IVF Reads / HSG Test: What It Shows, What It Cannot, and What Happens Next

HSG Test: What It Shows, What It Cannot, and What Happens Next

Geeta
Written by GeetaPublished Updated
AI summary

A hysterosalpingogram (HSG) is an X-ray of the uterine cavity and fallopian tubes using contrast dye; NICE NG257 (2026) recommends it to screen for tubal blockage in women without known pelvic disease, and laparoscopy and dye for those with it. An open result makes blockage unlikely, but a proximal block is often not real: in a study cited by ASRM (2021), 24 of 40 women (60%) with a proximal block had open tubes on a repeat HSG about a month later (Dessole et al, 2000).

  • Against laparoscopy, HSG had pooled sensitivity 0.94 and specificity 0.92 for tubal occlusion in nine studies (582 women), against 0.95 and 0.93 for ultrasound-based HyCoSy (Maheux-Lacroix et al, Hum Reprod 2014).
  • In the H2Oil trial of 1,119 women, ongoing pregnancy within 6 months was 39.7% after oil-based contrast versus 29.1% after water-based (rate ratio 1.37, 95% CI 1.16-1.61); live births 38.8% versus 28.1% (Dreyer et al, NEJM 2017).
  • At 5 years in the same trial, cumulative live birth was 74.8% after oil versus 67.3% after water contrast (van Rijswijk et al, Fertil Steril 2020).
  • In the FOAM trial (1,026 women), mean pain was 5.4 out of 10 for HSG versus 3.1 for HyFoSy, with similar live birth rates after management based on either test (van Welie et al, Hum Reprod 2022).
  • A Cochrane review of 23 trials (1,272 women) found no evidence that oral non-opioid painkillers reduce pain during HSG; topical anaesthetic may help, on low-quality evidence (Hindocha et al, 2015).
  • NICE NG257 (2026) recommends laparoscopic salpingectomy or tubal occlusion for hydrosalpinges before IVF, and considering tubal catheterisation for proximal blockage after discussing IVF.

What does an HSG test show?

An HSG answers two questions: are the fallopian tubes open, and is the inside of the uterus a normal shape. It is an X-ray taken while contrast dye is passed through the cervix; if the tubes are open the dye spills out of their far ends, and if something sits in the cavity — a polyp, a fibroid bulging inward, a septum or scarring — it shows as a gap in the outline.

It is the test people often call the "dye test" for the uterus or tubes. NICE's 2026 fertility guideline (NG257) recommends it as the first tubal test for women with no known pelvic problems, because it is reliable for ruling out blockage and less invasive than laparoscopy.

What it cannot show matters as much:

  • Endometriosis. The dye only travels inside the uterus and tubes; endometriosis sits on the outside surfaces of pelvic organs.
  • The ovaries, cysts, egg reserve or whether you ovulate — those need ultrasound and blood tests.
  • Fibroids inside the muscle wall that do not push into the cavity.
  • Whether an open tube works well enough to pick up and carry an egg.

Waiting for a tube result can feel like waiting for a verdict on whether pregnancy is possible at all. It rarely is that: most results either reassure or lead to a further, more specific test.

How accurate is an HSG compared with laparoscopy and HyCoSy?

Laparoscopy with dye — keyhole surgery under general anaesthetic — is the reference standard. Against it, HSG and ultrasound-based tubal testing perform similarly. A 2014 meta-analysis (Maheux-Lacroix et al) pooled nine studies where women had all three tests: HSG correctly identified most blocked tubes and most open ones, with figures close to those for HyCoSy.

Those numbers hide an asymmetry that matters for the next decision:

  • An open result is trustworthy. ASRM's 2021 committee opinion says that if HSG suggests patent tubes, "tubal blockage is highly unlikely".
  • A blocked result near the uterus is often not real. The same document cites a study in which 24 of 40 women (60%) whose first HSG showed a proximal block had both tubes open on a second HSG about a month later (Dessole et al, 2000) — a small, single-centre study, but ASRM lists mucus plugs, debris and spasm of the tube opening as reasons a proximal block can be apparent rather than real.
  • Older pooled data are less flattering: a 1995 meta-analysis of 20 studies (4,179 women) estimated HSG sensitivity at 0.65 and specificity at 0.83 against laparoscopy (Swart et al).

NICE NG257 recommends going straight to laparoscopy and dye for anyone with a known risk — previous pelvic infection, an ectopic pregnancy, or endometriosis — so that tubes and other pelvic disease can be assessed together. Where the expertise exists, it accepts ultrasound tubal testing (HyCoSy) as an effective alternative to HSG; that is done as part of a pelvic sonography appointment, with no X-ray and a view of the ovaries at the same time.

HSG vs HyCoSy against laparoscopy (9 studies, 582 women)0.94 / 0.92 vs 0.95 / 0.93Pooled sensitivity / specificity for tubal occlusion, HSG versus hysterosalpingo-contrast sonography, in studies where women had both tests plus laparoscopy.Maheux-Lacroix S et al. Hysterosalpingosonography for diagnosing tubal occlusion in subfertile women: a systematic review with meta-analysis. Hum Reprod 2014;29(5):953-63.

Does an HSG hurt, and is there a painless HSG?

Most people feel period-like cramping, mainly while the dye stretches the uterus and spills from the tubes. In the FOAM trial, where 1,026 Dutch women had both tests, the average pain score was 5.4 out of 10 for HSG and 3.1 for HyFoSy, a foam-based ultrasound test. No version is reliably painless, but some are gentler.

What the evidence says about pain relief is narrower than common advice. A Cochrane review of 23 trials (1,272 women, Hindocha et al, 2015) found:

  • no evidence that oral non-opioid painkillers — the tablet often suggested an hour before — reduce pain during the procedure (low-quality evidence);
  • topical anaesthetic applied beforehand may reduce pain during the procedure (low-quality evidence);
  • intravenous opioids may help, weighed against side effects and a longer recovery.

So the useful questions to ask in advance are whether the clinic uses a local anaesthetic, whether HyCoSy or HyFoSy is available instead, and whether sedation is offered if a previous attempt had to be stopped. Taking a painkiller beforehand is common practice — the FOAM trial allowed paracetamol or naproxen — but the trials do not show it makes the test less painful.

When in the cycle is an HSG done?

After menstrual bleeding has stopped and before ovulation — the first half of the cycle. The FOAM trial protocol, for example, performed tests "within the two weeks of the follicular phase of the cycle after complete cessation of menstrual bleeding". The point is to avoid exposing a very early pregnancy to X-rays and dye. With irregular cycles, clinics usually rely on a pregnancy test rather than counting days.

NICE NG257 recommends screening for chlamydia before any instrumentation of the uterus, and considering preventive antibiotics if screening has not been done, because instruments passed through the cervix can carry an existing infection upward.

Does an HSG improve your chances of getting pregnant?

Possibly, and the contrast type matters. The H2Oil trial (Dreyer et al, New England Journal of Medicine 2017) randomised 1,119 women in 27 Dutch hospitals to oil-based or water-based contrast. Within 6 months, ongoing pregnancy was 39.7% with oil versus 29.1% with water, and live births 38.8% versus 28.1%. The likely mechanism is flushing debris out of the tubes.

The gap narrows but persists over time: after 5 years, cumulative ongoing pregnancy was 80.0% versus 75.0% and cumulative live birth 74.8% versus 67.3% (van Rijswijk et al, 2020).

Three cautions before treating this as a reason to choose oil:

  • The trial enrolled ovulating women aged 18 to 39 at low risk of tubal disease. It says nothing about women already suspected of blocked tubes.
  • Oil-based contrast carries far more iodine. A 2024 review by the European Society of Urogenital Radiology recommends testing thyroid function before an oil-based HSG and monitoring it for 6 months afterwards. It also notes that in five of nine trials the contrast used is no longer on the market.
  • Dye entering the blood vessels (intravasation) was reported in 2.7% of oil-based HSGs, pooled across 19,339 procedures, versus 2.0% with water-based contrast; 18 oil embolisms were recorded, none with lasting harm (Roest et al, 2021).
Ongoing pregnancy within 6 months, H2Oil trial39.7% vs 29.1%Oil-based versus water-based contrast at HSG: 220 of 554 versus 161 of 554 women; rate ratio 1.37 (95% CI 1.16-1.61). Live births 38.8% versus 28.1%.Dreyer K et al. Oil-Based or Water-Based Contrast for Hysterosalpingography in Infertile Women. N Engl J Med 2017;376(21):2043-2052.

What does an abnormal HSG result mean?

Reports use a small vocabulary:

  • Bilateral or free spill — both tubes open.
  • Unilateral spill — one tube open, one not.
  • Proximal (cornual) block — blocked where the tube joins the uterus; the result most likely to be a false alarm.
  • Distal block or hydrosalpinx — blocked at the far end, sometimes swollen with fluid.
  • Filling defect — something occupying the cavity, such as a polyp or a fibroid pushing inward.

A filling defect is usually confirmed by looking directly with a hysteroscopy, which can often remove a polyp in the same sitting. NICE NG257 advises hysteroscopy only when a uterine abnormality is suspected, which an HSG finding of this kind provides.

What happens after a blocked-tube result?

The next step depends on where the block is and on everything else in the infertility workup — age, ovarian reserve, the semen analysis. A blocked result does not automatically mean IVF.

  • Proximal block. Often repeated or confirmed first, because so many are not real. NICE NG257 says to consider opening the tube with a catheter under X-ray or hysteroscopic guidance, after discussing IVF as the alternative. In a meta-analysis cited by ASRM (2021), cumulative clinical pregnancy after tubal cannulation was 26.4% at 12 months, live birth 22%, and ectopic pregnancy 4%.
  • Hydrosalpinx. Fluid from a swollen tube can leak into the uterus. ASRM cites two meta-analyses in which pregnancy, implantation and delivery rates were about 50% lower with hydrosalpinges, and trials in which removing the tube restored IVF pregnancy rates. NICE NG257 recommends laparoscopic removal or blocking of the tube before IVF.
  • Mild tubal disease in a younger woman. NICE NG257 says tubal surgery can be considered where expertise exists; ASRM describes the ideal candidate as young, with no other significant fertility factor.
  • Both tubes severely damaged, or other factors present. IVF bypasses the tubes. ASRM (2021) says reparative surgery is not appropriate for severe disease or combined proximal and distal blockage, and prefers IVF for proximal blockage in older women or with significant male-factor infertility.

In India, genital tuberculosis is a cause worth naming. In a series of 87 women with confirmed genital TB at AIIMS, New Delhi (Sharma et al, 2022), HSG showed a blocked right tube in 69 of 87 (79%) and a blocked left tube in 67 of 87 (77%), while the uterine cavity looked normal in 49 of 87 (56%). The diagnosis in that study was made by microbiology, PCR, biopsy or laparoscopy — not from the HSG picture alone — and no woman had a flare of TB after the HSG. Treatment options for blocked tubes are covered in tubal blockages: diagnosis and treatment.

Proximal blocks open on a repeat HSG24 of 40 womenOf 40 women with a proximal tubal block on a first HSG, 24 (60%) had both tubes open on a second HSG about one month later, at a single university hospital. Cited by ASRM (2021).Dessole S et al. A second hysterosalpingography reduces the use of selective technique for treatment of a proximal tubal obstruction. Fertil Steril 2000;73(5):1037-9; ASRM Practice Committee, Fertil Steril 2021;115(5):1143-1150.

What are the risks, and when should you call the clinic?

  • Pelvic infection — the reason for chlamydia screening or antibiotics beforehand.
  • Reaction to iodine contrast — tell the team about any previous contrast or iodine reaction.
  • Fainting or feeling faint as the cervix is handled.
  • Thyroid changes after oil-based contrast, as above.
  • A small radiation dose, which is why the test is timed before ovulation.

Light spotting and cramping that settle over the day are expected. Fever or chills, bleeding heavier than a period, pain that gets worse instead of easing, or a foul-smelling discharge should be reported the same day.

What should you ask, and what comes next?

  1. Before: will it be oil- or water-based contrast, is a local anaesthetic used, and is HyCoSy or HyFoSy available instead?
  2. Before an oil-based HSG: should thyroid function be checked?
  3. After: did dye spill from both tubes, and was the cavity outline normal?
  4. If a block was seen: proximal or distal — and should a proximal one be repeated before any decision?
  5. Ask for the images as well as the report; a second opinion or a change of clinic then does not mean repeating the test.

Not sure what your HSG report means?

Upload your report and IVY will explain each finding in plain language, and tell you which questions are worth raising with your clinic.

Keep reading

13 Sources

  1. National Institute for Health and Care Excellence. Fertility problems: assessment and treatment. NICE guideline NG257, published 31 March 2026 (replaces CG156). Recommendations 1.18.13-1.18.16 (HSG, hysterosalpingo-contrast-ultrasonography, laparoscopy and dye, hysteroscopy) and 1.22.1-1.22.3 (chlamydia screening before uterine instrumentation). NICE
  2. National Institute for Health and Care Excellence. NICE guideline NG257 (2026), Management of female factor fertility problems: 1.34.1 tubal surgery, 1.34.2 tubal catheterisation, 1.34.3-1.34.4 hydrosalpinges before IVF. NICE
  3. Practice Committee of the American Society for Reproductive Medicine. Role of tubal surgery in the era of assisted reproductive technology: a committee opinion. Fertil Steril 2021;115(5):1143-1150. PMID 33642065. American Society for Reproductive Medicine
  4. Dessole S, Meloni GB, Capobianco G, et al. A second hysterosalpingography reduces the use of selective technique for treatment of a proximal tubal obstruction. Fertil Steril 2000;73(5):1037-9. PMID 10785234. 24 of 40 women with proximal obstruction patent on second HSG. Fertility and Sterility
  5. Dreyer K, van Rijswijk J, Mijatovic V, et al. Oil-Based or Water-Based Contrast for Hysterosalpingography in Infertile Women. N Engl J Med 2017;376(21):2043-2052. PMID 28520519. New England Journal of Medicine
  6. van Rijswijk J, van Welie N, Dreyer K, et al. Tubal flushing with oil-based or water-based contrast at hysterosalpingography for infertility: long-term reproductive outcomes of a randomized trial. Fertil Steril 2020;114(1):155-162. PMID 32553471. Fertility and Sterility
  7. Maheux-Lacroix S, Boutin A, Moore L, et al. Hysterosalpingosonography for diagnosing tubal occlusion in subfertile women: a systematic review with meta-analysis. Hum Reprod 2014;29(5):953-63. PMID 24578476. Human Reproduction
  8. Swart P, Mol BW, van der Veen F, et al. The accuracy of hysterosalpingography in the diagnosis of tubal pathology: a meta-analysis. Fertil Steril 1995;64(3):486-91. PMID 7641899. Fertility and Sterility
  9. van Welie N, van Rijswijk J, Dreyer K, et al. Can hysterosalpingo-foam sonography replace hysterosalpingography as first-choice tubal patency test? A randomized non-inferiority trial (FOAM). Hum Reprod 2022;37(5):969-979. PMID 35220432. Human Reproduction
  10. Hindocha A, Beere L, O'Flynn H, Watson A, Ahmad G. Pain relief in hysterosalpingography. Cochrane Database Syst Rev 2015;(9):CD006106. PMID 26387564. Cochrane Database of Systematic Reviews
  11. Geenen RWF, van der Molen AJ, Dekkers IA, et al. Contrast media for hysterosalpingography: systematic search and review providing new guidelines by the Contrast Media Safety Committee of the European Society of Urogenital Radiology. Eur Radiol 2024;34(10):6435-6443. PMID 38573340. European Radiology / ESUR
  12. Roest I, Rosielle K, van Welie N, et al. Safety of oil-based contrast medium for hysterosalpingography: a systematic review. Reprod Biomed Online 2021;42(6):1119-1129. PMID 33931367. Reproductive BioMedicine Online
  13. Sharma JB, Kumari S, Jaiswal P, et al. Hysterosalpingography Observations in Female Genital Tuberculosis with Infertility. J Hum Reprod Sci 2022;15(4):362-369. PMID 37033134. 87 women, AIIMS New Delhi. Journal of Human Reproductive Sciences

Frequently asked questions

Short answers to the questions people search for before and after an HSG.

Is there a painless HSG test?

Not reliably. Average pain in the FOAM trial was 5.4 out of 10 for HSG and 3.1 for HyFoSy, an ultrasound-based alternative. A Cochrane review found topical anaesthetic may reduce pain during HSG, while oral non-opioid painkillers showed no clear effect, so ask about local anaesthetic, HyFoSy or sedation in advance.

What does a normal HSG look like?

The report describes a smooth, triangular uterine cavity and dye spilling freely from both tubes, often written as bilateral free spill. ASRM (2021) notes that when HSG shows open tubes, blockage is highly unlikely, though it cannot confirm the tubes work normally.

Can an HSG unblock your tubes?

It does not open a truly blocked tube, but flushing may clear debris. In the H2Oil trial ongoing pregnancy within 6 months was 39.7% after oil-based contrast against 29.1% after water-based, in ovulating women at low risk of tubal disease. Proximal blocks are treated separately, with catheter techniques or IVF.

Can an HSG detect endometriosis, ovarian cysts or fibroids?

Not endometriosis or ovarian cysts: the dye never reaches the outer surfaces of the pelvic organs or the ovaries. It shows only fibroids that push into the uterine cavity. Ultrasound assesses the ovaries and uterine wall, and laparoscopy remains the way to diagnose endometriosis.

Is HyCoSy better than HSG?

For tubal accuracy they are similar against laparoscopy. HyCoSy uses ultrasound instead of X-rays, also images the ovaries, and was less painful in the FOAM trial, but it was inconclusive more often (9.5% versus 2.9% of 1,026 women). NICE accepts it where the expertise exists.

Can an HSG affect the thyroid?

Oil-based contrast contains a large iodine load. The European Society of Urogenital Radiology (2024) recommends checking thyroid function before an oil-based HSG and monitoring it for 6 months afterwards, which is worth raising if you already have a thyroid condition.