Skip main navigation

Gender Differences in Left Ventricular Growth

Originally published 1995;26:979–983


    Abstract Because the number of human cardiac myocytes is determined in infancy, subsequent increases in left ventricular (LV) muscle mass reflect cellular enlargement (hypertrophy). To determine whether the greater LV mass in adult men than in women reflects sex differences that are present throughout development or disproportionate LV growth during puberty in men, we compared echocardiographic LV mass in 333 female and 278 male normal-weight, normotensive subjects from 4 months to 70 years of age. Only a small sex difference in LV mass (mean=6%) existed before age 12 years, whereas in all older-age strata LV mass in men was 25% to 38% greater than that in women (P<.02 to P<.0001). The divergence in LV mass between male and female adolescents closely paralleled differences in height and weight and was due to proportional increases in LV chamber dimension and wall thickness in males (with no sex difference in relative wall thickness). LV mass grew less rapidly from infancy through childhood than did body size, assessed by body weight or height2.7, yielding a reduction of LV mass/body size ratios up to puberty, which was followed by gradual increases during adulthood. Indexation of LV mass by body weight or height2.7 but not by body surface area or height markedly reduced the sex differences in LV mass/body size ratios from puberty through the seventh decade of life. Thus, before puberty, LV mass is only modestly higher in boys than in girls. Most of the sex difference in adult LV mass follows differences in body size and is due to a greater “physiological” LV hypertrophy in men than in women.

    The number of human cardiac myocytes is considered to be determined within the first year after birth, when mitotic activity of normal cardiocytes appears to cease.1 Subsequent increases in heart size are primarily determined by changes in the size of myocytes (hypertrophy).2 Among normal adults, left ventricular (LV) mass is larger in men than in women,345 which might be due to (1) a sex difference in myocardial mass (and hence the number of myocytes) that was programmed in early life, (2) differential cardiac growth with relative hypertrophy of myocytes during development in males, or (3) both. Under the first scenario, a difference in LV mass between males and females would be evident in early childhood and would be maintained during body growth. Under the second scenario, LV mass would be similar in female and male children and would diverge during adolescent maturation, at a time long after the numbers of cardiocytes had been determined. In the third situation, LV mass would be higher in boys than in girls in early childhood, but this difference would increase with body growth.

    To test these possibilities, we compared LV mass in normal-weight female and male members of a large, normotensive healthy population.


    We analyzed echocardiographic and body size measurements in 611 normotensive, normal-weight subjects, 4 months to 70 years of age (333 males and 278 females), including 100 adults from Cornell Medical Center in New York, 87 adults from the Federico II University Hospital in Naples, Italy, and 424 children to young adults from Children’s Hospital Medical Center in Cincinnati. Detailed characteristics of this nonobese cohort have been reported previously.6

    LV mass was calculated with the use of American Society of Echocardiography measurements in an anatomically validated formula.7 LV mass values were also normalized by various measures of body size, including height, height2.7, body weight, and body surface area.

    Statistical Analysis

    Forward stepwise multiple regression analysis was used to determine whether age was related to LV mass independently of body size and sex both in children to adolescents (birth to 17 years) and adults (>17 years). The population sample was divided into deciles according to rank order of age (61 individuals in nine deciles and 62 in the highest one). When more than one person with the same age overlapped two adjacent deciles of population, the individuals were randomly allocated in the two deciles to maintain the total number of 61 subjects for each decile. Age characteristics and sex distribution of each decile are reported in Table 1 and Fig 1. Average LV mass, as absolute values or normalized for body size, as well as LV chamber diameter and wall thickness were compared between males and females in all deciles with the use of ANOVA after adjustment for age differences within deciles by ANCOVA. ANCOVA adjusting for age was also used to compare the gender effect on LV mass in the entire subgroup of children younger than age 11 years. Crude mean values are represented in the figures, but the probability values were obtained after adjustment for age. To obtain information about the rate of growth-related changes in LV mass and body size, the slopes of lines relating LV mass, body weight, height, height2.7, and body surface area to age were examined in children to adolescents (up to 17 years old) after standardization of variables to a mean of 1 and a standard deviation corresponding to the coefficient of variability of the original variables in the entire cohort. With this approach, the slopes of the lines relating LV mass and the different measures of body size to age could be compared with the use of F statistics based on the between-slopes sum of squares.


    Relation of LV Mass to Age

    Age was related to LV mass both in children to adolescents (β=0.21, P<.002) and adults (β=0.31, P<.0001) independent of height2.7 or weight and sex. Preadolescent children (from birth to 12 years) exhibited a close relation between LV mass and age (r=.77, P<.0001, SEE=13 g). The slope of this regression was similar in the 123 boys and in the 116 girls (5.7 and 4.9 g/y, respectively, P=.19). However, if boys and girls between 12 and 17 years were also considered in this relation (ie, as a subgroup undergoing peripubertal and postpubertal growth), the LV growth rate became significantly more rapid in boys (7.3 g/y, n=196) than in girls (5.4 g/y, n=187, P<.0001). In adulthood, the relation between age and LV mass was weaker (r=.31, P<.0001, SEE=35 g), and the slopes were not different between men (n=137) and women (n=91) (0.79 and 0.53 g/y, respectively, P=.35).

    Age-Related Sex Differences in LV Mass

    Table 2 shows that there was no statistical difference in left ventricular mass values between boys and girls in individual age strata through 12 years of age. Because puberty usually occurs between the ages of 11 and 13 years, LV mass was also examined in this age range and was found to be 78.8 g in 25 boys (38.5±8.4 kg body wt and 1.49±0.13 m height) and 75.6 g in 27 girls (39.2±8.6 kg body wt and 1.48±0.12 m height) (P>.5). The entire group of children up to age 11 years (prepubertal age) also was examined (ANOVA with age as covariate): LV mass did not differ significantly between boys (n=107, 45±18 g) and girls (n=101, 43±17 g, P=.065).

    Starting with the group between 12 and 14 years of age (5th decile), encompassing the immediate postpubertal period for most individuals, LV mass values diverged strikingly between the sexes because ventricular mass increased much more in adolescent boys than in girls (Fig 1). Fig 2 shows that this difference was due to both higher LV chamber dimension and wall thickness in males than in females: Relative wall thickness was indeed similar in men and women in all deciles of age. Although the sex difference in LV mass before puberty was not statistically significant, in each age group boys had 5% to 8% higher LV mass than girls. In contrast, the sex difference in LV mass between adult men and women in different age strata was 25% to 38%.

    The pattern of increase in LV mass with age in females and males was similar to that in body height or weight through puberty and adolescence (Fig 3). The sex difference in LV mass became stable, paralleling differences in body height, but was earlier than differences in body weight (Figs 1 and 3).

    In adulthood, age had little effect on sex differences in LV mass (Fig 1) or body size (Fig 3) (.05<P<.0001). Similarly, the variability of LV mass did not differ between sexes in each adult age stratum and did not change during aging.

    Effect of Body Size Normalization on Age-Related Gender Differences in LV Mass

    The rate of variation of LV mass in relation to age was compared with the rate of variation of body weight and height2.7 (ie, the two measures linearly related to LV mass in normal-weight, normotensive individuals6 ), after standardization, to compare the slopes of the regression lines. In the 383 children and adolescents the standardized slopes of the relations of body weight and height2.7 to age were higher (slopes=1.54 and 1.69) than the rate of change in LV mass (standardized slope=1.34, both P<.001). In contrast, body surface area and height to the first power exhibited lower rates of increase than LV mass in relation to age (slopes=0.69 and 0.41, both P<.001).

    As a consequence of those different rates of change with age, LV mass normalized for body surface area or height followed the same trend in relation to age as LV mass, with statistically significant sex difference in adulthood (from 5th to 10th decile) (Fig 4). In contrast, indexation of LV mass for body weight (which could be done because overweight individuals were excluded from this population sample6 ) or more evidently, height2.7, attenuated the sex differences in adulthood (Fig 5) and revealed a decrease of indexed LV mass values from birth through the pubertal period. Beginning with the 7th decile (age 16 to 22 years), LV mass indexed for both body weight and height2.7 increased slightly with age both in men and in women. Sex differences were still statistically detectable in the 5th, 6th, and 8th deciles of age for LV mass/body weight, whereas normalization of LV mass for height2.7 resulted in the elimination of every statistical sex difference but for the highest decile.


    Our data show that LV mass is not significantly different in boys and in girls during infancy and childhood, suggesting that the initial number of cardiac myocytes is likely to be similar in males and in females. A clear-cut sex difference in LV mass becomes evident at puberty, when sex-specific hormonal influences are imposed on the original anatomic pattern; this increases during adolescence and remains roughly constant during adulthood (Fig 1). The greater increase in ventricular mass after puberty in boys than in girls was due to symmetrical increase both in chamber dimension and in wall thickness in males, yielding no sex difference in relative wall thickness. The small sex difference in LV mass during the prepubertal age was due to a tendency to greater LV chamber dimension in boys, in the presence of comparable wall thickness.

    Sex differences in LV mass were associated with greater increase in body size in males (due in most organs to an increase in cell number [hyperplasia] but in the heart to cell growth [hypertrophy]), but it was evident before a stable difference in body weight was reached. Because the mitotic activity of normal human cardiac myocytes stops in the first year of life,1 the greater increase in ventricular mass in males reflects disproportionate increases in the size of myocytes (hypertrophy).

    Our results are consistent with those of the Bogalusa Heart Study,8 in which a small sex difference in LV mass was found in children 7 to 11 years old. The average values of LV mass in this range of age were virtually identical to the average values that we found in the subset between 9 and 12 years. In a different and larger group of children up to age 11 years (prepubertal age) studied in Cincinnati and in Naples (unpublished data), a 6% difference in LV mass between sexes attained statistical significance (52±22 g in 206 boys versus 49±20 g in 196 girls, P<.01). A slightly greater sex difference was found by Goble et al,9 who reported that boys have about 10% higher LV mass than girls (mean, 83 g versus 75 g) at the age of 11 years, although girls were slightly taller and heavier than boys. A higher exercise capacity and more frequent family history of heart disease or hypertension in boys than in girls might have contributed to the greater sex difference in LV mass in that study.

    Because the initial LV mass and probably the number of myocytes is similar although not equal in boys and in girls (the 5% to 10% difference observed in this and other studies attains statistical significance when the population sample is very large), the fact that “normal” adult myocardial mass in men is about 30% greater than in women indicates that a state of relative cardiac hypertrophy exists in apparently normal adult men. This sex difference is markedly attenuated when the different body sizes in women and men are taken into account with allometrically appropriate normalization of LV mass (ie, weight in nonoverweight subjects or height2.7). The marked reduction in sex difference obtained using indexation for height2.7 suggests that this measure of body size might be an estimate of lean body mass, which has been shown to eliminate sex differences in LV mass in previous studies.10

    As is evident in Figs 4 and 5, the various methods of indexation of LV mass for body size reveal markedly different patterns of LV growth in relation to body growth. This apparent inconsistency is indeed explained by the different rate of growth of the various measures of body size and LV mass. The rate of LV growth from infancy to adolescence is slower than that of body weight or height2.7, paralleling a physiological decline in metabolic rate per kilogram.11 Accordingly, LV mass indexed for either body weight or height2.7 decreases from birth to adolescence, a phenomenon that could not be detected using traditional indexations for body surface area or height to the first power. The different performance of height to the first power and height2.7 (a near-to-cube function) reflects the geometric relations with body and LV weight (both three-dimensional measures). Of note, Malcom et al12 found LV mass in children and adolescents to be related most closely to height2.5, and Urbina et al13 confirmed LV mass to be related to height to a power close to 2.7 in the cohort of the Bogalusa Heart Study, indicating reasonable stability of this allometric relation in different populations.

    Limitations of the Study

    This is a cross-sectional study rather than one based on long-term changes in single individuals. Furthermore, exclusive selection of apparently normal subjects precludes discrimination between physiological and pathological forms of LV hypertrophy. All the relations between age and related variables should be interpreted with these limitations in mind. Further longitudinal studies in which subjects are followed over time and the physiological changes occurring with pubertal development are more precisely determined should provide improved understanding of these relations. However, in the interval before such data can be derived from multiyear follow-up, cross-sectional studies can provide useful interim data concerning the increasingly important topic of relations between body and organ growth and adult diseases.14

          Figure 1.

    Figure 1. Plot shows mean values for left ventricular mass (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (61 subjects in each decile, 62 in the highest one) (horizontal axis). Sex differences in left ventricular mass only attain statistical significance after puberty.

          Figure 2.

    Figure 2. Top, Plot shows mean values for left ventricular (LV) diastolic dimension (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Bottom, Mean values of wall thickness (averages of posterior wall and septum, vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Similar to left ventricular mass, sex differences become significant in the 5th decile (between 12 and 14 years).

          Figure 3.

    Figure 3. Top, Plot shows mean values for body weight (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Sex differences start in the 5th decile but become comparable to differences in left ventricular mass later, as shown in Fig 1 (between 16 and 22 years). Bottom, Mean values for height (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Sex differences are similar to those in left ventricular mass shown in Fig 1.

          Figure 4.

    Figure 4. Top, Plot shows mean values for left ventricular mass/height (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Bottom, Mean values for left ventricular mass/body surface area (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Sex differences in left ventricular mass indexes are often smaller in magnitude but show a similar relation to age as found with unindexed values (see Fig 1).

          Figure 5.

    Figure 5. Top, Plot shows mean values for left ventricular mass/height2.7 (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Bottom, Mean values for left ventricular mass/body weight (vertical axis) in normotensive, normal-weight males (squares) and females (diamonds) in deciles of age (horizontal axis). Sex differences in left ventricular mass are substantially offset by these normalizations, especially using height2.7. Consistent with the different rate of change of left ventricular mass and body size in relation to age (see text), left ventricular mass indexed for approximately three-dimensional measures of body size decreases during body growth.

    Table 1. Age and Sex Distribution in Deciles of Population

    Sex DistributionAge RangeMean Age±1 SD
    First decile (n=61)26 F /35 M3 mo-4 y2.53 ±1.02
    Second decile (n=61)27 F /34 M4-6 y5.10 ±0.67
    Third decile (n=61)36 F /25 M6-9 y7.72±1.03
    Fourth decile (n=61)27 F /34 M9-12 y10.61±0.74
    Fifth decile (n=61)26 F /35 M12-14 y13.07±0.77
    Sixth decile (n=61)32 F /29 M14-16 y15.26±0.60
    Seventh decile (n=61)35 F /26 M16-22 y18.64±1.68
    Eighth decile (n=61)23 F /38 M22-38 y29.28±4.63
    Ninth decile (n=61)22 F /39 M38-49 y42.70±3.58
    Tenth decile (n=62)24 F /38 M49-70 y56.61 ±5.51

    Table 2. Left Ventricular Mass in Boys and Girls From Birth to 12 Years

    AgeLeft Ventricular Mass, gP
    ≤3 y19 /1427.2±5.823.5 ±8.8.19
    4-6 y50 /4839.7±9.338.0±9.3.38
    7-9 y25 /3157.1±17.353.7±13.5.43
    10-12 y38 /3576.6±19.874.3±23.6.64

    This study was supported in part by grant HL-18323 from the National Heart, Lung, and Blood Institute, Bethesda, Md. We would like to thank Virginia Burns for her assistance in preparation of the manuscript.


    Correspondence to Dr Giovanni de Simone, Division of Cardiology, Box 222, The New York Hospital-Cornell Medical Center, 525 E 68th St, New York, NY 10021.


    • 1 Hort W. Quantitative histologische Untersuchungen an wachsenden Herzen. Virchows Arch.1953; 323:223-242. CrossrefMedlineGoogle Scholar
    • 2 Zak R. Development and proliferative capacity of cardiac muscle cells. Circ Res.1974; 35:17-26. Google Scholar
    • 3 Devereux RB, Lutas EM, Casale PN, Kligfield P, Eisenberg RR, Hammond IW, Miller DH, Reis G, Alderman MH, Laragh JH. Standardization of M-mode echocardiographic LV anatomic measurements. J Am Coll Cardiol.1984; 4:1222-1230. CrossrefMedlineGoogle Scholar
    • 4 Levy D, Savage DD, Garrison RJ, Anderson KM, Kannel WB, Castelli WP. Echocardiographic criteria for left ventricular hypertrophy: the Framingham Heart Study. Am J Cardiol.1987; 59:956-960. CrossrefMedlineGoogle Scholar
    • 5 de Simone G, Devereux RB, Roman MJ, Ganau A, Chien S, Alderman MH, Atlas S, Laragh JH. Gender differences in left ventricular anatomy, blood viscosity and volume regulatory hormones in normal adults. Am J Cardiol.1991; 68:1704-1708. CrossrefMedlineGoogle Scholar
    • 6 de Simone G, Daniels SR, Devereux RB, Meyer RA, Roman MJ, de Divitiis O, Alderman MH. Left ventricular mass and body size in normotensive children and adults: assessment of allometric relations and impact of overweight. J Am Coll Cardiol.1992; 20:1251-1260. CrossrefMedlineGoogle Scholar
    • 7 Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol.1986; 57:450-458. CrossrefMedlineGoogle Scholar
    • 8 Burke GL, Arcilla RA, Culpepper WS, Webber LS, Chiang YK, Berenson GS. Blood pressure and echocardiographic measures in children: the Bogalusa Heart Study. Circulation.1987; 75:106-114. CrossrefMedlineGoogle Scholar
    • 9 Goble MM, Mosteller M, Moskowitz WB, Schieken RM. Sex differences in the determinants of left ventricular mass in childhood: the Medical College of Virginia Twin Study. Circulation.1992; 85:1661-1665. CrossrefMedlineGoogle Scholar
    • 10 Hammond IW, Devereux RB, Alderman MH, Laragh JH. Relation of blood pressure and body build to left ventricular mass in normotensive and hypertensive employed adults. J Am Coll Cardiol.1988; 12:996-1004. CrossrefMedlineGoogle Scholar
    • 11 Schmidt-Nielsen K. Scaling: Why Is Animal Size So Important? Cambridge, England: Cambridge University Press; 1984:56-89. Google Scholar
    • 12 Malcom DD, Burns TL, Mahoney LT, Lauer RM. Factors affecting left ventricular mass in childhood: the Muscatine Study. Pediatrics.1993; 92:703-709. CrossrefMedlineGoogle Scholar
    • 13 Urbina EM, Gidding SS, Bao W, Pickoff AS, Berdusis K, Berenson GS. Effect of body size, ponderosity and blood pressure on left ventricular growth in children and young adults in the Bogalusa Heart Study. Circulation.1995; 91:2400-2406. CrossrefMedlineGoogle Scholar
    • 14 Weder AB, Schork NJ. Adaptation, allometry and hypertension. Hypertension.1994; 24:145-156.LinkGoogle Scholar