1 Department of Obstetrics and Gynecology, University of Heidelberg and 2 Department of Obstetrics and Gynecology, University of Tübingen, Germany
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: falloposcopy/hysteroscopy/infertility/reproductive medicine/tubal cannulation
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Long before cannulation had thus gained its place in clinical infertility management, Mohri et al. was already undertaking trials to advance the blind technique into a method for visual endotubal access (Mohri et al., 1970). Along with the progress of cannulation, this idea was realized in 1990 with Kerin's description of falloposcopy as a microendoscopic technique for visual exploration of the human Fallopian tube from the uterotubal ostium to the fimbria using a transvaginal approach (Kerin et al., 1990
). Technical development culminated in two commercially available systems for falloposcopic access, linear everting balloon and co-axial catheter.
Despite an increasing number of reports giving evidence to the potential clinical value of the method (Kerin et al., 1992; Lower et al., 1992
; Risquez et al., 1992
; Dunphy et al., 1995; Porcu et al., 1997
; Dechaud, 1998) it has not yet gained widespread importance. Technical problems have limited the usefulness to an extent that the method does not yet qualify for routine clinical practice (Lundberg et al., 1998
).
With the hypothesis that technical reproducibility may be crucial, the present study investigates success rates and influencing factors, analysing the data of a large international multicentre trial.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Inclusion criteria consisted a history of primary or secondary sterility at an age between 18 and 40 years, and a given indication for laparoscopy based on the suspicion of tubal disorder or unexplained infertility in the absence of therapy refractory andrological, endocrinological or immunological reproductive failures.
A total of 367 patients with 639 tubes at a median age of 32 years (range 1840) after a 3 year median duration of infertility (range 119) fulfilled these criteria. As cannulation was intentionally not attempted in seven (contralateral) and visualization not attempted in 24 tubes (nine contralateral, three single present or only unilaterally successfully cannulated tubes, six times bilaterally), the final number of tubes intended for falloposcopy amounted to 608 and the corresponding number of patients to 358.
Operative procedure
Tubal cannulation was performed under laparoscopic control using hysteroscopic guidance. First, the uterotubal ostium was identified with a hysteroscope, which was then held in place by an articulating arm fixed to the operation table. A guiding catheter, introduced via the working channel, gave support to the actual tubal catheter between shaft tip and ostium in order to prevent intracavitary curling or kinking. Cannulation was then performed by advancing the Fallopian tube catheter loaded with its guide-wire towards the fimbria. Once the catheterization was completed or stopped in front of an obstruction, the guide-wire was replaced by the fibreoptic (diameter: 0.5 mm, resolution: 3000 pixel). By withdrawing the unit of catheter and falloposcope under continous irrigationdistension, the tubal interior was visualized in a retrograde manner. The falloposcopy system used in this study was supplied by Conceptus, Inc. (San Carlos, CA, USA). Falloposcopy was classified as successful when achieving clear endoscopic pictures.
Statistical analysis
In order to determine the method's overall technical performance, success and failure rates were calculated descriptively, both with reference to the number of examined tubes and patients.
For identification of influencing factors, the results were then analysed stepwise for the procedure components: hysteroscopy, cannulation and visualization; and failures were thus categorized. The role of tubal anatomy in terms of segmental particularities and that of the presence of pathology (defined by laparoscopy as gold standard) was evaluated using 2-test for the comparison of the parameters of different binomial distributions. The surgeon's influence was analysed in two ways. Inter-centre variability was evaluated using a
2-assimilation test on equal distribution with testing on contribution to possible heterogeneity according to Thompson (Thompson, 1993
). Learning curve effects were evaluated by stratified and individual logistic regression analyses. Significance level was defined at 0.05 for all biometrical tests.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Catheter navigation of the entire length of the tube or to the point of an obstruction was successful in 530/593 attempts, for a cannulation success rate of 89.4%. No attempts were made in seven tubes. Failures occurred in 63 tubes (10.6%). They were related on 23 occasions to kinked catheters or bent guidewires. Pinhole tubal perforation or tubal wall dissection occurred in 22 attempts, for a complication rate of 3.7%. Excessive tortuosity could not be negotiated in 11 tubes. Spasms occurred in six cases. On one occasion catheterization was stopped to prevent overdistension of a sactosalpinx.
Clear images from all cannulated segments of the tube were obtained in 423 of 506 attempts, for a visualization success rate of 83.6%. No further attempt was made in 20 tubes presenting intramural blocks at the level of the ostium during catheterization and in four tubes for unspecified reasons. Failures occurred in 83 tubes (16.4%). They were attributed 13 times to damaged falloposcopes, 20 times to catheter kinks making it impossible to advance the falloposcope and on one occasion not specified. Light reflexions by contact of the optic with the tubal wall (`white-out') disturbed the image in 37 attempts. No details were reported in 12 cases.
Outcome in relation to segmental tubal anatomy
Cannulation success varied from 568/586 attempts (96.9%) in the intramural, 481/502 (95.8%) in the isthmic and 398/417 (95.4%) in the ampullary segment to 344/347 (99.1%) in the fimbrial end. A significant difference was only found for fimbrial segment (P < 0.05) showing the best result. Segments of seven tubes (five success and two failure cases) could not be considered because they were not specified by the participating centres. The numbers of attempts decreased over the segments because of tubal blocks.
Visualization success varied from 423/474 attempts (89.2%) in the intramural, 402/426 (94.4%) in the isthmic and 339/363 (93.4%) in the ampullary to 236/279 (84.6%) in the fimbrial segment. There was a significantly increased failure rate in the fimbrial segment (P < 0.001). The calculation did not consider 32 tubes in which segmental analysis cannot apply for complete visualization failures. Numbers decreased over the segments consecutive to the catheterization results.
Outcome in the presence of laparoscopically defined tubal pathologies
Laparoscopic findings were available from 587 tubes intended for falloposcopic evaluation after successful hysteroscopic ostium access. Of these, 250 were laparoscopically normal, 256 were classified pathological, and 81 were diagnosed with peritubal pathology, such as endometriosis or adhesions. Neither catheterization nor visualization success rates were significantly affected by the presence of laparoscopically defined tubal pathology. Details are given in Table II
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The final success rate of 70% in our series, the largest study published so far, demonstrates once again the feasibility of the method (Dechaud and Hedon, 1996). Whereas this value, however, refers to examined tubes, it appears clinically more relevant to consider success with reference to patients, and 80% of the women may have profited from visualization of both or at least one tube, whereas in <20% the method failed completely. A complete bilateral evaluation (or in respective cases of the single present tube) was achieved in 57% only. If the definition of success for falloposcopy is a complete assessment of the patient's endosalpinx, this rate can hardly endorse the method for routine clinical practice.
In previous smaller series 8090% feasibility rates were reported for particular techniques: original coaxial (Kerin et al., 1992), linear everting balloon (Bauer et al., 1992
; Pearlstone et al., 1992
; Scudamore et al., 1992
) and also the improved co-axial catheter system (Surrey et al., 1997
) used in the present trial. But one early study also described a low 30% rate of clear images (Pennehouat et al., 1993
) and a more recent two-centre study blamed technical insufficiencies for poor visualization rates of 60% overall and 23% bilaterally, and classified no images of sufficient quality for clinical use (Lundberg et al., 1998
). The above-outlined strong clinical interest in falloposcopy, as a potential tool to obtain valuable information in infertility, justified a detailed analysis of our data with respect to those discrepancies.
The coaxial cannulation technique applied here incorporated hysteroscopic tubal ostium location. While the measured success of 94% can compare to that of other access techniques (Lower et al., 1992; Pennehouat et al., 1993
; Scudamore et al., 1992
), failures could further be reduced in the future by excluding intracavitary pathology before scheduling the patient for falloposcopy and correct timing of the intervention in the follicular phase when thin endometrium will not disturb vision.
With respect to cannulation, principal historical concerns refering to the narrow tubal anatomy were overcome. Cannula access succeeded in almost 90% of cases. Only the origin of spasms in little more than 1% of all cannulations can be considered as potentially provoked by the manipulation. True complications in terms of interstitial or complete wall perforations occurred in 3.7%, generally predisposed by obstructive lesions, as already reported earlier (Kerin et al., 1992). Little is known on potential long-term sequelae of falloposcopic false route formation; in our study no ectopic but one spontaneous intrauterine pregnancy, in the respective case despite interstitial perforation of the single present tube with significant intraoperative oedema formation by the distension fluid, could be documented, confirming a comparable previous observation (Wenzl et al., 1997
).
While small dimensions thus appear to some extent mastered by modern technology, other direct or indirect device-related problems occurred in 5.7% of all cannulations and as many as 14% of all visualization attempts.
Responsible for more than half of the cannulation failures, insufficient flexibility prevented the negotiation of excessively tortuous tubes in almost 2% of all attempts, and catheters kinked or bent guidewires in a little less than 4%. Catheter kinks were additionally responsible for 25% of all visualization failures, when despite successful cannulation the falloposcopy optic could not be advanced over the site of the kink in 4% of the attempts. As a consequence, material research bringing together the contradictory aspects of sliding capacity and elasticity seems important for future improvements within the coaxial falloposcopy concept.
Inherent to the thin fragile materials of the falloposcopy devices are also handling difficulties resulting in fibreoptic damage during application but also while storing, sterilizing or preparing in the environment of the operating room, leading to visualization failures of 2.6% in our trial.
The predominant cause for visualization problems, however, were light reflexions, so-called `white-out'. This occurred to a relevant extent in >7% after successful cannulation, equivalent to 45% of all visualization failures. While inadvertent contact between the fibreoptic tip and the surrounding epithelium produced the phenomenon in healthy tubes, the presence of peritubal adhesions or, above all, proximal obstruction appeared to exascerbate the problem in pathological conditions.
Although statistically not significant, the respective visualization rates were notably lower at 77 and 74%. Although adhesions were previously described to interfere mainly with cannula access by unphysiological tubal tortuosities (Flood and Grow, 1993), our data show a predominant effect on visualization subsequent to cannulation. The methodological relevance for proximal pathology, one of the main indications for falloposcopy (Grow et al., 1993
; Dunphy, 1994
; Scudamore et al., 1994
; Sueoka et al., 1998
; Schill et al., 1999
), becomes especially apparent when taking into account that 20 out of 24 dropped visualization attempts occurred in tubes suspected to be intramurally obstructed. It was anticipated that no additional information would be obtained further to cannulation.
Advancements in fibreoptic or catheter configuration should therefore follow the improvements of cannulation characteristics in order to provide constant distance from the fibre tip to the tubal wall, in fixed curves of the ampulla as well as in the narrow, hardly distendable proximal space.
When further looking at visualization results in detail, significant deficiencies of the fimbrial segment reveal that the actual outcome of falloposcopy was not in keeping with the originally formulated concept of a method `for the evaluation from the uterotubal ostium to the fimbria' (Kerin et al., 1990). In contrast to a sactosalpinx, the funnel-shape of the latter under physiological conditions does not allow sufficient fluid distension for reproducible imaging.
Laparoscopic assistance by gently occluding the fimbria with a forceps may render the most distal ampulla accessible for visualization, but a thorough evaluation when clinically engaged requires a combined fimbrioscopic (Nezhat et al., 1990) approach, questioning the frequently proposed concept of in-office falloposcopy (Bauer et al., 1992
; Scudamore et al., 1992
; Dunphy, 1994
).
The previously emphasized significance of experience and continued training for falloposcopy success (Kerin et al., 1992; Lower et al., 1992
; Scudamore, 1992, Dunphy, 1994
) finds confirmation in the present trial regarding the heterogeneous results in the different contributing centres ranging from 25 to 100% and the observance of individual positive and negative learning curves. While the final bias in our results was low, most likely, simplification would positively influence the procedure performance rates. In this regard, interpretation of our data may call into question the current stepwise concept, in which small surgical problems contribute to the overall burden.
In conclusion, our data confirm the feasibility but highlight the technical limitations currently qualifying falloposcopy, and suggest its use for selected indications rather than for routine clinical practice. Future device modifications should address the identified sources of failure, above all questions of material and the problem of optic tip control to avoid `white-out'. Consideration should also be given to reducing the method's complexity. Assessment of the fimbrial end may be better approached by laparoscopic techniques.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Notes |
---|
3 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Voßstrasse 9, D-69115 Heidelberg, Germany. E-mail: stefan_rimbach{at}med.uni-heidelberg.de
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Dechaud, H. and Hedon, B. (1996) Fallopian tube endoscopy in 1996. Contracept Fertil Sex, 24, 543548.[ISI][Medline]
Dechaud, H., Daures, J.P. and Hedon, B. (1998) Prospective evaluation of falloposcopy. Hum Reprod, 13, 18151818.[Abstract]
DeCherney, A.H. (1987) Anything you can do I can do better ... or differently! Fertil Steril, 48, 374376.[ISI][Medline]
Dunphy, B.C. (1994) Office falloposcopic assessment in proximal tubal occlusive disease. Fertil Steril, 61, 168170.[ISI][Medline]
Dunphy, B.C. and Greene, C.A. (1995) Falloposcopic cannulation, oviductal appearances and prediction of treatment independent intrauterine pregnancy. Hum Reprod, 10, 33133316.[Abstract]
Dunphy, B.C., Taenzer, P., Bultz, B. et al. (1994) A comparison of pain experienced during hysterosalpingography and in-office falloposcopy. Fertil Steril, 62, 6770.[ISI][Medline]
Flood, J.T. and Grow, D.R. (1993) Transcervical tubal cannulation: a review. Obstet Gynecol Surv, 48, 768776.[Medline]
Grow, D.R., Coddington, C.C. and Flood, J.T. (1993) Proximal tubal occlusion by hysterosalpingogram: a role for falloposcopy. Fertil Steril, 60, 170174.[ISI][Medline]
Kerin, J.F., Daykhovsky, L., Segalowitz, J. et al. (1990) Falloposcopy: a microendoscopic technique for visual exploration of the human fallopian tube from the uterotubal ostium to the fimbria using a transvaginal approach. Fertil Steril, 54, 390400.[ISI][Medline]
Kerin, J.F., Williams, D., San Roman, G. et al. (1992) Falloposcopic classification and treatment of fallopian tube lumen diseae. Fertil Steril, 57, 731741.[ISI][Medline]
Lower, A.M., Maguiness, S., Djahanbakhch, O. and Grudzinskas, J.G. (1992) Transcervical tubal endoscopy (falloposcopy) | a clinically useful tool? Gynaecol Endosc, 1, 155158.
Lundberg, S., Rasmussen, C., Berg, A.A. and Lindblom, B. (1998) Falloposcopy in conjunction with laparoscopy: possibilities and limitations. Hum Reprod, 13, 14901492.[Abstract]
Mohri, T., Mohri, C. and Yamadori, F. (1970) Tubaloscope: flexible glassfiber endoscope for intratubal observation. Endoscopy, 4, 226230.
Nezhat, F., Winer, W.K. and Nezhat, C. (1990) Fimbrioscopy and salpingoscopy in patients with minimal to moderate pelvic endometriosis. Obstet Gynecol, 75, 1517.[Abstract]
Pearlstone, A.C., Surrey, E.S. and Kerin, J.F. (1992) The linear everting catheter: a nonhysteroscopic, transvaginal technique for access and microendoscopy of the fallopian tube. Fertil Steril, 58, 854857.[ISI][Medline]
Pennehouat, G., Risquez, F., Naouri, M. et al. (1993) Transcervical falloposcopy: preliminary experience. Hum Reprod, 8, 445449.[Abstract]
Porcu, E., Dal Prato, L., Seracchioli, R. et al. (1997) Births after transcervical gamete intrafallopian transfer with a falloposcopic delivery system. Fertil Steril, 67, 11751177.[ISI][Medline]
Risquez, F. and Confino, E. (1993) Transcervical tubal cannulation, past, present, and future. Fertil Steril, 60, 211226.[ISI][Medline]
Risquez, F., Pennehouat, G., Foulot, H. et al. (1992) Transcervical tubal cannulation and falloposcopy for the management of tubal pregnancy. Hum Reprod, 7, 274275.[Abstract]
Schill, T., Bauer, O., Felberbaum, R. et al. (1999) Transcervical falloposcopic dilatation of proximal tubal occlusion. Is there an indication? Hum Reprod, 14 (Suppl. 1), 137144.[ISI][Medline]
Scudamore, I.W., Dunphy, B.C. and Cooke, I.D. (1992) Outpatient falloposcopy. Br J Obstet Gynaecol, 99, 829835.[ISI][Medline]
Scudamore, I.W., Dunphy, B.C. and Cooke, I.D. (1994) Falloposcopic comparison of unilateral and bilateral proximal tubal occlusive disease. Hum Reprod, 9, 340342.[Abstract]
Smith, T.W. (1849) New method of treating sterility, by the removal of obstructions of the fallopian tubes. Lancet, i, 529531.
Sueoka, K., Asada, H., Tsuchiya, S. et al. (1998) Falloposcopic tuboplasty for bilateral tubal occlusion. A novel infertility treatment as an alternative for in-vitro fertilization? Hum Reprod, 13, 7174.[Abstract]
Surrey, E.S., Adamson, G.D., Nagel, T.C. et al. (1997) Multicenter feasibility study of a new coaxial falloposcopy system. J Am Assoc Gynecol Laparosc, 4, 473478.[ISI][Medline]
Sweeney, W.J. (1962) The interstitial portion of the uterine tube | its gross anatomy, course and length. Obstet Gynecol, 19, 38.[ISI]
Thompson, S.G. (1993) Controversies in meta-analysis: the case of the trials of serum cholesterol reduction. Statist Meth Med Res, 2, 173192.
Wenzl, R., Lehner, R., Kiss, H. et al. (1997) Second look laparoscopy after fallopian tube perforation with the tubal endoscope. Gynäkol Geburtshilf. Rundsch, 37, 3940.
Submitted on July 6, 2000; accepted on January 31, 2001.