Am Fam Physician. 1998;58(9):2077-2084
See related patient information handout on IUDs, written by the author of this article.
The intrauterine device, a common form of birth control in the early 1970s, is now avoided by American physicians and women because of concern about complications. This concern is largely the result of the problems reported with use of an intrauterine device that is no longer manufactured. More recent intrauterine devices have an improved design, and reevaluation has shown them to be a safe, efficacious and cost-effective form of birth control. Careful patient selection and preinsertion counseling are crucial to success with the device. Recent studies conclude that the intrauterine device poses no increased risk of pelvic inflammatory disease or infertility when used by appropriately selected patients.
Interest in the IUD started in the 1960s, and its use in this country increased over the next decade. A 1974 study, however, linked the Dalkon Shield (manufactured by A.H. Robins) to maternal death and found it to have a disproportionately higher rate of infection than any other IUD.3 The cause of infection was the multifilament string (or tail), which was a modification of the monofilament tails used by other IUDs.4,5 This multifilament tail provided a pathway for bacteria, enabling them to bypass the immunologic barrier provided by the endocervix. This design flaw caused a fivefold increase in pelvic inflammatory disease (PID) and an increase in septic abortion.5,6 After the Dalkon Shield was removed from the market, the use of IUDs declined in the United States.
Two recent studies on contraception have shed new light on the IUD. One report compared the economic value of 15 different contraceptive methods. The IUD was the most cost-effective method, both in cost per year for the user and in savings for society in health care resources (Table 1).7 Another study found a higher than expected failure rate for tubal ligations, ranging from 0.7 to 5.4 percent (depending on technique) over 10 years.8 This finding would position the IUD as a potentially more effective contraceptive than tubal ligation.
|Device||Medication||Failure rate (%)||Effective period (years)|
|T 380A (ParaGard; Ortho-McNeil Pharmaceutical)||Copper, 380 mm2||< 1||10|
|Progesterone T (Progestasert; Alza Corporation)||Progesterone, 38 mg||2.9||1|
Mechanisms of Action
Many theories have been proposed to explain the contraceptive action of IUDs.12,13 One of the earliest theories was based on studies showing that the presence of a foreign body in the uterine cavity interfered with reproduction in all species tested.9 This result was attributed to a biochemical, noninfectious, inflammatory foreign body reaction that induced prostaglandin release. This release resulted in altered uterine activity and tubal motility, as well as having a direct effect on spermatozoa.9,14
Other proposed mechanisms of action include the migration of white blood cells into the uterine cavity, resulting in the phagocytosis of spermatozoa; the release of an endotoxin during white blood cell breakdown, causing destruction of spermatozoa; and the presence of copper ions (in the case of copper-bearing IUDs), which seem to have a direct toxic effect on spermatozoa.12,13
It is probably a combination of all these factors that makes the IUD such an effective contraceptive. IUDs containing progesterone produce similar effects and, in addition, thicken cervical mucus and suppress ovulation in some women because of systemic absorption of the hormone.9
It is important for the patient to understand that the IUD does not appear to be an abortifacient but, rather, prevents conception. This position is supported by studies that have sought the presence of human chorionic gonadotropin in IUD users and compared tubal flushings of IUD users with those of noncontraceptive control subjects.15,16 These reports are further supported by other research demonstrating that IUDs have a toxic effect on sperm and that women who use IUDs have fewer ectopic pregnancies than noncontraceptive-using control subjects.17,18
IUD Candidates: Patient Selection
The most important part of the decision to use an IUD is proper patient selection. It is critical that the physician know the patient's history and be aware of patient characteristics that increase the risk for complications. Special attention should be given to the patient's history concerning sexually transmitted diseases (STDs) and PID, her menstrual cycle and pattern, previous contraceptive failures and future childbearing plans. The main goal of patient selection is to prevent the inadvertent insertion of an IUD in a patient who has an STD or is at high risk for exposure to one. Table 3 summarizes the characteristics of recommended IUD candidates. Most physicians have not offered the IUD to nulliparous patients, fearing the risk of infection-related infertility; however, if the patient has no contraindications and understands the risks, many physicians have found the IUD an excellent contraceptive for these patients.1,11 Contraindications for both copper-bearing and progesterone-releasing IUDs are listed in Table 4.
|Desire for a reversible, long-term, cost-effective birth control method that allows spontaneity (first choice for these patients)|
|Failure of a previous birth control method due to forgetfulness|
|Not ready for tubal ligation|
|Low risk for sexually transmitted diseases|
|Postpartum or postabortion|
|Childbearing complete, but patient objects to sterilization|
|Contraindications to hormonal contraception (excluding progesterone-releasing IUDs)|
|Previous uncomplicated use of IUD|
|History or presence of acute PID in the past three months|
|Patient (or partner) with multiple sexual partners|
|Postpartum or postabortion endometritis in the past three months|
|Uterine abnormalities resulting in anatomic distortion|
|Uterine or cervical malignancy, including unresolved, abnormal Papanicolaou smear|
|Genital bleeding of unknown etiology|
|Untreated vaginitis or cervicitis, including bacterial vaginosis|
|Previously inserted IUD still in place|
|Conditions or treatments associated with increased susceptibility to infection with microorganisms (e.g., leukemia, type 1 diabetes mellitus, intravenous drug use, AIDS)|
|Previous ectopic pregnancy*|
|Postpartum or postabortion incomplete involution of uterus*|
|Known allergy to copper†|
|Uterus measuring < 6 or > 10 cm by sound*|
Planning Insertion of the IUD
Counseling is the key to patient compliance and satisfaction with contraception. A well-informed patient is more likely to continue with the IUD method and less likely to request premature removal. A good description of possible adverse effects is listed in the patient package insert, and these effects should be reviewed with the patient to ensure that she completely understands the effects and her risk of experiencing them.
The patient should be prepared to experience some cramping and discomfort during insertion and mild spotting and cramping for one to three days after insertion. The bleeding is usually minimal, and the cramping normally responds well to an over-the-counter analgesic such as ibuprofen or acetaminophen. The patient should also be told that if these symptoms become more severe or are not controlled by an analgesic, reevaluation by the physician is recommended.
It is important to instruct the patient to check for the presence of the tail after her first menstrual period and to notify the physician if she feels the stem of the IUD or does not feel a tail. The patient should be encouraged to check for the tail on a regular basis, at least after every menstrual period. Any extended period of amenorrhea should also prompt the patient to see her physician for evaluation.
Historically, placement of the IUD has required at least two patient visits. The first visit includes a thorough counseling session, a history and a pelvic examination, with a screen for gonorrhea and Chlamydia and a Papanicolaou smear. The need for a gonorrhea and Chlamydia screen before all IUD placements has been questioned; however, this precaution is recommended by the American College of Obstetricians and Gynecologists (ACOG).19 Clearly, a pelvic examination should be performed to rule out abnormalities of the genital tract, and cervical and uterine infection.
If the results of Chlamydia or gonorrhea screening are positive, the patient should be reevaluated as a candidate for IUD placement. Abnormal Pap smear results should be investigated with colposcopy and follow-up before IUD placement.
The IUD can be inserted at any time. Some reports recommend insertion during menstruation to verify that the patient is not pregnant, decrease discomfort during placement and avoid inconvenience to the patient by allowing postinsertion bleeding to mix with the menstrual flow.12,13 Many physicians have suggested that the cervical os is more patent during this time; however, no solid evidence substantiates this suggestion.
On the other hand, there are several reasons to avoid IUD placement during menstruation. The menstrual blood could provide a rich nutrient source for the growth of pathogenic organisms, possibly increasing the risk of bacterial contamination and subsequent PID. In addition, the delay in awaiting menses may result in unintended pregnancy. Most physicians continue to recommend that IUD insertion be delayed until six to eight weeks post-partum or until the uterus has completely involuted.12
In some instances, a two-visit requirement for IUD placement may be inappropriate or inconvenient. The physician must use clinical judgment as to the patient's possible risk for infection or cervical disease if placement is carried out at the initial visit. The physician and the patient should be prepared to remove the IUD if cervical disease or infection is found after placement.
Table 51,10 summarizes the steps in placement of an IUD. Using a sound, the distance to the top of the fundus can be measured; it should be between 6 and 9 cm.12,19 The ACOG Technical Bulletin states that a uterine cavity smaller than 6 cm or larger than 9 cm is a contraindication to IUD placement; however, the prescribing information for the ParaGard indicates that only uteri smaller than 6 cm are associated with an increased risk of adverse reactions.10,19 At the completion of the procedure, the tenaculum site should be checked for hemostasis. It is not uncommon for a small amount of bleeding to arise from this site, and simple pressure or silver nitrate can be used for hemostasis.
|1.||Administer preoperative nonsteroidal anti-inflammatory drug.|
|2.||Obtain informed consent.|
|3.||Set up instruments on a sterile field.|
|4.||Complete pelvic examination to check for vaginitis and uterine position.|
|5.||Insert speculum to localize cervix.|
|6.||Clean cervix with antiseptic.|
|7.||Apply tenaculum to anterior lip of cervix.|
|8.||Use sound to measure uterus and note cavity depth.|
|9.||Consider using a paracervical block for patients with cervical stenosis.|
|10.||Load device through sterile packaging.|
|11.||Set stop/flange to depth indicated by sound.|
|12.||Insert IUD until stop/flange is against cervix.|
|13.||Release IUD (method varies with type and design of IUD).|
|14.||Remove insertion device.|
|15.||Cut string(s) to 1 to 2 cm from the cervical os and note length in chart.|
|17.||Observe for bleeding.|
|19.||Discharge patient with appointment for position check and review of side effects after first postinsertion menses.|
Removal of the IUD can take place at any time and is usually sought because of adverse effects, medical complications, patient preference or expiration of the IUD. Complications that may necessitate removal of the IUD include pregnancy, PID, lower genital malignancy, dyspareunia, perforation, partial expulsion and precancerous lesions in the cervix or endocervical canal. Removal does not require any special preparation and is a simple office procedure (Table 6).1,10
|1.||Administer preoperative nonsteroidal anti-inflammatory drug.|
|2.||Insert speculum to localize cervix.|
|3.||Grasp string with forceps or clamp.|
|4.||Apply steady force until IUD is removed.|
|5.||Use cotton swab to probe endocervical canal if string is not visualized.|
|6.||Use IUD hook to probe lower uterine segment and grasp IUD if still unable to locate a string.|
|7.||Consider ultrasound to identify position of IUD if not located.|
|8.||Consider using paracervical block with gentle cervical dilatation if IUD is localized but removal with IUD hook is unsuccessful.|
|9.||Consider removal of IUD under anesthesia with hysteroscopy if still unsuccessful or if the patient is unable to tolerate removal.|
The patient should be counseled to expect some mild cramping and possible postprocedure spotting. A study comparing weighted annual removal rates for the Paragard IUD revealed that the most common reasons for removal were bleeding or pain (2.2 to 11.9 percent), expulsion (zero to 5.7 percent) and pregnancy (zero to 0.7 percent). In all categories, these removal rates were highest in the first year of use.1,10,20,21
|Adverse effects and complications||Characteristics||Treatment|
|Cramping/bleeding||Most common adverse effect (12% of removal requests). Menstrual blood flow usually increases by 10 to 35 mL. Rate of removal for bleeding and cramping increases the later in the menstrual cycle the IUD is inserted.||Administer NSAID. Evaluate for infection if symptoms are persistent or excessive. If symptoms persist for more than three to four cycles or fail to respond to treatment, consider removing the IUD.|
|Expulsion||Occurs in 1 to 7% of women in first year of use. Most frequent in first three months. More common in younger women and nulliparous women. Increased risk in women with painful menstruation or heavy menstrual flow. Replacement IUD inserted after expulsion is less likely to be expelled. Pregnancy may be the first sign of expulsion. Expulsion rate decreases the later in the menstrual cycle the IUD is inserted.||Remove partially expelled IUD. Replace IUD or use other form of birth control.|
|Failure/pregnancy||In the presence of an IUD, 50 to 60% of pregnancies spontaneously abort. One half of these abortions occur in the second trimester. This risk drops to 20% when the IUD is removed. Septic abortion is 26 times more common in patients with an IUD. Copper T IUDs protect against ectopic pregnancy, while progesterone-releasing IUDs increase the risk of ectopic pregnancy almost twofold. This risk decreases as the amount of progesterone increases. The failure rate increases the later in the menstrual cycle the IUD is inserted, especially after day 17.||Perform a pregnancy test if patient has any unusual delay in menses. Perform an ultrasound examination to determine location of the pregnancy and the IUD. Remove IUD as soon as pregnancy is diagnosed.|
|Uterine perforation||Occurs in 0.1 to 0.3% of IUD insertions. Usually happens at the time of insertion. May tear into or through uterine wall or through cervix into vagina.||Remove IUD surgically when perforation is recognized, by laparoscopy, if intra-abdominal.|
|Misplaced IUD string (tail)||May be due to expulsion or perforation. First sign of pregnancy can be a drawing up of the tail into the uterus.||Rule out pregnancy. If pregnancy test is negative, explore endocervical canal for tail. If unable to locate tail, obtain ultrasonograph or radiograph to locate IUD. If no evidence of an IUD, consider inserting new IUD or using another form of birth control. If IUD is present within the uterine cavity, continue with observation or replace it.|
Pelvic Inflammatory Disease
The greatest concern with IUDs is the potential for infection. This risk was overestimated in the 1970s. Reevaluation of data from that time shows considerably lower risk estimates of infection with IUD use.1,6 The overestimation was probably secondary to several factors, including poor choice of control groups (such as users of oral and barrier contraceptives, both of which decrease the risk of infection), overdiagnosis of PID in IUD users and inclusion of data from Dalkon Shield users in the risk assessment figures.12,13
Many factors have been suggested as contributing to the development of PID in IUD users. There has been concern that the presence of a tail provides an avenue by which vaginal pathogens can ascend into the upper genital tract. Electron microscopy has shown that a coating builds on the tail during 12 to 14 months of use.25 This coating is usually thin and contains mostly mucus, and cellular and bacterial debris. The intrauterine portion of the IUD rarely contains any live bacteria. However, as the coating becomes thicker, bacteria can be found on the tail within the uterus and on the device itself.24 Some have theorized that, compared with the earlier multifilament tail, the monofilament tail is associated with far less risk of ascending infection because the total surface of the tail is exposed to the cervical mucus, which is believed to have protective properties against ascending infection.25 This theory is consistent with studies showing that the Dalkon Shield presented a greater risk of ascending infection because its multifilament tail had far less exposure to the cervical mucus.5 Recent studies conclude that the overall contributing factors in the development of ascending genital infection in IUD users are the number of sexual partners and the increased incidence of STDs.1,4,6
Several investigations have shown that the peak in PID risk among IUD users occurs during the first 20 days after insertion.26 The risk following this initial period appears to be low and, in the case of copper-bearing IUDs, the risk is negligible in women in mutually monogamous relationships.4,6,26 A recent study found that patients in mutually monogamous relationships and at low risk for STDs appear to have little, if any, risk of PID with IUD use.12,13,26
Recently, interest has grown in the use of prophylactic antibiotics at the time of insertion to decrease the risk of infection. Several studies were undertaken, and although one study indicated a slight decrease in the risk of PID with the use of prophylactic antibiotics, similar trials failed to show any difference.27,28 ACOG reports that antibiotics “may be considered” and suggests the use of doxycycline, in a dosage of 200 mg orally one hour before insertion, or erythromycin, in a dosage of 500 mg orally one hour before and six hours after insertion.19 A 1992 review of the use of prophylactic antibiotics at the time of IUD insertion found no documented value in antibiotic prophylaxis in patients at low risk for STDs.1
The possibility that the IUD may increase the sexual transmission of the human immunodeficiency virus (HIV) has raised concern. One study found that IUD users were at higher risk for acquiring HIV infection, but research on this issue is not adequate for firm recommendations.29
The treatment of actinomyces discovered on a routine Pap smear is one of the most controversial areas on IUD use in the literature. This rare finding on cervical cytologic specimens is more prevalent in IUD users than in other women.30 There was concern that actinomyces was linked to IUD-associated PID, but a study found that the presence of actinomyces on a routine Pap smear did not correlate with an increased risk of PID or the development of actinomycotic disease.30 Observation of the asymptomatic patient and active management of the symptomatic patient by removing the IUD and administering penicillin or tetracycline is recommended.12,30 The use of antibiotics in asymptomatic patients has not been reviewed thoroughly and cannot be recommended at this time.
Malignancy and Lower Genital Neoplasia
Multiple studies have shown no increased risk for cervical or uterine malignancies in IUD users.12,13 Several reviews have indicated that women using copper IUDs actually have a decreased risk of cervical malignancies with increasing duration of use, suggesting that copper ions may protect against cervical malignancies.31
Women experiencing adverse effects or requesting IUD removal should be counseled about the rapid return of fertility and, if appropriate, be offered an alternative form of birth control immediately after removal of the IUD. Women attempting pregnancy after IUD removal conceive at a similar rate to those discontinuing other methods of contraception, with approximately 80 percent achieving pregnancy within one year.1,23 Two studies found that nulliparous women with tubal infertility were two to three times more likely than others to have used an IUD.1,32