Late pregnancy ultrasound: what’s the evidence?

Is there evidence that late pregnancy ultrasound is effective at picking up babies with problems?

And what are the costs and potential harms of this intervention, to women and babies?

In this blog post, I’m going to explain some of the recent studies on this topic, to answer those questions and help you decide whether late pregnancy scanning is right for you.

The IRIS Study

Let’s start with the landmark study on this topic; the IRIS study.

This was a randomised controlled trial carried out in The Netherlands by Henrichs et al (2019) and published in the British Medical Journal.

The researchers recruited 13,046 women with low-risk singleton pregnancies across 60 midwifery practices in the Netherlands. They randomised practices so that some offered two routine biometry scans at 28–30 and 34–36 weeks, while others carried on with usual care. Usual care meant that women would be offered fundal height measurement, with an ultrasound scan offered only if there was a clinical indication for this. For example, if the fundal height measurement suggested that the baby was very small or large.

The researchers found that routine third-trimester ultrasound was associated with higher antenatal detection of small-for-gestational-age fetuses, but not with a reduced incidence of severe adverse perinatal outcomes compared with usual care alone.

This distinction is important, and will be a theme in this blog post. It’s not uncommon to find that, while an intervention does detect more problems, a higher rate of detection doesn’t lead to better outcomes.

And the take home message from the IRIS study?

The findings do not support routine ultrasonography in the third trimester for low-risk pregnancies.

The Oxford Prospective Cohort Study

Another prospective study, this time a cohort study, was carried out in the UK (Aderoba et al 2023).

Maternity services in Oxfordshire introduced the offer of a routine scan at 35-36 weeks, and evaluated this.

At first glance, the findings look positive. Researchers announced that routine scans decreased extended perinatal deaths and severe morbidity (illness).

However, neither of these results were statistically significant.

Another more recent study at the same hospital (D’Alberti et al 2026) found that the majority of severe clinical adverse outcomes occurred in pregnancies classified at the 36-week scan as “normal” for gestational age.

This shows that, even when universal scanning is offered, the scan cannot reliably rule out the risk of adverse outcomes at term.

The Systematic Review

A systematic review by Caradeux et al (2024) found that routine third-trimester ultrasounds significantly improve detection rates of intrauterine growth retardation (IUGR) compared with symphysis-fundal height measurement, but that direct high-quality evidence supporting enhanced perinatal outcomes from routine scanning is lacking.

This is the same issue that I highlighted above, when discussing the IRIS study.

Yes, late pregnancy scanning finds more small (SGA) babies. But whether finding them improves the outcomes for those babies is unproven.

And, again, we see that late pregnancy ultrasound scans lead to more intervention.

The Review of Guidelines

Emam et al (2025) carried out a thorough review of all major international guidelines on late pregnancy ultrasound.

They found substantial variability in recommendations, which is often the sign that the evidence on a topic isn’t clear, or is debatable. That’s because, if the evidence were clear and uncontested, different countries would tend to make the same recommendations.

All six guidelines (ISUOG 2024, BMUS, ACOG, ASUM, CAR, and SIEOG) indicated that their writers felt that late pregnancy scans had clinical value in assessing fetal growth, amniotic fluid, and placental assessment. However, the evidence was variable, and there was no uniform consensus on whether late pregnancy scans should be routinely offered, or on when they should be performed.

Only two guidelines (ISUOG and SIEOG) specify the gestational age at which a late pregnancy scan should be performed (32–36 weeks).

Again, this study looked at that the guidelines say in the different countries, which is not necessarily the same as what the evidence says. Guidelines are often based on the opinions of doctors, rather than robust research evidence.

ISOUG Late Pregnancy Ultrasound Practice Guidelines

The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) is “…a scientific organization that encourages sound clinical practice and high-quality teaching and research related to diagnostic imaging in women’s healthcare.” (Khalil et al 2024).

This is a remarkable and refreshingly honest admission from the organisation most strongly advocating for ultrasound scans.

The Screening for LGA Analysis

Another recent study looked specifically at whether late pregnancy ultrasound was a useful method of screening for large for gestational age (LGA) babies (Lopian et al 2026).

Researchers set out to evaluate the diagnostic accuracy of routine third-trimester ultrasound for detecting large-for-gestational-age and macrosomic (over 4kg) neonates at term, and the impact of inaccurate prenatal diagnosis on obstetric management and perinatal outcomes.

The question of impact is an important one, because many women experience stress, anxiety, and induction after being told their baby is large. This is particularly problematic because most of them will go on to give birth to an average sized baby. More on large babies here.

The Australian ‘Real World’ Analysis

An Australian study assessed changes in obstetric and neonatal outcomes and service delivery following implementation of routine third trimester ultrasound (Kumar et al 2026).

The researchers analysed data from more than 9000 woman/babies and the findings showed that introducing routine late-pregnancy ultrasound (at 35–37 weeks) increased intervention, especially induction of labour, without improving important outcomes for babies (Kumar et al 2026).

Despite a huge increase in scanning — from 34% to 81% — the study results showed:

• no reduction in stillbirth
• no reduction in HIE
• no reduction in low Apgar scores
• no reduction in shoulder dystocia
• no improvement in composite neonatal outcomes (Kumar et al 2026).

Birth interventions, however, did rise, while the rate of vaginal birth fell.

After the introduction of routine scans:

• Induction rose from 33.5% to 38.2%
• Emergency caesareans rose from 14.4% to 18.0%
• Total caesareans rose from 30.1% to 33.9%
• Vaginal birth fell from 69.8% to 66.0% (Kumar et al 2026).

The Policy Debate

I feel it would be unhelpful to share the evidence on this topic without also mentioning the policy debate that is ongoing in medical and obstetric journals. By ‘policy debate’, I mean the conversations that obstetricians are having with each other about what women should be offered.

A recent example was published in the Australian and New Zealand Journal of Obstetrics and Gynaecology (Buntain et al 2026).

In a nutshell, the authors argue for wider adoption of the routine 36-week scan, very much taking the “pro” side of the debate.

They acknowledge concerns about cost-effectiveness and provoking unnecessary interventions, but argue these must be weighed against potentially improved perinatal outcomes.

However, as I have explained thus far, the evidence doesn’t show that routine late pregnancy scanning leads to improved perinatal outcomes.

This is another example of the obstetric, ‘but we’re gonna do it anyway’ approach.

Why the tension?

In fact, one of the papers I mentioned above indirectly reveals the tension within modern obstetrics, when the authors note that, “National Australian health policies aiming to reduce stillbirth … whilst simultaneously reducing late preterm and early term births … frequently present a clinical dilemma…”  (Kumar et al 2026).

This pressure on clinicians to detect possible risk and intervene in an attempt to prevent stillbirth isn’t limited to Australia. It’s clear in other countries — including the UK — that stillbirth reduction programmes are playing a part in increasing interventions, and causing other harms to woman and babies.

This may explain why obstetricians and hospitals are recommending late pregnancy ultrasound, despite the fact that there is no good evidence to support this recommendation.

Late pregnancy ultrasound: the summary…

The most rigorous RCT evidence (IRIS study) does not support routine third-trimester scanning in low-risk pregnancies to improve perinatal outcomes.

Even ISUOG’s own 2024 guidelines explicitly acknowledge the lack of convincing evidence for a universal low-risk scan.

The debate has shifted somewhat toward arguing that late scans (35–36 weeks) detect SGA better than fundal height measurement — but detection of SGA and improvement in outcomes are not the same thing, and that link remains unproven.

The most recent ‘real world’ evidence shows that scans don’t improve outcomes for babies, and they increase induction and caesarean rates, and lead to fewer women having vaginal births.

For more information on making the decisions that are right for you:

What’s Right For Me?

And if you’d like more information about what’s behind the increased push for induction and other interventions in late pregnancy, as well as an explanation of what the evidence actually says about this;

In Your Own Time

---

---

---

Books to help you become informed and make the decisions that are right for you…

---

---