Sie sind auf Seite 1von 7

ORIGINAL ARTICLE

Cervical Length and Androgens in Pregnant Women


With Polycystic Ovary Syndrome: Has Metformin
Any Effect?

Tone Shetelig Løvvik, Solhild Stridsklev, Sven M. Carlsen, Øyvind Salvesen,


and Eszter Vanky

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


Department of Obstetrics and Gynecology (T.S.L., S.S., E.V.) and Department of Endocrinology (S.M.C.),
St. Olav’s Hospital, Trondheim University Hospital, 7006 Trondheim, Norway; Department of Laboratory
Medicine, Children’s and Women’s Health (T.S.L., S.S., E.V.) and Department of Cancer Research and
Molecular Medicine (S.M.D.), Norwegian University of Science and Technology, 7491 Trondheim,
Norway; and Department of Public Health and General Practice (Ø.S.), 7491 Trondheim, Norway

Context: Women with polycystic ovary syndrome (PCOS) have increased risk of preterm delivery.
Shortening of the cervix is a sign of preterm delivery.

Objective: This study aimed to investigate potential effect of metformin on cervical length and
whether androgen levels correlate with cervical length in PCOS pregnancies.

Design and Setting: This was a sub-study of a randomized, placebo-controlled, multicenter study
(The PregMet study) performed at 11 secondary or tertiary centers from 2005 to 2009.

Participants: Two-hundred sixty-one pregnancies of 245 women with PCOS, age 18 – 42 years
participated.

Interventions: Participants were randomly assigned to metformin or placebo from first trimester
to delivery.

Outcome Measurements: We compared cervical length and androgen levels in metformin and
placebo groups at gestational weeks 19 and 32. We also explored whether cervical length corre-
lated with androgen levels.

Results: We found no difference in cervical length between the metformin and the placebo groups
at gestational week 19 and 32. Dehydroepiandrosterone (DHEAS) tended to be higher in the
metformin group. There were no correlations between androgens and cervical length at week 19.
At gestational week 32, androstenedione (P ! .02) and DHEAS (P ! .03) showed a trend toward
negative correlation to cervical length. High androstenedione level correlated with shortening of
cervical length from week 19 to 32 when adjusted for confounders (P ! .003). T (P ! .03), DHEAS
(P ! .02), and free testosterone index (P ! .03) showed a similar trend.

Conclusion: Metformin in pregnancy did not affect cervical length in women with PCOS. High
maternal androgen levels correlated with cervical shortening from the second to the third trimester
of pregnancy, as a sign of cervical ripening. (J Clin Endocrinol Metab 101: 2325–2331, 2016)

D espite extensive use of tocolytic agents, antibiotics,


and sick-leave, preterm delivery remains one of the
major challenges in obstetrics with effects on short- and
long-term health of the offspring. The prevalence of pre-
term deliveries varies between countries. Interestingly the
United States has a significantly higher prevalence of pre-

ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviations: BMI, body mass index; BP, blood pressure; CI, confidence interval; DHEAS,
Printed in USA dehydroepiandrosterone; FTI, free T index; IVF, in vitro fertilization; PCOS, polycystic ovary
Copyright © 2016 by the Endocrine Society syndrome; PregMet, Metformin Treatment in Pregnant PCOS Women Study.
Received September 17, 2015. Accepted January 28, 2016.
First Published Online February 2, 2016

doi: 10.1210/jc.2015-3498 J Clin Endocrinol Metab, June 2016, 101(6):2325–2331 press.endocrine.org/journal/jcem 2325
2326 Shetelig Løvvik et al Cervical Length and Androgens in PCOS Pregnancy J Clin Endocrinol Metab, June 2016, 101(6):2325–2331

term deliveries compared with other developed countries. inal ultrasound examinations or 2) on androgens, and 3)
An increase from 9% in 1982 to 12% in 2003 was re- to explore whether there was any correlation between an-
ported (1). In 2012 the rate of preterm birth in the United drogen levels and cervical length in PCOS pregnancies.
States decreased to 9.7% (2, 3). In Norway, approxi-
mately 6% of all deliveries occur before gestational week
37 (4). The prevalence seems to be slowly increasing in Materials and Methods
most Western countries.
Polycystic ovary syndrome (PCOS) is a common con- The PregMet study
The present study is a substudy of the Metformin Treatment
dition, and recent studies suggest a prevalence of 15% in
in Pregnant PCOS Women (PregMet) study, which was a pro-
a general population (5, 6) and up to 30% among women spective, randomized, double-blind, multicenter trial, compar-

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


who had experienced preterm delivery (7). They are also ing 2000 mg metformin daily with placebo from the first trimes-
more prone to suffer pregnancy complications than ter to delivery (18).
women without PCOS. Increased risk for preeclampsia, In short, the inclusion criteria were: 1) PCOS diagnosed be-
gestational diabetes mellitus, and preterm delivery is re- fore pregnancy according to The Rotterdam Criteria (25), 2) age
18 – 45 years, 3) gestational age between 5 and 12 weeks, and 4)
ported (8 –10). Women with PCOS have increased andro- a singleton viable fetus shown on ultrasonography. The exclu-
gen levels throughout life compared with controls (11) and sion criteria were alanine aminotransferase greater than 90 IU/L,
most of them are hyperandrogenic both in the nonpreg- serum creatinine concentration greater than 130 !mol/L, known
nant and pregnant state (5). Androgen levels in relation to alcohol abuse, previously diagnosed diabetes mellitus or fasting
cervix length are not described to any extent. In preg- serum glucose greater than 7.0 mmol/L at the time point of in-
clusion, treatment with oral glucocorticoids, or use of drugs
nancy, androgen levels tend to be higher than in the non-
known to interfere with metformin.
pregnant state (12). Androgen excess is only one of several In all, 273 pregnancies in 257 women were randomly as-
possible etiologic factors in preterm delivery and might be signed to either metformin or placebo treatment (16 women par-
a causal factor in cervix remodelling in women with ticipated twice). Randomization, blinding, and performed mea-
PCOS. Hyperandrogenicity is also associated with in- surements are described in detail elsewhere (18). All participants
received written and individual counseling on diet and lifestyle at
creased levels of inflammatory mediators (13) which may
inclusion. Thereafter, treatment with metformin 500 mg (met-
contribute to both shortening of the cervix and/or preterm formin hydrochloride, Metformin, Weifa AS) or identically
labor (14, 15). coated placebo tablets was initiated. The participants were in-
Metformin may have an effect on premature labor in structed to take one tablet twice daily during the first week and
PCOS pregnancies (16). As shortened cervix is a risk factor two tablets twice daily for the rest of the study period (ie, until
of preterm labor, a possible association between cervical delivery). Study medication was stopped at delivery.
Standardized interviewer-administered questionnaires were
length and metformin exposure is of importance to ex- used to obtain self-reported data on former medical and gynae-
plore. Pooling results from two randomized controlled cologic/obstetric history, ethnicity, employment, and smoking
trials [The Pilot study and the PregMet study (17, 18)] on habits. Biometric variables, including height, weight, and blood
metformin or placebo treatment in pregnant women with pressure (BP) were recorded. BP and heart rate were measured
PCOS, we found a decreased prevalence of preterm deliv- while the patient was in the sitting position after at least 10
minutes of comfortable rest in a chair. The BP was measured
ery in the metformin group compared with the placebo
three times, two minutes apart with digital devices. The mean of
group (16). Cervical length measured by ultrasound is the second and third measurements was calculated. Body weight
used to assess women with risk for preterm birth (19, 20). was recorded with light clothes on and without shoes.
Shortening of the cervix is a sign of cervical ripening and
is associated with increased risk of imminent preterm Ultrasound measurements
delivery. The length of the cervix was measured by ultrasound, in a
Hormonal regulation is an important but not fully un- longitudinal midsection plane, where the cervical channel is vi-
sualized. This was performed with a vaginal probe, without pres-
derstood mechanism of cervical remodelling (21). There is sure on the cervix. Funnelling was noted. The mean of three
increasing body of evidence that vaginal administration of measurements was registered. All ultrasound examinations were
progesterone decreases the prevalence of preterm delivery performed at the different study sites, with up-to-date ultrasound
in singleton pregnancies (22, 23). Also, alteration in es- equipment in clinical use present at each site. Gestational age was
trogen levels has been shown to affect late cervical fibro- determined by midpregnancy ultrasound, measuring biparietal
diameter, femur length, and mean abdominal diameter of the
blast function in vitro (21). There is also some evidence
fetus.
that androgens may play a role in regulating cervical func-
tion, but this has been little studied (24). Assays
The aims of this study were to investigate the potential Venous blood samples were drawn from an antecubital vein
effect of metformin 1) on cervical length measured by vag- between 0800 and 1100 hours after an overnight fast in both
doi: 10.1210/jc.2015-3498 press.endocrine.org/journal/jcem 2327

week 19 and week 32 of pregnancy. The specimens were cen- index (BMI), age, randomization to metformin/placebo, previ-
trifuged within 1 hour and stored at 5°C. Within 24 hours, the ous vaginal delivery, and cervix conisation. To adjust for mul-
specimens were frozen at "80°C. All blood samples were ana- tiple testing we chose a pragmatic approach and considered a P "
lyzed in 2009, after the last study participant had delivered. De- .01 as significant. P & .05 and ' .01 were considered as trends
hydroepiandrosterone (DHEAS) and SHBG were analyzed by toward significance. Results are given as #-coefficients with 95%
the ELISA technique with the reagents and calibrators supplied confidence intervals (CI), where CI not including zero suggests
by the manufacturer (DRG Instruments, GmbH). We used or- significance or a trend toward significance.
ganic solvent extraction (dichloromethane for T and ethyl ether
for androstenedione) prior to quantification to analyze serum T
and androstenedione. For quantification we used the ELISA
technique for T (DRG Instruments, GmbH) and Coat-A-Count Results
RIA kits (Diagnostic Products Corporation) for androstenedi-

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


one. The intra- and interassay coefficients of variation were 6.6 Baseline characteristics of the study population are shown
and 4.0% for DHEAS, 5.3 and 2.8% for androstenedione, and in Table 3. No differences between the metformin and
11.9 and 9.1% for T, respectively. Free testosterone index (FTI) placebo groups were observed.
was calculated; T (nmol/L) # 100/SHBG (nmol/L). Insulin was
measured by the ELISA technique using kits and reagents sup- Metformin effect on cervical length and hormone
plied by the manufacturer (DRG Instruments, GmbH). The in-
tra-and interassay coefficients of variation were 2.7 and 7.6%,
levels
respectively. We found no difference in cervix length between the
The study was approved both by the Regional Committees for metformin and placebo groups, either at gestational week
Medical and Health Research Ethics and The Norwegian Med- 19 or 32 (Table 2). Per-protocol analyses did not change
icines Agency. Written informed consent was obtained from each the results. Androgen and insulin levels at both gestational
patient before inclusion and the declaration of Helsinki was fol-
lowed throughout the study. The study was conducted according weeks 19 and 32, and gestational lengths were also similar
to principles of “Good Clinical Practice” and the trial is regis- in the metformin and placebo groups. When performing
tered at www.clinicaltrials.gov as trial number NCT00159536. per protocol analyses (Table 2) we found higher DHEAS
levels in the metformin group compared with the placebo
The present substudy group ("0.07 ["0.12; "0.02]; P ! .01) at gestational
In all, cervix measurement was performed in 261 of the preg- week 32.
nancies assessed in the PregMet study (pregnancies of 245
women), either in week 19 or week 32 of pregnancy, or both.
At gestational week 19, no correlations were found be-
These pregnancies comprise the study population in the present tween cervical length and androgen or insulin levels (Table
substudy. Measurements from both time points were available in 3). At gestational week 32 cervix length tended to correlate
236 women, 18 had their cervix measured only in week 19, and negatively with androstenedione ("0.2 [0.3; "0.03]; P !
seven only in week 32 of pregnancy. Of the 18 women who had .02) and DHEAS ("0.83 ["1.5; "0.1]; P ! .03), ie, the
their cervices measured only in week 19 of pregnancy, four
dropped out of the study and five had preterm deliveries before
higher androgen level, the shorter the cervix.
week 32. The remaining nine women met at the week-32 fol- At gestational week 32 androstenedione correlated
lowup but their cervixes were not measured. They were included positively with % cervix from week 19 to week 32 when
in the analysis, as were those who did not have their cervix mea- adjusted for cervical conisation, previous vaginal deliver-
sured in week 19, but in week 32 of pregnancy. Androgen anal- ies, randomization metformin/placebo, BMI, age, and for
ysis demands also available blood samples, thus leading to minor
variations in numbers included in each analysis (Tables 1 and 2).
interval between examinations (0.2 [0.05; 0.35]; P ! .01)
(Table 3). Equally adjusted analyses showed similar trends
Statistical analyses for DHEAS (1.01 [0.16; 1.88]; P ! .02), T (0.42 [0.04;
All statistical procedures were performed using the SPSS ver- 0.79]; P ! .03) and FTI (1.31 [0.11; 2.50]; P ! .03). We
sion 22.0 (SPSS, Inc). Observed data are reported as mean $ SD. found no correlation between insulin of SHBG and %
We used t test for independent samples to examine potential cervix.
differences between the metformin and placebo groups. To eval-
uate associations between cervix length and hormone levels in
gestational weeks 19 and 32 we used multivariate linear regres-
sion analysis, after testing the study material for normality and Discussion
assuming equal variances. Comparison of gestational length,
cervical length, androgens, and insulin levels in gestational weeks The most important findings of our study were 1) that
19 and 32 were done both as intention-to-treat and per protocol metformin had no effect on cervical length, and 2) that
analysis. We also explored a possible association between the shortening of the cervix (% cervix) from the second to the
change in cervical length between weeks 19 and 32 (cervical
length week 19 " cervical length week 32 ! % cervix) and hor-
third trimester correlated with increased levels of andro-
monal levels in week 32. In all, in our analysis we performed stenedione. Increased levels of DHEAS and FTI showed a
adjustments for the following possible confounders: body mass trend toward correlation to % cervix. Metformin exposure
2328 Shetelig Løvvik et al Cervical Length and Androgens in PCOS Pregnancy J Clin Endocrinol Metab, June 2016, 101(6):2325–2331

Table 1. Differences in Cervical Length, Gestational Age at Delivery, Androgen and Insulin Levels in Pregnant PCOS
Women According to Metformin and Placebo Treatment
Metformin, Placebo, Crude Adjusted
Mean (SD) Mean (SD) P-Value ! (95% CI)b P-Valueb
Intention to treat analysis
n ! 129 n ! 125
Cervix wk 19, mm 42 (6) 43 (8) .73 0.002 ("0.01; 0.01) .67
n ! 126 n ! 117
Cervix wk 32, mm 37 (8) 36 (8) .70 "0.01 ("0.01; 0.01) .81
n ! 131 n ! 130
Gestational length, d 278 (10) 276 (15) .23 "0.003 ("0.01; 0.02) .27

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


Gestational wk 19 n ! 123 n ! 128
Androstendione, nmol/L 10.1 (5.0) 10.9 (5.9) .20 0.01 ("0.01;0.02) .32
SHBG, nmol/L 297 (107) 320 (96) .08 &0.01 (&0.01; 0.001) .18
DHEAS, !mol/L 4.0 (1.9) 3.7 (1.8) .17 "0.02 ("0.06; 0.01) .18
T, nmol/L 3.9 (1.6) 4.3 (2.3) .11 0.03 ("0.01; 0.06) .12
FTI 1.4 (0.7) 1.4 (0.8) .99 0.01 ("0.07; 0.1) .79
Insulin, pmol/L 104 (76) 113 (71) .37 &0.01 (&0.01; 0.002) .14
Gestational wk 32 n ! 120 n ! 118
Androstendione, nmol/L 14.1 (7.9) 15.1 (9.3) .37 0.003 ("0.01; 0.01) .45
SHBG, nmol/L 345 (107) 367 (109) .13 &0.01 (&0.01; 0.001) .20
DHEAS, !mol/L 3.3 (1.6) 2.9 (1.4) .02 "0.05 ("0.09; 0.01) .02
T, nmol/L 4.7 (2.6) 5.3 (4.1) .19 0.01 ("0.01; 0.03) .19
FTI 1.5 (0.9) 1.5 (1.2) .72 0.01 ("0.05; 0.08) .65
Insulin, pmol/L 142 (78) 147 (78) .64 0.001 (&0.001; 0.002) .22
Per Protocol analysisb
n ! 97 n ! 97
Cervix wk 19 43 (7) 43 (8) .18 "0.002 ("0.012; 0.008) .65
n ! 96 n ! 91
Cervix wk 32 37 (8) 36 (8) .41 "0.005 ("0.15; 0.004) .27
n ! 99 n ! 101
Gestational length, d 278 (10) 274 (16) .08 "0.004 ("0.009; 0.001) .15
Gestational week 19 n ! 93 n ! 99
Androstendione, nmol/L 9.7 (5.1) 10.5 (5.4) .31 0.02 ("0.01; 0.06) .22
SHBG, nmol/L 299 (108) 314 (92) .30 0.005 ("0.01; 0.02) .52
DHEAS, !mol/L 4.1 (1.9) 3.6 (1.7) .10 &0.001 (&0.001; 0.001) .43
T, nmol/L 3.8 (1.6) 4.2 (2.2) .18 "0.03 ("0.08; 0.006) .10
FTI 1.4 (0.7) 1.4 (0.8) .73 0.02 ("0.8; 0.13) .64
Insulin, pmol/L 99 (79) 108 (60) .42 0.01 (&0.001; 0.002) .17
Gestational wk 32 n ! 94 n ! 92
Androstendione, nmol/L 13.1 (6.9) 14.8 (9.5) .17 0.006 ("0.003; 0.014) .22
SHBG, nmol/L 345 (112) 362 (111) .31 &0.001 (&0.001; 0.001) .43
DHEAS, !mol/L 3.3 (1.6) 2.8 (1.3) .01 "0.07 ("0.12; "0.02) .01
T, nmol/L 4.5 (2.2) 5.2 (4.1) .18 0.02 ("0.01; 0.04) .20
FTI 1.4 (0.7) 1.4 (0.8) .46 0.03 ("0.04; 0.1) .45
Insulin, pmol/L 133 (64) 143 (77) 0.31 0.001 (&0.001; 0.002) .11
a
Per protocol analysis, participant took medication as prescribed.
b
Adjusted for BMI, age, previous deliveries, and cervical conisation. P-values & .01 was considered significant and P & .05 and ' .01 were
considered borderline.

did not affect cervical length or androgen levels at gesta- gestational week 32. If metformin has a protective effect
tional week 19 or 32. on preterm deliveries it might be exerted through path-
This study comprised women from The PregMet study. ways other than androgen synthesis or shortening of the
An epianalysis of the Pilot Study and the PregMet study cervix. Metformin could modulate inflammation, by re-
showed a possible protective effect of metformin on the ducing inflammatory cell numbers or inflammatory mark-
combined end-point; preterm delivery and second trimes- ers. Metformin has been shown to regulate the endome-
ter miscarriages (16). Our study only evaluates cervical trial cell function directly (26). Another theoretical
length in week 19 and 32 of pregnancy. Most of the pre- mechanism could be lowering of contractility threshold of
term deliveries in the PregMet study were “late preterm” the myometrium: one study reports that metformin does
(ie, after gestational week 34), which may explain why we not exert effect on the spontaneous or oxytocin-induced
did not find any difference in absolute cervical length at myometrial contractions in an in vitro model (27).
doi: 10.1210/jc.2015-3498 press.endocrine.org/journal/jcem 2329

Table 2. Correlations Between Cervical Length or Change in Cervical Length, Androgen, and Insulin Levels in
Pregnant PCOS Women in Gestational Week 19 and 32
Crude Adjusted
Crude " (CI) P-Value Adjusted " (CI)a P-Valuea
Gestational wk 19, n # 245
T, nmol/L 0.05 ("0.42; 0.53) .82 0.29 ("0.18; 0.75) .23
Androstenedione, nmol/L "0.15 ("0.32; 0.02) .08 "0.05 ("0.22; 0.12) .57
DHEAS, !mol/L "0.31 ("0.82; 0.21) .24 "0.08 ("0.58; 0.42) .76
SHBG, nmol/L 0.002 ("0.01; 0.01) .68 0.002 ("0.01; 0.01) .73
FTI 0.20 ("1.04; 1.43) .76 0.77 ("0.44; 1.98) .21
Insulin, pmol/L 0.01 ("0.004; 0.02) .16 0.01 (&0.001; 0.03) .04

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


Gestational wk 32, n # 233
T, nmol/L "0.24 ("0.55; 0.08) .14 "0.2 ("0.5; 0.1) .25
Androstenedione, nmol/L "0.18 ("0.31; "0.06) .003 "0.2 ("0.3; "0.03) .02
DHEAS, !mol/L "0.91 ("1.61; "0.21) .01 "0.83 ("1.5; "0.1) .03
SHBG, nmol/L "0.002 ("0.12; 0.01) .69 "0.001 ("0.01; 0.01) .88
FTI "0.61 ("1.62; 0.40) .23 " 0.5 ( " 1.5 " 0.5) .33
Insulin, pmol/L 0.002 ("0.12; 0.02) .77 0.001 ("0.01; 0.02) .87
$ cervix, n # 227
T, nmol/L 0.35 ("0.02; 0.71) .07 0.42 (0.04; 0.79) .03
Androstenedione, nmol/L 0.17 (0.02; 0.31) .02 0.20 (0.05; 0.35) .01
DHEAS, !mol/L 0.77 ("0.06; 1.60) .07 1.01 (0.16; 1.88) .02
SHBG, nmol/L 0.001 ("0.01; 0.01) .89 &0.01 ("0.01; 0.01) .97
FTI 1.07 ("0.12; 2.25) .08 1.31 (0.11; 2.50) .03
Insulin, pmol/L "0.01 ("0.26; 0.01) .25 "0.003 ("0.02; 0.02) .72
Adjustments made for multiple testing, interpreting P & .01 as significant. P-values & 5 and ' .01 are considered borderline significant and are
considered as trends.
a
Adjusted for cervix conisation, previous vaginal deliveries, randomization metformin/placebo, BMI, age, and gestational age at examination (for %
cervix adjusted for interval between examinations).

In the third trimester of pregnancy, at week 32, high extracellular matrix composition and affect the mechan-
maternal androstenedione and DHEAS levels showed a ical strength of the cervix (12, 29).
tendency toward correlation to a short cervix. The cervix The idea of androgens affecting cervical ripening and
consists of fibroblasts, smooth muscle, epithelial, and im- by that causing preterm labor is supported by animal stud-
mune cells that secrete extracellular matrix (28). Andro- ies in which administration of supra-physiological doses
gens, estradiol, relaxin, and prostaglandin can alter the of androstenedione was injected in pregnant rhesus mon-

Table 3. Baseline Data of 261 Women With PCOS at Inclusion in First Trimester of Pregnancy
P
Alla Metformina Placeboa Valueb
n 261 131 130
Age, y 29.5 (4.4) 29.7 (4.3) 29.3 (4.4) .39
BMI, kg/m2 28.9 (6.8) 29.5 (7.1) 28.3 (6.4) .15
Systolic BP, mm Hg 118 (11.6) 119 (12.0) 117 (11.2) .39
Diastolic BP, mm Hg 73 (9.3) 73 (8.9) 72 (9.6) .52
Spontaneous pregnancy 151 [58] 75 [57] 76 [59] .84
Ovulation stimulation 68 [26] 39 [30] 29 [22] .17
IVF/ICSI 38 [15] 17 [13] 21 [16] .47
Smoking 20 [8] 14 [11] 6 [5] .07
Working/student 229 [88] 118 [90] 111 [85] .18
Unemployed 5 [2] 2 [2] 3 [2] .65
Cervical conisation 14 [5] 6 [5] 8 [6] .57
Parity
0 145 [56] 73 [56] 72 [55] .63
1( 116 [44] 58 [44] 58 [45] .96
Abbreviation: ICSI, .
a
Values given as mean (SD) or n [%] as appropriate.
b
t test independent samples, equality of means.
2330 Shetelig Løvvik et al Cervical Length and Androgens in PCOS Pregnancy J Clin Endocrinol Metab, June 2016, 101(6):2325–2331

keys causing preterm labor (30). This is in line with the may influence remodeling of the cervix, especially ripening
present observation that androstenedione correlated neg- of the cervix at term, and maybe also in preterm deliveries
atively with cervical length and change, whereas the other (31–34).
androgens investigated showed a tendency toward
correlation. Strengths and limitations
In other animal studies, administration of dihydrotes- Our study was performed in a setting with relatively
tosterone resulted in decreased cervical resistance, low prevalence of preterm labor (3, 4). If performed in a
whereas administration of the androgen receptor antag- setting with higher prevalence, such as in the United States
onist flutamide increased the resistance. Cervical resis- (2), it might have altered the results. This study is a sub-
tance was determined by the “cervical creep method,” an study of an randomized controlled trial, which might be a

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


experimental model to evaluate the effect of hormone ad- limitation as such. Also, the number of women included in
ministration on cervix fibers exposed to a certain load each arm of the study represents a limitation. We have
(24). Shortening of the cervix can be considered as a phys- performed multiple analyses, which increases the possi-
iological sign of decreased cervical resistance. bility of finding statistical associations by chance. By using
Human studies on DHEAS administration and cervix the pragmatic approach and setting our significance level
ripening, increased collagenase, or IL-8 activity, which at P ! .01 we adjusted for this. The strengths of our study
correlate with shorter time to delivery (31–34). These are that our findings are in line with former animal studies
studies were performed later in pregnancy than ours (ie, in and smaller clinical studies, and together they represent a
gestational week 38 – 41), they are mostly small, and rarely consistency. Few studies are done on PCOS in pregnancy
measure DHEAS levels, but investigate the effect of and, although being a substudy, the original study (18)
DHEAS injections late in normal pregnancies. One study was a randomized, placebo-controlled trial of good qual-
also reported that high DHEAS levels correlated with ity, ie, representing a strength per se (39).
higher Bishop score (more ripe cervix) in women with
spontaneous cervix ripening compared with women in Conclusion
need of induction of labor (35). Our results (Table 3) are Metformin does not seem to affect cervical length in preg-
in line with these findings as we found that increased nant women with PCOS. If metformin plays a role to pre-
DHEAS level tended to correlate with shortening of the vent preterm delivery, it is possibly by other mechanisms.
cervix, a marker of cervical ripening. DHEAS is a precur- In women with PCOS, shortening of the cervix in the latter
sor for most other androgen hormones and might explain half of the pregnancy, as a sign of cervical ripening, cor-
why other androgens are elevated too. Importantly, the related with elevated androgen levels.
circulating level of DHEAS is approximately 1000 times
higher than those of androstenedione and T. The role of
DHEAS in PCOS pregnancies has not been investigated to Acknowledgments
any extent. DHEAS, a precursor in the synthesis of an-
Address all correspondence and requests for reprints to: Tone Shet-
drogens, correlates with increased shortening of the cer-
elig Løvvik, Department of Obstetrics and Gynecology, St. Olav’s
vix. Why our participants showed higher DHEAS concen-
Hospital, University Hospital of Trondheim, Olav Kyrres gate 16,
tration in the metformin group than in the placebo group 7006 Trondheim, Norway. E-mail: Tone.Shetelig.Lovvik@stolav.no.
remains unclear. However, this observation is in line with This study was registered in ClinicalTrials.gov as trial number
at least one other study performed on women with PCOS NCT00159536.
and metformin treatment, where pre-in vitro fertilization This work was supported by the Norwegian University of
(IVF) treatment of patients with PCOS taking metformin Science and Technology, The Medical Faculty.
was shown to increase DHEAS levels while other andro- Disclosure Summary: The authors have nothing to disclose.
gens remained unchanged (36). DHEAS is converted to
androgens and estrogens intracellularly in the peripheral
tissues. Thus, they may exert their activity in the cells of References
origin (ie, intracrine effect) or on the surrounding cells (ie,
1. Raju TN, Higgins RD, Stark AR, Leveno KJ. Optimizing care and
paracrine effect) rather than appearing in the systemic cir- outcome for late-preterm (near-term) infants: A summary of the
culation (37, 38). Most androgens tend to increase with workshop sponsored by the National Institute of Child Health and
gestational age in the normal pregnancy. The reported Human Development. Pediatrics. 2006;118:1207–1214.
2. Duryea EL, McIntire DD, Leveno KJ. The rate of preterm birth in the
normal ranges are rather wide. Total T, free T, and SHBG
United States is affected by the method of gestational age assign-
increase from the second until the third trimester in normal ment. Am J Obstet Gynecol 2015;213:231 e231– e235.
pregnancy (12). It has been hypothesized that androgens 3. Blencowe H, Cousens S, Oestergaard MZ, et al. National, regional,
doi: 10.1210/jc.2015-3498 press.endocrine.org/journal/jcem 2331

and worldwide estimates of preterm birth rates in the year 2010 with 21. Timmons B, Akins M, Mahendroo M. Cervical remodeling during
time trends since 1990 for selected countries: A systematic analysis pregnancy and parturition. Trends Endocrinol Metab. 2010;21:
and implications. Lancet. 2012;379:2162–2172. 353–361.
4. Morken NH, Vogel I, Kallen K, et al. Reference population for 22. Hassan SS, Romero R, Vidyadhari D, et al. Vaginal progesterone
international comparisons and time trend surveillance of preterm reduces the rate of preterm birth in women with a sonographic short
delivery proportions in three countries. BMC Womens Health. cervix: A multicenter, randomized, double-blind, placebo-con-
2008;8:16. trolled trial. Ultrasound Obstet Gynecol. 2011;38:18 –31.
5. March WA, Moore VM, Willson KJ, Phillips DI, Norman RJ, Davies 23. Romero R, Yeo L, Chaemsaithong P, Chaiworapongsa T, Hassan
MJ. The prevalence of polycystic ovary syndrome in a community SS. Progesterone to prevent spontaneous preterm birth. Semin Fetal
sample assessed under contrasting diagnostic criteria. Hum Reprod. Neonatal Med. 2014;19:15–26.
2010;25:544 –551. 24. Ji H, Dailey TL, Long V, Chien EK. Androgen-regulated cervical
6. Tehrani FR, Simbar M, Tohidi M, Hosseinpanah F, Azizi F. The ripening: A structural, biomechanical, and molecular analysis. Am J
prevalence of polycystic ovary syndrome in a community sample of Obstet Gynecol 2008;198:543 e541– e549.

Downloaded from https://academic.oup.com/jcem/article-abstract/101/6/2325/2804744 by guest on 27 March 2019


Iranian population: Iranian PCOS prevalence study. Reprod Biol 25. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop
Endocrinol. 2011;9:39. Group. Revised 2003 consensus on diagnostic criteria and long-term
7. Eilertsen TB, Vanky E, Carlsen SM. Increased prevalence of diabetes health risks related to polycystic ovary syndrome. Fertil Steril. 2004;
and polycystic ovary syndrome in women with a history of preterm 81:19 –25.
birth: A case-control study. BJOG. 2012;119:266 –275. 26. Li X, Cui P, Jiang HY, et al. Reversing the reduced level of endo-
8. Boomsma CM, Fauser BC, Macklon NS. Pregnancy complications metrial GLUT4 expression in polycystic ovary syndrome: A mech-
in women with polycystic ovary syndrome. Semin Reprod Med. anistic study of metformin action. Am J Transl Res. 2015;7:574 –
2008;26:72– 84. 586.
9. Kjerulff LE, Sanchez-Ramos L, Duffy D. Pregnancy outcomes in 27. Hehir MP, Morrison JJ. Metformin and human uterine contractility.
women with polycystic ovary syndrome: A metaanalysis. Am J Ob- Endocrine. 2012;42:761–763.
stet Gynecol 2011;204:558 e551– e556. 28. Leppert PC. Anatomy and physiology of cervical ripening. Clin Ob-
10. Roos N, Kieler H, Sahlin L, Ekman-Ordeberg G, Falconer H, stet Gynecol. 1995;38:267–279.
29. Ghule VV, Gray C, Galimberti A, Anumba DO. Prostaglandin-in-
Stephansson O. Risk of adverse pregnancy outcomes in women with
duced cervical remodelling in humans in the first trimester is asso-
polycystic ovary syndrome: Population based cohort study. BMJ.
ciated with increased expression of specific tight junction, but not
2011;343:d6309.
gap junction proteins. J Transl Med. 2012;10:40.
11. Pinola P, Piltonen TT, Puurunen J, et al. Androgen profile through
30. Mecenas CA, Giussani DA, Owiny JR, et al. Production of prema-
life in women with polycystic ovary syndrome: A Nordic Multi-
ture delivery in pregnant rhesus monkeys by androstenedione infu-
center Collaboration Study. J Clin Endocrinol Metab. 2015;100:
sion. Nat Med. 1996;2:443– 448.
3400 –3407.
31. Kanayama N, El Maradny E, Goto J, Terao T. Effect of dehydro-
12. Makieva S, Saunders PT, Norman JE. Androgens in pregnancy:
epiandrosterone sulfate on interleukin-8 receptor during cervical
Roles in parturition. Hum Reprod Update. 2014;20:542–559.
ripening. Eur J Endocrinol. 1998;138:587–593.
13. González F, Sia CL, Shepard MK, Rote NS, Minium J. Inflammation
32. Takahashi K, Ikeno N, Hoshiai H, Suzuki M. The effect of dehy-
in response to glucose ingestion is independent of excess abdominal
droepiandrosterone sulfate on serum steroid concentration in sin-
adiposity in normal-weight women with polycystic ovary syndrome.
gleton and twin pregnancies and its effect on cervical canal-ripening
J Clin Endocrinol Metab. 2012;97:4071– 4079. in singleton pregnancies. Tohoku J Exp Med. 1984;142:289 –298.
14. Cappelletti M, Della Bella S, Ferrazzi E, Mavilio D, Divanovic S. 33. Mochizuki M, Honda T, Deguchi M, Morikawa H, Tojo S. A study
Inflammation and preterm birth. J Leukoc Biol. 2016;99(1):67–78. on the effect of dehydroepiandrosterone sulfate on so-called cervical
15. Vannuccini S, Clifton VL, Fraser IS, et al. Infertility and reproductive ripening. Acta Obstet Gynecol Scand. 1978;57:397– 401.
disorders: Impact of hormonal and inflammatory mechanisms on 34. Mochizuki M, Maruo T. Effect of dehydroepiandrosterone sulfate
pregnancy outcome. Hum Reprod Update. 2016;22(1):104 –115. on uterine cervical ripening in late pregnancy. Acta Physiol Hung.
16. Vanky E, DE Zegher F, Diaz M, Ibáñez L, Carlsen SM. On the 1985;65:267–274.
potential of metformin to prevent preterm delivery in women with 35. Koyuncu F, Aksoy S, Karanfil C, Demir N, Bozkurt K, Tuncay G.
polycystic ovary syndrome—An epi-analysis. Acta Obstet Gynecol [The effect of dehydroepiandrosterone sulfate on cervical matura-
Scand. 2012;91:1460 –1464. tion during pregnancy]. J Gynecol Obstet Biol Reprod (Paris). 1995;
17. Vanky E, Salvesen KA, Heimstad R, Fougner KJ, Romundstad P, 24:630 – 633.
Carlsen SM. Metformin reduces pregnancy complications without 36. Kjøtrød SB, Sunde A, von Düring V, Carlsen SM. Possible met-
affecting androgen levels in pregnant polycystic ovary syndrome formin effect on adrenal androgens during pretreatment and IVF
women: Results of a randomized study. Hum Reprod. 2004;19: cycle in women with polycystic ovary syndrome. Fertil Steril. 2009;
1734 –1740. 91:500 –508.
18. Vanky E, Stridsklev S, Heimstad R, et al. Metformin versus placebo 37. Labrie F, Luu-The V, Lin SX, et al. Intracrinology: Role of the family
from first trimester to delivery in polycystic ovary syndrome: A ran- of 17 beta-hydroxysteroid dehydrogenases in human physiology
domized, controlled multicenter study. J Clin Endocrinol Metab. and disease. J Mol Endocrinol. 2000;25:1–16.
2010;95:E448 –E455. 38. Labrie F, Luu-The V, Labrie C, Bélanger A, Simard J, Lin SX, Pel-
19. Crane JM, Hutchens D. Transvaginal sonographic measurement of letier G. Endocrine and intracrine sources of androgens in women:
cervical length to predict preterm birth in asymptomatic women at Inhibition of breast cancer and other roles of androgens and their
increased risk: A systematic review. Ultrasound Obstet Gynecol. precursor dehydroepiandrosterone. Endocr Rev. 2003;24:152–
2008;31:579 –587. 182.
20. Crane JM, Hutchens D. Follow-up cervical length in asymptomatic 39. Legro RS. Metformin during pregnancy in polycystic ovary syn-
high-risk women and the risk of spontaneous preterm birth. J Peri- drome: Another vitamin bites the dust. J Clin Endocrinol Metab.
natol. 2011;31:318 –323. 2010;95:5199 –5202.