A Cochrane review «»1 «Lopez LM, Grimes DA, Schulz KF, Curtis KM. Steroid...»2 included 6 studies (n=503) assessing the effect of long acting progestin contraceptives on bone mineral density (BMD). Two trials randomized women using depot medroxyprogesterone acetate DMPA (150 mg DMPA intramuscularly every 12 weeks) to an estrogen supplement or a placebo. Both studies showed BMD increases in the estrogen groups and decreases in the placebo groups (percent change in BMD in spine by 12 months, MD 2.90 95% CI 1.80 to 4.00; 1 study, n=69 and change in BMD in spine by 24 months, MD 0.04 95% CI 0.02 to 0.06; 1 study, n=26). In one study (n=22), results in serum alkaline phosphatase and osteocalcin favoured levonorgestrel implant compared to DMPA 150 mg every 12 weeks.
A study comparing etonogestrel-releasing implant with levonorgestrel-releasing implant reported decreases in BMD in both groups, however the effect was less in the levonorgestrel group (etonogestrel vs levonorgestrel: percent change in BMD in midshaft ulna by 18 months, MD - 0.39 95% CI -0.56 to -0.22; 1 study, n=111). In a poor quality study no difference was found between monthly injections of norethisterone enanthate 50 mg plus estradiol valerate 5 mg versus Nova-T IUD (copper intrauterine device).
Another Cochrane review «»2 «Lopez LM, Chen M, Mullins S et al. Steroidal contr...»1 included 2 case-control studies examining progestin-only contraceptives. One reported increased fracture risk for DMPA ever-use (reported OR 1.44, 95% CI 1.01 to 2.06), more than 4 years of use (reported OR 2.16, 95% CI 1.32 to 3.53), and women over 50 years old. The other analyzed 17 527 incident fracture cases and 70 130 control patients (DMPA exposure 11 and 8%, respectively). Compared with nonuse, current use of one to two, 3 to 9, or 10 or more DMPA prescriptions yielded adjusted odds ratio (OR) for fractures of 1.18 (95% CI 0.93 to 1.49), 1.36 (95% CI 1.15 to 1.60), and 1.54 (95% CI 1.33 to 1.78), respectively. An increased risk was noted for any past use, including one or two prescriptions (reported OR 1.17, 95% CI 1.07 to 1.29). Fracture risk was highest after longer treatment duration (over 2 to 3 years), and there was no difference in patients below and above the age of 30. For users of combined estrogen-containing oral contraceptives, the OR were around 1.
A multicenter, prospective, non-randomized observational study «Harel Z, Johnson CC, Gold MA et al. Recovery of bo...»3 in 98 healthy female adolescents (12-18 years) assessed BMD changes during DMPA use for 40 weeks and for up to 300 weeks after DMPA cessation. There were no control group. BMD at the lumbar spine (LS), total hip (TH) and femoral neck (FN) was assessed by dual-energy X-ray absorptiometry. 7% completed the full 5 years of DMPA use, average exposure was 2 years. At the time of their final DMPA injection, mean BMD declines from baseline of 2.7% (LS), 4.1% (TH) and 3.9% (FN) (p<.001 at all 3 sites) were found. Within 60 weeks of discontinuation of DMPA, mean LS BMD had returned to baseline levels, and 240 weeks after DMPA discontinuation, the mean LS BMD was 4.7% above baseline. Mean TH and FN BMD values recovered to baseline values more slowly: 240 weeks and 180 weeks, respectively, after the last DMPA injection. Post-DMPA BMD increases were significantly smaller in those participants who exhibited 5% or greater BMD loss during DMPA treatment than in participants who had less BMD decline during DMPA use. Of those with over 5% loss in BMD during treatment, none showed a LS deficit of that magnitude after 3 years stopping treatment, however 14% continued to show 5% or greater deficit at the hip relative to baseline.
In a longitudinal study «Clark MK, Sowers M, Levy B et al. Bone mineral den...»4 changes in BMD during 48 months in first-time DMPA users (n=178) during use and after discontinuation were compared with controls (n=145) not using hormonal contraception. The BMD of the hip and spine was measured at 3-month intervals, by dual energy roentgen absorptiometry in these women (18-35 years). Hip and spine BMD declined during 48 months of DMPA use by 7.7% +/- 0.11% (mean +/- SE) and 6.4% +/- 0.36%, respectively. The BMD of controls declined 1.6% or less, +/- 0.30%. After discontinuation, BMD increased from 0.3% to 2.0% per year depending on length of depot MPA use and bone site. The longest depot MPA users remained 4.7% and 2.9% lower than hip and spine baseline values, respectively, 18 months after discontinuation.
Comment: The quality of evidence is downgraded by study quality (unclear allocation concealment) and by imprecise results (limited study size for each comparison).