predictive accuracy of ST depression dur”>American Journal of Emergency Medicine (2012) 30, 1042-1047

Original Contribution

Predictive accuracy of ST depression during rapid atrial fibrillation on the presence of obstructive coronary artery disease

Rajesh Pradhan MD ^a,?, Ashok Chaudhary MD ^b, Anthony A. Donato MD ^b

^aDepartment of Medicine, Division of Cardiology, Thomas Jefferson University Hospital, Philadelphia, PA 19107, USA

^bDepartment of Medicine, The Reading Hospital and Medical Center, West Reading, PA 19611, USA

Received 15 May 2011; revised 26 June 2011; accepted 27 June 2011

Abstract

Background: Rapid Atrial fibrillation is commonly associated with ST-segment depressions. ST- segment depression during a chest pain episode or exercise Stress testing in sinus rhythm is predictive of obstructive coronary artery disease , but it is unclear if the presence or magnitude of ST-segment depression during rapid AF has similar predictive accuracy.

Methods: One hundred twenty-seven patients with rapid AF (heart rate >=120 beats per minute) who had cardiac catheterization performed during the same hospital admission were retrospectively reviewed. Variables to compute Thrombolysis in Myocardial Infarction risk score, demographic profiles, ST-segment deviation, cardiac catheterization results, and Cardiac interventions were collected.

Results: Thirty-five patients had ST-segment depression of 1 mm or more, and 92 had no or less than 1 mm ST depression. Thirty-one patients were found to have obstructive CAD. In the group with ST- segment depression, 11 (31%) patients had obstructive CAD and 24 (69%) did not. In the group with less than 1 mm ST-segment depression, 20 (22%) had obstructive CAD and 72 (78%) did not (P = .25). Sensitivity, specificity, and positive and negative predictive values for presence of obstructive CAD were 35%, 75%, 31%, and 78%, respectively. The presence of ST-segment depression of 1 mm or more was not associated with presence of obstructive CAD before or after adjustment of TIMI variables. The relationship between increasing grades of ST-segment depression and obstructive CAD showed a trend toward significance (P = .09), which did not persist after adjusting for TIMI risk variables (P = .36). Conclusion: ST-segment depression during rapid AF is not predictive for the presence of obstructive CAD.

(C) 2012

Introduction

* Corresponding author. Haverford, PA 19041, USA. Tel.: +1 (215) 955 5050; fax: +1 484 412 8607.

E-mail addresses: [email protected] (R. Pradhan), [email protected] (A.A. Donato).

Atrial fibrillation is the most common arrhythmia in the United States [1], with a prevalence of 1% in patients presenting to emergency care [2]. Although nearly half of those with AF present with chest pain and ST-segment changes, only 1 in 20 will be diagnosed with myocardial infarction during that visit [3]. Ischemic electrocardiographic

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.06.027

(ECG) changes are predictive of coronary artery disease (CAD) in patients in sinus rhythm during an episode of acute chest pain [4-10] or when seen during stress testing [11-13]. When patients present with AF and rapid heart rate along with ischemic-appearing ECG changes, those changes are often interpreted as the equivalent of a “failed stress test.” Whether the presence or magnitude of ST-segment depres- sion seen during rapid AF has similar predictive value as in sinus rhythm or a lower predictive value as in paroxysmal supraventricular tachycardia (6%-30%) [14,15], suggesting a false-positive result, is not definitively known. We hypothesized that, in patients with AF and rapid ventricular response, increasing ST-segment depression would be associated with a higher likelihood of

obstructive CAD.

Materials and methods

Patient selection

This was a retrospective cohort study of records from a single medical center. We identified patients admitted to 1 academic community hospital with admission diagnosis of AF by International Classification of Diseases, Ninth Revision, coding who had been billed for cardiac catheter- ization during the same hospital admission from April 1996 to December 2007. Patients were included in the study if their presenting rhythm was AF with rapid ventricular response (initial presenting ECG rate >=120 beats per minute) as has been defined by other authors [16,17]. Patients were excluded if their ST segments were “nondiagnostic” as defined by American College of Cardiology/American Heart Association (ACC/AHA) exercise stress testing protocol [18], because of Left bundle-branch block, left ventricular hypertrophy (LVH) with repolarization changes, Digoxin use, preexcitation, Wolff-Parkinson-White (WPW) syndrome, and if they had electronically paced ventricular rhythm.

Measurements

Patients had all clinical variables collected to determine Thrombolysis In Myocardial Infarction risk score [19], types of AF (defined as “new” if recorded history and consultations did not document a prior diagnosis of AF and defined as “paroxysmal” if described in admission history or cardiology consultation as terminating without intervention), baseline medication for rate or Rhythm control, ECG changes, final catheterization results, and intended or performed cardiac interventions (which included percutaneous coronary intervention and coronary artery bypass surgery referral). We defined obstructive CAD as the presence of 50% or more stenosis of left main coronary artery or 70% or more stenosis of at least one of the rest of the epicardial coronary arteries.

Electrocardiographic classification

Two separate spreadsheets were used to collect all 12-lead ECG findings by 2 authors independently who did not have access to demographic profiles and outcomes of patients. Measurement of ST-segment depression was done as per ACC/AHA guideline update for exercise testing [18]. ST- segment depression was measured as an absolute value at 0.5- mm increments. PQ junction was taken as the isoelectric point. Deviation of ST-segment was measured 80 milliseconds after the J point. It was specified that there should be a presence of at least 1-mm ST-segment deviation in 3 consecutive beats in 2 consecutive leads to be considered abnormal. If the heart rate was faster than 130 beats per minute, ST-segment deviation was measured 60 milliseconds after the J point. The lead with the highest ST-segment depression was used to measure the absolute value. An independent third reader settled differences in reading of the ST-segment depression between the 2 readers.

Statistical methods

Patient data were analyzed with STATA version 8.0 (StataCorp, College Station, TX) for statistical analysis. Fisher exact, ?2, and Student t tests were used to compare the baseline characteristics in the groups with ST-segment depression of 1 mm or more and less, as appropriate. ?2 Testing was used to compare groups with no or less than 1 mm ST-segment depression vs ST-segment depression of 1 mm or more for the presence of obstructive CAD. Interobserver agreement for presence or absence of ST-segment depression was calculated using ? statistic and a weighted ? to compare the degree of ST- segment depression. Simple logistic regression models were used to control for other TIMI risk scores variables including prior documented obstructive CAD to determine whether ST- segment depression was independently associated with CAD. Increasing grades of ST-segment depression were examined for presence of CAD with the Student t test and adjusted for TIMI risk score with simple logistic regression. Sensitivity, specificity, and predictive accuracy of ECG changes of 1 mm or more for the presence of obstructive CAD were determined. We also performed regression analysis to see if the ST- segment depressions were related to the degree of tachycardia.

The Institutional Review Board of the Reading Hospital and Medical Center approved this study.

Results

Two hundred ninety patients were identified with a principal diagnosis of AF who received cardiac catheter- ization during the same hospital admission. One hundred twenty-seven patients met inclusion criteria, with others excluded for initial heart rates under 120 beats per minute (n = 81), rhythms other than AF (n = 30), and nondiagnostic ECGs (n = 47) (Fig. 1).

ST depression ? 1 mm N = 35
Presence of Obstructive CAD N = 11				Absence of Obstructive CAD N = 24
Intervention required N = 0				Intervention required N = 0

ST depression <1 mm N = 92
Presence of Obstructive CAD N = 20			Absence of Obstructive CAD N = 72
Intervention required N = 1 (PCI)			Intervention required N = 0

Fig. 1 Enrollment and outcomes of patients.

Of the 127 included patients, 35 had ST-segment depression of 1 mm or more, and 92 had no or less than 1 mm ST-segment depression (Fig. 1). Baseline characteristics of patients in these 2 groups are presented in Table 1. There were no significant differences in the frequencies of types of AF, hypertension, presence of prior CAD, anginal events, diabetes, smoking, family history of premature CAD, use of aspirin or ?-blockers, and presence of hyperlipidemia, but overall TIMI score after exclusion of ECG changes was higher (1.3 vs 1.7) in the ST-segment depression group, a trend that neared statistical significance (P = .054). Eleven patients were on Calcium-channel blockers (CCBs) at admission, and only 1 patient was receiving a class 1 antiarrhythmic. No patients were on class 3 antiarrhythmics at presentation. Patients in the group with ST-segment depression were more often female, older, and had faster ventricular rate. Calcium-channel blocker usage was statis- tically higher in the ST-segment depression group; however, there were only a total of 11 patients on CCB in both groups, so we doubt a clinically significant contribution to ST-

Patients meeting inclusion criteria (heart rate ? 120 beats per minute on admission ECG)

N = 127

Patients excluded secondary to non diagnostic ECG and heart rate not meeting inclusion criteria

N = 163

Rate <120 bpm = 81

LBBB = 10

LVH = 19

Digoxin use = 15

Non atrial fibrillation SVT = 30

Paced ventricle = 3 No ECG available = 5

Patients with a diagnosis of Atrial Fibrillation who had cardiac catheterization performed during the same hospital stay

N = 290

segment depression. A regression analysis revealed strong positive correlation between faster ventricular rate and increased ST-segment depression (P <= .001), which persisted even after controlling for atrioventricular (AV) nodal blockers and class 1 antiarrhythmics (P b .001).

Prediction of CAD

A total of 31 of 127 patients were found to have obstructive CAD. Of these patients, 25 had 1-vessel disease (12 left anterior descending, 6 left circumflex, and 7 Right coronary artery stenosis), 3 had 2-vessel disease, and 3 had 3-vessel disease. No patient had an obstructive left main coronary artery stenosis. In the group with ST-segment depression, 11 patients (31%) had obstructive CAD and 24 (69%) did not. In the group with less than 1 mm ST-segment depression (Fig. 1), 20 (22%) had obstructive CAD and 72 (78%) did not (P = .25). Sensitivity, specificity, and positive and negative predictive values were calculated to be 35%,

Patient characteristics	AF with b1 mm ST depression (n = 92)	AF with >=1 mm ST depression (n	=	35)	P
Age, y (mean +- SD)	61.9 +- 12.2	68.9 +- 11.8	.004
Male:female (n [ratio])	59:33 (1.78)	14:21 (0.66)	.014
Hypertension, n (%)	53 (57.6)	24 (68.5)	.31
Family h/o premature CAD, n (%)	12 (13)	2 (5.7)	.34
Diabetes mellitus, n (%)	12 (13)	3 (8.6)	.75
Smoking, n (%)	15 (16.3)	4 (11.4)	.58
Angina,a n (%)	9 (9.8%)	2 (5.7%)	.72
Aspirin use,b n (%)	33 (35.9%)	11 (31.4%)	.64
?-Blocker use, n (%)	27 (29.3)	8 (22.8)	.46
CCB, n (%)	5 (5.4)	6 (14.2)	.036
Hyperlipidemia, n (%)	38 (41)	13 (37.1)	.67
Prior CAD,c n (%)	9 (9.7)	4 (11.4)	.75
Newly diagnosed AF, n (%)	77 (83.7)	29 (82.8)	.91
Paroxysmal AF, n (%)	20 (21.7)	8 (22.9)	.89
Heart rate (mean +- SD)	132.4 +- 19.4	154 +- 15.4	b.0001
ECG abnormality,d n (%)	30 (32.6)	34 (97.1)	b.001
TIMI score excluding ECG criterione	1.32 +- 1.16	1.77 +- 1.2	.054
TIMI scoref (mean +- SD)	1.64 +- 1.33	2.74 +- 1.26	b.0001
a Two anginal episodes in the preceding 24-hour period. b Use of aspirin in the past 7 days. c Known prior obstructive coronary disease as defined in the TIMI risk score for unstable angina/non-ST elevation myocardial infarction [19]. d Defined as ST elevation or depression of 0.5 mm or more or T inversion of 2 mm or more in 2 leads (definition taken from TIMI risk score for unstable angina/non-ST elevation myocardial infarction (STEMI) [19]. e Composite of 6 TIMI risk score variables including Age 65 years or older, 3 or more risk factors for CAD, prior coronary stenosis of 50% or more, at least 2 anginal events in the prior 24 hours, use of aspirin in the prior 7 days, and elevated serum cardiac marker. Electrocardiographic changes excluded as the comparison between groups were based on presence or absence of ST depression [19]. f Total of TIMI risk score for unstable angina/non-ST elevation myocardial infarction (STEMI) [19].

75%, 31%, and 78%, respectively. The presence of ST- segment depression of 1 mm or more was not associated with the presence of obstructive CAD before or after adjustment of TIMI variables (Table 2). Six TIMI risk score variables were used for this calculation, including age 65 years or older, 3 or more risk factors for CAD, prior coronary stenosis of 50% or more, at least 2 anginal events in the prior 24 hours, use of aspirin in the prior 7 days, and elevated serum cardiac markers [19]. The TIMI risk of ECG changes was not used because we were comparing the groups with and without ST-segment depression for the presence of obstructive CAD as our primary outcome. None of the patients in the ST-segment depression group required invasive management, whereas 1 patient in the less-than-1-mm-ST-segment depres-

Table 1 Baseline characteristics of patients with AF with or without 1 mm ST depression

Table 2 Association between ST-segment depression and obstructive CAD

Unadjusted Adjusted ^a

b1 vs >=1 mm ST depression P = .25 ^b

Increasing grades of ST depression P = .096 ^d

P = .95 ^c

P = .36 ^e

^a Adjusted for TIMI risk score variables with exclusion of ECG criteria.

^b ?² test.

^c Logistic regression.

^d Student t test.

^e Logistic regression.

sion group underwent Percutaneous coronary intervention (PCI). Increasing grades of ST-segment depression were not statistically associated with obstructive CAD, although a trend toward it was seen (P = .096). When these data were adjusted to control for TIMI variables (with the exclusion of ECG changes), there was neither association nor trend seen between increasing grades of ST-segment depression and presence of obstructive CAD (P = .36).

Discussion

In our study, we found that patients in rapid AF with ST- segment depression had a prevalence of obstructive CAD of 31% (11/35), similar to the prevalence of 32.5% found by Androulakis et al [20]. ST-segment depression was not found to be associated with obstructive CAD both before and after adjustment for TIMI variables. Although there was a weak association between increasing grades of ST-segment depression and presence of obstructive CAD in our unadjusted analysis (P = .09), the association did not persist after adjusting for TIMI risk score variables (P = .36). Our results are contrary to those reported by Zimetbaum et al [3], who found an association between myocardial infarction and high-grade (>=2 mm) ST-segment depression. However, their conclusions were based on only 4 patients with 2 mm or more

ST-segment depression and were determined using cardiac biomarker elevation as their outcome; hence, the presence of obstructive CAD in these patients can only be inferred.

Given that this was a single-center retrospective study, we may have been limited by the quality of collected data from hospital records. We chose the retrospective design for convenience, cost, and feasibility. However, the hard end points we chose (ST-segment deviation, cardiac catheteriza- tion results) are not particularly subject to bias. The cohort selected (patients receiving cardiac catheterization) may have been a sicker cohort than unselected patients with AF, but we purposely sought a cohort in whom the suspicion of CAD was high to best determine the diagnostic value of ST- segment depression in patients with AF and rapid ventricular response. The sample size of our study (N = 127) was similar to the sample sizes of previously reported articles on this subject [3,20]. However, the size of our study may be inadequate to detect some meaningful associations.

The sensitivity and specificity of ST-segment depression during exercise testing for presence of obstructive coronary disease are 68% and 77%, respectively [12,13]. Ischemic- appearing ST-segment depression is also frequently ob- served during paroxysms of supraventricular tachycardia, but the predictive accuracy of such ST-segment depression is significantly less (6%-30%) in diagnosing CAD [14,15]. Our results support those of Androulakis et al [20] that show a low prevalence of obstructive CAD in patients with ST depression in the setting of AF with rapid ventricular response, with poor sensitivity for obstructive disease (35%). Some researchers in the past have found a positive correlation between ST-segment depression and heart rate in Paroxysmal supraventricular tachycardia (SVT) [21,22]. We observed a strong positive correlation between heart rate and degree of ST-segment depression in our study as well. Based on these findings, we think that the ST-segment depression was most probably a rate-related phenomenon and not a marker of coronary ischemia. We therefore suggest that physicians rely on validated risk stratification tools (eg, the TIMI and Global Registry of Acute coronary events risk score algorithms) as per ACC/AHA guide- lines [23] to manage patients with chest pain in rapid AF, whereas regarding the ST-segment depressions as nondiag-

nostic for presence of ischemia.

References

1. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and Stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285(18):2370-5.
2. Scott PA, Pancioli AM, Davis LA, Frederiksen SM, Eckman J. Prevalence of atrial fibrillation and antithrombotic prophylaxis in emergency department patients. Stroke 2002;33(11):2664-9.
3. Zimetbaum PJ, Josephson ME, McDonald MJ, et al. Incidence and predictors of myocardial infarction among patients with atrial fibrillation. J Am Coll Cardiol 2000;36(4):1223-7.
4. 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;30:133-40.
5. Savonitto S, Ardissino D, Granger CB, et al. Prognostic value of the admission electrocardiogram in acute coronary syndromes. JAMA 1999;281:707-13.
6. Rude RE, Poole WK, Muller JE, et al. Electrocardiographic and clinical criteria for recognition of acute myocardial infarction based on analysis of 3,697 patients. Am J Cardiol 1983 Nov 1;52(8):936-42.
7. Nyman I, Areskog M, Areskog NH, Swahn E, Wallentin L. Very early risk stratification by electrocardiogram at rest in men with suspected unstable coronary heart disease. The RISC Study Group. J Intern Med 1993;234:293-301.
8. Holmvang L, Clemmensen P, Wagner G, Grande P. Admission standard electrocardiogram for early risk stratification in patients with unstable coronary artery disease not eligible for acute revascularization therapy: a TRIM substudy. Am Heart J 1999;137:24-33.
9. Cohen M, Hawkins L, Greenberg S, Fuster V. Usefulness of ST- segment changes in greater than or equal to 2 leads on the emergency room electrocardiogram in either unstable angina pectoris or non-Q- wave myocardial infarction in predicting outcome. Am J Cardiol 1991;67:1368-73.
10. Schechtman KB, Capone RJ, Kleiger RE, et al. Risk stratification of patients with non-Q wave myocardial infarction. The critical role of ST segment depression. The Diltiazem Reinfarction Study Research Group. Circulation 1989;80:1148-58.
11. Goldschlager N, Selzer A, Cohn K. Treadmill stress tests as indicators of presence and severity of coronary artery disease. Ann Intern Med 1976;85(3):277-86.
12. Gianrossi R, Detrano R, Mulvihill D, et al. Exercise-induced ST depression in the diagnosis of coronary artery disease. A meta- analysis. Circulation 1989;80(1):87-98.
13. Morise AP, Diamond GA. Comparison of the sensitivity and specificity of exercise electrocardiography in biased and unbiased populations of men and women. Am Heart J 1995 Oct;130(4):741-7.
14. S. Gulec, F. Ertas, R. Karaoouz, M. Guldal, A. Alpman, D. Oral. Value

of ST-segment depression during paroxysmal supraventricular tachy- cardia in the diagnosis of coronary artery disease. Am J Cardiol 1999; 83(3):458-60, A10.

1. Imrie JR, Yee R, Klein GJ, Sharma AD. Incidence and clinical significance of ST segment depression in supraventricular tachycardia. Can J Cardiol 1990;6(8):323-6.
2. Feld GK, Fleck RP, Fujimura O, Prothro DL, Bahnson TD, Ibarra M. Control of rapid ventricular response by radiofrequency catheter modification of the atrioventricular node in patients with medically Refractory atrial fibrillation. Circulation 1994;90(5):2299-307.
3. Demircan C, Cikriklar HI, Engindeniz Z, et al. Comparison of the effectiveness of Intravenous diltiazem and metoprolol in the management of rapid ventricular rate in atrial fibrillation. Emerg Med J 2005;22(6):411-4.
4. Gibbons RJ, Balady GJ, Bricker JT, et al. American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). ACC/AHA 2002 guideline update for exercise testing: summary article: a report of the American College of Cardiology/A- merican Heart Association Task Force on Practice Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). Circulation 2002;106(14):1883-92.
5. Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284:835-42.
6. Androulakis A, Aznaouridis KA, Aggeli CJ, et al. Transient ST- segment depression during paroxysms of atrial fibrillation in otherwise normal individuals: relation with underlying coronary artery disease. J Am Coll Cardiol 2007;50(19):1909-11.
7. Takayanagi K, Hoshi H, Shimizu M, et al. Pronounced ST segment depression during paroxysmal supraventricular tachycardia. Jpn Heart J 1993;34:269-78.
8. Kim YN, Sousa J, El-Attasi R, et al. Magnitude of ST segment depression during paroxysmal supraventricular tachycardia. Am Heart J 1991;122:1486-7.
9. Anderson JL, Adams CD, Antman EM, et al. American College of Cardiology; American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-ST- Elevation Myocardial Infarction); American College of Emergency Physicians; Society for Cardiovascular Angiography and Interven- tions; Society of thoracic surgeons; American Association of

Cardiovascular and Pulmonary Rehabilitation; Society for Academic Emergency Medicine. ACC/AHA 2007 guidelines for the manage- ment of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-ST-Elevation Myocardial Infarction) developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiog- raphy and Interventions, and the Society of Thoracic Surgeons endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. Circulation 2007;116:e148-304.