Rationale: The IUD prevents pregnancy if inserted before implantation (268).

• Immediately postplacental or during or immediately after a cesarean section (special training required).
• Prior to hospital discharge (up to 48 hours after delivery) special training required).

Recommendation: As early as 4 to 6 weeks postpartum, to accommodate women who come to the clinic for routine postpartum care and who request an IUD. Copper T IUDs may be safely inserted at this time. For other types of IUDs, it may be prudent to wait until 6 weeks postpartum.

Rationale: A copper T may be safely inserted at 4 or more weeks postpartum. The withdrawal technique for copper T insertion presumably helps minimize perforations when inserting IUDs at the routine 4- or 6-week postpartum visit. Other IUDs that have a different profile or a push insertion technique might have different perforation rates. Given the relative lack of information on other IUDs at 4 to 6 weeks postpartum, it is prudent to wait until 6 weeks for the insertion of IUDs other than copper Ts (37, 192, 203).

Recommendation: In breastfeeding women.

Rationale: It has been shown that IUDs can be safely used in breastfeeding women (77).

Rationale: Immediate insertions during cesarean sections by a properly trained provider have a lower expulsion rate than vaginal insertions immediately (within 10 minutes) after delivery. Studies also found that women with IUDs inserted at the time of cesarean section had longer continuation rates (317, 321).

Recommendation: If an IUD is not inserted at the time of the cesarean section, it is recommended that the IUD be inserted no earlier than 6 weeks after the cesarean section.

Rationale: Delayed postpartum insertions should take place no earlier than 6 weeks after cesarean section because of the risk of uterine perforation. Clients need careful assessment for presence of infection before insertion even at this time (189).

Rationale: With appropriate technique, IUDs can be safely inserted postabortion (spontaneous or induced). Expulsion rates vary greatly depending on both the IUD type and provider. To minimize risk of expulsion, only providers with proper training (according to relevant national or institutional standards) and experience should insert IUDs. Clients should be carefully trained to detect expulsions.

Fertility returns almost immediately postabortion (spontaneous or induced): within 2 weeks for first-trimester abortion and within 4 weeks for second-trimester abortion. Within 6 weeks of abortion, 75% of women have ovulated (164).

Recommendation: IUDs should not be inserted in the following situations:

• With confirmed or presumptive diagnosis of infection (signs of unsafe or unclean induced abortion, signs and symptoms of sepsis or infection, or inability to rule out infection), do not insert an IUD until risk of infection has been ruled out or infection has fully resolved (approximately 3 months).
• With serious trauma to the genital tract (uterine perforation, serious vaginal or cervical trauma, chemical burns), do not insert an IUD until trauma has healed.
• With hemorrhage and severe anemia, IUDs (inert or copper-bearing) are not advised until hemorrhage or severe anemia is resolved. However, progestin-releasing IUDs can be used with severe anemia (they decrease menstrual blood loss).
• Postabortion IUD insertion after 16 weeks gestation requires special training of the provider for correct fundal placement. If this is not possible, delay insertion for 6 weeks.

The client should be strongly encouraged to come to the clinic any time she has questions or problems, particularly if she has:

• Late period (possible pregnancy),
• Prolonged or excessive abnormal spotting or bleeding,
• Abdominal pain or pain with intercourse,
• Infection exposure (such as gonorrhea), abnormal vaginal discharge, or pelvic pain especially with fever, or
• Strings missing or strings seem shorter or longer.

Rationale: A follow-up visit at 3 to 6 weeks is prudent since the peak incidence of pelvic inflammatory disease (PID) post-IUD insertion is at 1 month. Thereafter, there is no need for a fixed follow-up schedule. The best quality of care is to focus clinic resources and attention on those women who come back to the clinic with complaints or problems (76, 131).

Rationale: There is no medical need for a pre-exam (separate visit); it may be difficult for a woman to make two visits, and she may be at risk of pregnancy during this interval.

To receive IUDs, women should not be at increased risk of STDs. Counseling should focus on the risk of STDs, PID, and the possible risk of resulting infertility.

Rationale: Young women statistically have a higher risk of PID. IUDs, in contrast to all other modern contraceptive methods, increase the risk of PID when a woman is infected with an STD. PID is a major risk factor of tubal infertility and ectopic pregnancies. Because young women may have patterns of sexual activity that lead to STD risk, the relative risk of PID in young IUD users may be high. Additionally, nulliparous women receiving IUDs may be at higher risk for expulsion.

The degree to which client values future fertility is an important factor in the choice of a contraceptive method. Studies have shown that the risk of PID and subsequent tubal-factor infertility is directly proportional to the risk of exposure to sexually transmitted disease. IUDs do not protect women against PID or other STDs.

Nevertheless, women should be allowed to make their own choice (7, 181, 215, 313).

• The woman requests removal,
• The woman develops precautions/contraindications, or
• The effective life of the IUD is reached.

Make sure removal of the first IUD is indicated (i.e., for reasons of partial expulsion without infection or expiration of the IUD).

Rationale: Even with proper technique, the removal and reinsertion of an IUD expose a woman to the risk of introduction of vaginal and endocervical canal micro-organisms into the upper genital tract. Therefore, removal and insertion at the same time avoid two separate exposures (76).

In an interval between removal of one IUD and insertion of another, the woman will not be protected against pregnancy by the method of her choice.

If PID, mucopurulent endocervical discharge, cervicitis, or clinically apparent vaginitis is present, do not insert an IUD but treat for infection. Consider other contraceptive methods, if an STD* is suspected.

* Not all clinically apparent vaginal infections are due to STDs.

Rationale: Currently available lab tests may be impractical and often unaffordable (even in the developed world) to rule out endocervical colonization by infectious agents capable of ascending and causing PID. Most chlamydia tests are only 80% to 90% sensitive, tests for mycoplasma and ureaplasma are not routinely available, and cervical gram stain is less sensitive for gonorrhea. However, where gonorrhea culture and chlamydia tests are affordable, negative test results provide reassurance to corroborate the woman's history (21, 154, 198).

If she remains at increased risk of PID, advise against IUD use.

Rationale: PID may take several weeks to resolve clinically, and, in the case of severe PID, waiting several months in theory allows healthy tissues (free of micro-abscesses) to form (267).

Women with prior PID are at increased risk of repeat PID. A woman who has had an episode of upper reproductive tract infection may be at increased risk of repeat episodes of nonsexually transmitted PID regardless of IUD use. Theoretically, a previous episode of upper reproductive tract infection may result in tubal damage increasing susceptibility of the fallopian tubes to opportunistic lower genital tract flora (139, 294).

Give nutritional advice on the need to increase the intake of iron-containing foods.

Rationale: In general, IUDs (especially inert IUDs) commonly increase the amount of menstrual blood loss, varying according to IUD type, particularly in the first few months postinsertion (42).

Copper IUDs may increase normal menstrual blood loss by 50%, which may be clinically significant for women who are already anemic. (Progestin-releasing IUDs decrease menstrual blood loss; the more progestin an IUD releases, the more effectively it decreases menstrual blood loss.) (69)

Recommendation: For mild to moderate bleeding and pain in the first month postinsertion, with no evidence of clinically apparent pelvic infection, and if reassurance is not sufficient but the woman wants to keep the IUD, a short course of a nonsteroidal anti-inflammatory agent other than aspirin (e.g., ibuprofen) may be given.

Rationale: Nonsteroidal anti-inflammatory drugs (e.g., ibuprofen*) decrease uterine bleeding and cramping (9).

* Nonsteroidal anti-inflammatory drugs (e.g. ibuprofen) should be used instead of aspirin because of aspirin's stronger and longer-lasting inhibitory effects on platelet aggregation (aspirin promotes bleeding) (5,83).

Recommendation: Bleeding generally decreases over time. If bleeding is heavy or the woman is anemic, treatment with oral iron can improve hemoglobin levels.

If bleeding or pain is severe, or the client wishes to discontinue use, remove the IUD.

If suspected, abnormal conditions that cause prolonged or heavy bleeding should be evaluated and treated as appropriate.

If pelvic infection is diagnosed, remove the IUD and treat with antibiotics. (In the case of mild uterine tenderness without any other evidence of pelvic infection, broad-spectrum antibiotics or chemotherapeutics may solve the problem; use clinical judgment regarding whether or not to remove the IUD).

Rationale: Nurses and midwives have been shown to have equal or superior competence in IUD insertion when compared with doctors (73).

• Appropriate handwashing by the provider and careful preparation of the cervix,
• Sterile (or high-level disinfected) IUDs and equipment,
• Correct decontamination of instruments, and
• Safe disposal of contaminated disposables.

Sterilization is the safest and most effective method for processing instruments that come in contact with the bloodstream, tissue beneath the skin, or tissues that are normally sterile. When sterilization equipment is either not available or not suitable, high-level disinfection (HLD) is the only acceptable alternative. HLD destroys all microorganisms, including viruses causing hepatitis B and AIDS, but it does not reliably kill all bacterial endospores. For example, in family planning facilities, either sterilization or HLD is acceptable for processing instruments and gloves used for pelvic exams and IUD insertion and removal, since problems with endospores (Clostridia species) have not been reported with IUD use. Regardless of the method selected, however, HLD can only be effective when used (soiled) instruments and gloves are first decontaminated, thoroughly cleaned, and rinsed before disinfection (272).

Contaminated wastes may carry high loads of microorganisms that are potentially infectious to any persons who contact or handle the waste. Incineration provides high temperatures and destroys microorganisms; therefore it is the best method for disposal of contaminated wastes. Incineration also reduces the bulk size of wastes to be buried. If incineration is not possible, all contaminated wastes must be buried to prevent scattering the waste materials (272).

There is a theoretical rationale for the practice of giving prophylactic antibiotics. PID rates in IUD users are highest in the first few weeks, and antibiotics could reduce those PID rates. While there is no statistically significant proof for reduction in PID rates, one study found a lower rate of IUD-related unplanned returns to the clinic.

Arguments against the use of prophylactic antibiotics include the insignificant impact of antibiotics on reducing the PID rates in IUD users demonstrated in previous studies. Also, although the rate of PID in IUD users is highest in the first few weeks after insertion, due to the long duration of use of IUDs the greatest numbers of PID cases will occur after the first few weeks after insertion. In addition, there is some concern about the programmatic feasibility and cost of prophylactic antibiotics.

Good infection control procedures, proper assessment of the client's risk for STDs, and proper insertions are very important to keep the rate of PID low in IUD clients.

Rationale: The scientific literature does not show any large reduction in PID rates by giving prophylactic antibiotics for IUD insertion (76, 160, 252, 287, 324). However, in each of the studies, infection-prevention procedures were followed, and the rates of PID were very low. Also, the sample sizes in the studies were small.

Although the differences were not statistically significant, three studies all showed some reduction in the PID rate in women given prophylactic antibiotics.

Sinei et al. found that the PID rates for the first month after IUD insertion in women who were given doxycycline was 1.3% compared with 1.9% in the women who received a placebo. They also found that the women who received a placebo returned to the clinic for IUD-related problems that were suggestive of subclinical PID more often than the treated women (252).

Zorlu et al. found infection rates to be 2.1% and 2.9% in doxycycline-treated and untreated women, respectively, within the first 3 months after IUD insertion (325).

Walsh et al. found that within the first 3 months after IUD insertion, 3.6% of the doxycycline group had the IUD removed for medical reasons compared with 4.5% of the placebo group (287).

Rationale: Cervical ectropion (the presence on the ectocervix of columnar epithelial cells from the endocervix) is a normal condition in adolescents and in pregnancy, and is distinct from cervical infection (209).

IUD insertions and continued use of the IUD have no relation to risk of cervical carcinoma (170).