1 Research Institute for Endocrinology, Reproduction and Metabolism, Division of Reproductive Endocrinology and Fertility, Vrije Universiteit Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, 2 Department of Epidemiology and Medical Statistics, Free University, Amsterdam, 3 Netherlands Organization for Applied Scientific Research (TNO), Division Public Health and Prevention, Leiden and 4 Department of Youth Health Care of the Public Health Care Service Amstelland-de Meerlanden, Amstelveen, The Netherlands
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: adolescents/follicle stimulating hormone/luteinizing hormone/polycystic ovary syndrome
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
In the original protocol, it was decided not to use girls with irregular menstrual cycles for venepuncture because we expected a collection of heterogeneous menstrual cycle patterns in this group with the majority of girls only shortly after menarche. As ~10% of all girls had irregular menstrual cycles in the first part of the study, we concluded that information about this group might be important for the interpretation of the study. So, in the second part of the study girls with irregular menstrual cycles were invited to participate, and 83 girls did so.
Oral contraceptive users (n = 247) and those who had not reached menarche (n = 133) were excluded. Girls who had reached menarche within the previous 6 months (n = 128) were excluded because this time span is too short to determine a menstrual cycle pattern. Participants using hormonal medication other than oral contraceptives or insulin (n = 15), girls of whom one or both parents was not of west European descent (n = 207) and girls whose questionnaires were incomplete (n = 10) were also excluded.
Physical examination
Weight was measured to the nearest 0.5 kg using a mobile spring balance (Seca, Hamburg, Germany) and height to the nearest 0.5 cm using a mobile measuring rod (Microtoise, Stanley mabo, Poissy, France). Waist circumference was measured to the nearest 0.5 cm with a plastic tape at the smallest frontal waist diameter, usually at the level of the umbilicus. Hip circumference was measured at the broadest part of the lower body, usually at the level of the trochanters (Westrate et al., 1989). Pubarche and thelarche were scored according to the stages described (Tanner, 1962
; Roede van Wieringen, 1985
). Body hair grading was assessed according to the Ferriman and Gallwey score (Ferriman and Gallwey, 1961
). Abnormal body hair was defined as a Ferriman and Gallwey score of
1, hirsutism as a Ferriman and Gallwey score of
8. Acne was assessed according to the Plewig and Kligman score (Plewig and Kligman, 1975
). Acne was defined as a Plewig and Kligman score of
1, serious acne as a Plewig and Kligman score
2.
Blood sampling
Blood samples were taken from a forearm vein by the venoject system, usually at school, between 1200 and 1700 h. In girls with polymenorrhoea, regular or irregular menstrual cycles the blood samples were taken between the first and the 10th day of the menstrual cycle. To exclude influence of the midcycle luteinizing hormone (LH) peak, the blood samples from girls in these groups should have been taken at least 18 days before the next menstruation. In the girls with secondary amenorrhoea, blood samples were not timed. In oligomenorrhoeic girls, the blood sample was taken during the specific oligomenorrhoeic phase (SOP) defined as the time span between 2 weeks after the first day of a period and at least 3 weeks before the next period. This procedure excludes the possible influence of peri-ovulatory changes and post-ovulatory progesterone production on LH and androgen concentrations, which may extend into the follicular phase of the next menstrual cycle. If this happens, lower LH and androgen concentrations result in the first 2 weeks after this menstruation (Minakami et al., 1988; van Hooff et al., 1998b
). LH, androstenedione, testosterone and oestradiol concentrations during the SOP are known to be analogous to those in normogonadotrophic amenorrhoea (van Hooff et al., 1998b
). The majority of women in the latter group were PCOS patients.
Hormone assays
Plasma LH and follicle stimulating hormone (FSH) concentrations were determined by immunofluorometric assays (Amerlite, Amersham UK). For the LH assay the lower limit of detection was 0.3 IU/l. The intra-assay coefficients of variation (CV) were 5% at the level of 10 IU/l and 3% at levels of 20 IU/l and 40 IU/l. The inter-assay CV was 10% at the level of 2 IU/l and 6% at the level of 40 IU/l. For the FSH assay, the lower limit of detection was 0.5 IU/l. The intra-assay CV was 6% at the level of 5 IU/l and 5% at levels of 15 IU/l and 40 IU/l. The interassay CV was 9% at the level of 3 IU/l and 5% at the level of 35 IU/l. Prolactin concentrations were measured by an immunoradiometric-assay (Medgenix Diagnostics, Fleurus Belgium) with a lower limit of detection of 0.05 IU/l. The intra-assay CV was 4% at the level of 0.25 IU/l and 6% at a level of 1.0 IU/l. The interassay CV was 9% at the level of 0.15 IU/l and 7% at the levels of 1.0 IU/l and 2.0 IU/l. Oestradiol concentrations were measured by means of a radioimmunoassay (Estradiol-2, Sorin Biomedica, Saluggia, Italy). The detection limit was 37 pmol/l. The intra-assay CV was 4% at the level of 110 pmol/l and 5% at the level of 1000 pmol/l, the interassay CV was 11% at the level of 70 pmol/l and 10% at the level of 400 pmol/l. Serum androstenedione and dehydroepiandrosterone sulphate (DHEAS) were determined by using a radioimmunoassay (Coat a Count, DPC, Los Angeles USA). For the androstenedione assay, the lower limit of detection was 0.4 nmol/l. The intra-assay CV was 5% at the level of 1 nmol/l and 8% at the level of 3 nmol/l. The interassay CV was 11% at the level of 3 nmol/l and 8% at the level of 11 nmol/l. For DHEAS, the detection limit was 0.2 µmol/l. The intra-assay CV was 4% at the level of 6 µmol/l and 5% at the level of 20 µmol/l, the interassay variation 8% at the level of 2.5 µmol/l and 6% at the level of 10 µmol/l. Testosterone concentrations were determined using a double antibody radioimmunoassay (Coat-A-Count, DPC, Los Angeles, USA). The lower limit of detection was 0.3 nmol/l; the intra-assay coefficient of variation was 12% at the level of 1.5 nmol/l and 4% at the level of 3 nmol/l. The interassay coefficient of variation was 12% at the level of 2 nmol/l.
Ethical considerations
The study was approved by the Committee on Ethics of Research involving Human Subjects of the Free University Hospital in Amsterdam. The adolescents were all under age. During the information meetings, it was clearly stated that the adolescents needed approval of their parents to participate in the study. They received an information letter to hand over to their parents. Parents gave informed consent by telephone or were present when the blood was taken. This policy was developed in collaboration with the Institutional Review Board.
Statistics and assessment of reference values
The gynaecological age (months) was calculated by subtracting the age at menarche (months) from the calendar age (months). For the statistical analysis, girls with secondary amenorrhoea were recorded as oligomenorrhoeic (Siegberg, 1987). The data were analysed with BMDP statistical software package (BMDP statistical software, Cork, Ireland). Within the regular menstrual cycle group a significant, positive correlation of gynaecological age with LH, FSH, testosterone, androstenedione and oestradiol was found. The mean hormone concentrations of girls with regular menstrual cycles versus those with other menstrual cycle patterns were compared by analysis of covariance to adjust for differences in gynaecological age between the various menstrual cycle pattern groups.
The 5th and 95th centiles of hormone concentrations adjusted for gynaecological age in girls with regular menstrual cycles were defined as the normal range. Values above the adjusted 95th centile of the regular menstrual cycle group were considered elevated. The normal range for body mass index by gynaecological age was determined with data from the 1394 girls with regular menstrual cycles (van Hooff et al., 1998a).
Logistic regression analysis was performed to calculate odds ratios adjusted for differences in gynaecological age of the proportions of girls in oligomenorrhoea, polymenorrhoea and the irregular menstrual cycle group with hormone levels above the 95th centile of the regular menstrual cycle group (Table V) or clinical characteristics of PCOS (Table IV
). Logistic regression analysis was also used to estimate whether LH, testosterone, androstenedione and DHEAS contributed in an independent mode to discriminate between the oligomenorrhoeic group and the regular menstrual cycle group.
2 analysis for linear trends was done to evaluate the increase in proportion of oligomenorrhoeic girls with high LH or androgen concentrations by gynaecological age.
|
|
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
A greater number of girls with oligomenorrhoea, polymenorrhoea or irregular menstrual cycles who agreed to blood sampling reported participation in sports for >8 h a week than girls without blood sampling (2, P
0.05). Girls from the irregular menstrual cycle group with blood sampling also reported nipple discharge and dysmenorrhoea more frequently (
2 test, P = 0.02 and P < 0.01 respectively). The mentioned P values are not adjusted for multiple comparisons.
Adjustment in the analysis for differences in acne, dysmenorrhoea, participation in sports or nipple discharge by logistic regression did not change the results.
Physical examination and hormone levels
Table III shows a comparison of the mean hormone concentrations adjusted for gynaecological age between participants with regular menstrual cycles, oligomenorrhoea, polymenorrhoea or irregular menstrual cycles whose blood was sampled. LH, testosterone, androstenedione, DHEAS and oestradiol concentrations were significantly higher in oligomenorrhoeic girls than in girls with regular menstrual cycles. FSH and prolactin concentrations were higher in the polymenorrhoea group than in the regular menstrual cycle group. Mean LH, testosterone, DHEAS and prolactin concentrations were higher in the irregular menstrual cycle group than in the regular group. Table IV
shows the proportion of girls with abnormal body hair, acne, and high or low body mass index. More oligomenorrhoeic girls had a Ferriman and Gallwey score >1 than those with regular menstrual cycles. Only seven girls had a Ferriman and Gallwey score >8. No relation between hirsutism and menstrual cycle pattern could be documented.
|
|
The trend in relation to gynaecological age of oligomenorrhoeic girls with endocrine values above the 95th centile of the control group is shown in Table VI.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
LH and androgens, the hormones that play a central role in the PCOS, were higher in the oligomenorrhoea group than in the regular menstrual cycle group. These findings agree with studies on the endocrinology of oligomenorrhoea and long cycles of girls with irregular menstrual cycles in outpatient clinic populations (Emans et al., 1980; Singh, 1981
; Moll and Rosenfield, 1983;
Siegberg, 1987
; Venturoli et al., 1987
).
Higher mean FSH and prolactin concentrations were found in polymenorrhoeic girls compared to those having regular menstrual cycles. The only other study on polymenorrhoea in adolescents found no endocrine difference between 13 polymenorrhoeic girls and 28 controls, except for a lower mean progesterone concentration during the luteal phase in polymenorrhoeic girls, due to a higher frequency of anovulatory cycles (Siegberg et al., 1987). Higher mean prolactin levels found in our polymenorrhoeic group may be explained by prolonged oestrogen stimulation in anovulatory subjects (Zacur and Foster, 1992
). However, LH and androgen values were not high in this group, as might be expected in anovulatory subjects. A clear explanation for the higher mean FSH concentrations in the polymenorrhoeic girls cannot be given. Theoretically, high FSH concentrations may lead to a short follicular phase. This cannot be verified with our data. A negative correlation between FSH concentrations on days 34 of the cycle and the length of the follicular phase has been described (Apter and Vihko, 1985
).
Except for the present study, only one report (Venturoli et al., 1985) has described the proportion of participants with hormonal values outside the normal range. They did not use adolescent controls, but adults who will have had higher hormone values than adolescents. In anovulatory cycles longer than 35 days, an LH concentration above the mean + 2 SD of adult controls was found in six out of 20 adolescents (30%) in the follicular phase around the fifth day of the cycle but in 14 out of 20 (70%) adolescents during the premenstrual phase. This illustrates an effect of timing of blood sampling on hormone values in the oligomenorrhoeic cycle. In oligomenorrhoeic girls, we took blood samples at least 2 weeks after the first day of a menstruation and at least 3 weeks before the next, according to the concept of the specific oligomenorrhoeic phase (see Materials and methods). The rules we followed exclude the influence of peri- and post-ovulatory hormonal changes which also effect the hormone concentration in the first 2 weeks of the next menstrual cycle (Minakami et al., 1988
; van Hooff et al., 1998b
). We did not document ovulation in the study cycle.
Although significant differences in mean weight, body mass index, and waist circumference were documented between oligomenorrhoeic girls with and without endocrine signs of PCOS, the proportion of girls with obesity, acne or hirsutism was comparable in these subgroups (data not shown). In our study, these physical characteristics had no value in discriminating between oligomenorrhoeic girls with or without endocrine abnormalities.
The proportion of oligomenorrhoeic girls with high LH or androgen values, or both, appeared to increase in relation to gynaecological age, but the trend was not significant. Nearly 60% of the oligomenorrhoeic girls have endocrine signs compatible with PCOS in adults. A high LH value is the most frequent abnormality in oligomenorrhoeic adolescents.
Girls with irregular menstrual cycles also had higher mean LH, testosterone, DHEAS and prolactin concentrations. The differences between the irregular and regular menstrual cycle group were small in contrast to the differences between the oligomenorrhoea and regular menstrual cycle group. Moreover, the proportion of individuals with high concentrations was not increased in the irregular menstrual cycle group. Cross-sectional data from the POMP study showed that the prevalence of irregular menstrual cycles significantly decreased with increasing gynaecological age (van Hooff et al., 1998a). The prevalence of oligomenorrhoea showed no significant decrease. These findings may be in accordance with the hypothesis that high LH and androgen concentrations in adolescents with irregular menstrual cycles are a functional step of maturation of the ovulatory system (Apter, 1980
; Venturoli et al., 1985
), but that high concentrations of these hormones in oligomenorrhoeic girls are not transient and may be an early sign of PCOS.
The implications of endocrine abnormalities in our random population sample of oligomenorrhoeic adolescents are not clear. However, results of earlier studies suggest that most adolescents with oligomenorrhoea continue to have this menstrual cycle pattern combined with subfertility in adulthood (Southam and Richart, 1966; Kimura et al., 1988
; Apter and Vihko, 1990
). It was shown (Southam and Richart, 1966
) that 60% of adolescent patients, who had oligomenorrhoea 2 years after menarche kept this pattern in the following 8 years. In one study (Kimura et al., 1988
), it was found that the majority of 17 adolescent patients with menstrual cycle disorders and high LH concentrations suffered from ovulatory disturbance in adulthood. It has been shown (Apter and Vikho, 1990) that adolescent serum testosterone and androstenedione concentrations are preserved into adulthood and are reflected in fertility patterns during the third decade of life. Higher serum androgen concentrations were associated with lower fertility. No cut-off level for individuals with a high chance for fertility problems was given.
Two studies (Singh, 1981; Gardner, 1983
) reported that doctors tend to reassure adolescents that their menstrual pattern disorder will be transient and is due to incomplete maturation of the hypothalamicpituitaryovarian axis. Our study and other studies suggest that oligomenorrhoea in adolescents is not a transient stage in the physiological maturation of the hypothalamicpituitaryovarian axis but an early sign of PCOS. Endocrine evaluation should be considered before reassuring oligomenorrhoeic adolescents or prescribing oral contraceptives to these girls.
![]() |
Acknowledgments |
---|
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Apter, D. and Vihko, R. (1977) Serum pregnenolone, progesterone, 17-hydroxyprogesterone, testosterone and 5-alpha-dihydrotestosterone during female puberty. J. Clin. Endocrinol. Metab., 45, 10391048.[ISI][Medline]
Apter, D. and Vihko, R. (1985) Hormonal patterns of the first menstrual cycles. In Venturoli, S., Flamigni, C. and Givens, J. (eds), Adolescence in Females. Year Book Medical Publishers, Chicago.
Apter, D. and Vihko, R. (1990) Endocrine determinants of fertility: serum androgen concentrations during follow-up of adolescents into the third decade of life. J. Clin. Endocrinol. Metab., 71, 970974.[Abstract]
Emans, S.J., Grace, E. and Goldstein, D.P. (1980) Oligomenorrhoea in adolescent girls. J. Pediatr., 97, 815819.[ISI][Medline]
Ferriman, D. and Gallwey, J.D. (1961) Clinical assessment of body hair growth in women. J. Clin. Endocrinol. Metab., 21, 14401447.[ISI]
Fletcher, R.H., Fletcher, S.W. and Wagner, E.H. (1988) Clinical Epidemiology. The Essentials, 2nd edition. Williams & Wilkins, Baltimore, pp. 5861.
Gardner, J. (1983) Adolescent menstrual characteristics as predictors of gynaecological health. Ann. Hum. Biol., 10, 3140.[ISI][Medline]
Kimura, K., Minakami, H., and Tamada, T. (1988) [A longitudinal study on the prognosis of ovulatory disturbance in teenage patients with high LH and normal FSH serum levels]. Nippon Naibunpi Gakkai Zasshi., 64, 10881101.[Medline]
Minakami, H., Abe, N., Izumi, A., and Tamada, T. (1988) Serum luteinizing hormone profile during the menstrual cycle in polycystic ovarian syndrome. Fertil. Steril., 50, 990992.[ISI][Medline]
Moll, G.W. Jr and Rosenfield, R.L. (1983) Plasma free testosterone in the diagnosis of adolescent polycystic ovary syndrome. J. Pediatr., 102, 461464.[ISI][Medline]
Munster, K., Helm, P. and Schmidt, L. (1992) Secondary amenorrhoea: prevalence and medical contact a cross-sectional study from a Danish county. Br. J. Obstet. Gynaecol., 99, 430433.[ISI][Medline]
Plewig, G. and Kligman, A.M. (1975) Acne Morphogenesis and Treatment. Springer-Verlag, Berlin.
Roede, M.J. and Van Wieringen, J.C. (1985) Growth diagrams 1980: Netherlands third nation wide survey. Tijds. Soc. Geneesk., 63, 134 (supplement).
Sackett, D.L., Haynes, R.B., Guyatt, G.H. and Tugwell, P. (1991) Clinical Epidemiology. A Basic Science For Clinical Medicine, 2nd edition. Little, Brown and Company, Boston, pp. 178180.
Siegberg, R. (1987) Serum sex hormone concentrations in adolescent secondary amenorrhoea. Ann. Chir. Gynaecol., 76, 176180.[ISI][Medline]
Siegberg, R., Nilsson, C.G., Stenman, U.H. and Widholm, O. (1986) Endocrinologic features of oligomenorrhoeic adolescent girls. Fertil. Steril., 46, 852857.[ISI][Medline]
Siegberg, R., Nilsson, C.G., Stenman, U.H., and Widholm, O. (1987) Serum sex hormone levels in adolescent girls with polymenorrhoea. Ann. Clin. Res., 19, 183186.[ISI][Medline]
Singh, K.B. (1981) Menstrual disorders in college students. Am. J. Obstet. Gynecol., 140, 299302.[ISI][Medline]
Southam, A.L. and Richart, R.M. (1966) The prognosis for adolescents with menstrual abnormalities. Am. J. Obstet. Gynecol., 94, 637645.[ISI][Medline]
Tanner, J.M. (1962) Growth at Adolescence. Blackwell Scientific Publications, Oxford.
van Hooff, M.H.A., Voorhorst, F.J., Kaptein, M.B.M. et al. (1998a) Relationship of the menstrual cycle pattern in 1417 year old adolescents with gynaecological age, body mass index and historical parameters. Hum. Reprod., 13, 22522260.[Abstract]
van Hooff, M.H.A., van der Meer, M., Lambalk, C.B. and Schoemaker, J. (1998b) Effect of timing of bloodsampling for LH determination on the detection of elevated LH levels in oligomenorrhoeic women. Fertil. Steril., Supplement IFFS '98: S210(Abstract).
Venturoli, S., Porcu, E., Fabbri, R. et al. (1985) Endocrine profiles and ovarian aspects in adolescent menstrual abnormalities. In Venturoli, S., Flamigni, C. and Givens, J. (eds), Adolescence in Females. Year Book Medical Publishers, Chicago.
Venturoli, S., Porcu, E., Fabbri, R. et al. (1987) Postmenarchal evolution of endocrine pattern and ovarian aspects in adolescents with menstrual irregularities. Fertil. Steril., 48, 7885.[ISI][Medline]
Westrate, J.A., Deurenberg, J. and Van Tinteren, H. (1989) Indices of body fat distribution and adiposity in Dutch children from birth to 18 years of age. Int. J. Obesity, 13, 465477.[ISI][Medline]
Zacur, H.A. and Foster, G.V. (1992) Hyperprolactinemia and polycystic ovarian syndrome. Semin. Reprod. Endocrinol., 10, 236245.[ISI]
Submitted on December 15, 1998; accepted on May 24, 1999.