Author : Philip Darney, MD, Department of Obstetrics, Gynecology and Reproductive Sciences, San Francisco General Hospital Division, UCSF

2008-06-13

Non Hormonal Contraception : (Sterilization, Barriers, and Natural Methods)

Many women who want to avoid pregnancy prefer not to use hormones. Male and female sterilization operations, vaginal barriers , and the “natural” methods all provide contraception without the use of hormones. In fact, these are the most frequently used methods around the world, including the USA, where sterilization and condoms provide protection against unintended pregnancy to more people than all the other methods combined.

Introduction

Many women who want to avoid pregnancy prefer not to use hormones. Male and female sterilization operations (vasectomy in men and various methods of closing the tubes or oviducts in women), vaginal barriers (condoms, diaphragms and crèmes or foams), and the “natural” methods (ovulation timing, abstinence, and breast feeding) all provide contraception without the use of hormones. In fact, these are the most frequently used methods around the world, including the USA, where sterilization and condoms provide protection against unintended pregnancy to more people than all the other methods combined.

Depending on the method chosen and how it is used, non-hormonal methods have a much wider range of effectiveness than the hormonal contraceptives (pills, patches, rings and implantable methods), but they avoid the side effects of taking hormones. The most effective of these methods (male and female sterilization) are not reversible; the reversible non-hormonal methods are usually much less effective than hormonal contraceptives, but some people achieve much greater success than “typical” users. Our description of these methods will begin with the most effective and proceed to the less effective ones.

Sterilization

Female Sterilization Operations (Tubal Ligation or Tubal Occlusion)

Key points:

Tubal sterilization remains a popular and highly effective method of contraception worldwide.
The 12 year U.S. Collaborative Review of Sterilization (CREST) study reported a 10 year cumulative pregnancy rate of 1.8%, with the lowest failure rate in post-partum partial salpingectomy and highest among bipolar electrocautery.
Other large series suggest 10 year failure rates for interval sterilization of 0.23% - 0.9%, with the Filshie clip having the lowest failure rate.
Successful bilateral placement of Essure coils occurs in 92-95% of procedures.
On the day of the procedure, current pregnancy should be excluded prior to tubal sterilization.
While no pregnancies were reported in Essure clinical trials, five year data suggest a failure rate of 2.6 per 1000, with no reported ectopic pregnancies; most of these pregnancies were in women who did not follow-up after the procedure. Confirmation of tubal occlusion is crucial, and this method should be reserved for patients who are likely to follow up and to use an interval method of contraception.

Tubal sterilization: an update on method failures

Though the last twenty years has seen a proliferation of hormonal birth control options for women, tubal sterilization remains one of the most widely used methods around the world.¹ As reported by the 2002 U.S. National Survey of Family Growth (NSFG)², tubal sterilization was the second most commonly employed method of contraception among women ages 15-44, at 23.2%; moreover, in women older than 35 it was the most commonly used birth control. Globally, surgical sterilization is used more than any other contraceptive method, with 180 million women relying on the procedure.³ Tubal sterilization refers to a number of technical approaches – including the newest transcervical, hysteroscopic micro-insert placement (Essure Permanent Birth Control, Conceptus Inc, San Carlos, CA) – each associated with its own benefits and drawbacks. Women who seek tubal sterilization do so because they no longer desire fertility. Thus, contraceptive efficacy is of prime importance to them. To understand the reported method failures, it is first important to understand the various methods of tubal sterilization.

What methods of tubal sterilization are available?

Post-partum sterilization, performed within 48 hours of delivery, is performed either at the time of cesarean or shortly after vaginal delivery through minilaparotomy; that is, a small incision (about one inch long). There are several techniques of cutting out part of the tube, with the most popular being the modified Pomeroy, named after Ralph Pomeroy, a prominent Brooklyn physician who developed it in 1929.

Interval sterilization (that is, not associated with delivery) is done through minilaparotomy, laparoscopy (a thin – less than half inch diameter - telescope placed into the belly through the skin) or transcervical methods in which a telescope is placed into the uterus through the vagina and mouth of the womb (cervix). An incision can also be made though the vagina to reach the tubes (oviducts) but this route is less and less frequently used because of the increased risk of infection. Laparoscopic options include silicone bands (Falope Ring), spring clip (Hulka clip), titanium clip (Filshie clip), bipolar electrocautery, and partial or complete salpingectomy (cutting out a section of the tube) The micro-insert device Essure is placed hysteroscopically into the lumen of the tubes where they join the uterus, with 3-8 coils trailing into the cavity of the uterus. The nickel-titanium outer coil stabilizes the inner layer of polyethylene terephthalate and, typically over the next three to six months, leads to blockage (occlusion) of the tubes as the extruded chemical elicits a scarring reaction to close the tubes near their opening into the uterus. The ingrowth of tissue takes approximately three months, and so a back-up method of contraception is required until tubal occlusion is demonstrated with an X-ray of the tubes and uterus. The technique is atypical, in that blockage is not immediate.

For women who have heavy bleeding for which they want surgical treatment, endometrial ablation may not provide permanent infertility, but simultaneous ablation and Essure placement appears to be safe and does not appear to impair, placement of the coils. In a small series of patients from 87% to 100% of coils could be properly placed at the time of ablation.^{28, 29} Of 21 patients who returned for hysterosalpingogram (the test that confirms occlusion) 90% had bilateral occlusion at three months and 100% by six months, with ablation not impairing the ability to perform or interpret the HSG.²⁹ Another transcervical method which has been used in a few countries and is currently in clinical trials in the US is chemical hardening of the tubes using quinacrine tablets that are put into the uterus like an IUD (see “Hormonal methods of family planning”). This approach is controversial because it requires two applications to reach acceptable efficacy and because quinacrine can be toxic at higher doses.

What are some of the mechanisms of pregnancy after tubal sterilization?

Several reasons have been proposed for why tubal sterilization fails in some instances. Most explanations for early tubal failures involve improper placement of the rings, clips or electrodes. In later failures, the tubes reopen Sometimes failures are due to unrecognized pregnancies at the time of the sterilization operation, which is why a preoperative pregnancy test is important.^{4, 30} Techniques such as silicone bands and Filshie and Hulka clips require the device to be completely placed around the tube, in the narrowest portion, and at right angles to the canal inside the tube..^{4, 8, 30} However, even when perfect placement has been confirmed, tubal patency (opening) has been reported.³¹ For bipolar coagulation, still the most common method of laparoscopic sterilization, the tubal tissue can dry out from the initial electric current; as this occurs in the outer layers, conduction of energy to the inner layers of the tube is impeded, despite a visual appearance of tubal destruction.⁶ .

How often do these different methods fail?

The most comprehensive data we have on tubal failures come from the CREST, a large, prospective, multicenter observational trial of 10,685 women undergoing both post-partum and interval tubal sterilization operations, but not the Filshie Clip or Essure, the two newest methods..⁴ Overall, with eight to 14 years of follow-up on each participant, 143 pregnancies occurred, indicating a 10 year cumulative incidence of tubal failure of 18 per 1000 procedures, or 1.8%, with 32% of these being ectopic pregnancies.⁵ Significant variation in failure rates was observed based on method of sterilization, age at time of procedure, and year of the operation. Procedures done with bipolar electric coagulation had the highest 10 year cumulative failure rates at 24.8 per 1000 procedures, with 54.3 per 1000 women ages 18-27.⁶ Sixty-five percent of these pregnancies after bipolar failure were ectopic. With only 7.5 pregnancies per 1000 women, post-partum partial salpingectomy and unipolar electric coagulation demonstrated the lowest failure rates, though the latter technique is rarely performed because of increased risk of injury to surrounding structures.⁴ Reflecting decreasing ability to become pregnant that occurs with increasing age, CREST investigators also demonstrated that women who were sterilized when they were younger than 28 had a higher rate of failure than older women for all methods except unipolar coagulation and interval partial salpingectomy. Another key finding was that the probability of pregnancy increased over time, rather than remaining constant. Finally, as demonstrated in a analysis of bipolar procedures by year, the risk of failure was higher in the earlier years of training in this technique (1978-1982) than in later years (1985-1987), 1.95% compared to 0.63%.⁶

While the results of CREST provide invaluable long-term information, it enrolled women between 1978 and 1986. The last 20 years have seen improvements in laparoscopic surgery in general; more devices for interval sterilization like the Filshie clip, which was not included in the CREST trial; and newer, less invasive transcervical techniques for sterilization like Essure.

Another source of long term data comes from a Quebec database of 321,929 sterilization procedures performed between 1980 and 1998;⁷ Filshie clips are the most commonly used technique in Quebec.⁸ The 15 year cumulative probability of pregnancy in this cohort was nine per 1000. While this rate is lower than that reported in CREST, the Quebec study was retrospective and unable to follow up all women who had tubal sterilizations. They also could not distinguish the different techniques used. Yet another large-scale study of 6639 Indian women who underwent either interval laparoscopic tubal ligation with Falope rings or interval minilap Pomeroy ligations demonstrated a cumulative probability of pregnancy of 3.7 per 1000.⁹ All 25 of their failures were within eight months of the procedure, and 16% of them were ectopic.

Ten year failure rates of the Filshie clip based on prospective studies done in Australia, the UK, Canada, and Malaysia compare favorably to all the methods observed in the CREST study,^10-12 ranging from 0.23% in Filshie’s original series of 434 women¹³ to 0.56% among 10,000 women in the UK.⁸ Post-partum application of the Filshie clip is associated with a two year failure rate of 1.01%^{10, 11} and 15 year failure rate of 0.9%.¹²

Does Essure hysteroscopic sterilization have a lower failure rate than other methods?

To date, approximately 170,000 women worldwide have had Essure placements.¹⁴ Essure’s pregnancy rates come both from multinational clinical trials performed since 1997 and international clinical experience. Essure is unique from other tubal sterilization methods not only because of its transcervical, hysteroscopic approach, but also because its contraceptive effect is delayed about three months. In addition, placement of the coils is not always feasible at the time of initial surgery. Accordingly, Essure “failures” should distinguish between inability to place the coils, pregnancy after tubal occlusion is documented, and pregnancy without confirmation of tubal occlusion..

Early phase II trials of Essure indicated first-attempt, bilateral placement rates of 85-88% ,^{15, 16} which improved to 90% in a later study.¹⁷ And in a post-approval study of 41 newly trained physicians, initial bilateral placement was achieved in 350 of 370 women (94.6%).¹⁸ Other studies not sponsored by the developer also report similar initial successful placements of about 92%^{19, 20} and, in the largest series to date, 812 of 857 women (95.7%).²¹ Factors contributing to successful placement include operator experience,²⁰ timing in the menstrual cycle,¹⁹ and unpredictable occurrence of tubal spasm. While some clinicians schedule procedures only within two weeks after menstruation—unless the woman is on hormonal contraception—there are no data that support or refute this routine practice. Additionally, ketorolac or other non-steroidal anti-inflammatory agents are often used pre-operatively in order to prevent tubal spasm; however, it is not known whether this intervention actually improves placement rates.

If bilateral placement is achieved, the next step in determining Essure’s contraceptive reliability is documenting tubal occlusion. It is recommended that this confirmation be done three months later, with the woman using a reliable interval form of contraception. Although plain X-ray and ultrasound have been shown to accurately localize Essure placement,^{22, 23} the manufacturer recommends a hysterosalpingogram in which a dye is pushed into the uterus and tubes.¹⁸ But even when correct placement is noted on x-ray, hysteroscopy and laparoscopy, tubes may not really be closed.²⁴ Bilateral occlusion rates in clinical trials and cohort studies range from 92%¹⁷ to 99%²⁰ at three months to 99.5-100% at six or 12 months^{17, 20}. When the three month tubal X ray does not demonstrate blockage of both tubes, the woman should continue to use another method of contraception, and have another X ray three months later, until occlusion is shown.

Logically, it seems that tubes shown to be occluded by X ray should protect a woman from pregnancy. In two studies with four year follow up with totals of 6015 and 9620 months in which women with Essure were at risk of pregnancies and in which none occurred..^{16, 17} However, there have been a number of cases reported in the literature of pregnancies occurring after Essure placement. These range from pregnancies when the woman did not follow up for X ray confirmation of tubal occlusion,^{25, 26} to two failures even when X ray showed blocked tubes.²⁶ The largest series to date describes 64 pregnancies which were reported to the manufacturer, out of 50,000 procedures, for a five year cumulative risk of 2.6 per 1000.²⁷ None of these pregnancies were ectopic.

Potential explanations for Essure failure in these cases include improper placement with too many or too few coils trailing in the uterus; perforation of the device through the tube or embedded into the myometrium; expulsion of the device; or misinterpretation of the HSG or tubal spasm during the test.^{25, 27} Most reported pregnancies have occurred when there was no follow-up..²⁷. In all techniques, re-canalization or openings of the tube into the abdomen can occur.³⁰

Cumulative probability of pregnancy*, by sterilization technique

	5 year	10 year
Essure	2.6	-
Unipolar	2.3	7.5
Postpartum partial salpingectomy	.63	.75%
Filshie clip	-	2.3-10.1
Silicone band	10	17.7
Quinacrine	-	10-130
Bipolar	16.5	24.8%
Hulka clip	31.7	36.5%

*per 1000 women

Techniques of Male Sterilization

Vasectomy is surgery to occlude the vas deferens (tube carrying sperm from the testicles to the penis) to prevent sperm passage into the ejaculate; it is almost always performed under local anesthesia. Male sterilization is safer and less invasive than female surgical sterilization and has fewer complications. The no-scalpel technique of reaching the vas is widely preferred to conventional scalpel incision. Occlusion of the vasa, the tube which caries sperm from the testicles into the penis, is achieved through ligation, excision, clips, clamps, sutures, or a combination – or through electrocautery. Pulling the sheath covering the vas deferens over one severed end and then sewing it shut to creat a natural tissue barrier (fascial interposition) appears to increase effectiveness.

Vasectomy is not immediately effective as contraception. Backup contraception is required until the ejaculate is sperm-free, a period that can last for 12 weeks or more. Evidence suggests that 60% to 80% of vasectomized men are azoospermic (no sperm in the semen) after 12 weeks and after 20 ejaculations, although study results vary. However, the time to azoospermia can be highly variable and may depend upon the type of occlusion used, as well as on the age of individual men. Repeat sperm testing may be required until azoospermia is achieved. Recanalization is possible, both early and some time after vasectomy. An additional advantage of vasectomy is that the operation is more easily and cheaply reversed than is female tubal ligation, especially if only a few years have elapsed since the vasectomy. There is no evidence that vasectomy has long-term health consequences or has any effect on sexual function. The sperm are a minor component of the ejaculate so there is very little change in ejaculate volume after a vasectomy.

Barrier Methods

Barrier methods of contraception have been the most widely used contraceptive techniques throughout recorded history. These methods are receiving more attention as we respond to the personal and social impact of sexually transmitted diseases (STDs). A new need for sexual safety has brought modern respect and new developments to the condom, while the other barrier methods continue to serve well for appropriate couples.

EFFICACY OF BARRIER METHODS

FAILURE RATES DURING THE FIRST YEAR OF USE, UNITED STATES3

Percent of women with pregnancy

METHOD LOWEST EXPECTED TYPICAL

No method 85% 85%

Spermicides 3.0% 21.0%

Cervical cap 6.0% 18.0%

Sponge:

Parous women 9.0% 28.0%

Nulliparous women 6.0% 18.0%

Diaphragm and spermicides 6.0% 18.0%

Condom 2.0% 12.0%

Risks and Benefits Common to all Barrier Methods

Barrier (condoms and diaphragms) and spermicide methods provide protection (about a 50% reduction) against STDs and pelvic inflammatory disease (PID).4–8 This includes infections with chlamydia, gonorrhea, herpes simplex, cytomegalovirus, human papillomavirus, and human immunodeficiency virus (HIV); however, only the condom has been proven to prevent HIV infection. STD protection has a beneficial impact on the risk of tubal infertility and ectopic pregnancy.6,9 There have been no significant clinical studies on STDs and cervical caps or the female condom, but these methods should be effective. Women who have never used barrier methods of contraception are almost twice as likely to develop cancer of the cervix.9,10 The risk of toxic shock syndrome is increased with female barrier methods, but the actual incidence is so low that this is not a significant clinical consideration.11 Women who have had toxic shock syndrome, however, should be advised to avoid barrier methods.

The Diaphragm

The first effective contraceptive method under a woman's control was the vaginal diaphragm. Distribution of diaphragms led to Margaret Sanger's (the founder of the Planned Parenthood movement) arrest in New York City in 1918. This was still a contentious issue in 1965 when the Supreme Court's decision in Griswold v. Connecticut ended the ban on contraception in that state. By 1940, one-third of American couples practicing contraception were using the diaphragm. This decreased to 10% by 1965 after the introduction of oral contraceptives and intrauterine devices, and fell to about 3% by 1988 and has continued to decline with the availability of a variety of more effective methods (see Hormonal Contraception).

Efficacy

Failure rates for diaphragm users vary from as low as 2% per year of use to a high of 23%. The typical use failure rate after one year of use is 18%.3 Older, married women who use it longer achieve the highest efficacy, but young women can use diaphragms very successfully if they are properly encouraged and counseled. There have been no adequate studies to determine whether efficacy is different with and without spermicides.14

Side Effects

The diaphragm is a safe method of contraception that rarely causes even minor side effects. Occasionally women report vaginal irritation due to the latex rubber, or the spermicidal jelly or cream used with the diaphragm. Less than 1% discontinue diaphragm use for these reasons. Urinary tract infections are approximately twice as common among diaphragm users as among women using oral contraception.15 Possibly the rim of the diaphragm presses against the urethra and causes irritation, which is perceived as infectious in origin, or true infection may result from touching the perineal area or incomplete emptying of the bladder. It is more probable that spermicides used with the diaphragm can increase the risk of bacteriuria with E coli, perhaps due to an alteration in the normal vaginal flora (the mixture of organisms naturally found at the vagina).16 Clinical experience suggests that emptying the bladder after sexual intercourse is helpful, and if necessary, a single postcoital dose of a prophylactic antibiotic will nearly eliminate the risk.

Improper fitting or prolonged retention (beyond 24 hours) can cause vaginal abrasion or mucosal irritation. There is no link between the normal use of diaphragms and the toxic shock syndrome.17 It makes sense, however, to minimize the risk of toxic shock by removing the diaphragm after 24 hours and during menses.

Benefits

Diaphragm use reduces the incidence of cervical gonorrhea,18 pelvic inflammatory disease,19 and tubal infertility.5, 9 This protection may be due in part to the simultaneous use of a spermicide. There are few data, as of yet, regarding the effect of diaphragm use on the transmission of the AIDS virus (HIV); the only study did not show an effect against HIV transmission. An important advantage of the diaphragm is low cost. Diaphragms are durable, and with proper care, can last for several years.

Choice and Use of the Diaphragm

There are three types of diaphragms, and most manufacturers produce them in sizes ranging from 50 to 105 mm diameter, in increments of 2.5 to 5 mm. Most women use sizes between 65 and 80 mm.

The diaphragm made with a flat metal spring or a coil spring remains in a straight line when pinched at the edges. This type is suitable for women with good vaginal muscle tone and an adequate recess behind the pubic arch. However, many women find it difficult to place the posterior edge of these flat diaphragms into the posterior cul-de-sac, and over the cervix.

Arcing diaphragms are easier to use for most women. They come in two types. The All-Flex type bends into an arc no matter where around the rim the edges are pinched together. The hinged type must be pinched between the hinges in order to form a symmetrical arc. The hinged type forms a narrower shape when pinched together, and thus may be easier for some women to insert. The arcing diaphragms allow the posterior edge of the diaphragm to slip more easily past the cervix and into the posterior cul-de-sac. Arcing diaphragms are used more successfully by women with poor vaginal muscle tone, cystocele (fallen bladder), rectocele (when the rectum bulges into the vagina), a long cervix, or an anterior cervix of a retroverted uterus.

Successful use of a diaphragm depends upon proper fitting. Fitting rings or diaphragms should be scrupulously disinfected by soaking in a bleach solution and be available in all diameters. At the time of the pelvic examination, the middle finger is placed against the vaginal wall and the posterior cul-de-sac, while the hand is lifted anteriorly until the pubic symphysis abuts the index finger. This point is marked with the examiner's thumb to approximate the diameter of the diaphragm. The corresponding fitting ring or diaphragm is inserted, the fit to be assessed by both clinician and patient.

If the diaphragm is too tightly pressed against the pubic symphysis so that it is uncomfortable, a smaller size is selected. If the diaphragm is too loose (comes out with a cough or bearing down), the next larger size is selected After a good fit is obtained, the diaphragm is removed by hooking the index finger under the rim behind the symphysis and pulling .The woman should then herself insert the diaphragm, practice checking for proper placement, and attempt removal. Weight loss, weight gain, vaginal delivery, and even sexual intercourse can change vaginal caliber, so the fit of a diaphragm should be assessed every year at the time of the regular examination.

The diaphragm should be inserted no longer than six hours prior to sexual intercourse. About a tablespoonful of spermicidal cream or jelly should be placed in the dome of the diaphragm prior to insertion, some of the spermicide should be spread around the rim with a finger. The diaphragm should be left in place for approximately six hours (but no more than 24 hours) after coitus. Additional spermicide should be placed in the vagina before each additional episode of sexual intercourse while the diaphragm is in place.

After removal, the diaphragm should be washed with soap and water, rinsed, and dried. Powders of any sort need not and should not be applied to diaphragm. It is wise to use water to periodically check for leaks. Diaphragms should be stored in a cool and dark location.

The Cervical Cap

The cervical cap was popular in Europe long before its re-introduction into the United States. There are several types of cervical caps, but only the cavity rim (Prentif) cap is approved in the U.S. U.S. trials have demonstrated the cervical cap to be about as effective as the diaphragm, but somewhat harder to fit (it comes in only four sizes) and more difficult to insert (it must be placed precisely over the cervix).20,21

Efficacy and Benefits

The cervical cap has several advantages over the diaphragm. It can be left in place for a longer time (up to 48 hours), and it need not be used with a spermicide. However, a tablespoonful of spermicide placed in the cap before application is reported to increase efficacy (to a 6% failure rate in the first year among those who follow the instructions precisely) and to prolong wearing time by decreasing the incidence of foul-smelling discharge (a common complaint after 24 hours).21

Choice and Use

The size of the cervix varies considerably from woman to woman, and the cervix changes in individual women in response to pregnancy or surgery. Proper fitting can be accomplished in about 80% of women. Women with a cervix that is too long or too short, or with a cervix that is far forward in the vagina, may not be suited for cap use. However, women with vaginal wall weakness (bladder or rectal bulging into the vagina) may be better suited to a cervical cap than to a diaphragm.

To remove the cap (at least eight hours after coitus), pressure must be exerted with a finger tip to break the seal. The finger is hooked over the cap rim to pull it out of the vagina. Bearing down or squatting or both can help to bring the cervix within reach of the finger.

The cervical cap can be left in place for two days, but some women experience a foul-smelling discharge by that time. Like the diaphragm, it must be left in place for at least eight hours after sexual intercourse in order to ensure that no motile sperm are left in the vagina.

The most common cause of failure is dislodgment of the cap from the cervix during sexual intercourse. There is no evidence that cervical caps cause toxic shock syndrome or precancerous changes in the cervical mucosa.22 It seems likely (although not yet documented) that cervical caps would provide the same protection from sexually transmitted diseases as the diaphragm.

The Fem Cap, made of nonallergic silicone rubber, is shaped like a sailor’s hat, a design that allows a better fit over the cervix and in the vaginal fornices.23 This cap may be easier to fit and use, provides better efficacy, and is easier to remove.

Lea’s Shield is a vaginal barrier contraceptive composed of silicone.56,57 This soft, pliable device comes in one size and fits over the cervix, held in place by the pressure of the vaginal wall around it. There is a collapsible valve that communicates with a nine millimeter opening in the bowl that fits over the cervix. This valve allows equalization of air pressure during insertion and drainage of cervical secretions and discharge, permitting a snug fit over the cervix. A thick U-shaped loop attached to the anterior side of the bowl is used to stabilize the device during insertion and for removal. The thicker part of the device is shaped to fill the posterior fornix, thus contributing to its placement and stability over the cervix. The addition of a spermicide, placed in the bowl, is recommended to enhance STD protection. Lea’s Shield is designed to remain in place for 48 hours after intercourse. Clinical studies regarding efficacy and side effects are not yet available.

The Contraceptive Sponge

The vaginal contraceptive sponge is a sustained release system for the spermicide, nonoxynol-9. The sponge also absorbs semen and blocks the entrance to the cervical canal. The "Today" sponge is a dimpled polyurethane disc impregnated with one gram of nonoxynol-9. About 20% of the nonoxynol-9 is released over the 24 hours the sponge is left in the vagina. Production of the “Today” sponge in the U.S. ceased in 1995, and availability awaits a new manufacturer.