Diagnostic and therapeutic value ofselective salpingography and tubalcatheterization in an unselected infertilepopulation

Spyros Papaioannou, M.D.,a Masood Afnan, M.D.,a Alan J. Girling, M.A.,b

Bolarinde Ola, M.D.,a Olufemi Olufowobi, M.D.,a Aravinthan Coomarasamy, M.D.,c andKaldoun Sharif, M.D.a

Birmingham Women’s Hospital, Birmingham, United Kingdom

Objective: To present diagnostic findings and fertility outcome after selective salpingography and tubalcatheterization in an unselected infertile population.

Design: Cohort study.

Setting: Tertiary reproductive medicine unit.

Patient(s): One hundred ten consecutive infertile women. No exclusion criteria were applied. Follow-upranged from 16 to 54 months.

Intervention(s): Selective salpingography and tubal catheterization under fluoroscopic guidance as theprimary test for the assessment of the fallopian tubes.

Main Outcome Measure(s): Incidence of tubal disease at selective salpingography, therapeutic effectivenessof tubal catheterization, and fertility outcome after the procedure.

Result(s): Tubal disease was present in 31.4% of the tubes examined. Of tubes proximally blocked at selectivesalpingography, 52.1% were found to be normal after tubal catheterization. Proximal tubal blockage (bilateralor unilateral) was detected in 34.8% of women. This was reduced to 5.5% after tubal catheterization.Spontaneous conceptions occurred in 21.9% of the women. In total, 36.2% conceived without IVF or ICSI.

Conclusion(s): Selective salpingography and tubal catheterization can be useful as a primary tubal assessmenttool in the investigation of infertility. In cases of proximal tubal blockage, an effective see-and-treat approachcan be adopted. More research into the possible therapeutic benefits of the procedure is justified. (Fertil Steril�2003;79:613–7. ©2003 by American Society for Reproductive Medicine.)

Key Words: Infertility, selective salpingography, survival analysis, tubal catheterization

Fallopian tube assessment, an essential partof an infertility investigation work-up, is stillsynonymous with the performance of either ahysterosalpingogram (HSG) or a laparoscopyand dye test. However, the diagnostic accuracyand fertility predictive value of HSG have beenconvincingly criticized (1–3). Even laparos-copy and dye test, considered to be the opti-mum method for assessing tubal patency (3),may not determine whether a visualized block-age is due to true anatomic occlusion or due totubal spasm (2, 4). Resistance differences be-tween the tubes and technical competence mayhave a significant bearing on the results ofeither investigation.

Transcervical selective salpingography andtubal catheterization (TC) under fluoroscopicguidance (SS-TC) is an established procedurefor the diagnosis and treatment of proximaltubal blockage (PTB). Several series of SS-TChave attested to its accuracy in the diagnosis ofPTB as well as to its effectiveness in recana-lizing the proximally blocked fallopian tubeand restoring fertility for these women (5–8).

Limited experience exists in the use ofSS-TC as a first-line tubal assessment in inves-tigation for infertility. We describe the diag-nostic findings of SS-TC and present a survivalanalysis of conceptions after the procedure, inan unselected population of women who un-

Received March 13, 2002;revised and accepted July12, 2002.Reprint requests: SpyrosPapaioannou, M.D., 142Harborne Park Road,Birmingham B17 0BS WestMidlands, United Kingdom(FAX: 0044-121-627-2701;E-mail:spyrospap@talk21.com).a Assisted Conception Unit,Birmingham Women’sHospital.b School of Mathematicsand Statistics, TheUniversity of Birmingham,Birmingham, UnitedKingdom.c Education ResourceCenter, BirminghamWomen’s Hospital.

FERTILITY AND STERILITY�VOL. 79, NO. 3, MARCH 2003Copyright ©2003 American Society for Reproductive MedicinePublished by Elsevier Science Inc.Printed on acid-free paper in U.S.A.

0015-0282/03/$30.00doi:10.1016/S0015-0282(02)04704-0

613

derwent SS-TC as part of their initial infertility investigationwork-up.

MATERIALS AND METHODS

From June 1996 to April 1999, SS-TC was attempted asa first-line tubal assessment in 110 consecutive infertilewomen. Institutional review board approval was obtained.The procedure was possible in 107 (97.2%) cases. In threecases, the procedure was not completed because of thefollowing reasons: one patient could not tolerate the specu-lum; in another the cervical os was found to be too tight forthe SS-TC catheter to be inserted; whereas a third patientcomplained of pain as the catheter was advanced through thecervix and requested that the procedure be terminated. Inaddition, one case was excluded because the woman indi-cated in her response that she stopped trying to conceiveshortly after the procedure.

Therefore, 106 women were included in the analysis.Average age (�SD) was 31.9 � 5.4 years, and medianlength of infertility at SS-TC was 2.4 years (range, 1.1 to16.8 years). Fifty-two women (49%) presented with primaryand 54 (51%) with secondary infertility. Five of 106 women(4.7%) had had a salpingectomy before. One had been diag-nosed with uterus didelphys, and another had undergonethree ovarian cyctectomies. Anovulation was diagnosed in17 (16%) of cases (information was missing in 13 cases).Male factor, using the World Health Organization definitions(9), was present in 27 (25.4%) of the couples (informationwas missing in 13 cases).

The method used has been described elsewhere (10, 11).In brief, the fallopotorque (Cook UK, Letchworth, Hertford-shire, U.K.) SS-TC catheter system was used. The fluoro-scopic unit employed was the mobile Phillips Optimus BV29with a C-arm (Phillips Corporation, Amsterdam, The Neth-erlands). The distal end of the selective salpingographycatheter was connected by polyethylene tubing to a syringepump (KMAR-400; Cook UK). Water-soluble contrast me-dium (Hexabrix 320; Mallinckrodt Medical, Round Spiney,U.K.) at a constant flow rate of 10 mL/min was used.

Selective salpingography and tubal catheterization wasperformed during the follicular phase of the cycle. Theselective salpingography catheter was forwarded through thecervix and was advanced by tactile sensation to the tubalostium. Its position was fluoroscopically checked, and ifsatisfactory, dye was injected (selective salpingography). Ifthe obstruction was overcome, the tubal contour was out-lined with contrast agent. If it persisted, a guidewire wasthreaded through the catheter inner cannula and was ad-vanced toward the obstruction (tubal catheterization). A gen-tle push was applied to overcome it. The guidewire was thenwithdrawn, and contrast medium was injected through theselective salpingography catheter to confirm patency.

Only the woman’s first conception during the study pe-riod was considered. Cumulative conception rates and con-ception hazard rates were obtained from Kaplan-Meier sur-vival analysis. Smooth estimates of these rates were obtainedfrom Weibull analysis (12). Initially, only spontaneous con-ceptions were counted as successes. Conceptions arisingafter medical intervention were considered as censored at thetime of conception. Cases for whom no conception wasreported were censored at the time of follow-up. In a sup-plementary analysis, the definition of success was extendedto include all first conceptions except those resulting fromIVF or ICSI treatments. Thus, in the supplementary analysis,conceptions resulting from ovulation induction and IUI bypartner or by donor insemination treatments were alsocounted as successes.

RESULTSFrom the 106 women included, information was collected

from 91 (response rate, 85.8%). Of them, 78 women (85.7%)returned a questionnaire, whereas 13 (14.3%) were inter-viewed by telephone. There were 18 spontaneous concep-tions among the 78 mail responders and 2 among the 13telephone responders (P�.72, 2-tailed Fisher’s exact test).

Selective salpingography (SS) was possible in 191(92.3%) of the 207 available fallopian tubes. Acute antever-sion or retroversion of the uterus or irregular uterine cavityenlargement due to fibroids were present when SS was nottechnically possible. Normal tubes were identified in 131(68.6%) cases; PTB, in 49 (25.7%) cases; middle block, in 6(3.1%) cases; a hydrosalpinx, in 2 (1%) cases; whereas therewas uncertainty about interpretation in 3 (1.6%) cases. Tubalcatheterization was attempted in 48 of the 49 proximallyblocked tubes. After TC, 25 (52.1%) of these were found tobe of normal opacification patterns. Proximal tubal blockagepersisted in nine (18.7%) cases; in five (10.4%), a hydrosal-pinx was present; in two (4.2%) cases, a middle block wasidentified; in three (6.2%) cases, a distal block was identi-fied; peritubal adhesions were present in two (4.2%) cases;and in two (4.2%) cases, TC could not be performed.

The cumulative conception rates for spontaneous only aswell as for all non–IVF-ICSI pregnancies are shown inFigure 1. The same figure also shows the probability ofconception as a function of time after SS-TC (conceptionhazard rate) for spontaneous conceptions only. Inspection ofthe smooth (Weibull) conception hazard rate shows that thisdecreases slightly only over the period after SS-TC.

Two age groups were studied: women �35 years of age(n � 64; spontaneous conceptions, 13; other non–IVF-ICSIconceptions, 10) and women �35 years of age (n � 27;spontaneous conceptions, 7; other non–IVF-ICSI concep-tion, 1). Rates were not significantly different between theage groups (P�.926 and P�.352 for spontaneous and for allnon–IVF-ICSI conceptions, respectively, log-rank test). At

614 Papaioannou et al. Salpingography in unselected infertile women Vol. 79, No. 3, March 2003

SS, eight women (7.5%) were diagnosed of having bilateraland 29 (27.3%) of having unilateral PTB. After TC, two(1.8%) women still had bilateral PTB, and four (3.7%) hadunilateral PTB. Pregnancy results by conception method,considering only the first conception per subject, are shownin Table 1.

There were five cases of tubal perforation (4.6%) forwhich no further treatment was necessary. One patient(0.9%) experienced a vasovagal reaction, which was ade-

quately managed with intravenous fluids. Intravasation wasnoted in eight (7.4%) cases, whereas one woman was admit-ted after SS-TC because of pain. She was discharged thesame day. There was one case of uterine perforation (Fig. 2).

DISCUSSIONThis study, for the first time, presents diagnostic findings

of SS-TC and survival analysis of conceptions after theprocedure in a population of unselected infertile women. Byfar, most previous SS-TC studies only included infertile

F I G U R E 1

(A), Cumulative conception rates for spontaneous (thick line)and non–IVF-ICSI (thin line) pregnancies. Solid lines showKaplan-Meier estimates; dotted lines are fitted Weibull val-ues. Time is measured from date of SS-TC. (B), Instanta-neous probability of conception for spontaneous concep-tions only expressed as an annual rate (conception hazardrate), as a function of time since SS-TC.

Papioannou. Salpingography in unselected infertile women. Fertil Steril 2003.

T A B L E 1

Pregnancy results by method of conception (only thefirst conception per subject is considered).

Method of conceptionNo. of

conceptions

% Among allwomen in study

(n � 106)

% Amongresponders(n � 91)

Spontaneous 20 18.9 21.9Artificial insemination

(partner)5 4.7 5.5

Clomiphene citrate 5 4.7 5.5Donor insemination 1 0.9 1.1IVF–ICSI 23 21.7 25.3

Note: Not conceived, 37 women; not responded, 15 women.

Papioannou. Salpingography in unselected infertile women. Fertil Steril 2003.

F I G U R E 2

A rare case of uterine perforation during selective salpingog-raphy and tubal catheterization. The guidewire can be ob-served perforating the uterine fundus. Some contrast me-dium has escaped into the peritoneal cavity.

Papioannou. Salpingography in unselected infertile women. Fertil Steril 2003.

FERTILITY & STERILITY� 615

women diagnosed with PTB. For most studies, the presenceof any additional infertility factors was an exclusion crite-rion. Only Hovsepian et al. (13) attempted to address the roleof SS-TC in an unselected infertile population. However,those investigators admit that referral bias led to most oftheir patients (n � 37) presenting with tubal disease (all hadundergone other tubal assessment tests before).

In our series, some tubal abnormality was detected inabout a third of the tubes assessed. From the tubes that wereproximally blocked at SS, about half were found to benormal after recanalization. Proximal tubal blockage (bilat-eral or unilateral) was diagnosed in about a third of ourpatients. This was reduced to about 1 in 20 after TC. Amongwomen for whom pregnancy data were available, 21.9%conceived spontaneously, whereas 36.2% in total conceivedwithout IVF-ICSI.

The diagnostic advantages of SS-TC are substantial. In aprospective randomized controlled trial, SS-TC was shownto be more accurate than laparoscopy and dye test for thediagnosis of PTB and equally accurate for the diagnosis ofdistal tubal blockage (14). Demonstration of a “cobblestone”mucosal pattern at SS-TC has been associated with irrevers-ible tubal damage and a poor prognosis and hence would bean indication for IVF, despite the fallopian tubes beingpatent (15). An unsuccessful attempt at recanalization by TCis also of diagnostic value. In these cases there is almostalways organic tubal pathology, sometimes multiple (16,17), and therefore further tubal assessment is unnecessary.With SS-TC, the opportunity exists to measure the tubalperfusion pressures (18) that when normal, mark an im-proved fertility prognosis after controlled ovarian hyper-stimulation (19).

In addition to reliable diagnosis, SS-TC offers the oppor-tunity for a see-and-treat approach to the management ofPTB. The effectiveness of this approach is demonstrated bythe significant reduction after tubal catheterization of theproportion of women with PTB in our population. Oncefallopian tube patency is reestablished, ovulation inductionand IUI by partner or donor become feasible treatmentoptions, a fact demonstrated by previous authors as well (7,20). These treatments are safer and cheaper than IVF whichwould be, in settings where SS-TC is not available, still theonly alternative for women with bilateral, at least proximal,tubal blockage.

It can be argued, however, that in some of these women,PTB was the result of tubal spasm and therefore no treatmentwas necessary, as tubal spasm is accepted as a normalphenomenon. Some evidence to the contrary does existthough. Pregnancy rates in a prospectively followed cohortof women with tubal spasm who received no infertilitytreatment were disappointing (21). Furthermore, tubal per-fusion pressures (TPPs) at selective salpingography inwomen diagnosed with tubal spasm were found, with fewexceptions, to be elevated (22). On the basis of these find-

ings, Gleicher and Karande (23) have suggested that inter-mittent tubal blockage, in the absence of technical failure(e.g., excessive injection pressures), is probably not a normalphysiological variation of uterotubal junction function.Therefore, even for women with tubal spasm, the perfor-mance of SS-TC might represent a useful intervention.

In fact, some evidence for a therapeutic benefit of SS-TCalready exists. Kamiyama et al. (24), in a retrospective casecontrol study, observed a significantly higher pregnancy rateamong infertile women with patent (on HSG) tubes whounderwent SS (under hysteroscopic guidance), in compari-son to a control group that did not undergo the procedure.Our group has shown that TPP reductions achieved by tubalcatheterization mark an improvement in fertility prognosis(25). These results would strengthen the case for furtherinvestigation of the role of SS-TC as a first-line tubal as-sessment test in the management of infertility. The preva-lence of high TPPs or even tubal spasm among infertilewomen is unknown. High TPPs, confirmed in cases of tubalspasm, might indicate tubal pathology undetectable by HSGor laparoscopy and dye test in some cases of otherwiseunexplained infertility.

Proximal tubal blockage has been reported in �20% ofHSGs performed for the investigation of infertility (26, 27).The incidence of PTB (34.8%) encountered in our study(n � 106) was substantially higher, without an obviousexplanation. One possibility might lie in the diverse ethniccomposition of our population, in contrast to the primarilyCaucasian population that other studies described. Anothermight be the fact that our center has established a reputationfor its SS-TC service, receiving referrals of women withPTB from across the United Kingdom. There is also apossibility that some women seeking infertility advice forthe first time came to us on the recommendation of womenwith PTB treated here before to whom they are related. Thediagnosis of PTB was not operator dependent.

The incidence of tubal perforation, the most commoncomplication of SS-TC, in our series was within the 0 to 10%range reported before (28, 29). In common with other stud-ies, no additional treatment was necessary for our patients (4,7, 29–31). The case of uterine perforation reported here is asfar as we know, the first reported in the literature. Noimmediate sequelae were noted for this patient either. How-ever, little is known about the possibility of long-term con-sequences for these women of tubal perforation (e.g., risk ofinfection, adhesion formation, or ectopic pregnancy), andtherefore one should not minimize the significance of suchevents.

An investment of physician and radiology suite time aswell as of money for the purchase of equipment is necessaryfor the implementation of an SS-TC service. However, asclinical teams and individual clinicians become more famil-iar with the technique, the time requirements lessen signifi-cantly (10). On the other hand, savings could be achieved by

616 Papaioannou et al. Salpingography in unselected infertile women Vol. 79, No. 3, March 2003

the reduction in the use of laparoscopy and dye test, with itsassociated risks, which Gleicher and Karande (23) reportedat 75%. The cost of SS-TC has been reported at $300 (32).The cost per live birth for different therapeutic approaches ofPTB has been estimated at $17,000 for tubal surgery,$12,000 for IVF, and $6,400 for SS-TC (33).

A possible source of bias in this study would be a higherresponse rate among women who conceived after SS-TC andtherefore the relative underrepresentation in the respondersof women who did not. However, the proportion of sponta-neous conceptions among women who returned a question-naire was not significantly higher than that of women whowere interviewed by telephone. Therefore our results areunlikely to be biased by women with negative outcomesbeing underrepresented. The response rate achieved (83.4%)was well over 70%, a percentage above which most biasesbecome minimal (34).

Having gained recognition by respected professional bod-ies (35, 36) in the management of PTB, SS-TC emerges as acandidate to become a first-line tubal assessment tool in themanagement of infertility. The combination of its diagnosticadvantages and the possibility of a therapeutic benefit war-rants further research on its potential.

References1. Swart P, Mol BW, van deer Veen F, van Beurden M, Redekop WK,

Bossuyt PM. The accuracy of hysterosalpingography in the diagnosis oftubal pathology: a meta-analysis. Fertil Steril 1995;64:486–91.

2. Mol BW, Collins JA, Burrows EA, van d V, Bossuyt PM. Comparisonof hysterosalpingography and laparoscopy in predicting fertility out-come. Hum Reprod 1999;14:1237–42.

3. World Health Organization. Comparative trial of tubal insufflation,hysterosalpingography, and laparoscopy with dye hydrotubation forassessment of tubal patency. Fertil Steril 1986;46:1101–7.

4. Flood JT, Grow DR. Transcervical tubal cannulation: a review. ObstetGynecol Surv 1993;48:768–76.

5. Thurmond AS. Selective salpingography and fallopian tube recanaliza-tion. AJR Am J Roentgenol 1991;156:33–8.

6. Ferraiolo A, Ferraro F, Remorgida V, Gorlero F, Capitanio GL, deCecco L. Unexpected pregnancies after tubal recanalization failure withselective catheterization. Fertil Steril 1995;63:299–302.

7. Capitanio GL, Ferraiolo A, Croce S, Gazzo R, Anserini P, de Cecco L.Transcervical selective salpingography: a diagnostic and therapeuticapproach to cases of proximal tubal injection failure. Fertil Steril1991;55:1045–50.

8. Woolcott R, Petchpud A, O’Donnel P, Stranger J. Differential impacton pregnancy rate of selective salpingography, tubal catheterization andwire-guide recanalization in the treatment of proximal fallopian tubeobstruction. Hum Reprod 1995;10:1423–5.

9. World Health Organization. Laboratory manual for the examination ofhuman semen and sperm-cervical mucus interaction. 4th ed. Cam-bridge: Cambridge University Press, 1999.

10. Papaioannou S, Afnan M, Ola B, Hammadieh N, Coomarasamy A,Sharif K. The learning curve of selective salpingography and tubalcatheterization. Fertil Steril 2002;77:1049–52.

11. Papaioannou S, Afnan M, Coomarasamy A, Ola B, Hammadieh N,Temperton DH, et al. Long term safety of fluoroscopically guidedselective salpingography and tubal catheterization. Hum Reprod 2002;17:370–2.

12. Collett D. Modeling survival data in medical research. London: Chap-man & Hall, 1994.

13. Hovsepian DM, Bonn J, Eschelman DJ, Shapiro MJ, Sullivan KL,Gardiner GA Jr. Fallopian tube recanalization in an unrestricted patientpopulation. Radiology 1994;190:137–40.

14. Woolcott R, Fisher S, Thomas J, Kable W. A randomized, prospective,controlled study of laparoscopic dye studies and selective salpingogra-phy as diagnostic tests of fallopian tube patency. Fertil Steril 1999;72:879–84.

15. Lang EK, Dunaway HE Jr. Salpingographic demonstration of “cobble-stone” mucosa of the distal tubes is indicative of irreversible mucosaldamage. Fertil Steril 2001;76:342–5.

16. Letterie GS, Sakas EL. Histology of proximal tubal obstruction in casesof unsuccessful tubal canalization. Fertil Steril 1991;56:831–5.

17. Hilgers TW, Yeung P. Intratubal pressure before and after transcervicalcatheterization of the fallopian tubes. Fertil Steril 1999;72:174–8.

18. Gleicher N, Parrilli M, Redding L, Pratt D, Karande V. Standardizationof hysterosalpingography and selective salpingography: a valuable ad-junct to simple opacification studies. Fertil Steril 1992;58:1136–41.

19. Karande VC, Pratt DE, Rabin DS, Gleicher N. The limited value ofhysterosalpingography in assessing tubal status and fertility potential.Fertil Steril 1995;63:1167–71.

20. Motta EL, Nelson J, Batzofin J, Serafini P. Selective salpingographywith an insemination catheter in the treatment of women with cornualfallopian tube obstruction. Hum Reprod 1995;10:1156–9.

21. Gleicher N, Confino E, Corfman R, Coulam C, DeCherney AH, HaasG, et al. The multicenter transcervical balloon tubaplasty study: con-clusions and comparison to alternative technologies. Hum Reprod1993;8:1264–71.

22. Gleicher N. Proximal tubal disease: diagnosis and endoscopic treat-ment. In: Grudzinskas JG, Chapman MG, Chard T, Djahanbakch O,(eds). The fallopian tube: clinical and surgical aspects. London: Spring-er-Verlag, 1994:133–49.

23. Gleicher N, Karande V. The diagnosis and treatment of proximal tubaldisease. Hum Reprod 1996;11:1825–8.

24. Kamiyama S, Miyagi H, Kanazawa K. Therapeutic value of selectivesalpingography for infertile women with patent fallopian tubes: theimpact on pregnancy rate. Gynecol Obstet Invest 2000;49:36–40.

25. Papaioannou S, Afnan M, Girling AJ, Coomarasamy A, Ola B, Olu-fowobi O. The effect in pregnancy rates of tubal perfusion pressuresreductions achieved by guide-wire tubal catheterisation. Hum Reprod2002;17:2174–9.

26. Thurmond AS, Novy M, Rosch J. Terbutaline in diagnosis of interstitialfallopian tube obstruction. Invest Radiol 1988;23:209–10.

27. Winfield AC, Pittaway D, Maxson W, Daniell J, Wentz AC. Apparentcornual occlusion in hysterosalpingography: reversal by glucagon. AJRAm J Roentgenol 1982;139:525–7.

28. Lang EK, Dunaway HE Jr, Roniger WE. Selective osteal salpingogra-phy and transvaginal catheter dilatation in the diagnosis and treatmentof fallopian tube obstruction. AJR Am J Roentgenol 1990;154:735–40.

29. Kumpe DA, Zwerdlinger SC, Rothbarth LJ, Durham JD, Albrecht BH.Proximal fallopian tube occlusion: diagnosis and treatment with trans-cervical fallopian tube catheterization. Radiology 1990;177:183–7.

30. Yoder IC, Hall DA. Hysterosalpingography in the 1990s. AJR Am JRoentgenol 1991;157:675–83.

31. Thurmond AS. Seminars in interventional radiology. Semin InterventRadiol 2000;17:303–8.

32. Martensson O, Nilsson B, Ekelund L, Johansson J, Wickman G. Se-lective salpingography and fluoroscopic transcervical salpingoplasty fordiagnosis and treatment of proximal fallopian tube occlusions. ActaObstet Gynecol Scand 1993;72:458–64.

33. Lang EK, Dunaway HH. Recanalization of obstructed fallopian tube byselective salpingography and transvaginal bougie dilatation: outcomeand cost analysis. Fertil Steril 1996;66:210–5.

34. Lydeard S. Commentary: avoid surveys masquerading as research. BrMed J 1996;313:733–4.

35. American Fertility Society. Guidelines for tubal disease. Birmingham,AL: American Fertility Society, 1993.

36. Royal College of Obstetricians and Gynecologists. The management ofinfertility in secondary care. London: RCOG, 1998.

FERTILITY & STERILITY� 617