Oral contraceptive use by teenage women does not affect peak bone mass: a longitudinal study.

OBJECTIVE: To determine the effect of oral contraceptive pill (OCP) use during adolescence on peak bone mass. DESIGN: Longitudinal observational study. SETTING: Academic clinical research center. PATIENT(S): Sixty-two non-Hispanic, white females in The Penn State Young Women's Health Study, who were studied for 8 years during ages 12-20. INTERVENTION(S): There were 28 OCP users, who used OCPs for a minimum of 6 months and were still using at age 20, and 34 nonusers who had never used OCPs. MAIN OUTCOME MEASURE(S): Total body bone, dedicated hip bone, and body composition measurements were made by dual-energy roentgenogram absorptiometry. RESULT(S): The OCP users and nonusers did not differ at entry in anthropometric, body composition, or total body bone measurements. By age 20, the average duration of OCP use by the user group was 22 months. At age 20, the groups remained indistinguishable in anthropometric, body composition, total body, and hip bone measures, and in age of menarche and sports exercise scores. CONCLUSION(S): Oral contraceptive pill use by healthy, white, teenage females does not affect acquisition of peak bone mass.

PIP: This longitudinal observational study determined the effect of oral contraceptive (OC) use during adolescence on peak bone mass (PBM). The sample comprised 62 non-Hispanic, White females in The Penn State Young Women's Health Study, who were studied for 8 years between the ages of 12 and 20. There were 28 OC users who used OCs for a minimum of 6 months and were still using them at age 20, and 34 nonusers who had never used the regimen. Total body bone, dedicated hipbone, and body composition measurements were made by dual-energy roentgenogram absorptiometry. There was no difference between OC users and nonusers in the anthropometric, body composition, or total body bone measurements. By age 20, the average duration of OC use by the user group was 22 months. At this age, the groups remained indistinguishable in anthropometric, body composition, total body, and hipbone measurements, and in age of menarche and sports exercise scores. These findings suggest that OC use by healthy, White, teenage females does not affect acquisition of PBM.

Collagen metabolism markers as a reflection of bone and soft tissue turnover during the menstrual cycle and oral contraceptive use.

Two different groups of women, 23 healthy young adults and 13 women with chronic posterior pelvic pain, were studied before and during use of oral contraceptives (OC). Collagen metabolism markers-here, the amino-terminal propeptide of type I procollagen, the carboxy-terminal telopeptide of type I collagen, and the amino-terminal of procollagen type III-as well as hormones and other endocrine factors indicating the balance between androgen expression/anabolism and catabolism of the subjects (testosterone, sex-hormone binding globulin, and insulin-like growth factor I were measured. Type I procollagen, the carboxy-terminal telopeptide of type I collagen, and the amino-terminal of procollagen type III were all significantly decreased during OC use. These findings implicate OC use-induced changes in collagen type I and III turnover. A shift in the anabolic/catabolic balance was also recorded indicating a less anabolic situation during OC use.

Evidence that treatment with monophasic oral contraceptive formulations containing ethinylestradiol plus gestodene reduces bone resorption in young women.

The aim of the study was to evaluate if a pill containing the same dose of the same type of progestin compound (gestodene, GES, 75 microg) but different doses of ethinylestradiol (EE2) (20 or 30 microg) differently influences specific markers of bone resorption (urinary levels of pyridinoline (PYR) and dexoxypyridinoline (D-PYR)). During the 12 months of the study a significant decrease of urinary levels of PYR and D-PYR was found in 2 groups of young post-adolescent women taking the pills with 20 and 30 microg of EE2 in comparison with control women (subjects of the same age group with normal menstrual cycle who did not use contraception). In women taking pills with 20 or 30 microg EE2, the levels of sex hormone-binding globulin (SHBG) significantly increased during treatment in comparison with baseline, whereas in the same time period no changes occurred in control women. These findings suggest that similar to the pill containing 30 microg EE2, the lower dosage of the EE2 pill (20 microg) is also capable of reducing bone resorption. Twenty and 30 microg EE2 pills exert the same biological estrogenic effect. In fact, SHBG levels significantly increased with both pills. However, an additional effect of the progestin compound that could act directly on progestin or estrogen receptors of bone cannot be excluded. Since contraception with a pill containing the lowest estrogen dose is associated with the lowest incidence of side effects, these findings further suggest a pill containing 20 microg EE2 for young post-adolescent women would be the best choice.

PIP: The aim of this study was to evaluate if a pill containing the same dose of the same type of progestin compound (gestodene, GES, 75 mcg) but different doses of ethinyl estradiol (EE2) (20 or 30 mcg) differently influences specific markers of bone resorption (urinary levels of pyridinoline (PYR) and dexoxypyridinoline (D-PYR). During the 12 months of the study a significant decrease of urinary levels of PYR and D-PYR was found in 2 groups of young postadolescent women taking the pills with 20 or 30 mcg EE2 in comparison with control women (subjects of the same age group with normal menstrual cycles who did not use contraception). In women taking pills with 20 or 30 mcg EE2, the levels of sex hormone-binding globulin (SHBG) significantly increased during treatment in comparison with baseline, whereas in the same time period no changes occurred in control women. These findings suggest that similar to the pill containing 30 mcg EE2, the lower dosage of the EE2 pill (20 mcg) is also capable of reducing bone resorption. 20 and 30 mcg EE2 pills exert the same biological estrogenic effect. In fact, SHBG levels significantly increased with both pills. However, an additional effect of the progestin compound that could act directly on progestin or estrogen receptors of bone cannot be excluded. Since contraception with a pill containing the lowest estrogen dose is associated with the lowest incidence of side effects, these findings further suggest that a pill containing 20 mcg EE2 would be the best choice for young postadolescent women.

Low-dose oral contraceptives and bone mineral density: an evidence-based analysis.

We reviewed studies of the association of oral contraceptive (OC) use and bone mineral density (BMD). We limited the review to studies of women using low-dose oral contraceptives and that measured BMD by bone densitometry. A total of 13 studies met the inclusion criteria. Nine of these showed a positive effect of OC use on BMD, and four did not show an association. However, none of the studies showed a decrease in BMD with OC use. We classified the level of evidence from each study according to the guidelines of the US Preventive Services Task Force. The level of evidence supporting a positive association between OC use and increased BMD is II-1. There is fair evidence (Category B) to support the position that OC use has a favorable effect on BMD. We made suggestions for a study design that could yield Level I evidence.

PIP: The authors reviewed studies of the association of oral contraceptive (OC) use and bone mineral density (BMD). They limited the review to studies of women using low-dose OCs and that measured BMD by bone densitometry. A total of 13 studies met the inclusion criteria; 9 of these showed a positive effect of OC use on BMD, and 4 did not show an association. However, none of the studies showed a decrease in BMD with OC use. They classified the level of evidence from each study according to the guidelines of the US Preventive Services Task Force. There is evidence supporting a positive association between OC use and increased BMD. There is fair evidence to support the position that OC use has a favorable effect on BMD. The authors made suggestions for a study design that could yield level I evidence.

Oral contraceptives and bone mineral density: A population-based study.

OBJECTIVE: We sought to test the hypothesis that exposure to oral contraceptives protects the skeleton. STUDY DESIGN: Multiple regression techniques were used to analyze data for a random sample of 710 Australian women (age range, 20-69 years). Bone mineral density was measured at the lumbar spine, proximal femur, whole body, and distal forearm. Oral contraceptive exposure was assessed by a questionnaire. RESULTS: Women exposed to oral contraceptives had a 3.3% greater mean bone mineral density adjusted for body mass index and age at the lumbar spine (partial r (2) = 0.009; P =.014). Adjusted mean vertebral bone mineral density was 3.3% greater for premenopausal women (partial r (2) = 0.008; P <.05), but the effect did not reach significance among postmenopausal women. Higher bone mineral density was associated with increased duration of exposure, with a mean increase of 3.2% associated with the first 5 years and a further 0.2% with >/=5 years of exposure. No association was detected at other sites. CONCLUSION: Exposure to oral contraceptives may be associated with higher lumbar spine bone mineral density.

Bone mineral density of the spine and femur in healthy Saudi females: relation to vitamin D status, pregnancy, and lactation.

Bone mineral density (BMD) measurements of the anterio-posterior lumbar spine and the proximal femur using dual-energy x-ray absorptiometry, as well as relevant clinical and biochemical parameters, were determined in 321 healthy Saudi females in order to establish reference values and to study the effects of physical and lifestyle factors on BMD. Mean +/- SD of age, body mass index (BMI), number of pregnancies, and total duration of lactation were 35.4 +/- 11.3 years, 26.5 +/- 5.2 kg/m2, 3.1 +/- 3.1, and 23.7 +/- 42.4 months, respectively. Mean +/- SD of serum calcium, 25-hydroxyvitamin D (25OHD), and PTH levels were 2.37 +/- 0.09 mmol/liter, 24.5 +/- 17.2 nmol/liter, and 52.0 +/- 30.8 pg/ml, respectively. Peak BMD values were observed around age 35 years at the spine and earlier at the femur. Compared with USA females, Saudi females had lower weight-matched Z scores at the spine (-0.126 +/- 1. 078, P = 0.04), femoral neck (-0.234 +/- 0.846, P < 0.0001), and Ward's triangle (-0.269 +/- 1.015, P < 0.0001). Further, the prevalence of osteopenia and osteoporosis in subjects >/=31 years old were 18-41% and 0-7%, respectively, depending on the site examined. Severe hypovitaminosis D (25OHD level

PIP: This paper examines the relationship between bone mineral density (BMD) of the spine and femur and vitamin D status, pregnancy, and lactation among women in Saudi Arabia. The aims of the study are the following: 1) establish normative data for BMD at the anterio-posterior lumbar spine and femur using dual x-ray absorptiometry; 2) compare the BMD of Saudi females and their US counterparts; and 3) examine the relation of BMD to vitamin D status, pregnancy, and lactation. Samples included 321 healthy Saudi females recruited from the city of Riyadh, Saudi Arabia. Results suggest that the mean standard deviation (SD) of age, body mass index, number of pregnancies, and total duration of lactation were, respectively, 35.4 +or- 11.3 years, 26.5 +or- 5.2 kg/sq. m, 3.1 +or- 3.1, and 23.7 +or- 42.4 months. Mean +or- SD of serum calcium, 25-hydroxyvitamin D (25OHD), and PTH levels were 2.37 +or- 0.09 mmol/liter, 24.5 +or- 17.2 nmol/liter, and 52.0 +or- 30.8 pg/ml, respectively. Peak BMD values were observed around age 35 years at the spine and earlier at the femur. Compared with US females, Saudi females had lower weight-matched Z scores at the spine, femoral neck, and Ward's triangle. On the other hand, the number of pregnancies and total duration of lactation correlated with weight-matched BMD Z scores at the spine. This made the BMD in healthy Saudi females significantly lower than their US counterparts. This may due to the increase number of pregnancies and longer duration of lactation together with prevalent vitamin D deficiency.

Long-term depot-medroxyprogesterone acetate and bone mineral density.

The association between long-term use of depot-medroxyprogesterone acetate (DMPA) and bone mineral density (BMD) has been controversial, as seen in three case-control studies in New Zealand, Thailand, and the United Kingdom. In the present case-controlled study of BMD, a group of 67 Chinese women who had used DMPA from 5-15 years was compared with 218 women of the same age range who had not used any steroidal hormones. DMPA users were found to have a significantly lower BMD at lumbar vertebra (L2-4) (0.93 g/cm2), neck of femur (0.69 g/cm2), trochanter (0.59 g/cm2), and Ward's triangle (0.58 g/cm2), as compared with the control group, whose corresponding BMD values were 1.03 g/cm2, 0.83 g/cm2, 0.71 g/cm2, and 0.78 g/cm2, respectively (p < 0.001). The average percentage of bone loss per year was estimated to be 1.1% in L2-4, 2.3% in neck of femur, 2.4% in trochanter, and 3.5% in Ward's triangle. The percentage of bone loss in L2-4 was found to be more pronounced with age. This study provided information that the use of DMPA in a Chinese group for > 5 years in associated with bone loss, and a prospective study is needed to confirm these data, which are different from two case-control studies.

PIP: The effect of long-term use of depot medroxyprogesterone acetate (DMPA) on bone mineral density remains controversial. The present study compared bone mineral densities in 67 long-term (5 years or more) DMPA users recruited consecutively from the Hong Kong (China) Family Planning Association with those in 218 age-matched controls recruited from 8 family health service clinics in Hong Kong. Mean age was 42.8 years (range, 34-46 years) in the DMPA group and 40.0 years (range, 34-46 years) among controls. Body mass index, calcium intake, and smoking were similar in both groups. The median duration of DMPA use was 6 years (range, 5-15 years). Long-term DMPA users had significantly lower bone mineral densities than controls at the lumbar vertebra (0.93 vs. 1.03 g/sq. cm), neck of femur (0.69 vs. 0.83 g/sq. cm), trochanter (0.59 vs. 0.71 g/sq. cm), and Ward's triangle (0.58 vs. 0.78 g/sq. cm). The percentage of bone loss in L2-4 was more pronounced with increasing age. For each year of DMPA use, the decrease in bone mineral density was estimated to be 0.011 g/sq. cm (1.1%) in L2-4, 0.0193 g/sq. cm (2.3%) in the neck of femur, 0.0169 g/sq. cm (2.4%) in the trochanter, and 0.0277 g/sq. cm (3.5%) in Ward's triangle.

Effects of levonorgestrel-releasing subdermal contraceptive implants on bone density and bone metabolism.

A prospective, randomized clinical trial observed the effects of Norplant long-term contraceptive implants and domestic implants similar to Norplant on bone mineral density and bone metabolism in female acceptors for 1 year. Bone mineral density (BMD) and bone mineral content (BMC) of lumbar 2-4 and proximal femur of 61 normal women of child-bearing age were measured by dual energy x-ray absorptiometry (DEXA) before and 12 months after implants insertion in both groups. BMD and BMC of lumbar 2-4 in both groups 12 months after implant insertion significantly increased (p < 0.01); with an average increase of 2.40% and 3.34%, respectively in the Norplant implant group, and 2.75% and 4.47%, respectively in the domestic implant group. Urine hydroxyproline and creatinine ratio (Hop/Cr) in the domestic implant group significantly decreased (p < 0.01). There was no significant differences in the effects on BMD and BMC of lumbar spine and femur and on bone metabolism between the two groups of contraceptive implants (p > 0.05). Levonorgestrel releasing contraceptive subdermal implants were not deleterious to the skeleton in women of child-bearing age. There was no significant effect on achieving maximum bone mass in young women.

PIP: This article presents a prospective, randomized clinical trial concerning the outcome of levonorgestrel-releasing subdermal contraceptive implants on bone density and bone metabolism during a 1-year study in China. A total of 61 women of childbearing age received either Norplant or domestic implants from July to November 1997. A dual energy x-ray absorptiometry before and 12 months after implant insertion was performed to assess the bone mineral density (BMD) and bone mineral content (BMC) of lumbar 2-4 and proximal femur. A significant increase in BMD and BMC (p 0.01) was observed in both groups, with an increase of 2.40% and 3.34%, respectively, in the Norplant group and 2.75% and 4.47%, respectively, in the domestic implant group. Included in the analysis was the measurement of the urine hydroxyproline and creatinine ratio, which was observed to decrease (p 0.01) in the domestic implant group. Insignificant changes in BMD and BMC on lumbar spine and femur were noted among women 35 years of age or older. This study concludes that levonorgestrel releasing contraceptive subdermal implants are not deleterious to the osseous framework of women in childbearing age.

Oral contraception and other factors in relation to back disorders in women: findings in a large cohort study.

The Oxford-Family Planning Association contraceptive study includes 17,032 women, initially aged 25-39 years, recruited at 17 British family planning centers during the interval 1968-1974 and subsequently followed-up for periods up to 26 years. This article examines the pattern of referral to hospital for back disorders among these women. Certain back disorders have been reported to occur more frequently in oral contraceptive users than in other women, and back pain has also been reported in some women consequent to using an intrauterine device. The disorders considered were spinal osteoarthritis, displaced cervical disc, displaced lumbar disc, other and unspecified displaced disc, cervicalgia, unspecified back pain, and sprains and strains of the back. Spinal osteoarthritis and unspecified backache were the only two conditions significantly related (both positively) to age. Displaced lumbar disc and other and unspecified displaced disc were strongly positively related to height and weight. Unspecified backache showed similar, but less striking (in terms of the magnitude of the relative risks), associations with height and weight. Little evidence was found of any association between oral contraceptive use and any of the back disorders, and the same was true for intrauterine device use.

Bone density in long term users of depot medroxyprogesterone acetate.

OBJECTIVE: To identify any adverse effect on bone density in long term users of depot medroxyprogesterone acetate (DMPA) for contraception. DESIGN: Cross-sectional measurement of bone density in users with amenorrhoea of more than one year or any woman using DMPA for more than five years. SETTING: Community Family Planning Clinics in Portsmouth and Manchester. POPULATION: One hundred and eighty-five women aged 17-52 years (mean 33.3 years) who had used DMPA for between 1 and 16 years and were attending the clinics for further injections, between August 1994 and August 1996. METHODS: Dual energy X-ray measurement of bone density of femoral neck and lumbar spine, and venous blood sample taken just prior to the next injection of DMPA. MAIN OUTCOME MEASURES: Bone density of femoral neck and lumbar spine and serum oestradiol in relationship to years of DMPA use and duration of amenorrhoea. RESULTS: Most women (n=153) had serum oestradiol levels < 150 pmol/l. Despite this, the mean bone density of the lumbar spine compared with the population mean for women aged 20-59 years gave a Z score (95% CI) of -0.332 (-0.510 to -0.154). There was no significant difference in the mean density of the femoral neck from the normal population mean. CONCLUSION: Despite amenorrhoea and low serum oestradiol, this sample of long term DMPA users had bone density only minimally below the normal population mean. We therefore found no clinically important adverse effect on bone density and therefore no reason to recommend bone conserving measures, such as add-back oestrogen.

PIP: The effects on bone density of long-term depot medroxyprogesterone acetate (DMPA) use were investigated in a cross-sectional study of 185 clients 17-52 years of age at family planning clinics in Portsmouth and Manchester, England, who had been receiving contraceptive injections for 1-16 years (median, 5 years). Dual energy x-ray measurements of bone density of the femoral neck and lumbar spine, as well as venous blood samples, were taken prior to the women's next DMPA injection (1994-96). 153 women had serum estradiol levels under 150 pmol/l--the value considered adequate to maintain bone density. The mean bone density of the lumbar spine compared with the population mean for women 20-59 years old yielded a Z score of minus 0.332 (95% confidence interval, -0.510 to -0.154; p 0.001). There was a weak, nonsignificant correlation between lumbar spine Z score and years of DMPA use. Mean density of the femoral neck did not differ significantly from the normal population mean. There was no significant correlation between serum estradiol level and either bone density score. Overall, these findings provide no evidence that DMPA-induced amenorrhea places women at significant risk of further bone loss or that supplemental estrogen is required.

Bone density among long-term users of medroxyprogesterone acetate as a contraceptive.

The bone density (BD) of 72 women using depot-medroxyprogesterone acetate (DMPA) for at least 1 year was compared with that of 64 women who were not users of hormonal contraceptives. The BD of lumbar spine, femoral neck, Ward's triangle, and trochanter was measured by dual energy X-ray absorptiometry (DEXA-LUNAR DPX). Estradiol (E2) concentrations were measured by radioimmunoassay (RIA). The mean age of DMPA users and nonusers was 31.8 and 31.1 years, respectively. Mean E, was 55.7 pg/mL for users and 149.9 pg/mL for controls (p < 0.001). The BD was significantly lower for DMPA users than for controls in all sites (p < 0.01). In addition, young adult T-scores in the lumbar spine were significantly lower among DMPA users than in controls (p < 0.01). Differences were maintained in a subsample of 47 women per group paired by age and body mass index (BMI). Multiple regression analysis showed that older age, lower BMI, and longer amenorrhea were associated with lower BD in the femoral neck, whereas lower BMI and use of DMPA were associated with lower BD in the lumbar spine.

PIP: Family planning programs that offer depot medroxyprogesterone acetate (DMPA) cannot be indifferent to the risk of lowered bone density. A study conducted at the Family Planning Clinic of the State University of Campinas (Sao Paulo, Brazil) compared bone densities in 72 women who had been using DMPA for at least 1 year (mean duration, 42 months) and 64 regularly menstruating nonusers. Mean age was 31 years in both groups; there were no significant differences between the 2 groups in terms of ethnicity, body mass index (BMI), or smoking. Mean serum estradiol concentrations were 55.7 +or- 40.5 pg/ml for DMPA users and 149.9 +or- 88.2 pg/ml for nonusers (p 0.001). The mean length of amenorrhea was 26.5 +or- 23.8 months among DMPA users. The mean bone density in DMPA users was significantly lower than that of controls at all sites evaluated (i.e., lumbar spine, femoral neck, Ward's triangle, and trochanter). 38 DMPA users, compared with only 17 controls, had a T-score in the lumbar spine lower than -1 standard deviation (p = 0.014). Multiple regression analysis identified BMI and DMPA use as variables significantly associated with bone density in the lumbar spine; in the femoral neck, these variables were BMI, age, and length of amenorrhea. Periodic bone densitometry should be considered for women over 40 years of age with low BMI who have more than 2 years of continuous amenorrhea.

Gonadotropin-releasing hormone analog plus an oral contraceptive containing desogestrel in women with severe hirsutism: effects on hair, bone, and hormone profile after 1-year use.

To evaluate the usefulness of D-Trp-6-luteinizing hormone-releasing hormone (LHRH) (triptorelin), a gonadotropin-releasing hormone (GnRH) analog (GnRHa), plus an oral contraceptive (OC) in the treatment of severe hirsutism, a total of 48 women between 19 and 35 years of age suffering from polycystic ovary syndrome (PCOS) with severe hirsutism were studied. Hyperandrogenism of adrenal origin was excluded in all subjects. Twenty-three patients received 3.75 mg D-Trp-6-LHRH intramuscularly monthly for 1 year plus an OC containing 30 micrograms ethinyl-estradiol and 150 micrograms desogestrel. A second group of 25 subjects received an OC containing 35 micrograms ethinyl-estradiol and 2 mg cyproterone acetate (CPA). Immediately before and after months 6 and 12 of therapy, bone mineral density (BMD) and Ferriman-Gallwey scores were evaluated and follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), prolactin (PRL), testosterone (T), androstenedione (A), dehydroepiandrosterone sulfate (DHEAS), 17-OH-progesterone (17-OHP), and sex hormone-binding globulin (SHBG) were determined. After 1 year of follow-up study, the combination of a GnRHa plus OC resulted in a decrease of hirsutism similar to that observed in the CPA group (41.9% v 40.5%) and in a suppression of gonadotropins and ovarian steroids in all treated women, without significant changes in bone density. The GnRHa-OC combination can potentially be used in the treatment of hirsutism and hyperandrogenism.

Effects on bone mineral density of low-dosed oral contraceptives compared to and combined with physical activity.

A cross-sectional study was designed to examine the influence of exercise compared to and in combination with low-dosed oral contraceptives (OCs) on bone mineral density (BMD). One hundred twenty-eight women (20 to 35 years of age) were assigned to four groups with respect to the years of exercise and OC intake. Influence factors were determined by a detailed questionnaire and interview. BMD for L2-4 and the femoral neck was assessed by DXA. The highest BMD values were found in the group of women characterized by long-term exercise (9.45 +/- 4.32 yr) and short use of OC (1.6 +/- 1.69 yr). No beneficial effect of exercise on BMD was found in the group with a long exercise period (10.4 +/- 4.14 yr) and long-term intake of OC (8.2 +/- 4.14 yr). Differences in mean BMD values between the two groups were significant in all regions assessed (p < 0.05). No differences in mean BMD were found in the groups with short-term exercise but long or brief histories of OC. The question arises as to whether active women taking low-dosed OC at an earlier age will develop an adequate BMD.

Bone metabolism in young women taking a monophasic pill containing 20 mcg ethinylestradiol: a prospective study.

The present one-year prospective study was performed to evaluate the effects of an oral contraceptive containing 20 mcg ethinylestradiol plus 0.150 mg desogestrel on bone metabolism in young women. Nineteen women aged 20 to 30 years completed the trial. Bone density was measured in the distal radius by dual photon absorptiometry before starting pill use and at the 3rd, 6th and 12th cycle. At the same time intervals, urinary hydroxyproline-to-creatinine ratio and serum alkaline phosphatase were evaluated. Bone density showed a slight, but not significant, increase at the end of the trial. Both urinary hydroxyproline-to-creatinine ratio and serum alkaline phosphatase showed a significant decrease. The results suggest that bone resorption is reduced, although bone density in the distal radius is not significantly increased in young women using an oral contraceptive containing only 20 mcg ethinylestradiol for one year.