Tubal Surgery and Treatment of Infertility

The fallopian tube plays a critical role in conception as the site of ovum transport, sperm capacitation, fertilization, and embryo transport.¹ The tube measures an estimated 9 to 11 cm, and narrows from the ciliated infundibulum and ampulla to the medial one-third of isthmus before funneling into the interstitial segment of the myometrium. The ampulla measures 5 to 9 cm and is necessary for fertilization and early embryogenesis.²

Tubal disease is identified in up to 30% of couples undergoing an infertility investigation, and both reconstructive surgery and in vitro fertilization (IVF) are treatment options. In the past decade, IVF success rates have increased from 10% in the 1980s to more than 30% according to national data banks.³ In contrast, success rates for most forms of tubal surgery have remained relatively constant during the same interval. Given improvements in IVF, it is now the treatment of choice for many types of tubal injury, particularly for couples with multiple infertility diagnoses. Surgery, however, remains a necessary reproductive choice for couples with ethical, religious, or financial concerns associated with the IVF process. The purpose of this review is to clarify the evaluation for and contemporary role of reconstructive surgical techniques in the management of infertility secondary to tubal disease.

For most forms of tubal surgery, a limited preoperative evaluation is necessary. Diagnostic tests include a semen analysis and documentation of ovulation (using basal body temperature records, urine luteinizing hormone [LH] predictor kits, or luteal serum progesterone assays). In women older than age 35 years, basal follicle-stimulating hormone (FSH) testing is recommended to identify those who may have a decreased ovarian reserve that would make them poor candidates for surgery. Hysterosalpingography and laparoscopy can be used to define the location, extent, and degree of tubal damage. A discussion of the use of these tools in the selection of candidates for tubal reconstructive procedures follows.

The United States National Survey of Family Growth estimates that approximately 1 million women undergo elective tubal sterilization procedures annually, and that 1% of these women will later request its reversal. In a prospective study that surveyed women after tubal ligation, up to 7% regretted the decision. Forty-percent of these women were between the ages of 25 and 29 years.⁴ Given these findings, it is likely that many couples will seek a surgical specialist for consultation. In couples who desire future fertility, tubal reanastomosis and IVF are both legitimate options that offer favorable outcomes.

A limited preoperative evaluation is necessary before considering tubal reanastomosis. Review of operative notes and pathology reports can determine the type and extent of the ligation procedure performed. Hysterosalpingography may be necessary in selected cases to assess the length of proximal oviduct and confirm the type of sterilization procedure performed. Extensive cautery or removal of large segments of the tube may predict a poor postoperative prognosis.

The use of preoperative laparoscopy adds an additional element of cost and risk and is not necessary in most cases. An algorithm proposed by Opsahl defines patients requiring diagnostic laparoscopy before laparotomy and tubal ligation reversal.⁵ Patients with previous Pomeroy, loop, Hulka clip, Irving, and single-burn cautery tubal ligation techniques were considered good candidates for laparotomy (without previous laparoscopy) when it could be anticipated that the final length of at least one tube would be 3 cm or more. In this series, patients with multiple-burn cautery or ligations by unknown procedures underwent diagnostic laparoscopy before laparotomy. Using this algorithm, only 3.8% of 185 patients who did not have laparoscopy were found to have inoperable tubes at laparotomy, with most cases involving a previous unilateral salpingectomy. These data and our own clinical experience indicate that preoperative laparoscopy is indicated when an operative note is unavailable, or when there is evidence of either extensive tubal resection or pelvic pathology at the time of tubal ligation.

Microsurgical techniques require magnification, meticulous hemostasis, minimal tissue handling, methods to prevent desiccation, minimally reactive suture, and avoidance of peritoneal irritants.⁶ Microsurgery results in superior outcomes for salpingolysis, fimbrioplasty, neosalpingostomy, midsegment anastomosis, tubouterine anastomosis, tubouterine implantation, and tubal reanastomosis when compared with macrosurgery.^7,8 In two small series, microsurgical outcomes using loupe magnification or the operating microscope were similar.^9,10

Success rates after tubal reanastomosis are dependent on multiple factors (Table 1). Using a logistic regression, te Velde and coauthors analyzed 215 surgical procedures and determined that the location of the anastomosis, the number of tubes repaired, and the presence of coexisting fertility factors were key variables that predicted success.¹¹ Other studies indicate that success rates are dependent on maternal age and postoperative tubal length.^12,13 Microsurgical treatment rates are higher with isthmic–isthmic anastomoses after fallope ring or clip application,^14,15 and lower when electrocautery was used for the tubal ligation procedure.¹⁶ In the case of extensive adhesions or insufficient tubal length, it may be possible to treat only one tube. Cumulative pregnancy rates are similar when one or both tubes are repaired (61% versus 56%), although the time until conception is longer after unilateral repair.¹³

Table 1. Efficacy of Tubal Ligation Reversal by Methodology

Laparoscopic tubal reanastomosis is an option for highly skilled surgeons experienced in the technique. Most series describe a steep learning curve with the first 10 to 20 procedures, but the procedure can be performed in less than 3 hours with experience.¹⁷ Laparoscopy in conjunction with robotic surgery is currently an area of active investigation. In a small series, Falcone and colleagues completed 10 robotically assisted laparoscopic tubal reanastomoses with a mean operating room time of 159 minutes, and a 50% pregnancy rate after 12 months.¹⁸

In carefully selected patients, success rates are excellent after surgical reanastomosis. In a large review, pregnancy rates after laparotomy ranged from 45% to 82% and were slightly lower for laparoscopic reversal (25%–53%). The average monthly fecundability rate ranges between 8% and 10%. The ectopic pregnancy rate is 1% to 7%, with higher rates observed when the distal segment consists of only ampulla.^19,20

The use of tubal reanastomosis in women older than age 40 is controversial. While intrauterine pregnancy rates of greater than 40% by laparotomy and as high as 71% by laparoscopy are reported,^17,21 other experienced surgeons report lower rates. Glock published surgical outcomes in 52 cases, reporting a conception rate of 43% and spontaneous abortion rate of 23.8%. The live birth rate of only 14.3 %, with no live births in women older than age 43; Winston reported similar results in women older than age 40.^22,23

Of all surgical treatments for tubal disease, sterilization reversal results in the highest pregnancy and fecundability rates. The procedure is appealing to couples who desire an extended family or who are not comfortable with IVF. The ectopic pregnancy risk and the need for future contraception are recognized disadvantages, as is the necessity of laparotomy in most cases. Based on the collective data, the primary variables for success are location of the anastomosis, postoperative tubal length, and maternal age. In carefully selected cases, microsurgery is an alternative to IVF, particularly in younger women. In older women, the role of microsurgery is less certain but could be considered in patients with a favorable prognosis or those who are reluctant to consider IVF.

Distal tubal injury without occlusion can be treated by fimbriolysis or fimbrioplasty. Fimbriolysis is the separation of attached or phimotic fimbria, whereas fimbrioplasty involves repair of partially occluded fimbria. The preoperative investigation is similar to that outlined for tubal reanastomosis, except that the diagnosis and staging of fimbrial disease generally requires laparoscopy because the diagnostic accuracy in defining fimbrial disease using hysterosalpingography is poor.

Surgical outcomes after tubal repair are inversely proportional to the severity of disease. The degree of tubo-ovarian adhesions, tubal thickness, and ciliary and ampullary damage are recognized variables that predict surgical outcome.¹ In the event of minimal disease, term pregnancy rates can surpass 50%,²⁴ but when disease is more severe, pregnancy rates are markedly lower (22%–25%) and the risk of ectopic pregnancy is 12%.

The monthly fecundability rate is estimated to range between 2% and 4%, and both laparoscopy and laparotomy can be used successfully with comparable pregnancy outcomes (Table 2). In younger women, laparoscopic-directed surgical repair is an alternative to IVF. When there is no conception after the first postoperative year, IVF should be considered. In older women, or in women with dense intraperitoneal adhesions and associated tubal disease, IVF is a more likely successful and, therefore better, option.

Table 2. Efficacy of Fimbrioplasty and Lysis of Adhesions in Selected Large Series

NA = not available

Salpingostomy is the repair of complete distal tubal occlusion. The distinction is important because electron microscopic studies have demonstrated more significant mucosal damage in tubes that are completely occluded, as compared with damaged but still patent tubes.⁶ The diagnosis is generally made by hysterosalpingography, but laparoscopy is necessary to define and stage the extent of tubal damage. Numerous staging paradigms designed to predict operative success have been devised.^25,26,27 Whereas nearly all staging systems are based on post hoc analyses and lack validation, they are nonetheless helpful for characterizing the extent of tubal disease and for selection of surgical candidates. In cases of mild disease that are characterized by hydrosalpinges having a diameter less than 15 mm with few associated adhesions and recognizable fimbriae, pregnancy rates approach 80%. In contrast, severe disease, characterized by larger hydrosalpinges exceeding 30 mm in diameter with dense associated adhesions and no visible fimbriae, cumulative pregnancy rates are poor (10%–15%). The extent of adhesions, macroscopic appearance of the endosalpingeal mucosa, and tubal wall thickness also correlate with operative success.^28,29,30

Overall pregnancy rates and ectopic pregnancy risks after laparotomy vary widely, ranging from 13% to 37% and from 2% to 22%, respectively (Table 3). Differences in operative skill, techniques, and the degree of preexisting pelvic pathology all contribute to the differing reports of outcomes. As a rule, pregnancy rates correlate with the degree of tubal damage. For mild distal tubal disease (25% of total cases), live birth rates range from 39% to 59%, and risk for ectopic pregnancy varies between 4% and 10%.^28,31 Pregnancy rates approach 15% at 1 year, with most conceptions occurring within the first 2 postoperative years. In contrast, the outcome for severe disease (more than 30% of cases) is uniformly poor, with pregnancy rates less than 15%. Moreover, ectopic pregnancy rates increase with the severity of tubal damage. Risk for ectopic pregnancy increases with the degree of ampullary dilatation, ranging from 5%, with diameters between 15 and 25 mm, to 12%, when ampullary dilatation exceeds 25 mm.²⁹ Although initial attempts at laparoscopic neosalpingostomy were not very effective,³² experienced surgeons now report pregnancy rates that compare favorably with those achieved by laparotomy.

Table 3. Efficacy of Salpingostomy, Selected Large Series

NA = not available

Tubal patency rates after neosalpingostomy are substantially higher than pregnancy rates.^1,33 In a small series, Daniell reported a 98% tubal patency rate after neosalpingostomy using C0₂ laser at laparotomy, but fewer than 20% achieved an intrauterine pregnancy and 2% had an ectopic pregnancy.³⁴ These and other data indicate that even though patency can most often be restored, severely damaged tubes may still not function because tubal ciliary regeneration is slow and often fails altogether.

Limited disease can be treated with neosalpingostomy in younger women, but IVF should be considered after the first postoperative year. In older women, IVF is most often the best option, particularly when one considers the low monthly fecundability (1%–2%) observed after distal tubal surgery. The role of surgery for moderate or severe disease is less clear. Although more severe disease can be diagnosed and treated at laparoscopy, IVF should be considered the preferred therapy because of the low probability of surgical success and the high risk of ectopic pregnancy or tubal reocclusion.³⁵

For women with severe distal tubal disease, the presence of large hydrosalpinges may adversely impact IVF success. In an early retrospective study of 118 patients with hydrosalpinges, pregnancy rates were lower and miscarriage rates higher than those of controls with tubal disease but no hydrosalpinges.³⁶ A variety of additional studies have since confirmed these findings, and a recent meta-analysis that included 5592 patients (1004 with hydrosalpinx and 4588 with tubal infertility without hydrosalpinx) found lower pregnancy, implantation, and live birth rates (odds ratio [OR]: .64; confidence interval [CI]: .56, .74) and a higher incidence of early pregnancy loss in women with hydrosalpinges.³⁷

Several pathophysiologic models have been suggested to explain the lower pregnancy rates achieved with IVF in women with a hydrosalpinges.³⁷ The first model suggests that fluid from the hydrosalpinx may flow toward the uterine cavity and thereby hinder implantation by mechanically washing the uterine cavity.³⁸ A second hypothesizes that hydrosalpingeal fluid may contain cellular or infectious debris, lymphocytes, cytokines, prostaglandins, leukotrienes, and catecholamines, any or all of which may have adverse inflammatory, infectious, or immunological effects.³⁷ Several experiments in animals indicate that fluid from human hydrosalpinges retards the development of mouse embryos, suggesting that such fluid is embryotoxic.^37,39,40,41 A third mechanism envisions that hydrosalpingeal fluid may lower levels of endometrial cell surface proteins (integrins) necessary for implantation, normal expression of which is restored after salpingectomy.^37,42 Finally, there is the possibility that hydrosalpinges may in some way have direct adverse effects on oocytes early in follicular recruitment.^39,43

Collectively, the available evidence argues strongly that hydrosalpinges decrease pregnancy, implantation, and live birth rates and increases risk for early pregnancy loss in women after IVF.³⁷ Although there have been no randomized controlled trials to test the assertion directly, salpingectomy is generally considered preferable to neosalpingostomy before IVF.⁴³

The intramural portion of the tube that traverses the uterine wall measures 1.5 to 2.5 cm in length and has a luminal diameter between .4 and 1 mm. In hysterectomy specimens, the course of this segment is tortuous in 69%, straight in 23%, and curved in 8%.⁴⁴ Proximal tubal occlusion represents approximately one-third of all cases of tubal occlusion diagnosed by hysterosalpingography,¹ but false-positive results are common, occurring in 16% to 40% of imaging studies.⁴⁵ Although repeat imaging decreases the false-positive rate, laparoscopy is usually necessary to establish an accurate diagnosis and is useful for treating any coexisting tubo-ovarian disease that may be identified in up to 20% of cases.⁴⁶

Proximal tubal obstruction results from many causes, including infection, endometriosis, and inflammation.⁴⁷ In a histologic review, obliterative fibrosis (38.1%) was the most common diagnosis, followed by salpingitis isthmica nodosa (SIN) (23.8%), endometriosis (14.3%), and chronic inflammation (21.4%).^27,48 Obliterative fibrosis occurs when dense, collagenous connective tissue obstructs the proximal tubal lumen as a consequence of infection or a foreign body.¹ In contrast, SIN is characterized by muscular hyperplasia and hypertrophy surrounding pseudoglands lined with tubal epithelium.¹ Evidence suggests that SIN is a progressive disease that frequently results in complete tubal occlusion.⁴⁹ The etiology of SIN remains controversial. Infectious, inflammatory, mechanical, hormonal, and congenital causes have been suggested.⁴⁹ Occasionally, intrauterine or intratubal polyps may occlude the oviduct. Although pregnancy rates are no lower in patients with polyps than in those with unexplained infertility,⁵⁰ their removal has been advocated by some.¹

Microsurgical resection and reanastomosis is a proven treatment for proximal tubal obstruction, with pregnancy rates ranging from 54% to 58% after the procedure.^27,47,51 Macrosurgical and implantation techniques result in inferior pregnancy rates and should no longer be used. Honoré recently summarized the world's surgical literature, reporting a 58.9% total pregnancy rate, 47.4% ongoing pregnancy rate, and 7.4% ectopic pregnancy rate.⁴⁹ Monthly fecundability ranges between 5% and 6%, but outcomes depend on the etiology of the obstruction. Better pregnancy rates follow treatment of SIN and range between 44% and 58%. However, the prognosis is poor when occlusion results from endometriosis or chronic inflammation, because reocclusion rates as high as 63% have been observed.^1,45

Proximal tubal cannulation using hysteroscopic- or fluoroscopic-directed methods is a proven alternative to abdominal surgery. Pregnancy rates with fluoroscopic tubal recanalization range from 48% to 85%, with reocclusion rates of 28%.⁵² In the largest review of fluoroscopically guided proximal tubal cannulations, including 11 studies and 253 patients, at least one fallopian tube was successfully cannulated in 79% of treated women, and risk of tubal perforation (5%) was low.⁴⁵ Postprocedure follow-up after 1 to 26 months yielded a 34% pregnancy rate, comparable with that achieved with tubocornual reanastomosis. The average fluoroscopy time was 7.5 minutes, yielding 1.2 rads of radiation. Others have reported lower radiation doses to the ovary (0.85 ± 0.56 rads) that are comparable with those associated with barium enema.⁵³ In a review of five studies involving a total of 25 patients treated by hysteroscopic-directed tubal cannulation, at least one tube was successfully cannulated in 92% of treated women. Among these, 24% subsequently achieved an intrauterine pregnancy and 12% had an ectopic pregnancy. Perforation occurred in 11% of cases but required no additional or specific treatment.

Proximal tubal cannulation may not be quite as effective as tubocornual anastomosis, but it is certainly a less morbid procedure. One proposed treatment algorithm⁵² suggests laparoscopy to exclude endometriosis, adhesions, and distal tubal disease, followed by hysteroscopic-directed tubal cannulation when findings of proximal tubal occlusion are confirmed. When laparoscopy has been performed previously, fluoroscopic cannulation is an attractive choice for treatment. Tubocornual anastomosis or IVF generally may be reserved for those in whom surgical treatment has proven unsuccessful.

Bipolar tubal disease involves coexistent proximal and distal tubal obstruction. In general, success rates achieved with surgical repair of ipsilateral proximal and distal occlusion are poor. In one reported series involving 31 patients, the conception rate at 2.5 years was only 12% and there were no live births.⁵⁴ Others⁶ have reported results that are only modestly better (28% term pregnancy, 3% ectopic pregnancy). In a multicenter trial of transcervical balloon tuboplasty for proximal tubal occlusion, pregnancy rates for women with both distal and proximal disease were significantly lower than for those with only proximal tubal disease (49% versus 12%).⁵⁵ In a study comparing tubal microsurgery and IVF, women with thick-walled hydrosalpinges or combined proximal and distal tubal disease achieved better results with IVF.⁵⁶ In sum, the available data indicate that IVF is a better option than microsurgical repair for women with bipolar tubal occlusive disease.

Advances in assisted reproductive technology now provide a very effective means for treatment of tubal disease, to an extent that tubal reconstructive surgery is in an era of decline. Nonetheless, tubal surgery remains as a highly effective treatment for women seeking pregnancy after a previous tubal sterilization and represents a legitimate alternative to IVF for those with mild distal tubal disease, particularly when they are young, and women with proximal tubal occlusion (Table 4). Tubal surgery should also be viewed as an effective adjuvant to IVF for women with hydrosalpinges.

Table 4. Efficacy of Treatments for Proximal Tubal Disease

NA = not available