Permanent Contraception

Aileen M. Gariepy, MD, MPH

Rebecca H. Allen, MD, MPH

The phrase “permanent contraception” recognizes a person’s voluntary decision to complete childbearing as an active and positive decision. Although “sterilization” has been used to refer to nonreversible methods, the term can connote coercive or involuntary procedures in some settings; “permanent contraception” has been proposed as an alternative.¹ Permanent contraception is now the predominant method of contraception in the world.² However, the rate varies widely according to region and country (Table 4.1).

History

Although James Blundell proposed performing tubectomy at the time of cesarean section to avoid the need for repeat sections in women at risk of obstructed labor due to a contracted pelvis,³ Samuel Lungren of Toledo, Ohio, published the first report of postpartum permanent contraception in 1881.⁴ The method of Ralph Pomeroy, a prominent physician in Brooklyn, New York, was first described to the medical profession by his associates in 1929, 4 years after Pomeroy’s death.⁵ Frederick Irving of the Harvard Medical School described his technique in 1924, and the Uchida method was reported in 1946.

Few female permanent contraception procedures were performed until the 1930s when “family planning” was first suggested as an indication for surgical sterilization by Baird in Aberdeen. He required women to be older than 40 and to have had eight or more children. Mathematical formulas of this kind persisted through the 1960s. At the time, the American College of Obstetricians and Gynecologists (ACOG) recommended that voluntary permanent contraception be restricted to women or men whose age multiplied by number of children was equal to or greater than 120. In 1965, Sir Dugald Baird delivered a remarkable lecture, entitled “The Fifth Freedom,” calling attention to the need to alleviate the fear of unwanted pregnancies and the important role of permanent contraception.⁶ By the end of the 1960s, voluntary permanent contraception was a popular procedure but still legally restricted in some states until 1972. On the other hand, forced and coercive sterilizations were practiced in the United States during this time for both eugenic and antipoverty reasons.⁷

Table 4.1 Permanent Contraception Use in Select Regions/Countries

Region/Country	Total	Female	Male
India	40.2%	39.0%	1.2%
China	32.6%	28.2%	4.4%
United Kingdom	29.0%	8.0%	21.0%
United States	24.5%	18.6%	5.9%
Sub-Saharan Africa	1.6%	1.6%	0
Sources: United States Data: Daniels K, Abma JC, Current contraceptive status among women aged 15-49: United States, 2015-2017, NCHS Data Brief (327):1-8, 2018. Other World Data: United Nations, Trends in contraceptive use worldwide [cited June 19, 2019]. http://www.un.org/en/development/desa/population/publications/pdf/family/trendsContraceptiveUse2015Report.pdf, 2015.

Laparoscopic methods to occlude the fallopian tubes were introduced in the early 1970s. By 1982, more sterilization operations were performed for women than for men due to dramatic decreases in costs, hospital time, and pain with laparoscopy compared to laparotomy.⁸ The use of laparoscopy for permanent contraception increased from only 0.6% of surgeries in 1970 to more than 35% by 1975.⁹ These methods allowed women to undergo operations at times other than immediately after childbirth or during major surgery.

Hysteroscopic methods were investigated as early as 1934 when Schroeder recorded the first two attempts at hysteroscopic tubal occlusion with electrocautery initially applying electrocautery to the intramural portion of the tube.¹⁰ As hysteroscopy advanced, this technique continued to be used but ultimately fell out favor in the 1970s due to high patency rates (20%) and risk of uterine perforation and thermal injury.¹¹ The neodysmium:yttriumaluminum-garnet (Nd:YAG) laser was also trialed, but patency rates were worse than those for electrocautery (74%).¹² Research then focused on implantable devices and various plugs and screws made of nylon, silicone, ceramic, and polyethylene.¹³^,¹⁴ A formed-in-place silicone plug marketed under the trade name Ovabloc™ was introduced in Europe in the 1990s but was never approved in the United States. Ultimately, the Essure® coil system was approved by the European Union in 2001 and by the United States Food and Drug Administration (FDA) in 2002. Adiana®, a combination of silicone insert with thermal damage, was approved by the FDA in 2009.

Male sterilization through vasectomy began as an alternative to castration for the punishment of certain crimes and then entered the eugenics movement as means to sterilize criminals and others deemed “ unimprovable.”¹⁵ As a voluntary family planning procedure, however, vasectomy first gained popularity internationally. India introduced vasectomy as a means of population control in 1952, and massive vasectomy camps were established that often used coercive techniques.¹⁵ In the United States, vasectomy for family planning became more common in the 1960s after legal restrictions were removed on voluntary permanent contraception. Early vasectomy was performed through the inguinal canal and later moved to the scrotum. The no-scalpel vasectomy technique was developed in 1973 in China.¹⁶

For a complete review of the history of permanent contraception, see the online chapter available in the eBook.

Female Permanent Contraception Techniques

Interval Procedures

Procedures performed at a time unrelated to the postpartum period are classified as “interval” in timing. In the United States and most of the developed world, laparoscopy is more popular than minilaparotomy for tubal occlusion.¹⁷ A minilaparotomy is an incision no more than a few centimeters in length. For both procedures, hospitalization is not required, and most patients are discharged within a few hours. However, in obese women, performing a true minilaparotomy can be difficult, an important consideration as obesity rates increase worldwide. Laparoscopy offers many advantages including minimal discomfort, small incision scars, and the ability to perform the procedure as an outpatient. In addition, the surgeon has an opportunity to inspect the pelvic and abdominal organs for abnormalities. The disadvantages of laparoscopic permanent contraception include the cost related to expensive equipment, the special training required, and the risks of inadvertent bowel or vessel injury. In addition, some women with a history of extensive pelvic adhesions, inability to tolerate Trendelenburg positioning, or a contraindication to general anesthesia may not be candidates for a laparoscopic procedure.

Minilaparotomy for interval procedures is an important approach if laparoscopy is not available or impossible due to adhesions. In this procedure, the surgeon makes a small suprapubic horizontal incision to access the peritoneal cavity. A uterine manipulator is used to position the fallopian tubes closer to the incision. The surgeon can place a tubal occlusion device or perform a partial or total salpingectomy. In low-resource settings, this procedure is often performed under local anesthesia alone or local anesthesia with moderate intravenous sedation.¹⁸

The risk of complications with interval permanent contraception procedures is low (0.9 to 1.6 per 100 procedures); risk factors for complications include obesity, prior abdominal or pelvic surgery, and diabetes.¹⁹ The most extensive single study of female permanent contraceptive procedures comes from the landmark U.S. Collaborative Review of Sterilization (CREST), a large, multicenter, prospective study that enrolled almost 10,000 women from 1978 to 1987 and followed them at periodic intervals until 1994.²⁰ The CREST study compared bipolar coagulation, unipolar coagulation, silicone rubber band (Falope ring), spring clip (Hulka), interval partial salpingectomy via laparotomy, and postpartum partial salpingectomy. There were no deaths among 9,475 women who underwent laparoscopic permanent contraception in this study.¹⁹

Currently, laparoscopic sterilization can be achieved with any of these methods:

1. Occlusion by bipolar electrocoagulation
2. Occlusion by mechanical means (titanium clips or silicone rubber rings)
3. Removal of a portion of the fallopian tube (partial salpingectomy)
4. Removal of the entire fallopian tube (salpingectomy)

The most common laparoscopic approach for tubal occlusion or excision utilizes an umbilical port for the laparoscope and a single midline suprapubic port site for the introduction of the electrosurgical system or clip/band applying instrument. A uterine manipulator is placed transcervically to allow for better visualization of the fallopian tubes. Bilateral salpingectomy typically requires at least one additional trocar site.²¹ Alternatively, “single site” laparoscopic surgery techniques allow procedures through a single umbilical port site using either a laparoscope with an operating channel (operative laparoscope) or laparoendoscopic single-site surgery (LESS) with a multichannel port.²² All can be used with the “open” laparoscopic technique in which the laparoscopic instrument is placed into the abdominal cavity under direct visualization to decrease the risk of bowel or blood vessel puncture on blind entry.²³

Tubal Occlusion by Electrosurgical Methods

The most common electrosurgical technique utilized in the United States is bipolar coagulation. Unipolar cautery fell out of favor due to higher rates of inadvertent thermal bowel injury.²⁴^,²⁵ The bipolar method of sterilization uses a specially designed forceps, typically the reusable Kleppinger bipolar forceps, to desiccate the fallopian tube at the junction of the isthmus and ampulla and destroy the tubal lumen (Figure 4.1). A number of single-use bipolar forceps are also available. With all of these instruments, one jaw of the forceps is the active electrode, and the other jaw is the return electrode. Current density is greatest at the point of forceps contact with tissue, and the use of a low-voltage, high-frequency current prevents the spread of electrons to nontarget tissues. As desiccation occurs at the point of high current density, tissue resistance increases, and the coagulated area eventually provides resistance to flow of the low-voltage current. Should the resistance increase beyond the voltage’s capability to push electrons through the tissue, incomplete coagulation of the endosalpinx can result.²⁶ Although data from CREST were first reported to show that bipolar tubal electrocoagulation had a higher failure rate than other methods, a reanalysis of the data focusing on procedures that ensured full fallopian tube desiccation by using electrocoagulation at three or more sites (3 cm total) indicated a comparable 5-year pregnancy rate of 3.2 pregnancies per 1,000 women.²⁷ The use of an ammeter will assist the surgeon in confirming complete desiccation.

Figure 4.1 Anatomy of the Fallopian Tube. An ovum released from the ovary is swept up by the fimbria and carried through the infundibulum, ampulla, isthmus, and intramural segments of the fallopian tube aided by beating cilia and contracting smooth muscle. (Reprinted from Patil E, Jensen JT, Permanent contraception for women. Semin Reprod Med 34(3):139-144, 2016, with permission from Thieme. Illustration by Robin Jensen.)

Tubal Occlusion with Clips and Rings

Female sterilization by mechanical occlusion eliminates the safety concerns with electrosurgery. However, mechanical devices are subject to flaws in material, defects in manufacturing, and errors in design, all of which can alter efficacy. Two mechanical devices are in current use, the titanium Filshie clip with silicone rubber lining (Figure 4.2) and the Silastic (silicone elastomer) band, also known as the Falope or Yoon ring (Figure 4.3). The Hulka (spring) clip is no longer in use. Each of these devices requires an understanding of its mechanical function, a working knowledge of the intricate applicator necessary to apply the device, meticulous attention to maintenance of the applicators, and skillful tubal placement. These devices are less effective when used immediately postpartum on dilated tubes.²⁸^,²⁹ Mechanical occlusion offers a better chance for tubal reversal surgery compared to electrosurgical methods that destroy more tube; however, caution should be used with this justification for method choice since permanent, nonreversible contraception is always the goal.³⁰

Filshie Clip

The Filshie clip is made of titanium lined with silicone rubber. The hinged clip is locked over the tube using a special applicator through a second incision or operating laparoscope. The clip must be applied perpendicular to the isthmic portion of the tube, 1 to 2 cm from the cornua of the uterus. Care must be taken to ensure the clip completely occludes the entire tube by visualizing the tip of the applicator through the mesosalpinx. The rubber lining of the clip expands on compression to keep the tube blocked. Only 4 mm of the tube is destroyed. Clips are left permanently in place. Failure rates 1 year after the procedure approximate 1 per 1,000 women. A 15-year follow-up study in Quebec reported a cumulative failure rate of 9 per 1,000 women, whereas the 10-year failure rate in the United Kingdom was 5.6 per 1,000 women.³¹^,³² While the Filshie clip is also marketed for postpartum application at the time of cesarean delivery or via minilaparotomy after vaginal delivery, concerns have been raised over efficacy in this circumstance. The postpartum fallopian tubes are often hypertrophied and edematous, making correct clip application more difficult. One multicenter randomized controlled trial of 1,400 women compared the Filshie clip to partial salpingectomy.²⁹ At 2 years, the cumulative probability of pregnancy was 0.017 with the clip (9 pregnancies) and 0.004 with partial salpingectomy (2 pregnancies) (p = 0.04).

Figure 4.2 Filshie Clip.

Figure 4.3 Falope Ring.

Silastic (Falope or Yoon) Ring

This nonreactive Silastic band has an elastic memory of 100% if stretched to no more than 6 mm for a brief time (a few minutes at most). A special applicator, 6 mm in diameter, can be placed through a second cannula or through a standard offset operating laparoscope. The applicator is designed to grasp a knuckle of tube and release the Silastic band onto a 2.5-cm loop of tube. Ten to fifteen percent of patients experience severe postoperative pelvic cramping from the tight bands (which can be reduced by the application of a local anesthetic to the tube before or after banding).³³

The ring should be placed at the junction of the isthmic and ampullary portion each fallopian tube. Necrosis occurs promptly and a 2- to 3-cm segment of the tube is destroyed. The ring stays permanently in place. Failure rates are about 1% after 2 years, and the 10-year cumulative rate is 1.77%.²⁰^,³⁴ Mesosalpinx bleeding is the most common complication of Silastic ring application. It usually occurs when the forceps grab not only the tube but also a vascular fold of mesosalpinx. The mesosalpinx can also be torn on the edge of the stainless steel cylinder as the tube is drawn into the applicator. If bleeding is noted, application of the band often controls it. If the placement of additional bands or electrocoagulation fails to stop bleeding, laparotomy may be required. Silastic rings are occasionally placed on structures other than the tube. While removal of an incorrectly positioned band from the round ligament or mesosalpingeal folds typically is easily accomplished by grasping the band with the tongs of the applicator and applying gradual, increasing traction, removal is not generally necessary or recommended. Providers should be prepared to use an alternative method like electrosurgery or salpingectomy in case bands or clips cannot be applied due to adhesions or bleeding.

Total Salpingectomy

Traditionally, tubal occlusion techniques were used for laparoscopic permanent contraception as the methods were easier and thought to be associated with fewer complications than partial or total salpingectomy. Laparoscopic total salpingectomy, however, is increasingly being offered as an alternative to tubal occlusion techniques because of the potential higher efficacy and decrease in future ovarian cancer risk, so-called opportunistic salpingectomy.³⁵^,³⁶^,³⁷^,³⁸ In one large California health system, a one-time education campaign focusing on total salpingectomy as a means to decrease ovarian cancer risk resulted in an increase in total salpingectomy rates for interval permanent contraception from 1% in 2011 to 78% in 2016.³⁹ This retrospective cohort study of 9,007 tubal occlusion procedures and 1,734 total salpingectomies found that operative time increased by 3 minutes for total salpingectomy compared to tubal occlusion with no difference in estimated blood loss, readmission within 30 days, or emergency department visit within 7 days. Another retrospective cohort study compared total salpingectomy using an electrothermal bipolar tissue-sealing instrument (Ligasure) and tubal occlusive techniques.²¹ The procedures included 68 tubal occlusions (19% Falope rings, 32% bipolar cautery, and 47% Filshie clips) and 81 bilateral salpingectomies. Surgical time was only 6 minutes longer for the salpingectomies, and similar rates of complications were noted.

Transcervical Approaches

Although current methods of permanent contraception are safe and effective, they require skillful surgeons and, in the case of laparoscopic and hysteroscopic operations, expensive equipment. Simpler approaches could make permanent contraception available and acceptable to more women and avoid surgery. During the 1970s and 1980s, a number of investigators developed approaches for transcervical delivery of agents to occlude the fallopian tubes.⁴⁰ While some involved the use of hysteroscopes to place occlusive devices, many techniques used sclerosing agents. Zipper developed the technique of quinacrine sterilization in which quinacrine pellets are placed to the fundus using a modified IUD inserter.⁴¹ The pellets caused epithelial damage resulting in occlusive collagen replacement to the intramural portion of the fallopian tube.⁴² Unfortunately, questions regarding potential toxicity of the approach emerged after the initiation of widespread human use in many developing nations.⁴³^,⁴⁴ Since the human use occurred in the absence of regulatory approval in the United States or Europe, these allegations led to a loss of confidence in the approach and withdrawal of support by governments and funding agencies. Although subsequently completed epidemiologic studies support the safety of quinacrine sterilization,⁴⁵^,⁴⁶ serious concerns regarding implementation of the program and a significantly higher (two to four times greater than surgical tubal ligation at 10 years) failure rate present significant obstacles to further research and development of the method.⁴⁷^,⁴⁸ The FDA placed a planned U.S. phase three clinical trial program on hold in the United States in 2006, and the sponsor withdrew support effectively ending the program.

New approaches to transcervical nonsurgical permanent contraception are under investigation. A system named FemBloc™ is being investigated in late phase clinical trials (Clinicaltrials.gov NCT03433911), but no published data exist. This system uses a proprietary catheter system to deliver a biopolymer to the intramural tube under ultrasound guidance. The transcervical administration of polidocanol foam, a sclerosing agent currently approved for the indicated for the treatment of incompetent great saphenous veins, accessory saphenous veins, and visible varicosities of the great saphenous vein system, shows potential in nonhuman primate studies.⁴⁹

Hysteroscopic Permanent Contraception

Two hysteroscopic methods have been approved and marketed, only to be subsequently withdrawn from the market. Essure was first approved in 2002 and withdrawn in December 2018.⁵⁰ Adiana was approved in 2009⁵¹ and withdrawn in 2012.⁵² Both methods induced collagen deposition in the intramural portion of the tube over a few months. Whereas Essure delivered polyethylene terephthalate fibers to the tubal ostia, held in place by nickeltitanium (nitinol) coils, Adiana involved introduction of a silicone matrix after delivery of controlled thermal damage to the tubal lumen.

Hysteroscopic permanent contraception has several potential advantages compared to laparoscopic techniques. These advantages include no incisions or abdominal entry and the option to perform in the office setting under local anesthesia only.⁵³ These features make the procedure potentially beneficial for individuals at high risk for complications with abdominal surgery or general anesthesia. Two disadvantages of the previously marketed methods were the lack of immediate effectiveness and the requirement for a delayed confirmation test. These multiple steps and delay expose patients to potential pregnancy.

Both Essure and Adiana required verification of occlusion using hysterosalpingogram (HSG) or transvaginal ultrasonography. Follow-up to confirm bilateral tubal occlusion was poor, with an estimated 13% to 94% of women returning for HSG depending on the study population.⁵⁴^,⁵⁵^,⁵⁶^,⁵⁷ In 2015, the FDA approved transvaginal ultrasonography as an alternative to HSG in selected patients (uncomplicated placement procedure lasting ≤15 minutes and 1 to 8 trailing coils visible). One study observed an increase in follow-up rates from 77.5% for HSG to 88% (p = 0.008) for transvaginal ultrasonography.⁵⁸

This noncompliance with HSG follow-up accounted for approximately half of all hysteroscopic tubal occlusion failures and resulting pregnancies.⁵³^,⁵⁴^,⁵⁹^,⁶⁰^,⁶¹ In one large national comparative study, at 2 years, the pregnancy rate was 2.4% among women undergoing hysteroscopic tubal occlusion versus 2.0% among those undergoing laparoscopic procedures (HR 20, 95% CI 1.09-1.33).⁵⁴ However, among women who underwent HSG (66%), the Essure pregnancy rate fell to 1.8% and was no different than laparoscopy (HR 0.90, 95% CI 0.80-1.02). Other reasons for hysteroscopic sterilization failures include incomplete occlusion (10%), incorrect HSG interpretation (33%), and an established pregnancy before the procedure (approximately 1%).⁶⁰^,⁶¹^,⁶²^,⁶³ Overall, of all women who choose hysteroscopic sterilization, at 3 months postprocedure, approximately 85% of women could rely on it for pregnancy prevention.⁵⁵^,⁶⁴

Unsuccessful placement of both Essure coils occurred in 4% to 24% of procedures.⁵⁵^,⁶⁵^,⁶⁶^,⁶⁷ In one large study of 27,724 hysteroscopic tubal occlusion procedures, 4% underwent a second attempt hysteroscopically and 4% underwent a subsequent laparoscopic sterilization surgery.⁵⁴ In general, hysteroscopic tubal occlusion has a 4% to 8% rate of repeat sterilization surgery compared with less than 2% for laparoscopic procedures.⁵⁴^,⁶⁷^,⁶⁸

Essure

Essure consists of a nickel-titanium outer coil and a stainless steel inner coil wrapped with polyethylene terephthalate fibers (Figure 4.4). The insert is approximately 4 cm in length and 0.8 mm in diameter, which expands to 2 mm when deployed. The hysteroscopic surgeon utilizes a specially designed, single-use, insertion catheter that uses an outer sheath to maintain the system in a wound-down state and introduce it through the tubal ostia, positioning this in the interstitial region of the tube. Withdrawal of the outer sheath allows the outer coils to expand, anchoring the system in place, spanning the uterotubal junction. The number of outer coils left trailing into the uterine cavity should ideally be 3 to 8, however, between 0 and 17 is acceptable. Adequate uterine distention and a thin endometrium (follicular phase or preparation with hormonal agents) are essential to be able to visualize the tubal ostia. Over several months, the polyethylene terephthalate fibers stimulate an inflammatory response that results in replacement of the tubal lumen with a fibrotic scar. Unlike laparoscopic tubal occlusion and salpingectomy procedures, this method does not provide immediate contraception. Women must use a back-up method of contraception until a confirmation of tubal occlusion by a HSG or transvaginal ultrasound after 3 months. If occlusion is not present at 3 months, contraception is continued and HSG is repeated 3 months later (6 months postprocedure). Contraindications include less than 6 weeks from abortion or delivery, active or recent pelvic infection, uterine or tubal pathology blocking access to tubal ostia, suspected unicornuate uterus, and known allergy to contrast media preventing ability to undergo HSG.

Figure 4.4 Essure Procedure. UTJ, uterotubal junction; SUTJ, serosa of uterotubal junction. (Reprinted from Essure Permanent Birth Control product information, 2018 U.S. label, with permission from Bayer Healthcare, Whippany, NJ.)

Serious complications with hysteroscopic tubal occlusion are rare, but there is a risk of tubal perforation and potential intra-abdominal placement of coils and bowel injury and malposition or expulsion of the coils.⁶⁹^,⁷⁰^,⁷¹ Limited data exist on best practices for removal of Essure coils after insertion if needed. Case series have reported that removal may be accomplished with hysteroscopy, laparoscopy, and hysterectomy.⁷²^,⁷³^,⁷⁴^,⁷⁵ The most common reason cited for removal is pelvic pain.

The Essure device received expedited review from the FDA since it offered a new method of female permanent contraception that had advantages over laparoscopy.⁷⁶ While categorized as a Class III (high risk) medical device, all components of the Essure coil use materials previously approved for use in well-tolerated medical devices such as artificial joints and cardiac stents. Similarly, although a randomized clinical trial was not required, Essure developers did complete open-label safety and efficacy studies for the device. Once available, the uptake of Essure after approval was rapid, and the manufacturer estimates that more than 750,000 units were sold worldwide.⁶⁷^,⁷⁷^,⁷⁸^,⁷⁹ For example, in New York State, the use of Essure increased from 0.6% of all permanent contraception procedures in 2005 to 25.9% in 2013.⁸⁰

Adiana

The Adiana device utilized controlled thermal damage to the lumen of the fallopian tube followed by insertion of a silicone matrix. Using hysteroscopy, a delivery catheter is inserted into the tubal ostia into the interstitial portion of the tube. The distal tip of the catheter then delivers radiofrequency energy to the lumen. The silicone matrix is then deployed in the region of thermal damage. Over the next few weeks, fibroblasts grow into the matrix, which serves as a scaffolding, and tubal occlusion occurs. HSG is performed 3 months after the procedure to confirm tubal occlusion. The company that manufactured Essure at the time, Conceptus, sued Adiana’s manufacturer, Hologic, for patent infringement winning an award of $18 million. Ultimately, Hologic agreed to stop selling Adiana if Conceptus would rescind the settlement. Adiana was then removed from the market in 2012.⁸¹

Studies Comparing Laparoscopic and Hysteroscopic Permanent Contraception

Several retrospective cohort studies conducted in the United States and Europe have been published comparing outcomes after hysteroscopic and laparoscopic permanent contraception. The first U.S. study examined 8,048 patients undergoing hysteroscopic procedures and 44,278 undergoing laparoscopic procedures between 2005 and 2013 in New York state using a statewide claims database.⁸⁰ Women undergoing hysteroscopic procedures were more likely to be 40 years or older (25.2% vs. 20.5%, p < 0.01), have a history of pelvic inflammatory disease (10.3% vs. 7.2%, p < 0.01), major abdominal surgery (9.4% vs. 7.9%, p < 0.01), and cesarean section (23.3% vs. 15.4%, p < 0.01) compared to women undergoing laparoscopic procedures. Of women undergoing hysteroscopic permanent contraception, 50% received general anesthesia. At 1, 2, and 3 years after the procedure, the rate of pregnancy was similar, but the rate of reoperation was higher in the hysteroscopic group (1 year: OR 10.16, 95% CI 7.47-13.81; 2 years: OR 7.96, 95% CI 6.00-10.57; 3 years: OR 5.88, 95% CI 4.44-7.79). Although database studies have limited ability to infer causality due to limited control for baseline confounding, the authors of this study adjusted these odds ratio estimates for age, race, insurance status, year of procedure, major comorbidities, and history of pelvic inflammatory disease, major abdominal surgeries, and cesarean section, and the OR estimates are robust supportive of a relationship. Although pregnancy detection relied on either an emergency room visit or prenatal labs, likely underestimating overall risk of pregnancy, there is less reason to support differential bias for this outcome.

A retrospective cohort study of U.S. claims data from commercial health (i.e., private) insurance compared 42,391 women who underwent laparoscopic procedures to 27,724 women who underwent hysteroscopic tubal occlusion between 2008 and 2012.⁵⁴ Women in the hysteroscopic group had a higher rate of subsequent hysteroscopic surgeries (diagnostic hysteroscopy, endometrial ablation, myomectomy) compared to the laparoscopic group (17.5% vs. 13.4%) at 5 years, after excluding repeat hysteroscopic sterilization attempts. More women in the hysteroscopic group were diagnosed with menstrual dysfunction (26.8% vs. 22.3% at 2 years, adjusted HR 1.23, 95% CI 1.20-1.27) and had a higher rate of subsequent hysteroscopic surgeries. In contrast, they had a lower rate of subsequent intra-abdominal gynecologic surgery (15.7% vs. 17.8%) at 5 years including hysterectomy and laparoscopy. Both groups had equal rates of subsequent salpingectomy by 5 years, 1.7% versus 2.2%. There was also a lower rate of pelvic pain diagnoses in the hysteroscopic group (21% vs. 25.6%, HR 0.83, 95% CI 0.80-0.85). Women in the hysteroscopic group were more likely to become pregnant (2.4% vs. 2.0%, adjusted HR 1.20, 95% CI 1.09-1.33) in the intention-to-treat analysis, which is likely due to only 66% of women obtaining HSG to confirm occlusion. Among women who received an HSG, there was no difference in pregnancy rates compared to women undergoing laparoscopic procedures (1.8% vs. 2% at 2 years, adjusted HR 0.90, 95% CI 0.80-1.02). While statistically significant, the differences between outcomes after hysteroscopic compared to laparoscopic sterilization are very small and not clinically significant.

Another group used Medicaid claims data to compare outcomes at 24 months for 3,929 hysteroscopic tubal occlusions and 10,875 laparoscopic tubal ligations.⁵¹ In this study, after controlling for age, race, baseline pain, and abnormal uterine bleeding diagnoses, hysteroscopic procedures were associated with a slightly lower risk of subsequent hysterectomy (OR 0.77, 95% CI 0.60-0.97) or pelvic pain diagnosis (OR 0.91, 95% CI 0.83-0.99) at 2 years postprocedure. There was no difference in abnormal uterine bleeding between groups. The inability to control for baseline confounding inherent in database studies, self-selection, and lack of information on attempted but failed hysteroscopic procedures represent major limitations of the study.

In the United Kingdom, investigators performed a comparative observational cohort study at a single hospital comparing prospectively collected data on 1,085 women who underwent hysteroscopic procedures with the retrospective data of 2,412 who had undergone laparoscopic tubal ligation with Filshie clip between 2005 and 2015.⁶⁷ Hysteroscopic tubal occlusion was successful on first attempt in 91.4% of cases, and laparoscopy was successful in 99.5% of cases (OR 18.8, 95% CI 10.2-34.4). Of the failed insertions, 6.5% resulted from device failure, 34.4% occurred due to difficulty visualizing the tubal ostia, 43% due to tubal spasm, and 16.1% due to inability of the patient to tolerate the procedure under local anesthesia. Of the failed laparoscopy procedures, 8.3% were due to mesosalpinx laceration and 91.7% due to pelvic adhesions. Most (97%) women who had successful hysteroscopic procedures returned for confirmatory testing, of whom 91% had tubal occlusion documented and could rely on Essure for contraception. In all, 902/1,085 (83.1%) of successfully performed hysteroscopic procedures with confirmatory testing were considered satisfactory, compared to 2,400/2,412 (99.5%) of laparoscopic procedures (OR 40.6, 95% CI 22.5-73.1). Women in the hysteroscopic group had six times greater odds of reoperation by 1 year postoperatively (2% vs. 0.3%, OR 6.2, 95% CI 2.8-14.0). Reoperations included removal of incorrectly placed devices, second-stage hysteroscopic procedures to achieve bilateral coil placement, and laparoscopic permanent contraception. The one-year pregnancy risk was similar, with three reported pregnancies after hysteroscopic permanent contraception and five after laparoscopic permanent contraception (OR 1.3, 95% CI 0.3-5.6). Possible limitations to external generalizability include the study population’s access to universal health care, which likely facilitates radiographic confirmation testing, and lack of information on race, ethnicity, and comorbidities. Furthermore, it is unclear if pregnancy rates after hysteroscopic procedures were only measured for successful procedures, and not by intention to treat.

In Finland, investigators utilized national registries to compare pregnancy rates and repeat operations from 2009 to 2014 following 4,425 laparoscopic tubal occlusion procedures with Filshie clips, 5,631 hysteroscopic tubal occlusion procedures with Essure, and 6,216 postpartum Pomeroy partial salpingectomies.⁷⁹ There was no significant difference in pregnancy rates between the three groups, but women in the hysteroscopic group had higher rates of repeat sterilization operations (Table 4.2). Similar to other European studies, this analysis is limited by its external generalizability for a U.S. population due to study subjects’ access to universal health care, which likely facilitates radiographic confirmation testing, and its dissimilarity to the U.S. diversity of race, ethnicity, comorbidities, and BMI, which are not reported in this study.

Finally, a large French study used national claims data to evaluate 105,357 women who underwent hysteroscopic or laparoscopic permanent contraception between 2010 and 2014.⁵² Outcomes included risk of surgical complications, gynecologic outcomes (sterilization failure, second sterilization procedure, pregnancy), and medical outcomes (allergy, autoimmune disease, thyroid disorder, use of analgesics, use of migraine medication, antidepressants, benzodiazepine use, sick days, outpatient visits, suicide attempt, death) within 1 to 3 years after the procedure. Overall, 71,303 (67.7%) women underwent hysteroscopic and 34,054 (32.3%) women underwent laparoscopic permanent contraception. The risk of immediate surgical complications was lower with the hysteroscopic approach compared to the laparoscopic approach (0.13% vs. 0.78%, adjusted OR 0.18, 95% CI 0.14-0.23). There were no deaths in either group. Women in the hysteroscopic group had a higher risk of undergoing a second permanent contraception procedure at 1 year compared to the laparoscopic group (4.10% vs. 0.16%, adjusted HR 25.99, 95% CI 17.84-37.86). In addition, at one year, the hysteroscopic group had a higher risk of tubal disorder or surgery including salpingectomy (0.70% vs. 0.23%, adjusted HR 2.98, 95% CI 2.17-4.10). In contrast, women in the hysteroscopic group had a lower risk of uterine disorders (1.28% vs. 1.50%, adjusted HR 0.85, 95% CI 0.74-0.98), abnormal vaginal bleeding (0.23% vs. 0.33%, adjusted HR 0.71, 95% CI 0.52-0.96), and hysterectomy (0.43% vs. 0.81%, adjusted HR 0.54, 95% CI 0.44-0.66) compared to the laparoscopic group. At 3 years, pregnancy rates did not differ between the hysteroscopic group and the laparoscopic group (0.48% vs. 0.57%, adjusted HR 1.04, 95% CI 0.83-1.30). A composite outcome of sterilization failure (including salpingectomy, second sterilization procedure, and pregnancy) was significantly higher at 1 year in the hysteroscopic group compared to laparoscopic group (4.83% vs. 0.69%, adjusted HR 7.11, 95% CI 5.92-8.54). This difference persisted through 3 years (5.75% vs. 1.29%, adjusted HR 4.66, 95% CI 4.06-5.34). For medical complications, there was no difference between the two groups in autoimmune disease and thyroid disorders. Hysteroscopic procedures were associated with lower use of analgesics, antidepressants, benzodiazepines, outpatient visits, and sick days at 3 years compared to laparoscopic procedures. There was no real difference in the risk of suicide attempt or death in the two groups at 3 years. At 1 year, risk of allergies was slightly higher in the subgroup of women with previous allergies in those that underwent hysteroscopic versus laparoscopic permanent contraception (43.2% vs. 40.0%, adjusted HR 1.10, 95% CI 1.03-1.17), but this small difference does not seem clinically important.

Table 4.2 Outcomes Evaluated After Hysteroscopic, Laparoscopic, and Open Permanent Contraception Procedures in Finland Between 2009 and 2014

Method	Hysteroscopic (Essure) n = 5,631	Laparoscopic (Filshie Clip) n = 4,425	Minilaparotomy (Pomeroy) n = 6,216
Second permanent contraception procedure^*	229 (4.07%)	82 (1.85%)	53 (0.85%)
Spontaneous pregnancies^†	38 (0.67%)	34 (0.77%)	92 (1.48%)
Spontaneous pregnancies (per 100 follow-up years)	0.20	0.25	0.26
Source: Jokinen E, Heino A, Karipohja T, Gissler M, Hurskainen R, Safety and effectiveness of female tubal sterilisation by hysteroscopy, laparoscopy, or laparotomy: a register based study, BJOG 124(12):1851-1857, 2017.
^*p < 0.0001 for hysteroscopic vs. laparoscopic.^†p = 0.67 for hysteroscopic vs. laparoscopic.