Abstract
In this brief report, we provide normal reference ranges for PR duration [unadjusted and heart rate adjusted] and P-wave indices [duration, amplitude and terminal force in V1] in individuals free of cardiovascular disease and its risk factors. We used automatically processed digital ECG data from 1252 US participants [mean age 59 (± 10) years, 738 women, 588 whites, 207 African-Americans, 217 Hispanics, 240 Chinese] from the Multi-Ethnic Study of Atherosclerosis [MESA]. In multivariable adjusted linear regression models with PR and each P-wave variable as a separate outcome, significant age, sex and race differences in these markers were observed. Subsequently, we report reference ranges for abnormal [2nd and 98th percentiles], borderline abnormal [5th and 95th percentiles] and mean [SD] values of PR and P-wave indices stratified by age [middle age (45–64 years) and seniors (65–84 years)], sex [men and women] and race [whites, African Americans, Hispanics and Chinese].
Keywords: P-wave indices, PR interval, MESA
INTRODUCTION
The idea that PR duration and P-wave morphology carry important prognostic information that could be utilized for prediction of cardiovascular disease [CVD] outcomes is gaining wide interest. Data from population studies showed that PR duration and P-wave indices [duration, amplitude and terminal force in V1] are strong predictors of atrial fibrillation, stroke and all-cause mortality (1–4). These findings have triggered several studies looking at associations and impact of different CVD risk factors on PR and P-wave indices (5–7), as well as examining their genetic background (8–10). These associations and determinants, however, would be better interpreted in the context of normal values expected in individuals free of CVD and its risk factors. Therefore, we used data from the Multi-Ethnic Study of Atherosclerosis [MESA], a US community-based cohort study, to establish age-sex-race specific reference ranges for PR (unadjusted and adjusted for heart rate; PRa) and P-wave indices in individuals free of CVD and its risk factors.
METHODS
The description of the MESA study is provided elsewhere (10). Briefly, between July 2000 and August 2002, 6,814 men and women aged 45 to 84 years old and free of clinically apparent CVD were recruited from six US communities: Baltimore City and Baltimore County, Maryland; Chicago, Illinois; Forsyth County, North Carolina; Los Angeles County, California; Northern Manhattan and the Bronx, New York; and St. Paul, Minnesota. For the purpose of this analysis, we excluded individuals who were smokers or with diabetes, hypertension, dyslipidemia or obesity. We also excluded those with major ECG abnormalities as defined by Minnesota ECG classification (12). After all exclusions, 1252 individuals free of CVD and CVD common risk factors remained and were included in this analysis. All participants gave written informed consent and the study was approved by the Institutional Review Boards at the 6 Field Centers.
Electrocardiography
Standard 12-lead ECGs were digitally acquired using a GE MAC 1200 electrocardiograph [GE, Milwaukee, WI] at 10 mm/mV calibration and speed of 25 mm/s. ECG reading was performed centrally at the Epidemiological Cardiology Research Center (EPICARE), Wake Forest School of Medicine, Winston Salem, NC. All ECGs were initially inspected visually for technical errors and inadequate quality. P-wave durations and amplitudes needed to calculate p-wave indices were automatically measured with the GE Marquette 12-SL program 2001 version [GE Marquette, Milwaukee, WI]. A global single measure of PR interval was calculated from the beginning of the P-wave to the beginning of the QRS. Heart rate corrected PR was calculated using Soliman-Rautaharju formula (13). P-wave duration and P-wave amplitude used in this report were the maximum value of these measures across the 12 leads. Finally, P-wave terminal force was defined as the duration in milliseconds of the terminal part [negative] of the P wave in lead V1 multiplied by its depth in microvolts.
Statistical analysis
Multivariable linear regression analysis was used to examine differences (Beta coefficient and 95% confidence interval) in PR and each of the P-wave variables across categories of age [middle age (45–64 years) vs. seniors (65 years or older)], sex [male vs. female] and race [whites vs. African Americans, Hispanics and Chinese, separately]. Reference ranges for abnormal values [2nd and 98th percentiles], borderline abnormal [5th and 95th percentiles] and mean [SD] across these categories were calculated.
RESULTS
This analysis included 588 whites [60.7% women], 207 African Americans [54.1% women], 217 Hispanics [57.6% women] and 240 Chinese [60.0% women]. Characteristics of the study participants stratified by race are detailed in Table 1.
Table 1.
Characteristics of the study participants
| All participants | Whites | African Americans | Hispanics | Chinese | |
|---|---|---|---|---|---|
| (n=1252) | (n=588) | (n=207) | (n=217) | (n=240) | |
| Age, years | 59.2 ± 10.2 | 60.2 ± 10.1 | 59.2 ± 10.2 | 57.7 ± 10.2 | 57.9 ± 10.1 |
| 45–64, n (%) | 842 (67.2) | 375 (63.8) | 137 (66.2) | 155 (71.4) | 175 (72.9) |
| 65–84, n (%) | 410 (32.8) | 213 (36.2) | 70 (33.8) | 62 (28.6) | 65 (27.1) |
| Female Sex, n (%) | 738 (59.0) | 357 (60.7) | 112 (54.1) | 125 (57.6) | 144 (60.0) |
| Body Mass Index, kg/m2 | 24.6 ± 3.0 | 24.6 ± 2.8 | 25.6 ± 2.9 | 25.9 ± 2.7 | 22.7 ± 2.7 |
| Fasting Glucose, mg/dL | 86.0 ± 9.1 | 84.5 ± 8.8 | 86.1 ± 8.8 | 86.9 ± 8.9 | 88.9 ± 9.6 |
| Diastolic Blood Pressure, mmHg | 67.7 ± 8.8 | 66.3 ± 8.8 | 71.2 ± 8.1 | 67.2 ± 8.5 | 68.5 ± 8.5 |
| Systolic Blood Pressure, mmHg | 112.4 ± 13.3 | 111.5 ± 13.5 | 115.7 ± 12.9 | 112.0 ± 12.8 | 112.2 ± 13.4 |
| HDL Cholesterol, mg/dL | 57.9 ± 13.8 | 59.7 ± 14.9 | 60.0 ± 15.1 | 53.8 ± 10.7 | 55.5 ± 11.0 |
| LDL Cholesterol, mg/dL | 112.7 ± 23.6 | 113.0 ± 23.7 | 110.8 ± 25.1 | 115.2 ± 22.8 | 111.3 ± 22.6 |
| Total Cholesterol, mg/dL | 188.9 ± 25.0 | 190.8 ± 24.9 | 186.0 ± 26.9 | 189.0 ± 24.4 | 186.8 ± 23.9 |
| Triglycerides, mg/dL | 91.5 ± 36.5 | 90.3 ± 35.7 | 75.8 ± 30.0 | 100.1 ± 36.4 | 100.4 ± 38.6 |
In a multivariable linear regression model with PR and each of P-wave variables as separate outcomes and age, sex and race as covariates, significant differences in the distribution of PR and P-wave indices were observed across categories of these demographics. As shown in Table 2, participants younger than 65 years [middle age] compared to those 65 years or older [seniors] had significantly shorter PR, PRa and P-wave durations [p-value for each <0.01], lower P-wave amplitude [p=0.047] and smaller P-wave terminal force in V1 [p-value <0.01]. Men, compared to women, had longer PR, PRa and P-wave durations [p-value for each <0.01], but lower P-wave amplitude [p=0.04] and no significant difference in P-wave terminal force in V1. The most significant racial differences were between African Americans and whites. Compared to whites, African Americans had longer PR, PRa and P-wave durations, higher P-wave amplitude, and larger P-wave terminal force in V1 [p-value for each <0.01]. Given these age, sex and race differences, we provide reference ranges of PR, PRa and P-wave indices stratified by these demographics in Tables 3 and 4.
Table 2.
Multivariable adjusted age, sex and race differences in PR and P-wave indices*
| Age (middle age versus senior)** | Sex (men vs. women) | Race (Reference group = white) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| African American | Hispanic | Chinese | ||||||||
| Difference (95%CI) | p | Difference (95%CI) | p | Difference (95%CI) | p | Difference (95%CI) | p | Difference (95%CI) | p | |
| PR duration (ms) | −5.1 (−7.8, −2.4) | 0.001 | 9.6 (7.0, 12.2) | <0.001 | 5.8 (2.1, 9.4) | 0.002 | −4.3 (−7.9, −0.7) | 0.02 | −4.7 (−8.2, −1.2) | 0.01 |
| PRa duration (ms) | −4.3 (−7.0, −1.6) | 0.002 | 8.6 (6.0, 11.2) | <0.001 | 6.1 (2.5, 9.7) | 0.001 | −3.6 (−7.2, −0.1) | 0.05 | −3.7 (−7.2, −0.2) | 0.04 |
| P-wave duration (ms) | −5.8 (−7.3, −4.4) | <0.001 | 6.1 (4.7, 7.4) | <0.001 | 3.8 (1.9, 5.7) | <0.001 | −0.4 (−2.3, 1.4) | 0.65 | −2.4 (−4.2, −0.6) | 0.01 |
| P-wave amplitude (μV) | −4.1 (−8.1, −0.1) | 0.05 | −4.0 (−7.9, −0.2) | 0.04 | 11.7 (6.3, 17.1) | <0.001 | −1.5 (−6.8, 3.7) | 0.57 | 4.1 (−1.0, 9.2) | 0.11 |
| PTF-V1 (μV.ms) | −682.4 (−871.7, −493.1) | <0.001 | 99.9 (−80.3, 280.1) | 0.28 | 580.9 (328.4, 833.4) | <0.001 | −23.7 (−272.1, 224.6) | 0.85 | 327.4 (87.8, 567.0) | 0.007 |
Differences calculated using multivariable linear regression model with each PR/P-wave variable as the outcome in separate models and age, sex, race as covariates in all study population
Middle age= 45–64; Senior 65 and older
PRa= Heart rate adjusted PR duration
PTF-V1= absolute value of P-wave terminal force in V1
Table 3.
Age, sex and race specific reference ranges for PR interval
| Age group | Whites | African Americans | Hispanics | Chinese | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Men (n=231) | Women (n=357) | Men (n=95) | Women (n=112) | Men (n=92) | Women (n=125) | Men (n=96) | Women (n=144) | |||
| PR duration (ms) | Mean (SD) | Middle | 167 (25) | 155 (23) | 174 (24) | 165 (22) | 159 (19) | 155 (21) | 162 (19) | 154 (18) |
| Senior | 176 (32) | 162 (22) | 178 (31) | 160 (19) | 162 (17) | 163 (18) | 160 (24) | 158 (19) | ||
| 5%→95% | Middle | 134→212 | 120→194 | 144→228 | 134→208 | 131→197 | 120→196 | 124→198 | 124→188 | |
| Senior | 132→234 | 132→202 | 130→248 | 136→186 | 138→190 | 138→200 | 130→212 | 128→198 | ||
| 2%→98% | Middle | 132→244 | 114→214 | 140→240 | 122→210 | 128→206 | 114→212 | 120→204 | 120→192 | |
| Senior | 128→246 | 124→212 | 128→260 | 134→218 | 130→196 | 136→204 | 128→220 | 124→214 | ||
| PRa duration (ms) | Mean (SD) | Middle | 163 (25) | 153 (22) | 170 (24) | 162 (22) | 157 (19) | 153 (21) | 160 (19) | 152 (18) |
| Senior | 171 (33) | 159 (22) | 175 (32) | 157 (18) | 159 (17) | 161 (18) | 157 (24) | 156 (19) | ||
| 5%→95% | Middle | 130→207 | 118→189 | 137→224 | 131→205 | 130→193 | 119→197 | 124→194 | 121→188 | |
| Senior | 128→225 | 132→198 | 128→245 | 134→181 | 135→186 | 139→197 | 126→207 | 124→196 | ||
| 2%→98% | Middle | 130→241 | 113→213 | 136→238 | 120→206 | 126→200 | 112→208 | 118→199 | 118→189 | |
| Senior | 120→246 | 120→209 | 125→252 | 133→210 | 125→190 | 138→204 | 124→220 | 122→210 | ||
Middle age= 45–64; Senior 65 and older
PRa= Heart rate adjusted PR duration
Table 4.
Age, sex and race specific reference ranges for P-wave indices
| Age group | Whites | African Americans | Hispanics | Chinese | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Men (n=231) | Women (n=357) | Men (n=95) | Women (n=112) | Men (n=92) | Women (n=125) | Men (n=96) | Women (n=144) | |||
| P-wave duration (ms) | Mean (SD) | Middle | 104 (14) | 98 (13) | 108 (13) | 103 (10) | 105 (10) | 97 (12) | 104 (10) | 94 (10) |
| Senior | 111 (13) | 105 (13) | 112 (14) | 106 (9) | 108 (12) | 105 (10) | 106 (11) | 104 (10) | ||
| 5%→95% | Middle | 78→124 | 80→120 | 84→130 | 86→120 | 88→122 | 78→114 | 88→116 | 76→110 | |
| Senior | 90→130 | 83→125 | 96→144 | 84→122 | 88→128 | 90→122 | 92→124 | 86→128 | ||
| 2%→98% | Middle | 72→144 | 76→126 | 84→130 | 84→124 | 86→122 | 72→116 | 86→136 | 72→112 | |
| Senior | 82→136 | 78→138 | 72→146 | 84→122 | 84→134 | 88→126 | 78→126 | 82→128 | ||
| P-wave amplitude (μV) | Mean (SD) | Middle | 111 (29) | 119 (36) | 127 (38) | 127 (37) | 115 (28) | 111 (27) | 122 (33) | 119 (35) |
| Senior | 114 (31) | 123 (38) | 124 (43) | 138 (36) | 120 (34) | 121 (26) | 122 (37) | 122 (33) | ||
| 5%→95% | Middle | 63→161 | 73→185 | 73→205 | 78→205 | 81→175 | 78→161 | 73→180 | 78→185 | |
| Senior | 68→170 | 78→190 | 87→229 | 92→239 | 83→185 | 78→161 | 73→205 | 73→205 | ||
| 2%→98% | Middle | 53→170 | 63→200 | 63→219 | 73→214 | 78→180 | 68→175 | 58→195 | 63→229 | |
| Senior | 68→195 | 63→200 | 78→268 | 78→239 | 58→224 | 68→170 | 63→205 | 73→214 | ||
| PTF-V1 (μV.ms) | Mean (SD) | Middle | 1507 (1519) | 1350 (1510) | 1954 (1900) | 2049 (1424) | 1680 (1513) | 1238 (1198) | 1769 (1489) | 1538 (1487) |
| Senior | 1932 (1810) | 2143 (1849) | 2766 (1604) | 2431 (1964) | 1804 (1920) | 2137 (1714) | 2786 (1725) | 2557 (1131) | ||
| 5%→95% | Middle | 0→4410 | 0→4410 | 0→5976 | 0→4095 | 0→4551 | 0→3120 | 0→4347 | 0→4399 | |
| Senior | 0→4788 | 0→5905 | 0→5976 | 0→5772 | 0→5694 | 0→5229 | 0→5610 | 0→4399 | ||
| 2%→98% | Middle | 0→5146 | 0→5168 | 0→6612 | 0→4836 | 0→4891 | 0→3969 | 0→4420 | 0→5810 | |
| Senior | 0→6351 | 0→7812 | 0→7081 | 0→6059 | 0→7387 | 0→5304 | 0→7084 | 0→4453 | ||
Middle age= 45–64; Senior 65 and older
PTF-V1= absolute value of P-wave terminal force in V1
DISCUSSION
In a large multiracial population of adults without evidence of CVD or its risk factors, we report age-, sex- and race-specific normal reference ranges for PR and P-wave indices. With the increasing utilization of these easy-to-obtain ECG markers, there is need for reference values to distinguish normal from abnormal limits which would subsequently help in better interpretation of risk factor associations and impact of different diseases on these markers. The observed age, sex and race differences in the distribution of PR and P-wave indices in our study call for caution when setting a cut-point for abnormal values since what could be normal in one group may be actually abnormal in another. Nevertheless, given the small magnitude of these differences despite being statistically significant, it is not clear at this stage if these differences are of clinical importance or mirror potential differences in the prognostic significance.
The classical cut-points defining PR or P-wave abnormalities such as 200 ms for prolonged PR or 120 ms for prolonged P wave duration or 4000 μV.ms for large p-wave terminal force in V1 were derived with no consideration to age, sex, or race differences. They were also derived mostly using non-digital ECG data that were processed manually. As could be observed from our results, these classical cut-points are not similar to those derived from digital automatically processed ECG in multi-racial setting. With the availability of automated ECG interpretation systems, it should be feasible to consider age, sex and race differences in the definition of P-wave and PR abnormalities.
To our knowledge, this is the first report that establishes normal reference ranges for PR and P-wave indices utilizing digital ECG data collected from a large sample of individuals from 4 different ethnic/race groups. A previous report from the Framingham Heart Study provided reference ranges for PR, P-wave duration and P-wave dispersion (14). However, this was only in 295 whites, and more than half of the participants did not have all of the 12 ECG leads because of background noise that prohibited accurate measurement of P-wave indices. More importantly, P-wave indices in the Framingham study were measured by an operator and the process was not fully automated.
In our report we focused on the most commonly used P-wave indices as well as the recently developed heart rate corrected PR. We ignored P-wave area because of its rare use and inconsistency of the methods of its calculation using different automated software (15). We also ignored P-wave dispersion because of the conflicting reports about its prognostic significance (16, 17) and lack of consensus on what really dispersion in ECG represents (17, 18). There is no doubt that differences in the p-wave duration across leads exist, which is expressed as isolelectric interval in some leads due to different orientation. This isoelectric interval is not necessary due to heterogeneous atrial conduction (claimed to be reflected by P-wave dispersion) but could be simply explained by different orientation of the ECG leads. Without solid grounds linking difference in the duration of P-wave in the 12 ECG leads (i.e. P-wave dispersion) to specific atrial conduction abnormalities beyond the expected differences due to lead orientation, we believe it is not yet the time to use. Similarly, we used the maximum value of P-wave amplitude in the 12 leads instead of using individual values in each lead. To reasonably compare different populations, using a more repeatable measure such as maximum P-wave in the 12 leads that is less prone to random measurement error is needed. Nevertheless, given the known variations in the P-wave amplitude in different ECG leads, looking at individual leads may be warranted and useful for other purposes. Per lead P-wave amplitude data are available upon request from the authors.
The reference ranges we provide in this report are based on digital ECG data that were processed automatically. If PR and P-wave indices are to be measured manually by an operator using magnifying loupe, it would be appropriate to approximate the values of durations and amplitudes to the nearest 5 ms or 25 μv, respectively, which are probably the smallest measurements that could be reasonably made using visual non-automated measurement. On a related point, we did not compare our reference values from GE-Marquette to other automated software. However, it is believed that these measurements should be highly repeatable especially if high quality ECGs such as those in our study, are used.
Limitations
The reported reference ranges are for individuals aged 45 years and older, and therefore, may not be applicable to younger individuals. Also, caution is needed when comparing the reference ranges in this report with those previously published that were mainly based on lead II or using manual measurements.
We did not validate the automatic measurements of P-wave indices and PR which we used in this report. However, a statement of validation and accuracy of the Marquette 12-SL has been published elsewhere (19). Further, the GE Marquette 12-SL program is an FDA approved software for interoperation of ECG, and subsequently the results that come out of it should be reasonably valid. Therefore, we feel that validating the measurements ourselves would be beyond the scope of this paper.
Conclusions
In individuals free of CVD and its risk factors, there are differences by age, sex and race in the distribution of PR and P-wave indices. In this brief report, we provide age, sex and race specific references ranges for these prognostically important ECG markers using data from the MESA study, one of the major cohort studies in the US.
Acknowledgments
The MESA study was supported by contracts N01-HC-95159 through N01-HC-95169 from the National Heart, Lung and Blood Institute.
The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org.
Footnotes
Competing interests: None
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Soliman EZ, Prineas RJ, Case LD, Zhang ZM, Goff DC., Jr Ethnic distribution of electrocardiographic predictors of atrial fibrillation and its impact on understanding the ethnic distribution of ischemic stroke in the Atherosclerosis Risk in Communities Study (ARIC) Stroke. 2009;40:1204. doi: 10.1161/STROKEAHA.108.534735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Schnabel RB, Sullivan LM, Levy D, et al. Development of a risk score for atrial fibrillation (Framingham Heart Study): a community-based cohort study. Lancet. 2009;373:739. doi: 10.1016/S0140-6736(09)60443-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cheng S, Keyes MJ, Larson MG, et al. Long-term outcomes in individuals with prolonged PR interval or first-degree atrioventricular block. JAMA. 2009;301:2571. doi: 10.1001/jama.2009.888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Magnani JW, Johnson VM, Sullivan LM, et al. P wave duration and risk of longitudinal atrial fibrillation in persons ≥ 60 years old (from the Framingham Heart Study) Am J Cardiol. 2011;107(6):917–921. doi: 10.1016/j.amjcard.2010.10.075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Alonso A, Soliman EZ, Chen LY, Bluemke DA, Heckbert SR. Association of blood pressure and aortic distensibility with P wave indices and PR interval: the Multi-Ethnic Study of Atherosclerosis (MESA) J Electrocardiol. 2013 doi: 10.1016/j.jelectrocard.2013.01.009. (In press) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Babcock MJ, Soliman EZ, Ding J, Kronmal AR, Goff DC., Jr Pericardial fat and atrial conduction abnormalities in the Multiethnic Study of Atherosclerosis (MESA) Obesity (Silver Spring) 2011;19(1):179–84. doi: 10.1038/oby.2010.121. [DOI] [PubMed] [Google Scholar]
- 7.Magnani JW, Lopez FL, Soliman EZ, Maclehose RF, Crow RS, Alonso A. P wave indices, obesity, and the metabolic syndrome: the atherosclerosis risk in communities study. Obesity (Silver Spring) 2012;20(3):666–672. doi: 10.1038/oby.2011.53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Butler AM, Yin X, Evans DS, et al. Novel loci associated with PR interval in a genome-wide association study of 10 African American cohorts. Circ Cardiovasc Genet. 2012;5(6):639–46. doi: 10.1161/CIRCGENETICS.112.963991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Smith JG, Magnani JW, Palmer C, et al. Genome-wide association studies of the PR interval in African Americans. PLoS Genet. 2011;7(2):e1001304. doi: 10.1371/journal.pgen.1001304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pfeufer A, van Noord C, Marciante KD, et al. Genome-wide association study of PR interval. Nat Genet. 2010 Feb;42(2):153–9. doi: 10.1038/ng.517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bild DE, Bluemke DA, Burke GL, et al. Multi-ethnic study of atherosclerosis: Objectives and design. AmJ Epidemiol. 2002;156:871–881. doi: 10.1093/aje/kwf113. [DOI] [PubMed] [Google Scholar]
- 12.Prineas RJ, Crow R, Blackburn HW. The Minnesota Code manual of electrocardiographic findings: Standards and procedures for measurement and classification. Boston, Mass: J. Wright; 1982. pp. 44–46. [Google Scholar]
- 13.Soliman EZ, Rautaharju PM. Heart rate adjustment of PR interval in middle-aged and older adults. J Electrocardiol. 2012;45(1):66–9. doi: 10.1016/j.jelectrocard.2011.06.003. [DOI] [PubMed] [Google Scholar]
- 14.Magnani JW, Johnson VM, Sullivan LM, et al. P-wave indices: derivation of reference values from the Framingham Heart Study. Ann Noninvasive Electrocardiol. 2010;15(4):344–52. doi: 10.1111/j.1542-474X.2010.00390.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Soliman EZ. P-wave area. Response to Letter by Dewhurst and Adams. Stroke. 2011;42:e20. [Google Scholar]
- 16.Magnani JW, Johnson VM, Sullivan LM, et al. P wave duration and risk of longitudinal atrial fibrillation in persons ≥ 60 years old (from the Framingham Heart Study) Am J Cardiol. 2011;107:917–921. doi: 10.1016/j.amjcard.2010.10.075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Dilaveris PE, Gialafos EJ, Sideris SK, Theopistou AM, Andrikopoulos GK, Kyriakidis M, Gialafos JE, Toutouzas PK. Simple electrocardiographic markers for the prediction of paroxysmal idiopathic atrial fibrillation. Am Heart J. 1998;135:733–738. doi: 10.1016/s0002-8703(98)70030-4. [DOI] [PubMed] [Google Scholar]
- 18.Rautaharju PM. Dispersion in electrocardiography has become epidemic. Am J Cardiol. 2000;86:370. doi: 10.1016/s0002-9149(00)01154-1. [DOI] [PubMed] [Google Scholar]
- 19.GE Healthcare. Marquette™ 12SL™ ECG Analysis Program: Statement of validation and accuracy. Feb 1, 2007. 416791-003, Revision B. [Google Scholar]