Human Reproduction vol.15 no.10 pp.

21292132, 2000

Dehydroepiandrosterone supplementation augments

ovarian stimulation in poor responders: a case series

P.R.Casson1, M.S.Lindsay, M.D.Pisarska,

S.A.Carson and J.E.Buster
Division of Reproductive Endocrinology and Infertility, Department
of Obstetrics and Gynecology, Baylor College of Medicine,
6550 Fannin, Suite 801, Houston, Texas 77030, USA
1To

whom correspondence should be addressed at: Department of

Obstetrics and Gynecology, University of Vermont, Burlington,
VT 05405, USA. E-mail: pcasson@vtmednet.org

In patients with poor response to ovarian stimulation with

gonadotrophins, growth hormone (GH) is sometimes used
to increase paracrine insulin-like growth factor-1 (IGF-1)
effect. We postulated that dehydroepiandrosterone (DHEA)
administration to poor responders would augment gonadotrophin effect via a similar mechanism. Baseline ovarian
stimulation response to a cycle with DHEA in five healthy
non-smoking women <41 years old was compared with day
3 FSH <20 mIU/ml. All had documented poor response
to vigorous gonadotrophin administration. After day 2
ultrasounds, DHEA-sulphate (DHEA-S), FSH, human
chorionic gonadotrophin (HCG), and testosterone were
measured, and the women were given 80 mg/day of oral
micronized DHEA for 2 months. While still on DHEA,
they underwent ovarian stimulation with FSH given i.m.
twice a day, and HCG (10 000 IU) at follicular maturity,
followed by intrauterine insemination. Cycle parameters
assessed were peak oestradiol, and peak oestradiol/
ampoule. The DHEA/ovarian stimulation cycles occurred
between 4 and 24 months after the control cycles. After 2
months DHEA treatment, DHEA-S increased to 544 55
g/dl, and testosterone increased to 67.3 6.1 ng/dl. All
five subjects (six cycles; one subject had two DHEA cycles)
had increased responsiveness; peak oestradiol concentrations increased from 266.3 69.4 pg/ml to 939.8 418.9
pg/ml. The oestradiol/ampoule ratio increased in all six
cycles, by a mean of 2.94 0.50 fold (P 0.012). One of
the cycles resulted in a delivered twin pregnancy. In
this small series, DHEA improved response to ovarian
stimulation even after controlling for gonadotrophin dose.
Supplemental DHEA treatment during ovarian stimulation
may represent a novel way to maximize ovarian response.
Key words: androgens/dehydroepiandrosterone/gonadotrophins/
ovarian stimulation/poor responders

Introduction
In recent years, the widespread application of assisted reproductive technology has revolutionized the treatment of all forms of
infertility. With few exceptions, however, assisted reproductive
European Society of Human Reproduction and Embryology

technology depends on ovarian stimulation and concurrent

multiple oocyte development, induced by administration of
large quantities of exogenous gonadotrophins. Unfortunately for
some infertile women, gonadotrophin administration results in
desultory ovarian response. While this is commonly due to
diminished ovarian reserve, as indicated by advanced age and/
or elevated basal day 3 FSH concentrations, a subset of these
patients are 41 years old and have normal FSH concentrations.
To overcome this problem several strategies have been
reported, with limited success. These include gonadotrophinreleasing hormone (GnRH) flare protocols (Padillo et al.,
1996; Hugues and Durnerin, 1998), high dose gonadotrophin
administration (Hofmann et al., 1993), oestrogen pre-treatment
down-regulation (Check et al., 1990), and concomitant growth
hormone (GH) administration (Homburg et al., 1991). GH is
thought to amplify intra-ovarian insulin-like growth factor-I
(IGF-I) paracrine effect, which is expressed by granulosa cells
and enhances gonadotrophin action (Adashi et al., 1991).
However, the clinical utility of combined GH/ovarian stimulation is limited; responses, while present, are not dramatic, and
recombinant GH is extravagantly expensive.
Dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulphate (DHEA-S) are ubiquitous steroids of
primarily adrenocortical reticularis zonal origin. These hormones circulate in high amounts in female reproductive life;
however, concentrations fall progressively with age (Orentreich
et al., 1984), leading to speculation that replacement of DHEA
and DHEA-S in the elderly may have age-retardant effects
(Casson et al., 1998).
Two lines of circumstantial evidence support use of exogenous DHEA to augment ovarian stimulation in women aged
3540 years who are poor responders. First, well controlled
studies demonstrate marked augmentation of serum IGF-I
concentrations with oral administration of physiological DHEA
(Morales et al., 1994; Diamond et al., 1996; Casson et al.,
1998a). Second, in vivo, DHEA is a steroid prohormone for
ovarian follicular sex steroidogenesis (Haning et al., 1993).
On this basis, we postulated that in patients 41 years old,
with previously demonstrated poor response and normal FSH
concentrations, administration of oral DHEA in combination
with gonadotrophin stimulation would result in enhanced
ovarian response. Therefore, we designed and executed the
following prospective case series repeated here.

Materials and methods

With approval of our Institutional Review Board, five women with
unexplained infertility were identified from our tertiary clinical
infertility practice. All were 41 years old, had day 3 FSH concentra-

2129

P.R.Casson et al.

Table I. Comparison of control and DHEA-supplemented cycles

Cycle

Peak oestradiol

1
2a
3
4
5
6a,b
aTwo

No. follicles 15 mm

No. ampoules HMG (75IU)

Peak oestradiol/ampoule (pg/ml amp)

Control

DHEA

Control

DHEA

Control

DHEA

Control

DHEA

218
479
178
155
89
479

362
2883
403
349
327
1315

2
1
1
1
0
1

2
3
2
3
1
2

28
26
33
54
36
26

24
36
32
26
60
34

7.8
18.4
5.4
2.9
2.5
18.4

15.1
80.1
12.6
13.4
5.5
38.7

cycles in same subject.

in twin pregnancy.

bResulted

Figure 1. The fold increase in peak oestradiol concentration

attained in control versus subsequent DHEA-supplemented ovarian
stimulation cycles. Six cycles, performed in five subjects, are
presented. The mean ( SEM) fold increase was 3.10 0.69
(P 0.02).
tions 20 mIU/ml, had unexplained infertility, and had had previous
poor response to vigorous gonadotrophin stimulation (peak oestradiol
attained was 500 pg/ml, the number of mature follicles was 2).
All had documentation of the most recent gonadotrophin cycle,
meeting the above criteria of poor response. One patient subsequently
had an additional DHEAovarian stimulation cycle, which was also
included in the analysis.
After informed consent, the subjects had baseline ultrasound scans
on cycle day 2, and blood was drawn for serum DHEA-S, FSH,
HCG, testosterone assays, and liver function tests. All subjects had
regular cycles, and normal liver, thyroid and kidney function. The
women were then given 80 mg/day of oral micronized DHEA (Belmar
Pharmacy, Lakewood, CO, USA) for 2 months. Monthly repeat
DHEA-S, testosterone, liver function tests, and ultrasound scans were
performed. After 2 months of DHEA pretreatment, and while still
remaining on this hormone, the subjects had a repeat ovarian
stimulation cycle. The stimulation protocol was started on day 2 and
consisted of two ampoules of 75 IU recombinant FSH (rFSH,
Follistim; Organon, West Orange, NJ, USA) given i.m. twice a day
for 5 days. One subject (no. 2) used purified urinary FSH (Metrodin;
Serono Laboratories Inc., Randolph, MA, USA) in both her control
and DHEA cycles. On day 7 repeat ultrasound and oestradiol
measurements were performed and rFSH dose was subsequently
adjusted for maximal response. At follicular maturity (1 or more
follicles of 16 mm average diameter) ovulation was induced with

2130

Figure 2. The fold increase in peak oestradiol concentration,

controlled for the number of 75 IU ampoules of gonadotrophin
used (peak oestradiol/no. of ampoules) in control versus subsequent
DHEA-supplemented ovarian stimulation cycles (six cycles, five
subjects). The mean ( SEM) fold increase was 2.94 0.50
(P 0.012).
HCG (10 000 IU i.m.), followed at 36 h by intrauterine insemination.
The ovarian response to gonadotrophins was assessed with transvaginal ultrasound and with serum oestradiol concentrations at 0800
hours. To assess the differences between the control and DHEA
cycles, peak oestradiol concentrations, number of follicles 15 mm
average diameter, and change in peak oestradiol obtained per ampoule
of gonadotrophin were compared. The fold increase in peak oestradiol
concentrations and oestradiol per ampoule of gonadotrophin was also
compared for the five patients (six cycles) using paired t-tests with
post hoc (Bonferronis) correction.

Results
The six DHEA/ovarian stimulation cycles occurred between 4
and 24 months after the control ovulation induction cycles;
the age range of the subjects at the time of DHEA/ovarian
stimulation was 3540 years. The subjects baseline serum
hormone values were: day 3 FSH, 10.7 1.9 IU/ml (mean
SEM); DHEA-S 122 51.5 g/dl; testosterone, 34.2 2.1
ng/dl. After 2 months of DHEA supplementation serum steroid
concentrations at 0800 hours were: DHEA-S, 544 55 g/
dl; and testosterone, 67.7 6.1 ng/dl. The subjects control
ovarian stimulation cycles had a mean peak oestradiol of

DHEA supplementation in ovarian stimulation

266 69 pg/ml, attained with 35.4 5.0 ampoules of

gonadotrophins, a stimulation duration of between 7 and 11
days, and a mean number of mature follicles of 1.0. With
DHEA supplementation, all five subjects had increased
responsiveness to gonadotrophins. Peak oestradiol concentrations obtained were 940 419 pg/ml attained with 35.6
6.5 ampoules of gonadotrophin which represented 3.1 6.4
fold increase over control cycles (P 0.02). The cycle data
are presented in Table I. The fold increase in peak oestradiol
concentrations is presented in Figure 1. The mean number of
mature follicles also increased from 1.0 to 2.2. Assuming
equivalent biopotency between the gonadotrophin preparations,
an assumption borne out by the literature (Hedon et al., 1995;
Follistim, package insert), the mean oestradiol per no. of 75
IU ampoules ratio increased in all five patients (six cycles) by
2.94 0.50 fold (P 0.012). These increases are presented
in Figure 2.
One patient conceived (cycle no. 6, Table I) and delivered
a twin pregnancy as a result of one of the DHEA-augmented
ovarian stimulation cycles; those patients who did not achieve
pregnancy returned to their previous normal menstrual cycle
patterns.
Discussion
In these five subjects, all of whom were 41 years old, had
normal FSH concentrations, yet had poor response to ovarian
stimulation, it was found that concurrent oral DHEA
supplementation improved gonadotrophin response by approximately two-fold. On the basis of this preliminary data, we
conclude that use of adjuvant DHEA in poor responders may
represent a novel way to maximize ovarian response in ovarian
stimulation, and reduce gonadotrophin dose.
The mechanism by which DHEA supplementation exerts
this gonadotrophin-augmentation effect is uncertain. We, and
others, have shown that DHEA supplementation enhances
serum free IGF-I concentrations by ~150%, probably independently of changes in GH secretion (Morales et al., 1994;
Diamond et al., 1996; Casson et al., 1998). Perhaps this
indicates that DHEA amplifies hepatic and end organ IGF-I
response to GH, which, in the milieu of the ovarian follicle,
may potentiate gonadotrophin action. Another possible mechanism by which DHEA exerts its effect is based on other work
(Haning et al., 1993) demonstrating that circulating DHEA-S
acts as a prohormone for much of ovarian follicular sex
steroidogenesis. In the five subjects reported here, baseline
DHEA-S concentrations were relatively low (122 51.5 g/
dl). DHEA supplementation may therefore provide a more
readily available pool of ovarian steroidogenic prohormone,
facilitating follicular function and growth.
The literature regarding DHEAS, DHEA, and ovulation
induction is scant. In natural cycles, endogenous DHEAS
concentrations did not vary between luteal phases in nonconception and conception cycles (Castracane et al., 1998).
Dexamethasone suppression of elevated endogenous DHEAS
concentrations did not improve outcome in IVF (Rein et al.,
1996), but the addition of suppressive doses of this drug did
improve outcome in clomiphene-resistant ovulatory subjects

(Trott et al., 1996). The interplay between adrenal androgens

and ovarian function/stimulation is clearly complex.
Supplementation of this steroid may only be beneficial in
certain subgroups, such as the subjects in this case series.
The results of this study must be considered preliminary.
First, it is possible that oral DHEA administration, which we
have previously demonstrated is extensively metabolized to
the down stream androgenic steroids, may very well result in
production of a metabolite that cross reacts with the oestrogen
assay used and artificially elevates serum oestradiol concentrations. However, the baseline oestradiol concentrations in these
subjects were not elevated, even during concurrent DHEA
supplementation. Additionally, in other studies of administration of this dose range of oral DHEA in postmenopausal
women, circulating oestradiol concentrations do not appear to
be increased (Casson et al., 1998).
The second contentious issue arises from the different
gonadotrophin preparations used in the historical control cycles
(mainly i.m. HMG) and in the study cycles (mainly i.m. rFSH).
Perhaps the rFSH, by virtue of its greater purity, is more
potent. However, both preparations are designed to be bioequivalent; the recombinant product simply has much less protein
(Hendon et al., 1995; Follistim, package insert). Also, the
gonadotrophin used in the subject with the most dramatic
response (patient no. 2; purified FSH) was the same in the
control and DHEA cycle. However, it may be that nonrecombinant preparations of HMG contain some inhibitory
substance that may worsen ovarian response, compared to
rFSH. Clearly, a randomized controlled trial would address
these questions.
DHEA does appear to augment ovulation induction in poor
responders, particularly patients who are aged 3540 years
and have normal FSH concentrations. This effect may have
great clinical potential. Not only would it allow for successful
ovulation induction in patients with previous poor response,
but it may, in normal patients, allow for dose reduction of
gonadotrophin. The effect clearly bears further investigation.

Acknowledgements
We thank Organon for their support and supply of Follistim for this
study, and Belmar Pharmacy for supplying the DHEA tablets.

References
Adashi, E.Y., Resnick, C.E., Hernandez, E.R. et al. (1991) Insulin-like growth
factor I as an intra-ovarian regulator: basic and clinical implications. Ann.
N. Y. Acad. Sci., 626, 161167.
Buster, J.E., Casson, P.R., Straughn, A.B. et al. (1992) Postmenopausal steroid
replacement with micronized dehydroepiandrosterone: preliminary oral
bioavailability and dose proportionality studies. Am. J. Obstet. Gynecol.,
166, 11631170.
Casson, P.R., Carson S.A., Buster J.E. et al. (1998a) Replacement
dehydroepiandrosterone in elderly: rationale and prospects for the future.
Endocrinologist, 8, 187194.
Casson, P.R., Santoro, N.F., Elkind-Hirsch, K. et al. (1998b) Postmenopausal
dehydroepiandrosterone (DHEA) administration increases free insulin-like
growth factor-I (IGF-I) and decreases high density lipoprotein (HDL): a
six month trial. Fertil. Steril., 70, 107110.

2131

P.R.Casson et al.
Castracane, V.D., Stewart, D.R., Gimpel, T. et al. (1998) Maternal serum
androgens in human pregnancy: early increases within the cycles of
conception. Hum. Reprod., 13, 460464.
Check, J.H., Nowroozi, K., Chase, J.S. et al. (1990) Ovulation induction and
pregnancies in 100 consecutive women with hypergonadotropic amenorrhea.
Fertil. Steril., 70, 107110.
Diamond, P., Cusan, L., Gomez, J.L. et al. (1996) Metabolic effects of
12-month percutaneous dehydroepiandrosterone replacement therapy in
postmenopausal women. J. Endocrinol., 150, 543550.
Haning, R.V. Jr., Hackett, R.J., Flood, C.A. et al. (1993) Plasma
dehydroepiandrosterone sulfate serves as a prehormone for 48% of follicular
fluid testosterone during treatment with menotropins. J. Clin. Endocrinol.
Metab., 76, 13011307.
Hedon, B., Out, J.H., Hughes, J.N. et al. (1995) Efficacy and safety of
recombinant follicle stimulating hormone (Puregon) in infertile women
pituitary-suppressed with triptorelin undergoing in-vitro fertilization: a
prospective, randomized, assessor-blind, multicentre trial. Hum. Reprod.,
10, 31022106.
Hofmann, G.E., Toner, J.P., Muasher, S.J. et al. (1993) High-dose follicle
stimulating hormone (FSH) ovarian stimulation in low responder patients
for in-vitro fertilization J. In Vitro Fert. Embryo Trans., 6, 285289.
Homburg, R., Eshel, A., Abdulla, H.I. et al. (1988) Growth hormone facilitates
ovulation induction by gonadotrophins. Clin. Endocrinol., 29, 113115.
Hugues, J. and Durnerin, I. (1998) Revisiting gonadotrophin-releasing hormone
agonist protocols and management of poor ovarian responses to
gonadotrophins. Hum. Reprod. Update, 4, 83101.
Ibrahim, Z.H., Lieberman, B.A., Matson, P.L. et al. (1991) The use of
biosynthetic growth hormone to augment ovulation induction with buserelin
acetate/human menopausal gonadotrophin during ovarian stimulation for
in vitro fertilization in women with a poor ovarian response. Fertil. Steril.,
55, 202205.
Morales, A.J., Nolan, J.J., Nelson, J.C. et al. (1994) Effects of replacement
dose of dehydroepiandrosterone in men and women of advancing age. Clin.
Endocrinol. Metab., 78, 13601367.
Orentreich, N., Brind, J.L., Rizer, R.L. et al. (1984) Age changes and
sex differences in serum dehydroepiandrosterone sulfate concentrations
throughout adulthood. J. Clin. Endocrinol. Metab., 59, 551555.
Padilla, S.L., Dugan, K., Maruschak, V. et al. (1996) Use of the flare-up
protocol with high dose human follicle stimulating hormone and human
menopausal gonadotrophins for in vitro fertilization in poor responders.
Fertil. Steril., 65, 796799.
Rein, M., Jackson, K., Sable, D. et al. (1996) Dexamethasone during ovulation
induction for in-vitro fertilization: a pilot study. Hum. Reprod., 11, 253255.
Trott, E., Plouffe, L., Hansen, K. et al. (1996) Ovulation induction in
clomiphene-resistant anovulatory women with normal dehydroepiandrosterone sulfate levels: beneficial effects of the addition of
dexamethasone during the follicular phase. Fertil. Steril., 66, 484486.
Received on February 8, 1999; accepted on June 7, 2000

2132