GENDER DEPENDENT NORMAL LIMITS OF st-t amplitudes

Peter W. Macfarlane
University of Glasgow, Department of Medical Cardiology,
Royal Infirmary, Glasgow G31 2ER, SCOTLAND

Abstract: The normal limits of the ECG have been known to be gender dependent for many years yet little diagnostic use is made of this finding.   This paper highlights such differences particularly with respect to the T wave of the ECG both in Caucasians and Chinese

INTRODUCTION

It has long been known that there are sex dependent differences in the normal limits of ECG measurements both with respect to amplitude and duration [1,2].   On the other hand such changes are often submerged in tabular form in publications to the extent that they are overlooked by the vast majority of clinicians and even those responsible for development of software for ECG interpretation by computer.

The author has had a long-standing interest on this topic and the purpose of this paper is to highlight known differences in repolarisation parameters in the normal ECG of adult males and females and to consider whether there are any racially dependent changes.

METHODS

Electrocardiograms were recorded from 1,338 apparently healthy individuals recruited from local government in the West of Scotland.   There were 731 males and 590 females.  The age range was from 17 to 65 years.   There was a predominance of women in the younger age group 18 to 29 years but otherwise males predominated in the other deciles 30 to 39, 40 to 49 and 50 and above.   ECGs were recorded in digital form and transferred to a central ECG management system for analysis using software described elsewhere [3].  

Table 1. Age and sex distribution of the Caucasian cohort.

AGE/SEX	MALE	FEMALE
18-29 yrs	266	318
30-39yrs	221	119
40-49yrs	119	73
50-59yrs	125	80
Total	731	590

Standard statistical techniques were used to obtain the mean + standard deviation of each ECG measurement together with the 96% ranges obtained by excluding 2% of measurements at each end of a set of measurements.

Separately, ECGs were recorded from 503 Chinese individuals with a normal cardiovascular system.  In this cohort, there were 255 males and 248 females with almost equal numbers of each sex in the corresponding deciles as above. These individuals were recruited from the population of Taipei in Taiwan.

Table 2. Age and sex distribution of the Chinese cohort

AGE/SEX	MALE	FEMALE
18-29yrs	56	47
30-39yrs	50	59
40-49yrs	50	50
50-59yrs	50	48
> 60yrs	49	44
Total	255	248

ECGs were recorded on to digital cassette using a Siemens Mingorec and the cassettes were forwarded to Glasgow for analysis using the same software as that for the Caucasian population.   Statistical measurements were also made in a similar way.

RESULTS

This presentation concentrates on T wave amplitudes and to a much lesser extent on ST junctional measurements.   Figure 1 shows the upper limit of normal T wave amplitude from V1 to V6 in caucasian males and females aged from 18 to 29 years. 

Fig 1. Upper normal limit of T wave amplitudes in V1 – V6 in healthy Caucasian males and females aged 18 – 29 years.

Figure 2 provides similar data for males and females aged 50 and over.

Fig 2. Upper limit of normal T wave amplitude in males and females aged 50 years and over.

Figure 3 shows the dependence of the mean T wave amplitude in V2 on gender and to a lesser extent on age.

Fig. 3. Mean T wave amplitude in V2 in healthy Caucasian male and female adults.

Figure 4 shows a similar dependence on gender in particular with respect to the ST junctional amplitude in V3 in Caucasians. 

Fig.4. The upper limit of normal amplitude of the ST junction in V3 in healthy male and female Caucasians.

Figure 5 exhibits gender differences in the lower limits of normal T wave amplitude in middle aged Caucasian men and women. 

Fig.5. Lower limit of normal T amplitude V2-V6  for healthy Caucasian males and females aged 40-49 years.

Gender differences in Chinese individuals are examined in Figure 6 where the maximum T amplitude in the precordial leads is shown.

Fig 6. Upper limit of normal T amplitude V1-V6 for healthy Chines males and females aged 40-49 years.

Fig. 7 shows the maximum T wave amplitude in V1-V6 in middle aged Caucasian and Chinese males while Fig 8 shows the mean values of T in V3–V6 in Caucasian and Chinese females.

Fig 7. A comparison of the upper limit of normal T in healthy Chinese and Caucasian males aged 40-49 years.

Fig 8. A comparison of mean T wave amplitudes in V3-V6 in healthy Chinese and Caucasian females aged 40-49 years.    

DISCUSSION

There has been a renaissance in the study of gender differences in electrocardiography, which have been known about for over 40 years [1] while some of the data on normal limits highlighted here has been tabulated previously [1,4].  The figures show clearly that there are significant differences in the upper limits of normal T wave amplitudes between males and females.  This holds true for Caucasians and Chinese.  The diagnostic relevance of this is perhaps limited to detection of abnormalities such as hyperkalemia or perhaps even myocardial infarction.  On the other hand, the lower limit of T wave abnormality is also gender dependent as in V4 (see Fig 5) but little if any use of this is made in ECG interpretation.   For example, the lower limit of normal T wave amplitude in V4 in males 50 and over is 0.138mV yet only 0.061mV in women of the same age.   In other words, the lower limit of normal T wave amplitude in males in this lead is over twice that of females.   Indeed, the same trend can be seen in males and females under 30 years of age where corresponding values are 0.248mV and 0.153mV.

The data show that gender dependence for ECG measurements holds true for Chinese and Caucasians but that there is no clear difference in T wave measurements between the races (see Figs 7,8).  This is important for the application of computer programs developed in one population and applied in another as is the case with many commercial ECG machines.  However, it should be noted that there are definite differences in maximum R and S wave amplitudes in precordial leads between Caucasians and Chinese, e.g. the mean S amplitude in V2 in 30-39 year old Caucasians is 1.93mV but only 1.68mV in Chinese of a similar age. This needs to be acknowledged in such software.

Non specific T wave abnormalities is one of the most commonly used expressions in ECG reporting.   This is undoubtedly due to the many causes of T wave changes but it could indeed also be a reflection of a lack of understanding of the significance of such changes by electrocardiographers.   For example, there is undoubtedly prognostic value in electrocardiographic T waves as shown again recently by Larsen et al [5].   In another study with ECGs analysed in this laboratory [6], it was shown that there was prognostic value in the T wave amplitude in Lead I when only a single lead ECG was recorded as part of the Tromso Health Study.  In this analysis, no limits of T wave normality were used but the conclusion drawn was that the lower the T wave amplitude, the worse the prognosis.  There is ample epidemiological evidence that the T wave has significant prognostic value and it is possible that this information could be enhanced with more gender specific criteria.

This author has already pointed out the importance of gender differences in normal ST amplitude limits [7] and has suggested [7] that criteria for impending myocardial infarction recently published [8] should be sex dependent.

There is still much information in the electrocardiogram that has never been harnessed in order to enhance the prognostic value of the ECG and to assist with therapeutic decisions. A more widespread understanding of the gender based differences in the ST-T segment might perhaps remedy this shortcoming to a certain extent.

REFERENCES

[1] E. Simonson.   Differentiation between normal and abnormal in electrocardiography.  St. Louis, The C.V.Mosby Co., 1961.

[2] P.W. Macfarlane. T.D.V.Lawrie. Comprehensive electrocardiology.  Oxford, Pergamon Press, 1989.

[3] P.W. Macfarlane, B.Devine, S.Latif, S.McLaughlin, D.B.Shoat, M.P.Watts. “Methodology of ECG interpretation in the Glasgow program”. Meth Inform Med  vol 29, pp. 354-361, 1990.

[4] C.Y.Chen, B.N.Chiang. P.W.Macfarlane.  “Normal limits of the electrocardiogram in a Chinese population”.  J Electrocardiology vol 22, pp 1-15, 1989.

[5] C.T.Larsen, J.Dahlin, H.Blackburn, H.Scharling, M.Appleyard, B.Sigurd, P.Schnohr. “Prevalence and prognosis of electrocardiographic left ventricular hypertrophy, ST segment depression and negative T wave.  Eur Heart J  vol 23, pp 315-324, 2002

[6] M-L. Lochen, K Rasmussen, P.W. Macfarlane, E. Arnesen.  “Can single lead computerized electrocardiography predict myocardial infarction in young and middle-aged men? The Tromso study”.  J. Cardiovasc Risk vol 6, pp273-278, 1999.

[7] P.W.Macfarlane. “Age, sex and the ST amplitude in health and disease”.  J. Electrocardiology vol 34 (Suppl ), pp 235 – 241, 2001.

[8] Myocardial infarction redefined – a consensus document of the joint European Society of Cardiology/American Cardiology  of Cardiology Committee for the redefinition of myocardial infarction. Eur Heart J, vol 21, pp1502-1513, 2000.