On the potential of metformin to prevent preterm delivery in women with polycystic ovary syndrome an epi-analysis

Transcription

1 A C TA Obstetricia et Gynecologica AOGS SHORT RESEARCH REPORT On the potential of metformin to prevent preterm delivery in women with polycystic ovary syndrome an epi-analysis ESZTER VANKY 1,2, FRANCIS DE ZEGHER 3,MARTADÍAZ 4,5, LOURDES IBÁÑEZ 4,5 & SVEN M. CARLSEN 6,7 1 Department of Obstetrics and Gynecology, St Olav s Hospital, Trondheim University Hospital, Trondheim, Norway, 2 Institute for Laboratory Medicine, Children s and Women s Health, Norwegian University of Science and Technology, Trondheim, Norway, 3 Pediatric Endocrinology, University Hospital Gasthuisberg, Leuven, Belgium, 4 Endocrinology Unit, Hospital Sant Joan de Déu, University of Barcelona, Esplugues, Barcelona, Spain, 5 Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas, Instituto de Salud Carlos III, Madrid, Spain, 6 Unit for Applied Clinical Research, Institute of Cancer Research and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway, and 7 Department of Endocrinology, St Olav s Hospital, Trondheim, University Hospital, Trondheim, Norway Key words Metformin, polycystic ovary syndrome, preterm delivery, miscarriage, pre-eclampsia, gestational diabetes, pregnancy Correspondence Eszter Vanky, Department of Obstetrics and Gynecology, St Olav s Hospital, University Hospital of Trondheim, Olav Kyrres gt 16, 7006 Trondheim, Norway. eszter.vanky@ntnu.no Conflicts of interest The authors have stated explicitly that there are no conflicts of interest in connection with this article. Please cite this article as: Vanky E, De Zegher F, Díaz M, Ibáñez L, Carlsen SM. On the potential of metformin to prevent preterm delivery in women with polycystic ovary syndrome an epi-analysis. Acta Obstet Gynecol Scand 2012;91: Abstract Our aim was to re-evaluate whether metformin may reduce late miscarriage/preterm delivery, pre-eclampsia and gestational diabetes in women with polycystic ovary syndrome (PCOS). We performed an epi-analysis of two randomized controlled trials. The participants were 313 women aged years with PCOS who had singleton pregnancies. They were randomized to metformin or placebo treatment from first trimester until delivery. We analysed the prevalence of late miscarriage/preterm delivery, pre-eclampsia and gestational diabetes according to both the intentionto-treat principle and per protocol analysis. The metformin-treated patients had less late miscarriage/preterm delivery; five (3%) vs. 18 (11%) in the placebo group (p < 0.01). There was no difference in the prevalence of gestational diabetes and pre-eclampsia between the metformin and the placebo group. In this epi-analysis, metformin treatment during pregnancy seems to reduce early delivery in women with PCOS. We believe that further randomized studies should be performed before firm conclusions can be drawn. GDM, gestational diabetes mellitus; PCOS, polycystic ovary syn- Abbreviations: drome. Received: 29 February 2012 Accepted: 13 September 2012 DOI: /aogs Introduction A meta-analysis and a population-based cohort study have recently confirmed that women with polycystic ovary syndrome (PCOS) are at higher risk for pregnancy complications (1,2). So far, the potential of metformin to reduce pregnancy complications in women with PCOS has been explored in only two randomized, placebo-controlled trials; first in a single-center pilot study (3) and then in the larger multicenter PregMet study (4). The smaller study suggested that metformin reduces serious pregnancy complications in women with PCOS (3), but this reduction was not corroborated by the larger PregMet study. In the PregMet study, we found no statistical difference in the prevalence of gestational diabetes, pre-eclampsia and preterm delivery between the metformin- and the placebo-treated groups (4). Metformin is an insulin-sensitizing drug that is a first choice to treat adolescents and women with PCOS (5). Its use is also recommended in insulin-resistant PCOS women 1460

2 E. Vanky et al. Metformin to reduce pregnancy complications (6), and there is evidence that metformin is beneficial as a pretreatment before in vitro fertilization in non-obese PCOS patients (7). The present epi-analysis was prompted by the observation of a tendency toward fewer untimely deliveries in both trials (3,4), and by the comment that each study, when analysed separately, may have insufficient power to detect metformin effects (8). Furthermore, it may be argued that there was no strong rationale for not pooling second-trimester miscarriage with preterm delivery into a primary end-point in the PregMet study (4). Indeed, the 22 week cut-off level between late miscarriage and extreme preterm birth is essentially set and used for legal, not biological purposes. Here we report an epi-analysis of the two randomized controlled trials, the endpoints being second-trimester miscarriage/preterm delivery, pre-eclampsia and gestational diabetes. Material and methods Both clinical studies have been described in detail (3,4). In short, the pooled results relate to 313 singleton pregnancies in women aged years with PCOS diagnosed by Rotterdam criteria (9). They were assigned in a double-blind manner to receive either metformin [850 mg twice daily (3) or 1000 mg twice daily (4)] or placebo twice daily. Compliance was defined prior to unblinding of the results. Good/acceptable compliance was defined as medication taken as intended or reduced to one or two tablets/day for a maximum of four weeks and/or no tablets for a maximum two weeks. Poor compliance was defined as less medication intake than described above. Second-trimester miscarriage was defined as pregnancy loss between and weeks. Preterm delivery was definedasdeliverybetweenweeks22 +0 and 37 +0,basedon mid-pregnancy ultrasound scan, performed between gestational weeks 17 and 20. Pre-eclampsia was diagnosed as blood pressure 140/90 mmhg measured on two occasions after gestational week 20, and albuminuria +2 dipstick on one occasion or +1 dipstick on two occasions. Gestational diabetes mellitus (GDM) was diagnosed as fasting plasma glucose > 7.0 mmol/l and/or 2 h serum glucose > 7.8 mmol/l after an oral glucose tolerance test (75 g glucose dissolved in 300 ml water). Here we present data according to both the intention-to-treat principle and per-protocol analysis. Statistical analysis The data were analysed according to the intention-to-treat principle using SPSS version 17.0 for Windows (SPSS Inc., Chicago, IL, USA). The differences between the study groups werecompared withtwo-tailedt-tests for independent samples for normally distributed variables. Non-parametric tests for independent samples were applied for variables with skewed distributions. Values are reported as means and standard deviations. Two-tailed P < 0.05 was considered significant. Categorical variables are presented as a frequency and percentage in each group. A chi-squared test was used to test differences between the groups. If the smallest expected value in a cell was less than five, we used Fisher s exact test. Results Baseline data were comparable in the two study groups (Table 1). By intention-to-treat analysis, 18 (11%) miscarriages or preterm births occurred in the placebo group, and five (3%) in the metformin group (p = 0.008; Table 2). There was no difference in the prevalence of pre-eclampsia or GDM between the metformin- and placebo-treated women. Onehundred and twenty-six (82%) women had good compliance in the metformin group and 133 (83%) had good compliance in the placebo group. Also by per-protocol analysis (Table 2), metformin intake was accompanied by a lower prevalence of late miscarriage or preterm delivery (p = 0.003). No difference in pre-eclampsia and GDM was observed between the study groups. The 14 earliest deliveries occurred in the subgroup not receiving the tested metformin dose (i.e. placebo or poor compliance), at gestational age (weeks + days); 17 +5,19 +6,21 +5,29 +1,30 +0,30 +4,30 +6,31 +0,32 +1,32 +2, 32 +6,34 +3 and Of note, none of metformin-compliant women gave birth before gestational week 35. Discussion We re-evaluated the potential of metformin to reduce the prevalence of pregnancy complications in women with PCOS. Firstly, we brought the data from our two randomized, placebo-controlled studies (3,4) together in an epi-analysis in order to gain statistical power. Secondly, we defined the three end-points as preterm deliveries/second-trimester miscarriages, pre-eclampsia and GDM. Thirdly, we performed per-protocol analyses and we also compared metformin-compliant vs. non-metformincompliant/placebo-treated women. We observed that none of the metformin-compliant women gave birth before week 35. Intake of the tested metformin dose (1700 or 2000 mg/day) was thus accompanied by a lower prevalence of early deliveries. Both meta-analyses and large epidemiological studies report an increased incidence of preterm births among PCOS pregnancies (2,10). The prevalence is estimated to be 2.2 times higher among women with PCOS compared with control women. The prevalence of singleton preterm birth in a Norwegian reference population is 6.4% (11), compared with 11% in the present study among placebo-treated women with PCOS. This harmonizes well with former findings (2,10). 1461

3 Metformin to reduce pregnancy complications E. Vanky et al. Table 1. Baseline characteristics of the participants according to intention to treat. Characteristic Metformin n = 153 a (SD) Placebo n = 160 b (SD) p-value Age (years) 29.5 (4.4) 29.1 (4.3) 0.45 Weight (kg) 83.5 (20.2) 79.7 (18.5) 0.08 Body mass index (kg/m 2 ) 29.8 (7.0) 28.6 (7.3) 0.13 Systolic blood pressure (mmhg) 119 (12) 117 (12) 0.32 Diastolic blood pressure (mmhg) 74 (12) 73 (10) 0.14 Heart rate (beats/min) 76 (10) 74 (10) 0.15 Fasting glucose (mmol/l) 4.60 (0.46) 4.66 (0.56) h glucose (mmol/l) 5.54 (1.51) 5.70 (1.66) 0.37 Gestational length at inclusion (days) 73 (14) 73 (14) 0.94 n = 153 (%) n = 160 (%) p-value GDM at inclusion (n) 11 (7.2) 17 (10.6) 0.33 Metformin use at conception (n) 47 (30.7) 47 (29.4) 0.81 Mode of conception (n) 0.34 Spontaneous 77 (57.0) 80 (58.0) Clomiphene citrate 40 (29.6) 31 (22.5) IVF/ICSI 18 (13.3) 22 (15.9) Others c 0(0) 5(3.6) Parity (n) (58.2) 89 (55.6) 1 52 (34.0) 54 (33.8) (7.8) 17 (10.6) Experienced spontaneous abortions (n) d 52 (34.0) 54 (33.8) 0.97 Smoking (n) 16 (10.5) 15 (9.4) 0.85 Caucasian (n) 148 (96.7) 158 (98.8) 0.27 Abbreviations: GDM, gestational diabetes mellitus; ICSI, intracytoplasmic sperm injection; and IVF, in vitro fertilization. a Eighteen patients from the pilot study and 135 patients from the PregMet study; in total, 153 women with polycystic ovary syndrome. b Tewnty-two patients from the pilot study and 135 patients plus one early miscarriage and two late miscarriages (these three patients could not be evaluated for pregnancy complications such as pre-eclampsia and preterm delivery and were therefore not included in the analysis in the original PregMet Study); in total, 160 women with polycystic ovary syndrome. c Bariatric surgery, acupuncture, major weight loss. d Number of those who had been pregnant earlier. We are not aware of any other prospective, placebocontrolled studies reporting on the effect of metformin on preterm birth. The present epi-analysis appears to be the first with sufficient statistical power to discern a significant effect of metformin on the prevalence of miscarriages and preterm deliveries. As our key findings are derived from an epi-analysis and a redefined end-point, they should be interpreted with due caution. In our opinion, the present results should not be handled as conclusive proof, but rather as a basis for further study of the preterm-birth-preventing potential of metformin, which is an old, safe and cheap drug. The mechanisms of a possible metformin effect to prevent preterm birth can only be speculated on. Utero-placental blood flow, androgen levels and markers of inflammation at critical time points in pregnancy are the factors that should be investigated. Also, an approach to explore genetic differences in metformin response may lead us closer to answers. It is difficult to explain why an antidiabetic drug, such as metformin, does not have any effect on the incidence of GDM in women with PCOS. We have no plausible explanation. The incidence of GDM in the present study was high (24%). We do not think it is a type 2 error. Former studies showing a positive effect of metformin on GDM were not randomized and had shortcomings in their design (12 14). Concerning pre-eclampsia, there is one large randomized controlled trial, the MIG study, where women with GDM were randomized to metformin or insulin. No difference was found in the prevalence of pre-eclampsia (15). In conclusion, it is too early to dismiss metformin treatment in PCOS pregnancies. The present epi-analysis strengthens the idea that metformin intake in pregnancy may reduce the risk of early delivery in women with PCOS, and supports further study of the potential of metformin in this field. Funding The Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology (NTNU) funded this study. 1462

4 E. Vanky et al. Metformin to reduce pregnancy complications Table 2. Epi-analysis of effects of metformin on pregnancy complications in women with polycystic ovary syndrome according to study groups. Intention-to-treat analyses Metformin Placebo 5/ /159 a Pre-eclampsia 12/ /157 b New gestational diabetes 27/142 c 19 26/141 c Per-protocol analyses Metformin Placebo Good compliance n = n = Poor compliance n = n = / /114 a Pre-eclampsia 11/ /123 b New gestational diabetes 23/94 c 24 23/92 c Metformin compliant n = 125 (40) Non-compliant or not on metformin n = 188 (60) 3/ /167 a Pre-eclampsia 11/ /177 b New gestational diabetes 23/94 c 24 30/137 c a One first-trimester miscarriage is not included in the analysis. b Three miscarriages (one first trimester and two late miscarriages) are not included the analysis. c Patients with 2 h oral glucose tolerance test indicating gestational diabetes mellitus at inclusion (before the start of medication) are not included in the analysis. Acknowledgments Metformin and placebo tablets were delivered, free of charge, by Weifa A/S, Oslo, Norway. Weifa had no involvement in the study. References 1. Kjerulff LE, Sanchez-Ramos L, Duffy D. Pregnancy outcomes in women with polycystic ovary syndrome: a metaanalysis. Am J Obstet Gynecol. 2011;204:558.e Roos N, Kieler H, Sahlin L, Ekman-Ordeberg G, Falconer H, Stephansson O. Risk of adverse pregnancy outcomes in women with polycystic ovary syndrome: population based cohort study. BMJ. 2011;343:d Vanky E, Salvesen KA, Heimstad R, Fougner KJ, Romundstad P, Carlsen SM. Metformin reduces pregnancy complications without affecting androgen levels in pregnant polycystic ovary syndrome women: results of a randomized study. Hum Reprod. 2004;19: VankyE,StridsklevS,HeimstadR,RomundstadP,SkogøyK, Kleggetveit O, et al. Metformin versus placebo from first trimester to delivery in polycystic ovary syndrome: a randomized, controlled multicenter study. J Clin Endocrinol Metab. 2010;95:E Ibáñez L, López-Bermejo A, Díaz M, Marcos MV, de Zegher F. Early metformin therapy (age 8 12 years) in girls with precocious pubarche to reduce hirsutism, androgen excess, and oligomenorrhea in adolescence. J Clin Endocrinol Metab. 2011;96:E Franks S. When should an insulin sensitizing agent be used in the treatment of polycystic ovary syndrome? Clin Endocrinol (Oxf). 2011;74: Kjotrod SB, Carlsen SM, Rasmussen PE, Holst-Larsen T, Mellembakken J, Thurin-Kjellberg A, et al. Use of metformin before and during assisted reproductive technology in non-obese young infertile women with polycystic ovary syndrome: a prospective, randomized, double-blind, multi-centre study. Hum Reprod. 2011;26:

5 Metformin to reduce pregnancy complications E. Vanky et al. 8. Legro RS. Metformin during pregnancy in polycystic ovary syndrome: another vitamin bites the dust. J Clin Endocrinol Metab. 2010;95: Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod. 2004;19: Boomsma CM, Fauser BC, Macklon NS. Pregnancy complications in women with polycystic ovary syndrome. Semin Reprod Med. 2008;26: Morken NH, Vogel I, Kallen K, Skjaerven R, Langhoff-Roos J, Kesmodel US, et al. Reference population for international comparisons and time trend surveillance of preterm delivery proportions in three countries. BMC Womens Health. 2008;8: Begum MR, Khanam NN, Quadir E, Ferdous J, Begum MS, Khan F, et al. Prevention of gestational diabetes mellitus by continuing metformin therapy throughout pregnancy in women with polycystic ovary syndrome. J Obstet Gynaecol Res. 2009;35: Glueck CJ, Wang P, Kobayashi S, Phillips H, Sieve-Smith L. Metformin therapy throughout pregnancy reduces the development of gestational diabetes in women with polycystic ovary syndrome. Fertil Steril. 2002;77: Glueck CJ, Pranikoff J, Aregawi D, Wang P. Prevention of gestational diabetes by metformin plus diet in patients with polycystic ovary syndrome. Fertil Steril. 2008;89: Rowan JA, Hague WM, Gao W, Battin MR, Moore MP;MiG Trial Investigators. Metformin versus insulin for the treatment of gestational diabetes. N Engl J Med. 2008;358: