FAQs

 
E1 INTERGROWTH 21st
The Perinatal Institute gets occasional enquiries about the INTERGROWTH 21st standards for birthweight and fetal weight. We advise against their use as they can cause harm, because they
  • underestimate the diagnosis of SGA, with only about 4-5% of the maternity population identified as <10th centile - as a result of which they miss significant proportions of at-risk fetuses and infants; and
  • over-estimate the incidence of LGA, with about 20% of pregnancies considered to have a baby >90th centile, which can result in maternal anxiety and unnecessary interventions.

Potential users in the UK ought to be aware that Intergrowth is based on data from 8 countries, with only about 15% from the UK. The other countries were Brazil, China, India, Italy, Kenya, Oman, and USA. Worrying evidence has been emerging about INTERGROWTH from researchers in these countries, even when comparisons were made with conventional, population based (i.e. un-customised) fetal or neonatal standards. Here are some examples, with link to publication and a key quote:

  • UKPoon etal, 2016: The INTERGROWTH-21st standard underestimates the proportion of SGA live births and stillbirths in our population.
  • Brazil Da Matta et al, 2024: The INTERGROWTH standard was found to be inadequate to classify small-for-gestational-age infants
  • China Zhu et al, 2019: Adverse perinatal outcomes were more accurately predicted when EFW [fetal weight] was measured by the Hadlock standard than by INTERGROWTH-21
  • China Zhang et al, 2023: Adoption of Intergrowth could result in misclassifications
  • India (All India Institute of Medical Sciences) Anand et al 2022: Intergrowth 21st runs the risk of over-diagnosing SGA neonates who may not be at a higher risk of short-term morbidities.
  • Italy Stampaliya et al, 2020:Immediate application into clinical practice of international growth charts might result in an under-diagnosis of small for gestational age fetuses and, especially, in an over-diagnosis of large for gestational age fetuses with major consequences for clinical practice.
  • Africa - we found no publications from Kenya, but there was a meta-analysis of malaria studies from Tanzania and Malawi Mtove et al, 2022: The prevalence of SGA may be overestimated and the impact of malaria in pregnancy underestimated when using Intergrowth-21.
  • US (Washington State) Fay et al, 2021: GROW identified more small for gestational age infants with adverse outcomes than INTERGROWTH-21st, including more stillbirths, perinatal deaths, low Apgar scores, glucose instability, newborn seizure, and transfers to a higher level of care.

There are also recent studies from other countries – e.g. Australia, Canada, Denmark, Norway, Lithuania, Poland and Euro-Peristat (15 countries) - debunking the idea of a one size-fits-all standard, and/or reporting problems with the INTERGROWTH standard within their maternity populations /datasets.

We are not aware of any published evidence that either the fetal weight or the birthweight standards by INTERGROWTH have shown any significant benefit in terms of identification of adverse pregnancy outcome. In contrast, GROW charts have been validated in a number of studies including a 2018 review in AJOG. We also keep an on-line, regularly updated list of studies with head-to-head comparisons between Intergrowth and customised GROW charts, none of which favour Intergrowth.

A summary document (pdf) with an evidence based appraisal of INTERGROWTH vs GROW in the NHS is available here.

E2 K2 / Athena Maternity System ‘Customised Charts’
We have been asked by NHS users of K2 Maternity Systems to look at the charts incorporated by K2 into their Athena Maternity System, which they claim to be equivalent to the Perinatal Institute’s customised charts. Our GAP and statistical teams have undertaken an analysis of these charts. They have concluded that these charts cannot be recommended for clinical use as they are untested and unvalidated, and are likely to produce the wrong cut-offs for the identification of babies at risk, with potentially severe clinical consequences. For a detailed analysis, please see here.
E3 We have implemented GROW and use customised centiles to assess whether the birthweight is SGA. However our neonatologists use the WHO standard to screen for SGA as a risk factor for hypoglycaemia. The two assessments are often out of sync. Which one should we use?
The UK-WHO standard, as used in the parent held Red Book for neonates, is an internationally derived population reference which does not adjust for individual constitutional factors that affect the normal range of birthweight, as recommended by the RCOG.

The resultant discrepancies and confusion can have negative implications for parents, professionals and the quality of care. We examined the evidence and set out the argument for standardised assessment of birthweight in paper which we submitted in May 2019 to the British Association of Perinatal Medicine (BAPM) with a request to review their recommendations.
E4 What standard for birthweight should be used for babies assessed in the Perinatal Mortality Review Tool (PMRT)?
MBRRACE have recently recommended the use of their TIMMS birthweight centile calculator based on the birthweight curves by Norris et al [1]. We examined this standard and advise against its use as 1. it is not customised, contrary to RCOG guidelines [2], and 2. it applies a neonatal (instead of fetal) weight standard to preterm births which results in substantial underestimation of SGA at preterm gestations. For further details please see our analysis here: https://bit.ly/2KV1fyM

As a general rule, it is also important that the same standard is used to assess adverse outcome as that which was applied in antenatal care. For units using GROW for assessment of fetal growth and birthweight, the same standard should be used when assessing outcome. The birthweight centile can be either calculated postnatally in the GROW App, or with the customised centile calculator https://icc.growservice.org/754731/

    1. Norris T, Seaton SE, Manktelow BN, Baker PN, Kurinczuk JJ, Field D, et al. Updated birth weight centiles for England and Wales. Arch Dis Ch - Fetal & Neonatal 2018;103(6):F577–82.
    2. Royal College of Obstetricians and Gynaecologists. The investigation and management of the small-for-gestational-age fetus. Green Top Guideline No 31 - RCOG 2013.
E5 GAP effectiveness in Scotland
Stamatina Iliodromiti, Gordon Smith et al question the usefulness of the GAP program on the basis that stillbirth rates dropped in Scotland before GAP was introduced there: https://doi.org/10.1002/uog.21999

This is a good example of how statistics can be used to fit preconceptions. The research methodology is misleading in several ways. For example, the authors curtailed the period of study to exclude the years when Scottish stillbirth rates rose again, making the apparent decline statistically insignificant. They also did not present subsequent data which showed that stillbirths fell again after the GAP program had been introduced and had time to embed in Scotland. There were other flaws in methodology and conclusions which are detailed in our response: https://doi.org/10.1002/uog.22100         
E6 The DESIGN Trial

The DESiGN study was a London based randomised controlled trial investigating the effectiveness of the GAP program in 5 NHS Trusts. The results have recently been published in PLOS Medicine and have been circulated with the headline that ‘GAP has had no effect on antenatal detection of SGA’.

We believe that this conclusion is incorrect, and that the headline should instead reflect the most important finding of the study: the significant reduction in stillbirths in the GAP units in the trial - see our PLOS Medicine response. This is also the conclusion of Professor Lesley McCowan, one of the co-authors of the study, in her own published response, in which she estimates that this could translate to 400 fewer stillbirths a year in the UK.

There were various problems with the study, any of which could explain why the authors found no improvement in SGA detection rates between the two study arms. These included that:

  • Implementation of GAP was incomplete and did not follow the minimal requirements agreed before the trial.
  • The evaluation period stopped much too early, not allowing the new practice to embed.
  • Practice in the control arm (which should stay the same during an RCT) changed substantially following implementation of the SBL Care Bundle v1, with ultrasound scan rates increasing by 73%, while those in GAP units remained the same.
  • While there was extensive imputation required to address missingness of data, no information was provided on accuracy and level of ascertainment of ultrasound scan data. Missing scan reports, or missing information to obtain or calculate the EFW value within a report, can of course affect the accuracy of detection rates. This problem is described in more detail in item 1. of our 25.7.2022 response to the paper.

Additional comments about the trial have been posted on the same PLOS Medicine website:
https://journals.plos.org/plosmedicine/article/comments?id=10.1371/journal.pmed.1004004

See also thread summarising our response to the DESiGN Trial on Twitter (@PerinatalUK).

Recording of the lunchtime Lecture and Q&A about the Perinatal Institute’s response to the DESiGN Trial, held on 13 July 2022, is available here.

E7 French cluster RCT

Serial plotting of symphysis–fundal height and estimated fetal weight to improve the antenatal detection of infants small for gestational age: A cluster randomised trial. Ego et al, 2023 https://doi.org/10.1111/1471-0528.17399

This cluster randomised trial from France was published recently with the conclusion that ‘serial plotting of SFH and EFW measurements on customised growth charts did not improve the antenatal detection of FGR among SGA births’.

We will formally comment in relation to the external validity and relevance to the GAP programme. In the meantime, here are some preliminary observations:

  1. Problems with compliance. Started with 5 clusters in each arm; in intervention arm, one cluster dropped out as not adhering to protocol; others had varying uptake, incl. one as low as 41.8%. There is also wide variation in the number of charts which had the minimum number of measurements recorded.
  2. The standard used for measuring the fundal height in this study was different to the one used in GROW, and not customised. It is based on ‘Fournié’s rule’ (DOI: 10.1016/j.jgyn.2007.01.011) which is described as ‘an increasing SFH of +1 cm/week from 18 to 28 cm between 22 and 32 weeks of gestation, and +1  cm/2 weeks from 28 to 33 cm, between 32 and 42 weeks of gestation. (see SFH chart in Supplementary File S1).
  3. Antenatal SGA detection was calculated using a one-size fits all, population based birthweight standard. A customised antenatal standard is attuned to detect customised SGA birthweight.
  4. Detection rate was 40.0% (intervention) vs 37.1% (control arm). However the control group’s detection rate increased from 21.7% before the trial (https://doi.org/10.1111/1471-0528.13148), based on similar referral indications.
  5. The control arm of the study also had a different SGA rate, preterm birth rate and different gestational age at first antenatal visit compared to the intervention arm.

Some of these points are similar to those we made in response to the DESIGN trial in London (see FAQ E6) and highlight the challenges of assessing antenatal growth surveillance programmes through RCTs, especially if there is limited compliance and/or a steady performance in the control arm cannot be achieved.

In this regard, longitudinal studies assessing whether detection rates have improved following implementation are more informative and reliable. Two recent ones applying GROW charts and GAP principles, analysed independently but not referenced in the French RCT paper, are https://doi.org/10.1111/ajo.12902 and https://doi.org/10.1111/ajo.13283.