DMPA (Depot Medroxyprogesterone Acetate)

DMPA can be used for contraception in women who have contraindications to estrogen therapy. It provides continuous contraception for three months, improving compliance and efficacy. DMPA contains medroxyprogesterone, a synthetic long-acting progesterone. It can be given as a 150 mg intramuscular injection or as 104 mg subcutaneous injection every three months. Following IM injection, plasma levels increase for approximately three weeks peaking at seven ng/mL, then declining to undetectable between 120 to 200 days post-injection.^85,86 Ovulation usually resumes once DMPA levels are less than 0.1 ng/ml.⁸⁷ Amenorrhea occurs in approximately 50% of chronic DMPA users.16*

Subcutaneous injection has a slower, more sustained absorption than IM; peak levels are reduced by approximately half while generating the same length of contraceptive protection.⁸⁸ Subcutaneous injections are associated with an increased risk of injection site reactions compared to IM injections (8.0% vs. 0.4%). The reported reactions include dimpling of the skin, redness at the site of infection, rash, itching, bruising, blistering, and lipoatrophy.⁸⁹ The unintended pregnancy rate with DMPA is approximately 6% during the first year of use.17* With perfect use, the breakthrough pregnancy rate is 0.2%.17*

DMPA’s efficacy is not reduced in patients with an elevated BMI.⁹⁰ DMPA can cause significant menstrual changes, including spotting or unscheduled bleeding, primarily upon initiation of therapy. Unscheduled bleeding decreased from 57% at 12 months to 32% at 24 months of use.86* Persistent spotting or unscheduled bleeding is a leading cause of patients switching to another form of contraception. Heavy or bothersome bleeding warrants a referral to a physician for further evaluation.

DMPA can be initiated anytime as long as pregnancy has been reasonably excluded. If more than seven days have elapsed since the beginning of the menstrual cycle, barrier contraception or abstinence is recommended. If unprotected or inadequately protected intercourse occurred during that window, emergency contraception could be considered, and pregnancy testing done in one to two weeks. DMPA, like other progesterone-only hormonal contraceptives, is not associated with significant weight gain.28* DMPA is not associated with changes in mood or depressive symptoms.⁹¹ DMPA as a contraceptive in women aged 18 to 45 outweighs the risk of impacting bone density.12* When users of DMPA are compared to non-users, bone mineral density (BMD) at the hip and spine decreases by 0.5% to 3.5% after one year, then 5.7% to 7.5% after two years, and 5.2% to 5.4% after five years of use.^92-95 However, this was reversed after discontinuation of the drug. DMPA use does not appear to reduce peak bone mass nor increase the risk of osteoporotic fractures later in life for those at average risk for developing osteoporosis.95* Another retrospective cohort study with 1.7 million person-years found that the fracture risk in DMPA patients did not increase significantly after starting the contraceptive (incidence ratio 1.01 95% CI 92-1.11).⁹⁶ DMPA has a black box warning that women who use DMPA may lose significant BMD and recommends that DMPA not be used for longer than two years unless other birth control methods are inadequate.

Data on VTE and thrombosis risk with DMPA is lacking. USMEC recommendations state DMPA is a category two option for women with a history of DVT/PE or known thrombogenic mutations.16* DMPA is category three for women with multiple CVD risk factors, poorly controlled hypertension, and previous history of ischemic cardiovascular disease. DMPA use may acutely lower HDL levels and elevate LDL levels, however, this returns to baseline after approximately six months of therapy.⁹⁷ DMPA use is associated with mild increases in fasting insulin and glucose levels over baseline in some women; however, these changes are unlikely to be clinically significant.⁹⁸

When switching from another contraceptive to DMPA, the previous method should be discontinued seven days after the first DMPA injection; otherwise, backup contraception is necessary for seven days. If the patient was previously using an IUD for contraception, the DMPA injection should be given seven days before IUD removal. For seven days before IUD removal, abstinence or barrier contraception can be recommended to prevent unwanted pregnancy. Women two weeks late for their DMPA injection should be given a pregnancy test and use a backup contraceptive method for seven days after the next DMPA injection.5*

Fertility return can be delayed significantly, and women who desire pregnancy within a year should be referred to another type of contraceptive. Within ten months of the last injection, half of the women who stop DMPA to become pregnant will conceive; however, a return to fertility may take up to 18 months after the final injection.⁹⁹

Etonogestrel Subdermal Implant

Contraception is available via a subdermal implant placed in the upper arm. Clinicians who want to administer the device must complete a three-hour training. Information regarding the training can be found on the manufacturer’s website. The implant is radio-opaque, and insertion failures are rare as the cap will not open if the implant is not present at the tip of the needle. A finger lever ensures the implant completely discharges under the skin once the device is fully advanced. The implant is a 40 mm by 2 mm rod containing etonogestrel. It releases 60 to 70 mcg daily for the first year, decreasing to 45 to 35 mcg per day in year one, 30 to 40 mcg per day in two, and plateauing at 25 to 30 mcg per day from year three to five.¹⁰⁰ Etonogestrel implants need to be changed after five years; however, the manufacturer recommends changing after three years. A prospective trial indicates that the mean serum concentration of progestin is sufficient to maintain contraception efficacy for five years.¹⁰¹ The etonogestrel implant shares the same advantages as other long-acting and progestin-only contraceptives reviewed previously. The disadvantages are similar to long-acting progestin contraceptives; there is an increased risk of unscheduled or breakthrough bleeding compared to COCs, but less than progestin-containing IUDs.

Etonogestrel subdermal implant is highly effective for preventing unwanted pregnancies, with a Pearl Index of 0.38 pregnancies per 100 women-years; this is equivalent to other long-acting forms of contraception.¹⁰² Local adverse reactions occur in approximately 9% of women and include local erythema, bruising, pain, swelling, itching, irritation, fibrosis, and hematoma formation.102* Systemic side effects are similar to other progestin-only contraceptives, including nausea, vomiting, and breast tenderness. Unscheduled, irregular, or heavy bleeding was reported in 11% to 14.8%, with a mean number of bleeding days per 90-day period of 7.3 days.102*,103 The highest number of bleeding days occurred during the first three months, decreased over the next year, and then plateaued in year two or three.

Reduced insulin levels, HOMA-IR, total cholesterol, HDL, and LDL, were reported in 21 women followed for three years.¹⁰⁴ The risk of arterial thrombosis was not increased among 24,954 implant users compared to nine million combined oral hormonal contraceptive non-users.¹⁰⁵ Another study in postpartum patients receiving etonogestrel implants reported similar 30-day VTE admission rates between those who received the implant and those who did not.¹⁰⁶ Etonogestrel implant’s efficacy does not seem to be impacted by BMI.¹⁰⁷ Barrier contraception or abstinence is recommended for seven days after insertion if the implant is placed five or more days after the start of the patient’s menstrual period. The patient’s previous method of contraception can be used for seven days as a backup method. When switching from an IUD, it is advised to retain the IUD for seven days after insertion or use backup contraception for seven days before removing the IUD and inserting the implant. In this scenario, if five or more days have passed since a new menstrual cycle, backup contraception is recommended for seven days, or use emergency contraception at the time of removal and use backup contraceptive methods after insertion of an etonogestrel implant for seven days. When the rod is removed, fertility can reoccur quickly. Etonogestrel levels are undetectable within one week, and approximately 90% of women will ovulate within four weeks of removal. One-year pregnancy rates are as high as 90%.¹⁰⁸ Patients requesting a subdermal implant should be referred to primary care physician who offers subdermal contraception.

Intrauterine Devices (IUDs) 

IUD users account for 23% of female contraceptive users, and IUD use is increasing amongst women in the US.^109,110 IUDs offer long-acting contraception and are the most effective method for preventing unwanted pregnancies. Continuation rates for IUD usage are higher than the other forms of contraception. For LNG IUDs, the three-year continuation rate is 69.8%, copper IUDs 69.7%, etonogestrel implants 56.2%, and 31.0% combined for the non-long-acting reversible contraceptive methods.¹¹¹ After 2.1 years of use, IUDs become the most cost-effective method of contraception.¹¹²

Two types of IUDs are available for contraceptive use, Copper and LNG. The copper IUD is a T-shaped device that contains 380 mm2 that provides ten years of contraceptive protection with a single application.¹¹³ At the base of the device is a 3mm ball to decrease the risk of cervical perforation. A white polyethylene monofilament string is wound through this ball to facilitate removal and serve as an alert that the IUD is still in place. The device contains barium to make it radiopaque and is latex-free. A copper IUD can elevate serum copper levels, but this is only clinically relevant for patients with copper overload or an allergy to copper.¹¹⁴

The pregnancy rate in the first year of use is 0.5% to 0.8%, similar to surgical sterilization.^115,116 Non-contraceptive benefits include continued menstruation and reduced risk of cervical cancer and ovarian cancer.^117,118 Cu-IUD may also reduce the risk of endometrial cancer based on observations of reduced endometrial mitotic activity and estrogen receptor concentration.¹¹⁹ There are several mechanisms of action of Cu-IUD to prevent unwanted pregnancy. Foreign bodies in the uterus cause a sterile inflammatory reaction with the activation of inflammatory mediators and enzymes that impair sperm motility, reduce sperm viability, and provide a toxic environment for fertilization and implantation.¹²⁰ Copper enhances cytotoxic inflammatory responses in the endometrium, impairing sperm viability, movement, and implantation.¹²¹ Ovulation is not interrupted by Cu-IUD.

The risk of expulsion varies between 3% to 10%.^122,123 Menses may be heavier, longer, more painful, or dysregulated, especially during the first year. At three months, Cu-IUD users, compared to LNG-IUD users, reported more cramping (63% vs. 32%), increased volume of bleeding (71% vs. 12%) and increased bleeding frequency (41% vs. 33%). These symptoms improved with time, and by six months, the results were comparable for cramping (14% vs. 12%), increased bleeding volume (19 vs. 8%), and increased bleeding frequency (13% vs. 11%).¹²⁴ Nonsteroidal anti-inflammatory drugs (NSAIDs) may help decrease total blood loss and duration of bleeding, more so in women with a history of heavy menstrual bleeding.¹²⁵ The volume of monthly blood loss may increase by approximately 55%; however, this does not result in anemia for most users. Hemoglobin decreased by 0.36 to 0.94 g/dL in a 12-month period for not anemic users.¹²⁶

Currently, there are four LNG IUDs approved for use in the US. This information is included for completeness, however, IUDs are not prescribed or placed by pharmacists.

Table 10 – LNG-IUDs

The lower-dose LNG-IUDs are smaller and may be more practical for women with small uterine cavities. LNG 13.5 and 19.5 have a silver ring that facilities identification on ultrasound, and all four contain barium which makes them radiopaque. As previously discussed, LNG-IUDs add the progestin effect of increasing cervical mucus and decreasing endometrial viability. The LNG acts mainly at the endometrium level; systemic levels are lower than other forms of LNG contraception. The plasma concentration is highest seven days after insertion, with plasma levels between 100 and 250 pg/ml. This is approximately half of the plasma levels with etonogestrel implants (350 pg/mL) and far lower than progestin-only pills (1500 to 2000 pg/mL).^131,132 LNG IUDs can impact ovulation depending on the dose; 45% of cycles were ovulatory for women using 52 mg LNG-IUD compared to 88 to 97% for lower dose IUDs.127-130*

Amenorrhea at one year occurs in 20% of 52 mg LNG users, 12% of 19.5 mg users, and 6% of LNG 13.5 users. After three years, this increases to 30% to 50% of LNG 52 mg users, 20% of LNG 19.5 users, and 12% of LNG 13.5 mg users. Women who desire a monthly menstrual cycle will benefit from a lower dose LNG IUD while women who wish to avoid menstruation should be referred to a 52 mg LNG IUD.127-130* The first-year pregnancy rate is 0.1 to 0.2%, with a five-year cumulative pregnancy rate of 0.7%.^133,134

In head-to-head comparisons, failure rates with both versions of IUDs are low. Cu-IUD failure rates may be slightly higher, although the data is conflicting. A study with 58,000 participants demonstrated a one-year rate of 0.52 vs. 0.06 pregnancies per 100 woman-years with Cu-IUD and LNG-IUDs.115* An important caveat is the Cu-IUD arm used over 30 types of IUDs, with several containing less than 380 mm2of copper. Cu-IUDs below this threshold are less effective.¹³⁵ Another study compared laparoscopic tubal ligation with LNG-IUD or a 380 mm2 Cu-IUD. The unintended pregnancy rates are similar between tubal ligation and Cu-IUDs (0.92, 95% CI 0.82-1.05) and lower with LNG-IUDs (0.72, 95% CI 0.64-0.82). Non-contraceptive benefits include reducing menstrual bleeding and endometriosis-related pain, endometrial hyperplasia, cervical cancer, endometrial cancer, and ovarian cancer.117*,136 Expulsion during the first year of use was reported to occur in between 3% to 6% of IUD users.122,123*

The overall risk of pelvic inflammatory disease (PID) is low and is relatively similar to developing PID using COCs and the etonogestrel subdermal implant.¹³⁷ Other studies concluded the risk of PID was the same or even lower for IUD users than non-users.^138,139

The overall risk of ectopic pregnancy while using an IUD is lower than that of the general population because of the high efficacy that IUDs possess in preventing pregnancy. Women without contraception have a ten-fold higher risk of ectopic pregnancy compared to IUD users (3.25 to 5.25 per 1000 woman-years vs. 0 to 0.5 per 1000 woman-years).^140,141 If breakthrough pregnancy does occur, IUD users have a higher risk of developing an ectopic pregnancy than non-users.¹⁴² LNG-IUDs are less likely than copper IUDs to undergo contraceptive failure, but if pregnancy occurs, LNG-IUDs are more likely to be ectopic (27% vs. 50%). The adjusted hazard ratio for ectopic pregnancy for IUD use is 0.26 (95% CI 0.10-0.66).115*,143

Special contraindications to IUD therapy are known distortions of the uterine cavity, active PID, known or suspected pregnancy, Wilson’s disease, or copper allergy (Cu-IUD), unexplained uterine bleeding, and breast cancer (LNG-IUD). The usual precautions mentioned earlier for progestins apply to LNG-IUDs. Women with IUDs can safely undergo MRI procedures, but they should inform the radiologist that an IUD is present as this may change various aspects of the study. The manufacturer of the copper IUD states that 1.5 Tesla is safe, while the 52 mg LNG is safe for magnetic fields less than 3.0 Tesla.113*,133 Longitudinal studies report a higher prevalence of bacterial vaginosis (BV) assessed by Nugent score in Cu-IUD users than the general population (27% vs. 49%) at time zero compared to six months.¹⁴⁴ Similarly, a longitudinal study reported a higher incidence of BV by Nugent score for LNG-IUD users (37%) compared to CHC users (19.3%).¹⁴⁵

IUD malposition is common and may affect up to 10% of the IUD population. Uterine perforation is a rare event, with an incidence rate of 0.01%. This risk appears to be increased in breastfeeding women. A six-fold increase (RR 6.1 95% CI 3.9-9.6) was found in a prospective study of 61,000 women breastfeeding with an IUD. The overall risk remained small at 1%.¹⁴⁶ The most common reasons for discontinuation of IUD therapy within the first six months are uterine pain (28% to 35% of discontinuations) and irregular bleeding (10% to 19% of discontinuations). Current guidelines support using IUDs in women with leiomyomas; however, there is a possible increased risk of expulsion (11% vs. 3%).¹⁴⁷ The overall discontinuation rate is low, 7.3% for LNG-IUDs and 8% for Cu-IUDs.¹⁴⁸

Women who desire to return to fertility will need the IUD removed. After removal, pregnancy rates are similar between former users and non-users except for a reduced pregnancy rate in African American women. It is unknown how much time is necessary for the endometrium to revert to its previous status.^149,150 Both Cu-IUD and LNG-IUDs need to be inserted aseptically by a healthcare professional. For LNG-IUDs, evaluate the patient four to six weeks after insertion and then yearly. For Cu-IUD, examine the patient after first menses to ensure the device is still inserted appropriately. If a patient is switching from an IUD to an oral contraceptive method, the woman should retain the IUD for two days after starting POPs or five days after beginning COCs. Abstain from sexual intercourse or use barrier contraception for seven days before IUD removal and COC initiation. If there has been unprotected or inadequately protected sexual intercourse within seven days of IUD removal, it is appropriate to use emergency contraception when the IUD is removed to prevent pregnancy.50* Women should be educated on the feel of the IUD string and see a provider if they cannot feel the string and use backup contraception in the meantime.