Human Reproduction vol.12 no.6 pp.1152–1155, 1997
Utility of follicle stimulating hormone (FSH), luteinizing
hormone (LH), oestradiol and FSH:LH ratio in
predicting reproductive age in normal women
You Kon Kim1,2, Samuel K.Wasser1,3,
Victor Y.Fujimoto1, Nancy A.Klein1,
Donald E.Moore1 and Michael R.Soules1
1Department
of Obstetrics and Gynecology, Division of
Reproductive Endocrinology and Infertility, Box 354793, University
of Washington School of Medicine, Seattle, WA 98195, USA
2Present
address: Department of Obstetrics and Gynecology,
Hallym University School of Medicine, Seoul, South Korea
3To
whom correspondence should be addressed
The relative efficacy of follicle stimulating hormone (FSH),
luteinizing hormone (LH), FSH:LH ratio and oestradiol is
evaluated as a predictor of ovarian reserve (reproductive
age) in normal women. Serum levels of FSH, LH, oestradiol
and FSH:LH ratios were measured during menstrual cycle
days 1–4 in younger (20–25 years; n J 23) and older (40–
45 years; n J 32) reproductive age women with regular
menstruation and normal reproductive function. On days
1–4, mean levels of FSH, oestradiol and FSH:LH ratios
were significantly higher in older compared with younger
women. FSH increased in concentration across cycle days
in both age groups. A significantly lower LH value in
younger versus older women was found only on day 1.
Oestradiol showed no change across days in the younger
group, but increased significantly from day 1 to day 4 in
the older group. FSH values on days 1 or 2 were the best
single predictor of age differences. However, the best
prediction of age differences was obtained by using the
combination of FSH and LH (as opposed to the FSH:LH
ratio) on day 1 of the menstrual cycle.
Key words: FSH:LH ratio/gonadotrophins/reproductive ageing
Introduction
At birth the human ovary has between 0.5 and 13106 primordial
follicles, with regression to near zero by the time of menopause
(Richardson et al., 1987; Faddy et al., 1992; Gougeon et al.,
1994). By the end of the reproductive lifespan, the ovary loses
its functional capacity (ovulation efficiency and hormone
secretion). An early secondary effect of this attenuation of
ovarian secretory capacity is a monotropic rise in early
follicular (EF) phase follicle stimulating hormone (FSH)
(Sherman and Korenman, 1975; Sherman et al., 1976; Reyes
et al., 1977; Lee et al., 1988; Klein et al., 1996a) associated
with a shortened follicular phase (Lenton et al., 1984; Klein
et al., 1996a,b; Santoro et al., 1996). The elevation of early
follicular phase FSH represents a standard clinical marker of
reduced ovarian reserve and diminished responsiveness of the
1152
ovary to ovulation induction (Scott et al., 1989; Toner et al.,
1991). Several methods have been used to evaluate ovarian
reserve, such as cycle day 3 (EF phase) serum FSH values
(Scott et al., 1989; Toner et al., 1991), the clomiphene citrate
challenge test (Loumaye et al., 1990; Tanbo et al., 1992; Scott
et al., 1995), the gonadotrophin releasing hormone (GnRH)
agonist stimulation test (Winslow et al., 1991), and EF phase
oestradiol levels (DeGeyter et al., 1993; Licciardi et al., 1995).
Although these tests have reasonable predictive values, normal
results do not necessarily ensure optimal responses to exogenous gonadotrophin stimulation. Furthermore, the assignment
of day 3 as the most sensitive day for clinical testing appears
to be arbitrary. Accordingly, we examined changes in endocrine
measures of ovarian reserve on days 1–4 of the menstrual cycle.
The pituitary gland secretes FSH and luteinizing hormone
(LH) in a pulsatile manner in response to GnRH. While the
FSH rise occurs relatively early in reproductive ageing, a
subtle rise in LH is only observed at a later stage and to a
lesser degree (Lee et al., 1988), and LH does not demonstrate
a sustained increase until after the menopause (Sherman et al.,
1976; Lee et al., 1988). However, a recent study in older
ovulatory women showed a decreased LH response to GnRH
stimulation compared with younger women (Fujimoto et al.,
1996). This finding suggests that LH synthesis and secretion
may be different in older ovulatory women, despite the fact
that no significant changes in basal serum LH levels were
demonstrated. This being the case, both FSH and LH may
contribute information relevant to ovarian reserve. In support
of this concept, it has recently been reported that day 3
FSH:LH ratio is useful in predicting the outcome of an in-vitro
fertilization (IVF) cycle in women with a normal day 3 FSH
(Mukherjee et al., 1996). Another study has shown a subtle
increase in LH with age (Ebbiary et al., 1994). In order to
understand better the relationship between FSH and LH as it
pertains to the ageing female, we evaluated the EF phase FSH,
LH, oestradiol and FSH:LH ratio in normal ovulatory younger
(control group) and older (study group) women on days 1–4
of the menstrual cycle.
Materials and methods
Subjects
Sixty-six normally cycling women age 20–25 [group Y (younger),
n 5 29] and 40–45 years [group O (older), n 5 37] were recruited
for a series of studies on reproductive ageing (Sherman and Korenman,
1975; Reyes et al., 1977). All subjects were healthy and were of
normal weight for height (body mass index of 18–24 kg/m2). All
subjects had evidence of regular ovulation based on (i) regular
menstrual cycles every 25–35 days, (ii) a prestudy biphasic basal
© European Society for Human Reproduction and Embryology
Age-related changes in early follicular phase
body temperature graph, and (iii) serum progesterone of .10 nmol/
l. Evidence of dominant follicle development in the study cycle was
confirmed by serial ultrasound examinations of follicle growth and
daily measurement of serum oestradiol. Exclusion criteria included
any medications (including no exogenous hormones within 6 weeks
of the study), any past or current reproductive endocrine problems
(e.g. galactorrhoea, hirsutism) or infertility, and participation in .5
h/week of aerobic exercise. All subjects had normal serum prolactin
(,20 µg/l) and testosterone (,3.5 nmol/l) levels in the cycle
immediately preceding the study cycle. Informed consent was obtained
from each participant. The study protocol was approved by the
University of Washington Human Subjects Review Committee.
Hormone assay
In the study cycle, daily blood samples were obtained between 0700
and 1000 h by venipuncture beginning with the first day of menstrual
bleeding (day 1) and continuing daily until ovulation. For practical
reasons, an arbitrary cut-off of 2200 h was used for day 1. Subjects
who did not have blood drawn on day 1 were excluded from the
analysis (six younger and five older subjects). Serum was isolated
and frozen in aliquots at –4°C for subsequent analysis. All samples
from an individual were analysed in a single assay. Samples were
analysed in duplicate for LH (Second International Standard) and FSH
by an immunoradiometric method (MAIA clone; Serono Laboratories,
Geneva, Switzerland). The inter-assay coefficients of variation were
12 and 14%; intra-assay coefficients of variation for LH and FSH
were 5 and 9% respectively.
The double-antibody radioimmunoassay for oestradiol was performed in duplicate using reagents supplied by ICN Biomedical, Inc.
(Costa Mesa, CA, USA). The antibody cross-reacts 20% with oestrone,
1.5% with oestriol, and ,1% with all other steroids. The inter- and
intra-assay coefficients of variation were 16 and 7% respectively.
Serum progesterone concentrations were determined in duplicate by
solid phase radioimmunoassay using reagents supplied by Diagnostic
Systems Laboratories, Inc. (Webster, TX, USA). The antibody crossreactivity is ,5% with all other steroids. The inter-assay and intraassay coefficients of variation were 13 and 11% respectively.
Statistical analysis
Separate repeated measures analyses of variance were used to examine
the relation between each of the dependent variables: FSH, LH, the
FSH:LH ratio, and oestradiol concentrations against the main effects
of age (i.e. younger versus older women) and cycle day (i.e. across
the first 4 days of the menstrual cycle). The interaction between age
and cycle day also was examined in each analysis. Simple regression
analyses were then used to generate standardized scores that allowed
direct comparison between each of these measures, on each day of
the menstrual cycle, to determine which single measure and day best
discriminated between these two age groups. These standardized
scores reflect the amount of variance explained using simple regression
analyses to predict younger versus older subjects based on FSH, LH,
the FSH:LH ratio, and oestradiol values, on each day of the menstrual
cycle. A hierarchical multiple regression analysis was then used to
predict age differences based on FSH and LH concentrations, in that
order, on each cycle day.
Results
The mean ages of groups Y and O were 23.3 6 1.5 (n 5 23)
and 42.3 6 1.7 years (n 5 32) respectively. The repeated
measures analyses of variance used to compare FSH, LH, the
FSH:LH ratio and oestradiol values in younger versus older
women across the first 4 days of the menstrual cycle are shown
in Figure 1A–D respectively. For FSH, significant main effects
were found for age (F 5 14.56; P , 0.004) and cycle day
(F 5 7.78; P , 0.0001). FSH was higher in older versus
younger women across all cycle days, and increased in concentration across days for younger and older women (Figure 1A).
For LH, a significant main effect of cycle day (F 5 3.79;
P , 0.02) was found along with a significant interaction
between age and cycle day (F 5 8.74; P , 0.0001). This
pattern resulted largely from a substantially lower LH value
in younger versus older women on day 1 only (Figure 1B).
For the FSH:LH ratio, the only significant main effect was
age (F 5 4.99; P , 0.03), with the ratio being consistently
higher in older versus younger women (Figure 1C). For
oestradiol, there were significant main effects of age (F 5
6.63; P , 0.02) and cycle day (F 5 9.87; P , 0.0001), as
well as a significant interaction between age and cycle day
(F 5 8.40; P , 0.0001). Oestradiol was consistently higher
in older versus younger women, and showed an increased
concentration across days of the menstrual cycle in older
women only (Figure 1D).
The simple regression-generated standardized scores
reflecting the amount of age-related variance in FSH, LH, the
FSH:LH ratio and oestradiol values, on each day of the
menstrual cycle, are shown in parentheses in Figure 1A–D.
FSH consistently explained more of the variance in younger
versus older women than did any of the other three measures
on any given day of the menstrual cycle (Figure 1). Moreover,
FSH values on days 1 and 2 were better predictors of
these age differences than were FSH values on days 3 or 4
(Figure 1A).
A significant age-related variance was found in LH values,
on day 1 only (Figure 1B). However, because this difference
was in the same direction as the age difference in FSH values
on day 1 (Figure 1A), this age difference was not reflected by
the FSH:LH ratio on day 1 (Figure 1C). We therefore used a
hierarchical multiple regression analysis, adding the LH value
into the regression following FSH, to determine whether the
addition of LH explained more of the variance in age than did
FSH alone. The addition of LH values into the regression
explained an additional 9% of the variance (P 5 0.01), from
0.23 (for FSH alone) to 0.32 (for FSH and LH combined) on
day 1 only. Thus, the best predictor of age differences in this
study was obtained by using the combination of FSH and
LH (as opposed to the FSH:LH ratio) on day 1 of the
menstrual cycle.
Discussion
This study showed that FSH was the best single predictor
of younger versus older normal ovulatory women on any
of the first 4 days of the menstrual cycle. Moreover, FSH
values on days 1 and 2 were better predictors of this age
difference than were FSH values on days 3 or 4. This
finding is not surprising, considering that older women have
an earlier onset of the inter-cycle FSH rise, associated with
an earlier onset of the acute follicular phase oestradiol rise
and dominant follicle development, leading to higher early
1153
Y.Kon Kim et al.
Figure 1. Differences between younger (20–25 years) and older (40–45 years) normally cycling women during days 1–4 of the menstrual
cycle in (A) follicle stimulating hormone (FSH), (B) luteinizing hormone (LH), (C) FSH:LH ratio and (D) oestradiol, generated by repeated
measures analyses of variance. Solid bars, younger women; hatched bars, older women. Sample sizes: younger, 23 women; older, 32
women. Values in parentheses represent amount of age-related variance explained (r2) in the measure shown in that graph, on each cycle
day, by simple regression analyses. *P , 0.05; **P , 0.001.
follicular phase oestradiol levels (Klein et al., 1996a,b).
This early rise in oestradiol may in turn result in early
suppression of pituitary FSH secretion. Basal EF phase FSH
is the most commonly utilized predictor of ovarian reserve.
A subtle day 3 rise in FSH suggests declining ovarian
reserve in spite of regular menses, and portends decreased
fertility rates in both infertile and normal women (Scott
et al., 1989; Toner et al., 1991). While basal day 3 FSH
has become a standard to evaluate ovarian reserve, our
results suggest that it may be more useful to evaluate FSH
on day 1 or 2 instead of day 3 and theoretically to begin
ovulation induction earlier in the menstrual cycles of older
infertile women.
The serum LH level was significantly elevated in the
older group on cycle day 1. This finding may be related to
the fact that there is an earlier onset of the inter-cycle FSH
rise (Klein et al., 1996a) associated with earlier dominant
follicle development and ovulation in older ovulatory women
(Lenton et al., 1984; Klein et al., 1996a,b). (While older
ovulatory women have more advanced follicular development
on any given follicular phase cycle day, this finding may
be due to either earlier onset or an accelerated rate of
follicle development.) Examining LH on cycle day 1, in
addition to FSH, may improve predictive value even further,
but not when examined as the FSH:LH ratio. Because the
LH difference on cycle day 1 was in the same direction as
the age difference in FSH values on day 1, this age
1154
difference was not reflected by the FSH:LH ratio on day
1. This finding contradicts a recent report in the medical
literature that prognosis was generally poor in patients
undergoing IVF cycles if the day 3 FSH:LH ratio was
elevated to ù3.6, even though the day 3 FSH value was
normal (Mukherjee et al., 1996). However, their cut-off
value was more than 4.0 standard deviations above the ratio
for our younger age group (1.40 6 0.54 versus 3.6), which
was also greater than the highest FSH:LH ratio (3.42) in
our older age group. These study differences may have
resulted from their testing infertility patients rather than
normal healthy subjects as in our study.
We have demonstrated that an EF phase serum FSH level
and a cycle day 1 LH level have additive predictive values
for chronological age in two groups of normal women.
Thus, these changes in gonadotrophin secretion patterns can
be interpreted as signs of reproductive ageing. Whether such
changes are indicative of premature ovarian ageing and
predictive of relative subfertility and gonadotrophin resistance
remains to be determined in a clinical setting.
Acknowledgements
We wish to thank Gretchen Davis for assistance in all phases of
this study and Cydney Foote for preparation of the manuscript.
This study was supported in part by grants from the National
Institutes of Health (R01-HD18967 and P50-HD-12629) and by
the Hallym University Medical Center.
Age-related changes in early follicular phase
References
DeGeyter, C., DeGeyter, M., Castro, E. et al. (1993) Predictive parameters
for ovarian response to hyperstimulation with exogenous gonadotropins
after suppression of gonadotropin secretion of the pituitary using a longacting GnRH agonist. Eur. J. Obstet. Gynecol. Reprod. Biol., 51, 139–147.
Ebbiary, N., Lenton, E. and Cooke, I. (1994) Hypothalamic–pituitary ageing:
progressive increase in FSH and LH concentrations throughout the
reproductive life in regularly menstruating women. Clin. Endocrinol., 41,
199–206.
Faddy, M., Gosden, R., Gougeon, A. et al. (1992) Accelerated disappearance
of ovarian follicles in mid-life: implications for forecasting menopause.
Hum. Reprod., 7, 1342–1346.
Fujimoto, V., Klein, N., Battaglia, D. et al. (1996) The anterior pituitary
response to a gonadotropin-releasing hormone challenge test in normal
older reproductive age women. Fertil. Steril., 65, 539–544.
Gougeon, A., Ecochard, R. and Thalabard, J. (1994) Age-related changes of
the population of human ovarian follicles: increase in the disappearance
rate of non-growing and early-growing follicles in aging women. Biol.
Reprod., 50, 653–663.
Klein, N., Battaglia, D., Fujimoto, V. et al. (1996a) Reproductive aging:
accelerated follicular development associated a monotropic follicle
stimulating hormone rise in normal older women. J. Clin. Endocrinol.
Metab., 81, 1038–1045.
Klein, N., Battaglia, D., Miller, P. et al. (1996b) Ovarian follicular development
and the follicular fluid hormones and growth factors in normal women of
advanced reproductive age. J. Clin. Endocrinol. Metab., 81, 1946–1951.
Lee, S., Lenton, E., Sexton, L. et al. (1988) The effect of age on the cyclical
patterns of plasma LH, FSH, oestradiol and progesterone in women with
regular menstrual cycles. Hum. Reprod., 3, 851–855.
Lenton, E., Landgren, B., Sexton, L. et al. (1984) Normal variation in the
length of follicular phase of the menstrual cycle: effect of chronological
age. Br. J. Obstet. Gynaecol., 91, 681–684.
Licciardi, F., Liu, H. and Rosenwaks, Z. (1995) Day 3 estradiol serum
concentrations as prognosticators of ovarian stimulation response and
pregnancy outcome in patients undergoing in vitro fertilization. Fertil.
Steril., 64, 991–994.
Loumaye, E., Billion, J., Mine, J. et al. (1990) Prediction of individual
response to controlled ovarian hyperstimulation by means of a clomiphene
citrate challenge test. Fertil. Steril., 53, 295–301.
Mukherjee, T., Copperman, A., Lapiinski, R. et al. (1996) An elevated day
three follicle-stimulating hormone:luteinizing hormone ratio (FSH:LH) in
the presence of a normal day 3 FSH predicts a poor response to controlled
ovarian hyperstimulation. Fertil. Steril., 65, 588–593.
Reyes, F., Winter, J. and Faiman, C. (1977) Pituitary–ovarian relationships
preceding the menopause. I. A cross-sectional study of serum folliclestimulating hormone, luteinizing hormone, prolactin, estradiol and
progesterone levels. Am. J. Obstet. Gynecol., 129, 557–564.
Richardson, S., Senikas, V. and Nelson J. (1987) Follicular depletion during
the menopausal transition: evidence for accelerated loss and ultimate
exhaustion. J. Clin. Endocrinol. Metab., 65, 1231–1237.
Santoro, N., Brown, J., Adel, T. et al. (1996) Characterization of reproductive
hormonal dynamics in the perimenopause. J. Clin. Endocrinol. Metab., 81,
1495–1501.
Scott, R., Toner, J., Muasher, S. et al. (1989) Follicle-stimulating hormone
levels on cycle day 3 are predictive of in vitro fertilization outcome. Fertil.
Steril., 51, 651–654.
Scott, R., Opsahl, M., Leonardi, M. et al. (1995) Life table analysis of
pregnancy rates in a general infertility population relative to ovarian reserve
and patient age. Hum. Reprod., 10, 1706–1710.
Sherman, B. and Korenman, S. (1975) Hormonal characteristics of the human
menstrual cycle throughout reproductive life. J. Clin. Invest., 55, 699–706.
Sherman, B., West, J. and Korenman, S. (1976) The menopausal transition:
analysis of LH, FSH, estradiol, and progesterone concentrations during
menstrual cycles of older women. J. Clin. Endocrinol. Metab., 42, 629–636.
Tanbo, T., Dale, P., Lunde, O. et al. (1992) Prediction of response to controlled
ovarian hyperstimulation: a comparison of basal and clomiphene citratestimulated follicle-stimulating hormone levels. Fertil. Steril., 57, 819–824.
Toner, J., Philput, C., Jones, G. et al. (1991) Basal follicle-stimulating hormone
level is a better predictor of in vitro fertilization performance than age.
Fertil. Steril., 55, 784–791.
Winslow, K., Tonor, J., Brzyski, R. et al. (1991) The gonadotropin-releasing
hormone agonist stimulation test – a sensitive predictor of performance in
the flare-up in vitro fertilization cycle. Fertil. Steril., 56, 711–717.
Received on October 15, 1996; accepted on April 2, 1997
1155