HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Spring, D. B. Articles by Pruyn, S. C. Search for Related Content PubMed PubMed Citation Articles by Spring, D. B. Articles by Pruyn, S. C.

Potential Therapeutic Effects of Contrast Materials in Hysterosalpingography: A Prospective Randomized Clinical Trial¹

¹ From the Departments of Diagnostic Imaging (D.B.S.) and Obstetrics and Gynecology (S.C.P.), Kaiser Permanente Medical Center, 280 W MacArthur Blvd, Oakland, CA 94611-5693, and Health Services Research, Berkeley, Calif (H.E.B.). Received December 23, 1998; revision requested February 19, 1999; revision received April 20; accepted July 29. Supported in part by Kaiser Permanente Medical Care Program–Northern California Innovation Project grant no. 930198. Address reprint requests to D.B.S. (e-mail: David.Spring@kp.org).

	Abstract

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
PURPOSE: To evaluate the influence of the contrast material used in hysterosalpingography (HSG) on subsequent reproductive success, independent of other therapeutic interventions.

MATERIALS AND METHODS: In a prospective, multisite, randomized trial, 666 women who had been infertile for more than 1 year and were scheduled to undergo HSG as part of their evaluation were assigned to one of three groups: those receiving water-soluble contrast material (WSCM) (n = 260), those receiving oil-soluble contrast material (OSCM) (n = 273), and those receiving both OSCM and WSCM (n = 133). Possible causes of infertility and therapeutic interventions were abstracted from the medical records. Data on conception within 1 year and the outcome of conception were ascertained from multiple sources.

RESULTS: Of 666 women, 204 (30.6%) had at least one pregnancy, and 136 (20.4%) had live births. The rates of live births were 20.4% (54 of 260) after HSG with WSCM, 19.4% (53 of 273) after HSG with OSCM, and 21.8% (29 of 133) after HSG with both WSCM and OSCM. Differences in reproductive outcome among contrast material groups were not statistically significant (²₈ = 6.08, P = .64). Whatever the cause of infertility, the use of different contrast materials led to no significant differences in the rates of live births.

CONCLUSION: There is no evidence to suggest that the choice of contrast material affects the rate of term pregnancy.

Index terms: Abortion, 854.8252 • Contrast media, comparative studies • Contrast media, therapeutic effects • Pregnancy, 854.131 • Pregnancy, ectopic, 854.823 • Uterus, radiography, 854.1282

	Introduction

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
Continuing debate has focused on whether the choice of contrast material used in hysterosalpingography (HSG) influences subsequent reproductive success, independent of other therapeutic interventions (1,2). HSG most often has the goal of aiding efforts to induce normal term pregnancy by helping identify treatable causes of infertility. Because oil-soluble contrast material (OSCM) may promote granulomatous inflammation in the presence of obstructed or inflamed fallopian tubes, DeCherney and colleagues (3) recommend as an option the initial use of water-soluble contrast material (WSCM) to help confirm tubal patency, followed by use of OSCM "as a therapeutic modality." In the present study, we compared outcomes in a randomly selected sample of women undergoing HSG when a combination of WSCM and OSCM was used.

In a recent large, randomized clinical trial (4), the rate of successful term pregnancy after the use of OSCM for HSG was three times that found after the use of WSCM, independent of other therapeutic interventions. If this outcome is true, HSG (the diagnostic procedure) would have an independent therapeutic effect, and the use of OSCM would be preferable in the absence of substantial adverse effects. By using a larger sample size, we sought to test the validity of the potentially important findings of Rasmussen et al (4).

	MATERIALS AND METHODS

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
We conducted a prospective, multisite, randomized controlled clinical trial within a large health maintenance organization. The study was approved by the Northern California Institutional Review Board of the Kaiser Permanente Medical Care Program, and informed consent was obtained from all patients. During the study period (December 15, 1993, to July 1, 1996), women who had been infertile for more than 1 year were recruited from those referred for HSG as part of their infertility evaluation. All women were informed of the nature of the study and the risks associated with both types of contrast material; they were informed that they might be assigned to a group that could prove to have a lower rate of term pregnancy. Six hundred seventy-three women agreed to participate in this study. We excluded from the study six women who could not be contacted to confirm whether conception or term pregnancy had occurred. We also excluded one woman who became pregnant within 12 months after HSG but for whom we could not determine the outcome of pregnancy.

The 666 women who consented to participate were randomly assigned to one of three groups: those receiving WSCM (52.7% diatrizoate meglumine and 26.8% iodipamide meglumine, Sinografin; Bracco Diagnostics, New Brunswick, NJ) (n = 260), those receiving OSCM (ethiodized oil, Ethiodol; Savage Laboratories, Melville, NY) (n = 273), or those receiving sequential WSCM and OSCM (n = 133). For the sequential (combined) contrast material group, WSCM was administered first; OSCM was then administered if the fallopian tubes were not obstructed (3). At each of 10 separate facilities, a randomized sequence was generated in blocks of nine patients, thus allowing randomization within each separate medical center, as well as for all women. Physicians performing HSG and women undergoing the studies were aware of the contrast material selected with this randomization scheme.

For calculations, all statistical estimates were based on contrast material assignment (ie, according to intention to treat), regardless of whether all of the planned contrast material was administered (particularly for the combined contrast materials group). More than 80 radiologists and gynecologists participated in the Infertility Work Group. In most cases, radiologists provided patient information, obtained consent for the study, conducted the examinations, and interpreted the examination findings. In many instances, radiologists and gynecologists participated in and shared responsibilities for this procedure. In all instances, a radiologist offered a final interpretation. Procedure techniques varied among physicians but were standard.

For HSG, balloon catheters were placed in the uterus or cervix and often included the use of cannulas and tenacula or flange-tipped metal cannulas placed into the external cervical os. In some instances, prone imaging was performed when tubes were not visualized at HSG. The decisions regarding technical variations, including the acquisition of delayed images, were made by the examining physicians. Usually, about 10 mL of contrast material was used: The mean amount of WSCM used was 9.4 mL (range, 2–75 mL), and the mean amount of OSCM used was 8.6 mL (range, 1–55 mL). In studies with both contrast materials, means of 8.2 mL (range, 1–30 mL) of WSCM and 6.0 mL (range, 1–20 mL) of OSCM were used.

For all studies, the physician who performed HSG provided demographic and infertility information about patients, information about the HSG technique used, and results of the procedure. In addition, patients were interviewed about their infertility history, their subjective experience with HSG, any adverse consequences of HSG, and their subsequent reproductive experience. We tried to contact all women within 8 weeks after HSG and 1 year after HSG. We followed up all women who became pregnant during that year to learn the outcome of each pregnancy. All women thus were followed up for as long as 21 months.

Outcomes were classified as follows: (a) no conception within 12 months after HSG, (b) conception within 12 months after HSG followed by spontaneous abortion, (c) conception within 12 months after HSG with ectopic pregnancy, (d) conception within 12 months after HSG with therapeutic abortion, and (e) conception followed by live birth.

For all participants, we supplemented the above data by reviewing medical and infertility records and noted pertinent medical history, medications, and associated medical conditions. We also gathered information from the patients' medical records concerning medical and surgical interventions during the year of observation. We tabulated known data concerning the possible role of their partners' infertility.

In the study by Rasmussen et al (4), rates of term pregnancy were approximately 30% with OSCM and 10% with the WSCM diatrizoate meglumine (similar rates were found for two other low-osmolality WSCMs, ioxaglate and iohexol). We designed our study so that we could identify smaller differences—specifically, to distinguish if rates were 20% for OSCM and 10% for WSCM. We calculated (5,6) that 237 subjects per contrast material group would be required to achieve a power of 0.80 to detect these differences at a statistical significance level of less than .05. The ² test, analysis of variance (F test), and Fisher exact test were used to test for differences.

With the approval of the institutional review board, we discontinued the arm of the study that used the contrast material combination (ie, WSCM and OSCM) soon after recruiting 100 patients for that combination, because initial recruiting had been slower than expected. For the combined group, statistical data adequate for comment had been obtained. The statistical power to identify a 10% variation (as we planned for OSCM and WSCM), however, was not sought or obtained.

	RESULTS

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
Tabulated results were based on the 666 women we were able to observe for at least 12 months after HSG (21 months after HSG if they became pregnant).

Rates of live birth varied slightly among the patients in the three contrast material groups, but differences between these rates were not statistically significant (Table 1) (²₂ = 6.08, P = .64). Four hundred sixty-two (69.4%) women did not conceive during the 12 months after HSG; 204 (30.6%) did conceive (Table 1). Of the 204 pregnancies, 136 (66.7%) resulted in live birth; 59 (28.9%), in spontaneous abortion, seven (3.4%), in ectopic pregnancy; and two (1.0%), in therapeutic abortion.

We noted statistically significant differences between assigned contrast material groups in the techniques used for the examinations. An acorn-tipped cannula, usually with a tenaculum, was used more often in the OSCM group (54 [19.8%] of 273) and the combination group (26 [19.5%] of 133) than in the WSCM group (31 [12.0%] of 259) (²₂ = 6.81, P = .033).

For women who became pregnant and had a subsequent live birth, the estimated median time to conception during the 1-year observation period was approximately 3 months after HSG. Approximately 10% of the women who became pregnant after HSG became pregnant within the menstrual cycle during which HSG was performed. Similar results were obtained for women in the WSCM-OSCM combination group.

Of the 666 women, 238 (35.7%) had primary infertility (Table 2). Women with primary infertility were more often in the WSCM group, although the differences in prevalence of primary infertility among the assigned groups were not significant. In the WSCM group, 97 (37.1%) of 261 had primary infertility; in the OSCM group, 97 (35.0%) of 277 had primary infertility; and in the WSCM-OSCM combined group, 47 (34.8%) of 135 had primary infertility (²₂ = 0.34, P = .77). The likelihood of pregnancy was no greater in women with primary infertility than that in women with secondary infertility. However, women who became pregnant had significantly different pregnancy outcomes, depending on whether they had primary or secondary infertility: Women with primary infertility were less likely either to have a spontaneous abortion (20 [25.3%] of 79 vs 39 [31.2%] of 125) or to have an ectopic pregnancy (zero [0.0%] of 79 vs seven [5.6%] of 125) and were more likely to have a live birth (²₄ = 10.71, P = .03). Women who had previously had ectopic pregnancies and/or women who had ectopic pregnancies during the study were randomly distributed among the three contrast material groups (for the WSCM group, four [1.5%] of 260; for the OSCM group, two [0.7%] of 273; for the combined group, one [0.8%] of 133), with no significant differences among groups (²₂ = 0.97, P = .61).

In 44.4% (193 of 435) of the women for whom information was provided, the medical history included mention of a therapeutic abortion. The proportion of women who had previously undergone therapeutic abortion in each of the 5-year cohorts between the ages of 20 and 45 years was relatively constant. The rate of live birth for this group was 15.0% (29 of 193 patients), whereas the rate for women without prior therapeutic abortion was 23.2% (101 of 435 patients) (²₄ = 10.94, P = .027).

A history of sexually transmitted disease was common among infertile women: 50 had a history of gonorrhea, six had a history of syphilis, 35 had a history of chlamydia, and 49 had a history of pelvic inflammatory disease. A reported history of pelvic inflammatory disease (cause unspecified) in four (8.2%) of 49 women was significantly associated (²₄ = 10.27, P = .036) with a lower likelihood of live birth; 132 (21.2%) of 624 women with no prior pelvic inflammatory disease had a lower likelihood of live birth. Among the 50 women with a history of gonococcal infection, 11 (22%) were found to have bilateral (n = 7) or unilateral (n = 4) hydrosalpinx at HSG (²₂ = 12.01, P < .003). A history of syphilitic or chlamydial infection, however, was not significantly associated with hydrosalpinx.

We could not identify any statistically significant association between contrast material group and causal factors stated in the medical records or identified during our study. Assignment of causal factors was often based on notes recorded in the medical record, and the strength of support from the reported data varied. We specifically noted several causal factors at least once (sometimes in combination): tubal (n = 94), male (n = 84), ovulatory (n = 63), age (n = 43), uterine (n = 27), peritoneal (n = 22), endocrine (n = 18), cervical (n = 2), immunologic (n = 1), and drug-associated (n = 1) factors. In four cases, the causal factor was specifically listed as unexplained (7).

Artificial insemination was performed in 166 of the 666 women (64 [24.6%] of 260 in the WSCM group, 69 [25.3%] of 273 in the OSCM group, and 33 [24.8%] of 133 in the WSCM-OSCM combined group). Presumed or suspected infertility due to a male factor (eg, varicocele or disorders of spermatogenesis, including oligozoospermia and azoospermia) had been specified in the medical record in 40 of these 166 women. Reproductive outcome was similar for women who did and for those who did not undergo artificial insemination.

	DISCUSSION

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
The potentially therapeutic role of OSCM during HSG has been the subject of controversy for many years (1,2). Results from the large, well-designed study by Rasmussen and colleagues (4) supported the belief that the use of OSCM had a therapeutic effect on reproductive outcome. Findings from our study, however, which used a larger sample and was more sensitive, failed to confirm those results. We cannot fully explain this difference. We note that Rasmussen et al (4) found a significantly higher rate of spontaneous abortion for women in whom WSCM was used. This relationship might account for some of the difference. In addition, higher rates of hydrosalpinx in women receiving WSCM might account for the lower rates of live birth (8). Hydrosalpinx shown at HSG, laparoscopy, or both, did not occur more frequently in any of the contrast material groups in our study.

For all women in our study, the overall rate of pregnancy within 12 months after HSG was 30.6%. Intergroup differences in post-HSG pregnancy rates were not statistically significant (²₈ = 6.08, P = .64). The overall rate of live birth was 20.4%, which helps confirm the fact that approximately one-third of pregnancies were terminated—most due to spontaneous abortion (9). As in the study by Rasmussen et al (4), women who received WSCM had a higher rate of spontaneous abortion than did those who received OSCM. In our larger sample, however, this trend was not significant (²₂ = 2.37, P = .31).

The rate of spontaneous abortion was far lower overall among women aged 20–24 years than it was in older women. Among women aged 20–24 years, randomization resulted in assignment of slightly more women to receive WSCM than OSCM; similarly, randomization resulted in assignment of slightly more women aged 35–39 years to receive OSCM. These assignments might have affected each group's statistical chance for pregnancy and live birth, although the difference was not significant (²₁₂ = 18.6, P = .1).

The overall rate of ectopic pregnancy was 1.1% (seven of 666 patients). The rate of ectopic pregnancy after HSG in women without a history of ectopic pregnancy was 0.3% (two of 579). The rate of ectopic pregnancy in women with a prior ectopic pregnancy was significantly higher at 12.2% (Fisher exact test, P < .001). Women who had previously had more than one ectopic pregnancy had a more than 40-fold greater likelihood of another ectopic pregnancy (Table 3).

Oocyte degeneration has been cited as an explanation for declining rates of pregnancy and live birth after age 35 years (10,11). This effect is clearest in women after age 40 years, although the effect begins earlier. Our data helped confirm the lower rates of pregnancy and live birth among women aged 40–44 years who participated in our study.

In a large, multiinstitutional meta-analysis (2), data were pooled from several studies that had evaluated the potentially therapeutic role of OSCM. The authors of that study found that the difference in rates of live birth is limited to couples in whom the infertility was classified as unexplained. In our study, we classified infertility according to assumed causes that were based on independent classification by treating physicians. We made no attempt to validate or modify these classifications. Women whose infertility was classified as unexplained were not more likely to have higher rates of OSCM-related (ie, "therapeutic") live birth than other women (²₈ = 4.28, P = .83). The finding from the meta-analysis regarding HSG and unexplained infertility may have resulted from a lack of comparability and the variation in the complexity of diagnostic evaluation among the studies included in the meta-analysis. The value and limitations of meta-analyses have been acknowledged (12,13). Large-scale, prospective studies that have been randomized and controlled are preferable when possible, however.

We did not compare WSCM-enhanced and OSCM-enhanced HSG with nonenhanced HSG. Mackey et al (14) performed such a comparison but did so retrospectively. The current standard of care did not allow us to include a group undergoing nonenhanced HSG, inasmuch as HSG was performed to provide potentially useful morphologic information, as well.

In conclusion, despite using a sample size with statistical power sufficient to detect a difference between outcome (term pregnancy) rates of 10% and 20%, we found no significant difference in the proportions of pregnancies resulting in live births after the use of OSCM, WSCM, or both. We thus found no evidence to support the belief that specific contrast materials themselves increase rates of pregnancy and live birth. Our results validate the finding of a much higher risk of ectopic pregnancy in women with a history of ectopic pregnancy.

	Acknowledgments

 
The authors acknowledge the important contribution of the participating couples. The authors acknowledge special contributions of facility coordinators and liaisons from among more than 50 gynecologists and 30 radiologists participating in the Kaiser Permanente Medical Care Program Infertility Work Group (all in Calif): Sharon Abrams, MD (Martinez), Edward Bailey, MD (Santa Clara), Victoria Flavell, MD (Walnut Creek), Seth Feigenbaum, MD (San Francisco), Janie Hirata, MD (Oakland), Isaac Kaplan, MD (Vallejo), Gordon Manashil, MD (San Rafael), David Luke Pascoe, MD (Roseville), Adam Petras, MD (San Rafael), Paul Stevens, MD, (Richmond), and Virginia Weiss, MD (Santa Clara). Lystra Gonzalez, BA, ART, reviewed the medical charts; Sue Paulson, Marsha Bryant, CMA, and Laura Konitzer, PharmT, assisted with patient interviews. The medical editing department of the Kaiser Foundation Research Institute provided editorial assistance.

	Footnotes

 
Abbreviations: HSG = hysterosalpingography OSCM = oil-soluble contrast material WSCM = water-soluble contrast material

Author contributions: Guarantors of integrity of entire study, D.B.S., H.E.B.; study concepts, D.B.S., H.E.B., S.C.P.; study design, D.B.S., H.E.B.; definition of intellectual content, D.B.S., H.E.B., S.C.P.; literature research, D.B.S., H.E.B.; clinical studies, D.B.S., H.E.B., S.C.P.; data acquisition, D.B.S., H.E.B., S.C.P.; data analysis, D.B.S., H.E.B.; statistical analysis, H.E.B.; manuscript preparation and editing, D.B.S., H.E.B.; manuscript review, D.B.S., H.E.B., S.C.P.

	References

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Soules MR, Spadoni LR. Oil versus aqueous media for hysterosalpingography: a continuing debate based on many opinions and few facts. Fertil Steril 1982; 38:1-11.[Medline]
2. Watson A, Vandekerckhove P, Lilford R, Vail A, Brosens I, Hughes E. A meta-analysis of the therapeutic role of oil soluble contrast media at hysterosalpingography: a surprising result?. Fertil Steril 1994; 61:470-477.[Medline]
3. DeCherney AH, Kort H, Barney JB, DeVore GR. Increased pregnancy rate with oil-soluble hysterosalpingography dye. Fertil Steril 1980; 33:407-410.[Medline]
4. Rasmussen F, Lindequist S, Larsen C, Justesen P. Therapeutic effects of hysterosalpingography: oil- versus water-soluble contrast media—a randomized prospective study. Radiology 1991; 179:75-78.[Abstract/Free Full Text]
5. Fleiss JL. Statistical methods for rates and proportions New York, NY: Wiley, 1981.
6. Cohen J. Statistical power analysis for the behavioral sciences 2nd ed. Hillsdale, NJ: Erlbaum, 1990.
7. Collins JA. Unexplained infertility. In: Keye WR, Jr, Chang RJ, Rebar RW, Soules MR, eds. Infertility: evaluation and treatment. Philadelphia, Pa: Saunders, 1995; 249-262.
8. Spring DB, Barkan HE. Potential therapeutic effect of oil-soluble contrast media in hysterosalpingography (letter). Radiology 1992; 185:283-284.[Medline]
9. Wilcox AJ, Weinberg CR, O'Connor JF, et al. Incidence of early loss of pregnancy. N Engl J Med 1988; 319:189-194.[Abstract]
10. Lim AST, Tsakok MFH. Age-related decline in fertility: a link to degenerative oocytes. Fertil Steril 1997; 68:265-271.[Medline]
11. Berkowitz GS, Skovron ML, Lapinski RH, Berkowitz RL. Delayed childbearing and the outcome of pregnancy. N Engl J Med 1990; 332:659-664.
12. LeLorier J, Grégoire G, Benhaddad A, Lapierre J, Derderian F. Discrepancies between meta-analyses and subsequent large randomized, controlled trials. N Engl J Med 1997; 337:536-542.[Abstract/Free Full Text]
13. Bailar JC, III. The promise and problems of meta-analysis (editorial). N Engl J Med 1997; 337:559-561.[Free Full Text]
14. Mackey RA, Glass RH, Olson LE, Vaidya R. Pregnancy following hysterosalpingography with oil and water soluble dye. Fertil Steril 1971; 22:504-507.[Medline]

This article has been cited by other articles:

				L. Lindborg, J. Thorburn, C. Bergh, and A. Strandell Influence of HyCoSy on spontaneous pregnancy: a randomized controlled trial Hum. Reprod., May 1, 2009; 24(5): 1075 - 1079. [Abstract] [Full Text] [PDF]

				N. P. Johnson, W. E. Hadden, L. W. Chamley, D. B. Spring, H. E. Barkan, A. J. Yun, and P. Y. Lee Fertility Enhancement by Hysterosalpingography with Oil-Soluble Contrast Media: Reality Not Myth Am. J. Roentgenol., December 1, 2005; 185(6): 1654 - 1657. [Full Text] [PDF]

				A. J. Yun, D. B. Spring, and H. E. Barkan Enhanced Fertility After Diagnostic Hysterosalpingography: The Debate Continues Am. J. Roentgenol., January 1, 2005; 184(1): 347 - 348. [Full Text] [PDF]

				D. B. Spring and H. E. Barkan Enhanced Fertility After Diagnostic Hysterosalpingography May Be a Myth Am. J. Roentgenol., December 1, 2004; 183(6): 1728 - 1728. [Full Text] [PDF]

				J. J. W. Sandstede, T. Johnson, K. Harre, M. Beer, S. Hofmann, T. Pabst, W. Kenn, W. Voelker, S. Neubauer, and D. Hahn Cardiac Systolic Rotation and Contraction Before and After Valve Replacement for Aortic Stenosis: A Myocardial Tagging Study Using MR Imaging Am. J. Roentgenol., April 1, 2002; 178(4): 953 - 958. [Abstract] [Full Text] [PDF]

				A. L. Valentini, L. Muzii, R. Marana, G. F. Catalano, V. Summaria, F. Felici, A. Rossetti, and C. Destito Improvement of Hysterosalpingographic Accuracy in the Diagnosis of Peritubal Adhesions Am. J. Roentgenol., October 1, 2000; 175(4): 1173 - 1176. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Spring, D. B. Articles by Pruyn, S. C. Search for Related Content PubMed PubMed Citation Articles by Spring, D. B. Articles by Pruyn, S. C.