1 University of Auckland Department of Obstetrics and Gynaecology, National Women's Hospital, Auckland, 2 Fertility Plus, National Women's Hospital, Auckland, 3 University Specialists, Mercy Specialist Centre, Epsom, Auckland and 4 Auckland Radiology Group, 641 Manukau Road, Royal Oak, Auckland, New Zealand
5 To whom correspondence should be addressed at: University of Auckland Department of Obstetrics and Gynaecology, National Women's Hospital, Auckland, New Zealand. Email: n.johnson{at}auckland.ac.nz
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: endometriosis/lipiodol/oil soluble contrast media/randomized trial/unexplained infertility
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Diagnostic HSG were originally performed with OSCM. Their use was gradually replaced by WSCM for a number of reasons: lower cost; better imaging of the tubal mucosal folds and ampullary rugae than OSCM (Soules and Spadoni, 1982); lower viscosity and more prompt demonstration of tubal patency (reducing the need for a delayed film); less likelihood of persistence of contrast medium within the pelvic cavity and of complications such as intravasation resulting in allergic reactions or anaphylaxis or long-term lipogranuloma formation. There were several reports of deaths after the use of oily media in radiology (before the use of fluoroscopy screening) but none was reported after 1967 (Lindequist et al., 1991
). It is reassuring that the advent of fluoroscopy screening appears to have abolished severe adverse reactions following the use of oil-based media in radiology (Lindequist et al., 1991
) and the safety of HSG with OSCM in this context has been confirmed (Nunley et al., 1987
). Lipogranuloma formation has not been reported in women with patent Fallopian tubes following OSCM studies and is not known to have long-term consequences (Acton et al., 1988
). In general, many fertility centres have moved towards the use of laparoscopy with tubal patency testing rather than HSG, owing to the increased information available concerning tubo-peritoneal abnormalities (including endometriosis) from laparoscopy and the possibility of treating endometriosis and adhesions surgically (Johnson et al., 2000
). Currently those centres performing HSG tend to use WSCM rather than OSCM for the reasons outlined above.
The first systematic review of tubal flushing suggested that OSCM have a greater fertility-enhancing effect than WSCM and that the effect is most pronounced in couples with unexplained infertility (Watson et al., 1994). The effect in women with endometriosis has not been examined previously. The most recent update of the systematic review of randomized controlled trials (RCT), performed in 2002, has highlighted the problems with these trials (Johnson et al., 2004
). First, many of the data were from trials where the interventions were performed as diagnostic tests rather than as therapeutic interventions. Second, data on sexual frequency were not available for the flushing versus no treatment trials, hence the notion that the increased pregnancy rate might be due simply to an increased sexual frequency in the group who received treatment. Third, although five RCT evaluated the effect of OSCM versus WSCM (Alper et al., 1986
; De Boer et al., 1998
; Letterie and Rose, 1990
; Rasmussen et al., 1991
; Spring et al., 2001
) and three RCT evaluated the effect of adding OSCM to WSCM (Yang et al., 1989
; Spring et al., 2001
; Steiner et al., 2003
), there were only two RCT which assessed flushing with OSCM versus no intervention, one with small numbers in couples with unexplained infertility (Nugent et al., 2002
), the other in couples for whom the cause for infertility was not specified (Ogata et al., 1993
). Finally, there has also been a trend towards IVF as the panacea for all causes of infertility. The conclusion of the review was that there were insufficient data to support the routine use of tubal flushing as a fertility treatment (Johnson et al., 2004
).
The objective of this trial was to ascertain the effectiveness of lipiodol flushing for enhancing fertility in women with unexplained infertility.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Protocol
The inclusion criteria for women in the study population were as follows: unexplained infertility (or endometriosis where the Fallopian tubes and ovaries were unaffected by endometriotic disease, in the context of otherwise unexplained infertility) of duration 12 months; full investigation for the cause of infertility completed; age 1839 years inclusive; early follicular FSH level of
10 IU/l; mid-luteal progesterone level of
25 mmol/l in a spontaneous cycle; bilateral tubal patency confirmed either by dye studies at laparoscopy or by HSG. The partner's semen analysis had to be normal by World Health Organization (1992)
criteria. Exclusion criteria for women were: abnormal Fallopian tubes or a history of tubal ectopic pregnancy; laparoscopic evidence of endometriosis which had affected either the Fallopian tubes or ovaries; iodine allergy. Eligible couples who had given consent were stratified into two populations prior to randomization: group A comprised women with unexplained infertility; group B comprised women with endometriosis in the context of normal Fallopian tubes and ovaries and otherwise unexplained infertility. Women in the endometriosis population had previously had a laparoscopy at which endometriosis was visualized.
Women were randomized to receive lipiodol flushing performed by a HSG technique with fluoroscopic X-ray screening, or to no intervention. Both groups received the same fertility information sheet, which reinforced the principle of the pre-ovulatory fertile phase of the cycle, the timing of sexual intercourse to the fertile phase and the optimal sexual frequency (at least every 48 h) in the fertile phase. The lipiodol contrast medium was Lipiodol Ultra Fluide (Guerbet, France), an iodized poppy seed oil, obtained by substitution of ethyl esters for the glyceryl esters of lipiodol. One millilitre of Lipiodol Ultra Fluide contains 0.48 g iodine. Lipiodol flushing was carried out by one of two authors (W.H., n=45; N.J., n=27) in the follicular phase of the cycle between the end of menses and day 12 of the cycle. With the woman in the left lateral or supine position, using a no-touch technique after antiseptic solution application, the cervix was cannulated by a LeechWilkinson cannula (Downs Distributors, New Zealand) under speculum visualization. In the supine position and with intermittent fluoroscopic X-ray guidance, pre-warmed (37°C) lipiodol was slowly instilled into the uterine cavity. Typically 10 ml of lipiodol was instilled with the instillation being stopped once unequivocal bilateral spill of contrast from the Fallopian tubes into the peritoneal cavity had been observed. If no peritoneal spill was observed after use of 10 ml, further lipiodol was instilled. If intravasation of lipiodol was observed on X-ray (contrast apparent in the venous system), the instillation was immediately stopped.
The primary outcomes were clinical pregnancy (a positive urine or serum pregnancy test in association with an intrauterine gestation sac on ultrasound scan or histological evidence of trophoblastic tissue in the context of a miscarriage or ectopic pregnancy) at 6 months following randomization and live birth (once data for pregnancy outcomes became available). Secondary outcomes were miscarriage, ectopic pregnancy, multiple pregnancy, any other complications including a diagnosis of lipogranuloma, and a comparison of pre- and post-randomization sexual frequency and focus to the fertile stage of the cycle. These latter sexual behaviour variables were based on an estimate of sexual frequency before randomization and at the completion of follow-up. (For couples achieving a pregnancy during the 6 month follow-up phase, these estimates of sexual behaviour were based on that occurring prior to pregnancy.) A subjective increase in sexual frequency and a greater focus of sexual activity to the fertile phase of the cycle were both defined as that estimated by the woman at the 6 month follow-up; an objective increase in sexual frequency was defined by a comparison of the woman's estimate of sexual frequency at the 6 month follow-up compared to her estimate prior to randomization. For those undergoing lipiodol HSG, data were collected for the HSG appearances, procedural pain, intravasation and other lipiodol HSG-related complications.
A power calculation had been undertaken within the overall population to determine an appropriate sample size to demonstrate what was deemed to be a clinically important treatment effect of lipiodol flushing (Johnson et al., 2004). For couples with unexplained infertility (which could include women without confirmed endometriosis and women with confirmed endometriosis in the context of otherwise unexplained infertility), in order to have 80% power at the 95% confidence level to detect an increase in the pregnancy rate at 6 months post-randomization from 7% with no treatment to 25% with lipiodol flushing, 150 participants would be required for analysis. It was determined to allow a maximum of 3 years to attain this number, but recruitment was stopped once 158 couples had been recruited.
Data were collected from study participants by a telephone consultation with a research nurse; data concerning the lipiodol flushing procedure were collected in the presence of the study participant by the clinician performing the procedure. Numbers are presented as counts (percentages) for dichotomous data; continuous data are presented as either median (interquartile range) or mean (SD). Pregnancy outcomes for the unexplained infertility population (group A) and endometriosis population (group B) are presented separately. The results are also pooled in a total population, according to the a priori specification upon which the power calculation was based. Statistical analyses were performed using a 2-test or Fisher's exact test for dichotomous data and either Student's t-test or MannWhitney test for continuous data using SPSS software. The primary analysis was conducted a priori on an intention-to-treat (ITT) basis, but an exploratory per-protocol analysis was also carried out.
Assignment
Women were recruited from publicly funded and private secondary level gynaecology clinics and a tertiary level fertility clinic in Auckland, New Zealand. Some women from other similar clinics throughout New Zealand contacted the trial co-ordinators to express interest and were evaluated for eligibility. Having given informed consent to participate, women were randomized in the menstrual phase of the cycle in which they had committed to attend for a lipiodol HSG if their randomization allocated them to receive the treatment procedure.
Randomization was performed using two computer-generated random number sequences (unknown to the research nurse, the executor of the assignment), known as group A (women with unexplained infertility without confirmed endometriosis) and group B (women with endometriosis in the context of otherwise unexplained infertility) at the beginning of the cycle on which it had been determined that a lipiodol flushing procedure would be performed. Allocation concealment was securely maintained by storage in sealed, sequentially numbered opaque envelopes until the interventions were assigned. The randomization sequences were unblocked with up to 110 available in group A (96 actually randomized) and 90 available in group B (62 randomized). Allocation was strictly maintained sequentially, all envelopes in the sequence being used, with the allocated groups analysed on an ITT basis.
Blinding
It was not possible to blind participants to treatment allocation since the treatment involved a HSG procedure and the control involved no intervention. There was also no blinding of the executor of the assignment, the clinician performing the lipiodol flushing procedure, nor of the assessor at follow-up.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Three women were lost to follow-up. There were two known deviations from the protocol post-randomization, both in women randomized to lipiodol flushing. One woman from group A underwent a further laparoscopy and chromotubation procedure (which confirmed bilateral tubal patency) after a lipiodol HSG had shown neither fill nor spill of contrast from the Fallopian tubes, then conceived with IVF treatment within the 6 month follow-up phase. One woman from group B conceived prior to the flushing procedure, the presumed menstrual period at the time of randomization being an implantation bleed, pregnancy then being confirmed prior to the planned flushing procedure. These two women both had live births.
The baseline characteristics of the women at entry to the trial are presented in Table I. The groups randomized to lipiodol flushing compared to no treatment were similar regarding all factors known to be influential on fertility.
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
When our data are considered in light of data from previous RCT (Johnson et al., 2004), the effectiveness of lipiodol flushing is confirmed not only in the endometriosis population, but also in the population with unexplained infertility in a meta-analysis when appropriate pooling of RCT is carried out. In women with unexplained infertility, meta-analysis of women with unexplained infertility without confirmed endometriosis from our trial and those with unexplained infertility from the trial of Nugent et al. (2002)
gives a RR for pregnancy of 2.05 (95% CI 1.073.93) for lipiodol flushing versus no intervention (compared to RR 1.60, 95% CI 0.813.16 from this trial alone). The estimate of the number needed to treat (NNT) to achieve one additional pregnancy in women with endometriosis, based on our trial data, is
3 (95% CI 26; control pregnancy rate 10.8%); NNT for one additional pregnancy in women with unexplained infertility, based on meta-analysis data, is
6 (95% CI 339; control pregnancy rate 15.4%).
The main strengths of this study were the randomized design with secure allocation concealment and the analysis on both ITT and per protocol basis. The main weaknesses were that the randomization was not blocked and the absence of blinding. It was our a priori intention to present pooled data for the total population (group A and group B) as the primary analysis and the power calculation for sample size was based on the pooling of these populations. However, the data suggest that the populations were dissimilar in baseline fertility (the no-intervention pregnancy rates were 10.8% for women with endometriosis and 20.8% for women with unexplained infertility without confirmed endometriosis), which is consistent with the reduced natural fecundity (by a factor of approximately one-half) previously reported in women with minimal or mild endometriosis (Jansen, 1986; Toma et al., 1992
), and dissimilar in the response to lipiodol treatment. Moreover, owing to unblocked randomization, there was an over-representation of women with endometriosis in the no-intervention group (43.5%) compared to the lipiodol flushing group (34.2%) in the total population that has the potential to bias the pooled results in favour of lipiodol flushing. Although results for the pooled total population are presented, they must be interpreted cautiously.
It is recognized that lack of understanding of the fertile phase of the cycle and the optimal timing and frequency of intercourse to maximize fertility is widespread even in couples attending for tertiary assisted conception treatment (Blake et al., 1997). Our policy of providing a fertility information sheet to all trial participants was designed to maximize the number of pregnancies in the treatment and control groups. With our non-blinded design, it was possible that women undergoing lipiodol flushing could have increased their sexual frequency compared to the control group. However, our data do not support a difference in sexual behaviour (in terms of an increase in sexual frequency or an improved focus of sexual activity to the fertile phase of the cycle) amongst women assigned to lipiodol flushing treatment compared to those assigned to no intervention. Thus the increased pregnancy rate following lipiodol flushing is not explained by changes in sexual behaviour following treatment.
An exploratory per-protocol secondary analysis was deemed appropriate owing to the possibility of bias in favour of lipiodol flushing from the primary ITT analysis, given that two pregnancies in women randomized to receive lipiodol flushing could not be attributed to this intervention (one woman conceived prior to lipiodol flushing and another conceived in IVF treatment following lipiodol flushing). The conclusions from the trial were the same, whether based on an ITT or per-protocol analysis.
This RCT is one of two RCT assessing tubal flushing known to have been completed since the most recent update of the systematic review of RCT in 2002 (Johnson et al., 2004). The other trial suggested that the beneficial effect of adding OSCM to a WSCM, which shortened the time to pregnancy, attenuated over time and had disappeared by 18 months (Steiner et al., 2003
). Whilst most RCT showed a significant benefit of OSCM over WSCM in terms of the subsequent pregnancy rate (Johnson et al., 2004
), one of the largest and methodologically most robust trials comparing OSCM versus WSCM (Spring et al., 2001
) failed to demonstrate any relative benefit and this question remains unresolved.
There is now little doubt of the efficacy of lipiodol flushing versus no intervention and the effect is unlikely to be due to changes in sexual behaviour following the procedure. The mechanism of the fertility-enhancing effect of OSCM is unknown. One theory was that OSCM may be highly effective at flushing out debris from otherwise undamaged tubes (Watson et al., 1994). Such debris may not necessarily block the Fallopian tube, but may hinder conception or embryo transport along the Fallopian tube. An alternative, immunological hypothesis was that OSCM may enhance fertility for women with unexplained infertility or mild endometriosis by affecting peritoneal macrophages (Johnson et al., 1992
)OSCM have been shown to alter interleukin and prostaglandin production by peritoneal macrophages (Sawatari et al., 1993
) and to modulate macrophage activity in phagocytosis of sperm (Mikulska et al., 1994
). However, the pregnancy-enhancing effect may simply lie at the level of the endometrium. For most couples having unsuccessful IVF treatment, the outcome hinges on failed implantation: it stands to reason that a treatment which substantially increases the likelihood of conception is likely to have some effect on endometrial receptivity. This theory is supported by two spontaneous pregnancies in our series where lipiodol was not seen to spill from either Fallopian tube (although bilateral tubal patency had previously been confirmed). There is increasing evidence that the infertility related to mild endometriosis may be related to implantation failure owing to impaired endometrial receptivity (Kao et al., 2003
). The greater treatment effect in women with endometriosis compared to those with unexplained infertility in our study might result from a mechanism where lipiodol corrects an endometrial implantation dysfunction. The effect of lipiodol on the endometrium merits further investigation.
The current options for management of unexplained infertility include expectancy, use of empirical clomiphene citrate which may approximately double the per-cycle chance of conception (Hughes et al., 2004), intrauterine insemination (IUI) or IVF. Data are sparse on the relative advantages of these treatment options (Pandian et al., 2004
). Women with endometriosis have the further option of laparoscopic surgical treatment of endometriosis (Jacobson et al., 2004
). Lipiodol flushing now presents an alternative treatment option which may be more appealing to many couples. The advantages of lipiodol flushing are that the technique is less invasive than IVF or laparoscopic surgery, is relatively low cost (in New Zealand, comparable to the cost of a single IUI cycle), pregnancy is achieved by sexual intercourse thus it is regarded as more natural and there is no increased risk of multiple pregnancy, a problem which has been associated with many other fertility treatment options.
In conclusion, lipiodol flushing is effective in enhancing fertility for women with unexplained infertility and women with mild endometriosis in the context of otherwise unexplained infertility. The greatest benefit is apparent in women with endometriosis. It should be considered as a possible first-line fertility treatment for such women, especially in circumstances where resources for other assisted reproductive technologies are limited.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Alper MM, Garner PR, Spence JEH and Quarrington AM (1986) Pregnancy rates after hysterosalpingography with oil- and water-soluble contrast media. Obstet Gynecol 68, 69.[Abstract]
Barwin N (1971) Hysterosalpingography in infertility. Ulster Med J 41, 6165.[Medline]
Blake D, Smith D, Bargiacchi A, France M and Gudex G (1997) Fertility awareness in women attending a fertility clinic. Aust NZ J Obstet Gynaecol 37, 350352.[Medline]
De Boer AD, Vemer HM, Willemsen WNP and Saunders FBM (1998) Oil or aqueous contrast media for hysterosalpingography: a prospective, randomized, clinical study. Eur J Obstet Gynecol Reprod Biol 28, 6568.
DeCherney AH, Kort H, Barney JB and DeVore GR (1980) Increased pregnancy rate with oil-soluble hysterosalpingography dye. Fertil Steril 33, 407410.[Medline]
Gillespie H (1965) The therapeutic aspect of hysterosalpingography. Br J Radiol 38, 301302.[Medline]
Hughes E, Collins J and Vandekerckhove P (2004) Clomiphene citrate for unexplained subfertility in women (Cochrane Review). In The Cochrane Library, Issue 2. John Wiley & Sons Ltd, Chichester, UK.
Jacobson TZ, Barlow DH, Koninckx PR, Olive D and Farquhar C (2004) Laparoscopic surgery for subfertility associated with endometriosis (Cochrane Review). In The Cochrane Library, Issue 2, John Wiley & Sons Ltd, Chichester, UK.
Jansen RP (1986) Minimal endometriosis and reduced fecundability: prospective evidence from an artificial insemination by donor program. Fertil Steril 46, 141143.[Medline]
Johnson J, Montoya I and Olive D (1992) Ethiodol oil contrast medium inhibits macrophage phagocytosis and adherence by altering membrane electronegativity and microviscosity. Fertil Steril 58, 511517.[Medline]
Johnson NP, Taylor K, Nadgir AA, Chinn DJ and Taylor PJ (2000) Can diagnostic laparoscopy be avoided in routine investigation for infertility? Br J Obstet Gynaecol 107, 174178.
Johnson N, Vandekerckhove P, Watson A, Lilford R, Harada T and Hughes E (2004) Tubal flushing for subfertility (Cochrane Review). In The Cochrane Library, Issue 2, John Wiley & Sons Ltd, Chichester, UK.
Kao LC, Germeyer A, Tulac S, Lobo S, Yang JP, Taylor RN, Osteen K, Lessey BA and Giudice LC (2003) Expression profiling of endometrium from women with endometriosis reveals candidate genes for disease-based implantation failure and infertility. Endocrinology 144, 28702881.
Letterie GS and Rose GS (1990) Pregnancy rates after the use of oil-based and water-based contrast media to evaluate tubal patency. South Med J 83, 14021403.[Medline]
Lindequist S, Justesen P, Larsen C and Rasmussen F (1991) Diagnostic quality and complications of hysterosalpingography: oil- versus water-soluble contrast mediaa randomized prospective study. Radiol 179, 6974.
Mackey RA, Glass RH, Olson LE and Vaidya R (1971) Pregnancy following hysterosalpingography with oil and water soluble dye. Fertil Steril 22, 504507.[Medline]
Massouras HG (1970) Presentation of an insufflator-instillator and the use of it. Fertil Steril 21, 407410.[Medline]
Mikulska D, Kurzawa R and Rozewicka L (1994) Morphology of in vitro sperm phagocytosis by rat peritoneal macrophages under influence of oily contrast medium (lipiodol). Acta Eur Fertil 25, 203206.[Medline]
Nugent D, Watson AJ, Killick SR, Balen AH and Rutherford AJ (2002) A randomized controlled trial of tubal flushing with lipiodol for unexplained infertility. Fertil Steril 77, 173175.[CrossRef][Medline]
Nunley W, Bateman B, Kithchin J and Pope T (1987) Intravasation during hysterosalpingography using oil-base contrast mediuma second look. Obstet Gynecol 70, 309312.[Abstract]
Ogata R, Nakamura G, Uchiumi Y, Yokoyamay M, Watanabe Y, Nozaki M, Sano M and Nakano H (1993) Therapeutic efficacy of hysterosalpingography (HSG) in infertility, a prospective, randomized, clinical study. Jap J Fertil Steril 38, 9194.
Pandian Z, Bhattacharya S, Nikolaou D, Vale L and Templeton A (2004) In vitro fertilisation for unexplained subfertility (Cochrane Review). In The Cochrane Library, Issue 2, John Wiley & Sons Ltd, Chichester, UK.
Rasmussen F, Lindequist S, Larsen C and Justesen P (1991) Therapeutic effect of hysterosalpingography: oil- versus water-soluble contrast mediaa randomized prospective study. Radiology 179, 7579.[Abstract]
Sawatari Y, Hori T and Hoshiai H (1993) Oily contrast medium as a therapeutic agent for infertility because of mild endometriosis. Fertil Steril 59, 907911.[Medline]
Soules MR and Spadoni LR (1982) Oil versus aqueous media for hysterosalpingography: a continuing debate based on many opinions and few facts. Fertil Steril 28, 111.
Spring DB, Barkan HE and Pruyn SC (2001) Potential therapeutic effects of contrast materials in hysterosalpingography: a prospective randomized clinical trial. Radiology 214, 5357.
Steiner AZ, Meyer W, Clark RL and Hartmann KE (2003) Oil-soluble contrast during hysterosalpingography in women with proven tubal patency. Obstet Gynecol 101, 109113.
Toma SK, Stovall DW and Hammond MG (1992) The effect of laparoscopic ablation or Danocrine on pregnancy rates in patients with stage I or II endometriosis undergoing donor insemination. Obstet Gynecol 80, 253256.[Abstract]
Watson A, Vandekerckhove P, Lilford R, Vail A, Brosens I and Hughes E (1994) A meta-analysis of the therapeutic role of oil soluble contrast media at hysterosalpingography: a surprising result? Fertil Steril 61, 470477.[Medline]
Weir WC and Weir DR (1951) Therapeutic value of salpingograms in infertility. Fertil Steril 2, 514522.[Medline]
World Health Organization (1992) WHO Laboratory Manual for the Examination of Human Semen and Sperm-Cervical Mucus Interaction, 3rd edn. Cambridge University Press, Cambridge, UK.
Yaegashi N, Kuramoto M, Nakayama C and Nakano M (1987) Pregnancy rates after hysterosalpingographycomparing water soluble contrast medium with oily contrast medium. Acta Obstet Gynaecol Japon 39, 18121814.
Yang KN, Yeh NG and Pan SB (1989) The therapeutic effects of oil-soluble hysterosalpingography contrast medium following water-soluble hysterosalpingography contrast medium. Chin Med J 44, 293297.
Submitted on March 22, 2004; accepted on June 24, 2004.