In-utero androgen exposure and 2nd to 4th digit length ratio—comparisons between healthy controls and females with classical congenital adrenal hyperplasia

J.J. Buck, R.M. Williams, I.A. Hughes and C.L. Acerini1

Department of Paediatrics, University of Cambridge, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK

1 To whom correspondence should be addressed. e-mail: cla22{at}cam.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
BACKGROUND: Soft tissue measurements from the hand reveal lower second to fourth finger digit ratios (2D:4D) in males compared with females. The relatively longer 4th finger observed in males may be related to increased fetal exposure to androgens influencing the expression of Hox genes. METHODS: We have measured 2D:4D ratios in 69 healthy females [median age 9.3 (range 1.9–17) years], 77 control males [median age 13.86 (2.1–20.3) years] and in 66 females with classical virilizing congenital adrenal hyperplasia (CAH) (median age 8.5 (1.1–16.2) years] who are known to be exposed to high concentrations of androgens in utero. Measurements were determined from X-rays of the left hand using vernier callipers. Intra-observer variability in measurement technique was 0.01%. RESULTS: Control males had a significantly lower mean (SD) 2D:4D ratio [0.918 (0.029)] compared with female patients [0.927 (0.029), ANOVA P = 0.02]. No difference in 2D:4D ratio was observed between CAH females [0.925 (0.021)] and control females [0.927 (0.029)]. In contrast, 2D:4D ratio in males were significantly lower compared with CAH females (P = 0.03). CONCLUSIONS: 2D:4D ratios determined directly from radiographs of the left hand confirm significant differences between males and females. However, female patients with virilizing CAH do not have a male digit ratio pattern suggesting that in the left hand digit ratio development is not influenced by in-utero exposure to androgens.

Key words: 2D:4D digit ratio/androgens/congenital adrenal hyperplasia (CAH)/testosterone/X-ray hand


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
Homeobox-containing genes or Hox genes are evolutionarily conserved transcription factors known to be important for coordinating the temporal and spatial regulation of embryonic development in all multicellular organisms, including humans (Krumlauf, 1994Go). In mammals, Hoxa and Hoxd genes play a pivotal role in the normal differentiation of the reproductive tract (Hoxa-9 to 13) (Taylor et al., 1997Go) and skeletal systems (Hoxd-11 and 13) (Goff and Tabin, 1997Go). In mice, mutations of Hoxd-11 and Hoxa-13 result in phenotypes characterized by skeletal abnormalities (Mortlock et al., 1996Go; Boulet and Capecchi, 2002Go). Similarly, in humans, loss of function mutations of Hoxa-13 result in shortening of the digits and urogenital tract defects (Hand–Foot–Genital syndrome) (Mortlock and Innis, 1997Go). Hormones, such as the sex steroids, that act through nuclear hormone receptors have an important role in finely regulating gene expression processes involved in development, growth and reproduction. Endogenous sex steroids, including 17{beta}-estradiol and progesterone, and the exogenous non-steroidal estrogen diethylstilbestrol have been shown to directly modulate Hox gene expression in the reproductive tract (Taylor et al., 1999Go; Block et al., 2000Go; Cermik et al., 2001Go).

Recently, relationships between digit length and sex have been observed (Manning et al., 1998Go; Williams et al., 2000Go). Indirect soft tissue measurements from the hand have revealed significant differences in 2nd to 4th finger digit ratios (2D:4D) between the sexes with a higher ratio seen in females (females 1.00, versus males 0.98) (Manning et al., 1998Go). Male patients with infertility, low sperm counts and low testosterone levels have higher 2D:4D ratios (Manning et al., 1998Go). It has been proposed that sexual dimorphism in skeletal development is established early in fetal life (1st trimester), and that differences in androgen exposure in utero with high concentrations of fetal testosterone lead to expression of a low 2D:4D ratio, possibly by way of altered Hox gene expression (Manning et al., 1998Go).

Patients with congenital adrenal hyperplasia (CAH) due to 21 hydroxylase enzyme deficiency may be exposed to high concentrations of adrenal androgens in utero (New, 1998Go). Females with the classical salt wasting or simple virilizing forms of CAH are exposed to high concentrations of testosterone from early fetal life, leading to masculinization of the external genitalia, which is usually evident at birth. Given the hypothesis proposed by Manning and colleagues (Manning et al., 1998Go), females with virilizing CAH may be expected to have a lower or male pattern of 2D:4D digit ratio compared with healthy control females. We therefore report the results of a study in which 2D:4D digit ratio measurements were performed in healthy control females and males and were then compared with those in females with classical virilizing CAH.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
Second and fourth finger length measurements were determined directly from X-rays of the left hand and wrist using vernier callipers accurate to 0.5 mm. Measurements were made from the tip of the distal phalanx to the base of the proximal phalanx at a point midway across a line perpendicular to the base. X-rays from females with classical, virilizing CAH were obtained from the paediatric endocrinology clinics based at Cambridge (Addenbrooke’s Hospital), Oxford (John Radcliffe Hospital), Bristol (Royal Hospital for Sick Children), Newcastle (Royal Victoria Infirmary), Nottingham (Queen’s Medical Centre) and London (St George’s Hospital). All females with CAH presented with significant virilization of the genitalia at birth requiring corrective surgery and exhibited the characteristic biochemical profiles (e.g. elevated serum 17 hydroxy-progesterone and 21-deoxycortisol levels; increased urinary adrenocorticosteroid metabolites) consistent with a diagnosis of classical CAH due to 21-hydroxylase-enzyme deficiency. Controls were selected from an ethnically homogenous (white Caucasian) group of healthy males and females who had been seen at the paediatric endocrine clinic in Cambridge. They had been referred because of concerns about their growth and their height. Significant short stature and other causes of organic or genetic growth impairment were excluded following assessment and investigations, where necessary. Controls were matched for age and were known to have normal growth rates for age and puberty stage. Children with known chromosomal abnormalities, intrauterine growth retardation, congenital urogenital anomalies or skeletal dysplasias were excluded.

All X-rays were obtained from the left hand and wrist for standard bone age estimation, in CAH subjects as part of their routine clinical care and monitoring of treatment, and in control subjects as part of their initial evaluation at presentation. X-rays were excluded if clear demarcation of the extremities was not possible. X-rays of the left hand were used because the age and sex related standards for bone age measurement are based on radiographs of the left hand (Greulich et al., 1971Go; Tanner et al., 1983Go). Measurements were performed by a single observer (J.B.). In females with CAH, X-rays of the left hand were taken >=2 years after diagnosis of CAH in the neonatal period and following the commencement of therapy with physiological replacement doses of hydrocortisone and fludrocortisone. Ethical approval was obtained from Cambridge local research ethics committee.


    Statistics
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
To detect a ratio difference of 0.02 at 5% with a power of 90% it was calculated that a minimum of 54 subjects would be needed in each group. Simple linear regression analysis was used to observe the relationship between 2D:4D ratio and age. A univariate ANOVA model incorporating 2D:4D ratio as the dependent variable was used to compare differences between groups. Data are expressed as mean (± SD) unless otherwise stated. P-values <0.05 were considered to be significant. SPSS (SPSS version 10, Chicago, IL, USA) was used to perform all analyses.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
A total of 229 X-rays was examined (female = 76; male = 82; CAH = 71). Of these, measurements were obtained in 69 healthy females [median 9.3 (range 1.9–17) years], 77 males [13.86 (2.1–20.3) years] and in 66 females with classical virilizing CAH [8.5 (1.1–16.2) years]. Seventeen X-rays (female = 7; male = 5; CAH = 5) were rejected because of poor quality films. Repeat measurements were performed in 27 randomly selected patients and the intra-observer variability was found to be 0.01%. Table I shows the mean (± SD) 2D:4D ratios for each group.


View this table:
[in this window]
[in a new window]
 
Table I. 2D:4D ratio (mean (SD) in control females, control males and in females with virilizing congenital adrenal hyperplasia (CAH)
 
A trend for 2D:4D ratio to increase with age was observed in each of the three groups, which overall did not quite reach statistical significance (P = 0.08, Figure 1). Allowing for any possible effect of age, mean 2D:4D ratio in males was significantly lower compared with control females (P = 0.02) and to females with virilizing CAH (P = 0.03) (Table I). No significant difference in the 2D:4D ratio was observed between healthy control females and those with CAH (Table I).



View larger version (16K):
[in this window]
[in a new window]
 
Figure 1. Distribution of individual 2D:4D ratios according to age and to group (control males, control females and females with virilizing congenital adrenal hyperplasia (CAH).

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
This study confirms previous observations that there is a significant sex difference in 2D:4D ratio in the healthy population with a greater ratio found in females (Manning et al., 1998Go; Williams et al., 2000Go). We elected to measure digit length directly from X-rays, thus avoiding potential inaccuracies in finger measurement, which can be anticipated with those techniques that rely on the identification of surface anatomical landmarks either directly or from photo images of the hand. The use of X-rays to determine 2D:4D ratios may explain why overall ratio values were lower in our study (male 0.918/female 0.927) compared with other reports, where alternative digit measuring techniques have been used (male 0.980/female 1.00) (Manning et al., 1998Go; Williams et al., 2000Go). This is supported by observations made in a study of young Jamaican children of both sexes where lower 2D:4D ratio values were seen using X-rays compared with measurements made from photocopies of the hand (Manning et al., 2000Go). In the Jamaican study, however, sexual dimorphism in 2D:4D ratio was not observed with measurements made from X-rays, but was evident in photocopied images of the hand.

The disparity between X-ray and photocopy-determined 2D:4D ratios reflects the fundamental differences underlying the different type of measurements and raises concerns about the significance and accuracy of soft tissue measurements as a marker of true digit length and bone development. The lack of sexual dimorphism in X-ray-determined 2D:4D ratios seen in the Jamaican cohort study is in contrast with our study, due perhaps to ethnic or genetic differences between the two study populations. Furthermore, our control subjects were carefully selected and were known to be healthy individuals with normal growth patterns in whom the possibility of underlying confounding factors, such as skeletal abnormalities, chromosomal aberrations, endocrinopathies and urogenital disorders, had been excluded.

It has been reported that differences in 2D:4D ratio between the sexes is more pronounced when measurements are obtained from the right hand compared with the left (Manning et al., 1998Go). In our study, we were only able to make measurements from the left hand as X-rays of this hand are routinely used in the determination of bone age, and in order to allow comparison with age and sex, standards for bone age were determined from the left hand. X-rays of the right hand were therefore not available and a prospective study of the right hand X-rays would have been unethical in view of the added risks of radiation exposure. Whilst we have no data on 2D:4D ratio from the right hand, there is no reason to believe that a similar and perhaps even more significant sexually dimorphic pattern in 2D:4D ratio would have been seen in right hand X-rays

It has been hypothesized that sexual dimorphism in 2D:4D ratio is due to differences in androgen level exposure between the sexes in early fetal life (Manning et al., 1998Go). Exposure to higher concentrations of androgen in the male fetus is thought to determine a lower 2D:4D ratio, and that the 2D:4D ratio itself predicts testosterone concentrations and spermatogenesis in later life (Manning et al., 1998Go). Females with the virilizing (classical) form of CAH due to 21-hydroxylase-enzyme deficiency are exposed to high concentrations of adrenal androgens early in the first trimester, which is maintained throughout pregnancy (New, 1998Go). Indeed, serum concentrations of testosterone in the newborn are often higher than the male adult range (unpublished data). Affected females undergo significant virilization and masculinization of the external genitalia to a degree that gender determination at the time of birth is often not possible or incorrectly assigned as male. Females with CAH therefore seem suitable to test the hypothesis.

In a recent report Brown and colleagues, using photocopied images of the hand, have reported that females with CAH have a smaller 2D:4D ratio compared with non-CAH females (Brown et al., 2002Go). This observation was much more significant in the right hand (0.957 versus 0.981, P = 0.03) compared with the left (0.952 versus 0.968, P = 0.1) and is in keeping with the proposal that the relationship between 2D:4D ratio and testosterone concentration is more evident in the right hand than the left (Manning et al., 1998Go). Manning and colleagues observed that right hand digit ratio predicted relationships with testosterone, estrogen, LH, prolactin and sperm counts. In contrast, left hand 2D:4D ratio did not predict testosterone concentrations, although a weakly significant relationship with LH was observed (Manning et al., 1998Go). Nevertheless, a relationship between left hand 2D:4D ratio and testosterone cannot be entirely discounted from Manning’s observations given that in males, LH primarily stimulates testosterone production from the Leydig cells of the testis. Females with CAH in our study were known to be significantly virilized at birth, yet having been exposed to increased prenatal concentrations of adrenal androgens they did not demonstrate any lowering of the 2D:4D ratio in the left hand, and had greater adrenal androgen levels than control males. This is in contrast to the observations of Brown et al. (2002) where both left (0.952) and right (0.957) hand 2D:4D ratios in CAH females were similar to control males (0.955 and 0.957) suggesting an effect of prenatal androgens on digit 2D:4D ratio development (Brown et al., 2002Go). Nevertheless, it is difficult to draw any conclusions from their study given that only a very small number of CAH females were studied (n = 13). Furthermore, their report does not provide any detail regarding the type and severity of CAH in the affected cases. However, it is unlikely that they were more severely affected than the cases in our study, considering that all our subjects presented with significant genital virilization at birth.

It is possible that our results in CAH females were influenced by the effects of glucocorticoid treatment on 2D:4D ratio development. This would occur either through pituitary–adrenal suppression or by a direct effect on bone metabolism. While this cannot be discounted, such effects are unlikely. The 2D:4D ratio is established and thought to be fixed from an early gestational age (Garn et al., 1975Go). To our knowledge none of the mothers of the female CAH subjects in our study had received antenatal dexamethasone therapy as part of a prenatal virilization prevention programme. Postnatally, extreme care is taken with glucocorticoid replacement therapy in CAH in order to titrate doses at physiological replacement levels from an early stage, although a tendency to over-treat in early infancy has been recognized in the past (Jaaskelainen and Voutilainen, 1997Go; Van der Kamp et al., 2002Go). Although exposure to excessive amounts of glucocorticoids is known to have direct suppressive effects on bone growth (Brook et al., 1974Go), there is no evidence that the effect is specific to selective digits.

The results of our study raise the possibility that factors other than androgens may be more important in determining the sexual dimorphism observed in digit development. The presence of estrogen receptors in both osteoclasts and osteoblasts has suggested a direct role for these hormones in early skeletal development. In addition, estrogens can modulate the expression of genes belonging to the Hox family (Taylor et al., 1999Go). Fetal estrogen exposure in animal experiments results in significant changes in bone mineralization and in a shortening of bone length (Migliaccio et al., 1996Go). Perhaps non-hormonal factors determined by the sex chromosomes may be directly involved in digit length determination. It is well recognized that females with Turner’s syndrome (45 XO karyotype) are often observed to have abnormal shortening of the metacarpals (Preger et al., 1968Go) and relatively longer 4th digit phalanges (Necic and Grant, 1978Go). The skeletal abnormalities of Turner’s syndrome have been attributed to haplo-insufficiency of the SHOX gene (Clement-Jones et al., 2000Go). Other X-chromosome (non-SHOX) related syndromes are also known to be associated with congenital skeletal abnormalities, and hypoplasia of the phalanges and shortening of the metacarpal bones of the hand have been described (Lizcano-Gil et al., 1994Go; Donovan et al., 1996Go). The possibility that digit length determination is due to the expression and interaction with a sex chromosome dependent gene cannot be completely discounted.

In conclusion, we have confirmed, using radiological techniques, that a significant difference in 2D:4D ratio exists between healthy males and females when measured in the left hand. Females with virilizing classical CAH did not express a lower or male pattern 2D:4D ratio in this hand. The results suggest that, at least in the left hand, in-utero exposure to androgens is not responsible for the observed sexual dimorphic pattern, and that alternative mechanisms may be responsible for the difference in 2nd and 4th finger length development between the sexes.


    Acknowledgements
 
We would like to thank Dr Elizabeth Crowne (Royal Hospital for Sick Children, Bristol), Dr Tim Cheetham (Royal Victoria Infirmary, Newcastle), Dr Assunta Albanese (St Georges Hospital, London), Dr Derek Johnston (Queens Medical Centre, Nottingham) and Dr Justin Warner (John Radcliffe Hospital, Oxford) for their assistance in collecting X-rays from girls with CAH. We are also grateful to Dr Rakesh Amin (Department of Paediatrics, University of Cambridge) for his advice and help with data analysis.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Statistics
 Results
 Discussion
 References
 
Block, K., Kardana, A., Igarashi, P. and Taylor, H.S. (2000) In utero diethylstilbestrol (DES) exposure alters Hox gene expression in the developing mullerian system. FASEB, 14, 1101–1108.[Abstract/Free Full Text]

Boulet, A.M. and Capecchi, M.R. (2002) Duplication of the Hoxd11 gene cause alterations in the axial and appendicular skeleton of the mouse. Dev. Biol., 249, 96–107.[CrossRef][ISI][Medline]

Brook, C.G., Zachmann, M., Prader, A. and Murset, G. (1974) Experience with long-term therapy in congenital adrenal hyperplasia. J. Pediatr., 85, 12–19.[ISI][Medline]

Brown, W.M., Hines, M., Fane, B. and Breedlove, N.J. (2002) Masculinized finger length ratios in human males and females with congenital adrenal hyperplasia. Horm. Behav., 42, 380–386.[CrossRef][ISI][Medline]

Cermik, D., Karaca, M. and Taylor, H.S. (2001) HOXA10 expression is repressed by progesterone in the myometrium: differential tissue-specific regulation of the HOX gene expression in the reproductive tract. J. Clin. Endocrinol. Metab., 86, 3387–3392.[CrossRef]

Clement-Jones, M., Schiller, S., Rao, E., Blaschke, R.J., Zuniga, A., Zeller, R., Robson, S.C., Binder, G., Glass, I., Strachan, T. et al. (2000) The short stature homeobox gene SHOX is involved in skeletal abnormalities in Turner syndrome. Hum. Mol. Genet., 9, 695–702.[Abstract/Free Full Text]

Donovan, W.D., Brill, P.W., Winchester, P., Giampietro, P.F., Davis, J.G. and Harbison, M. (1996) Congenital fusion of the fourth and fifth metacarpals associated with primary gonadal failure. Skeletal Radiol., 25, 269–271.[CrossRef][ISI][Medline]

Garn, S.M., Burdi, A.R., Babler, W.J. and Stinson, S. (1975) Early prenatal attainment of adult metacarpal-phalangeal rankings and proportions. Am. J. Phys. Anthropol., 43, 327–332.[ISI][Medline]

Goff, D.J. and Tabin, C.J. (1997) Analysis of Hoxd-13 and Hoxd-11 misexpression in chick limb buds reveals that Hox genes affect both bone condensation and growth. Development, 124, 627–636.[Abstract/Free Full Text]

Greulich, W.W., Pyle, S.I. and Waterhouse, A.M. (1971) A radiographic standard of reference for the growing hand and wrist. Case Western Reserve University, Chicago.

Jaaskelainen, J. and Voutilainen, R. (1997) Growth of patients with 21-hydroxylase deficiency: an analysis of the factors influencing adult height. Pediatr. Res., 41, 30–33.[Abstract]

Krumlauf, R. (1994) Hox genes in vertebrate development. Cell, 78, 191–201.[ISI][Medline]

Lizcano-Gil, L.A., Garcia-Cruz, D., Cantu, J.M. and Fryns, J.P. (1994) The facio-digito-genital syndrome (Aarskog syndrome): a further delineation of the distinct radiological findings. Genet. Couns., 5, 387–392.[Medline]

Manning, J.T., Scutt, D., Wilson, J. and Lewis-Jones, D.I. (1998) The ratio of 2nd to 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and estrogen. Hum. Reprod., 13, 3000–3004.[Abstract]

Manning, J.T., Trivers, R.L., Thornhill, R. and Singh, D. (2000) The 2nd:4th digit ratio and asymmetry of hand performance in Jamaican children. Laterality, 5, 121–132.[CrossRef][ISI][Medline]

Migliaccio, S., Newbold, R.R., Bullock, B.C., Jefferson, W.J., Sutton, F.G., Jr., McLachlan, J.A. and Korach, K.S. (1996) Alterations of maternal estrogen levels during gestation affect the skeleton of female offspring. Endocrinology, 137, 2118–2125.[Abstract]

Mortlock, D.P. and Innis, J.W. (1997) Mutation of HOXA13 in hand-foot-genital syndrome. Nature Genet., 15, 179–180.[ISI][Medline]

Mortlock, D.P., Post, L.C. and Innis, J.W. (1996) The molecular basis of hypodactyly (Hd): a deletion in Hoxa13 leads to arrest of digital arch formation. Nature Genet., 13, 284–289.[ISI][Medline]

Necic, S. and Grant, D.B. (1978) Diagnostic value of hand X-rays in Turner’s syndrome. Acta Paediatr. Scand., 67, 309–312.[ISI][Medline]

New, M.I. (1998) Diagnosis and management of congenital adrenal hyerplasia. Annu. Rev. Med., 49, 311–328.[CrossRef][ISI][Medline]

Preger, L., Steinbach, H.L., Moskowitz, P., Scully, A.L. and Goldberg, M.B. (1968) Roentgenographic abnormalities in phenotypic females with gonadal dysgenesis. A comparison of chromatin positive patients and chromatin negative patients. Am. J. Roentgenol. Radium Ther. Nucl. Med., 104, 899–910.[ISI][Medline]

Tanner, J.M., Whitehouse, R.H., Cameron, N., Marshall, W.A., Healy, M.J. R. and Goldstein, H. (1983) Assessment of skeletal maturity and prediction of adult height. Academic Press. London, UK.

Taylor, H.S., Vanden Heuvel, G. and Igarashi, P. (1997) A conserved hox axis in the mouse and human reproductive system: late establishment and persistent expression of the hoxa cluster genes. Biol. Reprod., 57, 1338–1345.[Abstract]

Taylor, H.S., Igarashi, P., Olive, D.L. and Arici, A. (1999) Sex steroids mediate HOXA11 expression in the human peri-implantation endometrium. J. Clin. Endocrinol. Metab., 84, 1129–1135.[Abstract/Free Full Text]

Van der Kamp, H.J., Otten, B.J., Buitenweg, N., De Muinck Keizer-Schrama, S.M., Oostdijk, W., Jansen, M., Delemarre-de Waal, H.A., Vulsma, T. and Wit, J.M. (2002) Longitudinal analysis of growth and puberty in 21-hydroxylase deficiency patients. Arch. Dis. Child., 87, 139–44.[Abstract/Free Full Text]

Williams, T.J., Pepitone, M.E., Christensen, S.E., Cooke, B.M., Huberman, A.D., Breedlove, N.J., Breedlove, T.J., Jordan, C.L. and Breedlove, S.M. (2000) Finger-length ratios and sexual orientation. Nature, 404, 455–456.[CrossRef][ISI][Medline]

Submitted on October 11, 2002; accepted on January 13, 2003.