Human Reproduction Vol.17, No.3 pp. 641–646, 2002
Follicle stimulating hormone measured in unextracted
urine throughout the menstrual cycle correlates with age
and ovarian reserve
G.Jurjen E.Oosterhuis1,3,4, Istvan Vermes2, Henri W.B.Michgelsen1, Joop Schoemaker3 and
Cornelis B.Lambalk3
Departments of 1Obstetrics and Gynaecology and 2Clinical Chemistry, Medisch Spectrum Twente Hospital, Enschede and
3Department of Obstetrics and Gynaecology, Free University Medical Centre, Amsterdam, The Netherlands
4To
whom correspondence should be addressed at: VU Medical Centre, Department of Obstetrics and Gynaecology, PO Box 7057,
1007 MB Amsterdam, The Netherlands. Email: [email protected]
BACKGROUND: A method was previously described to measure FSH reliably in unextracted urine. The aim of
the current study was to establish the course of FSH measured in urine throughout the cycle. METHOD: Daily
urinary FSH (uFSH) concentrations were determined in 14 regularly menstruating volunteers aged 23–39 years
during one complete menstrual cycle. RESULTS: In each subject, mean daily uFSH measured in urine, as gold
standard for FSH tone, correlated significantly with FSH in early follicular phase fixed to menstruation on cycle
day 3 (r ⍧ 0.75, P ⍧ 0.002), or fixed to ovulation 9 days before the pre-ovulatory FSH surge (r ⍧ 0.87, P ⍧
0.0001), or when selected as being the highest follicular phase value (r ⍧ 0.91, P ⍧ 0.0001). Age correlated
significantly with mean daily uFSH (r ⍧ 0.67, P ⍧ 0.009), highest follicular phase uFSH (r ⍧ 0.60, P ⍧ 0.024),
uFSH on cycle day 3 (r ⍧ 0.80, P ⍧ 0.0006), and uFSH 9 days before FSH surge (r ⍧ 0.65, P ⍧ 0.0016). The
uFSH was also measured on cycle day 3 in 104 IVF patients in a cycle prior to pituitary down-regulation. The
uFSH correlated significantly with numbers of follicles (P ⍧ 0.02) and oocytes (P ⍧ 0.024). CONCLUSION: It is
concluded that cycle day 3 uFSH is a good reflection of the mean uFSH of the complete cycle, and there is a highly
significant correlation between uFSH and age and ovarian reserve. Measurement of FSH in urine on cycle day 3
seems to be a reliable and non-invasive tool for determining ovarian reserve in IVF.
Key words: age/assay/FSH/ovarian reserve/urine
Introduction
In reproductive endocrinology, basal FSH is measured in order
to detect women with ovarian failure. FSH measured in serum
on day 3 of the menstrual cycle is probably the most widely
used test to determine reproductive potential in women (Toner,
1993; Scott and Hofmann, 1995). Although much is known
about prognostic factors for success in assisted reproduction
and even the general population, knowledge about fecundity
influencing factors in the general population is still very low
(Oosterhuis et al., 1997). It is questioned whether the influence
of basal FSH for fertility potential in the general infertility
population is as large as it is in patients treated with IVF (Van
Montfrans et al., 2000).
Furthermore in the climacteric, FSH concentrations are used
to detect women at risk of climacteric symptoms or high bone
mineral density loss (Ebeling et al., 1996; Garton et al., 1996;
Ohta et al., 1996; Sowers et al., 1998; Wilbur et al., 1998).
Epidemiological health studies of large groups of people
require optimal cooperation of potential participants in order
to prevent bias as much as possible. Blood sampling, for
© European Society of Human Reproduction and Embryology
example, might introduce bias in excluding people not willing
to participate if invasive tests are required in a study.
In earlier studies on urinary FSH (uFSH), the applied
radioimmunoassays and fluoroimmunometric assays require
extraction of urine, which is less practical to use routinely in
the laboratory (Santner et al., 1981; Demir et al., 1995; Santoro
et al., 1996; Sachdev et al., 1998).
In order to create a tool to measure FSH in a non-invasive
way, we have adapted an assay to determine FSH in unextracted
urine (Oosterhuis et al., 1998a). In the current study, reference
values are provided for basal FSH in urine and fluctuations
during the menstrual cycle. The uFSH concentrations have
also been determined on cycle day 3 in 104 patients undergoing
IVF, and ovarian response has been correlated with uFSH.
Materials and methods
Subjects
Participants were 14 healthy volunteers, women recruited among
nurses, doctors, and friends who were likely to have ovulatory
641
G.J.E.Oosterhuis et al.
Figure 1. FSH measured in urine in 14 volunteers. All cycles are synchronized for FSH surge. Mean FSH ⫾ SD (A). The correlation
coefficient (B) and the P-value of the correlation (C) between mean daily uFSH concentrations and uFSH on any cycle day. Please note the
log scale in (C).
menstrual cycles. None of the participants had any history of infertility,
none had any reproductive organ surgery in their history, and none
had taken any oral contraception or had been breastfeeding in the
3 months prior to the study. All volunteers agreed to collect one urine
sample each day during a complete menstrual cycle. All cycles were
ovulatory.
Patients were 104 women participating in our IVF programme for
infertility treatment. On the third day of a menstrual cycle before
pituitary down-regulation, uFSH was determined as part of a fertility
work-up. They underwent follicular hyperstimulation for reasons of
infertility caused by endometriosis, tubal disorders, male factor
infertility or by unknown factors. The stimulation protocol was
described earlier (Oosterhuis et al., 1998b,c). In short, after entering
the IVF programme, women took oral contraceptives for 2–6 weeks
prior to pituitary down-regulation. The women then received gonadotrophin-releasing hormone (GnRH) analogue (Decapeptyl; Ferring
B.V., Hoofddorp, The Netherlands) in daily subcutaneous dosages.
After 1 week, a vaginal ultrasound was used to prove absence of
ovarian follicles ⬎5 mm. One week later, subcutaneous FSH (Metrodin
HP or Gonal F; Serono Benelux, The Hague, The Netherlands)
administration was commenced using 150–225 IU daily in a fixed
dosage for ovarian stimulation. Women ⬎35 years of age, and women
642
who in previous IVF cycles responded poorly to ovarian stimulation,
received 225 IU daily, others received 150 IU daily. One week later,
follicle size was measured by ultrasound every 2 days until at least
three follicles ⬎16 mm were obtained upon which oocyte retrieval
was planned. Human chorionic hormone (HCG), 10 000 IU (Profasi;
Serono) was injected subcutaneously 35 h before oocyte retrieval.
Oocyte retrieval was performed transvaginally using ultrasound for
visualization. In the laboratory, oocytes were isolated from the
follicular fluid, counted, and scored. If embryos were transferred (a
maximum of two embryos were transferred), a urinary pregnancy test
was performed if menses had not started 14 days later.
Study endpoints were number of ampoules of FSH administered
during treatment, number of days necessary to reach an adequate
number of follicles, number of preovulatory follicles, number of
oocytes isolated, and ongoing pregnancy (an intact intrauterine
pregnancy determined by transvaginal ultrasound).
Specimen handling
The urine samples were random void samples. The urine was collected
in a plastic container, one for each day in the volunteer study, without
any addition and stored in the refrigerator. Within 3 weeks after
collection in the volunteer study and on the same day in the IVF
FSH in urine correlates with age and ovarian reserve
Figure 2. FSH measured in urine in 14 volunteers. All cycles are synchronized for menstruation. Mean FSH ⫾ SD (A). The correlation
coefficient (B) and the P-value of the correlation (C) between mean daily urinary FSH concentrations and urinary FSH on any cycle day.
group, FSH was measured in urine as previously described (Oosterhuis
et al., 1998a). An AxSYM random access immunoassay analyser
(Abbott Laboratories, Abbott Park, IL, USA) with a MEIA (microparticle enzyme immunoassay) reagent kit was used. In order to
correct for dilution variability, creatinine was measured in all urine
specimens, and the uFSH results were normalized for creatinine
concentration by dividing the FSH concentrations by creatinine
concentrations.
Ethics
All volunteers gave written informed consent before entering the
study. All patients gave verbal informed consent. The study protocol
was approved by the ethics committee of Medisch Spectrum Twente
Hospital Group, Enschede, The Netherlands.
Statistics
Results are presented as the mean ⫾ standard deviation (SD).
Variables were normally distributed; otherwise a log transformation
was carried out to correct for skewness. Linear regression analysis
and Pearson’s correlation coefficient was calculated where appropriate.
Comparison of means was done with the unpaired t-test. Significance
was set at P ⬍ 0.05.
Results
The volunteers were aged 23–39 years. All 14 volunteers
completed collection of daily urine sampling during one
complete menstrual cycle. A periovulatory FSH surge defined
as a significant rise of FSH in the middle of the cycle was
detected in each volunteer (Figure 1). Cycle lengths varied
from 25 to 32 days.
For the purpose of this study, it was decided to consider
mean daily uFSH as best estimate (gold standard) of ‘FSH
tone’. This was calculated for each volunteer by dividing
the sum of daily urine FSH concentrations by number of
cycle days. The mean daily uFSH concentration correlates
significantly with uFSH on any cycle day, except during
two periovulatory days after synchronization for menses
(Figure 2).
643
G.J.E.Oosterhuis et al.
Figure 3. Correlation between age and urinary FSH (uFSH) measured on cycle day 3, mean daily urinary FSH, highest follicular phase
urinary FSH, and uFSH 9 days before the FSH surge.
Table I. Correlation between urinary FSH concentrations and age (Pearson’s correlation coefficient shown)
in 14 regularly menstruating women
Mean daily FSH
(P-value)
Highest follicular phase FSH
FSH on cycle day 3
FSH 9 days before FSH surge
Age
0.91
0.75
0.87
0.67
(0.0001)
(0.002)
(0.0001)
(0.009)
In particular, mean daily uFSH correlated significantly with
uFSH values of the early follicular phase, either related to the
start of menses (cycle day 3) or to the occurrence of ovulation
(FSH surge –9 days) or the highest follicular phase value
(Table I).
Age was significantly correlated with all uFSH parameters
(Figure 3).
In the IVF patients, the mean age was 32.9 years (range:
23–40). The uFSH concentration on cycle day three was
1.15 ⫾ 0.58 IU/mol creatinine. An average of 12.7 ⫾ 2.2 days
were necessary for follicular stimulation, and a mean of
644
Highest follicular
phase FSH
(P-value)
FSH on cycle
day 3
(P-value)
FSH 9 days
before FSH surge
(P-value)
0.63 (0.016)
0.85 (0.0003)
0.60 (0.024)
0.72 (0.0063)
0.80 (0.0006)
0.65 (0.016)
32.2 ⫾ 9.6 ampoules of FSH (75 IU FSH/ampoule) were
administered. A mean of 9.8 ⫾ 6.4 follicles and 7.1 ⫾ 4.7
oocytes were found per patient. Of the patients, 32 (31%)
became pregnant, and 24 (23%) had an ongoing pregnancy of
longer than 12 weeks, or had given birth to a live baby. The
uFSH concentrations on cycle day 3 correlated significantly
with the number of follicles (r ⫽ –0.23, P ⫽ 0.02) and with
the number of oocytes (r ⫽ –0.22, P ⫽ 0.024), but not with
the number of ampoules of FSH used or the number of days
of ovarian stimulation. Also, uFSH did not correlate with the
occurrence of pregnancy.
FSH in urine correlates with age and ovarian reserve
Discussion
This is the first report to describe the measurement of FSH in
unextracted urine during complete menstrual cycles in normally
cycling, healthy volunteers. There is a consequent significant
remarkably high correlation between age and all relevant
parameters of uFSH, in particular those of the early follicular phase.
In infertility populations, FSH is measured in serum to
detect age-dependent ageing ovaries (Cahill et al., 1994).
Inhibin B has also been shown to be a useful marker of ovarian
failure, and may be lower in women with diminished ovarian
reserve even before day 3 serum FSH is elevated (Seifer et al.,
1999). Ovarian failure is age- and FSH-related, and one
would expect an overt relationship between FSH and age.
Nevertheless, very strong correlations between serum FSH and
age in normal women ⬍40 years of age with regular menstrual
cycles have not yet been reported (Scott et al., 1989; Pearlstone
et al., 1992; Magarelli et al., 1996; El-Nemir et al., 1997;
Levi Setti et al., 1997; Broekmans et al., 1998). It has been
proposed (Marcus et al., 1993) to use uFSH as a marker for
ovarian failure. So far, only one study has reported age-related
differences in uFSH by comparing regularly cycling women
with those who were perimenopausal (Santoro et al., 1996).
FSH in urine reflects an integral of serum FSH, with fewer
fluctuations, and thus is perhaps a better steady state estimate
that apparently correlates well with ageing parameters.
Day 3 serum FSH estimates in IVF patients correlates with
numbers of growing follicles and obtained oocytes (Toner
et al., 1991; Sharif et al., 1998). This is believed to reflect the
reproductive age dependent decline of the cohort size and
ovarian oocyte reserve. It has been argued that the lack of
success in IVF in women with high basal FSH concentrations
is largely due to low numbers of available embryos, whilst
the age-related effect is due to declining implantation rates
(Toner et al., 1993). According to our findings in IVF patients
the same holds with regard to day 3 uFSH concentration.
However, the number of IVF cycles studied (only 104) limits
the strength of FSH measured in urine to predict IVF outcome.
A significant correlation is only seen with the numbers of
follicles and oocytes, and not with pregnancy. This is possibly
caused by the fact that a practical, clinical situation was used
to test the value of uFSH. Therefore, women ⬎35 years of
age and women who have shown signs of diminished ovarian
reserve received a higher daily dosage of FSH compared with
other women. In this situation, women who are more likely to
have diminished ovarian reserve receive more FSH compared
with other women and therefore respond better to ovarian
stimulation than they would have if they received the same
dosage of FSH as other women. This means that the correlation
found between uFSH and response to the IVF treatment
(number of follicles, number of oocytes, and pregnancy) is
less likely to be statistically significant than if all women were
given the same amount of FSH.
It has been argued that cycle day 3 FSH as estimate for
ageing is an arbitrary choice and, for instance, the highest
FSH concentration reached during the follicular phase would
better reflect ovarian reproductive age dependent changes
(Schipper et al., 1998). This is probably true. But it requires
multiple samples, and would be unpractical, invasive, and
costly. Hansen et al. have already shown that cycle day 3 is
as good as cycle day 2 or 4 to determine basal serum FSH
(Hansen et al., 1996). It has now been shown that the value
of cycle day 3 uFSH concentrations is just as good as the
value of any other cycle day as parameter of FSH secretion,
including the day of the highest follicular phase FSH, no
matter whether first day of menses or ovulation was taken as
reference point.
In conclusion, normal values have been established for FSH
measured in unextracted urine in daily samples throughout the
cycle in regularly menstruating women. Cycle day 3 values
are a good representation of uFSH secretion throughout the
cycle. Finally, cycle day 3 uFSH correlates well with the
inverse of the number of follicles ready to be stimulated as
well as with the numbers of oocytes obtained in IVF.
References
Broekmans, F.J., Scheffer, G.J., Bancsi, L.F. et al. (1998) Ovarian reserve
tests in infertility practice and normal fertile women. Maturitas, 30, 205–214.
Cahill, D.J., Prosser, C.J., Wardle, P.G. et al. (1994) Relative influence of
serum follicle stimulating hormone, age and other factors on ovarian
response to gonadotrophin stimulation. Br. J. Obstet. Gynaecol., 101,
999–1002.
Demir, A., Dunkel, L., Stenman, U.H. and Voutilainen, R. (1995) Age-related
course of urinary gonadotropins in children. J. Clin. Endocrinol. Metab.,
80, 1457–1460.
Ebeling, P.R., Atley, L.M., Guthrie, J.R. et al. (1996) Bone turnover markers
and bone density across the menopausal transition. J. Clin. Endocrinol.
Metab., 81, 3366–3371.
El-Nemir, A., Sabatini, L., Wilson, C. et al. (1997) Age, cigarette smoking
and early follicular phase FSH concentrations in infertile women. Hum.
Reprod., 12, 161–162.
Garton, M., Martin, J., New, S. et al. (1996) Bone mass and metabolism in
women aged 45–55. Clin. Endocrinol. (Oxf.), 44, 563–570.
Hansen, L.M., Batzer, F.R., Gutmann, J.N. et al. (1996) Evaluating ovarian
reserve: follicle stimulating hormone and oestradiol variability during cycle
days 2–5. Hum. Reprod., 11, 486–489.
Levi Setti, P.E., Vucetich, A., Sulpizio, P. et al. (1997) Patient age and FSH
concentration as prognostic factors in IVF outcome. Hum. Reprod., 12,
166–167.
Magarelli, P.C., Pearlstone, A.C. and Buyalos, R.P. (1996) Discrimination
between chronological and ovarian age in infertile women aged 35 years
and older: predicting pregnancy using basal follicle stimulating hormone,
age and number of ovulation induction intra-uterine insemination cycles.
Hum. Reprod., 11, 1214–1219.
Marcus, M., Grunfeld, L., Berkowitz, G. et al. (1993) Urinary folliclestimulating hormone as a biological marker of ovarian toxicity. Fertil.
Steril., 59, 931–933.
Ohta, H., Sugimoto, I., Masuda, A. et al. (1996) Decreased bone mineral
density associated with early menopause progresses for at least ten years:
cross-sectional comparisons between early and normal menopausal women.
Bone, 18, 227–231.
Oosterhuis, G.J.E., Michgelsen, H.W.B. and Vermes, I. (1997) Assessment of
ovarian reserve. J. Clin. Ligand Assay, 20, 316–324.
Oosterhuis, G.J.E., Lambalk, C.B., Michgelsen, H.W.B. et al. (1998a) Folliclestimulating hormone measured in unextracted urine: a reliable tool for easy
assessment of ovarian capacity. Fertil. Steril., 70, 544–548.
Oosterhuis, G.J.E., Michgelsen, H.W.B., Lambalk, C.B. et al. (1998b)
Apoptotic cell death in human granulosa-lutein cells: a possible indicator
of in vitro fertilization outcome. Fertil. Steril., 70, 747–749.
Oosterhuis, G.J.E., Vermes, I., Lambalk, C.B. et al. (1998c) Insulin-like
growth factor (IGF)-I and IGF binding protein-3 concentrations in fluid
from human stimulated follicles. Hum. Reprod., 13, 285–289.
Pearlstone, A.C., Fournet, N., Gambone, J.C. et al. (1992) Ovulation induction
in women age 40 and older: the importance of basal follicle-stimulating
hormone level and chronological age. Fertil. Steril., 58, 674–679.
645
G.J.E.Oosterhuis et al.
Sachdev, R., Von Hagen, S., Kamnani, A. and Santoro N. (1998) Persistent
pregnanediol glucuronide secretion after gonadotrophin suppression
indicates adrenal source of progesterone in premature ovarian failure. Hum.
Reprod., 13, 2061–2063.
Santoro, N., Brown, J.R., Adel, T. and Skurnick J.H. (1996) Characterization of
reproductive hormonal dynamics in the perimenopause. J. Clin. Endocrinol.
Metab., 81, 1495–1501.
Santner, S.J., Santen, R.J., Kulin, H.E. and Demers, L.M. (1981) A model for
validation of radioimmunoassay kit reagents: measurement of follitropin
and lutropin in blood and urine. Clin. Chem., 27, 1892–1895.
Schipper, I., De Jong, F.H. and Fauser, B.C. (1998) Lack of correlation
between maximum early follicular phase serum follicle stimulating hormone
concentrations and menstrual cycle characteristics in women under the age
of 35 years. Hum. Reprod., 13, 1442–448.
Scott, R.T. and Hofmann, G.E. (1995) Prognostic assessment of ovarian
reserve. Fertil. Steril., 63, 1–11.
Scott, R.T., Toner, J.P., Muasher, S.J. et al. (1989) Follicle-stimulating hormone
levels on cycle day 3 are predictive of in vitro fertilization outcome. Fertil.
Steril., 51, 651–654.
Seifer, D.B., Scott, R.T., Jr, Bergh, P.A. et al. (1999) Women with declining
ovarian reserve may demonstrate a decrease in day 3 serum inhibin B
before a rise in day 3 follicle-stimulating hormone. Fertil. Steril., 72, 63–65.
646
Sharif, K., Elgendy, M., Lashen, H. and Afnan, M. (1998) Age and basal
follicle stimulating hormone as predictors of in vitro fertilisation outcome.
Br. J. Obstet. Gynaecol., 105, 107–112.
Sowers, M., Randolph, J.F. Jr, Crutchfield, M. et al. (1998) Urinary ovarian
and gonadotropin hormone levels in premenopausal women with low bone
mass. J. Bone Miner. Res., 13, 1191–1202.
Toner, J.P. (1993) The significance of elevated FSH for reproductive function.
Baillieres Clin. Obstet. Gynaecol., 7, 283–295.
Toner, J.P., Philput, C.B., Jones, G.S. and Muasher, S.J. (1991) Basal folliclestimulating hormone level is a better predictor of in vitro fertilization
performance than age. Fertil. Steril., 55, 784–791.
Toner, J.P., Veeck, L.L. and Muasher, S.J. (1993) Basal follicle-stimulating
hormone level and age affect the chance for and outcome of pre-embryo
cryopreservation. Fertil. Steril., 59, 664–667.
Van Montfrans, J.M., Hoek, A., Van Hooff, M.H.A. et al. (2000) Predictive
value of basal FSH concentrations in a general subfertility population.
Fertil. Steril., 74, 97–103.
Wilbur, J., Miller, A.M., Montgomery, A. and Chandler, P. (1998)
Sociodemographic characteristics, biological factors, and symptom reporting
in midlife women. Menopause, 5, 43–51.
Submitted on October 30, 2000, resubmitted on 2 May, 2001; accepted on
29 October, 2001