (P < 0.01).
3.5. H-FABPc and myoglobin ratios in relation to treatment method in the reperfused group There were no significant differences in H-FABPc ratios between patients in the reperfused group treated with direct PTCA and those treated with thrombolysis at 15 (4.5-+2.1 vs. 3.8-+2.3), 30 (5.5___1.9 vs. 4.5_+3.0), or 60 rain (5.0---1.9 vs. 4.4_+2.2) after reperfusion. There were also no significant differences in myoglobin ratios between patients in the reperfused group treated by direct PTCA and those treated by thrombolysis at 15 (5.5_+1.7 vs. 4.5___2.7), 30 (6.6---2.8 vs. 5.4_3.3), or 60 min (6.2_+2.3 vs. 5.3_+2.4) after reperfusion. All patients who received direct PTCA had an H-FABPc ratio > 1.8 and a myoglobin ratio > 2.4. In the thrombolysis group, one patient had an H-FABPc ratio --< 1.8 and a myoglobin ratio -< 2.4 at 15 min after reperfusion, one patient had an H-FABPc ratio --- 1.8 at 15 min after reperfusion and a myoglobin ratio -< 2.4 at 15 and 30 min after reperfusion, and one patient had an H-FABPc ratio --< 1.8 and a myoglobin ratio -----2.4 at 15 and 30 min after reperfusion. All three of these patients had TIMI grade 2 reperfusion and showed a more marked delay on coronary angiograms than the other four patients in the thrombolysis group who had TIMI grade 2 perfusion.
3.6. Time course of changes in H-FABPc and myoglobin ratios H-FABPc ratio increased sharply after reperfusion, peaking at 41_+18 min after reperfusion, and decreased rapidly thereafter. The H-FABPc was 3.4_+ 1.5 at 120 min after reperfusion, 2.7--+1.5 at 180 min, and 2.1___1.3 at 240 rain. The H-FABPc ratio was --- 1.8 in 17% of the reperfused group at 120 min, 23% at 180 min, and 40% at 240 min after reperfusion. The H-FABPc ratio in the non-reperfused group increased slowly after initiation of treatment. When reperfusion was achieved by rescue PTCA, the H-FABPc ratio increased sharply, as seen in the reperfused group (Fig. 2). The myglobin ratio showed a similar time course of changes in H- FABPc ratio in the reperfused and non-reperfused groups. The myoglobin ratio peaked at 39_-_17 min after reperfusion. The myoglobin ratio was 3.7___1.9 at 120 min
22
J. lshii et al. I Clinica Chimica Acta 262 (1997) 13-27
H-FABPc ratio
8
6
4
0
' ' 1~o 60 120 time after onset of reperfusion (min)
H-FABPc ratio 8-
6
2 rescue PTCA 0
I
0
i
I
I
60 120 180 time after initiation of treatment (min)
Fig. 2. Time course of changes in H-FABPc ratio. Upper panel: changes in the H-FABPc ratio after reperfusion in the reperfused group: 0 , reperfused group (n = 30). Values are mean+__S.D. Lower panel: changes in the H-FABPc ratio after the initiation of treatment in the non-reperfused group: O, non-reperfused group (n = 15); [2, successful rescue PTCA (n = 11); I1, unsuccessful rescue PTCA (n = 2) or rescue PTCA not attempted (n = 2). Values are mean-+S.D.
after reperfusion, 3.0 + _ 1.5 at 180 min, and 2.5+- 1.3 at 240 min. T h e m y o g l o b i n ratio w a s --- 2.4 in 2 0 % o f the r e p e r f u s e d g r o u p at 120 min, in 2 7 % at 180 min, and in 4 3 % at 240 m i n after reperfusion.
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
23
4. Discussion The H-FABPc ratio was significantly correlated with the myoglobin ratio in the present study. Both the H-FABPc and myoglobin ratios increased sharply after reperfusion, peaking at 41 ± 18 and 3 9 +- 17 min, respectively, and decreased rapidly thereafter. Thus, the blood kinetics of the cytosolic low molecular weight proteins, H-FABPc and myoglobin, may be similar after reperfusion. Both H-FABPc and myoglobin ratios accurately detected reperfusion as early as 15 min after successful reperfusion and were highly accurate in detecting coronary reperfusion within 60 min of the onset of reperfusion. The accuracy of the H-FABPc ratio was similar to that of the myoglobin ratio for early detection of reperfusion within 60 min of initiation of treatment. However, because H-FABPc and myoglobin concentrations peaked soon after reperfusion, and were cleared rapidly from the circulation, the sensitivity of both ratios may decrease beyond 60 min after reperfusion. The H-FABPc and myoglobin ratios were significantly higher in patients with TIMI grade 3 than in patients with TIMI grade 2, suggesting that the washout of H-FABPc and myoglobin after reperfusion was limited by the presence of severe residual stenosis after reperfusion. There was no significant difference in the H-FABPc and myoglobin ratios between patients who received intracoronary thrombolysis and those who received direct PTCA in the reperfused group, suggesting that the increase in the serum concentration of H-FABPc and myoglobin is similar after direct PTCA and after intracoronary thrombolysis. There was a greater variation in H-FABPc and myoglobin ratios in the reperfused group than in the non-reperfused group. This variation may have been related to differences in infarct size, in the time between the onset of infarction and reperfusion, and in the extent of reperfusion injury, as well as to the degree of residual stenosis after reperfusion. Skeletal muscle injury, such as frequent intramuscular injections, repeated courses of electric defibrillation or muscle trauma, may have less of an effect on the serum concentration of H-FABPc than on the myoglobin concentration: the skeletal muscle content of myoglobin is approximately twice that in the heart [20,21], while the H-FABPc content in striated muscle is only 10-30% of that in human cardiac muscle [22,23]. One of the explanations for the similar accuracy of H-FABPc and myoglobin ratios is that no patients had skeletal muscle injury in the present study. To determine the usefulness of H-FABPc ratio for the early detection of successful coronary reperfusion in patients with skeletal muscle injury, an additional protocol that includes serial blood sampling is necessary in such patients. In the present study, the H-FABPc ratio detected reperfusion 36-+ 13 min after initiation of treatment. The sensitivity, specificity and predictive accuracy of an H-FABPc ratio > 1.8 were 97, 100 and 98%, respectively, when reperfusion
24
J. lshii et al. / Clinica Chimica Acta 262 (1997) 13-27
was evaluated 60 min after initiation of treatment. An H-FABPc ratio > 1.8 may not be an accurate indicator of reperfusion more than 60 min after initiation of treatment because the H-FABPc ratio increases even in patients without reperfusion in the course of the natural evolution of acute myocardial infarction. The H-FABPc ratio decreases more than 60 min after reperfusion, and there is interindividual variation in the occurrence of reperfusion after intravenous thrombolytic therapy [28,29]. Thus, measurement of H-FABPc at only two time points (before and 90 or 120 min after initiation of treatment), as is typical, may be insufficient. Our results suggest that serial measurements of H-FABPc are needed for accurate detection of coronary reperfusion more than 60 min after initiation of intravenous thrombolytic therapy. Theoretically, if the H-FABPc concentration is measured every 30 min after initiation of treatment and the H-FABPc concentration is divided by the value obtained 30 or 60 min earlier, successful reperfusion would be indicated by a ratio of H-FABPc > 1.8. This method would be applicable for evaluation of repel'fusion even 90 or 120 rain after the administration of a fibrinolytic agent. Application of the H-FABPc ratio as an early index of successful coronary repel'fusion requires a serial blood sampling scheme and a rapid assay procedure for H-FABPc. The competitive enzyme immunoassay based on a monospecific polyclonal anti-human H-FABPc antibody used in the present study is too complicated and time-consuming for use in an emergency setting [14,16]. However, a rapid and sensitive assay system is being developed to facilitate H-FABPc analysis in clinical practice [30,31]. There are limitations to the clinical application of the H-FABPc ratio for detecting coronary reperfusion. In the present study, false-negative H-FABPc ratios were obtained at 15 min in two patients and at both 15 and 30 min in one patient. False-negative myoglobin ratios were also obtained in these three patients. The degree of delay was more marked in these three patients than in the other reperfused patients who also had TIMI grade 2 perfusion. Thus, in the presence of a TIMI grade 2 perfusion associated with a severe delay on angiography, the washout of H-FABPc and myoglobin may be markedly decreased, and the H-FABPc and myoglobin ratios may not reach the cutoff level. However, this limitation is not necessarily serious. Patients with severe residual stenosis have a high frequency of re-occlusion [32,33]. Thus, patients with an H-FABPc ratio <-- 1.8 at 15 min after reperfusion may be at increased risk of re-occlusion and may require further therapy, such as rescue PTCA or a more aggressive pharmacological approach. It may be difficult to apply our method in patients with intermittent coronary occlusion [34], spontaneous reperfusion before treatment, early re-occlusion after successful reperfusion or the no-reflow phenomenon [35], because the release profiles of intramyocardial proteins, including H-FABPc and myoglobin, may vary in these patients. Washout of H-FABPc and myoglobin is due, to some
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
25
degree, to the presence of a rich collateral circulation [36]; thus, a rich collateral circulation may also lead to misdiagnosis of the perfusion status. None of our patients showed this condition. Like myoglobin concentrations, H-FABPc concentrations are difficult to interpret in the presence of renal failure because H-FABPc is normally cleared rapidly by the kidney [37]. In patients with renal failure, it may be difficult to evaluate the presence of coronary reperfusion using the H-FABPc ratio.
Acknowledgments We thank Mrs. Kumiko Asayama, Mr. Yasuhiko Ohkaru, Mr. Hiroo Ishii and Mr. Shinzo Nishimura for their technical assistance.
References [1] Garabedian HD, Gold HK, Yasuda T et al. Detection of coronary artery reperfusion with creatine kinase-MB determinations during thrombolytic therapy: correlation with acute angiography. J Am Coil Cardiol 1988;11:729-734. [2] Lewis BS, Ganz W, Laramee P, Cercek B, Hod H, Shah PK, Lew AS. Usefulness of a rapid initial increase in plasma creatine kinase activity as a marker of repeffusion during thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1988;62:20-24. [3] Ohman EM, Christenson RH, Califf RM et al. Noninvasive detection of reperfusion after thrombolysis based on serum creatine kinase MB changes and clinical variables. Am Heart J 1993;126:819-826. [4] Abe S, Arima S, Yamashita T et al. Early assessment of reperfusion therapy using cardiac troponin T. J Am Coil Cardiol 1994;23:1382-1389. [5] Ellis AK, Little T, Masud ARZ, Liberman HA, Morris DC, Klocke FJ. Early noninvasive detection of successful reperfusion in patients with acute myocardial infarction. Circulation 1988;78:1352-1357. [6] Abendschein DR, Ellis AK, Eisenberg PR, Klocke FJ, Sobel BE, Jaffe AS. Prompt detection of coronary recanalization by analysis of rates of change of concentrations of macromolecular markers in plasma. Coron Art Dis 1991;2:201-212. [7] Zabel M, Hohnloser SH, K6ter W, Prinz M, Kasper W, Just H. Analysis of creatine kinase, CK-MB, myoglobin, and troponin T time-activity curves for early assessment of coronary artery reperfusion after intravenous thrombolysis. Circulation 1993;87:1542-1550. [8] Ishii J, Nomura M, Ando T et al. Early detection of successful coronary reperfusion based on serum myoglobin concentration; comparison with serum creatine kinase isoenzyme MB activity. Am Heart J 1994;128:641-648. [9] Miyata M, Abe S, Arima S e t al. Rapid diagnosis of coronary reperfusion by measurement of myoglobin level every 15 rain in acute myocardial infarction. J Am Coil Cardiol 1994;23:1009-1015. [10] Seacord LM, Abendschein DR, Nohara R, Hartzler G, Sobel BE, Jaffe AS. Detection of reperfusion within 1 hour after coronary recanalization by analysis of isoforms of the MM creatine kinase isoenzyme in plasma. Fibrinolysis 1988;2:151- 156.
26
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
[11] Puleo PR, Perryman MB. Noninvasive detection of reperfusion in acute myocardial infarction based on plasma activity of creatine kinase MB subforms. J Am Coil Cardiol 1991;17:1047-1052. [12] Laperche T, Steg PG, Benessiano J e t al. Patterns of myoglobin and MM creatine kinase is0forms release early after intravenous thrombolysis or direct percutaneous transluminal coronary angioplasty for acute myocardial infarction, and implications for the early noninvasive diagnosis of reperfusion. Am J Cardiol 1992;70:1129-1134. [13] Abe S, Nomoto K, Arima Set al. Detection of reperfusion 30 and 60 minutes after coronary recanalization by a rapid new assay of creatine kinase isoforms in acute myocardial infarction. Am Heart J 1993;125:649-656. [14] Tanaka T, Hirota Y, Sohmiya K, Nishimura S, Kawamura K. Serum and urinary human heart fatty acid-binding protein in acute myocardial infarction. Clin Biochem 1991;24:195-201. [15] Kleine AH, Glatz JFC, Van Nieuwenhoven FA, Van der Vusse GJ. Release of heart fatty acid-binding protein into plasma after acute myocardial infarction in man. Mol Cell Biochem 1992;116:155-162. [16] Tsuji R, Tanaka T, Sohmiya K et al. Human heart-type cytoplasmic fatty acid-binding protein in serum and urine during hyperacute myocardial infarction. Int J Cardiol 1993;41:209-217. [17] Bass NM. The cellular fatty acid binding proteins: aspects of structure, regulation, and function. Int Rev Cytol 1988;111:143- 184. [ 18] Glatz JFC, Van der Vusse GJ. Cellular fatty acid-binding proteins: current concepts and future directions. Mol Cell Biochem 1990;98:237-251. [19] Veerkamp JH, Peeters RA, Maatman RGHJ. Structural and functional features of different types of cytoplasmic fatty acid-binding proteins. Biochim Biophys Acta 1991;1081:1-24. [20] Sylvtn C, Jansson E, B66k K. Myoglobin content in human skeletal muscle and myocardium: relation to fiber size and oxidative capacity. Cardiovasc Res 1984;18:443-446. [21] Lin L, Sylvtn C, Sotonyi P, Somogyi E, Kaljser L, Jansson E. Myoglobin content and citrate synthase activity in different parts of normal human heart. J Appl Physiol 1990;69:899-901. [22] Yoshimoto K, Tanaka T, Sohmiya K et al. Human heart-type cytoplasmic fatty acid-binding protein as an indicator of acute myocardial infarction. Heart Vessels 1995;10:304-309. [23] Van Nieuwenhoven FA, Kleine AH, Wodzig KWH et al. Discrimination between myocardial and skeletal muscle injury by assessment of the plasma ratio of myoglobin over fatty acid-binding protein. Circulation 1995;92:2848-2854. [24] Ishii J, Nomura M, Ando T, Hasegawa H, Kimura M, Kurokawa H, Kondo T, Watanabe Y, Hishida H. Heart fatty acid-binding protein can provide a higher level of accuracy than myoglobin for the early detection of acute myocardial infarction [abstract]. J Am Coil Cardiol 1995;25:147A. [25] Judkins MP. Selective coronary arteriography, part I: a percutaneous transfemoral technic. Radiology 1967;89:815-824. [26] The Thrombolysis in Myocardial Infarction Study Group. The Thrombolysis in Myocardial Infarction (TIMI) trial: phase I findings. N Eng J Med 1985;312:932-936. [27] Uji Y, Okabe H, Sugiuchi H, Sekine S. Measurement of serum myoglobin by a turbidimetric latex agglutination method. J Clin Lab Anal 1992;6:7-11. [28] Anderson JL, Rothbard RL, Hackworthy RA et al. Multicenter reperfusion trial of intravenous anisoylated plasminogen streptokinase activator complex (APSAC) in acute myocardial infarction: controlled comparison with intracoronary streptokinase. J Am Coll Cardiol 1988;11:1153-1163. [29] Chesebro JH, Knatterud G, Roberts R et al. Thrombolysis in myocardial infarction (TIMI) trial, phase I: a comparison between intravenous tissue plasrninogen activator and intravenous streptokinase. Circulation 1987;76:142-154.
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
27
[30] Ohkaru Y, Asayama K, Ishii H et al. Development of a sandwich enzyme-linked immunosorbent assay for the determination of human heart fatty acid-binding protein in plasma and urine by using two different monoclonal antibodies specific for human heart fatty acidbinding protein. J Immunol Methods 1995;178:99-111. [31] Roos W, Eymann E, Symannek M e t al. Monoclonal antibodies to human heart fatty acid-binding protein. J Immunol Methods 1995;183:149-153. [32] Harrison DG, Ferguson DW, Collins SM et al. Rethrombosis after reperfusion with streptokinase: importance of geometry of residual lesions. Circulation 1984;69:991-994. [33] Wall TC, Mark DB, Califf RM et al. Prediction of early recurrent myocardial ischemia and coronary reocclusion after successful thrombolysis: a qualitative and quantitative angiographic study. Am J Cardiol 1989;63:423-428. [34] Hackett D, Davies G, Chierchia S, Maseri A. Intermittent coronary occlusion in acute myocardial infarction. N Engl J Med 1987;317:1055-1059. [35] Kloner RA, Alker KJ. The effect of streptokinase on intramyocardial hemorrhage, infarct size, and the no-reflow phenomenon during coronary reperfusion. Circulation 1984;70:513521. [36] Hirai T, Fujita M, Sasayama S et al. Importance of coronary collateral circulation for kinetics of serum creatine kinase in acute myocardial infarction. Am J Cardiol 1987;60:446-450. [37] Sohmiya K, Tanaka T, Tsuji R et al. Plasma and urinary heart-type cytoplasmic fatty acid-binding protein in coronary occlusion and reperfusion induced myocardial injury model. J Mol Cell Cardiol 1993;25:1413-1426.
Clinica Chimica Acta 262 (1997) 13-27
Early detection of successful coronary reperfusion based on serum concentration of human heart-type cytoplasmic fatty acid-binding protein Junnichi Ishii a'*, Youichi Nagamura b, Masanori Nomura a, Jian-hua Wang a, Shinn Taga a, Masatomo Kinoshita a, Hiroshi Kurokawa a, Masatsugu Iwase a, Takeshi Kondo a, Yoshihiko Watanabe ~, Hitoshi Hishida ~, Takao Tanaka ~, Keishiro Kawamura c aDepartment of Internal Medicine, Fujita Health University School of Medicine, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake, Aichi 470-11, Japan bDepartment of Clinical Chemistry, Fujita Health University, School of Health Sciences, Toyoake, Aichi 470-11, Japan CThird Division, Department of Internal Medicine, Osaka Medical, College, Takatsuki, Osaka 569, Japan Received 9 August 1996; revised 27 December 1996; accepted 30 December 1996
Abstract
Both human heart-type cytoplasmic fatty acid-binding protein (H-FABPc) and myoglobin are low molecular weight proteins that are abundant in the cytoplasm of myocardial cells. Unlike myoglobin, H-FABPc content in the skeletal muscle is less than in cardiac muscle. To investigate the usefulness of the serum concentration of H-FABPc in the early detection of successful coronary reperfusion, we measured serum concentrations of H-FABPc and myoglobin in 45 patients with acute myocardial infarction treated with intracoronary thrombolysis or direct percutaneous transluminal coronary angioplasty. Coronary angiography was performed every 5 rain for reperfusion therapy to identify the onset of reperfusion. Reperfusion, defined as a TIMI grade 2 or 3, was achieved within 60 rain of the initiation of reperfusion therapy in 30 patients (the reperfused group), but was not achieved in 15 patients (the non-reperfused group). Blood samples were obtained before initiation of treatment and 15, 30 and 60 min after initiation of treatment in the non-reperfused group. In the reperfused group, samples were obtained before reperfusion and 15, 30 and 60 min after reperfusion. The H-FABPc ratio (the ratio of value after to value before the initiation of treatment or reperfusion) increased sharply after the onset of Corresponding author. Tel.: + 81 562 932312; fax: + 81 562 932315. 0009-8981/97/$17.00 Copyright © 1997 Elsevier Science B.V. All rights reserved Pll S0009-8981 (97)06547-9
14
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
reperfusion, peaking at 41 _-+_18 re.in, and decreased rapidly thereafter. The predictive accuracy of an H-FABPc ratio of > 1.8 for the detection of reperfusion within 60 min of the initiation of treatment was 93% at 15 min after reperfusion, 98% at 30 min, and 100% at 60 rnin. Similar rates of predictive accuracy were observed for a myoglobin ratio > 2.4. The H-FABPc ratio detected successful reperfusion as early as 15 rain after the onset of reperfusion and was highly accurate in detecting reperfusion within 60 min of the onset of reperfusion. The predictive accuracy of the H-FABPc ratio was similar to that of the myoglobin ratio for the early detection of successful coronary reperfusion. © 1997 Elsevier Science B.V.
Keywords: Human heart-type cytoplasmic fatty acid-binding protein; Myoglobin; Coronary reperfusion
1. Introduction
The usefulness of methods based on the washout phenomenon of biochemical markers, such as creatine kinase-2 (CK; EC 2.7.3.2) [1-4], myoglobin [5-9], troponin T [4], and CK-2 and CK-3 isoforms [6,10-13], for early evaluation of the results of reperfusion therapy has been investigated. Myoglobin may be an especially useful biochemical marker for early detection of myocardial cell damage and coronary reperfusion because of its low molecular weight, cytosolic localization and rapid assay method. We previously found that the serum concentration of myoglobin increased sharply after the onset of reperfusion, peaking at 44+--18 rain, and that measurements of serum myoglobin concentration accurately predicted the success of coronary reperfusion as early as 15 min after the onset of reperfusion [8]. Heart-type cytoplasmic fatty acid-binding protein (H-FABPc) has been proposed as an early marker of acute myocardial infarction [14-16]. This small (15 kDa) cytoplasmic protein is abundant in myocardial cells and is believed to be involved in the uptake, transport and metabolism of fatty acids, but its precise role has not been determined [17-19]. Unlike myoglobin, H-FABPc content in the skeletal muscle is less than in cardiac muscle [20-23]. Therefore, H-FABPc would be expected to be a more suitable marker than myoglobin for early detection of myocardial cell damage and coronary reperfusion. H-FABPc has been found to provide a higher level of accuracy than CK, CK-2 [14-16] and myoglobin [24] for early diagnosis of acute myocardial infarction. In the present study, we compared the predictive accuracy of serum concentrations of HFABPc and myoglobin for early diagnosis of coronary reperfusion to determine the diagnostic usefulness of the serum concentration of H-FABPc. We attempted to exactly determine the onset of reperfusion and to confirm the perfusion status of the infarct-related coronary artery by obtaining coronary angiograms every 5 min for 60 min after the initiation of reperfusion therapy.
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
15
We excluded from analysis those patients in whom angiography showed spontaneous reperfusion and those who demonstrated re-occlusion within 60 min of the initiation of treatment. In the present study, to observe the washout phenomenon of H-FABPc after reperfusion precisely and to evaluate how early H-FABPc can detect the success of reperfusion after the onset of reperfusion, we compared changes in H-FABPc after reperfusion in the reperfused group with those after the start of treatment in the non-reperfused group. Most previous studies observed the changes in biochemical markers after the start of reperfusion therapy in reperfused and non-reperfused patients. Therefore, to examine the accuracy of H-FABPc for the detection of successful reperfusion at 60 rain after the initiation of treatment, we also analyzed changes in H-FABPc at 60 min after the start of treatment in the reperfused and non-reperfused groups.
2. Methods
2.1. Study patients We studied 61 Japanese patients with an acute myocardial infarction and no history of previous myocardial infarction who received emergency coronary angiography within 6 h of the onset of chest pain. The diagnosis of acute myocardial infarction was based on the following criteria: (1) typical chest pain that lasted more than 30 min and which was not relieved by sublingual nitroglycerin, and (2) ST segment elevations of 0.1 mV or higher in at least two adjacent electrocardiographic leads. We excluded two patients with persistent shock from the present study because it was difficult to perform the study protocol completely. In addition, patients who showed spontaneous reperfusion of the infarct-related coronary artery before treatment (10 patients) and intermittent re-occlusion within 60 min of the initiation of reperfusion therapy (four patients) were excluded by means of serial coronary angiography. The remaining 45 patients (36 men and nine women, aged 41-76 years, mean (_S.D.) age, 60.0---9.8 years) had total occlusion of the infarct-related coronary artery that corresponded to electrocardiographic changes. No patients had frequent intramuscular injections, repeated courses of electric defibrillation, trauma, muscle disorders or renal failure. We explained the purpose of the study to the patients and the members of their families and obtained informed consent before beginning the study.
2.2. Reperfusion therapy Total occlusion of the infarct-related coronary artery was confirmed by coronary angiography, using Judkins' technique [25], before initiation of
16
J. Ishii et al. I Clinica Chimica Acta 262 (1997) 13-27
reperfusion therapy. Direct percutaneous transluminal coronary angioplasty (PTCA) was performed in seven patients. Intracoronary thrombolysis was administered to 38 patients, in whom tissue-type plasminogen activator (t-PA; tisokinase, in a total dose of 320-640 x 104 units) was infused into the coronary artery over 20-40 min. Rescue PTCA was performed according to clinical indications if reperfusion was not achieved within 60 min of initiation of thrombolytic therapy. All patients received an intravenous infusion of heparin for at least 48 h after completion of reperfusion therapy. The activated partial thromboplastin time was maintained at 1.5-2.0 times the upper limit of normal.
2.3. Coronary angiography All patients underwent coronary angiography every 5 min for 60 min after the start of reperfusion therapy to identify the onset of reperfusion and to confirm the perfusion status of the infarct-related artery. Coronary reperfusion was defined as an increase in the Thrombolysis in Myocardial Infarction Trial (TIMI) flow grade from 0 to 2 or 3 [26]. The time at which reperfusion was achieved was defined as the point at which reperfusion of the infarct-related coronary artery was initially confirmed by coronary angiography. Patients who achieved reperfusion within 60 min of the initiation of intracoronary thrombolysis or direct PTCA were assigned to the reperfused group (30 patients); patients in whom reperfusion was not achieved within 60 min were assigned to the non-reperfused group (15 patients).
2.4. Serum markers To determine the serum concentrations of H-FABPc and myoglobin, 6-ml blood samples were obtained just before initiation of treatment and 15, 30, 60, 120, 180 and 240 min after the initiation of treatment in the non-reperfused group. In the reperfused group, samples were obtained before the onset of reperfusion and 15, 30, 60, 120, 180 and 240 min after the onset of reperfusion. Blood samples were drawn via a sheath placed in the femoral vein, centrifuged at 1000 × g for 15 min, and stored at - 8 0 ° C until assayed. The serum concentration of H-FABPc was determined by a recently developed competitive enzyme immunoassay using an anti-human H-FABPc monospecific polyclonal antibody, the accuracy and reproducibility of which have been described previously [14]. The normal H-FABPc concentration is below 3.0/zg/1 [16]. The serum concentration of myoglobin was measured by the turbidimetric latex agglutination method (Mb-latex Seiken, Denka Seiken Co., Ltd., Tokyo, Japan) using an automatic chemical analyzer (30R, Toshiba Medical Corp., Tokyo, Japan) [8,9,27]. The time required to determine the
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
17
myoglobin concentration was approximately 10 min [8,9]. The upper limit of normal for myoglobin is 96/zg/1. The H-FABPc and myoglobin ratios were calculated as indices of coronary reperfusion. In the reperfused group, the H-FABPc and myoglobin ratios were obtained by dividing the H-FABPc and myoglobin concentrations at 15, 30, 60, 120, 180 and 240 min after the onset of reperfusion by the respective values obtained before the onset of reperfusion. In the non-reperfused group, the H-FABPc and myoglobin concentrations obtained at 15, 30, 60, 120, 180 and 240 min after initiation of treatment were divided by the values obtained just before initiation of treatment. In addition, in the reperfused group, the H-FABPc and myoglobin concentrations obtained at 60 min after initiation of treatment were also divided by the values obtained just before initiation of treatment. 2.5. Statistical analysis Data are the mean_S.D. The results are expressed in terms of sensitivity (number of true positive test results in all patients of the reperfused group), specificity (number of true negative test results in all patients of the nonreperfused group), and predictive accuracy (number of true positive and negative test results of all positive and negative test results observed). Differences in clinical characteristics between the reperfused and non-reperfused groups were analyzed by the nonpaired t-test, the X2-test and Fisher's exact test. Differences in the H-FABPc and myoglobin ratios between groups were compared by the nonpaired t-test or the Welch test. Simple regression analysis was used to determine the correlation between the H-FABPc and myoglobin ratios. The differences in sensitivity, specificity and predictive accuracy between H-FABPc and myoglobin ratios were evaluated by the sign test. A P level < 0.05 was accepted as statistically significant.
3. Results
3.1. Clinical characteristics There were no significant differences in age, gender, the location of the infarct-related coronary artery, or the time from the onset of symptoms to initiation of treatment between the reperfused (30 patients) and the nonreperfused (15 patients) groups (Table 1). The TIMI perfusion grade 60 min after the initiation of treatment was 0 in 14 patients, 1 in one patient, 2 in seven patients, and 3 in 23 patients. In the reperfused group, coronary reperfusion was achieved 21-+-13 min after the initiation of treatment. Reperfusion was observed at 25_-_12 min after the start of
18
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
Table 1 Patient characteristics
Number Age (years) Gender (M/F) Time to therapy after onset of chest pain (h) Time to reperfusion after onset of chest pain (h) Infarct-related artery LAD LCX LMT RCA Treatment direct PTCA t-PA
Reperfused group
Non-reperfusedgroup
30 61.0___9.9 23/7 3.5+_1.6 3.9 +_1.5
15 58.0+_9.2 13/2 3.6+_1.3
13 4 1 12
7 2 0 6
6 24
1 14
Values are mean+__S.D. LAD, left anterior descending coronary artery; LCX, left circumflex coronary artery; LMT, left main trunk; RCA, right coronary artery; PTCA, percutaneous transluminal coronary angioplasty; t-PA, tissue-type plasminogen activator. thrombolysis in 24 patients: seven patients had TIMI grade 2 and 17 patients had TIMI grade 3 at 60 min after the initiation of treatment. In the six patients in whom direct PTCA was successful, a TIMI grade 3 flow was obtained 5.3+-2.5 min after insertion of the guidewire. In the reperfused group, blood samples were obtained 4 5 - 7 2 min (mean, 60+-7 min) after initiation of therapy because blood sampling was performed 15, 30 and 60 min after the onset of reperfusion. In the non-reperfused group, 13 of 15 patients received rescue PTCA after the 60-min blood sample was obtained. Reperfusion was subsequently achieved in 11 of these 13 patients. There was no significant difference between the time from the onset of acute myocardial infarction to reperfusion in the reperfused group and the time from the onset of acute myocardial infarction to treatment in the non-reperfused group. There were no significant differences in the H-FABPc and myoglobin concentrations before reperfusion in the reperfused group (25.8___ 15.9 and 300+__260/zg/1, respectively) and before initiation of treatment in the non-reperfused group (20.5+-12.4 and 256+-200 /zg/1, respectively). 3.2. H - F A B P c and myoglobin ratios in the reperfused and the non-reperfused groups
The H-FABPc ratio was significantly correlated with the myoglobin ratio (H-FABPc ratio = 0.74 X myoglobin ratio + 0.44, r = 0.88, n = 300; P < 0.01). In the reperfused group, the H-FABPc ratio was 3.9+--2.2 at 15 min after reperfusion, 4.6+-2.8 at 30 min and 4.6+-2.3 at 60 min; these values were
19
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
significantly (P < 0.01) higher than those obtained after treatment in the nonreperfused group: 1.2___0.2, 1.3__+0.2 and 1.4___0.2, respectively (Fig. 1). The myoglobin ratio 15 min after reperfusion in the reperfused group was 4.9__-3.1, that 30 min after reperfusion was 6.1 __+4.2, and that 60 min after reperfusion was 5.9-+-3.9, which was also significantly (P < 0.01) higher than the corresponding values in the non-reperfused group (1.4+0.4, 1.4-+-0.4 and 1.7-+-0.4, respective]y). The H-FABPc ratio at 60_+7 min after initiation of treatment in the reperfused group (4.7_+2.8) was significantly (P < 0.01) higher than the H-FABPc ratio at
H-FABPc ratio
I
I
"i
.
I
I
I
15
10 •
8
|
e
! "
•
T
18
.
|
' • = 1
,.T ../
v/
.I.I ~,.i
..
5 o
1 . 8 =~
. . . . . .
,,,I ~I
. . . . . . . . .
I
15min
.
"T
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I
30min
I
60rain
L reperfused group - I time after onset 1 of reperfusion J
. . . . . . .
I
15min
. . . . . . . .
I
30min
I
60min
L non reperfusedgroupJ rtime after initiation] L of treatment ,J
Fig. 1. H-FABPc ratio in the reperfused and non-reperfused groups. H-FABPc ratio at 15, 30 and 60 rain after reperfusion in the reperfused group, and the H-FABPc ratio at 15, 30 and 60 min after the initiation of treatment in the non-reperfused group. Values are mean_S.D.; *P < 0.01.
20
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
60 min after initiation of treatment in the non-reperfused group (1.4±0.2). The myoglobin ratio at 60±7 min after initiation of treatment in the reperfused group (6.0±3.9) was also significantly (P < 0.01) higher than the ratio at 60 min after initiation of treatment in the non-reperfused group (1.7±0.4).
3.3. Detection of coronary reperfusion There were few overlaps in H-FABPc ratio at all evaluation periods between the reperfused and non-reperfused groups, indicating that the H-FABPc ratio could accurately detect successful reperfusion 15, 30 and 60 min after reperfusion (Fig. 1). When an H-FABPc ratio > 1.8, which represented the mean + 2S.D. of the H-FABPc ratio at 60 min in the non-reperfused group, was used as the criterion for coronary reperfusion within 60 min of initiating treatment, the sensitivity, specificity and predictive accuracy were 90, 100 and 93%, respectively, at 15 min after reperfusion; 97, 100 and 98%, respectively, at 30 min; and 100, 100 and 100%, respectively, at 60 min. The sensitivity, specificity and predictive accuracy of an H-FABPc ratio > 1.8 for detection of reperfusion at 60+7 rain after the start of treatment were 97, 100 and 98%, respectively. Like H-FABPc ratio, few overlaps were found in myoglobin ratio at all evaluation periods between the reperfused and non-reperfused groups, indicating that the myoglobin ratio also could accurately detect successful reperfusion 15, 30 and 60 min after reperfusion. A myoglobin ratio > 2 . 4 had similar sensitivity, specificity and predictive accuracy as a marker of successful coronary reperfusion within 60 min of initiation of treatment (Table 2).
3.4. H-FABPc and myoglobin ratios in relation to TIMI perfusion grades 2 and 3 The H-FABPc ratios in patients with a TIMI grade 3 at 15, 30 and 60 min after reperfusion were significantly higher than in patients with a TIMI grade 2 Table 2 Sensitivity, specificity and predictive accuracy in detection of coronary reperfusion within 60 min of start of treatment Sensitivity Time after reperfusion (min): H-FABPc ratio (%) Myoglobin ratio (%)
15 90 90
Time after start of treatment (min): H-FABPc ratio (%) Myoglobin ratio (%)
60---7 97 93
30 97 93
Specificity 60 100 100
15 100 100 60+--7 100 100
30 100 100
Predictive accuracy 60 100 100
15 93 93 60-+7 98 96
30 98 96
60 100 100
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
21
flow: 4.5_+2.3 vs. 2.2_+0.3 at 15 min, 5.2_+2.9 vs. 2.7_+0.8 at 30 min, and 5.1---2.4 vs. 3.1 _+0.6 at 60 min (P < 0.01 ). The myoglobin ratios in patients with a TIMI grade 3 at 15, 30 and 60 min after reperfusion were also significantly higher than in patients with a TIMI grade 2 flow: 5.4___2.5 vs. 2.4_+0.5 at 15 min (P < 0.01), 6.3_+3.3 vs. 3.6_+0.6 at 30 min (P < 0.05), and 6.0_+2.5 vs. 3.8_+0.8 at 60 rain (P < 0.01).
3.5. H-FABPc and myoglobin ratios in relation to treatment method in the reperfused group There were no significant differences in H-FABPc ratios between patients in the reperfused group treated with direct PTCA and those treated with thrombolysis at 15 (4.5-+2.1 vs. 3.8-+2.3), 30 (5.5___1.9 vs. 4.5_+3.0), or 60 rain (5.0---1.9 vs. 4.4_+2.2) after reperfusion. There were also no significant differences in myoglobin ratios between patients in the reperfused group treated by direct PTCA and those treated by thrombolysis at 15 (5.5_+1.7 vs. 4.5___2.7), 30 (6.6---2.8 vs. 5.4_3.3), or 60 min (6.2_+2.3 vs. 5.3_+2.4) after reperfusion. All patients who received direct PTCA had an H-FABPc ratio > 1.8 and a myoglobin ratio > 2.4. In the thrombolysis group, one patient had an H-FABPc ratio --< 1.8 and a myoglobin ratio -< 2.4 at 15 min after reperfusion, one patient had an H-FABPc ratio --- 1.8 at 15 min after reperfusion and a myoglobin ratio -< 2.4 at 15 and 30 min after reperfusion, and one patient had an H-FABPc ratio --< 1.8 and a myoglobin ratio -----2.4 at 15 and 30 min after reperfusion. All three of these patients had TIMI grade 2 reperfusion and showed a more marked delay on coronary angiograms than the other four patients in the thrombolysis group who had TIMI grade 2 perfusion.
3.6. Time course of changes in H-FABPc and myoglobin ratios H-FABPc ratio increased sharply after reperfusion, peaking at 41_+18 min after reperfusion, and decreased rapidly thereafter. The H-FABPc was 3.4_+ 1.5 at 120 min after reperfusion, 2.7--+1.5 at 180 min, and 2.1___1.3 at 240 rain. The H-FABPc ratio was --- 1.8 in 17% of the reperfused group at 120 min, 23% at 180 min, and 40% at 240 min after reperfusion. The H-FABPc ratio in the non-reperfused group increased slowly after initiation of treatment. When reperfusion was achieved by rescue PTCA, the H-FABPc ratio increased sharply, as seen in the reperfused group (Fig. 2). The myglobin ratio showed a similar time course of changes in H- FABPc ratio in the reperfused and non-reperfused groups. The myoglobin ratio peaked at 39_-_17 min after reperfusion. The myoglobin ratio was 3.7___1.9 at 120 min
22
J. lshii et al. I Clinica Chimica Acta 262 (1997) 13-27
H-FABPc ratio
8
6
4
0
' ' 1~o 60 120 time after onset of reperfusion (min)
H-FABPc ratio 8-
6
2 rescue PTCA 0
I
0
i
I
I
60 120 180 time after initiation of treatment (min)
Fig. 2. Time course of changes in H-FABPc ratio. Upper panel: changes in the H-FABPc ratio after reperfusion in the reperfused group: 0 , reperfused group (n = 30). Values are mean+__S.D. Lower panel: changes in the H-FABPc ratio after the initiation of treatment in the non-reperfused group: O, non-reperfused group (n = 15); [2, successful rescue PTCA (n = 11); I1, unsuccessful rescue PTCA (n = 2) or rescue PTCA not attempted (n = 2). Values are mean-+S.D.
after reperfusion, 3.0 + _ 1.5 at 180 min, and 2.5+- 1.3 at 240 min. T h e m y o g l o b i n ratio w a s --- 2.4 in 2 0 % o f the r e p e r f u s e d g r o u p at 120 min, in 2 7 % at 180 min, and in 4 3 % at 240 m i n after reperfusion.
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
23
4. Discussion The H-FABPc ratio was significantly correlated with the myoglobin ratio in the present study. Both the H-FABPc and myoglobin ratios increased sharply after reperfusion, peaking at 41 ± 18 and 3 9 +- 17 min, respectively, and decreased rapidly thereafter. Thus, the blood kinetics of the cytosolic low molecular weight proteins, H-FABPc and myoglobin, may be similar after reperfusion. Both H-FABPc and myoglobin ratios accurately detected reperfusion as early as 15 min after successful reperfusion and were highly accurate in detecting coronary reperfusion within 60 min of the onset of reperfusion. The accuracy of the H-FABPc ratio was similar to that of the myoglobin ratio for early detection of reperfusion within 60 min of initiation of treatment. However, because H-FABPc and myoglobin concentrations peaked soon after reperfusion, and were cleared rapidly from the circulation, the sensitivity of both ratios may decrease beyond 60 min after reperfusion. The H-FABPc and myoglobin ratios were significantly higher in patients with TIMI grade 3 than in patients with TIMI grade 2, suggesting that the washout of H-FABPc and myoglobin after reperfusion was limited by the presence of severe residual stenosis after reperfusion. There was no significant difference in the H-FABPc and myoglobin ratios between patients who received intracoronary thrombolysis and those who received direct PTCA in the reperfused group, suggesting that the increase in the serum concentration of H-FABPc and myoglobin is similar after direct PTCA and after intracoronary thrombolysis. There was a greater variation in H-FABPc and myoglobin ratios in the reperfused group than in the non-reperfused group. This variation may have been related to differences in infarct size, in the time between the onset of infarction and reperfusion, and in the extent of reperfusion injury, as well as to the degree of residual stenosis after reperfusion. Skeletal muscle injury, such as frequent intramuscular injections, repeated courses of electric defibrillation or muscle trauma, may have less of an effect on the serum concentration of H-FABPc than on the myoglobin concentration: the skeletal muscle content of myoglobin is approximately twice that in the heart [20,21], while the H-FABPc content in striated muscle is only 10-30% of that in human cardiac muscle [22,23]. One of the explanations for the similar accuracy of H-FABPc and myoglobin ratios is that no patients had skeletal muscle injury in the present study. To determine the usefulness of H-FABPc ratio for the early detection of successful coronary reperfusion in patients with skeletal muscle injury, an additional protocol that includes serial blood sampling is necessary in such patients. In the present study, the H-FABPc ratio detected reperfusion 36-+ 13 min after initiation of treatment. The sensitivity, specificity and predictive accuracy of an H-FABPc ratio > 1.8 were 97, 100 and 98%, respectively, when reperfusion
24
J. lshii et al. / Clinica Chimica Acta 262 (1997) 13-27
was evaluated 60 min after initiation of treatment. An H-FABPc ratio > 1.8 may not be an accurate indicator of reperfusion more than 60 min after initiation of treatment because the H-FABPc ratio increases even in patients without reperfusion in the course of the natural evolution of acute myocardial infarction. The H-FABPc ratio decreases more than 60 min after reperfusion, and there is interindividual variation in the occurrence of reperfusion after intravenous thrombolytic therapy [28,29]. Thus, measurement of H-FABPc at only two time points (before and 90 or 120 min after initiation of treatment), as is typical, may be insufficient. Our results suggest that serial measurements of H-FABPc are needed for accurate detection of coronary reperfusion more than 60 min after initiation of intravenous thrombolytic therapy. Theoretically, if the H-FABPc concentration is measured every 30 min after initiation of treatment and the H-FABPc concentration is divided by the value obtained 30 or 60 min earlier, successful reperfusion would be indicated by a ratio of H-FABPc > 1.8. This method would be applicable for evaluation of repel'fusion even 90 or 120 rain after the administration of a fibrinolytic agent. Application of the H-FABPc ratio as an early index of successful coronary repel'fusion requires a serial blood sampling scheme and a rapid assay procedure for H-FABPc. The competitive enzyme immunoassay based on a monospecific polyclonal anti-human H-FABPc antibody used in the present study is too complicated and time-consuming for use in an emergency setting [14,16]. However, a rapid and sensitive assay system is being developed to facilitate H-FABPc analysis in clinical practice [30,31]. There are limitations to the clinical application of the H-FABPc ratio for detecting coronary reperfusion. In the present study, false-negative H-FABPc ratios were obtained at 15 min in two patients and at both 15 and 30 min in one patient. False-negative myoglobin ratios were also obtained in these three patients. The degree of delay was more marked in these three patients than in the other reperfused patients who also had TIMI grade 2 perfusion. Thus, in the presence of a TIMI grade 2 perfusion associated with a severe delay on angiography, the washout of H-FABPc and myoglobin may be markedly decreased, and the H-FABPc and myoglobin ratios may not reach the cutoff level. However, this limitation is not necessarily serious. Patients with severe residual stenosis have a high frequency of re-occlusion [32,33]. Thus, patients with an H-FABPc ratio <-- 1.8 at 15 min after reperfusion may be at increased risk of re-occlusion and may require further therapy, such as rescue PTCA or a more aggressive pharmacological approach. It may be difficult to apply our method in patients with intermittent coronary occlusion [34], spontaneous reperfusion before treatment, early re-occlusion after successful reperfusion or the no-reflow phenomenon [35], because the release profiles of intramyocardial proteins, including H-FABPc and myoglobin, may vary in these patients. Washout of H-FABPc and myoglobin is due, to some
J. lshii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
25
degree, to the presence of a rich collateral circulation [36]; thus, a rich collateral circulation may also lead to misdiagnosis of the perfusion status. None of our patients showed this condition. Like myoglobin concentrations, H-FABPc concentrations are difficult to interpret in the presence of renal failure because H-FABPc is normally cleared rapidly by the kidney [37]. In patients with renal failure, it may be difficult to evaluate the presence of coronary reperfusion using the H-FABPc ratio.
Acknowledgments We thank Mrs. Kumiko Asayama, Mr. Yasuhiko Ohkaru, Mr. Hiroo Ishii and Mr. Shinzo Nishimura for their technical assistance.
References [1] Garabedian HD, Gold HK, Yasuda T et al. Detection of coronary artery reperfusion with creatine kinase-MB determinations during thrombolytic therapy: correlation with acute angiography. J Am Coil Cardiol 1988;11:729-734. [2] Lewis BS, Ganz W, Laramee P, Cercek B, Hod H, Shah PK, Lew AS. Usefulness of a rapid initial increase in plasma creatine kinase activity as a marker of repeffusion during thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1988;62:20-24. [3] Ohman EM, Christenson RH, Califf RM et al. Noninvasive detection of reperfusion after thrombolysis based on serum creatine kinase MB changes and clinical variables. Am Heart J 1993;126:819-826. [4] Abe S, Arima S, Yamashita T et al. Early assessment of reperfusion therapy using cardiac troponin T. J Am Coil Cardiol 1994;23:1382-1389. [5] Ellis AK, Little T, Masud ARZ, Liberman HA, Morris DC, Klocke FJ. Early noninvasive detection of successful reperfusion in patients with acute myocardial infarction. Circulation 1988;78:1352-1357. [6] Abendschein DR, Ellis AK, Eisenberg PR, Klocke FJ, Sobel BE, Jaffe AS. Prompt detection of coronary recanalization by analysis of rates of change of concentrations of macromolecular markers in plasma. Coron Art Dis 1991;2:201-212. [7] Zabel M, Hohnloser SH, K6ter W, Prinz M, Kasper W, Just H. Analysis of creatine kinase, CK-MB, myoglobin, and troponin T time-activity curves for early assessment of coronary artery reperfusion after intravenous thrombolysis. Circulation 1993;87:1542-1550. [8] Ishii J, Nomura M, Ando T et al. Early detection of successful coronary reperfusion based on serum myoglobin concentration; comparison with serum creatine kinase isoenzyme MB activity. Am Heart J 1994;128:641-648. [9] Miyata M, Abe S, Arima S e t al. Rapid diagnosis of coronary reperfusion by measurement of myoglobin level every 15 rain in acute myocardial infarction. J Am Coil Cardiol 1994;23:1009-1015. [10] Seacord LM, Abendschein DR, Nohara R, Hartzler G, Sobel BE, Jaffe AS. Detection of reperfusion within 1 hour after coronary recanalization by analysis of isoforms of the MM creatine kinase isoenzyme in plasma. Fibrinolysis 1988;2:151- 156.
26
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 13-27
[11] Puleo PR, Perryman MB. Noninvasive detection of reperfusion in acute myocardial infarction based on plasma activity of creatine kinase MB subforms. J Am Coil Cardiol 1991;17:1047-1052. [12] Laperche T, Steg PG, Benessiano J e t al. Patterns of myoglobin and MM creatine kinase is0forms release early after intravenous thrombolysis or direct percutaneous transluminal coronary angioplasty for acute myocardial infarction, and implications for the early noninvasive diagnosis of reperfusion. Am J Cardiol 1992;70:1129-1134. [13] Abe S, Nomoto K, Arima Set al. Detection of reperfusion 30 and 60 minutes after coronary recanalization by a rapid new assay of creatine kinase isoforms in acute myocardial infarction. Am Heart J 1993;125:649-656. [14] Tanaka T, Hirota Y, Sohmiya K, Nishimura S, Kawamura K. Serum and urinary human heart fatty acid-binding protein in acute myocardial infarction. Clin Biochem 1991;24:195-201. [15] Kleine AH, Glatz JFC, Van Nieuwenhoven FA, Van der Vusse GJ. Release of heart fatty acid-binding protein into plasma after acute myocardial infarction in man. Mol Cell Biochem 1992;116:155-162. [16] Tsuji R, Tanaka T, Sohmiya K et al. Human heart-type cytoplasmic fatty acid-binding protein in serum and urine during hyperacute myocardial infarction. Int J Cardiol 1993;41:209-217. [17] Bass NM. The cellular fatty acid binding proteins: aspects of structure, regulation, and function. Int Rev Cytol 1988;111:143- 184. [ 18] Glatz JFC, Van der Vusse GJ. Cellular fatty acid-binding proteins: current concepts and future directions. Mol Cell Biochem 1990;98:237-251. [19] Veerkamp JH, Peeters RA, Maatman RGHJ. Structural and functional features of different types of cytoplasmic fatty acid-binding proteins. Biochim Biophys Acta 1991;1081:1-24. [20] Sylvtn C, Jansson E, B66k K. Myoglobin content in human skeletal muscle and myocardium: relation to fiber size and oxidative capacity. Cardiovasc Res 1984;18:443-446. [21] Lin L, Sylvtn C, Sotonyi P, Somogyi E, Kaljser L, Jansson E. Myoglobin content and citrate synthase activity in different parts of normal human heart. J Appl Physiol 1990;69:899-901. [22] Yoshimoto K, Tanaka T, Sohmiya K et al. Human heart-type cytoplasmic fatty acid-binding protein as an indicator of acute myocardial infarction. Heart Vessels 1995;10:304-309. [23] Van Nieuwenhoven FA, Kleine AH, Wodzig KWH et al. Discrimination between myocardial and skeletal muscle injury by assessment of the plasma ratio of myoglobin over fatty acid-binding protein. Circulation 1995;92:2848-2854. [24] Ishii J, Nomura M, Ando T, Hasegawa H, Kimura M, Kurokawa H, Kondo T, Watanabe Y, Hishida H. Heart fatty acid-binding protein can provide a higher level of accuracy than myoglobin for the early detection of acute myocardial infarction [abstract]. J Am Coil Cardiol 1995;25:147A. [25] Judkins MP. Selective coronary arteriography, part I: a percutaneous transfemoral technic. Radiology 1967;89:815-824. [26] The Thrombolysis in Myocardial Infarction Study Group. The Thrombolysis in Myocardial Infarction (TIMI) trial: phase I findings. N Eng J Med 1985;312:932-936. [27] Uji Y, Okabe H, Sugiuchi H, Sekine S. Measurement of serum myoglobin by a turbidimetric latex agglutination method. J Clin Lab Anal 1992;6:7-11. [28] Anderson JL, Rothbard RL, Hackworthy RA et al. Multicenter reperfusion trial of intravenous anisoylated plasminogen streptokinase activator complex (APSAC) in acute myocardial infarction: controlled comparison with intracoronary streptokinase. J Am Coll Cardiol 1988;11:1153-1163. [29] Chesebro JH, Knatterud G, Roberts R et al. Thrombolysis in myocardial infarction (TIMI) trial, phase I: a comparison between intravenous tissue plasrninogen activator and intravenous streptokinase. Circulation 1987;76:142-154.
J. Ishii et al. / Clinica Chimica Acta 262 (1997) 1 3 - 2 7
27
[30] Ohkaru Y, Asayama K, Ishii H et al. Development of a sandwich enzyme-linked immunosorbent assay for the determination of human heart fatty acid-binding protein in plasma and urine by using two different monoclonal antibodies specific for human heart fatty acidbinding protein. J Immunol Methods 1995;178:99-111. [31] Roos W, Eymann E, Symannek M e t al. Monoclonal antibodies to human heart fatty acid-binding protein. J Immunol Methods 1995;183:149-153. [32] Harrison DG, Ferguson DW, Collins SM et al. Rethrombosis after reperfusion with streptokinase: importance of geometry of residual lesions. Circulation 1984;69:991-994. [33] Wall TC, Mark DB, Califf RM et al. Prediction of early recurrent myocardial ischemia and coronary reocclusion after successful thrombolysis: a qualitative and quantitative angiographic study. Am J Cardiol 1989;63:423-428. [34] Hackett D, Davies G, Chierchia S, Maseri A. Intermittent coronary occlusion in acute myocardial infarction. N Engl J Med 1987;317:1055-1059. [35] Kloner RA, Alker KJ. The effect of streptokinase on intramyocardial hemorrhage, infarct size, and the no-reflow phenomenon during coronary reperfusion. Circulation 1984;70:513521. [36] Hirai T, Fujita M, Sasayama S et al. Importance of coronary collateral circulation for kinetics of serum creatine kinase in acute myocardial infarction. Am J Cardiol 1987;60:446-450. [37] Sohmiya K, Tanaka T, Tsuji R et al. Plasma and urinary heart-type cytoplasmic fatty acid-binding protein in coronary occlusion and reperfusion induced myocardial injury model. J Mol Cell Cardiol 1993;25:1413-1426.