Healthy premenopausal women: no effect of oral contraceptives on bone mineral density
| Study design | Reference | No of patients | OC exposure | Measurement of BMD/bone metabolism | Results |
|---|---|---|---|---|---|
| OC, Oral contraceptive; BMD, bone mineral density; RCT, randomised controlled trial; EE, ethinyl oestradiol; CA, cyproterone acetate; DXA, dual energy x ray absorptiometry; PYR, pyridinoline; D-PYR, deoxypyridinoline; qCT, quantitative computed tomography; BSAP, bone specific alkaline phosphatase; NTx, N-telopeptides; DPA, dual photon absorptiometry; SPA, single photon absorptiometry; DMPA, deoxymedroxyprogesterone acetate; SXA, single energy x ray absorptiometry; PTH, parathyroid hormone; Cr, creatinine; NS, non-significant. | |||||
| RCT (level 1b,26,27,29 level 2b28) | Castelo-Branco et al26 | 67 women (ages 19–29) | 35 μg EE + 2 μg CA (n = 35) v 30 μg EE + 150 μg desogestrel (n = 32) for 24 months | DXA | No changes in BMD from baseline in either group |
| Nappi et al27 | 60 women (ages 22–34) | 20 μg EE +75 μg gestodene (n = 20) v 15 μg EE +60 μg gestodene (n = 20) v control (n = 20) for 12 months | Lumbar spine DXA; urinary PYR, D-PYR, serum osteocalcin | No changes in BMD from baseline in any group; decrease in PYR, D-PYR in OC treated groups suggesting decreased resorption | |
| Endrikat et al28 | 48 women (ages 20–38) | 30 μg EE + 150 μg levonorgestrel (n = 25) v 20 μg EE +100 μg levonorgestrel (n = 23) for 36 months | Lumbar spine qCT; serum BSAP, urinary NTx | No changes in BMD from baseline in either group; decrease in NTx in both groups (suggesting decreased resorption) | |
| Nappi et al29 | 71 women (ages 22–34) | 30 μg EE+3 mg drospirenone (n = 24) v 30 μg EE+75 μg gestodene (n = 24) v control (n = 23) for 12 months | Lumbar spine DXA; serum & urinary Ca2+, serum osteocalcin, urinary PYR, D-PYR | Decrease in PYR, D-PYR in both OC treated groups from baseline (suggesting decreased resorption); trend to increased BMD in EE+drospirenone group | |
| Cohort (level 2b) | Mazess & Barden30 | 300 women (ages 20–39) | 50% past/current OC users, 50% never users | Lumbar spine DPA, radius SPA | No association between OC use and BMD |
| Cromer et al31 | 48 women (ages 12–21) | 30 μg EE + 150 μg desogestrel (n = 9) v Norplant (n = 7) v Depo-Provera (n = 15) v control (n = 17) for 12 months | Lumbar spine DXA | No significant difference between change in BMD in OC treated group (1.5% increase in BMD) v control (2.9% increase in BMD) | |
| Lloyd et al32 | 62 white women (followed from age 12–20 years) | OC users (“low dose monophasic”) (n = 28) v non-users (n = 34) | Proximal femur DXA | No effect of OC treatment on peak bone mass or rate of acquisition | |
| Cohort (level 4,33,34 level 2b35–38) | Reed et al33 | 245 women (ages 18–39) | Current OC users (80% on 30–35 μg EE) (n = 89) v control (n = 156) | Lumbar spine, proximal femur, total body DXA | No change in BMD from baseline in either group |
| Lara-Torre et al34 | 148 women (ages 11–21) | New OC users (n = 71) v new DMPA users (n = 58) v control (n = 19) over 24 months | Lumbar spine DXA | No change in BMD from baseline in OC users | |
| Lloyd et al35 | 80 women (ages 12–22) | OC users (for ⩾6 months, and still using at age 22) (n = 33) v non-users (n = 17) | Total body, bilateral proximal femur DXA | No difference in BMD between OC users and non-users | |
| Berenson et al36 | 191 women (ages 18–33) | OC (35 μg EE+1 mg norethindrone or 30 μg EE+0.15 mg desogestrel) (n = 86) v DMPA (n = 47) v control (n = 58) for 24 months | Lumbar spine DXA | No difference in BMD change from baseline between OC groups and control (decrease in BMD from baseline in DMPA group v control) | |
| Paoletti et al37 | 54 women (ages 20–30) | 30 μg EE+3 mg drospirenone (n = 28) v control (n = 26) for 6 months | Heel DXA+laser; serum osteocalcin, BSAP, urinary PYR, D-PYR | No change in BMD from baseline in any group; decrease in osteocalcin, BSAP, PYR in OC group (suggesting decreased bone turnover) | |
| Rome et al38 | 370 women (ages 12–18) | 20 μg EE+100 μg levonorgestrel (n = 165) v DMPA (n = 53) v control (n = 152) for 12 months | Lumbar spine, hip DXA; serum BSAP, urinary D-PYR | Increase in BSAP in control v OC, but no difference in BMD between groups | |
| Cross sectional | Sowers et al39 | 86 women (ages 20–35) | OC users (for >2 months) (n = 78) v non-users (n = 8) | Bone mass by 125I photon absorptiometry | No difference in bone mass between ever v never users or between current v past users |
| Hreschyshyn et al40 | 352 women (pre- and post-menopausal; ages 24–79) | Ever OC users (n = 116) v never users (n = 236) | Lumbar spine, femoral neck DPA | No difference in BMD between ever OC users and never users | |
| Lloyd et al41 | 25 women | OC users (minimum 50 μg mestranol/day) (n = 14) v non-users (n = 11) | Lumbar spine qCT | No difference in BMD between OC users and non-users | |
| Stevenson et al42 | 284 white women (112 pre-, 172 post- menopausal) | OC users v non-users | Lumbar spine, proximal femur DPA | No association between OC use and BMD in premenopausal women | |
| Hall et al43 | 165 women (pre- and post-menopausal; ages4–80) | Ever OC users (n = 69) v never users (n = 96) | Lumbar spine DXA | No difference in BMD between ever OC users and non-users in any age group | |
| Murphy et al44 | 841 women (229 pre-, perimenopausal, 583 postmenopausal, 29 unknown) | Ever OC users (n = 159 pre-, perimenopausal; n = 182 postmenopausal; n = 11 unknown) v never users (n = 70 pre-, peri-menopausal; n = 401 postmenopausal; n = 18 unknown) | Lumbar spine, hip DXA | No difference in BMD between ever OC users and non-users | |
| Garnero et al45 | 208 women (ages 35–49) | OC users (combined pills with 30 μg EE, n = 41; combined pills with 50 μg EE, n = 3; sequential combined pills, n = 5; progestative contraceptives, n = 3) (total n = 52) v non-users (n = 156) | Lumbar spine, total body, hip, distal radius DXA; serum osetocalcin, BSAP, C terminal propeptide of type I collagen, urinary NTx and PYR | No difference in BMD between OC users and non-users; decrease in markers of both formation and resorption in OC users v non-users (suggesting decreased bone turnover) | |
| Ulrich et al46 | 25 women (mean age 41) | Ever OC users v never users | Axial, peripheral BMD by DXA | No difference in BMD between ever OC users and never users | |
| Petitti et al47 | 2474 women (ages 30–34) | Ever OC users (82% >30 but <50 μg oestrogen, 15% ⩾50 μg oestrogen, <1% <30 μg oestrogen, 2% unknown dose) (n = 819) v ever DMPA users (n = 350) v ever levonorgestrel users (n = 610) v control (n = 695) | Distal radius, midshaft ulna SXA | No difference in BMD between ever users of hormonal contraception v never users | |
| Ott et al48 | 227 women (ages 18–39) | OC users (53.6% 35 μg EE +0.5–1 mg norethindrone, 18% 35 μg EE + 1 mg levonorgestrel or 1 mg ethynodiol diacetate, 13.7% 30 μg EE +1.5 mg norethindrone, 9.7% 20 μg EE + levonorgestrel or norethindrone) (n = 39) v DMPA (n = 116) v control (n = 72) | Lumbar spine, total body, total hip DXA; serum Ca2+, PTH, osteocalcin, urinary NTx | No difference in BMD between any of the groups; decrease in osteocalcin and NTx in OC users than in non-users (suggesting decreased bone turnover) | |
| Perotti et al49 | 189 women (ages 30–34) | OC users (for ⩾2 years) (n = 63) v DMPA users (for ⩾2 years) (n = 63) v control (no hormonal contraception) (n = 63) | Non-dominant radius SXA | No difference in BMD between any of the groups | |
| Hawker et al50 | 830 women (ages 19–35) | Current OC users (n = 223) v past OC users (n = 512) v never users (n = 95) | Non-dominant radius SXA | No association between OC use and BMD | |
| Wanichsetakul et al51 | 155 women (ages 30–34) | OC users (n = 59) v DMPA (n = 34) v control (n = 62) | Lumbar spine, femoral neck, Ward’s triangle, greater trochanter, radius, ulna DPA | No difference in BMD between OC users and control | |
| Afghani et al52 | 39 Hispanic pre-/peri-menopausal women (ages 22–51) | Current OC user v non-user | Whole body DXA | No relation between current OC use and BMD (but no info re duration of use, past use, dose, etc) | |
| Meyer et al53 | 61 women (40 athletes (19 eumenorrhoeic, 21 oligoamenorrhoeic) 21 eumenorrhoeic non-athletes; mean age 26 years) | Current OC user v non-user | Areal BMD of whole body, lumbar spine, proximal femur, femoral neck, greater trochanter | No association between OC use and areal BMD in athlete group | |
| Case series (level 4) | Mais et al54 | 19 women (ages 20–30) | 20 μg EE + 0.15 mg desogestrel for 12 months | Distal radius DPA; serum BSAP, urinary hydroxyproline:Cr | NS increase in BMD; decrease in BSAP, hydroxyproline (suggesting decreased bone turnover) |