Eclamptic Seizure at 34 Weeks in a Hospital Without a Neonatal Unit
A 26-year-old woman, gravida 2 para 1, at 34 weeks by her last menstrual period, is brought by her husband after a 2-minute generalised tonic-clonic seizure in the ambulance. She has had no antenatal care this pregnancy. On arrival she is postictal, responding to painful stimuli, with blood pressure 176/112 mmHg, heart rate 108/min, SpO2 94% on 15 L via a non-rebreather, and bilateral pitting oedema to the knees. Urine dipstick shows 4+ protein. Fundal height is 32 cm, the fetal heart rate is 148 beats/min, and there is no vaginal bleeding or fluid loss. Random glucose 6.2 mmol/L. The facility is a district hospital with a functioning labour ward, an adult ICU with two ventilated beds and no neonatal intensive care unit. The tertiary obstetric referral centre is 2 hours away by road; the ambulance has a midwife and a ventilator. Her husband asks whether she can be transferred.
1. Write your management of the first 30 minutes, with specific drugs and doses, and state what you would avoid.
Read the worked answer
This is eclampsia: control the seizure, control the pressure, protect the airway.
- Left lateral position, suction, high-flow oxygen, continuous SpO2, ECG and blood pressure, two large-bore cannulae, and a urinary catheter with hourly output. Do not force anything into the mouth.
- Magnesium sulfate: loading dose 4 g IV over 20 minutes (some protocols use 6 g over 20-30 minutes), then 1 g/hour infusion. If she seizes again, give a further 2 g IV bolus over 5 minutes and consider increasing the infusion. Magnesium is the anticonvulsant of choice; it is superior to phenytoin and diazepam for both control and fetal outcome.
- Blood pressure: labetalol 20 mg IV, repeated or doubled every 10-15 minutes (40 mg, then 80 mg) to a maximum of 300 mg, or hydralazine 5-10 mg IV every 20 minutes, or oral nifedipine 10 mg. Target 130-150/80-100 mmHg. Bring the pressure down over minutes to hours, not instantly - a sharp fall causes placental hypoperfusion and fetal distress.
- Fetal monitoring once maternal stabilisation allows; continuous CTG if available.
- Bloods: full blood count, platelets, liver enzymes, creatinine, urate, coagulation, group and save, and catheter urine protein.
Avoid: diazepam or phenytoin as first-line (worse maternal and neonatal outcomes), ergometrine later, NSAIDs, and rapid bolus antihypertensives such as sublingual nifedipine or IV hydralazine boluses pushed quickly. Do not give a fluid load 'to improve placental perfusion' - she is at risk of pulmonary oedema. Do not omit the magnesium infusion after the loading dose; seizures recur when it is stopped too early.
2. Transfer or deliver here? Give your decision, the criteria you use, and what you would do if she deteriorates before departure.
Read the worked answer
My default is to stabilise and transfer her in utero, but only after she has been seizure-free and pressure-controlled for a period, and only if the fetus is not compromised.
Why transfer: at 34 weeks the baby will likely need neonatal intensive care, which this hospital does not have, and she may need obstetric high-dependency care and neurosurgical/ICU support she cannot get here. Delivery is the definitive treatment for eclampsia, but delivery in a unit without neonatal intensive care exposes a 34-week infant to avoidable risk. Antenatal transfer is safer than neonatal transfer.
Criteria to transfer: no seizure for at least 2 hours on magnesium, systolic BP below 160 and diastolic below 110 mmHg on treatment, improving conscious level, stable maternal observations, no evidence of abruption or non-reassuring fetal heart trace, and a receiving unit that has accepted her. Give betamethasone 12 mg IM twice, 24 hours apart, before departure if delivery is anticipated, and continue magnesium throughout the journey with reflexes, respiratory rate and urine output monitored.
Deliver locally if: seizures recur despite adequate magnesium, blood pressure cannot be controlled, there is abruptio placentae, pulmonary oedema, HELLP syndrome with deteriorating liver or coagulation, non-reassuring CTG, or she needs intubation and ventilation. Delivery is then the priority and the neonate is managed with the available team; arrange urgent neonatal retrieval after birth.
Document the discussion with the husband: transfer is to give her and the baby the care this hospital cannot provide, not because she is too sick to stay.
3. Four hours into the magnesium infusion her respiratory rate is 11/min, deep tendon reflexes are absent, and urine output has fallen to 15 mL/hour. What is happening, how do you treat it, and how would you have prevented it?
Read the worked answer
This is magnesium toxicity. Absent deep tendon reflexes are the earliest sign (around 10 mEq/L), followed by respiratory depression (12-15 mEq/L), then cardiac arrest at higher levels. Oliguria reduces magnesium clearance, so toxicity develops even at a standard 1 g/hour infusion.
Immediate actions:
- Stop the magnesium infusion.
- Give calcium gluconate 1 g IV (10 mL of 10%) over 10 minutes as the antidote, with cardiac monitoring; calcium chloride is more irritating and more dangerous peripherally.
- Oxygen and airway support; be prepared to ventilate if the respiratory rate falls further or saturation drops.
- Recheck renal function, magnesium level if available, and urine output; catheterise if not already done.
- Restart magnesium at a reduced rate (0.5 g/hour) only once reflexes have returned and the respiratory rate is above 12/min, and consider whether she still needs it - if delivery has occurred and she is stable, it may be stopped.
Prevention: the infusion is only safe with a monitoring chart - hourly respiratory rate, deep tendon reflexes and urine output (the 'toxicity triad'), with a target urine output above 30 mL/hour and continued vigilance for at least 24 hours after delivery. Magnesium should be prescribed with the same seriousness as an insulin infusion: a written rate, a documented monitoring plan, and a named person responsible for checking. Calcium gluconate must be immediately available wherever magnesium is running; in this case the falling urine output should have prompted a dose review before the reflexes disappeared.