Prior studies, including one from this institution, have demonstrated that the use of depot medroxyprogesterone acetate during the reproductive years is associated with a loss of bone mineral density (Lara-Torre E, Edwards CP, Perlman S, Hertweck SP. Prospective study evaluating bone mineral density in adolescent females using depo-Provera [depot medroxyprogesterone acetate] [abstract]. J Pediatr Adolesc Gynecol 2003;16:180).1–10 Only a small number of these studies, however, have used a longitudinal design to examine this relationship and even fewer have followed women beyond 12 months of contraceptive use. Furthermore, almost all of the studies with follow-up periods greater than 1 year have methodological limitations, such as a small sample size, which affect their generalizability. As a result, the few published studies with longer follow-up periods conflict in their results.1,9,11 For example, Cromer9 noted a 3.1% decrease in the lumbar spine after 24 months of depot medroxyprogesterone acetate use, and Busen et al1 observed a mean loss of 2.5%. Lara-Torre et al (J Pediatr Adolesc Gynecol 2003;16:180) observed a larger loss of 6.5%. These studies, however, were limited by small numbers of women who used depot medroxyprogesterone acetate for 2 years and a lack of subjects greater than 21 years of age. In contrast, Merki-Feld et al12 observed no significant change in cortical bone mass among 23 depot medroxyprogesterone acetate users followed up for 2 years, and Tang et al11 observed only a small loss of 0.44% per year in the lumbar spine. The former study, however, eliminated from analyses those women who lost more than 1% bone mineral density in the first 12 months. The study by Tang et al11 was limited to women aged between 37 and 49 years, who most likely would have already experienced some bone loss because of their age. Furthermore, the effects during the first 2 years of depot medroxyprogesterone acetate use could not be determined from their investigation because women who had been using the method at least 5 years were merged with new users. In the largest study published to date, Scholes et al13 reported that bone mineral density values decreased 6% on average after 4 years of depot medroxyprogesterone acetate use. This study, however, also merged continuous and new users, in addition to including users of birth control pills in the control group. Thus, precise data on the amount of bone mineral density loss experienced by women of reproductive age during the first 2 years of depot medroxyprogesterone acetate use are not available.
The purpose of this study was to extend our previous investigation so that we could measure the independent effects on bone mineral density of 24 months of consecutive use of depot medroxyprogesterone acetate compared with 2 different formulations of birth control pills and nonhormonal contraception. In this work, we have focused on bone mineral density changes occurring between 12 and 24 months and overall change from baseline to 24 months. Data from baseline to 12 months have been reported previously.3
MATERIALS AND METHODS
A detailed description of study methods is reported elsewhere.3 After obtaining institutional review board approval from the University of Texas Medical Branch in Galveston, Texas, and Wilford Hall Medical Center in San Antonio, Texas, women were recruited at Wilford Hall Medical Center or the Maternal and Child Health Clinic at the University of Texas Medical Branch between May 16, 1996, and January 20, 1999, to participate in a study on the relationship between hormonal contraceptive use and bone mineral density. Participants were aged between 18 and 33 years; of white, black, Hispanic, or Asian ethnicity; and were required to meet the minimum criteria for entry into the armed forces as a result of the funding source (Department of Defense). These criteria included a high school diploma or general education development certificate, no felony arrests, a body mass within 36% of ideal body weight for height, and a health status enabling completion of military training. Women who were currently pregnant or breastfeeding, had received an injection for contraception during the past 6 months or were taking birth control pills within the past month, or had a medical contraindication to hormonal contraception were not eligible.
After providing informed consent, women wishing to use hormonal contraception selected either injectable or oral hormonal contraception. Those selecting injectable contraception received 150 mg of depot medroxyprogesterone acetate (Depo-Provera; Pharmacia & Upjohn, North Peapack, NJ) every 3 months, whereas those selecting oral contraception were randomly assigned to receive pills containing either 0.035 mg of ethinyl estradiol and 1 mg of norethindrone (Ortho-Novum 1/35; Ortho Pharmaceutical Corporation, Brunswick, NJ) or pills containing 0.030 mg of ethinyl estradiol and 0.15 mg of desogestrel (Mircette; Organon Corporation, Oss, Netherlands). Pill assignment was blinded as previously described.3 Women who did not wish to use hormonal contraception were recruited to serve as nonhormonal controls. All controls were enrolled at the University of Texas Medical Branch site and frequency matched on age and race/ethnicity to the entire sample of hormonal contraceptive users.
During the baseline visit, height and weight were measured, and demographic information and medical history were recorded. Body mass index was calculated as weight in kilograms divided by the square of the height in meters. Smoking status (dichotomized: yes, smoker; no, nonsmoker) was determined using the question: “In the last 6 months, how many cigarettes did you smoke per day?” where the responses “not at all” and “less than 1 per day” were categorized “no” for smoking status and the responses “1–5 per day,” “1/2 pack per day,” “about 1 pack per day,” “about 1 1/2 packs per day,” and “2 or more packs per day” were categorized “yes” for smoking status. This variable differs from the smoking question “Have you ever smoked cigarettes?” which was asked only at baseline and reported in the previous article.3 We use the question reflecting smoking during the last 6 months in the present analysis because it was asked at each time point and therefore represents a more consistent longitudinal indicator of smoking. Engaging in weight-bearing or high-impact exercise as part of a regular exercise program was recorded as a dichotomous variable (yes/no). Dietary calcium intake was estimated from 24-hour dietary recalls conducted by trained research personnel at the baseline or 12-month study visit. Data obtained during the recall task were analyzed by using Menu Mizar 3.0 for Windows (Menu Systems, Ruffs Dale, PA) to estimate the calcium intake from foods and beverages consumed during the 24-hour period captured during the recall task.
Bone mineral density of the anterior-posterior lumbar spine (L1–L4) was determined with dual-energy X-ray absorptiometry at baseline and after 12 and 24 months of contraceptive use. All baseline scans were performed within 2 months of initiation of contraception. This 2-month “window” also was used at subsequent time points. Thus, follow-up scans were performed on women between 10 and 14 months after the baseline scan (12-month visit) and between 22 and 26 months after initiating contraception (24-month visit). Bone densitometry was performed at the University of Texas Medical Branch site with a Hologic QDR 1000-W machine (Hologic, Waltham, MA), and at Wilford Hall Medical Center with a Lunar DPX (Lunar, Madison, WI). In vivo precision, computed as the root-mean-square averages of standard deviations of repeated measurements14 was less than 1% at Wilford Hall Medical Center (across 10 repeat scans) and 1.2% at the University of Texas Medical Branch (across 20 repeat scans). Although strong correlations have been observed between measurements obtained on Hologic versus Lunar densitometers,15 direct comparison of bone mineral density values obtained with machines from different manufacturers is not ideal; therefore, data reflecting percent change from baseline are presented.
Continuous variables are expressed as means ± standard deviations. Group comparisons for demographic data and baseline variables were conducted using analysis of variance or independent group t tests. Group differences in categorical variables were analyzed by using χ2 or Fisher exact test. A 2-sided significance level of .05 was used to determine statistical significance. Our primary outcome was change in bone mineral density over the 2-year follow-up interval, which was computed as the mean percent change from baseline by using the formula: ([24-month bone mineral density − baseline bone mineral density]/baseline bone mineral density) × 100. In addition, we examined the percent change during the second year of hormonal contraceptive use by applying the formula: ([24-month bone mineral density − 12-month bone mineral density]/12-month bone mineral density) × 100. Analysis of covariance was conducted by using SAS statistical software (SAS Institute, Cary, NC). Covariate-adjusted least squares mean changes in bone mineral density from baseline are presented. Covariates included smoking status, calcium intake, and exercise at 12 months, race/ethnicity, age, site, and body mass index, with all covariates remaining in the model regardless of their P value. Bonferroni-adjusted group mean differences in bone mineral density and 95% confidence intervals of percent change from baseline at 24 months are reported, controlling the type I experimental error rate at the alpha level of significance of .05. To test for linearity among depot medroxyprogesterone acetate users, a repeated-measures mixed model, adjusted for covariates, was fit using data from baseline, 12 months, and 24 months.
Overall, 197 women had bone mineral density data at baseline and at least 1 follow-up time point (12 or 24 months) within the 2-month window of that time point. Specifically, 110 women had data at all 3 time points, 49 women had baseline and 12-month bone mineral density data, whereas 38 women had baseline and 24-month data but did not have 12-month measurements. Using the same criteria applied in our 12-month analysis,3 2 women were excluded from analysis as statistical outliers who had bone mineral density changes in the 24-month measurement exceeding 3 standard deviations from the group mean. Thus, a total of 6 women were excluded; this included 4 outliers identified previously in the 12-month analysis and 2 additional subjects meeting outlier criteria for the 24-month measurement. After omitting these 6 women from analyses, a total of 191 women made up the final sample: 86 users of oral contraception (38 norethindrone pills, 48 desogestrel pills), 47 users of depot medroxyprogesterone acetate, and 58 users of nonhormonal contraception (controls).
There were no significant differences between the 4 contraceptive groups in age, racial/ethnic composition, weight-bearing exercise, calcium intake, or body mass index (Table 1). Significantly fewer women reported smoking in the pill groups compared with women using depot medroxyprogesterone acetate or nonhormonal methods (P < .001).
Tables 2 and 3 demonstrate differences between groups after controlling for age, race/ethnicity, weight-bearing exercise, calcium intake, smoking status, and body mass index. Women using depot medroxyprogesterone acetate for 24 months experienced, on average, a 5.7% loss in bone density at the lumbar spine (L1–L4) from baseline, with a 3.2% loss occurring after the first year of use, that is, between months 12 and 24. A test for linearity among depot medroxyprogesterone acetate users reveals a strong overall linear trend among bone mineral density loss from baseline to 12 months and 12 to 24 months (P < .001). Users of desogestrel pills also experienced a loss in bone mineral density of approximately 2% between 12 and 24 months of pill use. No significant change in bone mineral density was noted among norethindrone users between 12 and 24 months or between baseline and 24 months. Women not using hormonal contraception experienced a 2.6% gain in bone mineral density during the second year of the study (months 12 to 24).
Analysis of percent change from baseline to 24 months determined that, on average, depot medroxyprogesterone acetate users lost nearly 6% bone mineral density at the lumbar spine over 2 years. On average, users of desogestrel pills lost 2.6% bone mineral density. However, when compared with controls in this study, the loss experienced by users of desogestrel pills was not statistically significant after 24 months (Table 3). Bonferroni-adjusted pairwise comparisons of the differences in percent change bone mineral density (adjusted for demographic and behavioral covariates) from baseline to 24 months demonstrated that the change in bone mineral density experienced by depot medroxyprogesterone acetate users differed significantly from the change in bone mineral density experienced by users of both pill types and the control group.
In this study, we observed approximately a 3% loss in bone mineral density at the lumbar spine during each of the first 2 years of depot medroxyprogesterone acetate use, for a total of almost 6% after 2 years. This rate of decline is similar to that noted in 1 prior investigation (Lara-Torre E, et al. J Pediatr Adolesc Gynecol 2003;16:180) but exceeds the amount reported by several others.1,9,12,13 Differences between studies in the amount of bone loss among depot medroxyprogesterone acetate users may be attributable to differences in important demographics, such as age or composition of the cohort, that is, combination of new and previous users. For example, studies by Tang et al6,11 were limited to long-term users aged more than 35 years who may have already lost a portion of their bone that is sensitive to estrogen, in contrast to our sample of women aged less than 35 years.
A prior study by Cundy et al16 suggests that losses that result from depot medroxyprogesterone acetate use may be reversible after discontinuation of the contraceptive. In their study, dual-energy X-ray absorptiometry scans were performed on 14 women approximately 12 and 24 months after they discontinued depot medroxyprogesterone acetate. A mean increase of 3% after 12 months and 6.4% after 24 months at the lumbar spine was observed. It is of interest that these numbers closely match the amount of loss we observed in this study after 12 and 24 months of depot medroxyprogesterone acetate use. Scholes et al13 also noted an increase in bone mineral density after discontinuation; the mean bone density for discontinuers over 21 years of age approached that of nonusers 30 months after stopping use of depot medroxyprogesterone acetate. Together, these studies suggest that use of depot medroxyprogesterone acetate in women aged more than 21 years is unlikely to result in long-term loss of bone mineral density or increase their risk of osteoporosis.
Even if changes in bone mineral density during the reproductive years are not completely reversible, loss of bone mineral density during the reproductive years does not necessarily increase women’s risk of osteoporosis. The most likely reason that women lose bone while using depot medroxyprogesterone acetate is that depot medroxyprogesterone acetate suppresses estrogen levels. All women are likely to become hypoestrogenic and experience some bone loss after menopause unless they take exogenous estrogen or another type of bone-building agent on a long-term basis. In a study of postmenopausal women, Cundy et al17 observed that women who were previous users of depot medroxyprogesterone acetate experienced attenuated rates of bone loss after going through menopause as compared with women who had not used depot medroxyprogesterone acetate. Thus, women who use depot medroxyprogesterone acetate for a number of years may simply be losing the estrogen-sensitive portion of their bone at an earlier age than they would have normally.
It may also be possible to prevent depot medroxyprogesterone acetate-related decreases in bone density. In fact, one study18 recently demonstrated that this could be accomplished by administering oral estrogen along with depot medroxyprogesterone acetate. In a double-blind, placebo-controlled trial of 38 premenopausal women who had used depot medroxyprogesterone acetate at least 2 years and had below-average baseline lumbar spine bone mineral density, mean lumbar spine bone mineral density increased 1% over a 2-year interval in the estrogen-treated group as compared with a 2.6% loss in the placebo group.18 Thus, supplemental estrogen could be offered to depot medroxyprogesterone acetate users as a way to decrease the risk of bone loss. However, this would not be an option for women who have a contraindication to estrogen and would decrease the convenience of the method. Another potential intervention may be calcium supplementation with or without an increase in weight-bearing exercise. Recent U.S. studies on depot medroxyprogesterone acetate users have reported a mean calcium intake among these women that is well below the Institute of Medicine’s recommendation of 1,000–1,300 mg/day for nonpregnant women.19 For example, Ott et al20 noted that women had an average intake of 831 mg/day whereas depot medroxyprogesterone acetate users in our study reported an average intake of 519 mg/day at baseline. Although it is obvious that these women are not consuming enough calcium, studies are still needed to determine the extent to which increasing calcium intake would decrease the amount of bone loss experienced as a result of depot medroxyprogesterone acetate use.
We also observed changes in bone mineral density among users of oral contraceptives. Those women who used the norethindrone-containing pill had a significant increase in bone mineral density after 12 months and those who used a desogestrel-containing pill had significant loss between 12 and 24 months. However, these results must be interpreted with caution because after a 24-month period, bone mineral density changes observed among the pill groups did not significantly differ from changes observed among the control group not using hormonal contraception. Thus, our study supports the work of others2,5,21–25 who have observed that use of oral contraceptives does not affect bone mineral density.
This study has several limitations. First of all, we had a fairly homogenous population. The majority of subjects were white, and none had body mass index that was greater than 20% of their ideal body weight. Furthermore, we did not include women aged less than 18 years, which may have been an important omission, because young women have been shown to be at increased risk of bone density changes associated with depot medroxyprogesterone acetate use. Finally, women enrolled in this study self-selected their birth control method, which may have introduced bias.
One difference that resulted from self-selection of their birth control method was that smokers were more likely to choose depot medroxyprogesterone acetate than birth control pills. Smoking has been associated with bone density loss in several studies of premenopausal women23,26,27; thus, some may speculate that the bone mineral density loss we observed among depot medroxyprogesterone acetate users was attributable to tobacco use. The degree to which smoking affects bone mineral density in premenopausal women is not clear, however, because other studies on young women have not demonstrated a significant association between tobacco use and bone mineral density. In fact, a meta-analysis of 29 studies noted that bone density was actually similar in smokers and nonsmokers before menopause and that current smokers did not exhibit greater bone loss than nonsmokers until after menopause.28 To ensure that this difference in smoking status between groups did not confound our results, we included smoking as a covariate in our multivariate analyses. These analyses demonstrated a difference in bone loss between groups even after accounting for the difference in smoking status between users of depot medroxyprogesterone acetate and pills.
It is important to note that we did not compare the rate of bone mineral density change experienced as a result of contraceptive use with that experienced during pregnancy. This factor was not included because pregnant women should not undergo dual-energy X-ray scans because of the risk of radiation exposure. This is an important omission, however, because women who do not use hormonal contraception are more likely to become pregnant than those who do. Prior investigators have demonstrated that both pregnancy and lactation are associated with short-term trabecular bone losses that equal or surpass those noted in this study. In fact, one study comparing prepregnancy dual-energy X-ray values to those obtained after delivery showed that bone mineral density at the lumbar spine decreased 2.1%29 whereas a second study observed a loss of 4.6% over this 9-month period.30 In addition, women may experience other medical problems, as well as social or economic consequences, as a result of pregnancy. For these reasons, providers should not deny women use of depot medroxyprogesterone acetate out of concern for bone health, especially if the physician and patient agree that this is the birth control method most likely to be successful for that particular patient.
In conclusion, the loss of bone mineral density associated with use of depot medroxyprogesterone acetate during the reproductive years appears to be linear during the first 2 years of use. Depot medroxyprogesterone acetate is intended for long-term use. Thus, additional studies that follow users of this method for extended periods of time and those that determine whether these changes are reversible are critical so that physicians can accurately counsel patients regarding this method.
1. Busen NH, Britt RB, Rianon N. Bone mineral density in a cohort of adolescent women using depot medroxyprogesterone acetate for one to two years. J Adolesc Health 2003;32:257–9.
2. Wanichsetakul P, Kamudhamas A, Watanaruangkovit P, Siripakarn Y, Visutakul P. Bone mineral density at various anatomic bone sites in women receiving combined oral contraceptives and depot-medroxyprogesterone acetate for contraception. Contraception 2002;65:407–10.
3. Berenson AB, Radecki CM, Grady JJ, Rickert VI, Thomas A. A prospective, controlled study of the effects of hormonal contraception on bone mineral density. Obstet Gynecol 2001;98:576–82.
4. Petitti DB, Piaggio G, Mehta S, Cravioto MC, Meirik O. Steroid hormone contraception and bone mineral density: a cross-sectional study in an international population. The WHO Study of Hormonal Contraception and Bone Health. Obstet Gynecol 2000;95:736–44.
5. Scholes D, Lacroix AZ, Ott SM, Ichikawa LE, Barlow WE. Bone mineral density in women using depot medroxyprogesterone acetate for contraception. Obstet Gynecol 1999;93:233–8.
6. Tang OS, Tang G, Yip P, Li B, Fan S. Long-term depot-medroxyprogesterone acetate and bone mineral density. Contraception 1999;59:25–9.
7. Paiva LC, Pinto-Neto AM, Faundes A. Bone density among long-term users of medroxyprogesterone acetate as a contraceptive. Contraception 1998;58:351–5.
8. Cundy T, Cornish J, Roberts H, Elder H, Reid IR. Spinal bone density in women using depot medroxyprogesterone contraception. Obstet Gynecol 1998;92:569–73.
9. Cromer BA, Blair JM, Mahan JD, Zibners L, Naumovski Z. A prospective comparison of bone density in adolescent girls receiving depot medroxyprogesterone acetate (Depo-Provera), levonorgestrel (Norplant), or oral contraceptives. J Pediatr 1996;129:671–6.
10. Cundy T, Evans M, Roberts H, Wattie D, Ames R, Reid IR. Bone density in women receiving depot medroxyprogesterone acetate for contraception [published correction appears in BMJ 1991;303:220]. BMJ 1991;303:13–6.
11. Tang OS, Tang G, Yip PS, Li B. Further evaluation on long-term depot-medroxyprogesterone acetate use and bone mineral density: a longitudinal cohort study. Contraception 2000;62:161–4.
12. Merki-Feld GS, Neff M, Keller PJ. A 2-year prospective study on the effects of depot medroxyprogesterone acetate on bone mass-response to estrogen and calcium therapy in individual users. Contraception 2003;67:79–86.
13. Scholes D. LaCroix AZ, Ichikawa LE, Barlow WE, Ott SM. Injectable hormone contraception and bone density: results from a prospective study [published correction appears in Epidemiology 2002;13:749]. Epidemiology 2002;13:581–7.
14. Gluer CC, Blake G, Lu Y, Blunt BA, Jergas M, Genant HK. Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int 1995;5:262–70.
15. Pocock NA, Noakes KA, Griffiths M, Bhalerao N, Sambrook PN, Eisman JA, et al. A comparison of longitudinal measurements in the spine and proximal femur using Lunar and Hologic instruments. J Bone Miner Res 1997;12:2113–8.
16. Cundy T, Cornish J, Evans MC, Roberts H, Reid IR. Recovery of bone density in women who stop using medroxyprogesterone acetate. BMJ 1994;308:247–8.
17. Cundy T, Cornish J, Roberts H, Reid IR. Menopausal bone loss in long-term users of depot medroxyprogesterone acetate contraception. Am J Obstet Gynecol 2002;186:978–83.
18. Cundy T, Ames R, Horne A, Clearwater J, Roberts H, Gamble G, et al. A randomized controlled trial of estrogen replacement therapy in long-term users of depot medroxyprogesterone acetate. J Clin Endocrinol Metab 2003;88:78–81.
19. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board, Institute of Medicine. Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Washington, DC: National Academy Press; 1999. p. 71–145.
20. Ott SM, Scholes D, LaCroix AZ, Ichikawa LE, Yoshida CK, Barlow WE. Effects of contraceptive use on bone biochemical markers in young women. J Clin Endocrinol Metab 2001;86:179–85.
21. Perrotti M, Bahamondes L, Petta C, Castro S. Forearm bone density in long-term users of oral combined contraceptives and depot medroxyprogesterone acetate. Fertil Steril 2001;76:469–73.
22. Lloyd T, Taylor DS, Lin HM, Matthews AE, Eggli DF, Legro RS. Oral contraceptive use by teenage women does not affect peak bone mass: a longitudinal study. Fertil Steril 2000;74:734–8.
23. Mazess RB, Barden HS. Bone density in premenopausal women: effects of age, dietary intake, physical activity, smoking, and birth-control pills. Am J Clin Nutr 1991;53:132–42.
24. Rodin A, Chapman M, Fogelman I. Bone density in users of combined oral contraception: preliminary reports of a pilot study. Br J Fam Plann 1991;16:125–129.
25. Lloyd T, Buchanan JR, Ursino GR, Myers C, Woodward G, Halbert DR. Long-term oral contraceptive use does not affect trabecular bone density. Am J Obstet Gynecol 1989;160:402–4.
26. Stevenson JC, Lees B, Devenport M, Cust MP, Ganger KF. Determinants of bone density in normal women: risk factors for future osteoporosis? BMJ 1989;298:924–8.
27. Franceschi S, Schinella D, Bidoli E, Dal Maso L, La Vecchia C, Parazzini F, et al. The influence of body size, smoking, and diet on bone density in pre- and postmenopausal women. Epidemiology 1996;7:411–4.
28. Law MR, Hackshaw AK. A meta-analysis of cigarette smoking, bone mineral density and risk of hip fracture: recognition of a major effect. BMJ 1997;315:841–6.
29. More C, Bettembuk P, Bhattoa HP, Balogh A. The effects of pregnancy and lactation on bone mineral density. Osteoporos Int 2001;12:732–37.
30. Naylor KE, Iqbal P, Fledelius C, Fraser RB, Eastell R. The effect of pregnancy on bone density and bone turnover. J Bone Miner Res 2000;15:129–37.
© 2004 The American College of Obstetricians and Gynecologists
This article has been cited