Clinical overview
Fluid management is one of the most consequential and most poorly executed decisions in operative obstetrics and gynaecology. The patient on your list is not a passive recipient of a standing maintenance order — she is a physiological system whose intravascular volume, electrolyte balance, oncotic pressure and tissue perfusion you are actively steering through a period of surgical stress, fasting, fluid shifts and, often, blood loss. Get it right and she wakes warm, perfused, lucid and passing urine. Get it wrong in either direction and the consequences are real: under-resuscitation drives hypovolaemic shock, acute kidney injury and lactic acidosis; over-resuscitation drives bowel oedema, ileus, anastomotic and wound-healing failure, pulmonary oedema and — in the pregnant or pre-eclamptic woman — sometimes fatal cardiogenic and permeability pulmonary oedema.
The verb in this objective is consider — you are being asked to reason from principles, not to recite a single recipe. There is no universal "10 mL/kg/h" that is safe for every patient. The same 2 litres of crystalloid that rescue a septic, ruptured ectopic will kill a severe pre-eclamptic with capillary leak. A registrar who understands the principles — what compartment is depleted, what fluid distributes where, what the heart and kidneys are telling you, and which patient phenotype is in front of you — will make safe individual decisions in theatre, on the post-operative ward and in the high-care unit. This chapter builds that reasoning. It pairs naturally with eras-principles, fluids-electrolytes-og, shock-management and arterial-blood-gas.
Core knowledge
Figure F5.1 — Where the litre goes: total body water across the ICF/ECF/plasma/interstitial compartments, and how balanced crystalloid, 5% dextrose and colloid distribute — distribution predicts expansion, acidosis and oedema.
Body water compartments and where fluids go
Total body water is classically ~60% of body weight in an adult (lower in the obese and elderly, higher in neonates) — standard physiology teaching. Of that, roughly two-thirds is intracellular and one-third extracellular, and of the extracellular fluid about a quarter is intravascular plasma and three-quarters interstitial. This 1:3 plasma:interstitial split is the single most important number for fluid prescribing, because it predicts where an infused fluid ends up.
- 0.9% saline and balanced crystalloids (Ringer's lactate, Plasma-Lyte) distribute across the whole extracellular space. Only about a quarter to a fifth stays intravascular after equilibration — so to expand plasma volume by 1 unit you must give roughly 3–4 units of crystalloid (standard teaching). The rest becomes interstitial oedema.
- 5% dextrose is effectively free water once the glucose is metabolised; it distributes across total body water, so only ~1/12 stays intravascular. It is a maintenance/free-water fluid, never a resuscitation fluid.
- Colloids (albumin, gelatins, starches) were designed to stay intravascular by oncotic pull. In health that logic holds; in the capillary-leak states common in our patients (sepsis, pre-eclampsia, major surgery) the barrier is leaky and colloid advantage shrinks.
Composition matters: the chloride problem
A registrar must know the difference between 0.9% "normal" saline (Na⁺ 154, Cl⁻ 154 mmol/L — both supraphysiological) and balanced solutions (Ringer's lactate / Hartmann's: Na⁺ ~131, Cl⁻ ~111, with lactate as a bicarbonate precursor; Plasma-Lyte similar with acetate/gluconate). Large volumes of 0.9% saline produce a hyperchloraemic metabolic acidosis and are associated with more renal vasoconstriction. The pragmatic landmark trials in this space — SMART and SALT-ED (balanced crystalloids vs saline in critically ill and non-critically ill adults; standard critical-care evidence) — showed a small favourable signal for balanced solutions on a composite of death, new renal replacement and persistent renal dysfunction. The practical principle: default to a balanced crystalloid for resuscitation and replacement; reserve 0.9% saline for specific indications (hypochloraemic alkalosis from vomiting, or hyponatraemia where you want the higher sodium). The major exception to "balanced by default" is the head-injured or hyponatraemic patient where the slightly hypotonic Ringer's is undesirable.
Maintenance vs resuscitation vs replacement — three different jobs
These are conceptually distinct and must never be conflated:
| Job | Question it answers | Typical fluid | Rough volume |
|---|---|---|---|
| Maintenance | What does a fasting patient need to stay in balance? | balanced crystalloid ± dextrose/K⁺ | ~25–30 mL/kg/day water, ~1 mmol/kg/day Na⁺/K⁺/Cl⁻, ~50–100 g/day glucose (standard teaching, per NICE IV-fluids principles) |
| Replacement | What ongoing abnormal losses must I match? | match the composition of the loss (e.g. balanced crystalloid for GI/third-space) | volume-for-volume |
| Resuscitation | Is there a perfusion deficit now? | balanced crystalloid bolus, then blood if bleeding | reassess after each ~250–500 mL bolus |
The classic ward error is running a "maintenance" rate to treat a resuscitation problem (too little, too slow) or, conversely, leaving a generous resuscitation-era rate running for days after the deficit is corrected (too much) — the latter is how post-operative gynaecology patients end up 5 litres positive, oedematous and ileus-bound.
The shift from "liberal" to "zero-balance" thinking
Two decades of evidence have moved us away from large pre-emptive fluid loading ("liberal"/supranormal) towards maintaining the patient at or near their pre-operative weight — euvolaemia, a near-zero fluid balance. This is the ERAS (Enhanced Recovery After Surgery) principle and it dovetails with this objective; see eras-principles. ERAS Society gynaecologic-surgery guidance emphasises avoiding both prolonged fasting and bowel-prep-induced dehydration, allowing clear fluids up to 2 hours pre-operatively and a carbohydrate drink, then giving goal-directed, restrictive maintenance intra-operatively and stopping IV fluids early once oral intake resumes. The large RELIEF trial (restrictive vs liberal fluids in major abdominal surgery; standard peri-operative evidence) is the key anchor: a truly restrictive zero-balance regimen produced more acute kidney injury than a modestly liberal one — the practical lesson is that the target is euvolaemia, not dryness. "Restrictive" must never mean "dehydrating."
Assessment
Phenotyping the patient before you prescribe
Reasoning from principles starts with which patient. The fluid plan for each of these is different:
- The hypovolaemic emergency — ruptured ectopic, massive PPH, ruptured ovarian cyst with haemoperitoneum, septic abortion. Intravascular volume is acutely down; the job is rapid restoration of perfusion and control of the source. (See ectopic-pregnancy-management, postpartum-haemorrhage, shock-management.)
- The well, elective gynaecology patient — euvolaemic, fasted, for laparoscopic or open benign or oncological surgery. The job is to keep her euvolaemic: minimal fasting, modest maintenance, goal-directed top-ups only for measured losses.
- The pre-eclamptic / severe-PET patient — this is the dangerous one. Capillary leak, low oncotic pressure, and a stiff vasoconstricted circulation mean she is simultaneously intravascularly underfilled and at high risk of pulmonary oedema. Fluids must be restricted, not liberal. (See pre-eclampsia-and-hellp, hypertension-in-pregnancy.)
- The septic patient — non-pregnancy-related sepsis and puerperal/post- surgical sepsis need early resuscitation but also have leaky capillaries and earn ICU-style de-escalation once perfusion is restored.
Clinical assessment of volume status
No single number suffices; integrate several:
- History — losses (vomiting, diarrhoea, blood, drains, bowel prep, insensible losses with fever), oral intake, comorbidity (cardiac, renal, pre-eclampsia).
- Examination — heart rate, blood pressure (and postural change), capillary refill, peripheral temperature, mucous membranes, skin turgor, JVP, and — the most useful bedside marker — examine the lung bases and look for sacral/peripheral oedema before adding more.
- Urine output — the cheap continuous monitor of renal perfusion. The conventional target is ≥0.5 mL/kg/h (standard teaching). Oliguria is a prompt to assess, not a reflex to bolus — in pre-eclampsia transient post-operative oliguria is expected and should not trigger aggressive fluid loading.
- Trends, not snapshots — a falling, narrowing or postural blood pressure with a rising heart rate and lactate is hypovolaemia until proven otherwise.
Investigations and dynamic monitoring
- Bloods — U&E/creatinine (baseline renal function, sodium, the chloride and bicarbonate that flag hyperchloraemic acidosis), full blood count (haemoglobin trend, though acute loss may not yet show), and a venous or arterial lactate as a global perfusion marker; serial lactate clearance is a better resuscitation endpoint than any single pressure. ABG interpretation is covered in arterial-blood-gas.
- Dynamic indices beat static ones. In the ventilated theatre patient, fluid responsiveness is better predicted by pulse-pressure / stroke- volume variation, or the response to a passive-leg-raise or a small fluid challenge than by CVP. A single central venous pressure reading is a poor guide to volume and should not be chased to a number. Where available, oesophageal-Doppler or pulse-contour goal-directed therapy titrates small boluses to stroke-volume optimisation — a core ERAS tool. In the South African district/regional setting most of this is unavailable; you titrate to heart rate, blood pressure, urine output, capillary refill, lactate and repeated clinical examination of the lung bases — and that is legitimate, principled practice.
Management
General principles in theatre
- Correct the deficit before induction where you can. Anaesthesia vasodilates; a patient resuscitated before induction tolerates it far better than one you chase afterwards.
- Replace measured blood loss appropriately — small losses with balanced crystalloid (~3:1), but switch to blood products rather than ever-larger crystalloid volumes once loss is significant; crystalloid alone dilutes clotting factors and platelets and worsens coagulopathy.
- Titrate maintenance to euvolaemia (ERAS / RELIEF principle), avoiding both the dry zero-balance extreme and reflex liberal loading.
- Keep her warm — hypothermia impairs coagulation and prolongs recovery; warm the fluids in major or rapid resuscitation.
- Stop the IV early post-operatively and return to oral intake as soon as safe — a cornerstone of eras-principles and the simplest way to avoid the "5 litres positive" disaster.
The obstetric/gynaecological haemorrhage drill (UNMISTAKABLE)
When the patient is actively bleeding and shocked — ruptured ectopic, major PPH, intra-operative vascular injury — fluid management is resuscitation and runs in parallel with surgical/uterotonic source control. Do not deliberate:
- Call for help. Declare a major haemorrhage. Activate the massive transfusion protocol; alert anaesthesia, theatre, blood bank, senior obstetrician.
- Two large-bore IV cannulae (14–16G). Bloods off the needle: FBC, U&E, coagulation/fibrinogen, cross-match (≥4 units), lactate.
- Start warmed balanced crystalloid while blood is coming — but switch to blood early; do not pour in litres of crystalloid. Aim to restore perfusion, not a normal blood pressure in uncontrolled bleeding (permissive hypotension principle until the source is secured).
- Tranexamic acid 1 g IV — for PPH, give early, within 3 hours of onset; the WOMAN trial (Lancet 2017) showed early TXA cuts death from bleeding by about a third. (See postpartum-haemorrhage.)
- Activate the massive transfusion / fixed-ratio pack (red cells : FFP : platelets in a balanced ratio per your blood-bank protocol) once loss is ongoing; transfuse to guided targets — RCOG GTG 47 (Blood Transfusions in Obstetrics) is the reference. Use point-of-care viscoelastic testing where available; replace fibrinogen (cryoprecipitate) as it falls.
- Treat the cause — uterotonics and mechanical/surgical measures for PPH (the E-MOTIVE bundle — early detection plus Massage, Oxytocics, TXA, IV fluids, Examine/escalate; NEJM 2023), clamp/ligate the vessel for surgical injury, control the ectopic.
- Monitor and reassess after every intervention — perfusion, lactate clearance, urine output, temperature, ongoing loss, repeat coagulation.
Crystalloid buys minutes; blood and source control save the life. The commonest fatal errors are delay in escalating and over-reliance on crystalloid producing dilutional coagulopathy.
Pre-eclampsia: restrict, do not flood (UNMISTAKABLE)
This deserves its own warning because the instinct to "give fluid for oliguria" kills here. In severe pre-eclampsia the NICE NG133 (Hypertension in pregnancy, 2019) principle, echoed by the SA National Integrated Maternal and Perinatal Care Guideline (NDoH, 2024), is to restrict total fluids (a commonly cited ceiling is around 80 mL/h including drug infusions — verify the exact figure against the current guideline before quoting it). Pulmonary oedema is a leading cause of pre-eclampsia death; transient oliguria is tolerated and managed by observation, not boluses. Magnesium sulphate is the anticonvulsant of choice (MAGPIE); fluid loading "to cover" regional anaesthesia is not warranted. (See pre-eclampsia-and-hellp.)
South African context
- The SA NDoH National Integrated Maternal and Perinatal Care Guideline, 5th ed (2024) is the obstetric source of truth — for PPH (oxytocin first-line uterotonic, misoprostol, carbetocin), pre-eclampsia fluid restriction, and referral.
- Saving Mothers / NCCEMD triennial reports show obstetric haemorrhage and hypertensive disease among the leading direct causes of maternal death in South Africa — and both are bound up with fluid decisions (under- resuscitation in haemorrhage, over-resuscitation/pulmonary oedema in pre-eclampsia). Non-pregnancy-related infection, predominantly HIV, remains a leading cause; HIV-associated cardiomyopathy, anaemia and sepsis change a patient's fluid tolerance, and the immunosuppressed septic patient needs early but carefully de-escalated resuscitation.
- Resources scale with level of care. District hospitals manage with clinical endpoints, balanced crystalloid (where stocked; often 0.9% saline and Ringer's lactate on the EML) and timely referral; regional/tertiary units add invasive monitoring, viscoelastic testing, blood-bank depth and high-care/ICU. Know what your level can and cannot do, and escalate/transfer the patient who outstrips it early, while she is still resuscitatable.
Red flags / pitfalls
- Treating a number, not the patient. Chasing a urine output or a CVP to a target value, or "topping up to a systolic of 120" in uncontrolled bleeding, causes harm. Treat perfusion and trends.
- Over-resuscitation. Bowel oedema, ileus, anastomotic and wound dehiscence, pulmonary oedema, prolonged stay. A patient several litres positive is a failure of stewardship — weigh, examine the bases, and stop the drip.
- Crystalloid for major haemorrhage. Litres of crystalloid dilute clotting factors and worsen coagulopathy. Switch to blood and a balanced transfusion ratio early; give TXA within 3 hours.
- Flooding the pre-eclamptic. The single most dangerous fluid error in obstetrics. Restrict; tolerate transient oliguria; watch the lungs.
- Large-volume 0.9% saline → hyperchloraemic metabolic acidosis and renal vasoconstriction. Default to a balanced crystalloid.
- 5% dextrose as a resuscitation fluid. It distributes into total body water and barely expands plasma — and risks hyponatraemia. It is maintenance/free water only.
- Forgetting potassium and glucose in prolonged fasting/maintenance, and iatrogenic hyponatraemia from hypotonic fluids — a recognised cause of post-operative morbidity, especially in women.
- Leaving the resuscitation rate running once the deficit is corrected. Re-prescribe fluids on every ward round; an IV order is a decision, not a default.
- Not warming fluids in rapid/major resuscitation → hypothermia and coagulopathy.
- Delaying escalation/referral beyond the point where the patient is salvageable — the recurring lesson of Saving Mothers.
Evidence anchors
- ERAS Society — Guidelines for perioperative care in gynaecologic/oncology surgery. Euvolaemia, minimal fasting, carbohydrate loading, goal-directed restrictive intra-operative fluids, early oral intake and early IV cessation. See eras-principles.
- RELIEF trial (restrictive vs liberal IV fluid in major abdominal surgery) — a truly restrictive regimen increased acute kidney injury; the target is euvolaemia, not dryness (standard peri-operative evidence).
- SMART and SALT-ED trials — small favourable signal for balanced crystalloids over 0.9% saline on renal/composite outcomes; default to balanced solutions (standard critical-care evidence).
- NICE NG133 — Hypertension in pregnancy (2019) — fluid restriction in severe pre-eclampsia; magnesium sulphate (MAGPIE) for seizure prophylaxis/treatment.
- SA NDoH National Integrated Maternal and Perinatal Care Guideline, 5th ed (2024) — SA obstetric source of truth for PPH and pre-eclampsia fluid management and referral pathways.
- RCOG Green-top Guideline 52 — Prevention and Management of Postpartum Haemorrhage and GTG 47 — Blood Transfusions in Obstetrics — resuscitation, balanced transfusion ratios and product targets in obstetric haemorrhage.
- WOMAN trial (Lancet 2017) — tranexamic acid 1 g IV within 3 hours of PPH onset reduces death from bleeding by ~one third. E-MOTIVE bundle (NEJM 2023) — early detection plus Massage/Oxytocics/TXA/IV fluids/Examine reduces severe PPH.
- Saving Mothers (NCCEMD) — obstetric haemorrhage and hypertensive disease as leading SA maternal-death causes; both fluid-management-sensitive.
- WHO Surgical Safety Checklist — anticipated blood loss and IV access are explicit pre-incision checklist items.