Tubal flushing with oil-or water-based contrast medium at hysterosalpingography for infertility: 3-year outcome of a randomized clinical trial. [Human Reproduction]

Abstract

Study question: What is the impact of the use of oil-or water-based contrast medium at hysterosalpingography (HSG) on fertility outcomes 3 years after randomization? Summary answer: Women who had their tubes flushed with oil-based con-trast had 3-year cumulative ongoing pregnancy rates of 83.2% versus 80.7% after the use of water-based contrast. What is known already: For decades it has been suggested that tubal flushing with oil-based contrast (Lipiodol) at HSG improved pregnancy, rates (Mohiyiddeen L, et al. 2015). We recently showed in a landmark trial published in the New England Journal of Medicine that in infertile women tubal flushing during HSG with oil-based contrast resulted in more ongoing pregnancies than HSG with the use of water-based contrast at 6 months after randomization(39.7% versus 29.1% (RR 1.37, 95%CI 1.16-1.61, p < 0.001)) (Dreyer et al. 2017). It is unclear whether this difference continues beyond 6 months. Study design, size, duration: This was a follow-up study of the H2Oil trial, a multi-center randomized controlled trial in the Netherlands in which 1,119 women were included and randomized to hysterosalpingography with oil-based contrast (n = 557) or water-based contrast (n = 562). Here, we present the fertility outcomes at three years after randomization. Participants/materials, setting, methods: All 1,119 participants were contacted 3-5 years after randomization, and were asked to fill in questionnaires about the fertility treatments and pregnancies. Additionally, we reviewed the medical files to record all pregnancies and fertility treatments. Primary outcome was the first ongoing pregnancy within 36 months after randomization. We also compared naturally conceived pregnancies, pregnancies conceived after IUI and IVF and time to ongoing pregnancy. Main results and the role of chance: In total, 383 ongoing pregnancies were accomplished in the oil-based contrast group versus 344 ongoing pregnancies in the water-based contrast group. The cumulative ongoing pregnancy rate 3 years after randomization was 83.2% (95%CI:79.3-87.1%) in the oil-based contrast group versus 80.7% (95%CI:76.4-85.0%) in the water-based contrast group. The Kaplan-Meier survival curves for time to ongoing pregnancy were significantly different in favor of the oil-based contrast group (median time to ongoing pregnancy 10.00 versus 14.83 months, p-value log-rank test 0.001). Of the 383 ongoing pregnancies in the oil-based contrast group, 207 (54.3%) were naturally conceived, 114 (29.8%) after IUI and 60 (15.7%) after IVF/ICSI. Of the 344 ongoing pregnancies in the water-based contrast group, 174 (51.2 %) were naturally conceived, 93 (27.0%) after IUI and 73 (21.2%) after IVF/ICSI. Conception of 2 versus 4 pregnancies was unknown. When follow-up is censored for starting IVF/ICSI treatment, the cumulative ongoing pregnancy rate was 74.2% (95%CI:68.1-80.3%) in the oil-based contrast group versus 70.1% (95%CI:63.0-77.2%) in the water-based contrast group, with the time to ongoing pregnancy significantly shorter in favor of the oil-based contrast group (median time to ongoing pregnancy 10.16 versus 17.07 months, p-value log-rank test <0.001). Limitations, reasons for caution: This study was limited to women at low risk for tubal pathology, <39 years and without known endocrinological dis-eases, the findings should therefore not be generalized. Here, we only report ongoing pregnancies and not live birth. Wider implications of the findings: Tubal flushing with oil-based contrast does not only result in higher 6-months ongoing pregnancy rates, but also in higher 3-year cumulative ongoing pregnancy rates. Time to ongoing pregnancy was significantly shorter in the oil-based contrast group. In our opinion, tubal flushing with oil-based contrast should be offered to infertile women.