doi: 10.1161/CIRCULATIONAHA.105.581942
(Circulation. 2006;113:464-466.)
© 2006 American Heart Association, Inc.
Editorial |
Electrocardiography in Women
Taking the Initiative
From the Division of Cardiology, Department of Medicine, Cornell Medical Center, New York, NY.
Correspondence to Peter M. Okin, MD, Cornell Medical Center, 525 E 68th St, New York, NY 10021. E-mail pokin{at}med.cornell.edu
Key Words: Editorials • electrocardiography • epidemiology • prognosis • sex
In the just over 100 years since the first ECG was performed,1 the ECG has become the most extensively used noninvasive diagnostic and prognostic tool in cardiology. Used both at rest and during provocative exercise, the 12-lead ECG has impressive, if imperfect, utility for rhythm analysis, detection of ischemic and hypertrophic heart disease, and outcome prediction in a variety of clinical settings, with a large body of literature that illustrates and supports these applications. The first observation of gender differences in the ECG was published 85 years ago by Bazett,2 demonstrating that women have significantly longer QT intervals than men despite having higher heart rates. However, despite a growing body of literature demonstrating significant gender differences in QRS amplitudes and duration,3–6 QT intervals,5,7,8 ST-segment deviation,9,10 and novel, computer-based measurements of T-wave complexity,8,11 few ECG criteria routinely use gender-specific diagnostic criteria, and there has been a relative paucity of data on the prognostic performance of ECG variables in women.
Articles pp 473 and 481
Two studies in the current issue of Circulation by Rautaharju and colleagues12,13 provide a wealth of new findings and impetus for further study of the ECG in women. Using well-validated computerized ECG methodology, the authors examined the value of a number of ECG variables for predicting incident coronary heart disease and its mortality, incident congestive heart failure, and total mortality in more than 38 000 women participating in the dietary modification trial of the Women’s Health Initiative. Although both studies reconfirm the predictive value of ECG evidence of prior Q-wave myocardial infarction (MI), the principle new findings are the strong predictive value of various ECG measures of repolarization, and in particular of the QRS/T angle,12,13 a measure of the spatial angle between mean QRS and T vectors. In the single-ECG multivariable models that adjusted for standard demographic and clinical risk factors, nearly all T-wave and ST-segment variables and the QRS/T angle, as well as novel measures of repolarization derived from singular value decomposition of the derived vectorcardiogram, were significant predictors of outcome. When all ECG variables were examined together in these models, the QRS/T angle retained its predictive value along with a number of other ECG measures, including rate-corrected QT interval, heart rate variability, and ECG evidence of prior Q-wave MI.
These 2 reports build on a number of previous studies that examined the prognostic value of the ECG in women alone14 or compared predictive value in men and women.15–17 Earlier reports that included large numbers of women and men15–17 predominantly demonstrated significant predictive value of standard ECG measures of old MI and major and minor repolarization abnormalities by Minnesota coding for cardiovascular and all-cause mortality. In all of these studies, however, identical criteria were used in men and women despite known gender differences in many of these variables, which potentially underestimated the value of these findings in women. In contrast, the present findings are strongly supported by a recent report in a small subset of the Women’s Ischemia Syndrome Evaluation (WISE) study.14 In 143 women undergoing diagnostic catheterization, QRS/T angle, QRS duration, rate-corrected QT interval, and ST depression in V5 all remained predictive of a combined outcome of cardiovascular disease end points that included cardiovascular death, congestive heart failure, and nonfatal MI; however, the small number of events among this subset of women (n=18) precluded the inclusion of all of the ECG variables in the final multivariable models, as was reported,14 which limits the conclusions that can be derived with respect to the independent value of these ECG variables in this population.
There are numerous strengths of the studies by Rautaharju et al.12,13 First, the size of the population and large number of events allowed adequate statistical power to assess specific outcomes in relation to multiple different ECG variables. Second, the authors used well-documented and readily reproducible computerized ECG measurements and criteria, as opposed to more qualitative and descriptive analyses that have been all too often employed in ECG-based studies.18 In addition, for the most part, the investigators took advantage of quantitative measurements of continuous variables to derive threshold partitions based on findings in the most abnormal decile or quartile of values, which allowed for the use of test partitions specific to women. It is somewhat unfortunate that the authors did not follow a similar approach to analysis of prior Q-wave MI but instead utilized well-validated Novacode criteria that were previously derived in both men and women.12,13
Several potential weaknesses of these studies should also be noted. First, although the strong predictive value of the QRS/T angle for events is one of the most important results, this measurement is not familiar to most clinicians and is not routinely available from computerized ECG analysis software currently in use, which potentially limits the applicability of this finding. Given the significant correlations of the QRS/T angle with other repolarization variables and the similar predictive value of these variables in the single-ECG models, it would have been informative to examine whether use of traditional measures of repolarization such as T-wave amplitudes and ST-segment deviation would have provided similar levels of prognostic value in the multi-ECG models if QRS/T angle was not included in the analyses. In addition, the greater risk concentration provided by the QRS/T angle may in part reflect the use of 2 levels of abnormality, borderline and high, with the greatest risk found in the widest-angle group. It is possible that similar risk stratification may have been observed if the ST segment in V5 had been further examined in the abnormal range at levels below –50 µV, particularly given that the partition value used for ST depression in V5 was not based on its distribution in the study population.12,13 Third, use of mean amplitude values from the orthonormal expansion and not more standard peak amplitude measurements that are routinely available and familiar to most physicians may limit applicability of the specific thresholds for abnormality in these studies. Finally, given the large number of black women in the population and known ethnic differences in ECG measures,5,19 it is unclear why ethnicity was not included as a covariate in their analyses. As noted by the authors,12 further evaluation of possible variations in their findings by ethnicity will be important.
The importance of continued examination of the ECG in women and the further development of gender-specific ECG criteria is highlighted by the fact that the use of identical diagnostic criteria in men and women can adversely affect performance of ECG criteria for left ventricular hypertrophy and for the detection of coronary disease during exercise testing.6,10 We and others have demonstrated that women, both with and without hypertrophy, have lower QRS voltages and shorter QRS durations than men, even after adjustment for gender differences in left ventricular mass and body weight.6 As a consequence, when non–gender-specific criteria are employed, the ECG exhibits higher specificity and lower sensitivity for the detection of hypertrophy in women than in men. Similarly, women with coronary disease have significantly less exercise-induced ST depression than men.10 As a result, non–gender-specific ST depression and heart-rate adjusted ST depression criteria have lower sensitivity in women than in men.10 Use of gender-specific test criteria with matched specificity in both men and women significantly improves sensitivity in women without reducing test sensitivity in men.10 Gender differences have also been demonstrated in the total cosine of the R-to-T,8 a measure of the vectorial deviation between the depolarization and repolarization wave front that is similar in concept to the QRS/T angle used by Rautaharju et al,12–14 and in other novel measures of T-wave complexity that have been demonstrated to stratify arrhythmic and mortality risk,8,11 but these findings are based on data in only a limited number of women. Further study of these variables in larger populations of women will be necessary to adequately assess potential differences between men and women.
Discussion of the possible mechanisms for the gender differences in ECG measures is beyond the scope of this editorial. However, a recent review20 of sex differences in cardiac repolarization provides an extensive discussion of the experimental data and potential gender differences in cellular-level ionic currents and the possible roles played by sex hormones in some of the better-characterized gender differences in cardiac repolarization.
The findings by Rautaharju and colleagues12,13 have important implications and provide direction for future investigation. Despite the widespread misconception that the ECG is of limited utility in women, these studies clearly demonstrate the value of the ECG for risk stratification in women, in particular the strong prognostic value of ECG measures of abnormal repolarization when using threshold criteria derived in women. Taken together with the wealth of information documenting the predictive value of the ECG in men, these findings strongly support the routine clinical application of computer-based ECG measurements for risk stratification in women. The known gender differences in QRS duration and amplitudes raise the question of whether sex-specific criteria for MI or bundle-branch blocks may be indicated to further enhance the predictive value of these ECG variables in women. Additional comparisons of ECG measurements in men and women will be necessary to more clearly delineate true mechanistic differences in ECG variables between men and women from differences that may be attributable to gender differences in left ventricular mass, body size, and composition, with application of these findings to improve accuracy of the ECG in both women and men.
In light of the low cost and widespread availability of the ECG and the increasing economic pressures on the practice of medicine, it is imperative that we continue to improve the diagnostic and prognostic performance of the ECG in women. Now that we have seized the initiative with respect to ECG research in women, we should not let it go to waste.
 |
Footnotes |
|---|
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
|
References |
|---|
 Top References |
|---|
1. Einthoven W. Galvanometrische registratie van het menschelijk electrocardiogram. In: Rosenstein SS, ed. Herinneringsbundel. Leiden, Netherlands: Eduard Ijdo; 1902: 102.
2. Bazett HC. An analysis of the time-relations of electrocardiograms. Heart. 1920; 7: 353–370.
3. Simonson E, Blackburn H, Puchner TC, Eisenberg P, Ribeiro F, Meja M. Sex differences in the electrocardiogram. Circulation. 1960; 22: 598–601.
4. Casale PN, Devereux RB, Alonso DR, Campo E, Kligfield P. Improved sex-specific criteria of left ventricular hypertrophy for clinical and computer interpretation of electrocardiograms: validation with autopsy findings. Circulation. 1987; 75: 565–572.
5. Wu J, Kors JA, Rijnbeek PR, van Herpen G, Lu Z, Xu C. Normal limits of the electrocardiogram in Chinese subjects. Int J Cardiol. 2003; 87: 37–51.[CrossRef][Medline] [Order article via Infotrieve]
6. Okin PM, Roman MJ, Devereux RB, Kligfield P. Gender differences and the electrocardiogram in left ventricular hypertrophy. Hypertension. 1995; 25: 242–249.
7. Merri M, Benhorin J, Alberti M, Locati E, Moss AJ. Electrocardiographic quantitation of ventricular repolarization. Circulation. 1989; 80: 1301–1308.
8. Smetana P, Batchvarov VN, Hnatkova K, Camm AJ, Malik M. Sex differences in repolarization homogeneity and its circadian pattern. Am J Physiol Heart Circ Physiol. 2002; 282: H1889–H1897.
9. Dellborg M, Herlitz J, Emanuelsson H, Swedberg K. ECG changes during myocardial ischemia: differences between men and women. J Electrocardiol. 1994; 27 (suppl): 42–45.[CrossRef][Medline] [Order article via Infotrieve]
10. Okin PM, Kligfield P. Gender-specific criteria and performance of the exercise electrocardiogram. Circulation. 1995; 92: 1209–1216.
11. Okin PM, Xue Q, Reddy S, Kligfield P. Electrocardiographic quantitation of heterogeneity of ventricular repolarization. Ann Noninvasive Electrocardiol. 2000; 5: 79–87.[CrossRef]
12. Rautaharju PM, Kooperberg C, Larson JC, LaCroix A. Electrocardiographic abnormalities that predict coronary heart disease events and mortality in postmenopausal women: the Women’s Health Initiative. Circulation. 2006; 113: 473–480.
13. Rautaharju PM, Kooperberg C, Larson JC, LaCroix A. Electrocardiographic predictors of incident congestive heart failure and all-cause mortality in postmenopausal women: the Women’s Health Initiative. Circulation. 2006; 113: 481–489.
14. Triola B, Olson MB, Reis SE, Rautaharju P, Bairey Merz CN, Kelsey SF, Shaw LJ, Sharaf BL, Sopko G, Saba S. Electrocardiographic predictors of cardiovascular outcome in women: the National Heart, Lung, and Blood Institute-sponsored Women’s Ischemia Syndrome Evaluation (WISE) study. J Am Coll Cardiol. 2005; 46: 51–56.
15. Liao Y, Liu K, Dyer A, Schoenberger JA, Shekelle RB, Collette P, Stamler J. Sex differential in the relationship of electrocardiographic ST-T abnormalities to risk of coronary death: 11.5 year follow-up findings of the Chicago Heart Association Detection Project in Industry. Circulation. 1987; 75: 347–352.
16. 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: 570–577.
17. Greenland P, Xie X, Liu K, Liu K, Colangelo L, Liao Y, Daviglus ML, Agulnek AN, Stamler J. Impact of minor electrocardiographic ST -segment and/or T-wave abnormalities on cardiovascular mortality during long-term follow-up. Am J Cardiol. 2003; 91: 1068–1074.[CrossRef][Medline] [Order article via Infotrieve]
18. Okin PM. QT interval prolongation and prognosis: further validation of the quantitative approach to electrocardiography. J Am Coll Cardiol. 2004; 43: 572–575.
19. Rautaharju PM, Zhou SH, Calhoun HP. Ethnic differences in North American white, black, and Hispanic men and women: effect of obesity and age. J Electrocardiol. 1994; 27 (suppl): 20–31.[CrossRef][Medline] [Order article via Infotrieve]
20. James AF, Choisy SCM, Hancox JC. Recent advances in understanding sex differences in cardiac repolarization. Prog Biophys Mol Biol. June 13, 2005. DOI: 10.1016/j.pbiomolbio.2005.05.010. Available at: http://www.sciencedirect.com/science/journal/00796107. Accessed September 6, 2005.
This article has been cited by other articles:
 |
|
 |
|
  S. Stern Electrocardiogram: Still the Cardiologist's Best Friend Circulation, May 16, 2006; 113(19): e753 - e756. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||