- Email: [email protected]
Heart-Type Fatty Acid Binding Protein as a marker of reperfusion after thrombolytic therapy
Clinica Chimica Acta 298 (2000) 85–97 www.elsevier.com / locate / clinchim
Heart-Type Fatty Acid Binding Protein as a marker of reperfusion after thrombolytic therapy a,
a
a
James A. de Lemos *, Elliott M. Antman , David A. Morrow , Joan Llevadot a , Robert P. Giugliano a , Stephanie A. Coulter a , Kristin C. Schuhwerk a , Shake Arslanian b , Carolyn H. McCabe a , C. Michael Gibson c , Nader Rifai b a
Department of Medicine, Brigham and Women’ s Hospital, Boston, MA, USA Department of Laboratory Medicine, Children’ s Hospital, Boston, MA, USA c Department of Medicine, University of California at San Francisco School of Medicine, San Francisco, CA, USA b
Received 23 November 1999; received in revised form 28 February 2000; accepted 8 March 2000
Abstract Accurate, rapid, and simple noninvasive measures of infarct-related artery (IRA) patency are needed to identify patients with failed coronary reperfusion for rescue percutaneous coronary intervention (PCI). Heart-type Fatty Acid Binding Protein (H-FABP) is a small, cytosolic protein found in high concentrations in the myocardium. We evaluated the efficacy of H-FABP as a marker for successful reperfusion after thrombolysis. Fifty-eight subjects from the TIMI 14 trial had H-FABP and myoglobin concentrations measured at baseline (immediately prior to thrombolysis) and 60, 90, and 180 min after thrombolysis. All patients underwent coronary angiography at 90 min. By 60 min after thrombolysis, median concentrations of H-FABP and myoglobin were significantly higher in patients with a patent IRA than in those with an occluded IRA (P , 0.01 for each). Similarly, the 60 and 90 min / baseline H-FABP and myoglobin ratios were significantly higher among patients with a patent IRA (P , 0.01 for each). There were no significant differences in marker concentrations or ratios between patients with TIMI grade 2 and TIMI grade 3 flow. The area under the ROC curve tended to be greater for the 60 and 90 min / baseline myoglobin ratios than for similar ratios of H-FABP (0.71 and 0.73 vs. 0.64 and 0.62; P 5 ns). In conclusion,
Abbreviations: H-FABP, Heart-type Fatty Acid Binding Protein; MI, myocardial infarction; IRA, infarct related artery; TIMI, Thrombolysis in Myocardial Infarction; PCI, percutaneous coronary intervention; ROC, receiver operating characteristic *Corresponding author. Tel.: 1 1-617-278-0145; fax: 1 1-617-734-7329. E-mail address: [email protected] (J.A. de Lemos) 0009-8981 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0009-8981( 00 )00259-X
86
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
successful reperfusion can be detected within the first 60 min after thrombolysis with either H-FABP or myoglobin. Despite a favorable kinetic profile, however, H-FABP does not appear to represent a significant advance over myoglobin in the noninvasive detection of reperfusion after thrombolysis. 2000 Elsevier Science B.V. All rights reserved. Keywords: Myocardial infarction; Thrombolysis; Reperfusion; Fatty-acid binding protein; Myoglobin
1. Introduction In patients with evolving ST elevation myocardial infarction (MI), thrombolytic therapy limits infarct size and improves survival [1,2], benefits which are due to the early restoration of infarct-related artery (IRA) patency [3–5]. In selected patients with failed reperfusion, rescue percutaneous coronary intervention (PCI) appears to improve outcomes [6,7]. As a result, accurate, simple, and rapid noninvasive measures of IRA patency are needed. We recently showed that although the presence of early resolution of STsegment elevation on the 12-lead ECG is indicative of a patent IRA, absence of ST resolution does not accurately predict an occluded IRA [8]. In these patients, an early rise of cardiac serum markers may help to identify additional patients who have achieved successful reperfusion [9]. Of the serum markers studied to date, myoglobin has been shown to be superior to CK-MB and the cardiac troponins for the early detection of reperfusion [10–13]. Heart-type fatty acid binding protein (H-FABP) is a cytosolic protein that plays an important role in the uptake and oxygenation of long-chain fatty acids in the heart [14]. Like myoglobin, H-FABP is a small (15 kD) macromolecule present in both skeletal and cardiac muscle [15]. Although H-FABP and myoglobin have similar release patterns, a much higher proportion of H-FABP is present in cardiac muscle [16]. This property leads not only to increased specificity for myocardial necrosis, but also to increased sensitivity, since the plasma reference normal range for H-FABP is lower than it is for myoglobin [17,18]. We postulated that these characteristics might offer superior performance in the detection of reperfusion after thrombolytic therapy.
2. Methods
2.1. Patient population The TIMI 14 trial design, including inclusion and exclusion criteria, have been reported previously [19]. The TIMI 14 trial was a phase II angiographic
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
87
trial comparing four different thrombolytic combinations in 888 patients with evolving ST elevation myocardial infarction of , 12 h duration: abciximab alone, accelerated alteplase alone, abciximab with reduced doses of alteplase, and abciximab with reduced doses of streptokinase. Subjects with known renal insufficiency were excluded from the TIMI 14 trial. All patients received concomitant aspirin and low-dose adjunctive intravenous heparin. Fifty-eight subjects were retrospectively identified for participation in this pilot substudy; individuals were selected to provide a balanced distribution of TIMI flow grades. Patients were excluded from this substudy if they underwent rescue angioplasty prior to the 60-min blood draw.
2.2. Blood collection Blood samples were collected by trained study personnel at baseline (immediately before initiation of thrombolytic therapy) and 60, 90, and 180 min after initiation of therapy. After coagulation at room temperature, serum samples were separated, frozen at 2 208C, and shipped to the Core Chemistry Laboratory at Children’s Hospital, where they were stored at 2 708C until analysis. Thawed frozen sera were used for all analyses. All assays were performed by a single operator (S.A.) blinded to angiographic results and clinical outcomes.
2.3. H-FABP and myoglobin Assays H-FABP was measured using a solid-phase sandwich ELISA assay (HyCult Biotechnology, Uden, The Netherlands). This assay can detect H-FABP levels as low as 0.1 mg l 21 . The day-to-day imprecision concentrations, reflected by coefficient of variation, at concentrations of 6.8 mg l 21 and 100.9 mg l 21 were 14% and 4%, respectively. The upper reference limit for H-FABP was 1.6 mg l 21 . Myoglobin was measured using a quantitative sandwich enzyme immunoassay technique (Myoglobin STAT, Roche Diagnostics) on the 1010 Elecsys immunoanalyzer (Roche Diagnostics). The lower detection limit of the assay was 15 mg l 21 and the day-to-day imprecision levels, reflected by coefficient of variation, at concentrations of 83.1 mg l 21 and 675.6 mg l 21 were 2.4% and 2.7%, respectively. The upper reference limit for this assay was 92 mg l 21 in men and 58 mg l 21 in women.
2.4. Angiographic analysis Coronary angiography was performed 90 min (range 80–100 min) after initiation of study drug. Except in cases of rapid and progressive hemodynamic
88
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
deterioration, coronary interventional procedures were performed at the treating physician’s discretion after the 90-min angiogram. All coronary angiograms were analyzed in an Angiographic Core Laboratory, blinded to treatment assignment, serum marker data, and clinical endpoints. Flow in the IRA was analyzed by a single observer (C.M.G) and reported using the TIMI flow grading system [20]. Patency was defined as TIMI grade 2 or 3 flow.
2.5. Data analysis Since mechanical reperfusion would be expected to alter the time-activity curves for both H-FABP and myoglobin, data from subjects who received rescue PCI were censored at the time PCI was performed. Since the data were not normally distributed, median values are reported unless otherwise specified. Comparisons of marker concentrations between groups were made with the Mann–Whitney U test. Correlation between myoglobin and H-FABP was calculated in a linear regression model using the least mean squares method. Comparisons of Receiver Operating Characteristic (ROC) curves were made using methods described by Hanley and McNeil [21].
3. Results Subjects received thrombolytic therapy 3.762.7 h after the onset of chest discomfort. Mean serum creatinine was 0.9960.19 mg dl 21 , and no subject had a serum creatinine . 1.4 mg dl 21 . TIMI flow grade 90 min after initiation of thrombolytic therapy was measured as grade 0 / 1 (occluded) in 18 patients (31%), grade 2 in 16 patients (28%), and grade 3 in 24 patients (41%). Patency (TIMI 2 or 3 flow) was achieved in 40 patients (69%). Myoglobin and H-FABP concentrations were significantly correlated over all time points (r 5 0.92; n 5 193; P , 0.0001) (Fig. 1).
3.1. Relationship between IRA flow and concentrations of H-FABP and myoglobin Median H-FABP and myoglobin concentrations were similar at baseline in patients with a patent vs. occluded IRA (Table 1). At 60 and 90 min, median H-FABP and myoglobin concentrations were significantly higher in patients with a patent IRA than in those with an occluded IRA (P , 0.01 for each)(Table 1). Among patients with a patent IRA, median H-FABP and myoglobin concentrations peaked at 90 min and were lower by 180 min (Fig. 2). Because only two patients with an occluded IRA had an evaluable H-FABP measurement at 180 min (due to rescue PCI before 180 min in the other 16 patients), a
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
89
Fig. 1. Correlation between H-FABP and myoglobin over all timepoints.
comparison between patients with a patent vs. occluded IRA at 180 min was not possible. There were no significant differences between patients with TIMI grade 2 and TIMI grade 3 flow in the median concentrations of H-FABP or myoglobin (Table 2). Median ratios of H-FABP and myoglobin at 60 min / baseline and 90 min / Table 1 Differences in H-FABP and myoglobin between patients with a patent vs. occluded IRA Patent IRA
Occluded IRA
P value
8 [5, 31] 180 [61, 328] 254 [97, 448] 194 [66, 301] 9.7 [4.3, 61.5] 130.9 [5.9, 65.5]
12 [3, 44] 49 [23, 111] 66 [28, 123] N /Aa 2.8 [1.5, 9.6] 4.5 [1.3, 11.0]
0.86 0.006 0.002 N /Aa 0.02 0.007
97 [57, 197] 889 [328, 1850] 1102 [462, 1937] 849 [302, 1476] 5.3 [2.1, 13.8] 8.2 [3.6, 23.4]
94 [44, 212] 217 [90, 450] 266 [108, 486] N /Aa 1.9 [1.2, 3.3] 2.4 [1.1, 4.0]
0.13 0.005 0.0003 N /Aa 0.001* 0.009*
21
H-FABP (mg l ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio Myoglobin (mg l 21 ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio a
Sixteen of 18 patients with an occluded IRA underwent rescue percutaneous coronary intervention between 90 and 180 min and thus were censored prior to the 180 min timepoint.
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
90
Fig. 2. Median H-FABP (mg l 21 ) and myoglobin (mg l 21 ) concentrations in patients with a patent vs. occluded infarct related artery 90 min after thrombolysis. Measurements were made at baseline, 60, 90, and 180 min after thrombolysis. Patients were censored at the time of percutaneous coronary intervention (PCI). Because 16 of 18 patients with an occluded IRA underwent rescue PCI between 90 and 180 min, data for patients with an occluded IRA is not shown for the 180-min timepoint. *P , 0.01 vs. occluded IRA group.
Table 2 Differences in H-FABP and myoglobin between patients with TIMI grade 2 and TIMI grade 3 flow TIMI 2 Flow
TIMI 3 Flow
P value
13 [5, 53] 144 [62, 236] 237 [96, 444] 228 [87, 508] 8.0 [4.3, 20.6] 13.7 [6.0, 80.1]
8 [5, 19] 190 [56, 421] 261 [119, 497] 157 [52, 281] 13.7 [4.0, 69.0] 41.5 [5.9, 69.6]
0.59 0.61 0.84 0.32 0.59 0.94
104 [60, 429] 687 [358, 1202] 1015 [451, 1970] 660 [302, 1628] 5.7 [2.8, 8.6] 8.0 [3.8, 13.6]
79 [57, 169] 912 [322, 1976] 1102 [570, 2395] 849 [318, 1476] 4.9 [1.9, 27.1] 13.5 [3.6, 34.3]
0.55 0.69 0.72 0.97 0.69 0.70
21
H-FABP (mg l ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio Myoglobin (mg l 21 ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
91
baseline were also significantly higher in patients with a patent IRA vs. those with an occluded IRA (P , 0.02 for each) (Table 1). There were no significant differences between patients with TIMI grade 2 and TIMI grade 3 flow in the 60 and 90 min / baseline H-FABP or myoglobin ratios (Table 2).
3.2. Use of H-FABP and myoglobin ratios to predict IRA patency ROC curves for H-FABP and myoglobin ratios at 60 and 90 min / baseline are shown in Fig. 3. The area under the ROC curve tended to be greatest for the 90 min myoglobin ratio (0.73), followed by 60 min myoglobin ratio (0.71), 60 min H-FABP ratio (0.64) and 90 min H-FABP ratio (0.62). These differences were not statistically significant. Fig. 4 shows the distribution of 60 and 90 min H-FABP and myoglobin ratios: among patients with an occluded IRA, more variability was seen in the ratio of H-FABP than the ratio of myoglobin, contributing to greater overlap in H-FABP ratios between subjects with patent and occluded infarct arteries. Optimal threshold values were chosen for H-FABP and myoglobin ratios at 60 and 90 min from the histograms and ROC curves. Table 3 lists sensitivity, specificity, positive and negative predictive values for each. A myoglobin ratio
Fig. 3. Receiver-operator curves for the ratios of H-FABP and myoglobin at 60 min / baseline and 90 min / baseline. The area under the curve tended to be greatest for the 90 min / baseline myoglobin ratio (P 5 ns).
92
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
Fig. 4. Frequency distribution plots for the ratios of H-FABP (panel a) and myoglobin (panel b) at 60 and 90 min / baseline. Horizontal solid lines mark the optimal thresholds for each ratio (see Table 3). Horizontal dashed lines mark the median values for each group. *P 5 0.02; 1 P , 0.01 for the comparison of median ratios between subjects with a patent and those with an occluded IRA.
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
93
Table 3 Operating characteristics for the prediction of 90-min IRA patency a
60 min H-FABP ratio . 8 60 min myoglobin ratio . 4 90 min H-FABP ratio . 12 90 min myoglobin ratio . 5 a
Sensitivity
Specificity
PPV
NPV
55 60 62 70
78 83 81 88
85 89 88 93
44 48 48 56
P 5 ns for all comparisons.
. 5 at 90 min tended to be the best discriminator of IRA patency at 90 min, with a sensitivity of 70%, a specificity of 88%, a positive predictive value of 93% and a negative predictive value of 56%.
4. Discussion The present study demonstrates that following successful coronary reperfusion, both H-FABP and myoglobin rise sharply, peaking within 90 to 180 min of the initiation of thrombolytic therapy. In patients with failed reperfusion, on the other hand, these markers rise at a much slower rate. We detected significant differences between patients with successful and unsuccessful reperfusion by the first measured timepoint (60 min) after thrombolysis; previous studies have shown that H-FABP and myoglobin washout occurs as early as 15 min after coronary recanalization [11,22]. A number of studies have documented the efficacy of serum myoglobin for the detection of successful thrombolysis [10–13,23]. In one previous study, Ishii et al. evaluated H-FABP in patients who had received intracoronary thrombolysis or direct coronary angioplasty, and reported 100% accuracy for the detection of IRA patency when a 60 min / baseline H-FABP ratio of . 1.8 was used as the criterion for reperfusion [22]. We noted much more overlap in H-FABP concentrations between those with successful and unsuccessful reperfusion, and in particular much more variation in H-FABP ratios among patients with failed reperfusion. As a result, H-FABP was a less accurate marker of epicardial reperfusion in our study than in the study reported by Ishii et al. The discordant findings between the two studies cannot be explained by differences in infarct location or time to onset to therapy. Ishii et al. performed cardiac catheterization prior to initiation of therapy, thus eliminating patients with subtotal coronary occlusion; in addition, they excluded patients with spontaneous reperfusion and intermittent re-occlusion. Our study, on the other hand, simulated clinical practice, in which patients were enrolled entirely based on presenting clinical and ECG criteria. The strict inclusion criteria in the prior study would serve to magnify differences between the reperfused and non-
94
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
reperfused groups, whereas our broader inclusion criteria would tend to minimize differences between the two groups. The discordant results between these two studies illustrates the difficulty of applying results from carefully controlled clinical experiments to ‘‘real world’’ clinical practice.
4.1. Comparison of H-FABP and myoglobin H-FABP and myoglobin were highly correlated at all timepoints, and the similarity between the two biochemical markers was striking. Several differences between the two markers were observed, however. H-FABP exhibited a greater proportional rise over baseline values than did myoglobin, as evidenced by higher 60 and 90 min / baseline ratios. This finding is likely due to the smaller size of H-FABP and the greater cardiac specificity of this marker (better signal to noise ratio). Despite these findings, myoglobin tended to perform slightly better than H-FABP at both 60 and 90 min, as evidenced by a trend towards a greater area under the ROC curve and a more clear threshold at both timepoints. This appears to be due to greater variation in H-FABP values, particularly among patients with failed reperfusion, resulting in overlap of H-FABP ratios between the patent and occluded IRA groups. As would be expected when comparing an automated assay with a manual assay, the imprecision was less for the myoglobin assay than for the H-FABP assay. These differences were not sufficient, however, to explain the variability in H-FABP ratios observed among subjects with an occluded IRA. Coronary reperfusion is a dynamic process, with cyclic changes in coronary flow occurring early after thrombolysis, often leading to transient episodes of reocclusion [24,25]. Given the increased sensitivity of H-FABP for myocardial necrosis, it is possible that transient changes in epicardial patency could lead to demonstrable H-FABP release in the absence of detectable myoglobin. Since we performed only a single static measurement of coronary patency at 90 min, we would not have detected the occurrence of transient reperfusion followed by persistent occlusion, a circumstance in which H-FABP might rise to a greater extent than myoglobin. This could lead to ‘‘false positive’’ elevations in H-FABP among patients with an occluded IRA at 90 min. A second possibility is that the timepoints we used (60 and 90 min after thrombolysis) to measure H-FABP were not optimal for this marker. The observation that H-FABP concentrations rose markedly by 60 min in some patients with failed reperfusion suggests that earlier or more frequent measurements of H-FABP might improve the diagnostic accuracy of this marker.
4.2. Clinical Implications Although concentrations of H-FABP and myoglobin were higher at 60 and 90 min in patients with TIMI 3 vs. TIMI 2 flow, these differences were not
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
95
significant, and neither marker could accurately differentiate TIMI 2 from TIMI 3 flow. This is a potential limitation in that TIMI grade 3 flow is clearly associated with better survival than is TIMI grade 2 flow [4]. On the other hand, rescue PCI does not appear to improve outcomes in the majority of patients with TIMI 2 flow [26], probably because TIMI 2 flow is a more a manifestation of extensive tissue injury and increased microvascular resistance than it is a marker of residual stenosis and thrombosis [27]. Therefore, patency, as defined by TIMI 2 or 3 flow, remains a valid endpoint for this study. Patients with a ratio of H-FABP or myoglobin above any of the defined thresholds at 60 or 90 min had a high probability of a patent IRA, and cardiac catheterization is probably not required to determine status of the IRA in these patients. On the other hand, those with a value below the identified threshold had only an | 50% probability of an occluded IRA. Therefore, if H-FABP or myoglobin ratios alone are used clinically, a significant number of patients may undergo unnecessary cardiac catheterization. This finding is similar to what has been observed when ST resolution alone is used to define IRA patency [8], and represents a significant limitation in the use of cardiac markers alone for the detection of successful reperfusion. While both ST resolution and cardiac markers can identify patients with a patent IRA accurately, neither can accurately determine which patients have an occluded IRA. An additional practical limitation of the use of cardiac markers for the diagnosis of reperfusion is an inherent delay in providing test results to the clinician (prolonged therapeutic turnaround time), particularly when these assays are performed in central laboratories. In general, decisions regarding reperfusion need to be made within 15 min. Point of care testing may obviate this problem and facilitate more rapid clinical decision-making. Using ratios of serum markers together with ECG and clinical variables (such as relief of chest pain), may improve overall accuracy for the noninvasive detection of reperfusion. Since ECG and serum marker measurements both have high positive but low negative predictive values, a strategy of using either ST resolution or a rapid increase in serum marker concentrations as evidence for successful reperfusion may allow more patients to be accurately classified as having a patent IRA, and reduce the number of patients who are incorrectly classified as having an occluded infarct artery [9]. Studies are needed that evaluate combinations of ECG, serum marker, and clinical variables for the noninvasive detection of the success or failure of reperfusion therapy.
5. Conclusions Diagnostic performance of H-FABP was not superior to that of myoglobin for the detection of successful epicardial reperfusion 90 min after thrombolytic therapy. Thus, despite its favorable kinetic profile, H-FABP does not appear to
96
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
represent a significant advance in the noninvasive detection of reperfusion after thrombolysis. Additional studies using earlier and / or more frequent sampling after the initiation of thrombolysis might show a relative benefit for H-FABP, and further investigation in this area may be warranted.
Acknowledgements This study was supported in part by a grant from Centocor, Malvern, Pennsylvania, and Eli Lilly, Inc., Indianapolis, IN.
References [1] Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;1:397–401. [2] Simoons M, Serruys P, van den Brand M et al. Early thrombolysis in acute myocardial infarction: limitation of infarct size and improved survival. J Am Coll Cardiol 1986;7:717– 28. [3] Braunwald E. Myocardial reperfusion, limitation of infarct size, reduction of left ventricular dysfunction, and improved survival: Should the paradigm be expanded? Circulation 1989;79:441–4. [4] The GUSTO Angiographic Investigators. The comparative effects of tissue plasminogen activator, streptokinase, or both on coronary artery patency, ventricular function and survival after acute myocardial infarction. N Engl J Med 1993;329:1615–22. [5] Gibson CM, Murphy SM, Rizzo MJ et al. Relationship between TIMI frame count and clinical outcomes after thrombolytic administration. Circulation 1999;99:1945–50. [6] Ellis S, da Silva ER, Heyndrickx G et al. Randomized comparison of rescue angioplasty with conservative management of patients with early failure of thrombolysis for acute anterior myocardial infarction. Circulation 1994;90:2280–4. [7] Ross A, Lundergan C, Rohrbeck S, Boyle D. Rescue angioplasty after failed thrombolysis: technical and clinical outcomes in a large thrombolysis trial. J Am Coll Cardiol 1998;31:1511–7. [8] de Lemos JA, Antman EM, McCabe CH et al. ST-segment resolution and infarct related artery patency and flow after thrombolytic therapy. Am J Cardiol 2000;85:299–304. [9] de Lemos JA, Antman EM, Rifai N et al. ST-segment resolution and serum myoglobin ratio predict reperfusion after thrombolytic therapy: a TIMI 14 substudy. J Am Coll Cardiol 1999;33(Suppl A):391A. [10] Christenson R, Ohmann E, Topol E et al. Assessment of coronary reperfusion after thrombolysis with a model combining myoglobin, creatine kinase-MB, and clinical variables. Circulation 1997;96:1776–82. [11] 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;128:641–8.
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
97
[12] Tanasijevic MJ, Cannon CP, Wybenga DR et al. Myoglobin, creatine kinase MB, and cardiac troponin-I to assess reperfusion after thrombolysis for acute myocardial infarction: results from TIMI 10A. Am Heart J 1997;134:622–30. [13] Tanasijevic MJ, Cannon CP, Antman EM et al. Myoglobin, creatine-kinase-MB, and cardiac troponin-I 60-min ratios predict infarct-related artery patency after thrombolysis for acute myocardial infarction: results from the Thrombolysis in Myocardial Infarction study (TIMI) 10B. J Am Coll Cardiol 1999;34:739–47. [14] Schaap FG, Binas B, Danneberg H, van der Vusse GJ, Glatz JF. Impaired long-chain fatty acid utilization by cardiac myocytes isolated from mice lacking the heart-type fatty acid binding protein gene. Circ Res 1999;85:329–37. [15] Glatz JFC, van der Vusse GJ. Cellular fatty acid-binding proteins: current concepts and future directions. Mol Cell Biochem 1990;98:237–51. [16] 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;92:2848–54. [17] Ishii J, Wang J, Naruse H et al. Serum concentrations of myoglobin vs. human heart-type cytoplasmic fatty acid-binding protein in early detection of acute myocardial infarction. Clin Chem 1997;43:1372–8. [18] Glatz JFC, van der Vusse GJ, Simoons ML, Kragten JA, van Dieijen-Visser MP, Hermens WT. Fatty acid-binding protein and the early detection of acute myocardial infarction. Clinic Chimica Acta 1998;272:87–92. [19] Antman EM, Giugliano RP, Gibson CM et al. Abciximab facilitates the rate and extent of thrombolysis: Results of TIMI 14 trial. Circulation 1999;99:2720–32. [20] TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) Trial; Phase I findings. N Engl J Med 1985;312:932–6. [21] Hanley JA, McNeil BJ. A method of comparing the areas under Receiver Operating Characteristic curves derived from the same cases. Radiology 1983;148:839–43. [22] Ishii J, Nagamura Y, Nomura M et al. Early detection of successful coronary reperfusion based on serum concentration of human heart-type cytoplasmic fatty acid-binding protein. Clin Chim Acta 1997;262:13–27. [23] Abe S, Arima S, Nomoto K et al. Early detection of coronary reperfusion by rapid assessment of plasma myoglobin. Int J Cardiol 1993;38:33–40. [24] Hackett D, Davies G, Chierchia S, Maseri A. Intermittent coronary occlusion in acute myocardial infarction. Value of combined thrombolytic and vasodilator therapy. N Engl J Med 1987;317:1055–9. [25] Kwon K, Freedman B, Wilcox I et al. The unstable ST segment early after thrombolysis for acute infarction and its usefulness as a marker of recurrent coronary occlusion. Am J Cardiol 1991;67:109–15. [26] Gibson M, Schweiger M, Sequeira RF et al. Outcomes of adjunctive PTCA / stenting for TIMI grade 2 flow following thrombolysis. J Am Coll Cardiol 1998;31(Suppl. A):231A. [27] Ito H, Okamura A, Isakura K et al. Myocardial perfusion patterns related to thrombolysis in myocardial infarction perfusion grades after coronary angioplasty in patients with acute anterior wall myocardial infarction. Circulation 1996;93:1993–9.
Heart-Type Fatty Acid Binding Protein as a marker of reperfusion after thrombolytic therapy a,
a
a
James A. de Lemos *, Elliott M. Antman , David A. Morrow , Joan Llevadot a , Robert P. Giugliano a , Stephanie A. Coulter a , Kristin C. Schuhwerk a , Shake Arslanian b , Carolyn H. McCabe a , C. Michael Gibson c , Nader Rifai b a
Department of Medicine, Brigham and Women’ s Hospital, Boston, MA, USA Department of Laboratory Medicine, Children’ s Hospital, Boston, MA, USA c Department of Medicine, University of California at San Francisco School of Medicine, San Francisco, CA, USA b
Received 23 November 1999; received in revised form 28 February 2000; accepted 8 March 2000
Abstract Accurate, rapid, and simple noninvasive measures of infarct-related artery (IRA) patency are needed to identify patients with failed coronary reperfusion for rescue percutaneous coronary intervention (PCI). Heart-type Fatty Acid Binding Protein (H-FABP) is a small, cytosolic protein found in high concentrations in the myocardium. We evaluated the efficacy of H-FABP as a marker for successful reperfusion after thrombolysis. Fifty-eight subjects from the TIMI 14 trial had H-FABP and myoglobin concentrations measured at baseline (immediately prior to thrombolysis) and 60, 90, and 180 min after thrombolysis. All patients underwent coronary angiography at 90 min. By 60 min after thrombolysis, median concentrations of H-FABP and myoglobin were significantly higher in patients with a patent IRA than in those with an occluded IRA (P , 0.01 for each). Similarly, the 60 and 90 min / baseline H-FABP and myoglobin ratios were significantly higher among patients with a patent IRA (P , 0.01 for each). There were no significant differences in marker concentrations or ratios between patients with TIMI grade 2 and TIMI grade 3 flow. The area under the ROC curve tended to be greater for the 60 and 90 min / baseline myoglobin ratios than for similar ratios of H-FABP (0.71 and 0.73 vs. 0.64 and 0.62; P 5 ns). In conclusion,
Abbreviations: H-FABP, Heart-type Fatty Acid Binding Protein; MI, myocardial infarction; IRA, infarct related artery; TIMI, Thrombolysis in Myocardial Infarction; PCI, percutaneous coronary intervention; ROC, receiver operating characteristic *Corresponding author. Tel.: 1 1-617-278-0145; fax: 1 1-617-734-7329. E-mail address: [email protected] (J.A. de Lemos) 0009-8981 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0009-8981( 00 )00259-X
86
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
successful reperfusion can be detected within the first 60 min after thrombolysis with either H-FABP or myoglobin. Despite a favorable kinetic profile, however, H-FABP does not appear to represent a significant advance over myoglobin in the noninvasive detection of reperfusion after thrombolysis. 2000 Elsevier Science B.V. All rights reserved. Keywords: Myocardial infarction; Thrombolysis; Reperfusion; Fatty-acid binding protein; Myoglobin
1. Introduction In patients with evolving ST elevation myocardial infarction (MI), thrombolytic therapy limits infarct size and improves survival [1,2], benefits which are due to the early restoration of infarct-related artery (IRA) patency [3–5]. In selected patients with failed reperfusion, rescue percutaneous coronary intervention (PCI) appears to improve outcomes [6,7]. As a result, accurate, simple, and rapid noninvasive measures of IRA patency are needed. We recently showed that although the presence of early resolution of STsegment elevation on the 12-lead ECG is indicative of a patent IRA, absence of ST resolution does not accurately predict an occluded IRA [8]. In these patients, an early rise of cardiac serum markers may help to identify additional patients who have achieved successful reperfusion [9]. Of the serum markers studied to date, myoglobin has been shown to be superior to CK-MB and the cardiac troponins for the early detection of reperfusion [10–13]. Heart-type fatty acid binding protein (H-FABP) is a cytosolic protein that plays an important role in the uptake and oxygenation of long-chain fatty acids in the heart [14]. Like myoglobin, H-FABP is a small (15 kD) macromolecule present in both skeletal and cardiac muscle [15]. Although H-FABP and myoglobin have similar release patterns, a much higher proportion of H-FABP is present in cardiac muscle [16]. This property leads not only to increased specificity for myocardial necrosis, but also to increased sensitivity, since the plasma reference normal range for H-FABP is lower than it is for myoglobin [17,18]. We postulated that these characteristics might offer superior performance in the detection of reperfusion after thrombolytic therapy.
2. Methods
2.1. Patient population The TIMI 14 trial design, including inclusion and exclusion criteria, have been reported previously [19]. The TIMI 14 trial was a phase II angiographic
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
87
trial comparing four different thrombolytic combinations in 888 patients with evolving ST elevation myocardial infarction of , 12 h duration: abciximab alone, accelerated alteplase alone, abciximab with reduced doses of alteplase, and abciximab with reduced doses of streptokinase. Subjects with known renal insufficiency were excluded from the TIMI 14 trial. All patients received concomitant aspirin and low-dose adjunctive intravenous heparin. Fifty-eight subjects were retrospectively identified for participation in this pilot substudy; individuals were selected to provide a balanced distribution of TIMI flow grades. Patients were excluded from this substudy if they underwent rescue angioplasty prior to the 60-min blood draw.
2.2. Blood collection Blood samples were collected by trained study personnel at baseline (immediately before initiation of thrombolytic therapy) and 60, 90, and 180 min after initiation of therapy. After coagulation at room temperature, serum samples were separated, frozen at 2 208C, and shipped to the Core Chemistry Laboratory at Children’s Hospital, where they were stored at 2 708C until analysis. Thawed frozen sera were used for all analyses. All assays were performed by a single operator (S.A.) blinded to angiographic results and clinical outcomes.
2.3. H-FABP and myoglobin Assays H-FABP was measured using a solid-phase sandwich ELISA assay (HyCult Biotechnology, Uden, The Netherlands). This assay can detect H-FABP levels as low as 0.1 mg l 21 . The day-to-day imprecision concentrations, reflected by coefficient of variation, at concentrations of 6.8 mg l 21 and 100.9 mg l 21 were 14% and 4%, respectively. The upper reference limit for H-FABP was 1.6 mg l 21 . Myoglobin was measured using a quantitative sandwich enzyme immunoassay technique (Myoglobin STAT, Roche Diagnostics) on the 1010 Elecsys immunoanalyzer (Roche Diagnostics). The lower detection limit of the assay was 15 mg l 21 and the day-to-day imprecision levels, reflected by coefficient of variation, at concentrations of 83.1 mg l 21 and 675.6 mg l 21 were 2.4% and 2.7%, respectively. The upper reference limit for this assay was 92 mg l 21 in men and 58 mg l 21 in women.
2.4. Angiographic analysis Coronary angiography was performed 90 min (range 80–100 min) after initiation of study drug. Except in cases of rapid and progressive hemodynamic
88
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
deterioration, coronary interventional procedures were performed at the treating physician’s discretion after the 90-min angiogram. All coronary angiograms were analyzed in an Angiographic Core Laboratory, blinded to treatment assignment, serum marker data, and clinical endpoints. Flow in the IRA was analyzed by a single observer (C.M.G) and reported using the TIMI flow grading system [20]. Patency was defined as TIMI grade 2 or 3 flow.
2.5. Data analysis Since mechanical reperfusion would be expected to alter the time-activity curves for both H-FABP and myoglobin, data from subjects who received rescue PCI were censored at the time PCI was performed. Since the data were not normally distributed, median values are reported unless otherwise specified. Comparisons of marker concentrations between groups were made with the Mann–Whitney U test. Correlation between myoglobin and H-FABP was calculated in a linear regression model using the least mean squares method. Comparisons of Receiver Operating Characteristic (ROC) curves were made using methods described by Hanley and McNeil [21].
3. Results Subjects received thrombolytic therapy 3.762.7 h after the onset of chest discomfort. Mean serum creatinine was 0.9960.19 mg dl 21 , and no subject had a serum creatinine . 1.4 mg dl 21 . TIMI flow grade 90 min after initiation of thrombolytic therapy was measured as grade 0 / 1 (occluded) in 18 patients (31%), grade 2 in 16 patients (28%), and grade 3 in 24 patients (41%). Patency (TIMI 2 or 3 flow) was achieved in 40 patients (69%). Myoglobin and H-FABP concentrations were significantly correlated over all time points (r 5 0.92; n 5 193; P , 0.0001) (Fig. 1).
3.1. Relationship between IRA flow and concentrations of H-FABP and myoglobin Median H-FABP and myoglobin concentrations were similar at baseline in patients with a patent vs. occluded IRA (Table 1). At 60 and 90 min, median H-FABP and myoglobin concentrations were significantly higher in patients with a patent IRA than in those with an occluded IRA (P , 0.01 for each)(Table 1). Among patients with a patent IRA, median H-FABP and myoglobin concentrations peaked at 90 min and were lower by 180 min (Fig. 2). Because only two patients with an occluded IRA had an evaluable H-FABP measurement at 180 min (due to rescue PCI before 180 min in the other 16 patients), a
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
89
Fig. 1. Correlation between H-FABP and myoglobin over all timepoints.
comparison between patients with a patent vs. occluded IRA at 180 min was not possible. There were no significant differences between patients with TIMI grade 2 and TIMI grade 3 flow in the median concentrations of H-FABP or myoglobin (Table 2). Median ratios of H-FABP and myoglobin at 60 min / baseline and 90 min / Table 1 Differences in H-FABP and myoglobin between patients with a patent vs. occluded IRA Patent IRA
Occluded IRA
P value
8 [5, 31] 180 [61, 328] 254 [97, 448] 194 [66, 301] 9.7 [4.3, 61.5] 130.9 [5.9, 65.5]
12 [3, 44] 49 [23, 111] 66 [28, 123] N /Aa 2.8 [1.5, 9.6] 4.5 [1.3, 11.0]
0.86 0.006 0.002 N /Aa 0.02 0.007
97 [57, 197] 889 [328, 1850] 1102 [462, 1937] 849 [302, 1476] 5.3 [2.1, 13.8] 8.2 [3.6, 23.4]
94 [44, 212] 217 [90, 450] 266 [108, 486] N /Aa 1.9 [1.2, 3.3] 2.4 [1.1, 4.0]
0.13 0.005 0.0003 N /Aa 0.001* 0.009*
21
H-FABP (mg l ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio Myoglobin (mg l 21 ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio a
Sixteen of 18 patients with an occluded IRA underwent rescue percutaneous coronary intervention between 90 and 180 min and thus were censored prior to the 180 min timepoint.
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
90
Fig. 2. Median H-FABP (mg l 21 ) and myoglobin (mg l 21 ) concentrations in patients with a patent vs. occluded infarct related artery 90 min after thrombolysis. Measurements were made at baseline, 60, 90, and 180 min after thrombolysis. Patients were censored at the time of percutaneous coronary intervention (PCI). Because 16 of 18 patients with an occluded IRA underwent rescue PCI between 90 and 180 min, data for patients with an occluded IRA is not shown for the 180-min timepoint. *P , 0.01 vs. occluded IRA group.
Table 2 Differences in H-FABP and myoglobin between patients with TIMI grade 2 and TIMI grade 3 flow TIMI 2 Flow
TIMI 3 Flow
P value
13 [5, 53] 144 [62, 236] 237 [96, 444] 228 [87, 508] 8.0 [4.3, 20.6] 13.7 [6.0, 80.1]
8 [5, 19] 190 [56, 421] 261 [119, 497] 157 [52, 281] 13.7 [4.0, 69.0] 41.5 [5.9, 69.6]
0.59 0.61 0.84 0.32 0.59 0.94
104 [60, 429] 687 [358, 1202] 1015 [451, 1970] 660 [302, 1628] 5.7 [2.8, 8.6] 8.0 [3.8, 13.6]
79 [57, 169] 912 [322, 1976] 1102 [570, 2395] 849 [318, 1476] 4.9 [1.9, 27.1] 13.5 [3.6, 34.3]
0.55 0.69 0.72 0.97 0.69 0.70
21
H-FABP (mg l ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio Myoglobin (mg l 21 ) median [25,75] Baseline 60 min 90 min 180 min 60 min / baseline ratio 90 min / baseline ratio
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
91
baseline were also significantly higher in patients with a patent IRA vs. those with an occluded IRA (P , 0.02 for each) (Table 1). There were no significant differences between patients with TIMI grade 2 and TIMI grade 3 flow in the 60 and 90 min / baseline H-FABP or myoglobin ratios (Table 2).
3.2. Use of H-FABP and myoglobin ratios to predict IRA patency ROC curves for H-FABP and myoglobin ratios at 60 and 90 min / baseline are shown in Fig. 3. The area under the ROC curve tended to be greatest for the 90 min myoglobin ratio (0.73), followed by 60 min myoglobin ratio (0.71), 60 min H-FABP ratio (0.64) and 90 min H-FABP ratio (0.62). These differences were not statistically significant. Fig. 4 shows the distribution of 60 and 90 min H-FABP and myoglobin ratios: among patients with an occluded IRA, more variability was seen in the ratio of H-FABP than the ratio of myoglobin, contributing to greater overlap in H-FABP ratios between subjects with patent and occluded infarct arteries. Optimal threshold values were chosen for H-FABP and myoglobin ratios at 60 and 90 min from the histograms and ROC curves. Table 3 lists sensitivity, specificity, positive and negative predictive values for each. A myoglobin ratio
Fig. 3. Receiver-operator curves for the ratios of H-FABP and myoglobin at 60 min / baseline and 90 min / baseline. The area under the curve tended to be greatest for the 90 min / baseline myoglobin ratio (P 5 ns).
92
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
Fig. 4. Frequency distribution plots for the ratios of H-FABP (panel a) and myoglobin (panel b) at 60 and 90 min / baseline. Horizontal solid lines mark the optimal thresholds for each ratio (see Table 3). Horizontal dashed lines mark the median values for each group. *P 5 0.02; 1 P , 0.01 for the comparison of median ratios between subjects with a patent and those with an occluded IRA.
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
93
Table 3 Operating characteristics for the prediction of 90-min IRA patency a
60 min H-FABP ratio . 8 60 min myoglobin ratio . 4 90 min H-FABP ratio . 12 90 min myoglobin ratio . 5 a
Sensitivity
Specificity
PPV
NPV
55 60 62 70
78 83 81 88
85 89 88 93
44 48 48 56
P 5 ns for all comparisons.
. 5 at 90 min tended to be the best discriminator of IRA patency at 90 min, with a sensitivity of 70%, a specificity of 88%, a positive predictive value of 93% and a negative predictive value of 56%.
4. Discussion The present study demonstrates that following successful coronary reperfusion, both H-FABP and myoglobin rise sharply, peaking within 90 to 180 min of the initiation of thrombolytic therapy. In patients with failed reperfusion, on the other hand, these markers rise at a much slower rate. We detected significant differences between patients with successful and unsuccessful reperfusion by the first measured timepoint (60 min) after thrombolysis; previous studies have shown that H-FABP and myoglobin washout occurs as early as 15 min after coronary recanalization [11,22]. A number of studies have documented the efficacy of serum myoglobin for the detection of successful thrombolysis [10–13,23]. In one previous study, Ishii et al. evaluated H-FABP in patients who had received intracoronary thrombolysis or direct coronary angioplasty, and reported 100% accuracy for the detection of IRA patency when a 60 min / baseline H-FABP ratio of . 1.8 was used as the criterion for reperfusion [22]. We noted much more overlap in H-FABP concentrations between those with successful and unsuccessful reperfusion, and in particular much more variation in H-FABP ratios among patients with failed reperfusion. As a result, H-FABP was a less accurate marker of epicardial reperfusion in our study than in the study reported by Ishii et al. The discordant findings between the two studies cannot be explained by differences in infarct location or time to onset to therapy. Ishii et al. performed cardiac catheterization prior to initiation of therapy, thus eliminating patients with subtotal coronary occlusion; in addition, they excluded patients with spontaneous reperfusion and intermittent re-occlusion. Our study, on the other hand, simulated clinical practice, in which patients were enrolled entirely based on presenting clinical and ECG criteria. The strict inclusion criteria in the prior study would serve to magnify differences between the reperfused and non-
94
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
reperfused groups, whereas our broader inclusion criteria would tend to minimize differences between the two groups. The discordant results between these two studies illustrates the difficulty of applying results from carefully controlled clinical experiments to ‘‘real world’’ clinical practice.
4.1. Comparison of H-FABP and myoglobin H-FABP and myoglobin were highly correlated at all timepoints, and the similarity between the two biochemical markers was striking. Several differences between the two markers were observed, however. H-FABP exhibited a greater proportional rise over baseline values than did myoglobin, as evidenced by higher 60 and 90 min / baseline ratios. This finding is likely due to the smaller size of H-FABP and the greater cardiac specificity of this marker (better signal to noise ratio). Despite these findings, myoglobin tended to perform slightly better than H-FABP at both 60 and 90 min, as evidenced by a trend towards a greater area under the ROC curve and a more clear threshold at both timepoints. This appears to be due to greater variation in H-FABP values, particularly among patients with failed reperfusion, resulting in overlap of H-FABP ratios between the patent and occluded IRA groups. As would be expected when comparing an automated assay with a manual assay, the imprecision was less for the myoglobin assay than for the H-FABP assay. These differences were not sufficient, however, to explain the variability in H-FABP ratios observed among subjects with an occluded IRA. Coronary reperfusion is a dynamic process, with cyclic changes in coronary flow occurring early after thrombolysis, often leading to transient episodes of reocclusion [24,25]. Given the increased sensitivity of H-FABP for myocardial necrosis, it is possible that transient changes in epicardial patency could lead to demonstrable H-FABP release in the absence of detectable myoglobin. Since we performed only a single static measurement of coronary patency at 90 min, we would not have detected the occurrence of transient reperfusion followed by persistent occlusion, a circumstance in which H-FABP might rise to a greater extent than myoglobin. This could lead to ‘‘false positive’’ elevations in H-FABP among patients with an occluded IRA at 90 min. A second possibility is that the timepoints we used (60 and 90 min after thrombolysis) to measure H-FABP were not optimal for this marker. The observation that H-FABP concentrations rose markedly by 60 min in some patients with failed reperfusion suggests that earlier or more frequent measurements of H-FABP might improve the diagnostic accuracy of this marker.
4.2. Clinical Implications Although concentrations of H-FABP and myoglobin were higher at 60 and 90 min in patients with TIMI 3 vs. TIMI 2 flow, these differences were not
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
95
significant, and neither marker could accurately differentiate TIMI 2 from TIMI 3 flow. This is a potential limitation in that TIMI grade 3 flow is clearly associated with better survival than is TIMI grade 2 flow [4]. On the other hand, rescue PCI does not appear to improve outcomes in the majority of patients with TIMI 2 flow [26], probably because TIMI 2 flow is a more a manifestation of extensive tissue injury and increased microvascular resistance than it is a marker of residual stenosis and thrombosis [27]. Therefore, patency, as defined by TIMI 2 or 3 flow, remains a valid endpoint for this study. Patients with a ratio of H-FABP or myoglobin above any of the defined thresholds at 60 or 90 min had a high probability of a patent IRA, and cardiac catheterization is probably not required to determine status of the IRA in these patients. On the other hand, those with a value below the identified threshold had only an | 50% probability of an occluded IRA. Therefore, if H-FABP or myoglobin ratios alone are used clinically, a significant number of patients may undergo unnecessary cardiac catheterization. This finding is similar to what has been observed when ST resolution alone is used to define IRA patency [8], and represents a significant limitation in the use of cardiac markers alone for the detection of successful reperfusion. While both ST resolution and cardiac markers can identify patients with a patent IRA accurately, neither can accurately determine which patients have an occluded IRA. An additional practical limitation of the use of cardiac markers for the diagnosis of reperfusion is an inherent delay in providing test results to the clinician (prolonged therapeutic turnaround time), particularly when these assays are performed in central laboratories. In general, decisions regarding reperfusion need to be made within 15 min. Point of care testing may obviate this problem and facilitate more rapid clinical decision-making. Using ratios of serum markers together with ECG and clinical variables (such as relief of chest pain), may improve overall accuracy for the noninvasive detection of reperfusion. Since ECG and serum marker measurements both have high positive but low negative predictive values, a strategy of using either ST resolution or a rapid increase in serum marker concentrations as evidence for successful reperfusion may allow more patients to be accurately classified as having a patent IRA, and reduce the number of patients who are incorrectly classified as having an occluded infarct artery [9]. Studies are needed that evaluate combinations of ECG, serum marker, and clinical variables for the noninvasive detection of the success or failure of reperfusion therapy.
5. Conclusions Diagnostic performance of H-FABP was not superior to that of myoglobin for the detection of successful epicardial reperfusion 90 min after thrombolytic therapy. Thus, despite its favorable kinetic profile, H-FABP does not appear to
96
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
represent a significant advance in the noninvasive detection of reperfusion after thrombolysis. Additional studies using earlier and / or more frequent sampling after the initiation of thrombolysis might show a relative benefit for H-FABP, and further investigation in this area may be warranted.
Acknowledgements This study was supported in part by a grant from Centocor, Malvern, Pennsylvania, and Eli Lilly, Inc., Indianapolis, IN.
References [1] Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;1:397–401. [2] Simoons M, Serruys P, van den Brand M et al. Early thrombolysis in acute myocardial infarction: limitation of infarct size and improved survival. J Am Coll Cardiol 1986;7:717– 28. [3] Braunwald E. Myocardial reperfusion, limitation of infarct size, reduction of left ventricular dysfunction, and improved survival: Should the paradigm be expanded? Circulation 1989;79:441–4. [4] The GUSTO Angiographic Investigators. The comparative effects of tissue plasminogen activator, streptokinase, or both on coronary artery patency, ventricular function and survival after acute myocardial infarction. N Engl J Med 1993;329:1615–22. [5] Gibson CM, Murphy SM, Rizzo MJ et al. Relationship between TIMI frame count and clinical outcomes after thrombolytic administration. Circulation 1999;99:1945–50. [6] Ellis S, da Silva ER, Heyndrickx G et al. Randomized comparison of rescue angioplasty with conservative management of patients with early failure of thrombolysis for acute anterior myocardial infarction. Circulation 1994;90:2280–4. [7] Ross A, Lundergan C, Rohrbeck S, Boyle D. Rescue angioplasty after failed thrombolysis: technical and clinical outcomes in a large thrombolysis trial. J Am Coll Cardiol 1998;31:1511–7. [8] de Lemos JA, Antman EM, McCabe CH et al. ST-segment resolution and infarct related artery patency and flow after thrombolytic therapy. Am J Cardiol 2000;85:299–304. [9] de Lemos JA, Antman EM, Rifai N et al. ST-segment resolution and serum myoglobin ratio predict reperfusion after thrombolytic therapy: a TIMI 14 substudy. J Am Coll Cardiol 1999;33(Suppl A):391A. [10] Christenson R, Ohmann E, Topol E et al. Assessment of coronary reperfusion after thrombolysis with a model combining myoglobin, creatine kinase-MB, and clinical variables. Circulation 1997;96:1776–82. [11] 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;128:641–8.
J. A. de Lemos et al. / Clinica Chimica Acta 298 (2000) 85 – 97
97
[12] Tanasijevic MJ, Cannon CP, Wybenga DR et al. Myoglobin, creatine kinase MB, and cardiac troponin-I to assess reperfusion after thrombolysis for acute myocardial infarction: results from TIMI 10A. Am Heart J 1997;134:622–30. [13] Tanasijevic MJ, Cannon CP, Antman EM et al. Myoglobin, creatine-kinase-MB, and cardiac troponin-I 60-min ratios predict infarct-related artery patency after thrombolysis for acute myocardial infarction: results from the Thrombolysis in Myocardial Infarction study (TIMI) 10B. J Am Coll Cardiol 1999;34:739–47. [14] Schaap FG, Binas B, Danneberg H, van der Vusse GJ, Glatz JF. Impaired long-chain fatty acid utilization by cardiac myocytes isolated from mice lacking the heart-type fatty acid binding protein gene. Circ Res 1999;85:329–37. [15] Glatz JFC, van der Vusse GJ. Cellular fatty acid-binding proteins: current concepts and future directions. Mol Cell Biochem 1990;98:237–51. [16] 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;92:2848–54. [17] Ishii J, Wang J, Naruse H et al. Serum concentrations of myoglobin vs. human heart-type cytoplasmic fatty acid-binding protein in early detection of acute myocardial infarction. Clin Chem 1997;43:1372–8. [18] Glatz JFC, van der Vusse GJ, Simoons ML, Kragten JA, van Dieijen-Visser MP, Hermens WT. Fatty acid-binding protein and the early detection of acute myocardial infarction. Clinic Chimica Acta 1998;272:87–92. [19] Antman EM, Giugliano RP, Gibson CM et al. Abciximab facilitates the rate and extent of thrombolysis: Results of TIMI 14 trial. Circulation 1999;99:2720–32. [20] TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) Trial; Phase I findings. N Engl J Med 1985;312:932–6. [21] Hanley JA, McNeil BJ. A method of comparing the areas under Receiver Operating Characteristic curves derived from the same cases. Radiology 1983;148:839–43. [22] Ishii J, Nagamura Y, Nomura M et al. Early detection of successful coronary reperfusion based on serum concentration of human heart-type cytoplasmic fatty acid-binding protein. Clin Chim Acta 1997;262:13–27. [23] Abe S, Arima S, Nomoto K et al. Early detection of coronary reperfusion by rapid assessment of plasma myoglobin. Int J Cardiol 1993;38:33–40. [24] Hackett D, Davies G, Chierchia S, Maseri A. Intermittent coronary occlusion in acute myocardial infarction. Value of combined thrombolytic and vasodilator therapy. N Engl J Med 1987;317:1055–9. [25] Kwon K, Freedman B, Wilcox I et al. The unstable ST segment early after thrombolysis for acute infarction and its usefulness as a marker of recurrent coronary occlusion. Am J Cardiol 1991;67:109–15. [26] Gibson M, Schweiger M, Sequeira RF et al. Outcomes of adjunctive PTCA / stenting for TIMI grade 2 flow following thrombolysis. J Am Coll Cardiol 1998;31(Suppl. A):231A. [27] Ito H, Okamura A, Isakura K et al. Myocardial perfusion patterns related to thrombolysis in myocardial infarction perfusion grades after coronary angioplasty in patients with acute anterior wall myocardial infarction. Circulation 1996;93:1993–9.