Fetal growth surveillance – Current guidelines, practices and challenges
Show less
Mandy Williams, Sue Turner, Emily Butler, Jason Gardosi First Published March 22, 2018 Review Article
https://doi.org/10.1177/1742271X18760657
Article information
Article has an altmetric score of 4 Free Access
Abstract
Antenatal surveillance of fetal growth is an essential part of good maternity care, as lack of detection of fetal growth restriction is directly associated with stillbirth and perinatal morbidity. New algorithms and guidelines provide care pathways which rely on regular third trimester ultrasound biometry and plotting of estimated fetal weight in pregnancies considered to be at increased risk, and their implementation has increased pressures on ultrasound resources. Customised growth charts have improved the distinction between constitutional and pathological smallness and reduced unnecessary referrals. Their introduction, together with clinicians' training, e-learning and audit as the key elements of the growth assessment protocol, has resulted in increased antenatal detection of small for gestational age babies and a reduction in avoidable stillbirths. However, missed case audits highlight that further improvements are needed, and point to the need to address quality assurance and resource issues in ultrasound services.
Keywords Estimated fetal weight, small for gestational age, fetal growth restriction, stillbirth
Fetal growth restriction and stillbirth
There has been a recent focus on stillbirth prevention, stimulated by the fact that stillbirth rates in the UK were among the highest in Western Europe and had not changed substantially over the last 20 years.1 The situation was not helped by the fact that the majority of stillbirths have until recently been classified as ‘unexplained’, which was taken to suggest that these deaths were unavoidable.2,3
However, independent case reviews into the care of pregnancies that ended with a stillbirth found that many were associated with substandard care and therefore potentially avoidable.4 The largest category were babies with growth restriction which had not been recognised antenatally. Many of these mothers never had a growth scan, and never even came to any special attention of the maternity services, until presenting one day – often late in pregnancy – with absent fetal movements.
These observations were reinforced by a better classification system which included a category for ‘fetal growth restriction’ (FGR), defined as small for gestational age (SGA) according to customised centiles. FGR was found to be the largest category, accounting for over half of all normally formed stillbirths.3 Subsequent population-based analysis confirmed that pregnancies with an SGA fetus had a seven-fold increased risk for stillbirth, and that most of them were not recognised antenatally.5 The report also found that antenatal recognition of FGR can halve stillbirth risk through earlier delivery (Figure 1).
figure
Figure 1. Stillbirth risk in pregnancies with and without antenatal (A/N) detection of fetal growth restriction (FGR), defined as < 10th customised centile. Illustration based on data in Gardosi et al.5
Risk factors and algorithms for surveillance
The Royal College of Obstetricians and Gynaecologists' guidelines6 for surveillance and management of pregnancies with an SGA fetal outlined a standard of care that included serial measurement and plotting of fundal height and ultrasound biometry in low- and high-risk pregnancies, respectively. Customised charts (which adjust growth curves according to maternal height, weight, parity and ethnic origin) were recommended for plotting both fundal height and estimated fetal weight (EFW).6
The RCOG guidelines reviewed the risk factors associated with FGR and developed an algorithm which set out the recommended assessment of early pregnancy risk factors and care pathways contingent on low risk (→ serial measurement and plotting of fundal height) and increased risk (→ serial measurement and plotting of estimated fetal weight). Risk factors were divided into major and minor, with major risk factors requiring serial scans, and three minor prompting early pregnancy assessment by uterine Doppler to establish whether risk was major. Many units found this protocol too complicated for everyday use. Therefore, we assisted NHS England, as part of their Saving Babies' Lives Care Bundle initiative to develop a simplified algorithm which identified a list of risk factors, the presence of any of which would be an indication for serial ultrasound biometry.7 The algorithm was adapted for the national GAP programme (see below) by including use of customised growth charts (Figure 2).
figure
Figure 2. The growth assessment protocol (GAP) algorithm for fetal growth surveillance. Adapted from NHS England.7
As most instances of FGR are ‘late onset’ in the third trimester (usually from 32 weeks),5,8 the algorithm recommends surveillance until delivery. In high-risk pregnancy, a three-weekly regime would result in 4.5 third trimester scans – at 28, 31, 34, 37 and 40 weeks (the 40-week scan being needed for only half of all pregnancies still ongoing at that gestation).
Resource implications for guideline based surveillance
For many units, a schedule of 4.5 scans for all high-risk pregnancies is challenging and difficult to sustain within current resources. A West Midlands audit of ‘current practice’ in scan regimes afforded to pregnancies with a past history of SGA birth – a clinically undisputed indication for serial scans in subsequent pregnancies – found that the median number of scans performed was between 2 and 3 only, and 21% of these high-risk pregnancies had no third trimester scan at all.9 The audit highlighted the importance of scan frequency: a one scan policy (usually at 32 weeks) or two scans (usually at 28 and 34 weeks) resulted in a detection rate of around 30%, which was little better than doing no scans at all.
We examined the proportion of the NHS population in the West Midlands that would be designated at increased risk requiring serial scans. According to the RCOG guidelines,6 25.5% of mothers would have either a major risk factor (20.6%) or 3 or more minor risk factors for SGA (4.9%).10 The figure is higher – 36% – when applying the categories in the NHSE algorithm11 (Table 1), mainly because all smokers, including those having <10 cigarettes per day, are included as high risk. In practice, many units struggle to provide scans for all patients with these indications. Typically heavy smokers, or any smokers, and mothers with a raised BMI do not tend to receive serial scans.
Table
Table 1. Prevalence of risk factors requiring serial ultrasound assessment of fetal growth according to new NHS England algorithm; West Midlands, N = 146,774
Table 1. Prevalence of risk factors requiring serial ultrasound assessment of fetal growth according to new NHS England algorithm; West Midlands, N = 146,774
View larger version
Cost–benefit analysis
We assessed the cost benefit of a serial scan regime for increased risk pregnancies10 according to RCOG defined risk factors.6 An estimated 25% of the population at higher risk of SGA translated to 600–700 additional scans per 1000 births, depending on current scan provision.10 A growth scan provided additionally to an already existing scanning service and infrastructure was estimated to cost £15 which, multiplied by the additional scan requirement, represented a cost of £10 per pregnancy, or £30,000 for a maternity unit with 3000 deliveries per annum. Against this, improved antenatal detection of IUGR has many benefits, including reduction in stillbirths (estimating 1 stillbirth saved per 1000 deliveries based on previous experience12), as well as reduced neonatal admissions, perinatal morbidity, cerebral palsy and litigation. We estimated that together, these will result in savings of at least £120 per pregnancy or £360,000 for a unit with 3000 births, representing a 12-fold return on investment.10 However, the value of benefits was likely to be gross under-estimates in the absence of good data. In the case of stillbirths, they were confined to £5000 per death, mainly for costs of investigations; however, a more recent, detailed attempt in Australia13 to estimate stillbirths' direct costs to the health service (in terms of investigations and counselling) as well as indirect costs including absenteeism and loss of productivity, resulted in an estimate of £36,000 per death.13 Quite apart from the immeasurable loss of a human life, savings recognised in this way would vastly exceed any investment required to improve ultrasound services in step with national guidelines for fetal growth surveillance.
Accuracy of growth scans at term
The NHS England algorithm prescribes scans in increased risk pregnancies to be continued until delivery. The argument for this approach is strengthened by findings of a study where a prospective, routinely collected EFW at ‘last scan’ – performed for any indication – was assessed for its ability to predict SGA at birth14 (Table 2). The data are analysed at the three most common gestational age points for a ‘routine’ third trimester scan – 34, 35 or 36 weeks. As shown in Table 2, pregnancies that had scans for any reason were at increased risk of SGA at delivery, with rates ranging from 14 to 19%. However, detection rates by EFW were overall poor, ranging from 19.0% at 34 weeks up to only 36.1% at 36 weeks. This may indicate poor scan technique or limitations of fetal weight estimation by ultrasound, but is more likely to be associated with slowing of growth towards an SGA birthweight late in third trimester, which is missed if the last scan is done too early.
Table
Table 2. Effectiveness of ultrasound biometry at 34–36 weeks in the detection of SGA at birth
Table 2. Effectiveness of ultrasound biometry at 34–36 weeks in the detection of SGA at birth
View larger version
When protocols for scanning up to delivery were first introduced, many ultrasound departments were reluctant because of concerns that scans at term are less inaccurate, although there seemed little evidence to support such a belief. We therefore investigated the accuracy of preterm and term scans, based on routine measurements collected from all 19 West Midlands regional maternity units.15 The cohort consisted of 2296 pregnancies where an ultrasound scan was performed within three days of delivery, and this included 606 preterm (<37 weeks) and 1690 term deliveries. EFW was calculated by the Hadlock formula programmed into most units’ ultrasound equipment. The assumed weight gain during the 1, 2 or 3 day delay between scan measurement and birth was adjusted using the previously described ‘proportionality formula’.16 The results showed that EFWs at term were at least as good, and in fact marginally better than scans done in the preterm period, with 73% of EFWs falling within a ±10% margin of error (Table 3).
Table
Table 3. Accuracy of ultrasound scan at preterm vs. term gestational age, with scan performed within three days before delivery preterm (n = 606) or term (n = 1690).15
Table 3. Accuracy of ultrasound scan at preterm vs. term gestational age, with scan performed within three days before delivery preterm (n = 606) or term (n = 1690).15
View larger version
Practitioners' concern about obtaining accurate fetal head measurement when it was fully engaged at term could be addressed by EFW formulae such as that by Hadlock,17 which relies on abdominal circumference (AC) and femur length (FL) only.
The growth assessment protocol: A comprehensive programme for surveillance
The case reviews of adverse outcomes and epidemiological analysis provided the rationale and urgent need to focus on antenatal detection of FGR, and the national guidelines provided the pathway to achieve this within routine antenatal care. We integrated these elements into the growth assessment protocol (GAP), a comprehensive programme including accreditation training for midwives, obstetricians and ultrasonographers, use of customised growth charts, birthweight centiles and audit. It included a recommended surveillance pathway which was broadly based on the NHS England algorithm,7 but included use of customised charts according to RCOG guidelines6 (Figure 2).
The emphasis is on longitudinal assessment, preferably by the same practitioner. In low-risk pregnancy, the main method of surveillance is serial fundal height measurement, with protocols prompting immediate referral when the first measurement is below the 10th customised centile, or the slope of sequential measurement does not follow the growth curves on the individually customised chart. Trained midwives can make autonomous referrals for scans if they have concerns about growth, without having their fundal height measurements ‘checked’ by other care provider(s).
Identification of increased risk pregnancy is contingent on primary care surveillance and referral, as well as risk assessment at the beginning of pregnancy to identify mothers who need serial scans. In either case, the required measure is the EFW plotted on a customised growth chart.
Customised growth charts
Underpinning the low and increased risk pathways (Figure 2) is the plotting of fundal height (FH) and EFW measurements on customised charts, as recommended by RCOG Green Top guidelines.6 A controlled study has shown that a standardised approach significantly increases detection rates while reducing unnecessary referrals for ultrasound scans.18 Customised charts also improve EFW assessment and reduce false positive diagnosis of SGA.19
The advantages of customised GROW (‘gestation related optimal weight’) charts become most apparent when looking at subgroups within the population, e.g. according to maternal size. There is better correlation between SGA defined by customised centiles and perinatal death than any population based, one-size-fits-all standard.20
GROW charts also improve the definition of SGA in multi-ethnic populations21: in a cohort of 10,405 South Asian mothers from India, Pakistan and Bangladesh, who had a third trimester scan, 1554 (14.9%) were SGA according to the Hadlock fetal weight curve; however, 864 of these (55.6%) were not SGA if the growth curves were customised according to the mother's characteristics (height, weight, parity and ethnic origin). This effect is illustrated in Figure 3(a) and (b). Significantly, these cases re-classified as not SGA had the same perinatal mortality risk as the general, non-SGA population.21
figure
Figure 3. (a) and (b) Examples of customised charts for British-European (a) and South Asian (b) mothers. A sample estimated fetal weight (EFW) of 2500 g at 37.0 weeks is plotted on each chart to illustrate different results and clinical implications; 56% of SGA EFWs in pregnancies of South Asian mothers are not SGA if plotted on their own customised charts, and are not associated with increased perinatal mortality risk.21
Recently, INTERGROWTH 21st has been promoted as an internationally applicable, one-size-fits all standard for birthweight22 and fetal weight.23 It was derived from low risk, well-nourished mothers in eight countries and was therefore meant to provide a prescriptive optimal standard for all pregnancies. However, this concept has been challenged on theoretical grounds,24 and when applied to detailed multi-ethnic maternity datasets, it performed poorly compared to a customised standard.25 In a recent study using a 10-country cohort of 1.25 million births, INTERGROWTH 21st showed unrepresentative rates of SGA (4.4%) and LGA (20.6%), and poor correlation with outcome.26
Plotting estimated fetal weight versus abdominal circumference measurement
Many ultrasonographers and clinicians are accustomed to plotting individual parameters like HC, AC and FL rather than EFW. While these need of course to be recorded, there are a number of reasons why calculation and plotting of EFW is preferred for the assessment of fetal growth:27
Although AC is the main component in the estimation of fetal weight, EFW is able to detect additional at-risk cases compared to AC alone.28 Serial EFW measurement is furthermore as good or better than serial AC in predicting adverse outcome.29
Fetal weight curves can be customised using a predicted 280 day endpoint together with fetal weight based ‘proportionality curve’,16,30 whereas individual biometry parameters cannot.
Scan error for EFW can be quantified and audited against birthweight (after adjusting for time delay between scan and delivery), while there is no such neonatal gold standard for AC or any other ultrasound biometry parameter.
The GROW chart is carried by the mother, and EFW means something to her in terms of the current size of her baby and the projected birthweight, while this is not the case for AC and other parameters.
Standardising fetal growth surveillance:
The Growth Assessment Protocol (GAP)31 combines customised growth charts with accreditation training, e-learning support and audit tools, and has the principal objective of improving antenatal awareness of intrauterine growth restriction, thereby initiating further investigation and management, ultimately reducing the risk of intrauterine death or other adverse outcomes.
A regional version of GAP was first introduced in the West Midlands in 2009, where implementation resulted in the drop in stillbirth rates to below the national level for the first time ever, with evidence that this reduction was due to fewer deaths associated with growth restriction.12 The programme was subsequently adopted in two additional NHS regions and resulted in further significant reductions in stillbirth rates which was confirmed to be causally associated with the intervention as no similar drop occurred in the other NHS regions.32 Since then GAP has been implemented in nearly 80% of all Trust and Health Boards in the UK.33 As illustrated in Figure 4, the national roll-out was associated with a year-on-year drop in ONS reported stillbirth rates in England to their lowest ever level of 4.35/1000 in 2016, a 19% reduction compared to the previous 10 year average (2000–2009: 5.35/1000).34 GAP has adapted the RCOG guidelines6 and the NHS England algorithm7 into a workable and successful quality improvement and stillbirth prevention programme.
figure
Figure 4. Trend in stillbirth rates in England. Stillbirth rates (per 1000) in England, according to ONS. The rate remained similar over a 10-year period (2000, 5.26; 2009, 5.29) and averaged 5.35; the fall to 4.35 by 2016 following the implementation of the GAP program represented a 19% drop (P<.01). CI: confidence interval; GAP: growth assessment protocol; ONS: office of national statistics. Reproduced from Gardosi et al.,34 with permission.
Auditing performance
Audit is an essential component of GAP and is facilitated by software integral to the customised chart (GROW App). Following delivery, information is entered about birthweight, sex and gestational age at delivery, as well as whether there was antenatal suspicion (referral) or diagnosis of fetal growth problems. This is then compared with outcome, i.e. whether the birthweight was small, normal or large for gestational age. Before a unit implements GAP, an audit is undertaken to establish baseline detection rates, which is followed in the majority of GAP units with ongoing recording of outcome, with high (>90%) ascertainment rates.
Figure 5 shows the baseline and national trend over the last 2 years, with an overall ‘GAP user average’ as well as the average for the 10 best performing units. The baseline detection rate of SGA birthweight averaged 18.7% which is consistent with published audits from other environments.35,36 There has been a steady increase in antenatal detection rates to 42% (a 2.5 fold increase from baseline), with the top 10 units averaging 56%.
figure
Figure 5. Antenatal detection of SGA. Trend of detection of newborn infants with SGA birthweight (<10th customised centile). Baseline rates, GAP user average (GUA) and average for top ten performing units is shown. GAP: growth assessment protocol; GUA: GAP user average; SGA: small for gestational age. Reproduced from Gardosi et al.,34 with permission.
A limitation of this audit is that there is no routinely collected neonatal ‘gold standard’ for fetuses that had intrauterine growth restriction without being SGA. Thus, auditing SGA birthweight as the denominator accounts for only a proportion of cases with growth problems that may have been detected. Similarly, it is difficult to assess false positive rates, as in many instances, scans may have identified clinically relevant slow growth, without the fetus or neonate being SGA.
Audit of missed cases
Alongside routine audit of referral and detection rates, GAP-enrolled Trusts and Health Boards also audit randomly selected cases where babies with SGA birthweight had not been recognised to be small antenatally. This is undertaken using bespoke software – GAP SCORE (standardised clinical outcome review and evaluation). The application systematically examines the care of each case by looking at:
Assessment of risk factors at booking and subsequent management plan.
Routine fetal growth surveillance of both high risk and low risk pregnancies including the number of third trimester scans.
Accuracy of EFW in relation to the neonatal birthweight.
The software then derives a taxonomy of the sub-standard care factors responsible for the SGA status not having been recognised.
Analysis of 2 years of data (n = 2977) from local audits undertaken in 64 UK Trusts and Health Boards participating in the GAP programme, showed that these missed cases fall into 4 main categories (Figure 6).
figure
Figure 6. Missed case audit of 2977 pregnancies with SGA newborn from 64 Trusts and Health Boards in the UK GAP programme. Data source: Perinatal Institute.
As the pie chart shows, low-risk cases managed by fundal height measurement which failed to be suspected of SGA and referred for a scan were one of the smaller categories, with 18.8% of the missed cases. It is important to acknowledge here that, in units with trained midwives and other health professionals, low-risk pregnancies screened with fundal height measurement contributed less than a fifth to the overall ‘burden’ of unrecognised SGA cases. This is relevant as some clinicians consider routine scanning of all pregnancies as a shortcut solution to replace fundal height measurement and referral pathways. While a routine 36-week scan was reported to reach an antenatal detection rate of 56% of SGA (<10th birthweight centile) within a research setting,37 this has not been demonstrated in routine practice14 (Table 2). Routine growth scans have also not shown to have benefits according to Cochrane reviews38 and may provide false reassurance. In general terms, longitudinal assessment of growth is better than a spot-check of fetal size.
The remaining three categories (Figure 6) all related to ultrasound scans, when they were done but still failed to detect that the baby was SGA, or not done despite being indicated according to the risk assessment. This points towards quality issues which need further investigation, but is also likely to be the result of insufficient scan provision reflecting resource issues. The largest category of missed cases, 39.3%, comprised pregnancies which had been recognised as high risk and had ‘serial’ (2 or more) third trimester growth scans, yet were still missed. The average number of ultrasound scans in these high risk pregnancies was only 2.4, and the average delay between the last scan and delivery was 3 weeks, which suggests that, contrary to protocol, serial scanning was usually not continued up to delivery and, not surprisingly, failed to recognise late onset fetal growth restriction.
Challenges and next steps
The widespread shortages in ultrasound resources have led to a Department of Health sponsored initiative co-ordinated by Health Education England, to train an additional 200 health care professionals (from a range of relevant professions) in obstetric ultrasound by the middle of 2018. This is an interim measure which needs to be followed by a more sustained effort to ensure that the service catches up with requirements, as currently clinical non-adherence to evidence based guidelines are putting babies at risk.
Further work is also needed to enhance quality assurance of growth scans. Fetal weight estimation is more sensitive and specific than other measures for detecting smallness for gestational age, but can have large random errors.39 Estimated fetal weight can be assessed against birthweight using the freely available EFW Audit Tool40 which adjusts for the time interval between scan and delivery using the proportionality growth formula16 and calculates the percentage error of the scan. There have been also useful efforts in local Trusts (e.g. Taunton & Somerset – D Downs, personal communication) to audit performance, feed-back to individual operators, and identify outliers requiring further training. Finally, research is also underway to develop automated fetal measurement systems which provide real-time feedback to sonographers on measurement technique, consistency and image selection.
Declaration of Interests
All authors are employees of the Perinatal Institute, a not-for-profit organisation which derives income from the provision of support services to healthcare organisations, including the ‘Growth Assessment Protocol’ reviewed in this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethics approval
None required – article is a review.
Guarantor
JG.
Contributors
JG conceived the review and wrote the first draft, and worked with MW on subsequent versions. All authors reviewed and approved the final version of the manuscript.
Acknowledgements
None declared.
References
1. Flenady, V, Wojcieszek, AM, Middleton, P Stillbirths: recall to action in high-income countries. Lancet 2016; 387: 691–702.
Google Scholar | Medline | ISI
2. Gardosi J. Clinical implications of ‘unexplained’ stillbirths. In: Maternal and Child health research consortium, ed: CESDI 8th annual report Confidential enquiry into stillbirths and deaths in infancy. 2001. pp. 40–7, www.pi.nhs.uk/pnm/CESDI%20SB%20commentary.pdf (accessed 7 January 2018).
Google Scholar
3. Gardosi, J . Classification of stillbirth by relevant condition at death (ReCoDe): population based cohort study. BMJ 2005; 331: 1113–1117.
Google Scholar | Medline
4. Perinatal Institute. Confidential enquiry into stillbirths with fetal growth restriction, www.pi.nhs.uk/rpnm/CE_SB_Final.pdf (2017, accessed 7 January 2018).
Google Scholar
5. Gardosi, J, Madurasinghe, V, Williams, M Maternal and fetal risk factors for stillbirth: population based study. BMJ 2013; 346: f108–f108.
Google Scholar | Medline
6. Royal College of Obstetricians and Gynaecologists. The investigation and management of the small for gestational age fetus. Green Top Guideline No 31. RCOG, London, 2013.
Google Scholar
7. NHS England. Saving babies lives: a care bundle for reducing stillbirth, www.england.nhs.uk/wp-content/uploads/2016/03/saving-babies-lives-car-bundl.pdf (2016, accessed 7 January 2018).
Google Scholar
8. Figueras, F, Gardosi, J. Intrauterine growth restriction: new concepts in antenatal surveillance, diagnosis, and management. Am J Obstet Gynecol 2011; 204: 288–300.
Google Scholar | Medline
9. Southam, M, Williams, M, Malik, A Effect of serial scan frequency on antenatal detection of fetal growth restriction. Arch Dis Child 2014; 99: A104–A105.
Google Scholar
10. Gardosi J and Williams M. Serial ultrasound scanning in pregnancies at risk of intrauterine growth restriction: cost benefit analysis. Perinatal Institute, www.perinatal.org.uk/gap/Resources/RCOG_serial_scans_cost_benefit_analysis.pdf (2014, accessed 7 January 2018).
Google Scholar
11. Francis, A, Giddings, S, Turner, S Prevalence of risk factors requiring serial ultrasound assessment of fetal growth according to new NHS England algorithm. BJOG 2016; 123: 87.
Google Scholar
12. Perinatal Institute. Stillbirths in the West Midlands: 2011 update, www.pi.nhs.uk/pnm/clusterreports/2011/WM_2011_Stillbirth_Update_Sept_2012.pdf (2012, accessed 7 January 2018).
Google Scholar
13. Stillbirth Foundation of Australia & PwC. The economic impacts of stillbirth in Australia, http://stillbirthfoundation.org.au/wp-content/uploads/2016/10/Economic-Impacts-of-Stillbirth-2016-PwC.pdf (2016, accessed 7 January 2018).
Google Scholar
14. Francis, A, Gardosi, J. Effectiveness of ultrasound biometry at 34 - 36 weeks in the detection of SGA at birth. BJOG 2016; 123: 86.
Google Scholar
15. Francis, A, Tonks, A, Gardosi, J. Accuracy of ultrasound estimation of fetal weight at term. Arch Dis Child 2011; 96(Suppl 1): Fa61.
Google Scholar
16. Gardosi, J, Mongelli, M, Wilcox, M An adjustable fetal weight standard. Ultrasound Obstet Gynecol 1995; 6: 168–174.
Google Scholar | Medline | ISI
17. Hadlock, FP . Estimation of fetal weight with the use of head, body, and femur measurements- a prospective study. Am J Obstet Gynecol 1985; 151: 333–337.
Google Scholar | Medline | ISI
18. Gardosi, J, Francis, A. Controlled trial of fundal height measurement plotted on customised antenatal growth charts. Br J Obstet Gynaecol 1999; 106: 309–317.
Google Scholar | Medline
19. Mongelli, M, Gardosi, J. Reduction of false-positive diagnosis of fetal growth restriction by application of customized fetal growth standards. Obstet Gynecol 1996; 88: 844–848.
Google Scholar | Medline | ISI
20. Gardosi, J, Clausson, B, Francis, A. The value of customised centiles in assessing perinatal mortality risk associated with parity and maternal size: value of customising centiles for parity and maternal size. BJOG 2009; 116: 1356–1363.
Google Scholar | Medline
21. Giddings, S, Clifford, S, Madurasinghe, V PFM.69 Customised vs uncustomised ultrasound charts in the assessment of perinatal mortality risk in the South Asian maternity population. Arch Dis Child 2014; 99(Suppl 1): A104–A104.
Google Scholar
22. Villar, J, Ismail, LC, Victora, CG International standards for newborn weight, length, and head circumference by gestational age and sex: the newborn cross-sectional study of the INTERGROWTH-21st project. Lancet 2014; 384: 857–868.
Google Scholar | Medline | ISI
23. Stirnemann, J, Villar, J, Salomon, LJ International estimated fetal weight standards of the INTERGROWTH-21st project. Ultrasound Obstet Gynecol 2017; 49: 478–486.
Google Scholar | Medline
24. Hanson, M, Kiserud, T, Visser, GHA Optimal fetal growth: a misconception?. Am J Obstet Gynecol 2015; 213: 332e1–4.
Google Scholar | Medline
25. Anderson, NH, Sadler, LC, McKinlay, CJD INTERGROWTH-21st vs customized birthweight standards for identification of perinatal mortality and morbidity. Am J Obstet Gynecol 2016; 214: 509e1–7.
Google Scholar
26. Francis, A, Hugh, O, Gardosi, J. Customized vs INTERGROWTH-21st standards for the assessment of birthweight and stillbirth risk at term. Am J Obstet Gynecol 2018; 218: S692–S699.
Google Scholar | Medline
27. Gardosi, J . Ultrasound biometry and fetal growth restriction. Fetal Matern Med Rev 2002; 13: 249–259.
Google Scholar
28. Chauhan, SP, Cole, J, Sanderson, M Suspicion of intrauterine growth restriction: use of abdominal circumference alone or estimated fetal weight below 10%. J Matern Fetal Neonatal Med 2006; 19: 557–562.
Google Scholar | Medline | ISI
29. Chang, TC, Robson, SC, Spencer, JAD Identification of fetal growth retardation: comparison of Doppler waveform indices and serial ultrasound measurements of abdominal circumference and fetal weight. Obstet Gynecol 1994; 45: 198.
Google Scholar
30. Gardosi, J, Chang, A, Kalyan, B Customised antenatal growth charts. Lancet 1992; 339: 283–287.
Google Scholar | Medline | ISI
31. Clifford, S, Giddings, S, Southam, M The growth assessment protocol: a national programme to improve patient safety in maternity care. MIDIRS Midwife Dig 2013; 23: 516–523.
Google Scholar
32. Gardosi, J, Giddings, S, Clifford, S Association between reduced stillbirth rates in England and regional uptake of accreditation training in customised fetal growth assessment. BMJ Open 2013; 3: e003942.
Google Scholar | Medline
33. Perinatal Institute. Growth assessment protocol – uptake, www.perinatal.org.uk/gap-uptake.aspx (2017, accessed 7 January 2018).
Google Scholar
34. Gardosi, J, Francis, A, Turner, S Customised growth charts: rationale, validation and clinical benefits. Am J Obstet Gynecol 2018; 218: S609–S618.
Google Scholar | Medline
35. Bakke, B, Nakling, J. Effectiveness of antenatal care: a population based study. BJOG 1993; 100: 727–732.
Google Scholar
36. Kean, L, Liu, D. Antenatal care as a screening tool for the detection of small for gestational age babies in the low risk population. J Obstet Gynaecol 1996; 16: 77–82.
Google Scholar
37. Sovio, U, White, IR, Dacey, A Screening for fetal growth restriction with universal third trimester ultrasonography in nulliparous women in the Pregnancy Outcome Prediction (POP) study: a prospective cohort study. Lancet 2015; 386: 2089–2097.
Google Scholar | Medline
38. Bricker, L, Medley, N, Pratt, JJ. Routine ultrasound in late pregnancy (after 24 weeks' gestation). Cochrane Database Syst Rev 2015; 6: CD001451.
Google Scholar
39. Dudley, NJ . A review of ultrasound fetal weight estimation in the early prediction of low birthweight. Ultrasound 2013; 21: 181–186.
Google Scholar | SAGE Journals
40. Perinatal Institute. GAP – ultrasound, www.perinatal.org.uk/gap/ultrasound.aspx (2017, accessed 7 January 2018).
Google Scholar
Show less
Mandy Williams, Sue Turner, Emily Butler, Jason Gardosi First Published March 22, 2018 Review Article
https://doi.org/10.1177/1742271X18760657
Article information
Article has an altmetric score of 4 Free Access
Abstract
Antenatal surveillance of fetal growth is an essential part of good maternity care, as lack of detection of fetal growth restriction is directly associated with stillbirth and perinatal morbidity. New algorithms and guidelines provide care pathways which rely on regular third trimester ultrasound biometry and plotting of estimated fetal weight in pregnancies considered to be at increased risk, and their implementation has increased pressures on ultrasound resources. Customised growth charts have improved the distinction between constitutional and pathological smallness and reduced unnecessary referrals. Their introduction, together with clinicians' training, e-learning and audit as the key elements of the growth assessment protocol, has resulted in increased antenatal detection of small for gestational age babies and a reduction in avoidable stillbirths. However, missed case audits highlight that further improvements are needed, and point to the need to address quality assurance and resource issues in ultrasound services.
Keywords Estimated fetal weight, small for gestational age, fetal growth restriction, stillbirth
Fetal growth restriction and stillbirth
There has been a recent focus on stillbirth prevention, stimulated by the fact that stillbirth rates in the UK were among the highest in Western Europe and had not changed substantially over the last 20 years.1 The situation was not helped by the fact that the majority of stillbirths have until recently been classified as ‘unexplained’, which was taken to suggest that these deaths were unavoidable.2,3
However, independent case reviews into the care of pregnancies that ended with a stillbirth found that many were associated with substandard care and therefore potentially avoidable.4 The largest category were babies with growth restriction which had not been recognised antenatally. Many of these mothers never had a growth scan, and never even came to any special attention of the maternity services, until presenting one day – often late in pregnancy – with absent fetal movements.
These observations were reinforced by a better classification system which included a category for ‘fetal growth restriction’ (FGR), defined as small for gestational age (SGA) according to customised centiles. FGR was found to be the largest category, accounting for over half of all normally formed stillbirths.3 Subsequent population-based analysis confirmed that pregnancies with an SGA fetus had a seven-fold increased risk for stillbirth, and that most of them were not recognised antenatally.5 The report also found that antenatal recognition of FGR can halve stillbirth risk through earlier delivery (Figure 1).
figure
Figure 1. Stillbirth risk in pregnancies with and without antenatal (A/N) detection of fetal growth restriction (FGR), defined as < 10th customised centile. Illustration based on data in Gardosi et al.5
Risk factors and algorithms for surveillance
The Royal College of Obstetricians and Gynaecologists' guidelines6 for surveillance and management of pregnancies with an SGA fetal outlined a standard of care that included serial measurement and plotting of fundal height and ultrasound biometry in low- and high-risk pregnancies, respectively. Customised charts (which adjust growth curves according to maternal height, weight, parity and ethnic origin) were recommended for plotting both fundal height and estimated fetal weight (EFW).6
The RCOG guidelines reviewed the risk factors associated with FGR and developed an algorithm which set out the recommended assessment of early pregnancy risk factors and care pathways contingent on low risk (→ serial measurement and plotting of fundal height) and increased risk (→ serial measurement and plotting of estimated fetal weight). Risk factors were divided into major and minor, with major risk factors requiring serial scans, and three minor prompting early pregnancy assessment by uterine Doppler to establish whether risk was major. Many units found this protocol too complicated for everyday use. Therefore, we assisted NHS England, as part of their Saving Babies' Lives Care Bundle initiative to develop a simplified algorithm which identified a list of risk factors, the presence of any of which would be an indication for serial ultrasound biometry.7 The algorithm was adapted for the national GAP programme (see below) by including use of customised growth charts (Figure 2).
figure
Figure 2. The growth assessment protocol (GAP) algorithm for fetal growth surveillance. Adapted from NHS England.7
As most instances of FGR are ‘late onset’ in the third trimester (usually from 32 weeks),5,8 the algorithm recommends surveillance until delivery. In high-risk pregnancy, a three-weekly regime would result in 4.5 third trimester scans – at 28, 31, 34, 37 and 40 weeks (the 40-week scan being needed for only half of all pregnancies still ongoing at that gestation).
Resource implications for guideline based surveillance
For many units, a schedule of 4.5 scans for all high-risk pregnancies is challenging and difficult to sustain within current resources. A West Midlands audit of ‘current practice’ in scan regimes afforded to pregnancies with a past history of SGA birth – a clinically undisputed indication for serial scans in subsequent pregnancies – found that the median number of scans performed was between 2 and 3 only, and 21% of these high-risk pregnancies had no third trimester scan at all.9 The audit highlighted the importance of scan frequency: a one scan policy (usually at 32 weeks) or two scans (usually at 28 and 34 weeks) resulted in a detection rate of around 30%, which was little better than doing no scans at all.
We examined the proportion of the NHS population in the West Midlands that would be designated at increased risk requiring serial scans. According to the RCOG guidelines,6 25.5% of mothers would have either a major risk factor (20.6%) or 3 or more minor risk factors for SGA (4.9%).10 The figure is higher – 36% – when applying the categories in the NHSE algorithm11 (Table 1), mainly because all smokers, including those having <10 cigarettes per day, are included as high risk. In practice, many units struggle to provide scans for all patients with these indications. Typically heavy smokers, or any smokers, and mothers with a raised BMI do not tend to receive serial scans.
Table
Table 1. Prevalence of risk factors requiring serial ultrasound assessment of fetal growth according to new NHS England algorithm; West Midlands, N = 146,774
Table 1. Prevalence of risk factors requiring serial ultrasound assessment of fetal growth according to new NHS England algorithm; West Midlands, N = 146,774
View larger version
Cost–benefit analysis
We assessed the cost benefit of a serial scan regime for increased risk pregnancies10 according to RCOG defined risk factors.6 An estimated 25% of the population at higher risk of SGA translated to 600–700 additional scans per 1000 births, depending on current scan provision.10 A growth scan provided additionally to an already existing scanning service and infrastructure was estimated to cost £15 which, multiplied by the additional scan requirement, represented a cost of £10 per pregnancy, or £30,000 for a maternity unit with 3000 deliveries per annum. Against this, improved antenatal detection of IUGR has many benefits, including reduction in stillbirths (estimating 1 stillbirth saved per 1000 deliveries based on previous experience12), as well as reduced neonatal admissions, perinatal morbidity, cerebral palsy and litigation. We estimated that together, these will result in savings of at least £120 per pregnancy or £360,000 for a unit with 3000 births, representing a 12-fold return on investment.10 However, the value of benefits was likely to be gross under-estimates in the absence of good data. In the case of stillbirths, they were confined to £5000 per death, mainly for costs of investigations; however, a more recent, detailed attempt in Australia13 to estimate stillbirths' direct costs to the health service (in terms of investigations and counselling) as well as indirect costs including absenteeism and loss of productivity, resulted in an estimate of £36,000 per death.13 Quite apart from the immeasurable loss of a human life, savings recognised in this way would vastly exceed any investment required to improve ultrasound services in step with national guidelines for fetal growth surveillance.
Accuracy of growth scans at term
The NHS England algorithm prescribes scans in increased risk pregnancies to be continued until delivery. The argument for this approach is strengthened by findings of a study where a prospective, routinely collected EFW at ‘last scan’ – performed for any indication – was assessed for its ability to predict SGA at birth14 (Table 2). The data are analysed at the three most common gestational age points for a ‘routine’ third trimester scan – 34, 35 or 36 weeks. As shown in Table 2, pregnancies that had scans for any reason were at increased risk of SGA at delivery, with rates ranging from 14 to 19%. However, detection rates by EFW were overall poor, ranging from 19.0% at 34 weeks up to only 36.1% at 36 weeks. This may indicate poor scan technique or limitations of fetal weight estimation by ultrasound, but is more likely to be associated with slowing of growth towards an SGA birthweight late in third trimester, which is missed if the last scan is done too early.
Table
Table 2. Effectiveness of ultrasound biometry at 34–36 weeks in the detection of SGA at birth
Table 2. Effectiveness of ultrasound biometry at 34–36 weeks in the detection of SGA at birth
View larger version
When protocols for scanning up to delivery were first introduced, many ultrasound departments were reluctant because of concerns that scans at term are less inaccurate, although there seemed little evidence to support such a belief. We therefore investigated the accuracy of preterm and term scans, based on routine measurements collected from all 19 West Midlands regional maternity units.15 The cohort consisted of 2296 pregnancies where an ultrasound scan was performed within three days of delivery, and this included 606 preterm (<37 weeks) and 1690 term deliveries. EFW was calculated by the Hadlock formula programmed into most units’ ultrasound equipment. The assumed weight gain during the 1, 2 or 3 day delay between scan measurement and birth was adjusted using the previously described ‘proportionality formula’.16 The results showed that EFWs at term were at least as good, and in fact marginally better than scans done in the preterm period, with 73% of EFWs falling within a ±10% margin of error (Table 3).
Table
Table 3. Accuracy of ultrasound scan at preterm vs. term gestational age, with scan performed within three days before delivery preterm (n = 606) or term (n = 1690).15
Table 3. Accuracy of ultrasound scan at preterm vs. term gestational age, with scan performed within three days before delivery preterm (n = 606) or term (n = 1690).15
View larger version
Practitioners' concern about obtaining accurate fetal head measurement when it was fully engaged at term could be addressed by EFW formulae such as that by Hadlock,17 which relies on abdominal circumference (AC) and femur length (FL) only.
The growth assessment protocol: A comprehensive programme for surveillance
The case reviews of adverse outcomes and epidemiological analysis provided the rationale and urgent need to focus on antenatal detection of FGR, and the national guidelines provided the pathway to achieve this within routine antenatal care. We integrated these elements into the growth assessment protocol (GAP), a comprehensive programme including accreditation training for midwives, obstetricians and ultrasonographers, use of customised growth charts, birthweight centiles and audit. It included a recommended surveillance pathway which was broadly based on the NHS England algorithm,7 but included use of customised charts according to RCOG guidelines6 (Figure 2).
The emphasis is on longitudinal assessment, preferably by the same practitioner. In low-risk pregnancy, the main method of surveillance is serial fundal height measurement, with protocols prompting immediate referral when the first measurement is below the 10th customised centile, or the slope of sequential measurement does not follow the growth curves on the individually customised chart. Trained midwives can make autonomous referrals for scans if they have concerns about growth, without having their fundal height measurements ‘checked’ by other care provider(s).
Identification of increased risk pregnancy is contingent on primary care surveillance and referral, as well as risk assessment at the beginning of pregnancy to identify mothers who need serial scans. In either case, the required measure is the EFW plotted on a customised growth chart.
Customised growth charts
Underpinning the low and increased risk pathways (Figure 2) is the plotting of fundal height (FH) and EFW measurements on customised charts, as recommended by RCOG Green Top guidelines.6 A controlled study has shown that a standardised approach significantly increases detection rates while reducing unnecessary referrals for ultrasound scans.18 Customised charts also improve EFW assessment and reduce false positive diagnosis of SGA.19
The advantages of customised GROW (‘gestation related optimal weight’) charts become most apparent when looking at subgroups within the population, e.g. according to maternal size. There is better correlation between SGA defined by customised centiles and perinatal death than any population based, one-size-fits-all standard.20
GROW charts also improve the definition of SGA in multi-ethnic populations21: in a cohort of 10,405 South Asian mothers from India, Pakistan and Bangladesh, who had a third trimester scan, 1554 (14.9%) were SGA according to the Hadlock fetal weight curve; however, 864 of these (55.6%) were not SGA if the growth curves were customised according to the mother's characteristics (height, weight, parity and ethnic origin). This effect is illustrated in Figure 3(a) and (b). Significantly, these cases re-classified as not SGA had the same perinatal mortality risk as the general, non-SGA population.21
figure
Figure 3. (a) and (b) Examples of customised charts for British-European (a) and South Asian (b) mothers. A sample estimated fetal weight (EFW) of 2500 g at 37.0 weeks is plotted on each chart to illustrate different results and clinical implications; 56% of SGA EFWs in pregnancies of South Asian mothers are not SGA if plotted on their own customised charts, and are not associated with increased perinatal mortality risk.21
Recently, INTERGROWTH 21st has been promoted as an internationally applicable, one-size-fits all standard for birthweight22 and fetal weight.23 It was derived from low risk, well-nourished mothers in eight countries and was therefore meant to provide a prescriptive optimal standard for all pregnancies. However, this concept has been challenged on theoretical grounds,24 and when applied to detailed multi-ethnic maternity datasets, it performed poorly compared to a customised standard.25 In a recent study using a 10-country cohort of 1.25 million births, INTERGROWTH 21st showed unrepresentative rates of SGA (4.4%) and LGA (20.6%), and poor correlation with outcome.26
Plotting estimated fetal weight versus abdominal circumference measurement
Many ultrasonographers and clinicians are accustomed to plotting individual parameters like HC, AC and FL rather than EFW. While these need of course to be recorded, there are a number of reasons why calculation and plotting of EFW is preferred for the assessment of fetal growth:27
Although AC is the main component in the estimation of fetal weight, EFW is able to detect additional at-risk cases compared to AC alone.28 Serial EFW measurement is furthermore as good or better than serial AC in predicting adverse outcome.29
Fetal weight curves can be customised using a predicted 280 day endpoint together with fetal weight based ‘proportionality curve’,16,30 whereas individual biometry parameters cannot.
Scan error for EFW can be quantified and audited against birthweight (after adjusting for time delay between scan and delivery), while there is no such neonatal gold standard for AC or any other ultrasound biometry parameter.
The GROW chart is carried by the mother, and EFW means something to her in terms of the current size of her baby and the projected birthweight, while this is not the case for AC and other parameters.
Standardising fetal growth surveillance:
The Growth Assessment Protocol (GAP)31 combines customised growth charts with accreditation training, e-learning support and audit tools, and has the principal objective of improving antenatal awareness of intrauterine growth restriction, thereby initiating further investigation and management, ultimately reducing the risk of intrauterine death or other adverse outcomes.
A regional version of GAP was first introduced in the West Midlands in 2009, where implementation resulted in the drop in stillbirth rates to below the national level for the first time ever, with evidence that this reduction was due to fewer deaths associated with growth restriction.12 The programme was subsequently adopted in two additional NHS regions and resulted in further significant reductions in stillbirth rates which was confirmed to be causally associated with the intervention as no similar drop occurred in the other NHS regions.32 Since then GAP has been implemented in nearly 80% of all Trust and Health Boards in the UK.33 As illustrated in Figure 4, the national roll-out was associated with a year-on-year drop in ONS reported stillbirth rates in England to their lowest ever level of 4.35/1000 in 2016, a 19% reduction compared to the previous 10 year average (2000–2009: 5.35/1000).34 GAP has adapted the RCOG guidelines6 and the NHS England algorithm7 into a workable and successful quality improvement and stillbirth prevention programme.
figure
Figure 4. Trend in stillbirth rates in England. Stillbirth rates (per 1000) in England, according to ONS. The rate remained similar over a 10-year period (2000, 5.26; 2009, 5.29) and averaged 5.35; the fall to 4.35 by 2016 following the implementation of the GAP program represented a 19% drop (P<.01). CI: confidence interval; GAP: growth assessment protocol; ONS: office of national statistics. Reproduced from Gardosi et al.,34 with permission.
Auditing performance
Audit is an essential component of GAP and is facilitated by software integral to the customised chart (GROW App). Following delivery, information is entered about birthweight, sex and gestational age at delivery, as well as whether there was antenatal suspicion (referral) or diagnosis of fetal growth problems. This is then compared with outcome, i.e. whether the birthweight was small, normal or large for gestational age. Before a unit implements GAP, an audit is undertaken to establish baseline detection rates, which is followed in the majority of GAP units with ongoing recording of outcome, with high (>90%) ascertainment rates.
Figure 5 shows the baseline and national trend over the last 2 years, with an overall ‘GAP user average’ as well as the average for the 10 best performing units. The baseline detection rate of SGA birthweight averaged 18.7% which is consistent with published audits from other environments.35,36 There has been a steady increase in antenatal detection rates to 42% (a 2.5 fold increase from baseline), with the top 10 units averaging 56%.
figure
Figure 5. Antenatal detection of SGA. Trend of detection of newborn infants with SGA birthweight (<10th customised centile). Baseline rates, GAP user average (GUA) and average for top ten performing units is shown. GAP: growth assessment protocol; GUA: GAP user average; SGA: small for gestational age. Reproduced from Gardosi et al.,34 with permission.
A limitation of this audit is that there is no routinely collected neonatal ‘gold standard’ for fetuses that had intrauterine growth restriction without being SGA. Thus, auditing SGA birthweight as the denominator accounts for only a proportion of cases with growth problems that may have been detected. Similarly, it is difficult to assess false positive rates, as in many instances, scans may have identified clinically relevant slow growth, without the fetus or neonate being SGA.
Audit of missed cases
Alongside routine audit of referral and detection rates, GAP-enrolled Trusts and Health Boards also audit randomly selected cases where babies with SGA birthweight had not been recognised to be small antenatally. This is undertaken using bespoke software – GAP SCORE (standardised clinical outcome review and evaluation). The application systematically examines the care of each case by looking at:
Assessment of risk factors at booking and subsequent management plan.
Routine fetal growth surveillance of both high risk and low risk pregnancies including the number of third trimester scans.
Accuracy of EFW in relation to the neonatal birthweight.
The software then derives a taxonomy of the sub-standard care factors responsible for the SGA status not having been recognised.
Analysis of 2 years of data (n = 2977) from local audits undertaken in 64 UK Trusts and Health Boards participating in the GAP programme, showed that these missed cases fall into 4 main categories (Figure 6).
figure
Figure 6. Missed case audit of 2977 pregnancies with SGA newborn from 64 Trusts and Health Boards in the UK GAP programme. Data source: Perinatal Institute.
As the pie chart shows, low-risk cases managed by fundal height measurement which failed to be suspected of SGA and referred for a scan were one of the smaller categories, with 18.8% of the missed cases. It is important to acknowledge here that, in units with trained midwives and other health professionals, low-risk pregnancies screened with fundal height measurement contributed less than a fifth to the overall ‘burden’ of unrecognised SGA cases. This is relevant as some clinicians consider routine scanning of all pregnancies as a shortcut solution to replace fundal height measurement and referral pathways. While a routine 36-week scan was reported to reach an antenatal detection rate of 56% of SGA (<10th birthweight centile) within a research setting,37 this has not been demonstrated in routine practice14 (Table 2). Routine growth scans have also not shown to have benefits according to Cochrane reviews38 and may provide false reassurance. In general terms, longitudinal assessment of growth is better than a spot-check of fetal size.
The remaining three categories (Figure 6) all related to ultrasound scans, when they were done but still failed to detect that the baby was SGA, or not done despite being indicated according to the risk assessment. This points towards quality issues which need further investigation, but is also likely to be the result of insufficient scan provision reflecting resource issues. The largest category of missed cases, 39.3%, comprised pregnancies which had been recognised as high risk and had ‘serial’ (2 or more) third trimester growth scans, yet were still missed. The average number of ultrasound scans in these high risk pregnancies was only 2.4, and the average delay between the last scan and delivery was 3 weeks, which suggests that, contrary to protocol, serial scanning was usually not continued up to delivery and, not surprisingly, failed to recognise late onset fetal growth restriction.
Challenges and next steps
The widespread shortages in ultrasound resources have led to a Department of Health sponsored initiative co-ordinated by Health Education England, to train an additional 200 health care professionals (from a range of relevant professions) in obstetric ultrasound by the middle of 2018. This is an interim measure which needs to be followed by a more sustained effort to ensure that the service catches up with requirements, as currently clinical non-adherence to evidence based guidelines are putting babies at risk.
Further work is also needed to enhance quality assurance of growth scans. Fetal weight estimation is more sensitive and specific than other measures for detecting smallness for gestational age, but can have large random errors.39 Estimated fetal weight can be assessed against birthweight using the freely available EFW Audit Tool40 which adjusts for the time interval between scan and delivery using the proportionality growth formula16 and calculates the percentage error of the scan. There have been also useful efforts in local Trusts (e.g. Taunton & Somerset – D Downs, personal communication) to audit performance, feed-back to individual operators, and identify outliers requiring further training. Finally, research is also underway to develop automated fetal measurement systems which provide real-time feedback to sonographers on measurement technique, consistency and image selection.
Declaration of Interests
All authors are employees of the Perinatal Institute, a not-for-profit organisation which derives income from the provision of support services to healthcare organisations, including the ‘Growth Assessment Protocol’ reviewed in this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethics approval
None required – article is a review.
Guarantor
JG.
Contributors
JG conceived the review and wrote the first draft, and worked with MW on subsequent versions. All authors reviewed and approved the final version of the manuscript.
Acknowledgements
None declared.
References
1. Flenady, V, Wojcieszek, AM, Middleton, P Stillbirths: recall to action in high-income countries. Lancet 2016; 387: 691–702.
Google Scholar | Medline | ISI
2. Gardosi J. Clinical implications of ‘unexplained’ stillbirths. In: Maternal and Child health research consortium, ed: CESDI 8th annual report Confidential enquiry into stillbirths and deaths in infancy. 2001. pp. 40–7, www.pi.nhs.uk/pnm/CESDI%20SB%20commentary.pdf (accessed 7 January 2018).
Google Scholar
3. Gardosi, J . Classification of stillbirth by relevant condition at death (ReCoDe): population based cohort study. BMJ 2005; 331: 1113–1117.
Google Scholar | Medline
4. Perinatal Institute. Confidential enquiry into stillbirths with fetal growth restriction, www.pi.nhs.uk/rpnm/CE_SB_Final.pdf (2017, accessed 7 January 2018).
Google Scholar
5. Gardosi, J, Madurasinghe, V, Williams, M Maternal and fetal risk factors for stillbirth: population based study. BMJ 2013; 346: f108–f108.
Google Scholar | Medline
6. Royal College of Obstetricians and Gynaecologists. The investigation and management of the small for gestational age fetus. Green Top Guideline No 31. RCOG, London, 2013.
Google Scholar
7. NHS England. Saving babies lives: a care bundle for reducing stillbirth, www.england.nhs.uk/wp-content/uploads/2016/03/saving-babies-lives-car-bundl.pdf (2016, accessed 7 January 2018).
Google Scholar
8. Figueras, F, Gardosi, J. Intrauterine growth restriction: new concepts in antenatal surveillance, diagnosis, and management. Am J Obstet Gynecol 2011; 204: 288–300.
Google Scholar | Medline
9. Southam, M, Williams, M, Malik, A Effect of serial scan frequency on antenatal detection of fetal growth restriction. Arch Dis Child 2014; 99: A104–A105.
Google Scholar
10. Gardosi J and Williams M. Serial ultrasound scanning in pregnancies at risk of intrauterine growth restriction: cost benefit analysis. Perinatal Institute, www.perinatal.org.uk/gap/Resources/RCOG_serial_scans_cost_benefit_analysis.pdf (2014, accessed 7 January 2018).
Google Scholar
11. Francis, A, Giddings, S, Turner, S Prevalence of risk factors requiring serial ultrasound assessment of fetal growth according to new NHS England algorithm. BJOG 2016; 123: 87.
Google Scholar
12. Perinatal Institute. Stillbirths in the West Midlands: 2011 update, www.pi.nhs.uk/pnm/clusterreports/2011/WM_2011_Stillbirth_Update_Sept_2012.pdf (2012, accessed 7 January 2018).
Google Scholar
13. Stillbirth Foundation of Australia & PwC. The economic impacts of stillbirth in Australia, http://stillbirthfoundation.org.au/wp-content/uploads/2016/10/Economic-Impacts-of-Stillbirth-2016-PwC.pdf (2016, accessed 7 January 2018).
Google Scholar
14. Francis, A, Gardosi, J. Effectiveness of ultrasound biometry at 34 - 36 weeks in the detection of SGA at birth. BJOG 2016; 123: 86.
Google Scholar
15. Francis, A, Tonks, A, Gardosi, J. Accuracy of ultrasound estimation of fetal weight at term. Arch Dis Child 2011; 96(Suppl 1): Fa61.
Google Scholar
16. Gardosi, J, Mongelli, M, Wilcox, M An adjustable fetal weight standard. Ultrasound Obstet Gynecol 1995; 6: 168–174.
Google Scholar | Medline | ISI
17. Hadlock, FP . Estimation of fetal weight with the use of head, body, and femur measurements- a prospective study. Am J Obstet Gynecol 1985; 151: 333–337.
Google Scholar | Medline | ISI
18. Gardosi, J, Francis, A. Controlled trial of fundal height measurement plotted on customised antenatal growth charts. Br J Obstet Gynaecol 1999; 106: 309–317.
Google Scholar | Medline
19. Mongelli, M, Gardosi, J. Reduction of false-positive diagnosis of fetal growth restriction by application of customized fetal growth standards. Obstet Gynecol 1996; 88: 844–848.
Google Scholar | Medline | ISI
20. Gardosi, J, Clausson, B, Francis, A. The value of customised centiles in assessing perinatal mortality risk associated with parity and maternal size: value of customising centiles for parity and maternal size. BJOG 2009; 116: 1356–1363.
Google Scholar | Medline
21. Giddings, S, Clifford, S, Madurasinghe, V PFM.69 Customised vs uncustomised ultrasound charts in the assessment of perinatal mortality risk in the South Asian maternity population. Arch Dis Child 2014; 99(Suppl 1): A104–A104.
Google Scholar
22. Villar, J, Ismail, LC, Victora, CG International standards for newborn weight, length, and head circumference by gestational age and sex: the newborn cross-sectional study of the INTERGROWTH-21st project. Lancet 2014; 384: 857–868.
Google Scholar | Medline | ISI
23. Stirnemann, J, Villar, J, Salomon, LJ International estimated fetal weight standards of the INTERGROWTH-21st project. Ultrasound Obstet Gynecol 2017; 49: 478–486.
Google Scholar | Medline
24. Hanson, M, Kiserud, T, Visser, GHA Optimal fetal growth: a misconception?. Am J Obstet Gynecol 2015; 213: 332e1–4.
Google Scholar | Medline
25. Anderson, NH, Sadler, LC, McKinlay, CJD INTERGROWTH-21st vs customized birthweight standards for identification of perinatal mortality and morbidity. Am J Obstet Gynecol 2016; 214: 509e1–7.
Google Scholar
26. Francis, A, Hugh, O, Gardosi, J. Customized vs INTERGROWTH-21st standards for the assessment of birthweight and stillbirth risk at term. Am J Obstet Gynecol 2018; 218: S692–S699.
Google Scholar | Medline
27. Gardosi, J . Ultrasound biometry and fetal growth restriction. Fetal Matern Med Rev 2002; 13: 249–259.
Google Scholar
28. Chauhan, SP, Cole, J, Sanderson, M Suspicion of intrauterine growth restriction: use of abdominal circumference alone or estimated fetal weight below 10%. J Matern Fetal Neonatal Med 2006; 19: 557–562.
Google Scholar | Medline | ISI
29. Chang, TC, Robson, SC, Spencer, JAD Identification of fetal growth retardation: comparison of Doppler waveform indices and serial ultrasound measurements of abdominal circumference and fetal weight. Obstet Gynecol 1994; 45: 198.
Google Scholar
30. Gardosi, J, Chang, A, Kalyan, B Customised antenatal growth charts. Lancet 1992; 339: 283–287.
Google Scholar | Medline | ISI
31. Clifford, S, Giddings, S, Southam, M The growth assessment protocol: a national programme to improve patient safety in maternity care. MIDIRS Midwife Dig 2013; 23: 516–523.
Google Scholar
32. Gardosi, J, Giddings, S, Clifford, S Association between reduced stillbirth rates in England and regional uptake of accreditation training in customised fetal growth assessment. BMJ Open 2013; 3: e003942.
Google Scholar | Medline
33. Perinatal Institute. Growth assessment protocol – uptake, www.perinatal.org.uk/gap-uptake.aspx (2017, accessed 7 January 2018).
Google Scholar
34. Gardosi, J, Francis, A, Turner, S Customised growth charts: rationale, validation and clinical benefits. Am J Obstet Gynecol 2018; 218: S609–S618.
Google Scholar | Medline
35. Bakke, B, Nakling, J. Effectiveness of antenatal care: a population based study. BJOG 1993; 100: 727–732.
Google Scholar
36. Kean, L, Liu, D. Antenatal care as a screening tool for the detection of small for gestational age babies in the low risk population. J Obstet Gynaecol 1996; 16: 77–82.
Google Scholar
37. Sovio, U, White, IR, Dacey, A Screening for fetal growth restriction with universal third trimester ultrasonography in nulliparous women in the Pregnancy Outcome Prediction (POP) study: a prospective cohort study. Lancet 2015; 386: 2089–2097.
Google Scholar | Medline
38. Bricker, L, Medley, N, Pratt, JJ. Routine ultrasound in late pregnancy (after 24 weeks' gestation). Cochrane Database Syst Rev 2015; 6: CD001451.
Google Scholar
39. Dudley, NJ . A review of ultrasound fetal weight estimation in the early prediction of low birthweight. Ultrasound 2013; 21: 181–186.
Google Scholar | SAGE Journals
40. Perinatal Institute. GAP – ultrasound, www.perinatal.org.uk/gap/ultrasound.aspx (2017, accessed 7 January 2018).
Google Scholar
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου