It is occasionally problematic to differentiate ST-segment

CLINICAL INVESTIGATION Differential Diagnosis of Acute Pericarditis From Normal Variant Early Repolarization and Left Ventricular Hypertrophy With Early Repolarization: An Electrocardiographic Study Ravindra Bhardwaj, MD, MPH, Chalak Berzingi, MD, Christopher Miller, MD, Gerald Hobbs, PhD, Wissam Gharib, MD, Robert J. Beto, MD, Bradford E. Warden, MD and Abnash C. Jain, MD Abstract: Background: Differentiation of ST-segment elevation on electrocardiogram (ECG) from acute pericarditis (AP), normal variant early repolarization (ER) and early repolarization of left ventricular hypertrophy (ERLVH) can be problematic. Hence, the authors evaluated the accuracy of the ST/T ratio in ECG to more optimally differentiate between AP, ST-segment elevation, ER and ERLVH. Methods: Between September 2006 and July 2010, 80 patients were enrolled in this study consisting of 25 individuals with AP, 27 with ER and 28 with ERLVH. Each ECG was analyzed in a systematic manner including the measurement of PR interval, QRS duration, QT-segment duration, PR-segment deviation, ST-segment deviation and the height of T wave. The ratio of the height of ST segment to the height of T wave was measured in leads I, II, III, avf and V 2 through V 6. Results: The mean ages of the patients with AP, ER and ERLVH were 32 6 16.5, 36 6 15.4 and 53 6 16 years, respectively. The ratio of the amplitude of ST segment to the amplitude of the T wave in leads I, V 4,V 5 and V 6 proved to be a significant discriminator at a value of $0.25 (P, 0.05 for all). Conclusions: Leads I, V 4,V 5 and V 6 can all be used to differentiate AP from ER and ERLVH. When ST elevation is present in lead I, the ST/T ratio has the best predictive value (0.82) to more accurately discriminate between AP, ER and ERLVH. Key Indexing Terms: ST/T ratio; Acute pericarditis; Normal variant early repolarization; Early repolarization of left ventricular hypertrophy. [Am J Med Sci 2013;345(1):28 32.] It is occasionally problematic to differentiate ST-segment elevation on electrocardiogram (ECG) from acute pericarditis (AP) and other causes of ST-segment elevation, such as a normal variant early repolarization (ER) and early repolarization of left ventricular hypertrophy (ERLVH). 1,2 When patients present to the emergency departments with chest pain, ECG can be a vital noninvasive test used to reach a clinical diagnosis. 3 5 However, one of the challenges of using ECG to diagnose chest pain is to determine the presence of AP compared with ER and ERLVH. Ginzton and Laks 3 distinguished AP from variants of ER by concluding that the former is more likely to occur when the ratio of ST/T in lead V 5 or V 6 is.0.25. These conditions can also be classified by ECG based on their wave characteristics. For example, Ginzton and Laks 3 indicated that AP occurs when the amplitude of a T wave From the Department of Medicine, Section of Cardiology (RB, CB, CM, WG, RJB, BEW, ACJ), and Department of Statistics (GH), West Virginia University School of Medicine, Morgantown, West Virginia. Submitted November 3, 2011; accepted in revised form January 19, 2012. Correspondence: Abnash C. Jain, MD, Department of Medicine, Section of Cardiology, West Virginia University School of Medicine, Morgantown, WV 26506 (E-mail: ajain@hsc.wvu.edu). in V 6 is,0.3 mv. Other challenges of ST-segment elevation include patients with left bundle branch block, hyperkalemia, pulmonary embolism and acute myocardial infarction. 4 8 There are several other ECG criteria used to distinguish AP from ER, including the degree of PR-segment deviation and ST elevation in V 6. 1,9 11 Undoubtedly, clinical presentation is most helpful, but simplified and easily measurable ECG criteria will serve to provide for a more accurate approach in diagnosing chest pain. Thus, the objective of this study was to evaluate the accuracy of the ratio of ST/T in ECG to more optimally differentiate between AP, ST-segment elevation, ER and ERLVH. MATERIALS AND METHODS Patient Selection This study was conducted at the West Virginia University Hospital, Morgantown, West Virginia, after receiving the approval of the Institutional Review Board for the Protection of Human Subjects for review of patients diagnosed with AP, ER and ERLVH. Between September 2006 and July 2010, a total of 80 patients were enrolled in the study and divided into 3 groups based on their clinical diagnosis. Group A consisted of 25 individuals with AP, group B of 27 individuals with ER and group C of 28 patients with ERLVH. Demographic data (ie, age, sex, weight, height, clinical risk factors and medications) were also gathered from the medical records and assessed. Group A consisted of patients with a serial ECG evidence of the evolution of acute pericarditis, pericardial friction rub, and echocardiographic evidence for pericardial effusion. After analyzing series of ECGs, the reference ECG with a maximum ST elevation was used for analysis (Figure 1A). It is important to note that unlike group A, group B comprised entirely of healthy adult patients with normal variant ER. These individuals did not have any clinical or historical evidence of cardiovascular disease (Figure 1B). Finally, group C included patients with an ER pattern and left ventricular hypertrophy using the Sokolow-Lyon index 12 and secondary ST-segment and T-wave changes (Figure 1C). Analysis of ECG All standard 12-lead ECGs were recorded at a paper speed of 25 mm/s. Each ECG was analyzed in a systematic manner including the measurement of PR interval, QRS duration, QT-segment duration, PR-segment deviation, ST-segment deviation and the height of T wave. All measurements were performed by the investigator using calipers and a magnifying lens (10-fold magnification). The ratio of the height of ST segment to the height of T wave was measured in leads I, II, III, avf and V 2 through V 6. The ST/T ratio was obtained by dividing the amplitude of the 28 The American Journal of the Medical Sciences Volume 345, Number 1, January 2013

ST/T Ratio in AP, ER and ERLVH FIGURE 1. Electrocardiogram patterns of acute pericarditis (A), early repolarization (B) and early repolarization of left ventricular hypertrophy (C). onset of the ST segment by the amplitude of the T wave. The ST/T ratio could not be calculated for the leads that had either flat or negative T waves. The end of PR segment was used as baseline for ST-segment and T-wave amplitudes. The peak of T wave was used for T-wave amplitude (Figure 2). The ECGs were reanalyzed and confirmed by a second cardiologist independently for interobserver variability. Statistical Analysis Data are presented as means and standard deviations for continuous variables for each of the 3 groups. The differences between groups were evaluated using the analysis of variance controlling for heart rate for continuous variables and the x 2 test for categorical variables. The results were considered statistically significant at a P value,0.05. The interrater agreement for T-wave and ST-segment amplitudes in various leads was evaluated by calculating Cohen s kappa coefficient. RESULTS Table 1 summarizes the demographics of the 3 groups including the ECG analysis. Group A comprised 19 men and 6 women with a mean age of 32 6 16.5 years, group B comprised 26 men and 1 woman with a mean age of 36 6 15.4 years, and group C comprised 22 men and 6 women with a mean Ó 2013 Lippincott Williams & Wilkins 29

Bhardwaj et al FIGURE 2. A diagrammatic representation of the different electrocardiogram segments. The end of PR segment was used as baseline for ST-segment and T-wave amplitudes, and the peak of T wave was used for T-wave amplitude. age of 53 6 16 years. Twenty-six patients in group C had hypertension and/or ischemic heart disease without any evidence of acute myocardial infarction, 1 had aortic regurgitation and 1 had mitral regurgitation. In groups A and B, the ST/T ratio was obtained in all patients in leads V 3,V 4,V 5 and V 6, and in 76% of the patients in lead I. In group C, the ST/T ratio was obtained in 90% of patients in lead V 3, 68% in leads V 4 and V 5, 47% in lead V 6, and 15% in lead I. The mean heart TABLE 1. Demographic and electrocardiographic characteristics of the 3 groups Mean 6 SD Group A (n 5 25) Group B (n 5 27) Group C (n 5 28) P a P b P c Mean age (yr) 32 6 16.5 36 6 15.4 53 6 16 0.37 0.00 0.00 Sex (male/female) 19/6 26/1 22/6 0.03 0.82 0.05 Mean heart rate (bpm) 87 6 27 63 6 15 66 6 12 0.00 0.00 0.32 PR depression 8 0 0 0.00 0.00 Mean PR interval (ms) 143 6 26 162 6 34 175 6 49 0.03 0.00 0.28 Mean QRS duration (ms) 94 6 18 92 6 9 96 6 17 0.69 0.67 0.31 T-wave amplitude (mm) I 2.8 6 0.66 3.4 6 0.85 3.3 6 1.7 0.02 0.16 0.90 II 3.5 6 2 4.6 6 1.4 5.3 6 1.1 0.81 0.24 0.29 III 3.4 6 1.9 3 6 3.9 4 6 2 0.14 0.90 0.43 avf 3.3 6 1.9 3.6 6 1.2 4.1 6 1.5 0.25 0.75 0.51 V 2 5.8 6 3 4.9 6 2.9 4.4 6 2.1 0.43 0.16 0.59 V 3 5.7 6 2.7 6.6 6 2.2 6.3 6 2.2 0.40 0.61 0.68 V 4 5.6 6 3.5 6.4 6 1.8 6.4 6 2.8 0.81 0.84 0.97 V 5 5 6 3.3 5.7 6 1.7 6 6 2.4 0.67 0.94 0.61 V 6 4.2 6 2.8 4.8 6 1.6 5.4 6 2.1 0.67 0.57 0.31 ST/T ratio I 0.36 6 0.14 0.19 6 0.07 0.17 6 0.02 0.00 0.04 0.83 II 0.4 6 0.2 0.22 6 0.08 0.19 6 0.08 0.02 0.03 0.61 III 0.49 6 0.26 0.26 6 0.13 0.15 6 0.08 0.11 0.02 0.19 avf 0.43 6 0.21 0.23 6 0.1 0.2 6 0.12 0.01 0.03 0.73 V 2 0.29 6 0.1 0.31 6 0.11 0.27 6 0.13 0.53 0.61 0.27 V 3 0.32 6 0.11 0.24 6 0.07 0.20 6 0.09 0.02 0.00 0.34 V 4 0.36 6 0.12 0.21 6 0.07 0.21 6 0.13 0.01 0.01 0.83 V 5 0.42 6 0.16 0.20 6 0.08 0.18 6 0.09 0.00 0.00 0.39 V 6 0.41 6 0.2 0.19 6 0.08 0.22 6 0.14 0.00 0.01 0.71 a P value between groups A and B. b P value between groups A and C. c P value between groups B and C. bpm, beats per minute; group A, patients with acute pericarditis; group B, individuals with early repolarization; group C, patients with early repolarization of left ventricle hypertrophy; SD, standard deviation. 30 Volume 345, Number 1, January 2013

ST/T Ratio in AP, ER and ERLVH rates in groups A, B and C were 87 6 27, 63 6 15 and 66 6 12 beats per minute, respectively. The Mean ST/T Ratio in Leads V 4,V 5 and V 6 The mean ST/T ratios for leads V 4,V 5 and V 6 were determined for all 3 groups. Group A had mean ST/T ratios of 0.36 6 0.12, 0.42 6 0.16 and 0.41 6 0.2, respectively. These values were found to be statistically significant, with P values of 0.01, 0.00 and 0.00, compared with those for group B where the mean ratios in all 3 leads were 0.21 6 0.07, 0.20 6 0.08 and 0.19 6 0.08, respectively. In group C, the ratios of ST/T remained statistically significant, with P values of 0.01, 0.00 and 0.01, compared with those for group A. This group also had ST/T ratios in all 3 leads of 0.21 6 0.13, 0.18 6 0.09 and 0.22 6 0.14, respectively. There was no significant difference in ST/T lead V 4,V 5 and V 6 ratios between groups B and C. The positive predictive values of an ST/T ratio $0.25 in leads V 4,V 5 and V 6 were 0.77, 0.75 and 0.77, whereas the negative predictive values were 0.95, 0.97 and 94, respectively (Figures 3B D). The Mean ST/T Ratio in Lead I The group A mean ST/T ratio in lead I was 0.36 6 0.14, which was statistically significant compared with that for group B, with a value of 0.19 6 0.07 (P 5 0.00). In group C, the ST/T ratio was statistically significant compared with that for group A, with mean of 0.17 6 0.02 (P 5 0.04). In groups B and C, there was no significant difference in the ST/T ratio in leads V 4, V 5 and V 6. In lead I, the positive predictive value indicating pericarditis was 0.82, and the negative predictive value was 0.95 (Figure 3A). The Mean ST/T Ratio in Leads II and avf The mean ST/T ratios in leads II and avf were 0.4 6 0.2 and 0.43 6 0.21 in group A, 0.22 6 0.08 and 0.23 6 0.1 in group B, and 0.19 6 0.08 and 0.2 6 0.12 in group C, respectively. A comparison of groups A and B indicated a statistical significance between the ST/T ratios in leads II and avf (P 5 0.02 and 0.01). This also occurred when analyzing groups A and C. In lead II, the positive predictive value of an ST/T ratio $0.25 was 0.72, and the negative predictive value was 0.90. The positive and negative predictive values for lead avf were 0.68 and 1, respectively. The interrater agreement calculated by Cohen s kappa coefficient was very good for both T-wave and ST-segment amplitudes (k 5 0.84 and 0.92). T-Wave Amplitude The T-wave amplitude in various leads was mostly not a significant discriminator between the 3 groups except in lead I, with a P value of 0.02 between groups A and B. The mean T-wave amplitude in different groups is shown in Table 1. DISCUSSION Ginzton and Laks 3 concluded that the ratio of ST/T in lead V 6 $0.25 is the best discriminator between the ECG of acute pericarditis and normal variant ER. Also, the amplitude of a T wave,0.3 mv was a significant discriminator in lead V 6.If lead V 6 was not available, then the ratios in leads I, V 4 and V 5 were also highly suggestive of AP. The ST/T ratios of leads V 4, V 5 and V 6 were all very useful in discriminating AP from ER and ERLVH in this study. Among the 25 patients in group A, 19 individuals had ST-segment elevation in lead I. This resulted in the greatest predictive value of 0.82. The T-wave amplitude, which is considered a very helpful discriminator by Ginzton FIGURE 3. An ST/T ratio in leads I (A), V 4 (B), V 5 (C) and V 6 (D) of $0.25 (in at least one of these leads) diagnosed all patients with pericarditis. The dashed lines represent the breakpoints used in calculating tests for predictive accuracy. AP, acute pericarditis; ER, early repolarization; ERLVH, early repolarization of left ventricular hypertrophy; NPV, negative predictive value; PPV, positive predictive value. Ó 2013 Lippincott Williams & Wilkins 31

Bhardwaj et al and Laks, 3 was not actually significant and varied from 0.2 to 1.5 mv in our study. In a few patients with AP, the ECG pattern may be indistinguishable from that of ER, which raises concerns about the accurate diagnosis. However, there are numerous ECG criteria that usually differentiate between the 2, including the degree and ubiquity of PR-segment deviations. 1,2,11 Marked and widespread PR depressions are not observed in ER. In this study, PR depression was only observed in 8 of the 25 patients in group A and not in group B or C. There was a small overlap in the ST/T ratio among the groups, but on the whole, there was very little room for misdiagnosis. We have confirmed Ginzton s observation of ECG discrimination for ST/T ratios in AP and ER and have also applied them to ERLVH. The ER variant seen on an ECG may be present in as many as 30% of young adults and is often confused with AP. 1 ER is characterized by ST elevation of the J point, which represents the junction between the end of the QRS complex (termination of depolarization) and the beginning of the ST segment (onset of ventricular repolarization). In ER and AP, ST elevation is most often present in the anterior and lateral chest leads (V 3 through V 6 ), although other leads may be involved. 1,10 The ECG in patients with ERLVH exhibits ST-elevation pattern, which is unlike the typical left ventricular hypertrophy pattern with secondary ST depression and T-wave inversion. Additionally, the J junction may form a depression and demonstrate a dome-like upward convexity. 2 This represents the combined and opposed interaction of repolarization changes as a result of left ventricular hypertrophy and ER. The QRS duration is normal in patients with ERLVH, 2 as was observed in our patient population. In this study population, our conclusions may have been limited because of the retrospective analysis of the ECG. To conclude, by comparing the ECGs of 3 different subject groups, it was found that ST/T ratios.0.25 in leads I, V 4,V 5 and V 6 can all be used to differentiate AP, ER and ERLVH. If ST elevation is present in lead I, an ST/T ratio.0.25 has the best predictive value (0.82) to discriminate between AP, ER and ERLVH. ACKNOWLEDGMENTS The authors sincerely thank Alan M. Ducatman, MD, and Shelley Layman, BA, for their help in this study. REFERENCES 1. Spodick DH. Differential characteristics of the electrocardiogram in early repolarization and acute pericarditis. N Engl J Med 1976;295:523. 2. Boineau JP. The early repolarization variant an electrocardiographic enigma with both QRS and J-STT anomalies. J Electrocardiol 2007; 40:3 4. 3. Ginzton LE, Laks MM. The differential diagnosis of acute pericarditis from the normal variant: new electrocardiographic criteria. Circulation 1982;65:1004 9. 4. Mansi I. Electrocardiographic diagnosis of pericarditis. JAMA 2003; 289:2942. 5. Wang K, Asinger RW, Marriott HJL. ST-segment elevation in conditions other than acute myocardial infarction. N Engl J Med 2003;349:2128 35. 6. Troughton RW, Asher CR, Klein AL. Pericarditis. Lancet 2004; 363:717. 7. Chandra S, Singh V, Nehra M, et al. ST-segment elevation in non-atherosclerotic coronaries: a brief overview. Intern Emerg Med 2011;6:129 39. 8. Brady WJ. ST segment and T wave abnormalities not caused by acute coronary syndromes. Emerg Med Clin North Am 2006;24:91 111. 9. Spodick DH. Acute pericarditis: current concepts and practice. JAMA 2003;289:1150 3. 10. Mehta M, Jain AC, Mehta A. Early repolarization. Clin Cardiol 1999; 22:59 65. 11. Klatsky AL, Oehm R, Cooper RA, et al. The early repolarization normal variant electrocardiogram: correlates and consequences. Am J Med 2003;115:171. 12. Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as obtained by unipolar precordial and limb leads. Am Heart J 1949;37:161 86. 32 Volume 345, Number 1, January 2013