Original research articleProbability of pregnancy after sterilization: a comparison of hysteroscopic versus laparoscopic sterilization☆,☆☆
Introduction
Female surgical sterilization is the most popular method of pregnancy prevention worldwide and is the most commonly used method of contraception among women age 35 years and older in the United States (U.S.) [1], [2]. Each year, 345,000 U.S. women undergo sterilization procedures and a total of 10.3 million U.S. women rely on female sterilization for pregnancy prevention [3], [4].
Since the introduction of a hysteroscopic approach in 2001, an increasing number of women are undergoing hysteroscopic sterilization instead of laparoscopic sterilization [5], [6], [7]. Hysteroscopic sterilization has several advantages over laparoscopic sterilization: it avoids abdominal entry, can be performed as an office procedure and may avoid general anesthesia [5]. More than 650,000 hysteroscopic sterilization procedures have been performed worldwide [5]. One U.S. academic center reported that the proportion of interval sterilization performed laparoscopically from 2002 to 2006 decreased by 50% with a corresponding increase in the proportion of procedures performed by hysteroscopic sterilization by 50% [6].
However, hysteroscopic sterilization has limitations as well. The likelihood of achieving successful bilateral coil placement on first attempt varies from 76% to 96% [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22]. In addition, unlike laparoscopic sterilization, hysteroscopic sterilization is not immediately effective; at least 3 months is required for tubal fibrosis and occlusion to occur for the procedure to be effective. During these 3 months, women need to use alternative contraception until they can undergo a post-procedure hysterosalpingogram (HSG) to confirm bilateral tubal blockage [23]. Prior research has shown that some (6–87%) women never return for their HSGs [8], [10], [12], [13], [14], [16], [17], [18], [19], [24] and that blockage does not occur in 5–16% of HSG evaluations 3 months post-procedure [8], [12], [13], [14], [16], [17], [18], [19]. The multiple steps involved in hysteroscopic sterilization, including the 3-month delay in possibly achieving sterilization, can increase the risk of patient non-compliance with this clinical care protocol and subject women to contraceptive failures (unintended pregnancies) during the process [23].
For any new method of contraception or sterilization, the most important variable to scrutinize is effectiveness. Unfortunately, the literature on hysteroscopic sterilization is limited by lack of such data [2], [25], [26], [27]. Most studies of its efficacy have excluded women who failed initial microinsert placement did not return for HSG or who became pregnant before their HSG [23]. Furthermore, there are no prospective studies comparing the effectiveness of hysteroscopic and laparoscopic sterilizations. Most studies that do retrospectively report pregnancies after hysteroscopic sterilization are limited by small study numbers, short follow-up duration, lack of reporting follow-up duration and high loss to follow-up [27], [28].
We sought to gain a more objective and comprehensive understanding of hysteroscopic sterilization outcomes, based on the published literature. In the absence of a prospective study directly comparing short- and long-term probability of pregnancy after hysteroscopic and laparoscopic sterilization procedures, the best way to understand the consequences of the two contraceptive approaches is through a decision analytic model. Decision analysis can account for the complexity of the multi-step process for coil placement and follow-up, can incorporate the variability in clinical outcomes reported in the literature and can compare the expected probability of pregnancy after hysteroscopic and laparoscopic sterilization. Also, by mapping out the clinical pathway of these procedures, it provides a unique opportunity to identify knowledge gaps in the current literature and help set priorities for future research.
Section snippets
Study design
We developed a Markov state-transition model (Fig. 1) to estimate the probability of pregnancy following three sterilization strategies: hysteroscopic, laparoscopic with silicone rubber band application (falope rings) and laparoscopic with bipolar coagulation. Laparoscopic sterilization was chosen as the comparator for the newer hysteroscopic sterilization procedure, since it is the standard of care for interval (not related to pregnancy) female sterilization [2]. Using yearly cycles, the
Results
In the base case analysis at all points in time after the sterilization procedure (ranging from 1 year to 10 years after procedure initiation), the expected cumulative risk of pregnancy after sterilization is higher in women opting for hysteroscopic than laparoscopic sterilization using silicone band application or bipolar coagulation. Pregnancy risk after hysteroscopic sterilization is primarily accrued in the first year after initiating the process. The expected pregnancy rates per 1000 women
Discussion
Based on best data currently available, our model suggests that hysteroscopic sterilization is not as effective as laparoscopic sterilization in preventing pregnancy when the complete clinical pathways of the procedures are considered. Our analysis improves upon prior studies of hysteroscopic sterilization by taking into account uncertainties in successful placement of coils, return for HSG and successful blockage of tubes. Reflecting these real-life circumstances, our base case estimates
Acknowledgment
The authors thank Lisbet Lundsberg, PhD, for her assistance with the literature review.
References (35)
- et al.
Trends in sterilization since the introduction of Essure hysteroscopic sterilization
J Minim Invasive Gynecol
(2009) Essure: a review six years later
J Minim Invasive Gynecol
(2009)- et al.
Essure: a new device for hysteroscopic tubal sterilization in an outpatient setting
Fertil Steril
(2004) - et al.
Post-Essure hysterosalpingography compliance in a clinic population
J Minim Invasive Gynecol
(2008) - et al.
Prospective analysis of office-based hysteroscopic sterilization
J Minim Invasive Gynecol
(2006) - et al.
Outcomes of transcervical hysteroscopic sterilization in an urban academic medical center
Contraception
(2009) - et al.
Microinsert nonincisional hysteroscopic sterilization
Obstet Gynecol
(2003) - et al.
A comparative study of hysteroscopic sterilization performed in-office versus a hospital operating room
J Minim Invasive Gynecol
(2006) - et al.
Cost-effectiveness analysis comparing the essure tubal sterilization procedure and laparoscopic tubal sterilization
J Obstet Gynaecol Can
(2008) - et al.
Hysteroscopic tubal sterilization: a systematic review of the Essure system
Fertil Steril
(2010)
Pregnancies after hysteroscopic sterilization: a systematic review
Contraception
Estimates of contraceptive failure from the 2002 National Survey of Family Growth
Contraception
Contraceptive failure in the United States
Contraception
A summary of reported pregnancies after hysteroscopic sterilization
J Minim Invasive Gynecol
Pregnancy after tubal sterilization with bipolar electrocoagulation. U.S. Collaborative Review of Sterilization Working Group
Obstet Gynecol
Sterilization
Obstet Gynecol
Cited by (32)
Female permanent contraception trends and updates
2022, American Journal of Obstetrics and GynecologyCitation Excerpt :Associated complications were rare, although procedures were often performed in patients with more medical and surgical comorbidities than patients who underwent laparoscopic procedures.12,17,18,19 The requirement for a 3-month tubal occlusion confirmation test delayed knowing whether one could rely on the method for contraception and thus was a key disadvantage.20 Bilateral placement of Essure coils was unsuccessful in up to 24% of procedures, and nonocclusion occurred in 5% to 16% at 3 months.21,22
Reviving permanent contraception: New medical procedures or new service delivery modalities?
2020, Best Practice and Research: Clinical Obstetrics and GynaecologyCitation Excerpt :Data do not include the contemporary tubal occlusion methods, such as the hysteroscopy devices Adiana, Ovabloc, and Essure. Essure is currently the only tubal insert on the market with an expected population risk of pregnancy of 9.6% [41]. Further studies of the hysteroscopy approaches are required.
Intrauterine fluid instillation to confirm tubal occlusion after transcervical permanent contraception: A pilot study<sup>,</sup>
2020, ContraceptionCitation Excerpt :While this verification provides evidence of occlusion, failure rates in real-world clinical use with Essure® appear higher than in clinical studies. That estimates of actual completion of the verification test vary widely (6–87%) likely explains this difference [4]. This suggests current strategies for confirmation do not meet the needs of clinicians and women.
Controversies in tubal sterilization
2016, Gynecologie Obstetrique et FertilitePregnancy following hysteroscopic sterilization at an Urban Abortion Clinic- A case series: Are We looking in the wrong places?
2016, ContraceptionCitation Excerpt :The manufacturer reported to Essure® be 99.97% and 99.83% effective at 1 and 5 years respectively [4] with the Adiana® system having similar 1-year pregnancy prevention rates [5]. Nonetheless, some evidence suggests that the typical use pregnancy rate following sterilization via the hysteroscopic approach might not be as low as previously thought, since most studies failed to do intention to treat analysis [6,7]. Given that nearly 40% of unintended pregnancies in the U.S. result in abortion [8], and approximately 70% of abortions in the U.S. are performed in outpatient abortion clinics [9] we sought to describe women with a history of HS presenting for abortion at a 3-site urban abortion clinic.
Hysteroscopic Essure Inserts for Permanent Contraception: Extended Follow-Up Results of a Phase III Multicenter International Study
2016, Journal of Minimally Invasive Gynecology
- ☆
Funding: A grant from the Society of Family Planning.
- ☆☆
Disclosure: The authors report no conflict of interest.