输卵管不育腹腔镜外科因素

Factors determining outcomes of the laparoscopic surgery of tubal infertility Anara Kudaiberdieva, Aygul Kangeldieva Kyrgyz Scientific Center of Human Reproduction, Bishkek, Kyrgyzstan Summary Background: We aimed to elucidate the factors affecting the occurrence of spontaneous pregnancy after laparo- scopic (LS) surgery in patients with tubal infertility. Material/Methods: Fifty patients with tubal infertility undergoing tubal LS surgery (adhesiolysis and salpingostomy) entered the study. The patients were evaluated by clinical testing, hysterosalpingography (HSG) before and after LS and diagnostic LS. Patients were followed-up for 1 year. Logistic regression analysis (LRA) was performed to determine the factors influencing reproductive outcome. Results: Fourteen patients (28%) became pregnant after LS surgery. LRA demonstrated that the best predictive model included a combination of LS chromopertubation score (OR – 8.76, p<0.003) and HSG tubal patency score with the site of occlusion (OR – 6.78, p< 0.009). Conclusion: Thus, the level and severity of tubal occlusion by HSG and the basal LS chromopertubation result influence the reproductive outcome in patients with tubal infertility undergoing tubal LS surgery. Key words: tubal infertility • laparoscopy • surgery • prognosis Full-text PDF: Full-text PDF-www.jgi-online.org/get_pdf.php?IDMAN=7434 Word count: 3915 Tables: 5 Figures: - References: 31 Author’s address: Anara Kudaiberdieva, Sevastopolskaya 20/2, 720040, Bishkek, Kyrgyzstan, E-mail: gkudaiberdieva@gmail.com Received: 2005.05.04 Accepted: 2005.10.10 Published: 2005.12.30 129 Original Articles: Basic and Clinical Research WWW.JGI-ONLINE.ORG© Pol J Gyn Invest., 2005; 8(4): 129-134 BACKGROUND Tubal adhesive disease is an important cause of infertility, being responsible for about 30% of all causes of infertili- ty [1, 2]. It is well known that inflammatory processes in the pelvis and infectious diseases are associated with the development of adhesions leading to tubal and peritoneal forms of infertility [3, 4]. Chlamydia antibodies testing along with hysterosalpingography (HSG) and laparosco- py (LS) are used now as the diagnostic tests for the diag- nosis of tubal infertility (TIF) [5, 6, 7]. Both HSG and LS have been found to be quite informative for this purpose [7]. The laparoscopy technique not only allows treatment of tubal abnormalities [4, 8, 9, 10, 11], but also selects can- didates for in-vitro-fertilization [12]. However, the pregnancy rates after tubal surgery and in- vitro-fertilization vary around 15%-35% and 25%, respec- tively [1, 8]. Although several factors such as age, duration of infertility, presence of endometriosis, ovulatory factors, HSG and LS data have been reported to have prognostic value [13-19], factors determining development of natu- ral pregnancy after correction of tubal pathology need to be addressed. We aimed to elucidate the factors affecting the develop- ment of natural pregnancy after laparoscopic surgery in patients with tubal infertility. MATERIAL AND METHODS Fifty women aged between 20 and 37 years (mean age 29.16+ 4.74 years) with the duration of infertility ranging from 2 to 15 years (mean duration 6.1±3.6 years) gave informed con- sent and entered the study. The diagnosis of TIF was con- firmed by HSG in all patients. Primary infertility was di- agnosed in 18 (36%) patients and secondary in 32 (64%) patients. Three patients had a history of ovarian surgery due to cysts and nonmalignant tumors, 2 - ectopic pregnancy and 9 - abdominal operations (appendectomy – 7 patients and peritonitis- 2 patients). Sixteen patients were positive for Chlamydia antibodies (fluorescent antibody testing), anovu- lation (based on daily rectal temperature analysis, hormone blood analyses – thyroid hormones, prolactin, testosterone, follicle-stimulating hormone and luteinizing hormone; and diagnostic tests explained below) was found in 11 (22%) pa- tients and cervical factor (colposcopy) - 5 patients. All patients underwent clinical examinations; blood test- ing, endocervical culture analyses for urogenital (includ- ing direct fluorescent testing and luminescent microscopy for Chlamydia trachomatis and mycoplasmosis) and bac- teriological infections. It should be noted that all women with positive culture analysis for Chlamydia infection prior to laparoscopic surgery received antibiotic treatment with Doxycycline (10-day courses 300 mg daily) until elimina- tion of the infection. Transvaginal ultrasound investigation was performed us- ing 5 MHz transducer and “Aloka SSD-500” machine (Mure, Japan). Hysterosalpingography of the Fallopian tubes was per- formed on the 7th – 9th days of the menstrual cycle in the early follicular phase to eliminate functional obstruction components using a “Shimadzu“ (Kyoto, Japan) roent- genographic machine and water-soluble contrast before the tubal surgery (basal HSG) and after tubal interven- tion (control HSG). We assessed the length of the tube, patency, contrast spillover, occlusion level and the pres- ence of adhesion cavities filled with contrast in the ab- dominal cavity. To define the prognostic significance of HSG parameters, we developed two scoring systems with and without inclusion of the occlusion level. HSG scor- ing without accounting for the level of the occlusion was encoded as: 1 – patent tube; 2- patent with peritubal ad- hesions; 3 – occlusion or hydrosalpinx. HSG scoring sys- tem which included the level of the occlusion was as fol- lowing: 1- patent; 2 – patent with peritubal adhesions; 3 – occlusion at isthmic level; 4 – occlusion at interstitial level; 5 - occlusion at ampullary level; 6 – occlusion at the fimbrial part; 7 – hydrosalpinx. HSG patency scores were estimated from the arithmetic sum of the score val- ues for both tubes. Additionally, we estimated the length of tubes before and after tubal surgery and the degree of absolute increase in tubal length after surgery. HSG pat- ency after surgery was also assessed as the presence or ab- sence of patency recovery. Laparoscopy was performed in all patients using endo- scopic equipment (“Karl Storz”; Tuttlingen, Germany) according to widely used technique [20]. The following parameters were assessed: tubal appearance, tubal color, presence of peritubal adhesions, presence of Morgagni hydatids presence of endometriosis and the result of dye chromopertubation. We also evaluated the extent of ad- hesions and the number of organs involved in the ad- hesive process including: anterior pouch and pouch of Douglas, uterosacral ligaments, both tubes, both ovaries, uterus, mesentery, small and large bowel, bladder and omentum. We considered the effect of coexisting peritu- bal and ipsilateral ovarian adhesions to be worth evalu- ating. The bilaterality of any tubal adhesive process and tubal occlusion according to LS data were included in the statistical analysis. We developed the scoring systems for evaluation of chromopertubation and the appearance and color of both tubes, to make grading easier to determine the prognostic significance of chromopertubation, and the color and ap- pearance of the tubes. Laparoscopic chromopertubation was assessed using a scoring system, which allowed the state of both tubes to be quantified and was based on the arithmetic sum of both tubal scores. Encoding was as following: 1- abundant dye spill; 2- limited (weak) dye spill; 3 – penetration without dye spill; 4 – no penetration, no spill. The prognostic scoring system of the color of the tubes was also equal to the arithmetic sum of both tubal scores and was as following: normal –1, hyperemia –2, cyano- sis (bluish) – 3. Similarly the appearance of the tubes was encoded: normal – 1, enlarged in the ampullary part – 2, enlarged throughout its whole length – 3. Patients were followed up for one year to ascertain preg- nancy. Original Articles: Basic and Clinical Research 130 © Pol J Gyn Invest., 2005; 8(4): 129-134 Statistical analysis was performed using statistical package SPSS for Windows version 10.00. (Chicago, Illinois, USA). Continuous data are represented as mean + standard devi- ation. The differences between groups of categorical var- iables were assessed using the Chi-square test and of con- tinuous variables by the unpaired Student-t test. Logistic regression analysis was used for the estimation of the factors determining the development of natural pregnancy after LS. The dependent variable was occurrence of pregnan- cy during the one year follow-up period, the independent variables were: age (<30 years and >30 years) (13), infertil- ity duration (<3 years and > 3 years) (13), presence or ab- sence of a history of pelvic inflammatory disease, positive or negative result for Chlamydia infection, presence or ab- sence of ovulation, primary or secondary infertility; HSG covariates - basal HSG tubal patency scores with and with- out the level of occlusion, recovery of tubal patency at con- trol HSG, length of tubes and absolute increase after tubal surgery at control HSG; LS covariates – number of or- gans involved in adhesive process, presence or absence of peritubal adhesions, mesenteric adhesions, bilaterality of peritubal adhesions, coexistence of peritubal and ovarian adhesions, presence or absence of endometriosis, chromo- pertubation, tubal colour and appearance scores; and type of surgical intervention. The forward stepwise regression analysis selected the most powerful predictors of positive reproductive outcome. RESULTS Overall 14 women became pregnant during one-year fol- low-up after LS surgery for TIF. All patients were divided into two groups: pregnancy-negative (PN) group - 36 women and pregnancy-positive (PP) group - 14 women. Out of the 50 women entering the study, laparoscopic ad- hesiolysis was performed in 41 patients; salpingostomy only - 1 (2%) patient, salpingostomy and adhesiolysis in 5 patients. adhesiolysis and coagulation of endometrio- sis - in 3 women. There were no differences between groups (Table 1) in age, duration of infertility, history of pelvic inflammatory dis- ease, Chamydia infection, type of infertility, and frequen- cy of anovulation. Analysis of HSG data (Table 2) revealed that PP patients had lower HSG patency scores taking the occlusion level into account (p<0.01 and p<0.01, respectively), as com- pared with PN patients. All PP patients had recovery of tubal patency as compared to PN patients at control HSG (p<0.01). The mean length of tubes increased sig- nificantly (left tube – p<0.004 and right tube – p<0.002) and their absolute change in length was greater (p<0.04) in women of the PP group as compared with patients of the PN group. Table 1. Clinical characteristics. Parameters Pregnancy+ Pregnancy - р Age, % (n) >30 years <30 years 28.6 (4) 71.4 (10) 33.3 (12) 66.7 (24) 0.746 Infertility duration, % (n) >3 years <3 years 57.1 (8) 42.9 (60) 75.0 (27) 25.0 (9) 0.216 Pelvic inflammatory disease, % (n) 64.3 (9) 61.1 (22) 0.83 Chlamydia infection, % (n) 42.9 (6) 27.8 (10) 0.305 Type of infertility, % (n) Primary Secondary 50.0 (7) 50.0(7) 30.6 (11) 69.4 (25) 0.198 Anovulation, % (n) 28.6 (4) 19.4 (7) 0.48 Table 2. Distribution of HSG (basal and control) factors. Parameters Pregnancy+ Pregnancy - р Basal HSG tubal patency score; points 4.5±1.2 5.3±0.8 0.01 Basal HSG tubal patency score with the level of occlusion; points 5.5±2.8 7.9±3.4 0.01 Recovery of tubal patency after LS correction, % (n) 100.0 (14) 69.4 (25) 0.01 Mean length of tube after LS correction, cm Right tube Left tube 10.8±0.7 10.7±0.6 8.0±4.7 8.6±4.0 0.002 0.004 Absolute increase of tubal length after LS correction; cm 7.8±4.4 5.6±5.0 0.04 LS – Laparoscopy, HSG – hysterosalpingography © Pol J Gyn Invest., 2005; 8(4): 129-134 131 Kudaiberdieva A et al – Factors determining outcomes... Analysis of LS data (Table 3) showed that the number of patients with bilateral peritubal adhesions was significantly greater in the PN group (p<0.008). Similarly, the number of women with coexisting tubal adhesions involving the ipsilateral ovary and other organs was greater in the PN group as compared with the PP group (p<0.02 for the left tube and p<0.08 for the right tube). The incidence of en- dometriosis did not differ between groups, but mesenteric adhesions were more often seen (p<0.02) in the PN group. The tubes were more often abnormal in color (bluish, cy- anotic) and appearance (enlarged) in the former group (p<0.01 for the tubal color score and p<0.02 for the tu- bal appearance score) than in the PP group. On the other hand, PP women had lower chromopertubation scores compared with those in the PN group (p<0.009). There was a slightly higher mean number of organs involved in adhesive processes in the PN group, but this was not sta- tistically significant (p<0.09). Logistic regression analysis (Table 4) demonstrated that among all clinical, HSG and LS variables entered into the analysis the highest odds ratio for occurrence of pregnan- cy was patency score at basal HSG with and without in- cluding the level of occlusion (OR - 4.73, p<0.02 and OR - 4.81, p<0.02), recovery of tubal patency at control HSG (OR - 4.74, p<0.02), presence of peritubal adhesions and their bilaterality (OR - 6.5, p<0.01 and OR - 3.9, p<0.04), followed by the coexistence of tubal and ovarian adhesions (OR 4.3, p<0.03), LS chromopertubation score (OR – 7.28, p<0.006) and tubal color and appearance scores (OR – 4.69, p<0.03 and OR – 4.3, p<0.03, respectively). Forward stepwise regression analysis (Table 5) showed that the best predictive model was that included a com- bination of the LS chromopertubation score (OR – 8.76, p<0.003) and HSG tubal patency score with the defined level of occlusion (OR – 6.78, p< 0.009). The patients with lower chromopertubation scores had an 8.76- fold great- er chance of becoming pregnant after tubal surgery and those with higher HSG scores (proximal bilateral occlu- sion) had a 6.78 greater risk of unfavorable reproductive outcome development after LS treatment of TIF. Table 3. Distribution of LS factors. Parameters Pregnancy+ Pregnancy - р Mean number of organs, involved in adhesive process 6.4±3.8 5.61±3.6 0.09 Presence of peritubal adhesions, % (n) Yes No 35.7 (5) 64.3 (9) 86.1 (31) 14.9 (5) 0.002 Bilateral peritubal adhesions, % (n) No Unilateral Bilateral 7.1 (1) 28.6 (4) 19.4 (5) 36.1 (13) 50.0 (18) 0.008 Сoexisting peritubal and ovarian adhesions, % (n) Left tube+left ovary Yes Other organs Right tube+right ovary Yes Other organs 28.6 (4) - 14.3 (2) 21.4 (3) 30.6 (11) 33.3 (12) 26.5 (9) 44.1 (15) 0.02 0.08 Mesenteric adhesions, % (n) 35.7 (5) 69.4 (25) 0.02 Endometriosis, % (n) 14.3 (2) 2.8 (1) 0.12 LS chromopertubation, points 4.2±1.8 5.9±2.0 0.01 Tube color, points 2.3±0.7 3.1±1.0 0.01 Tube appearance, points 2.8±1.2 4.0±1.4 0.009 LS – Laparoscopy Table 4. Logistic regression analysis data. Parameters Odds ratio р Presence of peritubal adhesions 6.5 0.01 Bilateral peritubal adhesions 3.9 0.04 Сoexisting peritubal and ovarian adhesions Left tube+left ovary Right tube+right ovary 4.3 2.98 0.03 0.08 LS chromopertubation score 7.28 0.006 LS tubal color score 4.69 0.03 LS tubal appearance score 4.3 0.03 Basal HSG tubal patency score without the level of occlusion 4.73 0.02 Basal HSG tubal patency score with the level of occlusion 4.81 0.02 Recovery of tubal patency at control HSG 4.74 0.02 LS – Laparoscopy, HSG - hysterosalpingography Original Articles: Basic and Clinical Research 132 © Pol J Gyn Invest., 2005; 8(4): 129-134 DISCUSSION Our study demonstrates that 28% of women with TIF laparoscopic interventions became pregnant. Logistic re- gression analysis showed that factors affecting the pregnan- cy were the patency score at basal HSG with and without the level of occlusion, recovery of tubal patency at control HSG, presence of bilateral peritubal adhesions, coexist- ing with ovarian adhesions, LS chromopertubation, tubal colour and appearance scores. Further stepwise regression analysis showed that a combination of chromopertubation score and HSG tubal patency score with the level of oc- clusion may predict occurrence of pregnancy during the follow-up period after LS correction of TIF. Previous investigations have revealed that the chance of positive outcome reduces with advancing age and in- creased duration of infertility. Hunault et al. reported that couples with less than 3 years’ duration of infertility have a 1.8-fold higher probability of conceiving than those with longer duration [13]. The probability of live birth is also 1.5-fold higher if the female partner is younger than 30 years and has had a previous pregnancy (secondary in- fertility) [13]. In our study, there were no differences be- tween groups in the number of women >30 years of age, with prolonged duration of infertility and the number of patients with primary or secondary infertility. The pres- ence of ovulation also did not differ between the groups who did or did not conceive. A history of pelvic inflammatory disease may influence reproductive outcome [5, 6, 21]. Though Hubacher et al. (22) did not find any correlation between a histo- ry of pelvic inflammatory disease and reproductive out- come, Taylor et al. [23] showed that a history of pre- vious pregnancies and absence of gonococcal infection may predict intrauterine pregnancy after LS salpingos- tomy in patients with distal occlusions. In our study, the presence of Chlamydia or other infection had no effect on the outcome. The lack of differences in the aforementioned clinical pa- rameters between groups with and without successful con- ception suggests that both groups were comparable. Similarly, in our study the presence of endometriosis and type of LS intervention did not influence the outcome. Although previous studies have shown that the presence of even mild endometriosis may reduce fertility rate [14, 15], both groups of patients were comparable in the frequency of concomitant endometriosis. However, our findings are in accordance with the study of Maruyama et al. [16], which revealed that the combination of LS tubal microsurgery and ablation of accompanying endometriosis improved fertility outcome. In our study, 3 patients underwent both adhesiolysis and endometriosis ablation. Previous studies have reported that salpingostomy and abla- tion of endometriosis may promote the prospect of pregnan- cy and even contribute to the success of in-vitro fertilization [9, 15, 24-28]; however, we found that in our group of pa- tients the type of intervention did not affect outcome. Several investigators have studied the prognostic signifi- cance of LS and HSG data on peritubal adhesions and distal tubal occlusion [17-19]. Thus, Mol et al. [17] indi- cated that the risk of fertile outcome significantly reduc- es from 0.81 to 0.30 in cases of bilateral tube occlusion as seen at HSG as compared with unilateral occlusion. The possibility of positive outcome in their study for the pres- ence of peritubal adhesions was 0.72. However, the groups of patients differed in age, duration of infertility and pres- ence of anovulation. The same investigators compared the predictive value of HSG and LS according to the results of the CITES study [19]. The authors [19] found that bi- lateral occlusion by LS had a higher predictive value than occlusion as shown by HSG. Our results of the prognostic value of HSG and LS data confirmed the findings of the previous investigations that the presence of peritubal adhesions, bilateral involvement and ovarian adhesions decrease reproductive outcome [17, 19, 29]. However, for the first time we used a scoring sys- tem of chromoper