Background
Whether admission myocardial wall motion score (WMS) in non–ST-segment elevation acute coronary syndromes might be a better predictor of 30-day mortality than currently recognized prognostic markers is unknown.
Methods
Admission echocardiographic and electrocardiographic data as well as coronary angiographic data were prospectively evaluated in 488 patients. Variables analyzed were clinical data, quantitative ST-segment depression, peak troponin I, WMS, ejection fraction, extent of coronary artery disease, and Thrombolysis In Myocardial Infarction (TIMI) risk score.
Results
Severity of WMS in quartiles was associated with peak troponin I (quartile 1, 5.2 μg/L; quartile 2, 9.4 μg/L; quartile 3, 11.7 μg/L; quartile 4, 23.7 μg/L; P < .001) and with the sum of all leads with ST-segment depression (quartile 1, −2.5 mm; quartile 2, −3.2 mm; quartile 3, −3.8 mm; quartile 4, −5.1 mm; P < .001). Thirty-day mortality was associated with increased worsening of WMS (quartiles 1, 0.7%; quartile 2, 3.4%; quartile 3, 3.8%; quartile 4, 11.5%; P = .001) and quantitative ST-segment depression (0 mm, 2.7%; <1.0 mm, 1.8%; 1.0–1.9 mm, 3.5%; 2.0–2.9 mm, 7.3%; ≥3.0 mm, 15.0%; P = .008). Mortality was also associated with age ( P = .002), diabetes ( P = .007), peripheral vascular disease ( P < .001), Killip class ≥ II ( P < .001), ejection fraction ( P < .001), troponin I level ( P < .001), three-vessel and/or left main coronary artery disease ( P < .001), and admission TIMI risk score ( P < .001). Nevertheless, WMS predicted 30-day mortality after adjusting for TIMI risk score (odds ratio per unit increase, 1.14; 95% confidence interval, 1.06–1.21; P < .001) or for TIMI score and Killip class > I (odds ratio per unit increase, 1.11; 95% confidence interval, 1.04–1.19; P = .004).
Conclusions
In comparison with quantitative ST-segment depression, troponin I, and TIMI risk score, WMS on admission is a better early predictor of 30-day mortality in patients with first non–ST-segment elevation acute coronary syndromes.
The admission electrocardiographic (ECG) pattern in patients with non–ST-segment elevation acute coronary syndromes (NSTEACS) has important prognostic implications for the extent to which ST-segment depression carries a worse prognosis than negative T waves. Moreover, quantitative ST-segment depression appears to be of additional prognostic value. Other recognized markers of risk are troponin levels, the ischemic threshold, the extent of coronary artery disease, and the Thrombolysis In Myocardial Infarction (TIMI) risk score. Left ventricular wall motion score (WMS) is also a prognostic marker in patients with acute myocardial infarctions, but existing series have not analyzed specifically its value in patients without ST-segment elevation, although they have included a number of these patients. Moreover, they have not excluded patients with previous myocardial infarctions, which confounds interpretation of the acute wall motion alterations, and they have not related their findings to the extent of ECG changes or peak levels of necrosis markers. As an attempt to further improve risk assessment on admission, we investigated the prognostic value of WMS as calculated using two-dimensional echocardiography in patients with first NSTEACS and compared our findings with other known risk markers, in particular with ECG changes, because they are among the strongest predictors available on admission.
Methods
Patients
The 529 consecutive patients initially included in this study were those with first NSTEACS admitted to our coronary care unit; those with other final diagnosis were excluded. Diagnosis of acute coronary syndromes was based on the presence of typical anginal pain responsive or nonresponsive to nitroglycerin and with or without admission ECG changes. Patients included those with prolonged (>20 min) pain at rest, new-onset (de novo) angina, or recent destabilization of previously stable angina (crescendo angina). However, patients with clearly atypical symptoms, those without chest pain in whom the onset of NSTEACS was unclear, those with known previous infarctions or with unknown infarctions but with frank Q waves on admission electrocardiography, and those with previous coronary artery bypass grafting (CABG) were excluded. Also excluded were those with transient ST-segment elevation, left bundle branch block, QRS duration ≥ 120 msec, or associated with other relevant valvular heart disease unrelated to NSTEACS, or hypertrophic or dilated cardiomyopathy. Finally, we also excluded patients in shock, because the performance of echocardiography is often not satisfactory in these patients. In addition, in patients with acute pulmonary edema, echocardiography was performed only when the patient was able to tolerate a recumbent position. Patients with non–ST-segment elevation myocardial infarctions were those with peak troponin I (TnI) levels > 1.0 μg/L, whereas those with unstable angina had a peak TnI levels ≤ 1.0 μg/L. The occurrence of reinfarction during hospitalization was defined by the development of prolonged anginal pain and/or acute ECG changes associated with a new increase in TnI level. In case of reinfarction after CABG, it was required that enzyme levels be ≥3 times our upper limit of normal (1.0 μg/L).
Electrocardiography
Standard 12-lead electrocardiography was performed on admission, daily thereafter, and during and after episodes of chest pain, whereas plasma levels of creatine kinase–MB (CK-MB) mass and TnI were measured every 4 to 6 hours during ≥24 hours and after anginal episodes ≥30 min in duration. ECG measurements were performed by consensus of two experts unaware of patients’ echocardiographic or angiographic findings. The admission electrocardiogram was used as a reference, and the magnitude of ST-segment depression was measured at 80 msec after the J point and expressed as maximum ST-segment depression in any one lead, the number of leads with ST-segment depression, and the sum of all leads with ST-segment depression (ΣST-segment depression). Five groups were established according to the presence and magnitude of maximum ST-segment depression on admission electrocardiography: (1) patients with no ST-segment depression, including those with negative T waves or normal or near normal ECG results (rectified ST segment or absence of repolarization changes); (2) those with ST-segment depression < 1.0 mm; (3) those with ST-segment depression of 1 to 1.9 mm; (4) those with ST-segment depression of 2.0 to 2.9 mm; and (5) those with ST-segment depression ≥ 3.0 mm.
Echocardiography
Two-dimensional echocardiography (Vivid 3 or Vivid i with harmonic imaging; GE Healthcare, Milwaukee, WI) was performed in all patients within the first 8 hours of their admission to the coronary care unit (within 4 hours in 450 patients [85%]) to assess ejection fraction and WMS of the left ventricle. Ejection fraction was calculated using the biplane Simpson’s method, and WMS was assessed in the parasternal short-axis and the apical two-chamber, four-chamber, and long-axis views, and wall thickening was considered in addition to endocardial motion for wall motion evaluation. Wall motion was distributed in 17 segments and was graded in each segment as follows: 0 = normal, 1 = hypokinesia, 2 = akinesia, and 3 = dyskinesia. The sum of all segments was expressed as WMS. The 41 patients in whom <14 segments were adequately identified were excluded; hence, 488 patients were ultimately included in the analysis. In the last 226 patients, we also collected data on wall thickness and left ventricular diameters.
Echocardiography was performed by on-call cardiologists of our cardiology department with advanced training in echocardiography. However, echocardiograms with doubtful findings were reviewed by a senior echocardiographer (A.E.). We performed an analysis of the intraobserver and interobserver variability in the assessment of myocardial regional contractility in 17 segments in 28 unselected patients (476 segments) with acute coronary syndromes and abnormal wall motion also admitted to our coronary care unit. Two observers performed the analysis of interobserver variability, and one observer also analyzed intraobserver variability. They were blinded to patients’ clinical profiles and were free to choose the images used for the analysis of regional contractility. We used κ statistics referring to global contractility, normokinesia, hypokinesia, and akinesia. The average summation score and the score index were also evaluated, and intraclass correlation coefficients for intraobserver and interobserver variability for contractility score were also assessed. The reliability of our echocardiographic measurements in a 16-segment model, which is more in keeping with the American Society of Echocardiography’s recommendations, was also assessed using the same analysis for myocardial regional contractility in 16 segments (excluding the apical segment) in the same 28 patients.
Coronary Angiography
Coronary angiography was performed within the first 4 days of pain onset, and findings were visually interpreted by consensus of two experienced observers. The culprit artery was identified according to the presence of thrombus, total occlusion, delayed anterograde flow, or eccentric type II stenosis. The extent of coronary artery disease was assessed as the number of main epicardial vessels with significant coronary stenosis (≥70%), and left main trunk stenosis was considered significant when ≥50%. The protocol complied with the Declaration of Helsinki and was approved by the hospital ethics committee, and informed consent was obtained before patients entered the study.
Statistical Analysis
The study was designed and carried out prospectively, and the variables included were conventional risk factors (active smoking, arterial hypertension requiring medication, diabetes mellitus also under medication, and total cholesterol level consistently >230 mg/dL), peripheral vascular disease, admission ECG changes, antecedent angina (exercise and rest angina, ≥3 and <30 days and >1 month), Killip class (I–IV), peak TnI, peak CK-MB mass, ejection fraction, WMS, admission TIMI risk score, and the number of vessels with significant stenosis. The central comparison was made between survivors and nonsurvivors, but an additional comparison of WMS and the rest of variables among patients with the five different ECG patterns (quantitative ST-segment depression) was also carried out. Student’s t tests were used for comparisons between two continuous variables with normal distributions and Mann-Whitney U tests for variables with skewed distributions. For comparisons of three or more groups, we used the analyses of variance for continuous variables with normal distributions and Kruskal-Wallis tests for variables with skewed distribution. Chi-square or Fisher’s exact tests were used for comparisons between categorical variables. The correlation between WMS, TnI level, and quantitative ST-segment depression was analyzed using χ 2 tests for trend and Pearson’s correlation coefficient and its relationship with mortality using the Jonckheere-Terpstra test.
The value of the contractility score in predicting 30-day mortality was assessed after adjusting for TIMI score or for TMI score and Killip class > I by means of logistic regression analysis. The analysis was performed using SPSS version 13.0 (SPSS, Inc., Chicago, IL). Data are expressed as mean ± SD or medians, and differences were considered significant at P < .05.
Results
Clinical and Angiographic Features
There were 117 patients with unstable angina and 371 with non–ST-segment elevation myocardial infarctions, and the mean durations of pain were 71 ± 89 and 176 ± 165 min, respectively. According to their ECG changes on admission, 304 patients had ST-segment depression and 184 had either negative T waves ( n = 83) or normal ECG results ( n = 101). Overall, patients with ST-segment depression were older than those without (66.1 ± 1.3 vs 60.7 ± 11.6 years, P < .001); had higher incidence of diabetes (35% vs 24.3%, P = .014), chronic obstructive pulmonary disease (21.9% vs 14.4%, P = .042), peripheral vascular disease (23.9 vs 8.8%, P < .001), and Killip class ≥ II (11.8% vs 3.9%, P = .003); and had higher peak TnI levels (14.5 ± 22.4 vs 8.9 ± 17.4 μg/L, P = .004) and CK-MB mass (63 ± 92 vs 36 ± 63 μg/L, P < .001).
Increasing severity of maximum ST-segment depression was associated with increasing age, male gender, diabetes, chronic obstructive pulmonary disease, peripheral vascular disease, and antecedent “chronic” effort angina (≥1 month), whereas the rate of “chronic” rest angina (≥1 month) was similar and low ( Table 1 ). Also, the rates of Killip class ≥ II ( P < .001), ΣST-segment depression, number of leads with ST-segment depression, and peak CK-MB mass and TnI were also progressively higher ( Table 1 ).
ST-segment depression (mm) | |||||||
---|---|---|---|---|---|---|---|
All patients | None | <1.0 | 1.0–1.9 | 2.0–2.9 | ≥3.0 | ||
Variable | ( n = 488) | ( n = 184) | ( n = 55) | ( n = 113) | ( n = 96) | ( n = 40) | P |
Age (y) | 64 ± 11.6 | 60.4 ± 11.6 | 64.0 ± 9.6 | 63.4 ± 11.7 | 68.1 ± 11.0 | 70.7 ± 9.9 | <.001 |
Women | 29.7% | 34.8% | 38.2% | 31.0% | 21.9% | 10.3% | .003 |
Hypertension | 58.2% | 57.1% | 60.0% | 58.4% | 60.4% | 55.0% | .855 |
Diabetes | 31.1% | 24.5% | 30.9% | 25.7% | 45.8% | 42.5% | .002 |
Total cholesterol > 230 mg/dL | 52.0% | 44.6% | 65.6% | 47.8% | 60.4% | 60.0% | .013 |
Active smoking | 34.2% | 33.7% | 38.2% | 32.7% | 38.5% | 25.0% | .588 |
COPD | 19.1% | 14.1% | 18.2% | 22.1% | 28.1% | 12.5% | .046 |
CVA | 6.8% | 7.1% | 5.5% | 6.2% | 6.3% | 10.0% | .929 |
PVD | 18.2% | 9.2% | 20.0% | 17.7% | 28.1% | 35.0% | <.001 |
Angina during exercise | 59.1% | 54.3% | 46.3% | 60.2% | 65.3% | 80.0% | .007 |
≥3 and <30 d | 13.6% | 15.4% | 18.5% | 11.8% | 10.6% | 10.5% | .580 |
≥1 mo | 43.0% | 37.0% | 22.2% | 47.8% | 50.5% | 65.0% | <.001 |
Angina at rest >2 d | 43.0% | 39.1% | 40.7% | 46.9% | 45.3% | 47.5% | .650 |
≥3 and <30 d | 19.8% | 17.9% | 20.8% | 17.7% | 23.2% | 25.0% | .717 |
≥1 mo | 12.5% | 10.6% | 9.3% | 17.4% | 10.0% | 17.2% | .417 |
TIMI risk score 3–7 | 22.7% | 23.4% | 65.5% | 62.8% | 79.2% | 85.0% | <.001 |
ST-segment depression | |||||||
Maximum (mm) | 1.0 ± 1.2 | 0 | 0.4 ± 0.9 | 1.1 ± 1.5 | 2.2 ± 0.8 | 3.8 ± 1.10 | <.001 |
ΣST-segment depression (mm) | 3.6 ± 4.6 | 0 | 1.5 ± 1.0 | 4.2 ± 1.8 | 7.1 ± 2.5 | 14.0 ± 4.8 | <.001 |
Number of leads | 3.3 ± 3.0 | 0 | 4.3 ± 1.7 | 5.0 ± 1.9 | 5.5 ± 1.8 | 6.9 ± 1.8 | <.001 |
Killip class | <.001 | ||||||
I | 90.8% | 96.2% | 96.4% | 92.0% | 83.3% | 72.5% | |
II | 3.7% | 1.1% | 0% | 2.7% | 6.3% | 17.5% | |
III | 5.5% | 2.7% | 3.6% | 5.3% | 10.4% | 10.0% | |
Peak TnI (μg/L) | 12.6 ± 21.1 | 9.0 ± 17.3 | 7.4 ± 8.1 | 11.1 ± 19.8 | 19.3 ± 23.7 | 24.4 ± 34.6 | <.001 |
Peak CK-MB mass (μg/L) | 53 ± 87 | 36 ± 64 | 33 ± 43 | 50 ± 74 | 79 ± 110 | 103 ± 122 | <.001 |
Ejection fraction (%) | 57.7 ± 9.7 | 59.6 ± 9.3 | 60.1 ± 9.5 | 59.7 ± 9.0 | 55.3 ± 9.7 | 51.4 ± 10.3 | <.001 |
WMS | 6.5 ± 6.0 | 5.5 ± 5.8 | 5.4 ± 5.2 | 5.9 ± 5.8 | 7.9 ± 5.8 | 11.0 ± 6.9 | <.001 |
Number of vessels with stenoses ≥ 70% | <.001 | ||||||
0 | 10.7% | 16.8% | 7.3% | 12.4% | 3.1% | 0% | |
1 | 37.9% | 47.8% | 54.5% | 38.9% | 16.7% | 17.5% | |
2 | 20.3% | 20.1% | 14.5% | 22.1% | 24.0% | 15.0% | |
3 | 18.2% | 8.7% | 16.4% | 20.4% | 31.3% | 27.5% | |
Left main coronary artery stenosis > 50% | 12.9% | 6.5% | 7.3% | 6.2% | 25.0% | 40.0% | |
Culprit coronary artery stenosis | <.001 | ||||||
Left main | 6.0% | 1.7% | 5.9% | 0% | 14 | 25.7% | |
LAD | 35.9% | 52.6% | 33.3% | 33.7% | 11.6% | 22.9% | |
CFX | 32.6% | 18.3% | 27.5% | 35.6% | 58.1% | 40.0% | |
RCA | 16.2% | 13.7% | 25.5% | 18.3% | 15.1% | 11.4% | |
<70% stenosis | 9.3% | 13.7% | 7.8% | 12.5% | 1.2% | 0% | |
Time admission to angiography (h) | 37.1 ± 27.1 | 39.5 ± 25.4 | 37.9 ± 26.3 | 39.7 ± 28.8 | 32.6 ± 29.1 | 27.9 ± 23.8 | .088 |
Medications | |||||||
β-blockers | 91.2% | 97.3% | 92.7% | 93.8% | 95.8% | 92.5% | .442 |
Nitrates | 99.2% | 100% | 98.2% | 97.3% | 100% | 100% | .092 |
ACE inhibitors | 8.8% | 8.2% | 9.7% | 9.7% | 7.3% | 12.5% | .882 |
Aspirin | 99.8% | 100% | 100% | 99.1% | 100% | 100% | .505 |
Heparin | 98.4% | 98.9% | 94.5% | 97.3% | 100% | 100% | .081 |
Clopidogrel | 50.6% | 54.7% | 58.5% | 52.7% | 41.9% | 35.9% | .064 |
Glycoprotein IIb/IIIa inhibitors | 23.2% | 19.6% | 16.4% | 28.3% | 21.9% | 37.5% | .057 |
Calcium antagonists | 8.4% | 9.8% | 10.9% | 13.4% | 4.2% | 7.5% | .239 |
Statins | 16.8% | 16.8% | 18.2% | 19.5% | 13.5% | 15.0% | .831 |
PCI | 49.2% | 54.9% | 58.2% | 52.2% | 42.7% | 37.5% | .097 |
CABG | 21.9% | 13.6% | 14.5% | 20.4% | 34.4% | 45.0% | <.001 |
Angina | 30.3% | 28.8% | 12.7% | 32.7% | 36.5% | 40.0% | .017 |
Reinfarction | 3.1% | 1.1% | 3.6% | 1.8% | 5.2% | 10.0% | .026 |
Death | 4.7% | 2.7% | 1.8% | 3.5% | 7.3% | 15.0% | .008 |
Cardiac | 3.3% | 1.6% | 1.8% | 2.7% | 4.2% | 12.5% | .006 |
Echocardiographic and Angiographic Data
The analysis of the intraobserver and interobserver variability, respectively, for the assessment of myocardial regional contractility in 17 segments yielded the following results (all with P values < .001): for global contractility, κ = 0.91 (95% confidence interval [CI], 0.87–0.96) and 0.87 (95% CI, 0.83–0.90); for normokinesia, κ = 0.89 (95% CI, 0.83–0.96) and 0.83 (95% CI, 0.78–0.88); for hypokinesia, κ = 0.68 (95% CI, 0.53–0.84) and 0.48 (95% CI, 0.35–0.61); and for akinesia, κ = 0.94 (95% CI, 0.88–1.00) and 0.91 (95% CI, 0.87–0.96). The average summation score was 24.750 ± 5.725, and the average score index was 1.456 ± 0.337. The intraclass correlation coefficients for intraobserver and interobserver variability for contractility score were 0.93 (95% CI, 0.88–0.99; P < .001) and 0.85 (95% CI, 0.77–0.92; P < .001), respectively. Likewise, the reliability of our echocardiographic measurements in a 16-segment model showed similar results for intraobserver and interobserver reliability, respectively (all with P values < .001): for global contractility, κ = 0.91 (95% CI, 0.86–0.96) and 0.86 (95% CI, 0.82–0.90); for normokinesia, κ = 0.89 (95% CI, 0.83–0.95) and 0.83 (95% CI, 0.77–0.88); for hypokinesia, κ = 0.68 (95% CI, 0.53–0.83); and for akinesia, κ = 0.94 (95% CI, 0.87–1.00). The average summation score was 22.906 ± 4.967, and the average score index was 1.432 ± 0.310, whereas the intraclass correlation coefficients for intraobserver and interobserver variability for the contractility score were 0.93 (95% CI, 0.88–0.99; P < .001) and 0.85 (95% CI, 0.77–0.92; P < .001), respectively.
There were 148 patients with normal (30%) and 341 with abnormal (70%) wall motion, 215 of whom showed one or more hypokinetic segment, 112 one or more akinetic segment, and 13 one or more dyskinetic segment. Also, 78 patients with (26%) and 70 without (38%) ST-segment depression had normal WMS. Patients with ST-segment depression had similar ejection fractions (57.2 ± 10.0% vs 58.7 ± 9.3%, P = .122) but higher WMS (7.0 ± 6.0 vs 5.5 ± 5.7, P = .006) than those without. WMS distributed in quartiles was well correlated with peak TnI ( Figure 1 ) and CK-MB levels ( P < .001). Moreover, increasing severity of maximum ST-segment depression was associated with progressively increasing WMS ( Figure 2 ) and lower ejection fraction ( Table 1 ).
