Acta Obstet Gynecol Scand 62:397 -401, 1983
A CLINICAL EVALUATION OF TREATMENT WITH ESTRIOL VAGINAL CREAM
VERSUS SUPPOSITORY IN POSTMENOPAUSAL WOMEN
Lars-Ake Mattsson and Goran Cullberg
From the Department of Obstetrics and Gynecology, University of Gbteborg, t)stra sjukhuset, GBteborg, Sweden
Abstract. Thirty postmenopausal women with vaginal
atrophy were treated with a vaginal cream or vaginal sup-
positories containing 0.5 mg estriol (OvestinR, Organon,
The Netherlands). The preparations were given daily for 14
days and then during the next 6 weeks a maintenance dose
was applied twice a week.
Cervical mucus, vaginal cytology and the endometrium
were studied. Blood samples were analysed for unconjugat-
ed and conjugated estriol, FSH, LH, prolactin and sex hor-
mone binding globulin capacity, before and after 2 weeks of
treatment. Subjective relief of vaginal symptoms was
reported by all patients. Restoration of vaginal mucosa was
confirmed by colposcopy. Some women noted a feeling of
transient "vaginal heat" during the first days of treatment;
otherwise no side effects were reported.
Higher values for Spinnbarkeit, ferning and karyopyk-
notic index were found for the cream group on day 14. No
endometrial stimulation was detected as judged by light mi-
croscopy in any of the samples. No significant differences
between the preparations were found for any of the hor-
mone variables, or sex hormone binding globulin capacity,
except for prolactin levels after 2 weeks of treatment. Possi-
ble reasons for the disparity between the two preparations
regarding peripheral variables are discussed.
After the menopause about 20% of women (8) com-
plain of vaginal atrophy and related problems such as
itching and dyspareunia due to estrogen deficiency.
The need for estrogen therapy is obvious but the risk
of developing endometrial hyperplasia or malignancy
must be considered. Estriol (E3) administered orally is
an effective treatment for the vaginal symptoms men-
tioned earlier (2).
Administered as a single oral low daily dose, estriol
has been reported to cause negligible proliferation in
the endometrium (2, 22). However, it has been re-
ported that daily divided doses for 2-3.5 months
may stimulate the endometrium ( 5 ) . Estrogens are ef-
fectively absorbed after vaginal application (8, 15, 18)
and have been widely used in the treatment of condi-
tions related to vaginal atrophy. Sustained and higher
serum levels of estrogens have been reported follow-
ing vaginal versus oral administration (15, 18).
Recent clinical studies with estriol cream and/or
suppositories (8, 21) have shown a sharp rise in
unconjugated E3 followed by a gradual decline over
an observation period of up to 8 hours, which is in
agreement with a previous pharmacological study by
Schiff et al. (19). It has also been claimed that the me-
dium used for vaginal preparations could be import-
ant for the absorption rate and clinical efficacy (15).
The present study was undertaken to compare clin-
ical, morphological and endocrinologicai effects of
estriol administered as a vaginal cream or as supposi-
tories in postmenopausal women.
PATIENTS AND METHODS
Thirty postmenopausal patients presenting with symptoms
of vaginal atrophy such as dyspareunia and vaginal dryness
volunteered after giving informed consent to take part in
this study. They were otherwise healthy and had not been
treated with estrogenic preparations for at least 6 months
prior to the study. The women were randomly allocated to
one of the two treatment groups receiving daily doses of 0.5
mg estriol administered either as a vaginal cream (group A)
or as a vaginal suppository (group B) (OvestinR cream or
suppositories, Organon International B.V., Oss, The Neth-
erlands). The preparations were given daily for 14 days and
during the following 6 weeks the same dose was given twice
a week. Mean age for group A was 59.5 years (range
52 - 72) and for group B 58 years (range 50- 70). The mean
intervals since the last menstrual bleeding were 6.6 years
(range 3 - 27) and 8 years (range 3 - 20) respectively for the
two groups.
The following clinical variables were studied before and
after 2 weeks of treatment:
1. General physical status including blood pressure and
body weight.
2. Cervical mucus properties such as Spinnbarkeit (cm) and
ferning, scored as 0, I , 2 or 3
3. Vaginal cytology expressed as karyopyknotic index @er-
centage of superficial cells)
4. Colour colpophotography
5 . Endometrial biopsies
Acla Obstet Gynecol Scand 62 (1983)
398 L.-A. Mattsson and G . Cullberg

In addition, parameters 1 and 2 were studied after 5 and 8

weeks of treatment.
Spinnbarkeit was assessed by taking cervical mucus with
ring forceps and by separating the arms of the forceps a
thread was formed and measured. Ferning was judged mi-
croscopically after cervical mucus had been left to crystal-
lize on a microscope slide. Vaginal cytology specimens were
obtained from the lateral fornix and the smears were stained
according to the Papanicolaou technique. Endometrial
biopsies were taken in 4 patients from each group from the
fundal region by means of a Randall curette under local
anesthesia, using paracervical blockade (Xylocain 1 Vo with-
out epinephrine).
:
01

Blood was drawn from an antecubital vein before and
after 14 days of treatment about 20 hours after the last vagi-
nal cream or suppository application. The blood was then
centrifuged and after separation serum was frozen at
- 26°C until analysed.
Radio-immunoassay methods were used to measure un-
conjugated and conjugated estriol (17), FSH and LH (3)
and estrone (14). Prolactin was determined by a homolog-
ous double antibody radio-immunoprecipitation technique
(13) and sex hormone binding globulin capacity (SHBG)
was measured ad modum Dennis et al. (4).
One patient in each group withdrew during the first 2
weeks of treatment due to a feeling of ”vaginal heat”. In
the suppository group one blood sample was lost before
analysis.
STATISTICS
To compare results within treatment groups, Student’s
paired t-test was performed. Differences between the two
groups were judged by the non-paired Student’s t-test. Ob-
servations below the limit of detection of the assay for un-
conjugated estriol (0.05 nmol/l) were ascribed the value
0.04 nmol/l. A p-value <0.05 was considered statistically
significant.
RESULTS
Subjective relief of vaginal symptoms such as dryness
and dyspareunia was experienced by all patients
completing the study. In 3 women treated with
cream, colpitis was still present after 2 weeks of treat-
ment and they were then in addition treated with con-
ventional antibiotics. In one of these women, diabet-
es mellitus was discovered and she was later referred
to a physician. In 4 cases subjective symptoms of
urethritis were present before treatment. These symp-
toms disappeared during treatment and in one case of
craurosis vulvae, subjective relief of itching was regis-
tered. Three women in the cream group and 2 women
in the suppository group noted a feeling of transient
”vaginal heat” but otherwise no side effects were
recorded by the patients. No changes in body weight
or blood pressure were noted during treatment.
5 1
I I 1
0 2 5 8
WEEKS
Fig. 1. Ferning and Spinnbarkeit (mean+SE) in postmeno-
pausal women treated with vaginal estriol cream (open
circles) and estriol suppositories (filled circles) on day four-
teen. Asterisks indicate differences between the two groups.
* = p <0.05, ** = p <0.01, *** = p <0.001
Cervical mucus. Induced changes in ferning and
Spinnbarkeit for the two groups are shown in Fig. 1.
A greater Spinnbarkeit for the cream group compar-
ed with the suppository group was noted after 2
weeks of treatment (p <0.05) and was even more ac-
centuated after 5 and 8 weeks (p<O.Ol). A similar
difference was seen in ferning (p <0.01) after 5 and 8
weeks.
Vaginal cytology. The figures for karyopyknotic in-
dex (KI) before and after 2 weeks of treatment in
group A were 0.5 and 43.4; the corresponding figures
for group B were 0.5 and 16.5. The difference in KI
after treatment was significant (p <0.01).
Histopathology. Endometrial biopsy samples prior to
treatment were atrophic and after 2 weeks of treat-
ment, curettage was repeated. Two patients in the
cream group (the other 2 did not collaborate) and 4
patients in the suppository group had an atrophic
inactive endometrium, as judged by light microscopy.

Acra Obsrel Gynecol Scand 62 (1983)

 Estriol cream vs. suppositoriesfor vaginal application 399

Fig. 2. Colpophotography of
the uterine cervix in one pa-
tient before and after 14 days
of treatment with a daily
dose of 0.5 mg estriol cream
for vaginal application.

Colposcopy. Colposcopy and photographs showed W

disappearance of surface capillaries, increased thick-
3 500-
P E 3, conjugated
ness of surface vaginal epithelium and reduced ten- -
dency to contact bleeding, which confirmed the sub- a
-
c
jective relief of symptoms expressed by all patients 300-
(Fig. 2). I
0
w -
Hormone variables and SHBG. Data on unconjugat- LL
E 3. unconjugated
ed and conjugated estriol, gonadotropins, prolactin, g 100 -
estrone and SHBG are set out in Table 1. The serum
concentration of unconjugated estriol was not detect-
2V
able in any of the patients before treatment. In 5 out I-
z
@ FSH
of 14 patients in group A and in 4 out of 13 patients -50-
IY
in group B, the unconjugated estriol level was below
the detection limit 20 hours after the last dose was
taken on day 14. A higher concentration, though not Fig. 3. Mean percentage change2SE of the pretreatment
concentration of FSH, unconjugated and conjugated estriol
statistically significant, of unconjugated estriol for after 14 days of treatment with 0.5 mg estriol per day, given
group A vis-a-vis group B was found after 2 weeks of either as a vaginal cream (open circles) or suppository (filled
treatment. circles) in postmenopausal women.

Table I. Serum hormone levels and SHBG (mean-cSE)in postmenopausal women before and after 2 weeks of
treatment with 0.5 mg estriol (Ed per day either as a cream (group A ) or as a vaginal suppository (group B).
Group A (n = 14) Group B (n = 13)
Pretreatment Treatment Pretreatment Treatment
~~~ ~ ~~

FSH, ug/l 14.2k0.4 13.5k0.4" 14.9k0.7 13.7k0.6'

LH, ug/l 9.220.5 8.O+O. 6 10.320.9 11.621 .O
E, unconj., nmol/l N.D. 0.23k0.08"' N.D. 0.1 lk0.02
E, conj., nmol/l 0.28k0.01 1.48k0.2"' 0.33+0.05 1.57kO.2"'
Prolactin, ug/l 9.2kO. 9 12.4k1.7' 13.7k1.6 17.1k2.3
SHBG, nmol/l 63.0k6.8 60.5k6.5 80.7k6.4 69.7k6.7
Estrone, nmol/l 0.1420.01 0.13-cO.01 0.1220.01 0.13k0.01

N.D. =not detectable (<0.05 nmol/l). Asterisks indicate the difference between pretreatment and treatment levels. =p<O.OS, **=p<0.01,
*** =p<O.ool.
Acta Obstet Gynecol k a n d 62 (1983)
400 L.-A. Mattsson and G. Cullberg
Serum levels of conjugated estriol rose significantly effects observed. Rigg et al. (15) found that a cream
(p<O.OOl)but there was no difference between the medium appeared to retard vaginal absorption of mi-
two formulations. A decrease in FSH was noted for cronized estradiol, compared with the same estrogen
both groups. The percentage changes in the pretreat- when suspended in saline. As the formulations used
ment concentrations of FSH, unconjugated and con- in the present study were previously shown to be ef-
jugated estriol are shown in Fig. 3. Increases in the fectively and equally absorbed (12)the media used
mean serum levels of prolactin were noted for both were probably not responsible for the discrepancy be-
groups. However the increase was significant tween the groups. It is uncertain to what extent the
(p <0.05) for group A only. volume of 0.5 and 2.5 ml for the cream and supposi-
tory respectively could be of importance for the ab-
sorption after some time of treatment. Furuhjelm et
DISCUSSION al. (6)reported a reduced vaginal absorption of con-
The absorption of vaginally applied estrogen prepara- jugated estrogens with a ripened mucosa vis-a-vis an
tions is well documented (8, 15, 18)and treatment of atrophic one and the same phenomenon has been
atrophy of the vaginal mucosa and related problems demonstrated for estriol cream (7). The different
by this route of administration is well established. modes of application for the cream and suppository
Since 1975 increased attention has been focused on might also explain the difference since the cream was
estrogen therapy and its possible connection with ma- administered by a specially calibrated applicator,
lignancy of the endometrium (11, 20). Vaginally ap- whereas the suppository was applied manually.
plied conjugated estrogens and estrone sulfate have It could be speculated that it may be difficult for
been shown to produce endometrial proliferation (10, menopausal women to properly apply the supposi-
24), whereas estriol for 16 weeks did not stimulate the tory manually at the top of the vagina. Thus the dif-
endometrium (8). ference in blood flow for different parts of the vagina
The present study of estriol as a vaginal cream or as could influence the result obtained. In the earlier ab-
a suppository confirmed previous reports on allevia- sorption study the suppository was applied by the in-
tion of local symptoms associated with vaginal vestigator. The importance of careful instructions to
atrophy (8, 10,21). The endometrium was not stimu- the patients should be emphasized and an applicator
lated after 2 weeks of treatment with a daily vaginal provided.
application, as judged by light microscopy. However, A reduction in FSH (but not for LH) was found on
at least for oral treatment, the application interval day 14,20 hours after application. It has been shown
and duration of treatment ( 5 ) might be of decisive im- previously (12)that LH was maximally depressed 5-6
portance for long-term estrogenic events such as hours after vaginal application of estriol and had "re-
endometrial proliferation. In the present study, turned" to baseline level within 24 hours. Higher pro-
clinical efficacy regarding subjective complaints was lactin levels after 2 weeks were found for the cream
maintained over an 8-week period with the same dose group but not for the suppository group. Exogenous
of estriol twice a week. It is very unlikely that this estrogens, e.g. high doses of ethinylestradiol, en-
maintenance dose, due to short receptor-binding (1, hance prolactin secretion (16). It is also well known
23), could produce endometrial stimulation even if that stress induced by venepuncture can give rise to
used for a longer period, as recently shown by KiCoviC misleading data on plasma prolactin levels (9).Estriol
et al. (8). Rapid absorption of estriol administered in- cream was earlier shown not to affect the function of
travaginally has been reported previously (7). the lactotroph either in a large number of patients
In an earlier study (12)we found equal serum levels treated for longer periods of time (8) or following
for both formulations used in this clinical experi- TRH stimulation (7). The present data are therefore
ment. However, significant differences for certain hard to evaluate.
parameters such as Spinnbarkeit, ferning and KI for A low dose (0.5 mg) of vaginally applied estriol had
the two formulations were found in the present study. beneficial effects on subjective complaints in the pre-
The mean concentration of unconjugated estriol - sent study. However, a transient feeling of "vaginal
which is the biologically active part - differed for the heat", probably due to altered vascularization, was
two groups. Because of a wide spread in vaginal ab- reported by some patients in either group. The pa-
sorption the difference was not statistically signifi- tients should be warned of this side effect so as to
cant, but this difference may explain the disparity in avoid unnecessary fear or treatment non-compliance.

Acta Obstet Gynecol Scand 62 (1983)

 Estriol cream vs. suppositoriesfor vaginal application
ACKNOWLEDGEMENTS
Thanks are due to Mr Kjeld Steffensen, Copenhagen for
valuable help in planning and evaluating the study and to
Mrs Berit Johansson for typing the manuscript.
REFERENCES
1. Bergink, E.W.: Oestriol receptor interactions: Their
biological importance and therapeutic implications.
Acta Endocrinol Suppl. 233, 94:9. 1980
2. Borglin, N.E.: Effect of oestriol on the female genital
tract. Acta Obstet Gynecol Scand 38:157, 1959
3. Burr, I.M., Grant, D.B., Sizonenko, P.C., Kaplan,
S.L. & Grumbach, M.M.: Some critical factors in dou-
ble antibody radioprecipitating sera for measurement
of human serum LH, FSH and HGH. J Clin En-
docrinol Metab 28:948, 1969
4. Dennis, M., Horst, H.J., Krieg, M. et al: Plasma sex
hormonebinding globulin binding capacity in benign
prostatic hypertrophy and prostatic carcinoma: Com-
parison with age dependent rise in normal human
males. Acta Endotrinol 84:207, 1977
5. Englund, D.E. & Johansson, E.D.B.: Endometrial ef-
fect of oral oestriol treatment in postmenopausal wom-
en. Acta Obstet Gynecol Scand 59:449, 1980
6. Furuhjelm, M., Karlgren, E. & CarlstrBm, K.: Intra-
vaginal administration of conjugated estrogens in pre-
menopausal and postmenopausal women. Int J Gynae-
col Obstet 17:335, 1980
7. Haspels, A.A., Luisi, M. & KiCoviC, P.M.: Endocri-
nological and clinical investigations in postmenopausal
women following administration of vaginal cream con-
taining oestriol. Maturitas 3:321, 1981
8. KitoviC, P.M., Cortes- Prieto, J., MilojeviC, S., Has-
pels, A.A. & Aljinovic, A.; The treatment of post-
menopausal vaginal atrophy with Ovestin vaginal
cream or suppositories: Clinical, endocrinological and
safety aspects. Maturitas 2:275, 1980
9. Koninckx, P.: Stress hyperprolactinaemia in clinical
practice. Lancet 1:273, 1978
10. Luisi, M., Franchi, F. & KiCoviC, P.M. Group com-
parative study of effects of Ovestin cream versus Pre-
marin cream in postmenopausal women with vaginal
atrophy. Maturitas 2:311, 1980
11. Mack, T.M., Pike, M.C., Henderson, B.E., Pfeffer,
R.I., Gerkins, V.R., Arthur, M. & Bronn, S.E.: Estro-
gens and endometrial cancer in a retirement communi-
ty. N Engl J Med 294:1262, 1976
12. Mattsson, L.A. & Cullberg, G.: Vaginal absorption of
two estriol preparations: A comparative study in post-
menopausal women. Acta Obstet Gynecol Scand
62:393, 1983.
40 1
13. McNeilly, A.S.: Radioimmunoassay of human prolac-
tin. Proceedinas of the Roval Societv of Medicine
I
66:863, 1973
14. Mikhail, G., Wu, C.H., Ferin, M. & Van de Wiele,
R.L.: Radioimmunoassav of olasma estrone and estra-
diol. Steroids 15:333, 19iO .
15. Rigg, L.A., Hermann, H. & Yen, S.S.C.: Absorption
of estrogens from vaginal creams. N Engl J Med
298:195, 1978
16. Robyn., C., Vekemans, M., Delvoye, P. & L’Hermite,
M.: Gonadotrophin and prolactin secretion during hor-
monal contraception. I n International Symposium on
Hormonal Contraception (ed. A.A. Haspels & C.W.
Kay), p. 98. Excerpta Medica, Amsterdam-Oxford,
1978
17. Rotti, K., Stevens, I., Watson, D. & Longcope, C.: Es-
trio1 concentrations in plasma of normal non-pregnant
women. Steroids 25:807, 1975
18. Schiff, I., Tulchinsky, D. & Ryan, K.J.: Vaginal ab-
sorption of estrone and 17 0-estradiol. Fertil Steril
28:1063, 1977
19. Schiff, I . , Wentworth, B., Koos, B., Ryan, K.J. &Tul-
chinsky, D.: Effect of estriol administration on the
hypogonadal woman. Fertil Steril 30:278, 1978
20. Smith, D.C., Prentice, R., Thompson, D.J. & Her-
mann, W.L.: Association of exogenous estrogen and
endometrial carcinoma. N Engl J Med 293:1167, 1975
21. Trevoux, R., Van der Velden, W.H.M. & PopoviC D.:
Ovestin vaginal cream and suppositories for the treat-
ment of menopausal vaginal atrophy. Reproduction
6:101, 1982
22. Tzingounis, V.V., Aksa, M.F. & Greenblatt, R.B.:
Oestriol in the management of the menopause. JAMA
293:1638, 1981
23. Van der Vies, J.: The pharmacology of oestriol. Matur-
itas 4:291, 1982
24. Widholm, 0. & Vartiainen, E.: The absorption of con-
jugated oestrogen and sodium oestrone sulphate from
the vagina. Ann Chir Gynaecol Fenn 63:186, 1974
Submitted for publication June 29, 1983
Accepted October I , 1983
Lars-Ake Mattsson, M.D.
Department of Obstetrics and Gynecology
Ostra sjukhuset
S-416 85 GBteborg, Sweden
Acta Obstet Gynecol Scand 62 (1983)