Critically appraise antihypertensive choice and the management of postpartum and mild hypertension in pregnancy

In one line

Treating non-severe hypertension to a target of <140/90 mmHg now improves outcomes (CHAP); the antihypertensive choice matters less than treating to target, controlling severe levels promptly, and not stopping at delivery — the postpartum window is where most preventable maternal deaths still occur.

This chapter assumes the diagnostic and first-line groundwork in hypertension in pregnancy and pre-eclampsia & HELLP basics; it spends its words on the drug-by-drug argument, the postpartum and mild-disease judgement calls, and the primary-literature appraisal. The expectant-vs-deliver and magnesium-and-fluid craft of severe disease lives in pre-eclampsia-early-onset-severe.

Assessment

Assume the Intermediate groundwork — definitions, pre-eclampsia diagnosis and HELLP — from hypertension in pregnancy and pre-eclampsia and HELLP. The decisive question is whether this is hypertension that merely needs a number lowered, or hypertension that signals an evolving multisystem disease.

• Classify before you treat. Chronic hypertension (pre-existing or <20 weeks), gestational hypertension (≥20 weeks, no proteinuria/organ involvement), and pre-eclampsia behave differently and carry different antihypertensive logic. A booking BP is gold — without it, a 20-week reading of 140/95 is uninterpretable.
• Severity is the trigger. Severe hypertension = ≥160/110 mmHg (SA NDoH and international consensus). This is an emergency irrespective of symptoms — the threshold above which maternal intracerebral haemorrhage risk climbs steeply, and the single commonest avoidable cause of direct maternal death.
• Look for the disease behind the number. Headache, visual disturbance, epigastric/RUQ pain, brisk reflexes, oliguria. Bloods: FBC (platelets), U&E/creatinine, ALT/AST, urate, urine protein:creatinine ratio. Rising urate or creatinine, falling platelets or transaminitis reclassify "mild hypertension" as severe disease in a normotensive envelope.
• Postpartum is not the all-clear. BP typically peaks days 3–6 postpartum. De-novo postpartum pre-eclampsia and eclampsia occur, so a woman discharged normotensive on day 1 still needs a plan and a follow-up BP.

The advanced assessment — the subtype distinction that changes the drug

Beyond recognising the clinical picture, the task is to classify the hypertensive subtype well enough to predict its trajectory and pick the right agent, and to recognise the atypical presentations where the "obstetric" label is wrong.

• Chronic hypertension is the highest-risk substrate, not just a baseline number. Pre-existing hypertension carries a roughly threefold increased risk of superimposed pre-eclampsia, and that risk is the reason CHAP-style treat-to-target matters most in this group. Distinguish superimposed pre-eclampsia (new/worsening proteinuria, organ dysfunction, a sudden escalation of BP requirements, rising urate or falling platelets) from chronic hypertension worsening with gestation — the first mandates the full pre-eclampsia pathway (magnesium, delivery planning), the second is a maintenance-dose problem. In a woman who books late with no first-trimester BP, hypertension plus proteinuria after 20 weeks is treated as pre-eclampsia until proven otherwise — you cannot retrospectively "un-diagnose" it.
• "White-coat", masked and gestational-then-resolving phenotypes. White-coat hypertension (raised in clinic, normal on ambulatory/home readings) still carries an elevated pre-eclampsia risk and is not benign; masked hypertension (normal in clinic, raised at home) is the trap that under-treats. Where home/ambulatory BP is available it refines classification, but in most SA district settings the pragmatic answer is serial standardised clinic readings with a correctly sized cuff.
• Severity is a trajectory, not a single reading. A BP creeping from 138/88 to 150/96 over a week in a woman with rising urate is more concerning than an isolated 162/112 in an otherwise stable, asymptomatic chronic hypertensive — the first is an evolving syndrome, the second a number to control. Track the envelope of disease (BP trend + bloods trend + symptoms), not the worst single value.
• The secondary-hypertension mimic that inverts the drug choice. A young woman with paroxysmal severe hypertension, palpitations, sweating, headache and labile pressures that respond badly or paradoxically to a beta-blocker should raise phaeochromocytoma — rare, but lethal if missed and managed as ordinary pre-eclampsia. It changes the drug entirely: alpha-blockade (phenoxybenzamine) must be established before any beta-blockade, because beta-blockade against unopposed alpha stimulation precipitates a hypertensive crisis. Labetalol's fixed oral α:β antagonist ratio (≈1:7) gives inadequate alpha blockade, so labetalol monotherapy in an unrecognised phaeochromocytoma can itself trigger crisis — a clean example of why the right drug depends on the right diagnosis. Other secondary causes (renal artery stenosis, primary aldosteronism, coarctation, intrinsic renal disease) deserve thought in early, severe or treatment-resistant chronic hypertension.

Management

Frame the answer immediate → ongoing → long-term.

Immediate (severe hypertension, ≥160/110). Treat within minutes; do not wait for proteinuria. SA first line is oral nifedipine 10 mg (immediate-release, repeat at 20–30 min to a max of ~3 doses) or IV labetalol where available; IV hydralazine is the third option. Methyldopa has no role acutely — too slow. Severe hypertension and pre-eclampsia also mandate magnesium sulphate for seizure prophylaxis: lowering BP is not the same as preventing eclampsia (MAGPIE).

The named acute regimens — exact doses and how they differ

All three first-line acute agents are acceptable; the right one is the one you can give safely and fast with the access and stock you have. The differences are in titration step, onset, and the failure modes:

A few further points:

• The nifedipine–magnesium myth. The classic teaching warning against giving nifedipine with magnesium (fear of neuromuscular blockade and profound hypotension) rests on a tiny number of historical case reports. Retrospective series — and the fact that ~30% of women in MAGPIE received nifedipine after entry with no signal — show no excess of serious magnesium-related effects; the combination is used routinely. Do not withhold the fastest available antihypertensive for a near-mythical interaction while a woman sits at 170/115 — uncontrolled severe hypertension is the real killer.
• Target for the acute phase. Lower BP to roughly ≤150/100 (or systolic 130–150 / diastolic 80–100, the Easterling target band) — control, not normalisation. Overshoot drops placental perfusion and can cause non-reassuring fetal status; the goal is to pull the woman off the haemorrhage cliff, not to make her normotensive in an hour.
• Acute control is not seizure prevention. Antihypertensives prevent stroke; magnesium sulphate prevents eclampsia — the regimens and monitoring are in pre-eclampsia-early-onset-severe. Pre-eclampsia with severe features needs both, concurrently.

Ongoing (maintenance, antepartum). Treat to target <140/90 mmHg (CHAP). Verified SA-available oral options:

ACE inhibitors and ARBs are contraindicated antepartum (fetopathy, oligohydramnios, renal dysgenesis). Atenolol is avoided (fetal growth restriction).

Choosing and escalating the maintenance agent — the consultant logic

• Mechanism → side-effect → who it suits. Methyldopa is a central α₂-agonist: cheapest and the longest pregnancy safety record, but sedating and depressogenic — a poor choice to continue into a postpartum already at risk of mood disorder. Labetalol is a combined α/β-blocker: effective and the international first line, but β-blockade is unsafe in asthma and significant conduction disease, and chronic β-blockade is associated with fetal growth restriction (the reason atenolol specifically is avoided). Nifedipine (a dihydropyridine calcium-channel blocker) is a clean maintenance and acute agent with reflex tachycardia and ankle oedema as its costs. The agent should be matched to the woman — asthmatic → not labetalol; mood-disorder history → not methyldopa; FGR fetus → think twice about a β-blocker.
• Why ACE-I/ARB are antepartum-poison but postpartum-gold. The fetopathy (oligohydramnios, renal tubular dysgenesis, skull ossification defects, the late-pregnancy renal failure) is a fetal renin-angiotensin effect of second- and third-trimester exposure — which is exactly why enalapril becomes a first-line, breastfeeding-compatible choice the moment the fetus is delivered. The same drug class moves from contraindicated to preferred at delivery, driven by this mechanistic switch.
• Combine before you max out. If a single agent at a reasonable dose does not reach <140/90, add a second class (e.g. methyldopa + nifedipine, or labetalol + nifedipine) rather than pushing one agent to toxic doses — and re-ask whether escalating antihypertensive requirement is itself a sign of evolving pre-eclampsia rather than just "harder-to-control" chronic hypertension.

Delivery timing. For controlled chronic/gestational hypertension reaching term, WILL supports offering planned birth at 38+0–38+3 weeks — it did not increase poor maternal or neonatal outcomes and reduced pre-eclampsia (NNTB 10). Pre-eclampsia with severe features remains a deliver-after-stabilisation decision on its own merits.

Postpartum. BP often rises before it settles. Continue or start an antihypertensive and review daily until controlled. Postpartum the formulary opens up: enalapril and amlodipine become first-line, breastfeeding-compatible choices, and methyldopa should be stopped within 24–48 h of birth (postnatal depression). Aim to wean over weeks, not abruptly.

Postpartum hypertension in depth — the most under-managed window

• Why it peaks days 3–6 and why that matters operationally. The mobilisation of the extravascular fluid accumulated in pregnancy, the loss of the placental low-resistance circuit, NSAID and ergometrine exposure, and pain all push BP up after delivery — and a woman discharged normotensive on day 1 is discharged before her peak. The fix is structural: a documented antihypertensive plan, a named threshold to re-present, and a guaranteed follow-up BP. De-novo postpartum pre-eclampsia and eclampsia are real entities — a first-ever seizure or severe headache in the puerperium is eclampsia/PRES until excluded, not "just a headache".
• The drug switch, mechanistically. Off the methyldopa (mood), onto enalapril (RAS-effective, milk-safe — the fetal-renal contraindication is gone) and/or amlodipine (long-acting CCB, milk-compatible). Labetalol can be continued. Wean over weeks against home/clinic readings, not abruptly, to avoid rebound.
• The NSAID question. Older teaching said avoid all NSAIDs postpartum in hypertensive disorders for fear of raising BP. The evidence is more nuanced: a small 2017 RCT (Vigil-De Gracia) suggested a higher rate of moderately raised BP after vaginal birth with ibuprofen, but a 2018 double-blind RCT (Blue) and subsequent meta-analyses found no difference in severe-range BP, antihypertensive need or end-organ markers. NSAIDs are not absolutely contraindicated in hypertensive disorders of pregnancy and are reasonable for analgesia in non-severe disease, but use caution (and prefer paracetamol-based regimens) where BP is severe, control is precarious, or there is AKI/oliguria — the renal-prostaglandin effect matters most in the failing kidney. The blanket "HDP = never NSAIDs" rule does not hold against the data.
• Adjuncts that are evidence-supported but not yet SA standard. A short postpartum furosemide course (FoR BP) speeds BP recovery, chiefly in non-severe disease, and app-supported physician-optimised self-titration (POP-HT) achieves better long-term control and better cardiac remodelling. Both demand infrastructure most districts lack; they are the direction of travel, not current district standard of care.

Long-term. A hypertensive pregnancy is a cardiovascular-risk flag for life: arrange a 6-week BP and proteinuria check, document the diagnosis for future risk stratification, and counsel on aspirin prophylaxis and BP surveillance in the next pregnancy.

Long-term, recurrence and counselling — the consultant exit conversation

• Recurrence and the next-pregnancy plan. A hypertensive disorder recurs in a substantial minority — higher and earlier the more severe and the earlier the index episode. The actionable output is a written plan for the next pregnancy: low-dose aspirin from the first trimester (the argument and dose are in pre-eclampsia-prevention-aspirin), early booking BP, and a low threshold for review. Persistent hypertension or proteinuria beyond 6–12 weeks postpartum is chronic hypertension or underlying renal disease unmasked by pregnancy — it needs a medical referral and a workup, not reassurance.
• Cardiovascular risk is the real legacy. A hypertensive pregnancy roughly doubles long-term risk of chronic hypertension, ischaemic heart disease and stroke — pre-eclampsia is a recognised female-specific cardiovascular risk marker. The exit conversation hands the woman (and her primary-care clinician) a lifelong surveillance plan: BP, weight, lipids, glucose, and lifestyle. In SA this is best anchored to the existing district chronic-care clinics rather than an obstetric follow-up that ends at 6 weeks.

The evidence & the controversy

The defining shift is CHAP (2022). For two decades, CHIPS-era teaching was that tighter control mainly prevented severe hypertension without changing hard outcomes, so many units tolerated BPs up to 150s/90s. CHAP overturned the threshold: treating mild chronic hypertension to <140/90 cut the composite of pre-eclampsia with severe features, indicated preterm birth <35 weeks, abruption and perinatal death by ~18% without increasing small-for-gestational-age babies — the historical fear that drove under-treatment. CHIPS established that tight control is safe for the fetus; CHAP established that it is beneficial for the mother. Read together they retire the "less-tight" strategy for chronic hypertension.

The arithmetic: CHAP's primary composite was 30.2% (active) vs 37.0% (control) — an absolute risk reduction of 6.8%, so the NNT ≈ 1 / 0.068 ≈ 15 (the trialists report 14.7) to prevent one component of the composite, at an adjusted RR 0.82 (95% CI 0.74–0.92). That single-digit-teens NNT makes treat-to-<140/90 a cheap, high-value intervention well suited to a resource-limited system. Note the boundary: CHAP studied mild chronic hypertension; it does not license aggressive normalisation in pre-eclampsia with severe features (where over-treatment compromises placental perfusion) and CHIPS's caution about over-tight control still informs the acute severe scenario.

On which drug, the honest consultant answer is that head-to-head data are thin and largely outcome-neutral. Easterling's 2019 Lancet RCT (Nagpur, n=894, mean ~36.5 weeks) found all three oral agents got most women to the target band (SBP 120–150, DBP 70–100) within 6 h, with nifedipine retard achieving target most often (84%) vs labetalol (77%) and methyldopa (76%), and methyldopa most often needing a second agent (~19%). This is directly relevant to SA district hospitals where IV access and labetalol stock are unreliable — it legitimises an oral pathway for severe hypertension. So choice is driven by availability, contraindications and the clinical phase (acute vs maintenance vs postpartum), not a proven outcome hierarchy.

Postpartum remains under-managed and under-researched. FoR BP showed a short furosemide course speeds BP recovery, chiefly in non-severe disease, and POP-HT showed app-supported, physician-optimised self-titration achieves better long-term control (lower 24-h ambulatory BP at 6–9 months: ~114/71 vs ~120/77 mmHg) and better cardiac remodelling on its imaging substudy — but neither is yet SA standard of care, and both demand infrastructure most districts lack. The defensible position: prioritise the basics that demonstrably save lives — daily review, not stopping treatment at the door, and a guaranteed follow-up. A separate controversy worth flagging is the WILL caveat: it was stopped early at ~37% of its recruitment target, so its "offer birth at 38 weeks" signal rests on an underpowered dataset — present it as a reasonable, guideline-aligned offer for controlled hypertension at term, not a hard rule, and keep the uncertainty explicit.

Landmark trials & key evidence

Exam traps & red flags

• Treating the number, missing the disease. A "mild" BP with rising urate, falling platelets or epigastric pain is pre-eclampsia with severe features. Reclassify and escalate.
• Confusing BP control with seizure prevention. Antihypertensives prevent stroke; magnesium sulphate prevents eclampsia. Pre-eclampsia with severe features needs both.
• Reaching for methyldopa in a crisis. It is a maintenance agent — useless for ≥160/110. Use nifedipine or IV labetalol/hydralazine acutely.
• Stopping all treatment at delivery. BP peaks days 3–6. Continuing methyldopa postpartum (depression risk) instead of switching to enalapril/amlodipine is a second, common error.
• Discharging without a follow-up BP. No 6-week check is a missed chance to diagnose persistent hypertension and flag lifelong CV risk.
• ACE inhibitor or ARB antepartum — fetotoxic; safe and first-line only after delivery.
• Quoting CHIPS to justify loose control. Post-CHAP, the target is <140/90.
• Withholding nifedipine because magnesium is running. The feared nifedipine–magnesium interaction is near-mythical (a handful of case reports; no signal in MAGPIE) — do not leave a woman at 170/115 to avoid it.
• Sublingual nifedipine. 10 mg swallowed, never crushed/sublingual — the precipitous drop causes fetal compromise.
• Labetalol monotherapy in an unrecognised phaeochromocytoma. Its fixed α:β ratio gives inadequate alpha blockade — alpha-blockade (phenoxybenzamine) must precede any beta-blockade, or you precipitate a crisis. Think of secondary causes in early, severe or resistant hypertension.
• Reflexively banning all NSAIDs postpartum. The modern evidence does not support an absolute contraindication in non-severe disease; reserve caution for severe/precarious BP or AKI.

Worked viva — how to structure the answer

Take a stem such as "a 33-year-old with chronic hypertension on methyldopa, now 35 weeks, BP 158/104 in clinic, urate rising, platelets 130." The answer runs:

1. Frame and classify — "Chronic hypertension with possible superimposed pre-eclampsia — the rising urate and a BP escalating beyond her maintenance dose are the clues; I would not call this 'just harder-to-control chronic hypertension' until I've excluded the syndrome."
2. Stratify and investigate — repeat standardised BP, FBC/platelet trend, U&E/creatinine, ALT/AST, urate, urine PCR; ask about symptoms of end-organ involvement; assess the fetus (growth, Dopplers).
3. Treat to target, pick the right drug — control to <140/90 maintenance; if it tips into severe range, acute nifedipine/labetalol/hydralazine to ≤150/100; add a second class rather than max one agent; reconsider methyldopa given the approaching postpartum.
4. Decide timing — for controlled chronic/gestational hypertension, offer planned birth around 38 weeks (WILL, with its early-stopping caveat); superimposed pre-eclampsia is delivered on its own merits after stabilisation.
5. Justify from evidence — CHAP for the <140/90 target and its NNT ≈ 15; CHIPS for fetal safety of tight control; Easterling for the oral pathway where IV access is unreliable; MAGPIE if it becomes pre-eclampsia.
6. Plan the postpartum and beyond — switch methyldopa → enalapril/amlodipine, daily review through the days 3–6 peak, a named re-presentation threshold, a 6-week check, and a lifelong cardiovascular-risk and next-pregnancy-aspirin conversation.

Evidence anchors

• CHAP — Treatment for Mild Chronic Hypertension during Pregnancy. N Engl J Med 2022;386:1781–1792
• CHIPS — Less-Tight versus Tight Control of Hypertension in Pregnancy. N Engl J Med 2015;372:407–417
• Easterling oral antihypertensive RCT. Lancet 2019;394:1011–1021
• HYPITAT — induction vs expectant monitoring for mild hypertensive disease ≥36 wk. Lancet 2009;374:979–988
• WILL trial. PLOS Med 2024;21:e1004481
• FoR BP — furosemide postpartum RCT. Hypertension 2021;77:1517–1524
• POP-HT — self-management RCT. JAMA 2023;330:1991–1999
• POP-HT cardiac-remodelling imaging substudy. Circulation 2024
• MAGPIE — magnesium sulphate for pre-eclampsia. Lancet 2002;359:1877–1890
• NICE NG133 — Hypertension in pregnancy: diagnosis and management
• SA NDoH National Integrated Maternal and Perinatal Care Guideline (October 2024) — severe HTN ≥160/110, acute oral nifedipine / IV labetalol, methyldopa for maintenance.