TECHNIQUES AND INSTRUMENTATION

Transcription

1 TECHNIQUES AND INSTRUMENTATION FERTILITY AND STERILITY VOL. 75, NO. 3, MARCH 2001 Copyright 2001 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Role of laparoscopic salpingostomy in the treatment of hydrosalpinx Rebecca C. Taylor, M.D., a Jonathan Berkowitz, Ph.D., b and Peter F. McComb, M.B., B.S. c University of British Columbia, Vancouver, British Columbia, Canada Received January 5, 2000; revised and accepted October 23, Supported by Vancouver Hospital and Health Sciences Centre, Vancouver, British Columbia, Canada. Presented in part at the Middle East Fertility Society Meeting, Sharm El Sheik, Egypt, November 24 26, Reprint requests: Peter McComb, M.B., B.S., Department of Obstetrics and Gynaecology, Vancouver Hospital and Health Sciences Centre, 805 West 12th Avenue, Willow Pavilion, Vancouver, British Columbia, Canada V5Z 1M9 (FAX: ; pmccomb@unixg.ubc.ca). a Department of Surgery, University of Ottawa, Ottawa, Ontario, Canada. b Berkowitz & Associates Consulting Inc., Vancouver, British Columbia, Canada. c Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynaecology, Vancouver, British Columbia, Canada /01/$20.00 PII S (00) Objective: To determine pregnancy rates after laparoscopic salpingostomy in occlusive distal tubal disease. To evaluate the relative impact of various historical, physical, and operative factors on pregnancy outcome using a multivariate statistical analysis. Design: Prospective cohort. Setting: University-affiliated tertiary care infertility clinic. Patient(s): One hundred thirty-nine infertile women with occlusive distal tube disease. Intervention(s): Laparoscopic salpingostomy. Main Outcome Measure(s): The occurrence of intrauterine (IUP) and ectopic pregnancy (EP). Result(s): The overall IUP and EP rates were 24.5% and 16.5%, respectively. Analysis of historical variables, assessed independently, demonstrated a significantly higher IUP rate with a positive history of gonorrhea and a significantly higher EP rate with a positive history of pelvic inflammatory disease, lack of history of intrauterine device (IUD) usage, or the performance of a bilateral procedure. The logistic regression model to predict intrauterine pregnancy had an overall predictive value of 77.5% and included the following significant variables: secondary infertility, positive history of gonorrhea, and the operative finding of moderate periadnexal adhesions. The logistic regression model to predict ectopic pregnancy had an overall predictive value of 89.0% and included the following significant variables: previous ectopic pregnancy, negative history of IUD use, positive history of PID, a bilateral procedure, and perihepatic adhesions. Conclusion(s): Operative laparoscopy may be effective for the correction of hydrosalpinges in selected patients. The probability of achieving an intrauterine or an ectopic pregnancy can be predicted based on combinations of significant variables. (Fertil Steril 2001;75: by American Society for Reproductive Medicine.) Key Words: Laparoscopy, fallopian tube diseases, infertility, adhesions, salpingostomy, neosalpingostomy, hydrosalpinx The role of terminal salpingostomy to treat distal tubal occlusion continues to be debated. Improvements in IVF success rates, in addition to the high ectopic pregnancy (EP) risk after salpingostomy, have led some to conclude that surgery for distal tubal obstruction should be abandoned (1, 2). By contrast, the relatively low morbidity and short hospital stay after laparoscopic salpingostomy and the potential for a number of subsequent pregnancies have lent support to the argument that surgery should be performed in all but the most severely damaged hydrosalpinges (3 5). To further complicate the issue, recent evidence suggests that the presence of hydrosalpinges has a negative effect on IVF outcome with reduced embryo implantation and increased risk of EP (6 9). Ideally, selected patients who have a reasonable probability of pregnancy after surgery should be offered a laparoscopic salpingostomy, whereas those with a very poor probability of an intrauterine pregnancy should proceed directly to IVF (with or without salpingectomy). Still others may benefit from a combination of salpingostomy followed by IVF with the view that either may result in a pregnancy and that resolution of the hydrosalpinx may improve IVF results (10 12). Still unresolved is the question of how to best classify patients into these prognostic groups to optimize the pregnancy outcome. To answer this question, the current study was designed to identify the prognostic variables that influence the occurrence of IUP and 594

2 EP after laparoscopic neosalpingostomy. Studies of patients who were selected for microsurgical salpingostomy by laparotomy have shown that the achievement of an IUP depends primarily on the preexisting pathology of the fallopian tube (13 29). Relatively few series have looked at the liberal use of the laparoscopic technique to treat hydrosalpinx (20 29). Most laparoscopic studies are inadequate for one of two reasons: they combine patients with partial tubal patency with those with total distal tubal occlusion, thus leading to falsely high pregnancy rates; or they have small numbers of patients and therefore lack the statistical power to draw conclusions. Moreover, none have considered an ectopic pregnancy to be a separate failure that warrants prediction. Finally, the statistical models previously used to establish the predictive value of a given variable are inadequate because they consider variables in isolation of one another. A large study designed to investigate the predictors for both the achievement of an intrauterine pregnancy and an ectopic pregnancy after laparoscopic neosalpingostomy for complete distal tubal occlusion, using a multivariate statistical model, has never before been reported. MATERIALS AND METHODS From April 24, 1984 to May 6, 1998, 172 consecutively referred patients underwent unilateral or bilateral laparoscopic neosalpingostomy, as defined by Gomel (30), for occlusive distal tubal disease. Infertility was primary in 65 cases and secondary in 74 cases. Thirty-three patients were excluded from the analysis for the following reasons: 23 had a contralateral patent tube, 7 had a previous salpingostomy by laparotomy, 2 had a congenital absence of the distal ampulla, and one had a previous tubocornual anastomosis. None of the patients had a history of genital tract tuberculosis. Institutional review board approval was not obtained because our study did not involve new surgical techniques. Since 1987, standardized demographic and clinical data considered to be relevant to tubal infertility have been entered prospectively in an ongoing administrative database system, dbase III (Ashton-Tate, Culver City, CA). From these data, the following historical variables were assessed: parity and previous ectopic pregnancy, duration of infertility, age at surgery, previous use of an intrauterine contraceptive device, history of pelvic inflammatory disease, sexually transmitted disease, postpartum endometritis, abdominal surgery including appendectomy, cigarette smoking, medical illness, medication use, history of male-factor infertility, endometriosis, menstrual dysfunction, and the use of antibiotics. The physical variables assessed were the following: hirsutism, galactorrhea, pelvic mobility, uterine size, and position. Before surgery, all patients underwent an infertility evaluation, including a semen analysis, an assessment of ovulatory status (by basal body temperature, luteal phase serum progesterone, and/or a timed endometrial biopsy), and hysterosalpingography. A male factor was defined by a sperm count of 10 million per milliliter or by mobility 40%. This definition of male factor is consistent with that of the University of British Columbia Fertility Clinic. Anovulation was defined by a monophasic basal body temperature record, a luteal-phase serum progesterone below 6.5 ng/ml, or an absence of secretory change in a timed endometrial biopsy. The laparoscopic salpingostomies were performed by one surgeon (P.F.M.) in all patients, using the sharp-incision technique previously described by McComb (31). Patients underwent a bilateral operation when both tubes terminated in operable hydrosalpinges and a unilateral operation when the contralateral tube was characterized by any of the following: an inoperable proximal block (5 cases), an inoperable hydrosalpinx (40 cases), a bipolar block (22 cases), or absence (19 cases). Inoperable hydrosalpinges typically were thick walled and encased in extensive, thick adhesions similar to a frozen pelvis. The operative findings were documented and graded so as to assess the following operative variables: presence and severity of endometriosis, presence of perihepatic scarring (Fitz-Hugh Curtis syndrome), American Fertility Society classification (32) of distal tubal occlusion, evidence of proximal tube disease including salpingitis isthmica nodosa, and the use of microsuture. Patients with asymmetrical distal lesions were included in the category corresponding to the least affected tube. This operative information was collected from the charts and entered into a Microsoft Excel file, which was then merged with the historical file. Patients were discharged home on the day of surgery and were requested to inform the surgeon about the fertility outcome. An extensive follow-up was performed in June Patients were traced through their previous telephone number, their family physicians, and the Internet. Repeated attempts were made to contact each patient, resulting in only six patients being lost to follow-up. In terms of analysis, these patients were considered not to have conceived. The primary outcome measures were the occurrence of an IUP, EP, or pregnancy of undetermined site. The EP rate was defined as the EP number per 100 patients undergoing salpingostomy as opposed to the usual expression of EP per 100 total pregnancies. An IUP was defined by the ultrasound presence of an intrauterine gestational sac and/or clinical evidence of an IUP. EP was defined by human chorionic gonadotropin level and the absence of ultrasound identification of an IUP and/or clinical evidence of an EP. The pregnancy site could not be determined where the hcg did not reach a level sufficiently high to allow ultrasound identification of an IUP. For each patient, only the first outcome was taken into consideration. For patients who conceived, the follow-up period was from the date of surgery to the last menstrual period before conception. For patients who did not become pregnant, the follow-up period was from the date of FERTILITY & STERILITY 595

3 surgery to the date of last contact, and patients were censored at this time. Eight patients were censored before the date of last contact because of an acquired inability to conceive, such as a hysterectomy. A secondary outcome measure was patency of the fallopian tubes at either follow-up hysterosalpingography or second-look laparoscopy using chromopertubation. The status of the fallopian tubes and any subsequent operations performed at the time of the second laparoscopy were recorded for analysis. Patients who underwent a second surgery, whether a repeat salpingostomy, a fimbrial phimosis repair, or a salpingovariolysis, were not censored at this time. Statistical Analysis Crude pregnancy rates were compared using the 2 test under assumption of a binomial distribution. The two-sample t-test was used for the comparison of age at surgery and time to achieve pregnancy. All statistical tests were two tailed. The 0.05 statistical significance level was used for each of the statistical tests. The cure rate and the monthly probability of pregnancy (fecundity rate) for those who achieved conception were calculated using the two-parameter exponential model of Guzick and Rock (33) and used to develop cumulative pregnancy curves. Prognostic variables affecting pregnancy outcome were examined in combination using multivariate logistic regression, with the outcome stratified into three groups. Models were assessed by overall predictive value (that is, correct classification rate). Final models were used to compute probabilities of IUP and EP for various combinations of statistically significant preoperative and postoperative variables. All statistical analyses were performed using SPSS Windows version 7.5. RESULTS Overall Pregnancy Rates The overall conception rate was 41.7% (18.0% live births, 6.5% spontaneous abortions, 16.5% ectopic pregnancies, 0.7% pregnancies of undetermined site). The mean time to conception was 17.7 months, with a range of 0.5 to 86.4 months. The mean time to IUP was not different from the mean time to EP. Of the patients who eventually did conceive, 55.2% did so within 12 months, and 84.5% did so within 36 months. The cumulative conception curve, EP curve, and IUP curve are shown in Figure 1. The cure rate to achieve an IUP was 22.0%, and the monthly fecundity rate was 6.2%. For EP, the corresponding figures were 17.3% and 4.4%. Relationship Between Independent Predictor Variables and Outcome The mean age of patients at surgery was years (range, 21.4 to 41.1 years). The mean age at surgery of patients who conceived ( years) was not significantly different from those who did not conceive ( years). However, the conception rate for patients 30 years FIGURE 1 Cumulative overall ( ), intrauterine ( ), and ectopic pregnancy ( ) rates. of age (53.2%, n 62) was significantly different (P 10.05) from the conception rate for patients between 30 and 35 years (26.5%, n 49) and for patients 35 years of age (32.1%, n 28). This reflected the significantly higher (P.05) EP rate in the youngest group. The average duration of infertility before surgery was 58 months (range, 2 to 212 mo). There was no difference in the duration of infertility among those patients who conceived and those who did not. Thirty-seven women had no identifiable etiologic factor. Women with a positive history for pelvic inflammatory disease had a similar IUP rate but a significantly higher (P.02) EP rate compared with women with no such history. Women with a positive history for IUD use also had a similar IUP rate when compared with women with a negative history, but they had a significantly lower (P.03) EP rate. The IUP rate for women with a positive history for gonorrhea was significantly higher (P.04) than for women without this history. No other etiologic factor had a significant effect on the fertility outcome. The surgery performed was unilateral in 86 patients and 596 Taylor et al. Pregnancy outcome after laparoscopic salpingostomy Vol. 75, No. 3, March 2001

4 bilateral in 53 patients. The IUP rate was similar in both groups, but the EP rate was significantly higher (P.001) after a bilateral operation as compared with after a unilateral operation. The presence of perihepatic adhesions was associated with an increased EP rate but was not statistically significant (P.06). There was no significant correlation between the overall AFS score and the occurrence of an IUP or EP. The IUP rate for patients with tubal wall of normal thickness was significantly higher (P.02) than patients with a moderately thick/edematous or thick/rigid tubal wall. None of the eight patients with a thick/rigid tubal wall conceived. The IUP rate was similar for all three mucosal grades, yet the EP rate was significantly higher (P.05) in patients with moderately damaged mucosa (24.3%, n 70) when compared with those with either mild or severe mucosal damage (9.4%, n 64). By contrast, the EP rate was significantly lower (P.038) in patients with moderately dense/vascular adhesions (0%, n 15) when compared to both those with none/filmy adhesion (21.9%, n 96) and those with severely dense adhesions (7.4%, n 27). Neither the ampullary diameter of the hydrosalpinx nor the extent of pelvic adhesions predicted subsequent pregnancy outcome. Of the 192 tubes operated on initially, 109 (56.8%) were reassessed postoperatively when pregnancy had not occurred. Tubal patency was observed in 84 tubes (77.1%), and 81% of patients had patency of at least one tube. Of these tubes, 92 were operated on after the introduction of sutures, and 17 were operated on before sutures were available. The tubal patency rate of these two groups was similar (76% and 82%, respectively) as were their pregnancy rates. Of the 59 patients who underwent a second-look laparoscopy, 11 (19%) had reformed adhesions, but in only three were the adhesions worse than those observed at initial laparoscopy. Adhesions were significantly (P.05) more likely to reform where there were dense adhesions present at the initial laparoscopy (36% reformed), as compared with flimsy or moderately dense/vascular adhesions (15% reformed). Relationship Between Multiple Predictor Variables and Outcome Logistic Regression Two outcomes, IUP and EP, were analyzed separately, using logistic regression, yielding two distinct statistical models. Variables included in each model significantly improve the overall predictive value of the model. Each of Tables 1 4 is classified according to information available from the patient s history, the preoperative model, and according to information gained later at laparoscopy, the postoperative model. The logistic regression model to predict IUP has an overall predictive value of 77.5%, a positive predictive value of 70.6%, a negative predictive value of 79.7%, a sensitivity of 26.5%, and a specificity of 94.2%. The variables included in the logistic regression model to TABLE 1 Calculated odds ratios and constants for the logistic regression model to predict IUP. Variable constant Standard error Odds ratio Preoperative conception Previous conception Gonorrhea Constant ( 0 ) Pre- and postoperative model Previous conception Gonorrhea Moderately dense adhesions Constant ( 0 ) predict IUP are presented with their calculated odds ratios, standard errors of the mean, and coefficients in Table 1. The logistic regression model to predict EP has an overall predictive value of 89%, a positive predictive value of 76.9%, a negative predictive value of 90.3%, a sensitivity of 45.4%, and a specificity of 97.4%. The variables included in the logistic regression model to predict EP are presented with their calculated odds ratios, standard errors of the mean, and coefficients in Table 2. The probabilities of achieving an IUP for each combination of significant variables, calculated using the logistic regression model, are presented in Table 3. The probabilities of achieving an EP for each combination of significant variables, calculated using the logistic regression model, are presented in Table 4. TABLE 2 Calculated odds ratios and constants for the logistic regression model to predict EP. Variable constant Standard error Odds ratio Preoperative model Previous EP Positive history for IUD use Positive history for PID Constant ( 0 ) Pre- and postoperative model Previous EP Positive history for IUD use Positive for PID Unilateral surgery Presence of perihepatic adhesions Constant ( 0 ) FERTILITY & STERILITY 597

5 TABLE 3 Clinical prediction (%) of IUP. Previous conception No previous conception Factors Gonorrhea No gonorrhea Gonorrhea No gonorrhea Preoperative Postoperative Moderate/vascular adhesions Mild/none or dense adhesions DISCUSSION Overall Pregnancy Rates It is difficult to compare our results with those reported in the literature because the nomenclature is confusing. Many reports combine salpingostomy in the overall fimbrioplasty category and do not distinguish between partial and total distal tubal occlusion, despite the fact that studies have repeatedly shown that fimbrial phimosis repair is associated with a significantly better prognosis (16, 23, 28). In addition, most studies either report on a small number of patients or combine laparoscopic salpingostomies performed using a variety of techniques and several surgeons. The strength of our study lies in its very clear definition of salpingostomy as the use of a single surgical technique performed by one surgeon and on a large number of patients. Despite the discrepancies between studies, our uncorrected IUP rate of 24.5% compares favorably with the IUP rate previously reported in the literature of 0 to 44% (20 29). Our uncorrected EP rate of 16% was slightly higher than the 0 to 14% (20 29) reported in previous laparoscopic studies, but it is well within the 0 to 21% range of EP rates reported for studies using a microsurgical technique (13 19, 34). It is anticipated that this high rate of EP may be mitigated by improved selection of cases for salpingostomy. Inoperable hydrosalpinges were excluded from our study, with potential bias. However, it is also pertinent that a surgeon may be more liberal in selecting cases for laparoscopic salpingostomy in contrast to selecting cases for microsurgical salpingostomy procedure by laparotomy. Uncorrected pregnancy rates, used as a means of describing outcome, have limited statistical validity because they do not take into account the length of follow-up. It is clear that the patients who have a shortened follow-up, because the procedure was undertaken late in the study or because the follow-up was terminated early because of an irreversible change in fertility, should not be compared directly with patients who have been followed since the beginning of the study. The best estimates of pregnancy rates are the cure rate TABLE 4 Clinical prediction (%) of EP. Historical factors Previous EP No previous EP IUD No IUD IUD No IUD Surgical factors PID No PID PID No PID PID No PID PID No PID Preoperative Postoperative Bilateral surgery FHC FHC FHC FHC FHC FHC FHC FHC No FHC No FHC No FHC No FHC No FHC No FHC No FHC No FHC Unilateral surgery FHC FHC FHC FHC FHC FHC FHC FHC No FHC No FHC No FHC No FHC No FHC No FHC No FHC No FHC Note: FHC Fitz-Hugh Curtis syndrome (perihepatic adhesions). 598 Taylor et al. Pregnancy outcome after laparoscopic salpingostomy Vol. 75, No. 3, March 2001

6 and monthly probability of pregnancy obtained using lifetable analysis and the monthly fecundity rate as described by Saleh et al. (35). We are unable to draw comparison with other studies because these parameters are not reported. Analysis of Independent Predictor Variables Age is recognized to have a negative effect on the fertility potential of women. In this study, the increase in pregnancy rate noted in younger patients was due to an increase in the number of EPs, with no change in the number of IUPs. In our study, we noted a much higher EP rate after a bilateral procedure, as compared with after a unilateral one. This has not been reported before. This may reflect an overall increase in fertilization potential conferred by bilateral surgery, albeit through damaged tubes. It has been repeatedly shown that the prognosis for fertility after laparotomy, with or without microsurgical technique, is governed by the type and extent of pelvic adhesions (13 20). Recent studies of laparoscopic salpingostomy, however, have not been able to similarly establish a role for pelvic adhesions in predicting pregnancy outcome (20, 21, 24, 36). In our study, adhesions had re-formed in only 19% of patients, and dense adhesions were significantly more likely to re-form. This might explain why the EP rate was significantly lower and the IUP rate was significantly higher (by multivariate analysis) in patients with moderately dense/ vascular adhesions as compared with dense adhesions. The factor universally recognized as being of prognostic value is the condition of the tube. Mage et al. (16) classified tubal lesions based on three factors: appearance of mucosal folds at hysterosalpingography, degree of distal tubal occlusion, and appearance of the tubal wall at laparoscopy. Both Canis et al. (20) and Dubuisson et al. (21) have shown that when tubes were assessed according to this scoring system, IUP rates were significantly better for patients with mildly to moderately damaged tubes as compared with those with severely damaged tubes. Dubuisson et al. (21, 22) also demonstrated that tubal mucosa alone performs just as well as a distal tubal score when establishing the prognosis, suggesting that a simplified assessment of tubal damage based on a mucosal score alone may be sufficient. Our results suggest that a thick tubal wall is incompatible with conception, which is consistent with the Mage scoring system. Unlike the previous studies, we did not demonstrate a significant correlation between mucosal appearance at laparoscopy and the IUP rate. Subsequent analysis reveals that only 6 of 139 patients were classified, according to the AFS scoring system, as having mildly damaged mucosa ( 75% preserved mucosa). Despite the 50% IUP rate in this category, it represents only three of the six patients. The small numbers limit the power of the chi-square test to detect significant differences, and therefore it is difficult to draw conclusions from our data. The mucosal score is important in predicting the probability of EP. Our results confirm the hypothesis put forth by Lavy et al. (2): patients with either mild or severe disease have lower EP rates than do those with moderate disease. The most likely explanation is that severe tubal dysfunction interferes with both IUP and EP because it does not allow ovum pickup to occur, whereas moderate dysfunction allows for ovum pickup, but the abnormal tube is still unable to transport the embryo into the uterus, resulting in an EP. Relationship Between Multiple Predictor Variables and Outcome Logistic Regression The clinical utility of drawing conclusions based on analyses that consider variables as independent predictors of outcome is questionable. Patients cannot be classified into a prognostic category based on a single variable because in practice, patients represent combinations of variables that may have conflicting prognostic significance. In spite of this, only a minority of studies have used multivariate analysis to predict pregnancy outcome after salpingostomy (13, 34). The clinical application of the logistic regression models will depend upon the prevalence of the determining factors. For example, within our study population, the fertility benefit of a previous gonococcal infection applies to only 17 of 139 women, whereas the other factors are more evenly distributed and thus apply more ubiquitously. Further, there will likely be different prevalences of factors between patient populations that will determine the clinical efficacy of this model. By application of the categories in Tables 3 and 4, individual women are afforded specific prognoses for benefit of IUP and risk of EP at the preoperative stage and, with better definition, postoperatively. A decision can then be made as to whether salpingostomy or IVF is the better first option. Salpingectomy or occlusion of the proximal tube may be deemed necessary for low IUP potential and/or high ectopic-potential patients who wish to undergo IVF. To this end, the chance of IUP can be estimated from 5% to 74%, and the chance of EP, from 5% to 95% when the historical and operative information is known. It should be noted that for certain combinations of variables, the actual numbers of cases in the sample with that combination may be small. Precision of the estimates will vary accordingly. We have modified our selection of patients for laparoscopic salpingostomy according to the likelihood of subsequent IUP and EP. In cases where there is a history of previous conception and/or gonococcal infection (factors that predict above-average IUP rates), salpingostomy is encouraged. In cases with a history of EP and IUD usage and with no history of PID (factors that predict below-average ectopic pregnancy rates), salpingostomy is encouraged. Further, if, at laparoscopy, perihepatic adhesions are discovered and the anatomic findings require bilateral versus unilateral surgery (factors that further augment the likelihood of EP), salpingectomy rather than salpingostomy is performed. FERTILITY & STERILITY 599

7 The role of laparoscopic salpingostomy in the treatment of complete distal fallopian tube occlusion will continue to change as IVF technology continues to improve. To better define this role, we have described a reliable and clinically applicable model that allows a physician to predict the probability of a given patient achieving an IUP and/or an EP. References 1. Gocial B. Primary therapy for tubal disease: surgery versus IVF. Int J Fertil Menopausal Stud 1995;40: Lavy G, Diamond M, DeCherney A. Ectopic pregnancy: its relationship to tubal reconstructive surgery. Fertil Steril 1987;47: Broadbent M, Magos A. Management of tubal infertility in the 1990 s. Br J Obstet Gynaecol 1991;98: DeCherney A. In vitro fertilization-embryo transfer (IVF-ET) versus tubal surgery. J Assist Reprod Genet 1992;9: Winston R. Tubal surgery or in vitro fertilization (IVF). J Assist Reprod Genet 1992;9: Zeyneloglu H, Arici A, Olive D. Adverse effects of hydrosalpinx on pregnancy rates after in vitro fertilization embryo transfer. Fertil Steril 1998;70: Murray D, Sagoskin A, Widra E, Levy M. The adverse effect of hydrosalpinges on in vitro fertilization pregnancy rates and the benefit of surgical correction. Fertil Steril 1998;69: Jones G, Trounson A, Lolatgis N, Wood C. Factors affecting the success of human blastocyst development and pregnancy following in vitro fertilization and embryo transfer. Fertil Steril 1998;70: Sharara F, Scott R, Marut E, Queenan J. In vitro fertilization outcome in women with hydrosalpinx. Hum Reprod 1996;11: Benadiva C, Kligman I, Davis O, Rosenwaks Z. In vitro fertilization versus tubal surgery: is pelvic reconstructive surgery obsolete? Fertil Steril 1995;64: Gomel V, Taylor P. In vitro fertilization versus reconstructive tubal surgery. J Assist Reprod Genet 1992;9: Marana R, Quagliarello J. Distal tubal occlusion: microsurgery versus in vitro fertilization. Int J Fertil 1988;33: Boer-Meisel M, te Velde E, Habbema J, Kardaun J. Predicting the pregnancy outcome in patients treated for hydrosalpinx: a prospective study. Fertil Steril 1986;45: Henry-Suchet J, Veluyre M, Pia P. Etude statistique des facteurs influencant le pronostic des plasties tubaires. J Gynecol Obstet Biol Reprod 1989;18: Hulka J. Adnexal adhesions: a prognostic staging and classification system based on a five year survey of fertility surgery results at Chapel Hill, North Carolina. Am J Obstet Gynecol 1982;144: Mage G, Pouly J, Bouquet de Joliniere J, Chabrand S, Riouallon A, Bruhat M. A preoperative classification to predict the intrauterine and ectopic pregnancy rates after distal tubal microsurgery. Fertil Steril 1986;46: Rock J, Katayama K, Martin E, Woodruff J, Jones H. Factors influencing the success of salpingostomy techniques for distal fimbrial obstruction. Obstet Gynecol 1978;52: Schlaff W, Hassiakos D, Damewood M, Rock J. Neosalpingostomy for distal tubal obstruction: prognostic factors and impact of surgical technique. Fertil Steril 1990;54: Vasquez G, Boeckx W, Brosens I. Prospective study of tubal mucosal lesions and fertility in hydrosalpinges. Hum Reprod 1995;10: Canis M, Mage G, Pouly J, Manhes H, Wattiez A, Bruhat M. Laparoscopic distal tuboplasty: report of 87 cases and a 4-year experience. Fertil Steril 1991;56: Dubuisson J, Chapron C, Morice P, Aubriot F, Foulot H, Bouquet de Joliniere J. Laparoscopic salpingostomy: fertility results according to the tubal mucosal appearance. Hum Reprod 1994;9: Dubuisson J, Morice P, Chapron C, Aubriot F, Foulot H, Bouquet de Joliniere J. Salpingostomies percoelioscopiques: analyse des resultats. Contracept Fertil Sex 1995;23: Dlugi A, Reddy S, Saleh W, Mersol-Barg M, Jacobsen G. Pregnancy rates after operative endoscopic treatment of total (neosalpingostomy) or near total (salpingostomy) distal tubal occlusion. Fertil Steril 1994; 62: Eyraud B, Erny R, Vergnet F. Chirurgie tubaire distale par coelioscopie. J Gynecol Obstet Biol Reprod 1993;22: Fayez J. An assessment of the role of operative laparoscopy in tuboplasty. Fertil Steril 1983;39: Gomel V. Laparoscopic tubal surgery in infertility. Obstet Gynecol 1975;46: Mettler L, Giesel H, Semm K. Treatment of female infertility due to tubal obstruction by operative laparoscopy. Fertil Steril 1979;32: Lavergne N, Krimly A, Roge P, Erny R. Resultats et indications de la coeliochirurgie tubaire distale. Contracept Fertil Sex 1996;24: Reich H. Laparoscopic treatment of extensive pelvic adhesions, including hydrosalpinx. J Reprod Med 1987;32: Gomel V. Classification of operations for tubal and peritoneal factors causing infertility. Clin Obstet Gynecol 1980;23: McComb P, Athanasios P. The intussusception salpingostomy technique for the therapy of distal oviductal occlusion at laparoscopy. Obstet Gynecol 1991;73: The American Fertility Society. The American Fertility Society classification of adnexal adhesions, distal tubal occlusion, tubal occlusion secondary to tubal ligation, tubal pregnancies, Mullerian anomalies and intrauterine adhesions. Fertil Steril 1988;49: Guzick DS, Rock JA. Estimation of a model of cumulative pregnancy following infertility therapy. Am J Obstet Gynecol 1981;140: Williams KM, Giffin W. Distal tuboplasty: is it appropriate? South Med J 1988;81: Saleh W, Dlugi A. Pregnancy outcome after laparoscopic fimbrioplasty in nonocclusive distal tubal disease. Fertil Steril 1997;67: Marana R, Rizzi M, Muzzi L, Catalano G, Caruana P, Mancuso S. Correlation between the American Fertility Society classification of adnexal adhesions and distal tubal occlusion, salpingoscopy, and reproductive outcome in tubal surgery. Fertil Steril 1995;64: Taylor et al. Pregnancy outcome after laparoscopic salpingostomy Vol. 75, No. 3, March 2001