Table 10.
DMPA injections and oral contraceptive use on bone health in adolescents
| Source | Study description | Population description | Number of subjects | End points | Results | ||
|---|---|---|---|---|---|---|---|
| Observational studies | |||||||
| Lloyd et al. 2000 [223] | An 8-year longitudinal study of white females in the Penn State Young Women’s Health Study. OC users were those individuals who had used low-dose monophasic OCs for a minimum of 6 months at age ~12 and were still using them at age 20 years. | Sex: 62 females Age: 11.9 ± 0.5 years at baseline Race: Caucasian Location: USA |
62 | Total body BMC (g) at 12 years | OC users | Nonusers | |
| 1463.0 ± 64.0 | 1402.5 ± 45.5 | ||||||
| Total body BMC (g) at 20 years | 2272.2 ± 55.9 | 2214.2 ± 45.1 | |||||
| Total body BMD (g/cm2) at 12 years | 0.92 ± 0.01 | 0.91 ± 0.01 | |||||
| Total body BMD (g/cm2) at 20 years | 1.13 ± 0.01 | 1.12 ± 0.01 | |||||
| Hip BMD (g/cm2) at 20 years | 1.01 ± 0.02 | 0.99 ± 0.02 | |||||
| The OC users did not differ in anthropometric, body, or total bone measurements at baseline or at age 20 years (no differences in age at menarche or exercise at age 20 years). OC pill use by healthy white teenage females did not affect acquisition of peak bone mass. |
|||||||
| Lara-Torre et al. 2004 [225] | Nonrandomized prospective study to examine BMD in control adolescent subjects vs those receiving DMPA injections or OCs over a 2-year period | Sex: 71 females Age: 11–19 years Race: Caucasian and black Location: USA |
148 | OC usersb | DMPA injectionb | Nonuserb | |
| % change in lumbar BMD at 6 months | 1.170 (−0.14, 2.48) | −0.249 (−1.25, −0.75)* | 2.768 (0.48, 5.05) | ||||
| % change in lumbar BMD at 12 months | 2.35 (0.16, 3.78) | −1.59 (−2.53, 0.59)* | 2.45 (−2.01, 5.99) | ||||
| % change in lumbar BMD at 18 months | 3.82 (−0.62, 6.41) | −2.91 (−3.64, −2.23)* | 0.73 (−1.01, 1.54) | ||||
| % change in lumbar BMD at 24 months | −1.01 (−1.23, 5.33) | −1.85 (−4.15, −0.23)* | 5.89 (5.37, 6.85) | ||||
| There was a statistically significant difference in BMD between DMPA users and the control at 6, 12, 18, and 24 months. There was no statistical difference between OC pill users and the control at any time point. | |||||||
| Lloyd et al. 2004 [224] | A 10-year longitudinal study of white females in the Penn State Young Women’s Health Study. OC users were those individuals who had used low-dose monophasic OCs for a minimum of 6 months at age ~12 years and were still using them at age 22 years. | Sex: 80 females Age: 21.7 ± 0.1 years Race: Caucasian Location: USA |
80 | OC users | Nonusers | ||
| Total body BMC (g) at 22 years | 2260 ± 55 | 2316 ± 66 | |||||
| Total body BMD (g/cm2) at 22 years | 1.14 ± 0.01 | 1.15 ± 0.02 | |||||
| Hip BMD (g/cm2) at 22 years | 1.0 ± 0.02 | 1.0 ± 0.02 | |||||
| Femoral neck section modulus (cm3) at 22 years | 1.29 ± 0.05 | 1.32 ± 0.05 | |||||
| Femoral neck section modulus (cm3) at 22 years | 1.70 ± 0.06 | 1.74 ± 0.07 | |||||
| OC pill use among adolescents was not correlated with bone or body composition measurements. | |||||||
| Rome et al. 2004 [227] | Prospective, observational design study to examine the effects of DMPA injections or OC (20 μg ethinyl estradiol/100 μg levonorgestrel) use on bone biochemical markers over 12 months | Sex: 165 females Age: 11–18 years Race: Caucasian and black Location: USA |
370 | Spine and femoral neck BMD | No significant differences reported between spine and femoral neck BMD for each group at each time point. Spine BMD was lower in the DMPA injection group vs the OC group. Exact values not reported | ||
| Cromer et al. 2008 [226] | Observational prospective cohort study to examine the effects of DMPA injections or OC (20 μg ethinyl estradiol/100 μg levonorgestrel) use on BMD over 24 months | Sex: 375 females Age: 12–18 years Race: Caucasian and black Location: USA |
433 | OC users | DMPA injection | Nonuser | |
| Spine BMD (g/cm2) | 1.03 ± 0.11 | 0.98 ± 0.09* | 0.98 ± 0.11 | ||||
| Femoral neck BMD (g/cm3) | 0.97 ± 0.14 | 0.92 ± 0.14* | 0.92 ± 0.15 | ||||
| Over 24 months, mean percent change in spine BMD was −1.5 % (DMPA injection), 4.2 % (OC use), and 6.3 % (control). Over 24 months, mean percent change in femoral neck BMD was −5.2 % (DMPA injection), 3.0 % (OC use), and 3.8 % (control). Adolescent girls receiving DMPA injections had significant loss of BMD. Clinical significance of this loss may be mitigated by the slowed loss after the first year of DMPA use. |
|||||||
| Pikkarainen et al. 2008 [228] | 4-year follow-up study examining the effects of estrogen-progestin OC use | Sex: 122 females Age: 12–19 years Race: Assumed predominantly Caucasian Location: Finland |
122 | OC user (1–2 years) | OC user (>2 years) | Nonuser | |
| Lumbar spine area (cm2) at baseline | 56.1 ± 6.1 | 56.7 ± 5.9 | 55.1 ± 5.9 | ||||
| Lumbar spine area (cm2) at 4 years | 58.6 ± 6.1 | 58.7 ± 0.7 | 59.5 ± 6.3 | ||||
| Femoral neck area (cm2) at baseline | 4.79 ± 4.33 | 4.64 ± 0.37 | 4.66 ± 0.40 | ||||
| Femoral neck area (cm2) at 4 years | 4.80 ± 0.3 | 4.7 ± 0.4 | 4.7 ± 0.40 | ||||
| Lumbar spine BMC (g) at baseline | 54.3 ± 11.8 | 57.5 ± 11.9 | 51.2 ± 10.2 | ||||
| Lumbar spine BMC (g) at 4 years | 59.10 ± 11.7 | 60.2 ± 12.3 | 59.2 ± 10.0 | ||||
| Femoral neck BMC (g) at baseline | 4.3 ± 0.8 | 4.4 ± 0.8 | 4.1 ± 0.6 | ||||
| Femoral neck BMC (g) at 4 years | 4.5 ± 0.8 | 4.4 ± 0.7 | 4.3 ± 0.6 | ||||
| There was a significant trend of a lesser increase of BMC in lumbar spine in the OC users (>2 years) than in the other two groups (P < 0.0046). The development of the femoral neck was significantly different between the OC users (1–2 years) and OC users (>2 years). The longer duration of OC use seemed to suppress normal BMC development (P for linearity = 0.038) |
|||||||
| Walsh et al. 2008 [230] | A case-controlled matched study aiming to examine whether the effects of DMPA injections are age-specific and to determine the effects of DMPA on hormones and bone turnover | Sex: 100 females (50 pairs) Age: 12–18 years Race: Caucasian and black Location: USA |
100 | DMPA injectiona | Nonusersa | ||
| Lumbar spine BMD (g/cm2) | 1.003 (0.972–1.035)* | 0.947 (0.950–0.977) | |||||
| Total hip BMD (g/cm2) | 0.983 (0.950–1.016)* | 0.931 (0.897–0.966) | |||||
| Distal forearm BMD (g/cm2) | 0.465 (0.446–0.483) | 0.474 (0.457–0.491) | |||||
| DMPA injections are associated with a bone density deficit at the spine and hip when used before peak bone mass. | |||||||
| Bonny et al. 2011 [222] | Prospective longitudinal study examining the relationship between weight and BMD and the correlation between weight change and BMD in premenarcheal girls reporting DMPA use or OC (20 μg ethinyl estradiol/100 μg levonorgestrel) use | Sex: 433 females Age: 12–18 years Race: Caucasian and black Location: USA |
433 | User | DMPA injection | Nonuser | |
| Correlation between absolute change in weight (kg) and absolute change in femoral neck BMD at 12 months | 1.000 | 1.000 | 1.000 | ||||
| Correlation between absolute change in weight (kg) and absolute change in femoral neck BMD at 24 months | 1.000 | 1.000 | 1.000 | ||||
| Correlation between absolute change in weight (kg) and absolute change in spine BMD at 12 months | −0.82 | −0.101 | 0.10 | ||||
| Correlation between absolute change in weight (kg) and absolute change in spine BMD at 24 months | −0.12 | 0.098 | 0.48 | ||||
| Body weight was significantly positively associated with femoral neck BMD and spine BMD regardless of the contraceptive method (P < 0.05). Change in body weight at 12 and 24 months was highly correlated with change in femoral neck BMD (P < 0.0001) for all treatment groups. No statistically significant correlation between change in weight and change in spine BMD was seen in the DMPA, OC, or control subjects at 12 or 24 months. |
|||||||
| Biason et al. 2015 [229] | Nonrandomized parallel-control study with 1-year follow-up assessing the effect of OC (20 μg ethinyl estradiol/150 μg desogestrel) users on bone density | Sex: 67 females Age: 12–19 years Race: Assumed Brazilian Location: Brazil |
67 | OC user | Nonuser | ||
| % variation lumbar BMD (g/cm2) | 2.07 | 12.16 | |||||
| % variation lumbar BMC (g) | 1.57* | 16.84 | |||||
| % variation subtotal BMD (g/cm2) | 0.56* | 5.28 | |||||
| % variation subtotal BMC (g) | 1.18* | 16.04 | |||||
| % variation whole body BMD (g/cm2) | 0.84 | 5.28 | |||||
| % variation whole body BMC (g) | 1.22* | 11.34 | |||||
| Use of low-dose OCs was associated with lower bone mass acquisition in adolescents over a 1-year period. OC users showed low bone mass acquisition in the lumbar spine and had BMD and BMC median variations of 2.07 and 1.57 %, respectively, between the measurements at baseline and 12 months. Nonusers showed median variations of 12.16 and 16.84 % for BMD and BMC, respectively, over the same period. |
|||||||
95 % CI confidence interval, BMC, bone mineral content, BMD bone mineral density, RCT randomized controlled trial
aMean (95 % CI)
bMedian (25 %, 75 %)
*P < 0.05 compared with control