Clinical overview
Recurrent pregnancy loss (RPL) is the loss of two or more pregnancies, and it is one of the most emotionally charged consultations in obstetrics. The couple in front of you have usually already grieved several times, may have been given conflicting advice, and frequently carry a heavy load of guilt and self-blame. Your task as a registrar is two-fold and inseparable: a systematic, evidence-based evaluation to find a treatable cause, and compassionate, honest counselling — because in roughly half of couples no cause is ever identified, and even then the prognosis for a future live birth is good.
The exam objective is to evaluate a woman with a history of RPL. That verb matters: this chapter is weighted to assessment — what to ask, what to examine, what to test, how to interpret the results, and how to translate them into a management plan. RPL is not a single disease but a clinical phenotype with a heterogeneous set of contributors: genetic, anatomical, endocrine, thrombophilic (autoimmune), and a large idiopathic group. The most important discipline is to investigate appropriately, not exhaustively — to order the tests with evidence behind them, interpret them in context, and avoid the twin traps of over-investigation (false-positive findings driving unnecessary, sometimes harmful treatment) and under-investigation (missing the eminently treatable antiphospholipid syndrome). In the South African setting you must also frame loss against the background of the national maternal-mortality picture and ensure you are not missing an evolving complication such as an ectopic pregnancy or sepsis at the time of an acute loss.
Core knowledge
Definitions
There is no single global definition, and the exam expects you to know the differences:
- RCOG (GTG 17, recurrent miscarriage) traditionally defines recurrent miscarriage as three or more consecutive first-trimester losses, while accepting that investigation may reasonably begin after fewer.
- ESHRE (Recurrent Pregnancy Loss guideline, 2022 update) defines RPL as the loss of two or more pregnancies and explicitly includes non-visualised pregnancy losses (biochemical and pregnancy-of-unknown-location losses), while excluding ectopic and molar pregnancies. ESHRE recommends that the losses need not be consecutive.
Hedge this in your mind: the trend, reflected in the ESHRE 2022 position, is to start investigating after two clinically recognised losses, particularly where the woman is older or anxious, because the yield is reasonable and the wait for a third loss is hard to justify.
A useful subdivision is primary RPL (no previous live birth) versus secondary RPL (losses after a previous viable pregnancy), and early (pre-12 weeks, the majority) versus late (12 weeks to viability) loss — the latter shifting suspicion toward cervical, anatomical, and thrombotic causes.
Epidemiology and the chance factor
Sporadic miscarriage is common — affecting roughly one in five recognised pregnancies — so two or three losses can occur by chance alone. This is the single most important concept to grasp: most isolated early losses are due to sporadic embryonic aneuploidy, and the probability of a chromosomally abnormal loss falls with each successive miscarriage, whereas the probability of an underlying maternal cause rises. Advancing maternal age is the dominant risk factor for loss, driven by the exponential rise in oocyte aneuploidy; paternal age contributes more modestly. This frames every counselling conversation: even after three unexplained losses, the prognosis for a subsequent live birth with supportive care is encouraging.
Established and probable causes
Genetic. Two patterns. First, embryonic aneuploidy — sporadic and the commonest single cause of any individual early loss. Second, a parental structural chromosomal rearrangement (most often a balanced reciprocal or Robertsonian translocation) found in a small percentage of couples; the carrier is phenotypically normal but produces unbalanced gametes, causing repeated loss.
Anatomical. Congenital uterine anomalies (notably the septate uterus, the anomaly most consistently associated with loss) and acquired lesions — submucosal fibroids, intrauterine adhesions (Asherman syndrome), and endometrial polyps. Cervical insufficiency classically presents as painless mid-trimester loss or rapid preterm delivery; see cervical-cerclage.
Endocrine and metabolic. Overt and poorly controlled diabetes and overt thyroid disease are associated with loss; thyroid autoimmunity and subclinical hypothyroidism are areas of genuine uncertainty. Polycystic ovary morphology and the associated metabolic milieu are linked epidemiologically. Obesity independently increases miscarriage risk.
Antiphospholipid syndrome (APS). This is the single most important treatable cause because effective therapy exists. APS is the association of persistent antiphospholipid antibodies with defined clinical events including recurrent early loss, one or more late losses, or severe early pre-eclampsia/placental insufficiency. The laboratory criteria require lupus anticoagulant, anticardiolipin antibodies, or anti-β2-glycoprotein-I antibodies, with the critical caveat that the abnormal result must be confirmed on a repeat sample, classically ≥12 weeks apart (standard teaching), because transient positivity is common after infection.
Inherited thrombophilias. Heritable thrombophilias (e.g. factor V Leiden, prothrombin gene variants, protein C/S deficiency) have a weak and inconsistent association with RPL, and — crucially — anticoagulation has not been shown to improve live-birth rates in this group. This drives the modern position of not routinely screening for them in the RPL work-up outside a research or specific thrombotic context.
Figure J13.1 — Definitions, chance aneuploidy, age risk and the main evidence-based recurrent pregnancy loss cause buckets, with APS highlighted as the key treatable diagnosis.
Assessment
This is the heart of the objective. Evaluate the couple together wherever possible — both partners contribute genetically and the loss is a shared event.
History
Take a meticulous, chronological reproductive history:
- Each pregnancy in turn: gestation at loss, whether the loss was visualised on ultrasound (fetal pole/heart seen) versus biochemical, the mode (spontaneous, surgical, medical), and whether products of conception were sent for histology/karyotype.
- Pattern: early versus late; the painless mid-trimester loss/rupture story that suggests cervical insufficiency; the late loss with growth restriction or severe pre-eclampsia that points toward APS or placental disease — see pre-eclampsia-and-hellp and intrauterine-growth-restriction.
- Symptoms of thrombophilia/autoimmunity: prior venous or arterial thrombosis, features of SLE (photosensitive rash, arthralgia, serositis), recurrent fetal growth restriction.
- Endocrine clues: menstrual pattern, galactorrhoea, symptoms of thyroid disease, known diabetes, weight change.
- Family history of recurrent loss, thrombosis, or known chromosomal rearrangement.
- Exposures and lifestyle: smoking, alcohol, recreational drugs, caffeine, occupational exposures, BMI. See substance-use-in-pregnancy.
- Medical and surgical history, including previous uterine surgery (caesarean, myomectomy, evacuation — adhesion risk).
- HIV and infection history. In the South African context HIV status is essential: counsel and test per the national programme. While there is no established causal link between well-controlled HIV and recurrent first-trimester loss, untreated HIV, syphilis, and other infections matter for the current and next pregnancy, and the booking screen is the safety net — see hiv-in-pregnancy and infections-in-pregnancy.
Examination
- General: BMI, blood pressure, signs of endocrine disease (thyroid, hyperandrogenism, acanthosis nigricans), and features of connective-tissue disease.
- Pelvic examination for uterine size/shape, cervical assessment, and signs of infection.
Investigations — what to order and why
Order tests with evidence behind them; resist the urge to "screen for everything".
1. Cytogenetic analysis of the products of conception. Where feasible, karyotyping or array analysis of the miscarried tissue is among the highest-yield investigations: an aneuploid result reassures (a sporadic chance event, good prognosis) and an abnormal result that is not a simple trisomy may flag a parental rearrangement, directing parental karyotyping. Availability is the limiting factor in the SA public sector — flag this and prioritise where the service exists.
2. Parental peripheral-blood karyotyping. Indicated particularly when products of conception show an unbalanced structural abnormality, or where there is a relevant family history. Routine universal parental karyotyping has a low yield; targeted testing guided by the loss cytogenetics is the evidence-aligned approach. Refer carriers for genetic counselling — see principles-of-inheritance.
3. Antiphospholipid antibody screen. Test lupus anticoagulant, anticardiolipin, and anti-β2-glycoprotein-I antibodies, and — this is the examiner's favourite point — repeat any positive result after the standard interval (classically ≥12 weeks) before diagnosing APS. Do not anticoagulate on a single positive result.
4. Pelvic ultrasound (transvaginal). First-line for uterine cavity and morphology, ovarian assessment, and to detect fibroids/polyps. Where a uterine anomaly is suspected, three-dimensional ultrasound is the modern non-invasive tool of choice for classifying the cavity (e.g. distinguishing septate from bicornuate); hysteroscopy, saline-infusion sonography, or MRI are used to confirm and treat as needed.
5. Endocrine tests, directed by history. Thyroid function (and thyroid peroxidase antibodies where indicated), and screening for diabetes where there are risk factors. Treat overt thyroid and glycaemic disease; the evidence for treating subclinical thyroid abnormality or thyroid autoimmunity specifically to improve live birth is uncertain — counsel honestly.
6. Inherited thrombophilia screen — generally NOT routine. Because anticoagulation does not improve outcomes in heritable thrombophilia with RPL, do not order these as a reflex. Reserve testing for a personal/strong family history of VTE, late losses with placental pathology, and always in discussion with haematology.
A succinct way to remember the core evidence-based panel:
| Domain | First-line test | Note |
|---|---|---|
| Autoimmune | Lupus anticoagulant, aCL, anti-β2-GPI | Repeat positive ≥12 wk before diagnosis |
| Anatomical | Transvaginal ± 3D ultrasound | Hysteroscopy/MRI to confirm/treat |
| Genetic (loss) | POC karyotype/array (if available) | Highest-yield where accessible |
| Genetic (parents) | Karyotype if POC unbalanced / FHx | Refer for genetic counselling |
| Endocrine | TFTs ± TPO; glucose if at risk | Treat overt disease |
| Inherited thrombophilia | Not routine | Only with VTE history + haematology |
Figure J13.2 — Evidence-based RPL work-up from couple consultation through core investigations, South African booking-screen safety checks and tests to avoid.
Investigations should be done between pregnancies where possible, and many results (anticoagulant assays, thyroid antibodies) are altered by pregnancy itself — interpret with caution if drawn during a current pregnancy.
Management
Management follows the cause; where none is found, supportive care and accurate counselling are themselves the treatment.
Antiphospholipid syndrome
This is the scenario where you can change the outcome, so know it cold. In a woman meeting clinical and confirmed laboratory criteria for obstetric APS, the established intervention is low-dose aspirin plus a prophylactic-dose heparin (low-molecular-weight heparin) commenced in early pregnancy, which improves live-birth rates compared with no treatment or aspirin alone. Doses and timing should follow the current guideline and local protocol — confirm against the prevailing RCOG/specialist haematology guidance rather than quoting from memory. These pregnancies need specialist, often joint obstetric–haematology, antenatal care with surveillance for pre-eclampsia and growth restriction, because APS is a placental disease.
Genetic / parental chromosomal rearrangement
Refer to clinical genetics. Options to discuss include natural conception with prenatal diagnosis (CVS/amniocentesis) in a future pregnancy, and IVF with preimplantation genetic testing, recognising that the cumulative live-birth rate with expectant management is often still favourable. The role of routine preimplantation testing for aneuploidy in unexplained RPL is not established — counsel against assuming it is a solution.
Anatomical
A uterine septum may be resected hysteroscopically, though the evidence that metroplasty improves live birth is less robust than once assumed — counsel about uncertainty and involve a unit experienced in the procedure. Submucosal fibroids distorting the cavity and significant intrauterine adhesions are reasonable to treat. For a history consistent with cervical insufficiency, discuss surveillance and cervical cerclage — see cervical-cerclage — and the role of progesterone in selected women.
Endocrine and lifestyle
Optimise glycaemic control before conception in diabetes, treat overt thyroid disease, and address modifiable factors: smoking cessation, alcohol avoidance, weight optimisation, and folate supplementation. These are universally applicable and harmless.
Unexplained RPL — what not to do
For unexplained RPL, the message is one of realistic optimism and restraint:
- Provide supportive care with reassurance and early-pregnancy scanning in the next pregnancy — psychological support and dedicated early-pregnancy follow-up are associated with good outcomes.
- Progesterone: the evidence is nuanced. Routine progesterone for all RPL is not supported, but vaginal progesterone has a place in the woman with RPL and early-pregnancy bleeding — discuss with the patient and follow current national/RCOG guidance; see progesterone-in-pregnancy.
- Do not prescribe heparin or aspirin for unexplained RPL (no APS) — trials do not show benefit and there is exposure to harm.
- Avoid unproven "add-ons": empirical steroids, immunoglobulin, leucocyte immunisation, and routine inherited-thrombophilia anticoagulation have no good evidence and may harm.
South African context
Frame the work-up against SA realities. The National Integrated Maternal and Perinatal Care Guideline (NDoH, 2024) governs antenatal care and the booking screen (HIV, syphilis, anaemia) into which the next pregnancy must be slotted; ensure the woman is established in antenatal care at the appropriate level of care and referred to a regional/tertiary unit for the RPL work-up where specialist genetic, immunology, and 3D-ultrasound services exist. Laboratory access (POC karyotyping, antiphospholipid assays) is uneven across districts — be explicit with the couple about what is and is not available, and prioritise the highest-yield, locally feasible tests. HIV must be addressed at every contact; an acute loss is also an opportunity to confirm status and link to care. The leading SA maternal-death causes (obstetric haemorrhage, hypertension, and non-pregnancy-related infection including HIV — Saving Mothers/NCCEMD) are a reminder that the acute loss event carries its own risks of haemorrhage and sepsis, and that the next pregnancy in a woman with APS or late loss is genuinely high-risk and belongs in a higher level of care.
The acute loss — emergency drill
Evaluation of RPL is usually elective, but a woman with a history of RPL may present mid-loss, and you must not let the chronic label blind you to an acute emergency. If she presents bleeding and/or in pain, treat it as a fresh obstetric emergency.
DRILL — bleeding/collapse in a presenting loss: Call for help. Airway–Breathing–Circulation. Two large-bore IV cannulae; bloods including FBC, group-and-save/crossmatch, βhCG. Resuscitate with crystalloid and blood as needed. Exclude ectopic pregnancy in any positive-βhCG woman with pain/bleeding — it is the diagnosis that kills. Assess for incomplete miscarriage with haemorrhage (uterine evacuation, uterotonics) and for sepsis (fever, tachycardia, offensive loss → cultures and broad-spectrum antibiotics per protocol, source control). Give anti-D to a non-sensitised RhD-negative woman per the isoimmunisation protocol. Stabilise first; the RPL work-up waits.
Figure J13.3 — Cause-directed RPL management and the acute presenting-loss drill, separating confirmed APS treatment from unexplained RPL restraint and emergency stabilisation.
Red flags / pitfalls
- Missing antiphospholipid syndrome — the one treatable cause. Always test for it, and always confirm a positive on a repeat sample before labelling and anticoagulating.
- Diagnosing APS on a single positive result — transient antiphospholipid positivity after infection is common; one result is not a diagnosis.
- Forgetting ectopic pregnancy in a presenting loss with a positive βhCG — a "miscarriage" with pain and bleeding may be an ectopic. See ectopic-pathophysiology.
- Over-investigation: routinely screening for inherited thrombophilias and then anticoagulating — no live-birth benefit, real harm.
- Empirical "add-ons" (steroids, IVIG, paternal leucocyte immunisation, routine PGT-A) sold as solutions for unexplained RPL — no good evidence; counsel against.
- Over-treating subclinical thyroid disease as if proven to fix RPL — honesty about uncertainty beats false reassurance.
- Omitting the products-of-conception karyotype when the service exists — it is high-yield and reframes prognosis.
- Neglecting the psychological dimension — failing to acknowledge grief, or quoting cold statistics without support, is a clinical failure, not just a courtesy lapse.
- Anti-D oversight — forgetting prophylaxis in an RhD-negative woman during an acute loss.
- Counselling the woman alone when the couple should be assessed together (parental genetics, shared decision-making).
Evidence anchors
- ESHRE Guideline: Recurrent Pregnancy Loss (2022 update) — defines RPL as ≥2 losses (need not be consecutive, includes non-visualised losses); frames the evidence-based investigation panel and the restrained position on inherited thrombophilia screening and unproven add-ons.
- RCOG Green-top Guideline No. 17 — Recurrent Miscarriage — traditional ≥3-consecutive-loss definition; antiphospholipid antibody testing with confirmatory repeat; aspirin + heparin for confirmed obstetric APS.
- NICE NG126 — Ectopic Pregnancy and Miscarriage: diagnosis and initial management (2019) — diagnosis and initial management of the acute loss and exclusion of ectopic; anti-D considerations in early pregnancy loss (cross-check current anti-D guidance, NICE TA156 / BSH).
- National Integrated Maternal and Perinatal Care Guideline (NDoH, 2024), SA NDoH — SA source of truth for booking screen, levels of care, referral pathways, HIV testing.
- Saving Mothers / NCCEMD (latest triennium) — SA maternal-mortality context (haemorrhage, hypertension, HIV-related infection) relevant to the acute loss and high-risk next pregnancy.
- South African National HIV / ART Consolidated Guidelines (2023) — HIV testing and treatment framing for every contact.
Notes on uncertainty: RPL definitions differ between RCOG (≥3 consecutive) and ESHRE (≥2). The "repeat antiphospholipid testing ≥12 weeks apart", standard miscarriage incidence (~1 in 5), and specific aspirin/heparin doses are stated as standard teaching and should be confirmed against the current RCOG/ESHRE and local protocols before prescribing — no doses are quoted here for that reason. The benefit of septum resection, progesterone, and treatment of subclinical thyroid/thyroid autoimmunity for RPL specifically remains contested and is hedged accordingly.