Major Finding: Within 60 weeks of DMPA discontinuation, mean BMD values for the lumbar spine had returned to baseline levels. At the hip, full recovery of mean BMD took 240 weeks and at the femoral neck, 180 weeks.
Data Source: BMD values for up to 300 weeks after cessation of DMPA in 98 healthy female adolescents who initiated the injections for contraception between the ages of 12 and 18.
Disclosures: The study was funded by a grant from Pfizer Inc. Editorial assistance for the publication of the results was provided by Anthemis Consulting Ltd. and funded by Pfizer. Dr. Harel disclosed financial relationships with Merck & Co., Teva/Duramed, Ortho-McNeil, GlaxoSmithKline, Novartis, and Warner Chilcott.
TORONTO — Bone mineral density loss in female adolescents receiving depot medroxyprogesterone acetate for contraception was substantially or fully reversible following discontinuation of the drug, according to study results.
In girls who lost more than 5% of bone mineral density (BMD) during treatment, however, complete recovery was less likely. Also, recovery was generally greater and faster in the lumbar spine than the hip.
“If someone loses about 2%-4% of bone mineral density on Depo, then when they stop the injections, their BMD will recover and there is no risk,” said principal investigator Dr. Zeev Harel, professor of pediatrics Brown University, Providence, R.I. “The longer they use Depo, the more shots they get, the more BMD they lose; it's somewhat more difficult to recover.”
Participants included 98 healthy female adolescents between the ages of 12 and 18 who initiated depot medroxyprogesterone acetate (DMPA, Depo-Provera) intramuscular injections for contraception and provided BMD data for up to 300 weeks after cessation of DMPA. BMD was assessed by dual-energy x-ray absorptiometry at the lumbar spine, hip, and femoral neck.
During the study period, 19.4% of participants received 17 or more injections of DMPA, 15.3% received 13–16 injections, 17.3% received 9–12 injections, 24.5% received 5–8 injections, and 23.5% received 4 or fewer injections. Overall, the median total number of DMPA injections received was nine, Dr. Harel reported in a poster presentation during the annual meeting.
At the time of DMPA cessation, participants showed mean BMD declines from baseline of 2.7% at the lumbar spine, 4.1% at the hip, and 3.9% at the femoral neck.
Within 60 weeks of DMPA discontinuation, mean BMD values for the lumbar spine had returned to baseline levels. By 240 weeks, they had increased by 4.7% above baseline. Recovery occurred more slowly at the hip and femoral neck, with full recovery of mean BMD not seen until 240 weeks in the hip and 180 weeks in the femoral neck.
Postcessation gains were smaller in girls who exhibited a 5% or greater BMD loss during treatment, with mean BMD remaining below baseline at 240 weeks, also published in Contraception (2010;81:281-91).
Participants who had a 5% or greater loss of BMD had received a significantly greater number of DMPA injections (median, 13) than did those with less than 5% loss (median, 5).
The investigators noted five patient factors that affected BMD loss during DPMA treatment. Dr. Harel explained that those who had adequate calcium intake, adequate vitamin D intake, no smoking, and lower alcohol intake tended to lose less BMD during treatment. “The fifth factor, and one we have no control over, was weight,” he said. “Those who were a bit more overweight tended to lose less BMD.”
A 15% gain in BMD was seen in the control group. “In girls this age, over about 8–9 years, they can gain 15% in bone mineral density,” said Dr. Harel.
My Take
Weigh the Risk-Benefit Ratio
I think Dr Harel's data are encouraging in that some of the BMD loss appears to be reversible. On the flip side, it doesn't appear that the losses were completely reversible, so one worries in a young person on Depo-Provera that she might have skeletal deficits as she reaches peak bone mass.
Having said that, I'm an adolescent medicine physician, and for some patients Depo-Provera keeps them from becoming pregnant most effectively, so I think a clinician has to very carefully weigh the risk-benefit ratio. As we know, pregnancy is a high bone turnover state, so they're breaking down their skeleton when they should be accruing it.
We don't have the long-term outcome data to say whether these girls go on to be at higher risk for osteoporosis and for fractures, but these data are sorely needed, and I think they stimulate us to want to design those long-term studies to look at the health outcomes that accompany those changes in bone density.