Introduction

Normal ECG standards in newborns, infants, and children have been collected and published [1,2,3,4,5,6,7]. Davignon et al. [3], Rijinbeek et al. [4], and our previous study [5] provided the most extensive collection of normal values of ECG parameters. The normal ECG values in the adolescent, however, were limited, lacking in male- and female-specific ECG values and in sex-difference beginning ages. This study was designed and carried out to establish the normal ECG standards in male and female adolescents.

Materials and Methods

The study population consisted of 898 school children and adolescents, who were screened with a questionnaire, then examined and confirmed, as healthy and of normal weight, height, and appearance, by a pediatrician and a pediatric cardiologist, during the period of 2003–2005, in Taipei, Taiwan. They were divided by age and sex, into 6–9 years (N = 99), 9–13 year (N = 289), and 13–18 year (N = 510) age-groups (Table 1). A 12-lead ECG was recorded on supine position, utilizing a Fukuda Denshi FCP-4301, MS-DOS/IBM-AT, ECG machine, at a paper speed of 25 mm/s and calibration 10 mm/mV. For large amplitude complexes, half-standard (5 mm/mV) was used. Lead V3R was not taken. Analog-to-digital conversion was performed by Fukuda signal acquisition module. The analog potential was digitalized into 5-UV units at a sampling rate of 500 Hz (once every 2 ms) [8, 9]. ECG chest electrodes were placed according to standard procedure [10].

Table 1 Study population
Full size table

Morphology measurements were made from the median voltages of the identical P-QRS-T cycles representative of a normal complex selected by the above-mentioned analysis program. Amplitude measurements were made using the PR segment as reference for the baseline. The onsets and offsets of the P-QRS-T wave were determined by an analysis of the simultaneous slopes in all 12 leads from the earliest onset in any lead to the latest deflection in any lead.

Visual verification using a magnifier and appropriate lighting was systematically performed by a pediatric cardiologist on all electrocardiograms, with measurements made to the nearest 0.1 mm. In instances of computer wave-recognition error and of more than 10% difference between visual and computer measurements, the visually determined value was substituted in the data file. ECG records discarded due to noises, baseline drifting, bundle branch blocks, WPW, and ECG rhythms other than sinus rhythm were around 54 in total. The data collected were processed and analyzed, for the mean, standard deviation, 2nd to 98th percentiles, 95% confidence intervals of heart rate, P-QRS-T intervals, duration, axis, amplitude and calculated values, such as R/S amplitude ratio and ventricular activation time (VAT). The percentiles were estimated by parametric method with normal distribution. The analysis program had been validated [5]. No transformation was made. The differences of male and female values and of the values from this study and the literature were analyzed using the independent t test. P Values of < 0.05 and < 0.01 were considered as significant and highly significant, respectively.

Results

Heart Rate, P, QRS, and T Wave Interval and Duration by Age-Group and Sex

The male- and female-specific mean, standard deviation, and 2nd–98th percentiles values of heart rate, PR interval, QT interval, QTc interval, QRS duration, and RR interval in each age-group are listed in Table 2. Sex difference occurred in heart rate, QTc interval, QRS duration, and RR interval, beginning in the 9–13 years and 13–18 years age-group.

Table 2 Heart rate, P, QRS, and T interval and duration by age-group and sex
Full size table

Frontal Plane P-QRS-T Axis by Age-Group and Sex

The male- and female-specific mean, standard deviation, 2nd–98th percentiles values, and 95% confidence intervals of frontal plain P axis, QRS axis, and T axis in each age-group were obtained as listed in Table 3. A highly significant sex difference of T axis values occurring in the 9–13 and 13–18 years age-groups was noted. No sex difference occurred in P axis and QRS axis.

Table 3 Frontal plane P-QRS-T axis (degrees) by age-group and sex
Full size table

P, Q, R, S, and T Wave Amplitude by Age-Group and Sex

The male- and female-specific mean, standard deviation, 2nd–98th percentiles, and 95% confidence intervals of frontal plain P axis, QRS axis, and T axis in each age-group were obtained as listed in Table 4. Significant sex differences which began to appear in 6–9 years age-group were R V1, S III, and S aVF. Sex differences which began to appear in the 9–13 years age-group were Q III, Q aVL, Q aVF, Q V5, Q V6, R aVR, R V2, R V4, R V5, R V6, S II, S V4, S V5, S V6, T II, T III, T aVR, T aVL,T aVF, T V4, T V5, and T V6. Sex differences which began in 13–18 years age-group were Q II. Q V4, S I, S aVL, S V1, S V2, T V1, and T V2.

Table 4 P, Q, R, S, and T wave amplitude (mV) by age-group and sex
Full size table

Calculated Values on R/S Wave Amplitude Ratio and Ventricular Activation Time (VAT) by Age-Group and Sex

The male- and female-specific mean, standard deviation, 2nd–98th percentiles, and 95% confidence intervals of R/S ratio and VAT values in each age-group were obtained as listed in Table 5. Sex differences that first occurred in 9–13 years age-group were Sum of R V3 and S V3, Sum of R V6, and S V1 and VAT V1. Sex differences that first occurred in the 13–18 years age-group were R/S V1, Sum of R V6 and S V2, VAT II, VAT aVF, VAT V2, VAT V4, VAT V5, and VAT V6.

Table 5 Calculated values on R/S amplitude ratio and ventricular activation time by age-group and sex
Full size table

Discussion

Of the 69 ECG parameters we studied, 49 (71.0%) parameters showed a significant male and female difference: 3 (6.1%) parameters beginning in the 6–9 years age-group were R V1, S III, and S aVF; 30 (61.2%) parameters beginning in the 9–13 years age-group were heart rate, QRS duration, RR interval, T axis, Q III, Q aVL, Q aVF, Q V5, Q V6, R aVR, R V2, R V4, R V5, R V6, S II, S V4, S V5, S V6, T II, T III, T aVR, T aVL, T aVF, T V4, T V5, T V6, Sum of R V3 and S V3, Sum of R V6 and S V1 and VAT V1; and 16 parameters (23.2%) beginning in the 13–18 years age-group were QTc interval, Q II, Q V4, S I, S aVL, S V1, S V2, T V1, T V2, R/S V1, S V2, VAT II, VAT aVF, VAT V2, VAT V4, VAT V5, and VAT V6. Sex difference did not occur in 20 (29.0%) parameters: PR interval, QT interval, P axis, and QRS axis; P wave amplitude in II and T wave in ; R/S ratio in I, II, III, aVR, aVL, aVF, V2, V3, V4, V5, V6; and VAT in I, III, and aVR.

ECG gender difference once occurred, it continued to occur along with the age growth, with only two exceptions in Q III and Q aVF wave amplitudes, which occurred in 6–9 age-group, as an isolated or non-sequential sex difference.

Normal ECG values established in this study were compared with those published in the literature [3, 4, 6]. A total of 14 male- and female-specific normal ECG values were collected from Davignon et al. [3] in a 12–15 years age-group, and Rijinbeek et al. [4] and Semizel et al. [6] in a 12–16 years age-group, as listed in Table 6. The male-specific R V1 values (#1) of this study were compared with those of Davignon et al. [3] which provided mean, standard deviation, and ranges, yielding no statistical difference (p = 0.867). Female-specific R V1 values (#2) showed also no difference (p = 0.742). R V6 values (#3 and #4) were also compared showing similar results (p = 0.661 and p = 0.719), indirectly confirming that the normal ECG values collected in this study were accurate and credible.

Table 6 A comparison of 14 normal sex-specific ECG values collected in this study with those published in the literature
Full size table

Typical male ECG could be differentiated from typical female ECG in five characteristic features: Higher takeoff of the ST segment, shorter period between J point and onset of T wave, steeper slope of ST segment, steeper ascent of T wave, and higher T wave amplitude [11, 12]. Surawicz et al. [12], Bidoggia et al. [13], and Vicente et al. [14] reported that male sex hormones or autonomic innervations played an important role in modulating the cardiac repolarization. Sex ECG amplitude difference was more prominent in the precordial leads than in extremity leads [15].

Adolescence is a transitional period from puberty to adulthood. Sex difference in the adolescent conveys the anatomical and physiological consequences of puberty [12]. Before puberty, sex does not appear to influence the ECG significantly [16, 17]. Girls begin puberty usually around ages 10–11 years, and boys around 11–12 years, but puberty may start 1–2 or more years earlier due to social, dietary, environmental, and many other factors [17,18,19,20].

Puberty begins, when sex hormone levels rise and secondary sex characteristics, such as pubic hair, enlarged breasts, and facial hair appear. The most obvious difference between male and female includes muscle mass, height, and hair distribution. Men have a heavier heart weight and larger volume than women [21, 22]. Females have a significantly higher heart rate, shorter conduction times, decreased P, Q, and T amplitudes, and longer repolarization time, and decreased P, Q, and T amplitudes [15, 21, 23, 24]. QTc interval begins to decrease in men at the time of puberty when testosterone levels increase affecting effects on calcium current, and begins to increase as men age and testosterone levels decrease [14, 15, 24,25,26].

Study Limitations

In this study, lead V3R ECG was not taken, and the number of subjects in 6–9 years age-group was only 99 (male 57, female 42), not over 100, as recommended by Davignon et al. [3].