- Email: [email protected]
Short- and long-term prognostic value of the corrected QT interval in the non–ST-elevation acute coronary syndrome
Journal of Electrocardiology 40 (2007) 180 – 187 www.elsevier.com/locate/jelectrocard
Short- and long-term prognostic value of the corrected QT interval in the non–ST-elevation acute coronary syndrome Javier Jime´nez-Candil, MD, PhD,4 Ignacio Cruz Gonza´lez, MD, PhD, Jose´ M. Gonza´lez Matas, MD, Carmen Albarra´n, MD, Pedro Pabo´n, MD, PhD, Jose´ Luis Morı´n˜igo, MD, PhD, Claudio Ledesma, MD, Francisco Martı´n, MD, Maximiliano Diego, MD, Ca´ndido Martı´n-Luengo, MD, PhD Department of Cardiology, University Hospital, Salamanca, Spain Received 21 June 2006; accepted 9 October 2006
Abstract
Background and Purpose: Myocardial ischemia prolongs the QTc interval. Very little data exists about its prognostic implications in the non–ST-elevation acute coronary syndromes (NST-ACS). Methods: This is and observational and prospective study in which we evaluated the prognostic implications of the QTc obtained at admission (AQTc) in the short- and long-term of the NST-ACS. The median of the follow-up was 17 months. Results: AQTc correlated adequately with the incidence of adverse events in the short- and longterm ( P b .001), with the best cut-off point in 450 milliseconds. Patients with AQTc z 450 presented higher frequency of in-hospital death: 8.8% vs 1.2%; P = .001, and MACE (death, recurrent ischemia, or urgent coronary revascularization): 72% vs 25%; P b .001. In a Cox regression analysis, we found 3 independent predictors of cardiovascular death after discharge: AQTc z 450 (14.7% vs 2.1%; P b .0001), age N 65 years and left ventricular ejection fraction b 40%. Coronary revascularization reduced the risk of posthospitalary cardiovascular death in AQTc z 450 milliseconds (5% vs 24%; P b .0001) but had no significant effect in AQTc b 450 milliseconds. Conclusion: These findings provide a new evidence supporting the prognostic value of the AQTc in predicting unfavorable events in the short- and long-term of the NST-ACS. D 2007 Elsevier Inc. All rights reserved.
Keywords:
QTc; Electrocardiography; Unstable; Angina; Acute myocardial infarction; Prognosis
Introduction Patients with non–ST-elevation acute coronary syndromes (NST-ACS) represent an increasing large number of people admitted to the emergency departments (EDs). Across this broad group of patients, the risk of subsequent adverse outcome varies widely.1 Accordingly, efforts have
Abbreviations: NST-ACS, Non–ST-elevation Acute Coronary Syndrome; QTc, Corrected QT interval; AQTc, QTc obtained on the hospital admission electrocardiogram; MACE, Death, recurrent ischemia or urgent coronary revascularization during hospitalization; LVEF, Left ventricular ejection fraction; ACS, Acute Coronary Syndrome; TfE, Time from the onset of the symptoms to the first electrocardiogram on admission; OD, Odds ratio; CI, Confidence Interval. 4 Corresponding author. Department of Cardiology. University Hospital, Paseo de San Vicente, 58-182, 37007 Salamanca, Spain. Tel.: +34 923 291356; fax: +34 923 270008. E-mail address: [email protected] 0022-0736/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jelectrocard.2006.10.006
been focused on earlier and better identification of high-risk patients who need more aggressive medical and interventional treatment. Because of its universal availability, low cost, and simplicity, the electrocardiogram is an attractive method to risk stratify patients2,3 in which the presence and magnitude of ST-segment depression on admission is an independent predictor of adverse events in the short and long term.4 - 6 Myocardial ischemia can prolong the heart rate corrected QT interval (QTc).7 Previous studies have reported that QTc is abnormally prolonged in patients with unstable angina,8 and is associated with an increase in the incidence of adverse events in the short-term.9 However, it is unknown whether the QTc prolongation occurs early in patients with NST-ACS (unstable angina and non-Q wave myocardial infarction) and whether the QTc on the hospital admission electrocardiogram (AQTc) could have prognostic implications in the short-term.
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
On the other hand, in the Q wave myocardial infarction, QTc prolongation increases the risk of sudden death on the follow-up.10,11 Nevertheless, little prospective information is available on the value of QTc in predicting the long-term outcome in patients with NST-ACS. We conducted an observational study to prospectively assess, in the NST-ACS, the relationship between the AQTc and the hospitalary and posthospitalary outcome.
Methods Study patients and in-hospital management Inclusion criteria: Patients older than 18 years, last episode of angina less than 24 hours before the hospital admission, and definitive diagnosis of NST-ACS, verified by cardiac biochemical markers, ST-depression, or T-wave inversion 0.2 mV or greater in at least 2 contiguous leads without the presence of concomitant Q waves. Exclusion criteria: Indication for thrombolysis, atrial fibrillation, atrial flutter with irregular ventricular response, chronic treatment with Ia or III type antiarrhythmic drugs, and significant hypokalemia (defined as a potassium serum level on admission b 3 mmol/ml). The management of the patients was performed according to the ACC/AHA Guideline for the Management of patients with Unstable Angina and non–st-segment elevation acute myocardial infarction.12 QTc measurements The QTc was determined using Bazett’s formula in all standard 12-lead electrocardiograms (25 mm/s and 1 mV) performed on our patients from hospital admission to discharge. The QT was measured from the onset of the QRS complex to the end of the T wave, defined as the point of return of the T wave to the isoelectric line or to the nadir between the T and U waves in cases where a U wave was present.13 The measurements were taken blind with no knowledge of the hospitalary course of the patients. In the first 114 patients, two independent and experienced investigators simultaneously measured the QTc, and the value used was the mean. After an interobserver variability of 1.9% F 0.9%, the next measurements were performed by one investigator. In each patient, we analyzed three QTc measurements: the QTc in the electrocardiogram obtained on admission (AQTc), the maximum QTc in the first 48 hours of hospitalization, and the maximum QTc during hospitalization. We further determined the difference between the AQTc and the maximum QTc during hospitalization. Definitions Myocardial infarction: Typical chest pain, dynamic or evolutive electrocardiographic changes and elevation of biochemical markers of myocardial injury. Unstable angina: Typical chest pain, dynamic or evolutive electrocardiographic changes and demonstration of inducible ischemia or significant narrowing in an epicardial
181
coronary artery, without elevation of biochemical markers of myocardial injury. Recurrent ischemia: New episode of typical chest pain with electrocardiographic changes or elevation of biochemical markers of myocardial injury. Urgent coronary revascularization: Coronary revascularization indicated by: recurrent ischemia, clinical instability or high-risk noninvasive stratification.12,14 Major cardiac event (MACE): A combination of death, recurrent ischemia and urgent coronary revascularization during hospitalization. High-risk stress test: In the presence of any of the following: exercise-induced ST elevation (aVR excluded), ST-segment depression z 2 mm, ST depression starting at b 6 METS, ST depression involving z 5 leads, ST depression persisting z 5 minutes into recovery, inadequate blood pressure response, duration of symptom limiting exercise b 6 METS, angina at low exercise workloads, reproducible sustained or symptomatic ventricular tachycardia, Duke treadmill score b 5.14,15 Endpoints The primary endpoints were in the short-term to determine the prognostic impact of the AQTc in the incidence of MACE during hospitalization. In the longterm: To determine the relationship between the AQTc and the incidence of cardiovascular death or nonfatal ACS after hospital discharge. The secondary endpoints was to study the relationship between AQTc and the principal prognostic variables of the NST-ACS: serum markers of cardiac injury, left ventricular ejection fraction (LVEF), degree of inducible ischemia and extent of coronary artery disease. Follow-up After discharge, the patients were periodically reviewed: first, at one month, and later, from three to six months depending on their clinical situation. In each revision, a full clinical analysis was performed. On the follow-up, a total of 8 patients were lost. The vital status of the remaining patients was determined in July 2005. Statistical analysis Analysis was performed using the SPSS 11.5 for Windows (SPSS Inc, Chicago, Illinois). Normal and continuous variables were described by mean and standard deviation, whereas categorical variables were summarized by number of patients and percentage. In order to establish the QTc cut-off points with the best sensitivity and specificity for clinical events, we determined the ROC curves. Comparison of categorical variables was performed with the chisquare test (or Fisher’s exact test if n b 5). Comparison of 2 normal variables (determined by the Kolgomorov-Smirnov test) and continuous variables was done with the t-Student test. Comparison of N2 continuous variables was done using the analysis of variance (ANOVA) test. Multivariate analysis was performed by the logistic regression test, including the variables with statistical significance in the univariate analysis and those with known
182
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
clinical impact. The Cox proportional hazards model was used to assess the impact of the selected risk factors on the survival of patients. A p value b 0.05 was considered statistically significant. Results Baseline characteristics and patient outcomes From 1-1-2002 to 12-31-2004, 523 patients were consecutively admitted to our institution for NST-ACS. Patients in atrial fibrillation (n = 69), taking Ia or III type antiarrhythmic drugs (n = 16) or with significant hypokalemia upon hospital admission (n = 11), were excluded from the analysis. Thus, our study sample was composed by 427 patients; their baseline demographic and admission clinical data are shown in Table 1. During hospitalization, 25 patients (5.9%) died and 230 patients (54%) suffered a MACE (Table 2). Table 1 Patients Characteristics Variable
Value
Age (years) Male gender Hypertension Diabetes mellitus Dislipemia Smoking Peripheral vascular disease Prior acute myocardial infarction Prior percutaneous revascularization Prior by-pass surgery Prior treatment with AAS Admission data: ST-segment depression z 0,5 mm T wave inversion z 0,2 mV TnI positive4 (non Q-wave myocardial infarction) Systolic blood pressure, mm Hg Heart rate, bpm Killip class N 1 QRS duration, ms ( N 120 ms) In-Hospital data: LVEF Stress test -High risk stress test Coronariography -No significant coronary artery disease -Multivessel disease Coronary revascularization -Percutaneous revascularization -By-pass surgery Medical therapy: -Aspirin -Clopidogrel -Nitrates -h-blockers -Calcium channel blockers -Low molecular weight heparin -IIb-IIIa glycoprotein receptor inhibitors Treatment at discharge: Aspirin Clopidogrel h-blockers Statins
70 F 10 289 (68%) 260 (61%) 117 (27%) 213 (50%) 151 (35%) 45 (10%) 130 (33%) 26 (6%) 34 (8%) 205 (48%)
4 Troponin I cut-off value: z 0,1 Ag/L.
132 98 307 123 87 73 91
(31%) (23%) (72%) F 71 F 22 (17%) F 17 (12%)
55 F 12 188 (44%) 67 (16%) 241 (56%) 21 (9%) 118 (49%) 155 (36%) 133 (31%) 22 ( 5%) 412 234 376 362 104 363 178
(96%) (55%) (88%) (85%) (24%) (85%) (42%)
347 215 323 293
(86%) (53%) (80%) (73%)
Table 2 Patient outcomes In hospital Death -Cardiac failure -Ventricular fibrillation -Bradyarrhythmia -Cardiac rupture Recurrent ischemia Urgent coronary revascularization MACE Cardiac failure Sustained ventricular tachyarrhythmia After the hospital discharge
25 19 4 1 1 155 149 230 85 7
(5.9%)
Cardiovascular death -ACS -Cardiac failure -Sudden death New non-fatal ACS Cardiovascular death or new non-fatal ACS Overall (median follow-up: 17 months) Cumulative incidence of Cardiovascular Death
40 (10.1%) 26 13 1 77 (19.7%) 117 (29.8%)
(36%) (35%) (54%) (18%) (1.5%)
65 (15.5%)
After discharge, on a mean follow-up of 17 F 8 months (median, 17 months), 40 patients (10.1%) died of cardiovascular causes (only 1 was a case of cardiac sudden death) and 77 (19.7%) patients suffered a new non-fatal ACS (Table 2). QTc evolution and measurements The time from the onset of the symptoms to the first electrocardiogram on admission (TfE) was: Median: 118 minutes; interquartile range: 88 minutes. The AQTc, the maximum QTc in the first 48 hours and the maximum QTc during hospitalization presented a normal distribution; their values did not show significant differences (472 F 61, 483 F 65 and 485 F 66, respectively; P N .1 in the ANOVA Tukey test). The longest AQTc was recorded in the leads V1-V4 in 255 patients (61%), in V5-V6 in 102 (24%) and in the frontal leads in 70 (16,5%). In 289 patients (69%), the AQTc was the longest QTc; in the remaining 135 patients, the QTc increased in the next few hours. In 98 patients (23%) the longest QTc was obtained in the first 48 hours after Hospital admission. The mean of the prolongation from the AQTc to the maximum QTc was 13 F 29 milliseconds (median: 0; interquartile range: 11 milliseconds). Table 3 Mean values of the AQTc in relation to other baseline characteristics Variable
AQTc, ms4
Male gender Age N 65 Hypertension Diabetes mellitus Dislipemia Smoking Peripheral vascular disease Prior myocardial infarction Prior percutaneous revascularization Prior by-pass surgery
473 475 476 483 474 480 495 479 477 493
4 Present versus absent.
F F F F F F F F F F
60 60 62 65 61 65 67 63 63 57
vs vs vs vs vs vs vs vs vs vs
P value 472 463 466 468 471 468 469 468 472 470
F F F F F F F F F F
61 62 59 59 62 59 60 60 61 60
0,8 0,08 0,08 .023 0,6 0,06 0,007 0,1 0,6 0,04
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
183
The AQTc was longer in patients with diabetes mellitus, peripheral vascular disease and prior by-pass surgery. Table 3. The AQTc did not change significantly with respect to the TfE. The mean AQTc was similar in patients with/without TfE z 118 minutes: 476 F 61 vs 466 F 60 milliseconds (95% CI of the difference: 21.7; 1.9; P = .1). Finally, the quartiles of the AQTc presented a similar mean heart rate: 87 F 21 bpm (first quartile), 88 F 26 bpm (second quartile), 83 F 14 bpm (third quartile) and 88 F 22 bpm (fourth quartile) ( P N .1 for all comparisons in the ANOVA-Tukey test). Primary endpoints: prognostic implications of AQTc During hospitalization The AQTc was significantly longer in patients who died or suffered a MACE (Fig. 1), and showed a significant correlation with the risk of death (Area Under ROC-Curve: 0.77; P b 0001), and MACE (area under ROC Curve: 0.77; P b .001 for both). The sensitivity and specificity of an AQTc z 450 milliseconds for death and for MACE was: 92%, 40%, and 80%, 64%, respectively. Fig. 2. In a univariate analysis, an AQTc z 450 milliseconds increased the risk of death (8.8% vs 1.2%; P = .001) and
Fig. 2. ROC curves of the AQTc for in-hospital death and MACE.
MACE (72% vs 25%; P b .001). Table 4. In a multivariate analysis (logistic regression) with the following variables considered: age N65, diabetes mellitus, hypertension, prior Table 4 Prognostic implications of the AQTc z 450 milliseconds during hospitalization. Univariate analysis
Fig. 1. Box-plots of the AQTc in relation to the hospitalary prognosis.
Variable
Frequency (%)4
Statistical analysis
Death
8.8 vs 1.2
Recurrent ischemia
49 vs 16
Urgent coronary revascularization MACE
45 vs 19 72 vs 25
Cardiac failure
25 vs 6
Sustained ventricular tachyarrhythmia
2.7 vs 0
OR: 7.9 (95% P = .001 OR: 3.2 (95% P b .001 OR: 3.6 (95% P b .001 OR: 7.6 (95% P b .001 OR: 5.2 (95% P b .001 OR: 1.6 (95% P = .046
4 AQTc z 450 ms versus AQTc b 450 ms.
CI: 1.8 - 34); CI: 3.2- 8.5); CI: 2.2 - 5.7); CI: 4.8 - 11.8); CI: 2.6 - 10.4); CI: 1.58 - 1.8);
184
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
Fig. 3. Kaplan-Meier plots of the incidence of death (A) and death or nonfatal ACS, after discharge (B), according to the AQTc.
by-pass surgery, ST depression z 0.5 mm, z 2 inverted T waves, QRS z 120 milliseconds, LVEF V 40%, peripheral vascular disease, serum elevation of the cardiac injury markers, multivessel disease and AQTc z 450 milliseconds, Table 5 Independent predictors of major events after hospital discharge in a multivariate analysis (Cox regression) Cardiovascular Death: - Age N 65
Hazard Ratio: 4.3 (95% CI: 1.3- 14.2); P = .016 - LVEF V 40% Hazard Ratio: 4.8 (95% CI: 1.6- 7); P = .004 - AQTc z 450 ms Hazard Ratio: 8.3 (95% CI: 1.9- 34.9); P = .001 Cardiovascular Death or non-fatal ACS: - Age N 65 - LVEF V 40% - AQTc z 450 ms
Fig. 4. Kaplan-Meier plots of the incidence of death (A) and death or nonfatal ACS, after the discharge (B), according to the AQTc and the coronary revascularization in the hospitalization index. A: AQTc z 450 milliseconds with coronary revascularization (n = 116); B: AQTc z 450 milliseconds without coronary revascularization (n = 123); C: AQTc b 450 milliseconds without coronary revascularization (n = 37); D: AQTc b 450 milliseconds with coronary revascularization (n = 124).
two persisted as independent predictors of MACE: multivessel disease: OR: 2.1 (95% CI: 1.1 - 4.18); P = .029, and AQTc z 450 milliseconds: OR: 3 (95% CI: 1.5 5.8); P = .001. An AQTc z 450 was associated with a higher incidence of MACE in patients with ST depression N 1mm: 79% vs 51% (OR: 3.5; 95% CI: 1.6-7.5; P = .001) and in patients without ST depression: 65% vs 17% (OR: 8.8; 95% CI: 4.915.8; P b .001). Similarly, the risk of MACE increased if the AQTcz 450 in patients with Troponin I release (72% vs
Hazard Ratio: 2.06 (95% CI: 1.2- 3.5); P = .006 Hazard Ratio: 2.04 (95% CI: 1.2- 3.2); P = .003 Hazard Ratio: 3.5 (95% CI: 2.07- 6.23); P b .0001
Variables included in the analysis: age N 65, diabetes mellitus, hypertension, prior by-pass surgery, ST depression, N 2 negative T waves z 0,2 mV, QRS z 120 milliseconds, LVEF V 40%, Troponin I release, high-risk stress test, multivessel disease, treatment with statins, treatment with h-blockers and AQTc z 450 milliseconds.
Fig. 5. Distribution of the frequency of some basic prognostic variables in relation to the AQTc. (Q: Quartiles of the AQTc).
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
185
Table 6 Relationship between some prognostic variables and the AQTc Variable
Quantitative analysis4
Troponin I release LVEF V 40% High-risk stress test No significant coronary artery disease Multivessel disease
484 F 57 vs 440 F 59; P 506 F 57 vs 465 F 59; P 497 F 59 vs 432 F 52; P 437 F 48 vs 490 F 54; P 494 F 53 vs 478 F 57; P
Qualitative analysis44 b b b b =
.001 .001 .001 .001 .035
85% vs 52%; P b .001 19% vs 6%; P b .001 62% vs 14%; P b .001 4% vs 22%; P b .001 53% vs 33%; P b .001
4 AQTc in the presence versus absence of the variable. 44 Frequency of the variable in AQTc z 450 ms versus AQTc b 450 ms.
35%; OR: 4.8; CI: 2.9-8.3; P b .001) and in patients with normal serum levels of Troponin I (70% vs 15%; OR: 13; CI: 5-32; P b .001). Finally, the prognostic value of the AQTc was independent to the QRS duration: in patients with narrow QRS ( b120 milliseconds), AQTc z 450 was associated with a significant higher frequency of in-hospital death (7.4% vs 1.3%; OR: 6.2; CI: 1.4-27; P = .006) and MACE (79% vs 20%; OR: 7.9; CI: 4.9-12.7; P b .001); these results were also objectified in patients with wide QRS (z 120 milliseconds): 16% vs 0% (OR: 1.19; IC: 1.05-46; P = .045), for in-hospital death, and 64% vs 14% (OR: 11; IC: 1.2-98; P = .03), for in-hospital MACE.
disease. Finally, the incidence of these variables was higher in patients with an AQTc z 450 milliseconds. Table 6.
After discharge On the follow-up (median: 17 months), the AQTc correlated with the risk of cardiovascular death and cardiovascular death or non-fatal ACS. For cardiovascular death: Area Under ROC-Curve: 0.71 (0.63- 0.78; P b .001), and for cardiovascular death or non-fatal ACS: Area Under ROC-Curve: 0.68 (0.62- 0.73; P b .001). An AQTc z 450 presented a sensitivity and specificity of 92% and 41% (for cardiovascular death), and 82% and 49% (for cardiovascular death or non-fatal ACS), respectively. In a Kaplan-Meier analysis, the AQTc z 450 milliseconds increased the incidence of cardiovascular death: 14.7% vs 2.1% ( P b .0001), and cardiovascular death or non-fatal ACS: 39% vs 12.9% ( P b .0001). Fig. 3. AQTc z 450 milliseconds remained an independent predictor of cardiovascular death ( P = .004) and cardiovascular death or non-fatal ACS ( P b .001), in a Cox regression analysis. Table 5. In patients with an AQTc z 450 milliseconds, coronary revascularization reduced the incidence of cardiovascular death after discharge (5% vs 24%; P b .001), and cardiovascular death or non-fatal ACS (22% vs 65%; P b .001). Nevertheless, in patients with an AQTc b 450 milliseconds, coronary revascularization presented a non significant trend towards an increase in the risk of cardiovascular death (6.5% vs 1.5; P = .11) and cardiovascular death or non-fatal ACS (26% vs 13%; P = .11). Fig. 4.
The QT interval includes the total duration of ventricular activation and recovery and, in general, corresponds to the duration of the ventricular action potential. Myocardial ischemia has been shown to prolong the ventricular action potential, in this setting QTc interval is also prolonged.7,8 According to our data, in patients with NST-ACS, this occurs early on: in 69% of the patients the longest QTc was the QTc obtained on admission (AQTc), and only in 8% of the cases the longest QTc was observed after more than 48 hours of hospitalization. It is interesting to note that the AQTc was independent of the delay time in performing the first electrocardiogram. These findings may support the use of the AQTc in the early risk stratification of NST-ACS.
Secondary end-points The frequency of elevated Troponin I, moderate to severe left ventricular dysfunction (LVEF V 40%), high-risk stress test and multivessel disease increased with the duration of the AQTc. Fig. 5. Mean AQTc was longer in patients with Troponin I release, LVEF V 40%, high-risk stress test and multivessel
Discussion In this study we prospectively assessed the relationship between the QTc and the short- and long-term prognosis in patients with NST-ACS. The QTc on hospital admission electrocardiogram (AQTc) was positively associated with the incidence of death and MACE during hospitalization and increased the risk of major events after discharge. QT interval and timing of myocardial ischemia
QTc on admission and hospitalary outcome In patients with unstable angina, Gadaleta et al9 showed that a maximum QTc N 460 milliseconds in V2 to V4, observed during hospitalization, increased the risk of death, myocardial infarction or need of urgent revascularization at 30 days. In relation to this, our study presents several differences: Firstly, we included patients with unstable angina and non-Q wave myocardial infarction. Secondly, we determined the QTc in all leads, not only in precordials. The QTc may vary between leads by up to 50 milliseconds,16 and is not always longer in precordial leads. In our study sample, the longest AQTc was recorded in other leads than V2-V4 in more than one third of the patients. Finally, we evaluated the prognostic implications of the QTc obtained on the hospital admission electrocardiogram. To our knowledge no study had systematically investigated this feature. The AQTc correlated with the in-hospital incidence of death and MACE: AQTc z 450 milliseconds was associated with a significant increase in death (8.8% vs 1.2%) and MACE (72% vs 25%). In a multivariate logistic
186
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
regression model, AQTc z 450 was found to be an independent predictor for both. The power of ST depression in predicting the short-term outcome of NST-ACS has been confirmed in different reports.4-6 Concerning this, our data show that, in patients with ST deviation, an AQTc z 450 milliseconds provides additional prognostic information, increasing the risk of death and MACE. On the other hand, 34% to 54% of patients with NST-ACS do not present ST depression on admission2,5; their short-term outcome varies widely. According to our findings, the AQTc may help in the early risk stratification of these patients.
restores the QTc to normal when it is prolonged secondarily to acute ischemia.8 According to our data, AQTc z 450 milliseconds seems to select a group of patients with more myocardial injury, left ventricular dysfunction and ischemia, in whom coronary revascularization may produce better results, reducing the incidence of cardiovascular death and cardiovascular death or non-fatal ACS on the follow-up. Study limitations 1-
QTc and long-term outcome Aberrations of ventricular repolarization, which manifest as ST segment depression, T waveform abnormalities, or corrected QT prolongation constitute a risk factor for cardiovascular death,17 because they can be a marker of subjacent ischemia, left ventricular hypertrophy or left ventricular dysfunction.18,19 The QTc prolongation has been associated with ventricular tachyarrhythmias and sudden cardiac death. Some epidemiologic studies have shown that a prolonged QTc increases the risk of cardiac sudden death.20,21 This feature was also reported in patients after a Q-wave myocardial infarction.10,11 However, the QTc interval has not been used in the NST-ACS as a prognostic tool in the long-term. Our data indicate that, in NST-ACS, an AQTc z 450 milliseconds is associated with a worse course after discharge, essentially by an increase in fatal and non-fatal ACS. We can offer two possible explanations: First of all, we propose that the QTc is related to the extension of the coronary artery disease in the angiography. Second of all, the QTc may be a maker of subclinical and non significant coronary atherosclerosis: Festa et al22 studied, in 912 nondiabetic subjects, the relationship of the QTc interval with subclinical atherosclerosis, as determined by ultrasonographic measurement of carotid intima-media thickness. They found a significant correlation between the carotid intima-media thickness of the common carotid artery and the QTc (r = 0.14). In a multiple regression analysis adjusting for demographic variables, the association of common carotid artery intima-media thickness to QTc interval remained highly significant, but adjustment for cardiovascular risk factors weakened the relationship. The basis mechanism for this point could be the effects on cardiac action potential of coronary or systemic cytokines associated with endothelial dysfunction in the atherosclerosis18; in this sense, Li et al23 found that the human recombinant interleukin-1 beta increased significantly the action potential duration and the effective refractory period in excised tissues and dissociated myocytes from guinea pig ventricles (these effects are mediated, at least in part, by changes in the conductance of calcium channels). On the other hand, Magyar et al24 reported that the human endothelin-1 inhibited both the L-type calcium current and the rapid component of the delayed rectifier potassium current in human ventricular myocytes. Finally, in NST-ACS, coronary revascularization improves the long-term outcome in high-risk patients3,25 and
2-
3-
4-
Our study is an observational study and, as such, the treatment strategies were not standardized. However, this is a prospective study with careful screening and follow-up of all consecutive patients to avoid any selection or reporting bias, thus giving an accurate picture of clinical outcomes in current practice. The QT measurements are done manually and using a conventional 12 lead electrocardiogram at 25 mm/s time. This could reduce their accuracy. On the other hand, although the simplest and most common approach for correcting the QT interval is to divide its value by the square root of the preceding RR interval expressed in seconds, i.e., by using Bazett’s formula, several studies have shown that Bazett’s correction formula is not optimal in the extremes heart rates. In our series, the mean heart rate on hospital admission electrocardiogram was normal and it did not present significant differences across the patients when they were classified by the quartiles of AQTc. In this setting, the heart rate correction according to BazettTs formula works reasonably well. However, it is possible that in some patients the differences observed in the QTc from admission to the electrocardiogram in which the maximum QTc was measured (mean: 13 F 29 milliseconds) were not due to changes in the measured QT (these differences are probably too short to be objectified manually) but to little changes in the heart rate. Low serum levels of ionic calcium and magnesium prolong the QT interval. The main objective of our study was to identify a new tool in the early management of the NST-ACS. Since the AQTc was determined at the ED, and in order to avoid any selection bias, patients were included in the study there, after a full clinical data and basic serum determinations were obtained. Due to the cardiac protocol of acute chest pain in the ED of our Institution does not include the determinations of the levels of ionic calcium and magnesium, we do not consider any cut-off point of calcium and magnesium as exclusion criteria. The limited sample size and absolute number of events may reduce the statistical power of our analysis.
Conclusions In the NST-ACS, the QTc on hospital admission electrocardiogram is a simple and useful clinical tool in
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
identifying the patients at high risk of suffering adverse events in the short- and long-term term (especially in the absence of ST segment depression or Troponin release). Coronary revascularization could be particularly effective in the presence of AQTc z 450 milliseconds. New studies that involve a larger number of patients are needed to elucidate the clinical value of our findings.
References 1. Anderson HV, Cannon CP, Stone PH, et al. One-year results of the Thrombolysis in Myocardial Infarction (TIMI) IIIB clinical trial. A randomized comparison of tissue-type plasminogen activator versus placebo and early invasive versus early conservative strategies in unstable angina and non–Q wave myocardial infarction. J Am Coll Cardiol 1995;26:1643. 2. Cannon CP, McCabe CH, Stone PH, et al. The electrocardiogram predicts one-year outcome of patients with unstable angina and non–Q wave myocardial infarction: results of the TIMI III Registry ECG Ancillary Study. Thrombolysis in Myocardial Ischemia. J Am Coll Cardiol 1997;30:133. 3. Holmvang L, Clemmensen P, Lindahl B, et al. Quantitative analysis of the admission electrocardiogram identifies patients with unstable coronary artery disease who benefit the most from early invasive treatment. J Am Coll Cardiol 2003;41:905. 4. Kaul P, Newby LK, Fu Y, et al. Troponin T and quantitative STsegment depression offer complementary prognostic information in the risk stratification of acute coronary syndrome patients. J Am Coll Cardiol 2003;41:371. 5. Kaul P, Fu Y, Chang WC, et al. Prognostic value of ST segment depression in acute coronary syndromes: insights from PARAGON-A applied to GUSTO-IIb. PARAGON-A and GUSTO IIb Investigators. Platelet IIb/IIIa Antagonism for the Reduction of Acute Global Organization Network. J Am Coll Cardiol 2001;38:64. 6. Savonitto S, Ardissino D, Granger CB, et al. Prognostic value of the admission electrocardiogram in acute coronary syndromes. JAMA 1999;281:707. 7. Cinca J, Figueras J, Tenorio L, et al. Time course and rate dependence of Q-T interval changes during noncomplicated acute transmural myocardial infarction in human beings. Am J Cardiol 1981;48:1023. 8. Shawl FA, Velasco CE, Goldbaum TS, Forman MB. Effect of coronary angioplasty on electrocardiographic changes in patients with unstable angina secondary to left anterior descending coronary artery disease. J Am Coll Cardiol 1990;16:325. 9. Gadaleta FL, Llois SC, Lapuente AR, Batchvarov VN, Kaski JC. Prognostic value of corrected QT-interval prolongation in patients with unstable angina pectoris. Am J Cardiol 2003;92:203.
187
10. Schwartz PJ, Wolf S. QT interval prolongation as predictor of sudden death in patients with myocardial infarction. Circulation 1978;57:1074. 11. Yi G, Guo XH, Reardon M, et al. Circadian variation of the QT interval in patients with sudden cardiac death after myocardial infarction. Am J Cardiol 1998;81:950. 12. Braunwald E, Antman E, Beasley J, et al. ACC/AHA guideline update for the management of patients with unstable angina and non–STsegment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366. 13. Lepeschkin E, Surawicz B. The measurement of the Q-T interval of the electrocardiogram. Circulation 1952;6:378. 14. Chaitman B. Exercise stress testing. In: Braunwald E, editor. Heart disease. A textbook of cardiovascular medicine. 6th ed. Philadelphia7 WB Saunders Company; 2001. p. 141. 15. Mark DB, Shaw L, Harrel FE. Prognostic value of a treadmill score in outpatients with suspected coronary artery disease. N Engl J Med 1991;325:849. 16. Mirvis D, Goldberger A. Electrocardiography. In: Braunwald E, editor. Heart disease. A textbook of cardiovascular medicine. 6th ed. Philadelphia7 WB Saunders Company; 2001. p. 82. 17. De Bacquer D, De Backer G, Kornitzer M, Blackburn H. Prognostic value of ECG findings for total, cardiovascular disease, and coronary heart disease death in men and women. Heart 1998;80:570. 18. Lehmann MH, Morady F. QT interval: metric for cardiac prognosis? Am J Med 2003;115:732. 19. Surawicz B, Kilmans T. Chou’s electrocardiography in clinical practice: adult and pediatric. Philadelphia, Pennsylvania7 WB Saunders Company; 2001. 20. Algra A, Tijssen JG, Roelandt JR, Pool J, Lubsen J. QTc prolongation measured by standard 12-lead electrocardiography is an independent risk factor for sudden death due to cardiac arrest. Circulation 1991; 83:1888. 21. Straus SM, Kors JA, De Bruin ML, et al. Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. J Am Coll Cardiol 2006;47:362. 22. Festa A, D’Agostino Jr R, Rautaharju P, et al. Is QT interval a marker of subclinical atherosclerosis in nondiabetic subjects? The Insulin Resistance Atherosclerosis Study (IRAS). Stroke 1999;30:1566. 23. Li YH, Rozanski GJ. Effects of human recombinant interleukin-1 on electrical properties of guinea pig ventricular cells. Cardiovasc Res 1993;27:525. 24. Magyar J, Iost N, Kortvely A, et al. Effects of endothelin-1 on calcium and potassium currents in undiseased human ventricular myocytes. Pflugers Arch 2000;441:144. 25. Fox KA, Poole-Wilson P, Clayton TC, et al. 5-year outcome of an interventional strategy in non–ST-elevation acute coronary syndrome: the British Heart Foundation RITA 3 randomised trial. Lancet 2005; 366:914.
Short- and long-term prognostic value of the corrected QT interval in the non–ST-elevation acute coronary syndrome Javier Jime´nez-Candil, MD, PhD,4 Ignacio Cruz Gonza´lez, MD, PhD, Jose´ M. Gonza´lez Matas, MD, Carmen Albarra´n, MD, Pedro Pabo´n, MD, PhD, Jose´ Luis Morı´n˜igo, MD, PhD, Claudio Ledesma, MD, Francisco Martı´n, MD, Maximiliano Diego, MD, Ca´ndido Martı´n-Luengo, MD, PhD Department of Cardiology, University Hospital, Salamanca, Spain Received 21 June 2006; accepted 9 October 2006
Abstract
Background and Purpose: Myocardial ischemia prolongs the QTc interval. Very little data exists about its prognostic implications in the non–ST-elevation acute coronary syndromes (NST-ACS). Methods: This is and observational and prospective study in which we evaluated the prognostic implications of the QTc obtained at admission (AQTc) in the short- and long-term of the NST-ACS. The median of the follow-up was 17 months. Results: AQTc correlated adequately with the incidence of adverse events in the short- and longterm ( P b .001), with the best cut-off point in 450 milliseconds. Patients with AQTc z 450 presented higher frequency of in-hospital death: 8.8% vs 1.2%; P = .001, and MACE (death, recurrent ischemia, or urgent coronary revascularization): 72% vs 25%; P b .001. In a Cox regression analysis, we found 3 independent predictors of cardiovascular death after discharge: AQTc z 450 (14.7% vs 2.1%; P b .0001), age N 65 years and left ventricular ejection fraction b 40%. Coronary revascularization reduced the risk of posthospitalary cardiovascular death in AQTc z 450 milliseconds (5% vs 24%; P b .0001) but had no significant effect in AQTc b 450 milliseconds. Conclusion: These findings provide a new evidence supporting the prognostic value of the AQTc in predicting unfavorable events in the short- and long-term of the NST-ACS. D 2007 Elsevier Inc. All rights reserved.
Keywords:
QTc; Electrocardiography; Unstable; Angina; Acute myocardial infarction; Prognosis
Introduction Patients with non–ST-elevation acute coronary syndromes (NST-ACS) represent an increasing large number of people admitted to the emergency departments (EDs). Across this broad group of patients, the risk of subsequent adverse outcome varies widely.1 Accordingly, efforts have
Abbreviations: NST-ACS, Non–ST-elevation Acute Coronary Syndrome; QTc, Corrected QT interval; AQTc, QTc obtained on the hospital admission electrocardiogram; MACE, Death, recurrent ischemia or urgent coronary revascularization during hospitalization; LVEF, Left ventricular ejection fraction; ACS, Acute Coronary Syndrome; TfE, Time from the onset of the symptoms to the first electrocardiogram on admission; OD, Odds ratio; CI, Confidence Interval. 4 Corresponding author. Department of Cardiology. University Hospital, Paseo de San Vicente, 58-182, 37007 Salamanca, Spain. Tel.: +34 923 291356; fax: +34 923 270008. E-mail address: [email protected] 0022-0736/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jelectrocard.2006.10.006
been focused on earlier and better identification of high-risk patients who need more aggressive medical and interventional treatment. Because of its universal availability, low cost, and simplicity, the electrocardiogram is an attractive method to risk stratify patients2,3 in which the presence and magnitude of ST-segment depression on admission is an independent predictor of adverse events in the short and long term.4 - 6 Myocardial ischemia can prolong the heart rate corrected QT interval (QTc).7 Previous studies have reported that QTc is abnormally prolonged in patients with unstable angina,8 and is associated with an increase in the incidence of adverse events in the short-term.9 However, it is unknown whether the QTc prolongation occurs early in patients with NST-ACS (unstable angina and non-Q wave myocardial infarction) and whether the QTc on the hospital admission electrocardiogram (AQTc) could have prognostic implications in the short-term.
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
On the other hand, in the Q wave myocardial infarction, QTc prolongation increases the risk of sudden death on the follow-up.10,11 Nevertheless, little prospective information is available on the value of QTc in predicting the long-term outcome in patients with NST-ACS. We conducted an observational study to prospectively assess, in the NST-ACS, the relationship between the AQTc and the hospitalary and posthospitalary outcome.
Methods Study patients and in-hospital management Inclusion criteria: Patients older than 18 years, last episode of angina less than 24 hours before the hospital admission, and definitive diagnosis of NST-ACS, verified by cardiac biochemical markers, ST-depression, or T-wave inversion 0.2 mV or greater in at least 2 contiguous leads without the presence of concomitant Q waves. Exclusion criteria: Indication for thrombolysis, atrial fibrillation, atrial flutter with irregular ventricular response, chronic treatment with Ia or III type antiarrhythmic drugs, and significant hypokalemia (defined as a potassium serum level on admission b 3 mmol/ml). The management of the patients was performed according to the ACC/AHA Guideline for the Management of patients with Unstable Angina and non–st-segment elevation acute myocardial infarction.12 QTc measurements The QTc was determined using Bazett’s formula in all standard 12-lead electrocardiograms (25 mm/s and 1 mV) performed on our patients from hospital admission to discharge. The QT was measured from the onset of the QRS complex to the end of the T wave, defined as the point of return of the T wave to the isoelectric line or to the nadir between the T and U waves in cases where a U wave was present.13 The measurements were taken blind with no knowledge of the hospitalary course of the patients. In the first 114 patients, two independent and experienced investigators simultaneously measured the QTc, and the value used was the mean. After an interobserver variability of 1.9% F 0.9%, the next measurements were performed by one investigator. In each patient, we analyzed three QTc measurements: the QTc in the electrocardiogram obtained on admission (AQTc), the maximum QTc in the first 48 hours of hospitalization, and the maximum QTc during hospitalization. We further determined the difference between the AQTc and the maximum QTc during hospitalization. Definitions Myocardial infarction: Typical chest pain, dynamic or evolutive electrocardiographic changes and elevation of biochemical markers of myocardial injury. Unstable angina: Typical chest pain, dynamic or evolutive electrocardiographic changes and demonstration of inducible ischemia or significant narrowing in an epicardial
181
coronary artery, without elevation of biochemical markers of myocardial injury. Recurrent ischemia: New episode of typical chest pain with electrocardiographic changes or elevation of biochemical markers of myocardial injury. Urgent coronary revascularization: Coronary revascularization indicated by: recurrent ischemia, clinical instability or high-risk noninvasive stratification.12,14 Major cardiac event (MACE): A combination of death, recurrent ischemia and urgent coronary revascularization during hospitalization. High-risk stress test: In the presence of any of the following: exercise-induced ST elevation (aVR excluded), ST-segment depression z 2 mm, ST depression starting at b 6 METS, ST depression involving z 5 leads, ST depression persisting z 5 minutes into recovery, inadequate blood pressure response, duration of symptom limiting exercise b 6 METS, angina at low exercise workloads, reproducible sustained or symptomatic ventricular tachycardia, Duke treadmill score b 5.14,15 Endpoints The primary endpoints were in the short-term to determine the prognostic impact of the AQTc in the incidence of MACE during hospitalization. In the longterm: To determine the relationship between the AQTc and the incidence of cardiovascular death or nonfatal ACS after hospital discharge. The secondary endpoints was to study the relationship between AQTc and the principal prognostic variables of the NST-ACS: serum markers of cardiac injury, left ventricular ejection fraction (LVEF), degree of inducible ischemia and extent of coronary artery disease. Follow-up After discharge, the patients were periodically reviewed: first, at one month, and later, from three to six months depending on their clinical situation. In each revision, a full clinical analysis was performed. On the follow-up, a total of 8 patients were lost. The vital status of the remaining patients was determined in July 2005. Statistical analysis Analysis was performed using the SPSS 11.5 for Windows (SPSS Inc, Chicago, Illinois). Normal and continuous variables were described by mean and standard deviation, whereas categorical variables were summarized by number of patients and percentage. In order to establish the QTc cut-off points with the best sensitivity and specificity for clinical events, we determined the ROC curves. Comparison of categorical variables was performed with the chisquare test (or Fisher’s exact test if n b 5). Comparison of 2 normal variables (determined by the Kolgomorov-Smirnov test) and continuous variables was done with the t-Student test. Comparison of N2 continuous variables was done using the analysis of variance (ANOVA) test. Multivariate analysis was performed by the logistic regression test, including the variables with statistical significance in the univariate analysis and those with known
182
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
clinical impact. The Cox proportional hazards model was used to assess the impact of the selected risk factors on the survival of patients. A p value b 0.05 was considered statistically significant. Results Baseline characteristics and patient outcomes From 1-1-2002 to 12-31-2004, 523 patients were consecutively admitted to our institution for NST-ACS. Patients in atrial fibrillation (n = 69), taking Ia or III type antiarrhythmic drugs (n = 16) or with significant hypokalemia upon hospital admission (n = 11), were excluded from the analysis. Thus, our study sample was composed by 427 patients; their baseline demographic and admission clinical data are shown in Table 1. During hospitalization, 25 patients (5.9%) died and 230 patients (54%) suffered a MACE (Table 2). Table 1 Patients Characteristics Variable
Value
Age (years) Male gender Hypertension Diabetes mellitus Dislipemia Smoking Peripheral vascular disease Prior acute myocardial infarction Prior percutaneous revascularization Prior by-pass surgery Prior treatment with AAS Admission data: ST-segment depression z 0,5 mm T wave inversion z 0,2 mV TnI positive4 (non Q-wave myocardial infarction) Systolic blood pressure, mm Hg Heart rate, bpm Killip class N 1 QRS duration, ms ( N 120 ms) In-Hospital data: LVEF Stress test -High risk stress test Coronariography -No significant coronary artery disease -Multivessel disease Coronary revascularization -Percutaneous revascularization -By-pass surgery Medical therapy: -Aspirin -Clopidogrel -Nitrates -h-blockers -Calcium channel blockers -Low molecular weight heparin -IIb-IIIa glycoprotein receptor inhibitors Treatment at discharge: Aspirin Clopidogrel h-blockers Statins
70 F 10 289 (68%) 260 (61%) 117 (27%) 213 (50%) 151 (35%) 45 (10%) 130 (33%) 26 (6%) 34 (8%) 205 (48%)
4 Troponin I cut-off value: z 0,1 Ag/L.
132 98 307 123 87 73 91
(31%) (23%) (72%) F 71 F 22 (17%) F 17 (12%)
55 F 12 188 (44%) 67 (16%) 241 (56%) 21 (9%) 118 (49%) 155 (36%) 133 (31%) 22 ( 5%) 412 234 376 362 104 363 178
(96%) (55%) (88%) (85%) (24%) (85%) (42%)
347 215 323 293
(86%) (53%) (80%) (73%)
Table 2 Patient outcomes In hospital Death -Cardiac failure -Ventricular fibrillation -Bradyarrhythmia -Cardiac rupture Recurrent ischemia Urgent coronary revascularization MACE Cardiac failure Sustained ventricular tachyarrhythmia After the hospital discharge
25 19 4 1 1 155 149 230 85 7
(5.9%)
Cardiovascular death -ACS -Cardiac failure -Sudden death New non-fatal ACS Cardiovascular death or new non-fatal ACS Overall (median follow-up: 17 months) Cumulative incidence of Cardiovascular Death
40 (10.1%) 26 13 1 77 (19.7%) 117 (29.8%)
(36%) (35%) (54%) (18%) (1.5%)
65 (15.5%)
After discharge, on a mean follow-up of 17 F 8 months (median, 17 months), 40 patients (10.1%) died of cardiovascular causes (only 1 was a case of cardiac sudden death) and 77 (19.7%) patients suffered a new non-fatal ACS (Table 2). QTc evolution and measurements The time from the onset of the symptoms to the first electrocardiogram on admission (TfE) was: Median: 118 minutes; interquartile range: 88 minutes. The AQTc, the maximum QTc in the first 48 hours and the maximum QTc during hospitalization presented a normal distribution; their values did not show significant differences (472 F 61, 483 F 65 and 485 F 66, respectively; P N .1 in the ANOVA Tukey test). The longest AQTc was recorded in the leads V1-V4 in 255 patients (61%), in V5-V6 in 102 (24%) and in the frontal leads in 70 (16,5%). In 289 patients (69%), the AQTc was the longest QTc; in the remaining 135 patients, the QTc increased in the next few hours. In 98 patients (23%) the longest QTc was obtained in the first 48 hours after Hospital admission. The mean of the prolongation from the AQTc to the maximum QTc was 13 F 29 milliseconds (median: 0; interquartile range: 11 milliseconds). Table 3 Mean values of the AQTc in relation to other baseline characteristics Variable
AQTc, ms4
Male gender Age N 65 Hypertension Diabetes mellitus Dislipemia Smoking Peripheral vascular disease Prior myocardial infarction Prior percutaneous revascularization Prior by-pass surgery
473 475 476 483 474 480 495 479 477 493
4 Present versus absent.
F F F F F F F F F F
60 60 62 65 61 65 67 63 63 57
vs vs vs vs vs vs vs vs vs vs
P value 472 463 466 468 471 468 469 468 472 470
F F F F F F F F F F
61 62 59 59 62 59 60 60 61 60
0,8 0,08 0,08 .023 0,6 0,06 0,007 0,1 0,6 0,04
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
183
The AQTc was longer in patients with diabetes mellitus, peripheral vascular disease and prior by-pass surgery. Table 3. The AQTc did not change significantly with respect to the TfE. The mean AQTc was similar in patients with/without TfE z 118 minutes: 476 F 61 vs 466 F 60 milliseconds (95% CI of the difference: 21.7; 1.9; P = .1). Finally, the quartiles of the AQTc presented a similar mean heart rate: 87 F 21 bpm (first quartile), 88 F 26 bpm (second quartile), 83 F 14 bpm (third quartile) and 88 F 22 bpm (fourth quartile) ( P N .1 for all comparisons in the ANOVA-Tukey test). Primary endpoints: prognostic implications of AQTc During hospitalization The AQTc was significantly longer in patients who died or suffered a MACE (Fig. 1), and showed a significant correlation with the risk of death (Area Under ROC-Curve: 0.77; P b 0001), and MACE (area under ROC Curve: 0.77; P b .001 for both). The sensitivity and specificity of an AQTc z 450 milliseconds for death and for MACE was: 92%, 40%, and 80%, 64%, respectively. Fig. 2. In a univariate analysis, an AQTc z 450 milliseconds increased the risk of death (8.8% vs 1.2%; P = .001) and
Fig. 2. ROC curves of the AQTc for in-hospital death and MACE.
MACE (72% vs 25%; P b .001). Table 4. In a multivariate analysis (logistic regression) with the following variables considered: age N65, diabetes mellitus, hypertension, prior Table 4 Prognostic implications of the AQTc z 450 milliseconds during hospitalization. Univariate analysis
Fig. 1. Box-plots of the AQTc in relation to the hospitalary prognosis.
Variable
Frequency (%)4
Statistical analysis
Death
8.8 vs 1.2
Recurrent ischemia
49 vs 16
Urgent coronary revascularization MACE
45 vs 19 72 vs 25
Cardiac failure
25 vs 6
Sustained ventricular tachyarrhythmia
2.7 vs 0
OR: 7.9 (95% P = .001 OR: 3.2 (95% P b .001 OR: 3.6 (95% P b .001 OR: 7.6 (95% P b .001 OR: 5.2 (95% P b .001 OR: 1.6 (95% P = .046
4 AQTc z 450 ms versus AQTc b 450 ms.
CI: 1.8 - 34); CI: 3.2- 8.5); CI: 2.2 - 5.7); CI: 4.8 - 11.8); CI: 2.6 - 10.4); CI: 1.58 - 1.8);
184
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
Fig. 3. Kaplan-Meier plots of the incidence of death (A) and death or nonfatal ACS, after discharge (B), according to the AQTc.
by-pass surgery, ST depression z 0.5 mm, z 2 inverted T waves, QRS z 120 milliseconds, LVEF V 40%, peripheral vascular disease, serum elevation of the cardiac injury markers, multivessel disease and AQTc z 450 milliseconds, Table 5 Independent predictors of major events after hospital discharge in a multivariate analysis (Cox regression) Cardiovascular Death: - Age N 65
Hazard Ratio: 4.3 (95% CI: 1.3- 14.2); P = .016 - LVEF V 40% Hazard Ratio: 4.8 (95% CI: 1.6- 7); P = .004 - AQTc z 450 ms Hazard Ratio: 8.3 (95% CI: 1.9- 34.9); P = .001 Cardiovascular Death or non-fatal ACS: - Age N 65 - LVEF V 40% - AQTc z 450 ms
Fig. 4. Kaplan-Meier plots of the incidence of death (A) and death or nonfatal ACS, after the discharge (B), according to the AQTc and the coronary revascularization in the hospitalization index. A: AQTc z 450 milliseconds with coronary revascularization (n = 116); B: AQTc z 450 milliseconds without coronary revascularization (n = 123); C: AQTc b 450 milliseconds without coronary revascularization (n = 37); D: AQTc b 450 milliseconds with coronary revascularization (n = 124).
two persisted as independent predictors of MACE: multivessel disease: OR: 2.1 (95% CI: 1.1 - 4.18); P = .029, and AQTc z 450 milliseconds: OR: 3 (95% CI: 1.5 5.8); P = .001. An AQTc z 450 was associated with a higher incidence of MACE in patients with ST depression N 1mm: 79% vs 51% (OR: 3.5; 95% CI: 1.6-7.5; P = .001) and in patients without ST depression: 65% vs 17% (OR: 8.8; 95% CI: 4.915.8; P b .001). Similarly, the risk of MACE increased if the AQTcz 450 in patients with Troponin I release (72% vs
Hazard Ratio: 2.06 (95% CI: 1.2- 3.5); P = .006 Hazard Ratio: 2.04 (95% CI: 1.2- 3.2); P = .003 Hazard Ratio: 3.5 (95% CI: 2.07- 6.23); P b .0001
Variables included in the analysis: age N 65, diabetes mellitus, hypertension, prior by-pass surgery, ST depression, N 2 negative T waves z 0,2 mV, QRS z 120 milliseconds, LVEF V 40%, Troponin I release, high-risk stress test, multivessel disease, treatment with statins, treatment with h-blockers and AQTc z 450 milliseconds.
Fig. 5. Distribution of the frequency of some basic prognostic variables in relation to the AQTc. (Q: Quartiles of the AQTc).
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
185
Table 6 Relationship between some prognostic variables and the AQTc Variable
Quantitative analysis4
Troponin I release LVEF V 40% High-risk stress test No significant coronary artery disease Multivessel disease
484 F 57 vs 440 F 59; P 506 F 57 vs 465 F 59; P 497 F 59 vs 432 F 52; P 437 F 48 vs 490 F 54; P 494 F 53 vs 478 F 57; P
Qualitative analysis44 b b b b =
.001 .001 .001 .001 .035
85% vs 52%; P b .001 19% vs 6%; P b .001 62% vs 14%; P b .001 4% vs 22%; P b .001 53% vs 33%; P b .001
4 AQTc in the presence versus absence of the variable. 44 Frequency of the variable in AQTc z 450 ms versus AQTc b 450 ms.
35%; OR: 4.8; CI: 2.9-8.3; P b .001) and in patients with normal serum levels of Troponin I (70% vs 15%; OR: 13; CI: 5-32; P b .001). Finally, the prognostic value of the AQTc was independent to the QRS duration: in patients with narrow QRS ( b120 milliseconds), AQTc z 450 was associated with a significant higher frequency of in-hospital death (7.4% vs 1.3%; OR: 6.2; CI: 1.4-27; P = .006) and MACE (79% vs 20%; OR: 7.9; CI: 4.9-12.7; P b .001); these results were also objectified in patients with wide QRS (z 120 milliseconds): 16% vs 0% (OR: 1.19; IC: 1.05-46; P = .045), for in-hospital death, and 64% vs 14% (OR: 11; IC: 1.2-98; P = .03), for in-hospital MACE.
disease. Finally, the incidence of these variables was higher in patients with an AQTc z 450 milliseconds. Table 6.
After discharge On the follow-up (median: 17 months), the AQTc correlated with the risk of cardiovascular death and cardiovascular death or non-fatal ACS. For cardiovascular death: Area Under ROC-Curve: 0.71 (0.63- 0.78; P b .001), and for cardiovascular death or non-fatal ACS: Area Under ROC-Curve: 0.68 (0.62- 0.73; P b .001). An AQTc z 450 presented a sensitivity and specificity of 92% and 41% (for cardiovascular death), and 82% and 49% (for cardiovascular death or non-fatal ACS), respectively. In a Kaplan-Meier analysis, the AQTc z 450 milliseconds increased the incidence of cardiovascular death: 14.7% vs 2.1% ( P b .0001), and cardiovascular death or non-fatal ACS: 39% vs 12.9% ( P b .0001). Fig. 3. AQTc z 450 milliseconds remained an independent predictor of cardiovascular death ( P = .004) and cardiovascular death or non-fatal ACS ( P b .001), in a Cox regression analysis. Table 5. In patients with an AQTc z 450 milliseconds, coronary revascularization reduced the incidence of cardiovascular death after discharge (5% vs 24%; P b .001), and cardiovascular death or non-fatal ACS (22% vs 65%; P b .001). Nevertheless, in patients with an AQTc b 450 milliseconds, coronary revascularization presented a non significant trend towards an increase in the risk of cardiovascular death (6.5% vs 1.5; P = .11) and cardiovascular death or non-fatal ACS (26% vs 13%; P = .11). Fig. 4.
The QT interval includes the total duration of ventricular activation and recovery and, in general, corresponds to the duration of the ventricular action potential. Myocardial ischemia has been shown to prolong the ventricular action potential, in this setting QTc interval is also prolonged.7,8 According to our data, in patients with NST-ACS, this occurs early on: in 69% of the patients the longest QTc was the QTc obtained on admission (AQTc), and only in 8% of the cases the longest QTc was observed after more than 48 hours of hospitalization. It is interesting to note that the AQTc was independent of the delay time in performing the first electrocardiogram. These findings may support the use of the AQTc in the early risk stratification of NST-ACS.
Secondary end-points The frequency of elevated Troponin I, moderate to severe left ventricular dysfunction (LVEF V 40%), high-risk stress test and multivessel disease increased with the duration of the AQTc. Fig. 5. Mean AQTc was longer in patients with Troponin I release, LVEF V 40%, high-risk stress test and multivessel
Discussion In this study we prospectively assessed the relationship between the QTc and the short- and long-term prognosis in patients with NST-ACS. The QTc on hospital admission electrocardiogram (AQTc) was positively associated with the incidence of death and MACE during hospitalization and increased the risk of major events after discharge. QT interval and timing of myocardial ischemia
QTc on admission and hospitalary outcome In patients with unstable angina, Gadaleta et al9 showed that a maximum QTc N 460 milliseconds in V2 to V4, observed during hospitalization, increased the risk of death, myocardial infarction or need of urgent revascularization at 30 days. In relation to this, our study presents several differences: Firstly, we included patients with unstable angina and non-Q wave myocardial infarction. Secondly, we determined the QTc in all leads, not only in precordials. The QTc may vary between leads by up to 50 milliseconds,16 and is not always longer in precordial leads. In our study sample, the longest AQTc was recorded in other leads than V2-V4 in more than one third of the patients. Finally, we evaluated the prognostic implications of the QTc obtained on the hospital admission electrocardiogram. To our knowledge no study had systematically investigated this feature. The AQTc correlated with the in-hospital incidence of death and MACE: AQTc z 450 milliseconds was associated with a significant increase in death (8.8% vs 1.2%) and MACE (72% vs 25%). In a multivariate logistic
186
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
regression model, AQTc z 450 was found to be an independent predictor for both. The power of ST depression in predicting the short-term outcome of NST-ACS has been confirmed in different reports.4-6 Concerning this, our data show that, in patients with ST deviation, an AQTc z 450 milliseconds provides additional prognostic information, increasing the risk of death and MACE. On the other hand, 34% to 54% of patients with NST-ACS do not present ST depression on admission2,5; their short-term outcome varies widely. According to our findings, the AQTc may help in the early risk stratification of these patients.
restores the QTc to normal when it is prolonged secondarily to acute ischemia.8 According to our data, AQTc z 450 milliseconds seems to select a group of patients with more myocardial injury, left ventricular dysfunction and ischemia, in whom coronary revascularization may produce better results, reducing the incidence of cardiovascular death and cardiovascular death or non-fatal ACS on the follow-up. Study limitations 1-
QTc and long-term outcome Aberrations of ventricular repolarization, which manifest as ST segment depression, T waveform abnormalities, or corrected QT prolongation constitute a risk factor for cardiovascular death,17 because they can be a marker of subjacent ischemia, left ventricular hypertrophy or left ventricular dysfunction.18,19 The QTc prolongation has been associated with ventricular tachyarrhythmias and sudden cardiac death. Some epidemiologic studies have shown that a prolonged QTc increases the risk of cardiac sudden death.20,21 This feature was also reported in patients after a Q-wave myocardial infarction.10,11 However, the QTc interval has not been used in the NST-ACS as a prognostic tool in the long-term. Our data indicate that, in NST-ACS, an AQTc z 450 milliseconds is associated with a worse course after discharge, essentially by an increase in fatal and non-fatal ACS. We can offer two possible explanations: First of all, we propose that the QTc is related to the extension of the coronary artery disease in the angiography. Second of all, the QTc may be a maker of subclinical and non significant coronary atherosclerosis: Festa et al22 studied, in 912 nondiabetic subjects, the relationship of the QTc interval with subclinical atherosclerosis, as determined by ultrasonographic measurement of carotid intima-media thickness. They found a significant correlation between the carotid intima-media thickness of the common carotid artery and the QTc (r = 0.14). In a multiple regression analysis adjusting for demographic variables, the association of common carotid artery intima-media thickness to QTc interval remained highly significant, but adjustment for cardiovascular risk factors weakened the relationship. The basis mechanism for this point could be the effects on cardiac action potential of coronary or systemic cytokines associated with endothelial dysfunction in the atherosclerosis18; in this sense, Li et al23 found that the human recombinant interleukin-1 beta increased significantly the action potential duration and the effective refractory period in excised tissues and dissociated myocytes from guinea pig ventricles (these effects are mediated, at least in part, by changes in the conductance of calcium channels). On the other hand, Magyar et al24 reported that the human endothelin-1 inhibited both the L-type calcium current and the rapid component of the delayed rectifier potassium current in human ventricular myocytes. Finally, in NST-ACS, coronary revascularization improves the long-term outcome in high-risk patients3,25 and
2-
3-
4-
Our study is an observational study and, as such, the treatment strategies were not standardized. However, this is a prospective study with careful screening and follow-up of all consecutive patients to avoid any selection or reporting bias, thus giving an accurate picture of clinical outcomes in current practice. The QT measurements are done manually and using a conventional 12 lead electrocardiogram at 25 mm/s time. This could reduce their accuracy. On the other hand, although the simplest and most common approach for correcting the QT interval is to divide its value by the square root of the preceding RR interval expressed in seconds, i.e., by using Bazett’s formula, several studies have shown that Bazett’s correction formula is not optimal in the extremes heart rates. In our series, the mean heart rate on hospital admission electrocardiogram was normal and it did not present significant differences across the patients when they were classified by the quartiles of AQTc. In this setting, the heart rate correction according to BazettTs formula works reasonably well. However, it is possible that in some patients the differences observed in the QTc from admission to the electrocardiogram in which the maximum QTc was measured (mean: 13 F 29 milliseconds) were not due to changes in the measured QT (these differences are probably too short to be objectified manually) but to little changes in the heart rate. Low serum levels of ionic calcium and magnesium prolong the QT interval. The main objective of our study was to identify a new tool in the early management of the NST-ACS. Since the AQTc was determined at the ED, and in order to avoid any selection bias, patients were included in the study there, after a full clinical data and basic serum determinations were obtained. Due to the cardiac protocol of acute chest pain in the ED of our Institution does not include the determinations of the levels of ionic calcium and magnesium, we do not consider any cut-off point of calcium and magnesium as exclusion criteria. The limited sample size and absolute number of events may reduce the statistical power of our analysis.
Conclusions In the NST-ACS, the QTc on hospital admission electrocardiogram is a simple and useful clinical tool in
J. Jime´nez-Candil et al. / Journal of Electrocardiology 40 (2007) 180–187
identifying the patients at high risk of suffering adverse events in the short- and long-term term (especially in the absence of ST segment depression or Troponin release). Coronary revascularization could be particularly effective in the presence of AQTc z 450 milliseconds. New studies that involve a larger number of patients are needed to elucidate the clinical value of our findings.
References 1. Anderson HV, Cannon CP, Stone PH, et al. One-year results of the Thrombolysis in Myocardial Infarction (TIMI) IIIB clinical trial. A randomized comparison of tissue-type plasminogen activator versus placebo and early invasive versus early conservative strategies in unstable angina and non–Q wave myocardial infarction. J Am Coll Cardiol 1995;26:1643. 2. Cannon CP, McCabe CH, Stone PH, et al. The electrocardiogram predicts one-year outcome of patients with unstable angina and non–Q wave myocardial infarction: results of the TIMI III Registry ECG Ancillary Study. Thrombolysis in Myocardial Ischemia. J Am Coll Cardiol 1997;30:133. 3. Holmvang L, Clemmensen P, Lindahl B, et al. Quantitative analysis of the admission electrocardiogram identifies patients with unstable coronary artery disease who benefit the most from early invasive treatment. J Am Coll Cardiol 2003;41:905. 4. Kaul P, Newby LK, Fu Y, et al. Troponin T and quantitative STsegment depression offer complementary prognostic information in the risk stratification of acute coronary syndrome patients. J Am Coll Cardiol 2003;41:371. 5. Kaul P, Fu Y, Chang WC, et al. Prognostic value of ST segment depression in acute coronary syndromes: insights from PARAGON-A applied to GUSTO-IIb. PARAGON-A and GUSTO IIb Investigators. Platelet IIb/IIIa Antagonism for the Reduction of Acute Global Organization Network. J Am Coll Cardiol 2001;38:64. 6. Savonitto S, Ardissino D, Granger CB, et al. Prognostic value of the admission electrocardiogram in acute coronary syndromes. JAMA 1999;281:707. 7. Cinca J, Figueras J, Tenorio L, et al. Time course and rate dependence of Q-T interval changes during noncomplicated acute transmural myocardial infarction in human beings. Am J Cardiol 1981;48:1023. 8. Shawl FA, Velasco CE, Goldbaum TS, Forman MB. Effect of coronary angioplasty on electrocardiographic changes in patients with unstable angina secondary to left anterior descending coronary artery disease. J Am Coll Cardiol 1990;16:325. 9. Gadaleta FL, Llois SC, Lapuente AR, Batchvarov VN, Kaski JC. Prognostic value of corrected QT-interval prolongation in patients with unstable angina pectoris. Am J Cardiol 2003;92:203.
187
10. Schwartz PJ, Wolf S. QT interval prolongation as predictor of sudden death in patients with myocardial infarction. Circulation 1978;57:1074. 11. Yi G, Guo XH, Reardon M, et al. Circadian variation of the QT interval in patients with sudden cardiac death after myocardial infarction. Am J Cardiol 1998;81:950. 12. Braunwald E, Antman E, Beasley J, et al. ACC/AHA guideline update for the management of patients with unstable angina and non–STsegment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366. 13. Lepeschkin E, Surawicz B. The measurement of the Q-T interval of the electrocardiogram. Circulation 1952;6:378. 14. Chaitman B. Exercise stress testing. In: Braunwald E, editor. Heart disease. A textbook of cardiovascular medicine. 6th ed. Philadelphia7 WB Saunders Company; 2001. p. 141. 15. Mark DB, Shaw L, Harrel FE. Prognostic value of a treadmill score in outpatients with suspected coronary artery disease. N Engl J Med 1991;325:849. 16. Mirvis D, Goldberger A. Electrocardiography. In: Braunwald E, editor. Heart disease. A textbook of cardiovascular medicine. 6th ed. Philadelphia7 WB Saunders Company; 2001. p. 82. 17. De Bacquer D, De Backer G, Kornitzer M, Blackburn H. Prognostic value of ECG findings for total, cardiovascular disease, and coronary heart disease death in men and women. Heart 1998;80:570. 18. Lehmann MH, Morady F. QT interval: metric for cardiac prognosis? Am J Med 2003;115:732. 19. Surawicz B, Kilmans T. Chou’s electrocardiography in clinical practice: adult and pediatric. Philadelphia, Pennsylvania7 WB Saunders Company; 2001. 20. Algra A, Tijssen JG, Roelandt JR, Pool J, Lubsen J. QTc prolongation measured by standard 12-lead electrocardiography is an independent risk factor for sudden death due to cardiac arrest. Circulation 1991; 83:1888. 21. Straus SM, Kors JA, De Bruin ML, et al. Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. J Am Coll Cardiol 2006;47:362. 22. Festa A, D’Agostino Jr R, Rautaharju P, et al. Is QT interval a marker of subclinical atherosclerosis in nondiabetic subjects? The Insulin Resistance Atherosclerosis Study (IRAS). Stroke 1999;30:1566. 23. Li YH, Rozanski GJ. Effects of human recombinant interleukin-1 on electrical properties of guinea pig ventricular cells. Cardiovasc Res 1993;27:525. 24. Magyar J, Iost N, Kortvely A, et al. Effects of endothelin-1 on calcium and potassium currents in undiseased human ventricular myocytes. Pflugers Arch 2000;441:144. 25. Fox KA, Poole-Wilson P, Clayton TC, et al. 5-year outcome of an interventional strategy in non–ST-elevation acute coronary syndrome: the British Heart Foundation RITA 3 randomised trial. Lancet 2005; 366:914.