| Metformin treatment for six months was significantly more effective than six-cycles of clomiphene citrate in improving fertility in non-obese anovulatory women with PCOS. This outcome occurred despite the fact that improvement in ovulation rates was similar between the two groups. Adverse event rates were similar. NNT of -3 indicates that for each 3 women treated, one less woman will remain infertile after six months of treatment. The relative risk of remaining anovulatory was reduced by approximately 80 percent. In the metformin-treated group, the cumulative rate rate of pregancy continued to increase with each subsequent cycle, whereas in the clomiphene group, it plateaued. This treatment strategy may be useful to primary care physicians wishing to help their PCOS patients achieve fertility, with possibly less risk of hyperstimulation of ovulation (risk of multiple gestation pregnancies.) Results from this study may not be extrapolatable to obese (BMI > 30) women with PCOS. |
Citation/s:
Prospective Parallel Randomized, double-blind, double-dummy controlled clinical trial comparing clomiphehe citrate and metformin as the first-line treatment for ovulation induction in nonobese anovulatory women with polycystic ovary syndrome. J Clin Endocrinol Metab 90:4068-4074, 2005.
Lead author's name and fax: Palomba, Stefano.
Three-part Clinical Question: In non-obese anovulatory women with polycystic ovary syndrome, is 6 month treatment with metformin as effective as 6 cycles of clomiphene citrate in improving fertility?
Search Terms: Journal Club presentation, November 2, 2005.
The Study:
Double-blinded concealed randomised controlled trial without intention-to-treat.
The Study Patients: Inclusion: Patients were 100 nonobese primary infertile anovulatory women with PCOS (dx made according to NIH criteria). Exclusion: age under 20, over 34; BMI over 30kg/Msquared; neoplastic, metabolic or cardiovascular concurrent illness, hypothyroid; cushing's syndrome; hyperprolactinemia; etc; no uterine bleeding after prgestin challenge; tubal or male infertility factor; organic pelvic disease
Control group (N = 50; 47 analysed): Double dummy; control group (group B) given clomiphene citrate 150 mg for 5 days starting on day 3 of progesterone-induced withdrawal bleed; and placebo tabs 2 tabs daily. Usual diet and exercise program.
Experimental group (N = 50; 45 analysed): Double dummy; experimental group (Group A) given metformin 850 mg bid; and 3 placebo tablets to be taken twice daily for 5 days starting on day 3 of progesterone-induced withdrawal bleed. Usual diet and exercise program. After 6 months, if still anovulatory, women received medications in a crossover fashion in the same doses as those described. (Crossover results were not included.) Ovulation was confirmed in both groups by transvaginal ultrasound and by serological confirmation
The Evidence:
| Outcome | Time to Outcome | CER | EER | RRR | ARR | NNT |
| Cumulative pregnancy rate | 6 months | 0.320 | 0.620 | -94% | -0.300 | -3 |
|---|---|---|---|---|---|---|
| 95% Confidence Intervals: | -152% to -35% | -0.487 to -0.113 | -9 to -2 | |||
| Post-six-month therapy persistent oligo- or amenorrhea | 0.320 | 0.060 | 81% | 0.260 | 4 | |
| 95% Confidence Intervals: | 36% to 100% | 0.115 to 0.405 | 2 to 9 | |||
Comments:
Limitations: small numbers of patients; 5 women in experimental group and 3 in control group were excluded after randomization for "missing first visit" or "poor compliance" or "significant body weight loss", so the study not Intention to Treat. Previous studies have looked at ovulation rates or return of menses, this study is one of few (if any others) that examines the POEM outcome of pregnancy. Metformin has not been shown to adversely affect the fetus in women who are taking it at the time they become pregnant. All women were also taking prenatal vitamins.
Appraised by: Linda Olson Douglas, MD
Journal Club led by Leah Schrupp, MD; Tuesday, November 01, 2005
Email: linda.douglas@phci.org
Kill or Update By: