Prognostic Value of Myocardial Salvage Achieved by Reperfusion Therapy in Patients with Acute Myocardial Infarction

Study Population

Eligible for this study were 913 patients with acute myocardial infarction who were randomized to treatment with coronary stenting, primary percutaneous coronary angioplasty, or thrombolysis, in the framework of 3 randomized trials: Stent Versus Thrombolysis for Occluded Coronary Arteries in Patients with Acute Myocardial Infarction (STOPAMI) trials 1 and 2 (7,8) and Stent or PTCA for Occluded Coronary Arteries in Patients with Acute Myocardial Infarction Ineligible for Thrombolysis (STOPAMI 3) (9). The diagnosis of acute myocardial infarction was based on the presence of chest pain lasting at least 20 min associated with typical electrocardiographic changes (ST-segment elevation of at least 0.1 mV in 2 or more extremity leads, ST-segment elevation of at least 0.2 mV in 2 or more contiguous precordial leads, or left bundle branch block of new onset) or elevation of creatine kinase activity in the serum by at least twice the upper normal limit. Exclusion criteria were onset of symptoms > 48 h before admission, stroke (within the previous 3 mo), active bleeding or bleeding diathesis, trauma or major surgery (within the last month), suspected aortic dissection, oral anticoagulation with coumarin derivatives, severe uncontrolled arterial hypertension (systolic value > 180 mm Hg), and noncompressible vascular puncture sites. The study protocols of the 3 randomized trials from which the patients came were approved by the institutional ethics committee. All patients gave written consent before recruitment in the studies.

Study Protocol and Treatment Assignment

Detailed information about the randomly assigned treatment has been published previously (7–9). Briefly, before assignment to therapy, all patients received aspirin (500 mg) and intravenous heparin (60 mg/kg of body weight, to a maximum of 5,000 U). In total, 457 patients were assigned to treatment with coronary stenting, 306 patients were assigned to treatment with percutaneous transluminal coronary angioplasty (PTCA), and 150 patients were assigned to treatment with thrombolysis. Coronary stent placement and primary percutaneous coronary angioplasty were performed as previously described (7). In the group with thrombolysis, alteplase (Actilyse; Boehringer Ingelheim) was used as a full-dose therapy in 69 patients or as a half-dose therapy plus abciximab in 81 patients. In all patients treated with stenting or PTCA, in the absence of contraindications, abciximab (ReoPro; Lilly Deutschland GmbH) was used as an adjunct therapy. Abciximab was used as a bolus of 0.25 mg/kg of body weight followed by a continuous infusion of 0.125 μg/kg/min for 12 h.

In the groups with stenting and PTCA, coronary angiography was performed before and at the end of the procedure. Digital angiograms were analyzed offline in the angiographic core laboratory with an automated edge-detection system (CMS; Medis Medical Imaging Systems).

^99mTc-Sestamibi SPECT

Before initiation of the assigned therapy, patients received an intravenous injection of 1,000 MBq (27 mCi) of ^99mTc-sestamibi. Single-photon emission CT was performed within 6–8 h after injection of the radioactive agent. Follow-up ^99mTc-sestamibi imaging was performed 7–14 d after the reperfusion therapy. Initial perfusion defect (area at risk) and final infarct size were determined according to previously described methods (7,8). All measurements were performed in the scintigraphic core laboratory by investigators unaware of the type of therapy received. Paired ^99mTc-sestamibi imaging enabled calculation of 3 parameters: initial perfusion defect and final infarct size (perfusion defects in the initial and follow-up ^99mTc-sestamibi imaging), both expressed as percentage of the left ventricle, and myocardial salvage index (initial perfusion defect minus final infarct size divided by the initial perfusion defect), which represents the proportion of the initial perfusion defect that was salvaged by reperfusion therapy or a myocardial salvage parameter that is normalized for the initial perfusion defect.

Paired ^99mTc-sestamibi imaging could not be performed on 148 of the 913 patients (74 patients in the group with stenting, 55 patients in the group with PTCA, and 19 patients in the group with thrombolysis). The remaining 765 patients (84%) with paired ^99mTc-sestamibi imaging (383 patients in the group with stenting, 251 patients in the group with PTCA, and 131 patients in the group with thrombolysis) represent the patient population of this study.

Definitions and Follow-up Protocol

The median of salvage index for the whole study population was 0.50. Patients were divided into 2 groups: the group with salvage index < 0.5 and the group with salvage index ≥ 0.5. The primary endpoint of the study was mortality at 6 mo after the index event.

After discharge, the patients were followed up clinically by phone interview at 1 mo and by an office visit at 6 mo after the acute event. Furthermore, patients were advised to contact our outpatient clinic or their referring physicians if chest pain or other cardiac symptoms arose.

Statistical Analysis

Data are presented as median (25th, 75th percentiles), counts, or proportions (percentages). Categoric data were compared by the χ² or Fisher exact test whenever an expected cell value was <5. Continuous data were compared with the Wilcoxon rank sum test. Survival was analyzed by the Kaplan–Meier method. Differences in survival were assessed with the log-rank test. We used the Cox proportional hazards model to assess the independent impact of myocardial salvage index on mortality while adjusting for other potentially confounding variables. All analyses were performed using the S-PLUS software package (Insightful Corp.). A P value < 0.05 was considered statistically significant.

Patient Characteristics

The initial perfusion defect and the salvage index (median [25th, 75th] percentiles) for the entire study population were 24.0% (13.7, 40.2) of the left ventricle and 0.50 (0.25, 0.79), respectively. No correlation between the initial perfusion defect and salvage index was found (r = 0.009; P = 0.80). Baseline characteristics of the patients are shown in Table 1. There were several differences between the groups. Patients with myocardial salvage index ≥ 0.5 had a greater proportion of women, less often had a history of myocardial infarction, and had smaller initial perfusion defects than did patients with salvage index < 0.5. In addition, PTCA and stenting were used more often in patients with myocardial salvage index ≥ 0.5 than in patients with salvage index < 0.5.

In the group of patients with myocardial salvage index ≥ 0.5, the values (median [25th, 75th percentiles]) of final infarct size, absolute salvage (both expressed as percentage of the left ventricle), and myocardial salvage index were 4.0% (0.3, 8.5), 16.5% (8.7, 27.0), and 0.78 (0.63, 1.0), respectively. In the group of patients with myocardial salvage index < 0.5, the values of final infarct size, absolute salvage, and myocardial salvage index were 22.0% (12.0, 33.5), 6.0% (1.1, 12.0), and 0.25 (0.06, 0.35), respectively. Peak values of creatine kinase activity (upper limit of normal = 80 U/L) were 814.5 U/L (352.8; 1480.8) in the group with salvage index < 0.5 and 513 U/L (293.5; 856.0) in the group with salvage index ≥ 0.5 (P < 0.001).

Clinical Outcome

During the 30 d after acute myocardial infarction, 4 deaths (1.1%) occurred in the group of patients with myocardial salvage index < 0.5 but none in the group of patients with myocardial salvage index ≥ 0.5 (P = 0.11). At 6 mo after the acute myocardial infarction, the differences in mortality between the 2 groups were further increased. Thus, at 6 mo, 19 deaths (5.1%) had occurred in the group of patients with myocardial salvage index < 0.5, compared with 4 deaths (1.0%) in the group of patients with myocardial salvage index ≥ 0.5 (odds ratio, 5.1; 95% confidence interval, 1.9–13.3; P = 0.001). Survival curves are shown in Figure 1. In the 23 patients who died during follow-up, salvage index was 0.19 (0.05, 0.37), which was significantly smaller than the salvage index of 0.50 (0.26, 0.80) in the 742 patients who survived the 6-mo follow-up (P = 0.0004; Fig. 2). In addition, if the absolute value of myocardial salvage (percentage of the left ventricle) is analyzed without normalization for initial perfusion defect, nonsurviving patients had a myocardial salvage of 6.7% (1.0, 15.8), compared with 11.0% (4.7, 20.0) for surviving patients (P = 0.08).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Survival curves of the 2 groups of patients defined by salvage index.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Box plot presentation of salvage index values in patients who died (nonsurvivors) and those who survived the 6-mo follow-up (survivors). Values are presented as median (horizontal lines) and interquartile range.

We adjusted for differences in baseline characteristics between the 2 groups of patients by using the multivariate Cox proportional hazards model. In this model, we included age, sex, previous myocardial infarction, initial perfusion defect, type of reperfusion therapy (thrombolysis or mechanical reperfusion), and salvage index as independent variables and mortality at 6 mo as a dependent variable. The c index of discrimination of the multivariate model was 0.80 (a c value of 0.5 indicates random prediction, whereas a c value of 1.0 indicates a perfectly discriminating model). The model showed that salvage index (χ² = 11.4; P = 0.0007), initial perfusion defect (χ² = 11.4; P = 0.0007), and age (χ² = 4.1; P = 0.04) were independently associated with 6-mo mortality.