Clinical overview
Intra-uterine growth restriction (IUGR), or fetal growth restriction (FGR), describes a fetus that has failed to reach its genetically determined growth potential, usually because of placental insufficiency. It is the single most important antenatal antecedent of stillbirth, and a major contributor to perinatal mortality and to long-term neurodevelopmental and cardiometabolic morbidity. In South Africa, where obstetric haemorrhage, hypertension and HIV-associated infection dominate maternal deaths, FGR is the quiet engine behind a large share of the fetal and early-neonatal deaths counted in perinatal audit — many of them avoidable through earlier detection and timely delivery.
The single most useful conceptual move a registrar can make is to stop thinking of FGR as one disease and instead split it into early-onset (diagnosed before 32 weeks) and late-onset (at or after 32 weeks) phenotypes. They share a cause — placental insufficiency — but diverge in severity, the order in which Doppler abnormalities appear, the speed of deterioration, the surveillance tool that detects them, and the obstetric decision that follows. Early-onset FGR is severe, deteriorates predictably over days to weeks, and is dominated by the conflict between prematurity and progressive hypoxia. Late-onset FGR is milder per fetus but vastly commoner, deteriorates fast and unpredictably, is easy to miss because the fetus is not small, and is responsible for a disproportionate share of unexplained term stillbirth. Distinguishing the two — and applying the correct surveillance and delivery logic to each — is the HOTS skill this objective tests.
The first essential distinction is between a small-for-gestational-age (SGA) fetus that is constitutionally small but healthy, and a growth-restricted fetus that is pathologically failing to grow. Not every small fetus is compromised, and — critically — not every compromised fetus is small. This is why modern definitions combine size with functional (Doppler) and growth-trajectory criteria rather than relying on a single estimated-weight centile.
Figure J21.1 — Early- and late-onset FGR are distinct placental-insufficiency phenotypes separated at 32 weeks, with different Doppler clues and timing risks.
Core knowledge
Definitions
- SGA: estimated fetal weight (EFW) or abdominal circumference (AC) below the 10th centile. A statistical, not a pathological, label.
- FGR: failure to achieve growth potential, usually evidenced by EFW/AC below a low centile combined with Doppler abnormality and/or slowing of growth across serial scans. The widely used Delphi consensus definition of FGR (referenced in RCOG GTG 31) separates early from late forms at 32 weeks and grades severity by EFW/AC centile (e.g. <3rd centile as a stand-alone criterion) plus umbilical-artery, uterine-artery and middle-cerebral-artery Doppler thresholds. Quote the framework, not invented cut-offs.
- Symmetrical vs asymmetrical: an older but still useful heuristic. Asymmetrical restriction (head-sparing, AC reduced more than head circumference) suggests placental insufficiency of later onset, where the fetus redistributes cardiac output to the brain. Symmetrical restriction (head and body equally small) is more typical of an early insult — aneuploidy, congenital infection or very early severe placental disease. The dichotomy is imperfect and should not override Doppler-based phenotyping.
Shared cause, divergent biology
Both phenotypes usually stem from placental insufficiency — see placental-insufficiency-response for the fetal physiology. The placenta fails to deliver enough oxygen and substrate, and the fetus mounts an adaptive response that ends, if unrelieved, in hypoxia, acidaemia and death.
Early-onset FGR reflects severe, often global, placental maldevelopment — failure of trophoblast invasion and spiral-artery remodelling, high-resistance uteroplacental flow, and frequently shared pathophysiology with pre-eclampsia (the two coexist commonly because both arise from defective placentation; see pre-eclampsia-and-hellp and hypertension-in-pregnancy). A large proportion of the villous tree is affected, so the umbilical artery — which integrates resistance across the whole placenta — shows abnormality early. The hallmark Doppler cascade is therefore detectable and sequential:
- Rising umbilical-artery (UA) resistance (raised PI), then
- Absent end-diastolic flow (AEDF), then
- Reversed end-diastolic flow (REDF) in the UA;
- Middle-cerebral-artery (MCA) vasodilation (falling PI) — "brain-sparing" — and a falling cerebroplacental ratio (CPR);
- Late, ominous deterioration of ductus venosus (DV) waveforms (absent or reversed a-wave) and ultimately abnormal CTG / biophysical profile.
This orderly progression is what makes early-onset FGR manageable: it gives the clinician a staged warning system over days to weeks within which to optimise neonatal readiness (corticosteroids, magnesium sulphate, transfer to an appropriate level of care) before delivering.
Late-onset FGR reflects milder, often focal or late-developing placental dysfunction. Because a smaller fraction of villous resistance is affected, the umbilical artery Doppler is frequently normal even in a genuinely compromised fetus — this is the great trap. The fetus that is mildly hypoxic at term has little reserve, and the predominant signal is cerebral redistribution: a low MCA PI and a low CPR may be the only Doppler abnormality. Deterioration to CTG abnormality or demise can be rapid and poorly predicted, which is why late-onset FGR contributes so heavily to unexplained term stillbirth. There is no long Doppler cascade to lean on; surveillance relies on CPR, growth velocity, liquor volume and fetal movements rather than the UA alone. See liquor-volume-abnormalities and decreased-fetal-movements.
Figure J21.2 — Doppler surveillance in FGR: early disease follows a UA-led cascade, while late disease may show only MCA/CPR redistribution before rapid deterioration.
Risk factors and other causes
Placental insufficiency is the commonest mechanism, but always consider:
- Maternal vascular disease: chronic hypertension, pre-eclampsia, renal disease, autoimmune disease (antiphospholipid syndrome).
- Maternal substrate/oxygen delivery: severe anaemia, cyanotic heart disease, high-altitude residence, malnutrition.
- Toxins: smoking, alcohol, cocaine — see substance-use-in-pregnancy.
- Fetal causes: aneuploidy (especially with early symmetrical FGR + structural anomaly), confined placental mosaicism, congenital infection (CMV, toxoplasmosis, rubella, syphilis), and multiple pregnancy with selective FGR — see multiple-pregnancy.
- South African context: maternal HIV and its treatment, syphilis (still a measurable cause of growth restriction and stillbirth where antenatal screening is missed), TB, and the high background prevalence of hypertensive disease all feed the FGR burden. HIV itself and some antiretroviral exposures are associated with lower birth weight, so booking, screening and ART optimisation matter — see hiv-in-pregnancy.
Assessment
Identifying the at-risk fetus
The screening backbone in SA primary and district care is accurate dating at booking (antenatal-booking, gestational-age-assessment), risk stratification (high-risk-pregnancy-risks), and serial symphysis-fundal height (SFH) measurement plotted on a customised or population growth chart. SFH is cheap and universal but insensitive; a static or falling SFH, or a measurement lagging behind gestation, mandates referral for ultrasound. The NDoH Maternal and Perinatal Care Guideline (NDoH, 2024) structures this primary-level screening and the referral thresholds up the levels of care.
Ultrasound — the diagnostic and surveillance tool
- Biometry: head circumference, abdominal circumference and femur length → EFW and centile. Serial scans (conventionally ≥2–3 weeks apart to avoid measurement-error noise) define growth velocity — a fetus dropping across centiles is restricted even if still above the 10th.
- Umbilical artery Doppler: the evidence-based surveillance tool in established FGR. Its abnormality drives early-onset management. Reduced/absent/reversed end-diastolic flow is a graded danger signal.
- MCA Doppler and CPR: most useful in late-onset/near-term FGR, detecting redistribution when the UA is still normal. A low CPR is the most sensitive single marker of the at-risk late fetus.
- Ductus venosus Doppler: the late, near-terminal marker in early-onset FGR; an absent/reversed a-wave signals imminent acidaemia and is a key delivery trigger in the very preterm fetus.
- Uterine artery Doppler: in the second trimester, persistent high resistance with notching predicts placental disease and the early-onset phenotype.
- Amniotic fluid volume (deepest pool / AFI) and biophysical profile add functional information; oligohydramnios is a worrying adjunct. See liquor-volume-abnormalities.
- CTG: late marker of compromise — reduced variability, decelerations. Computerised CTG short-term variation is used in some centres to time delivery in the very preterm. See ctg-interpretation and fetal-monitoring-methods.
The ISUOG Doppler standards underpin technique and reference ranges; do not invent threshold numbers — use the centre's validated reference charts.
Working out why
Once FGR is confirmed, the assessment must answer: is this placental insufficiency, or is it a fetus that is small for a non-placental reason that changes management entirely?
- Detailed anomaly ultrasound — a structural anomaly with early symmetrical FGR shifts suspicion towards aneuploidy and warrants offering invasive testing/karyotype (link down-syndrome-counselling for counselling principles).
- Infection screen where indicated (CMV, toxoplasmosis, syphilis serology, and ensure HIV status known and managed) — see infections-in-pregnancy.
- Maternal evaluation for pre-eclampsia: blood pressure, urinalysis/PCR, bloods; placental disease frequently announces itself first as FGR.
Management
There is no proven antenatal treatment that reverses established placental FGR. Aspirin reduces incidence in high-risk women (prophylaxis), but once FGR is established the only definitive intervention is delivery at the optimal time. Management is therefore the art of surveillance and timing: balancing the risk of leaving a hypoxic fetus in utero against the risk of iatrogenic prematurity. This balance is struck completely differently in the two phenotypes.
Figure J21.3 — Management of FGR centres on surveillance, right-level referral, fetal optimisation and timed delivery rather than reversal of established placental disease.
Prevention and shared antenatal care
- Low-dose aspirin for women at high risk of placental disease, started in the first trimester (from ~12 weeks). NICE NG133 supports aspirin 75–150 mg from 12 weeks in women at risk of pre-eclampsia (the same population at risk of placental FGR); the SA Maternity Guideline aligns. It is a preventive, not a treatment, measure.
- Stop smoking; optimise maternal disease; screen and treat syphilis and optimise HIV/ART.
- Once FGR is diagnosed: antenatal corticosteroids for fetal lung maturation if delivery before ~34 weeks is anticipated (RCOG GTG 74), and magnesium sulphate for fetal neuroprotection when delivery before ~30–32 weeks is imminent. See preterm-birth-and-pprom.
- Deliver at the appropriate level of care. A fetus likely to need delivery before ~32 weeks, or with abnormal UA Doppler, belongs at a regional/tertiary unit with neonatal intensive care — in-utero transfer is far safer than ex-utero transfer of a sick preterm neonate. This is the central SA referral decision and is set out in the levels-of-care framework of the NDoH guideline.
Early-onset FGR (<32 weeks)
The problem is prematurity versus progressive hypoxia, played out over a relatively predictable Doppler cascade.
- Surveillance is UA-Doppler-led, intensifying as flow deteriorates — from weekly, to twice-weekly, to daily/continuous monitoring with AEDF/REDF.
- Add ductus venosus Doppler and CTG as the fetus deteriorates; the DV a-wave and CTG variability become the delivery triggers at the earliest gestations, weighing each extra day of maturity against acidaemia risk (the principle tested by the TRUFFLE study, referenced in RCOG GTG 31, which compared timing strategies in early FGR).
- Give corticosteroids and magnesium sulphate before delivery, and ensure NICU readiness.
- Mode of delivery: with significant Doppler abnormality (AEDF/REDF) and a preterm, compromised, often unfavourable cervix, the fetus tolerates labour poorly — caesarean section is usual. See safe-caesarean-technique.
- The honest message to parents: at the earliest gestations the limiting factor may be neonatal viability, not the obstetric decision, and counselling is shared with neonatology.
Late-onset FGR (≥32 weeks)
The problem is a near-term fetus with little reserve that can deteriorate fast, often with a normal UA Doppler.
- Surveillance leans on MCA/CPR, growth velocity, liquor volume, CTG and fetal movements rather than the UA.
- The threshold to deliver is low because the cost of prematurity is small at these gestations and the cost of waiting (stillbirth) is real. Abnormal CPR, oligohydramnios, static growth or any CTG concern → deliver.
- Timing follows centre/guideline protocols (the DIGITAT trial, referenced in RCOG GTG 31, supports a low threshold for induction at term in suspected FGR).
- Mode of delivery: many late-FGR fetuses with reassuring Doppler can undergo induction of labour with continuous CTG monitoring, accepting a raised caesarean rate for intrapartum compromise. A fetus already showing redistribution has reduced intrapartum reserve — counsel accordingly and have a low threshold for caesarean.
Intrapartum and neonatal
The growth-restricted fetus tolerates the hypoxic stress of contractions poorly (see contractions-fetal-oxygenation). Deliver with continuous electronic fetal monitoring (ctg-interpretation); anticipate meconium and a depressed neonate; ensure a resuscitation-competent person is present (neonatal-resuscitation, initiation-of-respiration, neonatal-transition). The growth-restricted neonate is at risk of hypothermia, hypoglycaemia and polycythaemia — keep warm, feed early, monitor glucose.
Emergency drills — unmistakable
The growth-restricted fetus is a fetus on the edge. Treat these as obstetric emergencies:
REVERSED end-diastolic flow (REDF) in the umbilical artery, OR absent/reversed ductus venosus a-wave, OR a pathological CTG (reduced variability with decelerations) in an FGR fetus = imminent fetal death. Drill: call for senior obstetric help NOW → confirm with repeat Doppler/CTG → give antenatal corticosteroids and magnesium sulphate if <34 / <32 weeks and not already given → expedite delivery, usually by caesarean at a unit with neonatal capability → alert neonatal team. Do not send such a patient home or defer to a routine list.
Abnormal SFH or reduced fetal movements at a primary clinic = refer for ultrasound the same day, not "next antenatal visit." Decreased fetal movements in a known FGR pregnancy (decreased-fetal-movements) is a red flag for imminent compromise — do CTG and arrange senior review immediately.
In-utero transfer beats ex-utero transfer. If a fetus is likely to deliver before ~32 weeks or has abnormal UA Doppler, transfer the mother to a regional/tertiary unit with NICU before delivery, not the sick neonate afterwards.
Red flags / pitfalls
- Confusing SGA with FGR — and the opposite trap, the "normal-sized" growth-restricted fetus. A fetus on the 15th centile that has crossed down from the 50th, with a low CPR, is restricted and at risk despite being above the 10th centile. Relying on a single EFW centile misses late-onset FGR.
- Trusting a normal umbilical-artery Doppler at term. In late-onset FGR the UA is frequently normal. Use CPR/MCA, growth velocity and liquor — and a low delivery threshold.
- Wrong dates. FGR cannot be diagnosed reliably on uncertain dating. Confirm gestation from early ultrasound; otherwise "FGR" may be a mis-dated normal fetus, or true FGR may be missed. See gestational-age-assessment.
- Forgetting non-placental causes. Early, symmetrical FGR with a structural anomaly is aneuploidy/infection until proven otherwise — investigate before assuming placental insufficiency, because the prognosis and counselling differ entirely.
- Iatrogenic prematurity vs stillbirth — getting the balance backwards. Delivering an early-onset FGR fetus too soon causes avoidable prematurity morbidity; waiting too long in a late-onset FGR fetus causes avoidable stillbirth. The error direction differs by phenotype.
- Delivering the very preterm FGR fetus at the wrong level of care. Without NICU, neonatal mortality is high — arrange in-utero transfer.
- Missing the coexisting pre-eclampsia. FGR and pre-eclampsia share placental pathology; check the mother's BP, urine and bloods at every FGR encounter (pre-eclampsia-and-hellp).
- Treating FGR as fixable in utero. There is no medical therapy that makes the placenta work better once FGR is established; surveillance + timed delivery is the only lever. Aspirin is prevention, given too late to help an already-restricted fetus.
Evidence anchors
- RCOG Green-top Guideline No. 31 — The Investigation and Management of the Small-for-Gestational-Age Fetus / Growth-Restricted Fetus. Primary guideline for definitions, the early/late split, Doppler surveillance and timing of delivery; references the Delphi consensus FGR definition, and the TRUFFLE (timing in early FGR) and DIGITAT (induction vs expectant at term) trials.
- ISUOG Practice Guidelines — Doppler ultrasound (umbilical artery, MCA, ductus venosus, cerebroplacental ratio). Technique and reference standards for the surveillance described above.
- SA National Integrated Maternal and Perinatal Care Guideline (NDoH, 2024), NDoH. SA source of truth for SFH screening, referral thresholds, levels of care and in-utero transfer.
- NICE NG133 — Hypertension in Pregnancy (2019). Low-dose aspirin 75–150 mg from 12 weeks in women at risk of pre-eclampsia (the population at risk of placental FGR); magnesium sulphate use.
- RCOG GTG 74 — Antenatal Corticosteroids. Fetal lung maturation when preterm delivery is anticipated.
- RCOG GTG 57 — Reduced Fetal Movements. Decreased movements as a red flag in the at-risk fetus.
- Saving Mothers / NCCEMD and the SA perinatal audit. FGR-related fetal and early-neonatal deaths and the avoidable-factor analysis that frames SA surveillance and referral practice.
Where exact centile or Doppler PI cut-offs are needed, use the centre's validated reference charts and the Delphi/RCOG GTG 31 criteria — do not memorise invented numbers.