Seizure prophylaxis in hypertensive pregnancies: a framework for making clinical decisions

To describe a framework for generating therapeutic recommendations using seizure prophylaxis in hypertensive pregnancies as an example.

A decision-making framework was built using: 1. evidence of therapeutic benefit, with number needed to treat as the effect measure; 2. baseline rates of the target disorder that the treatment was designed to prevent; and 3. a treatment threshold, determined by weighting the potential risks against the potential benefits of the treatment.

Evidence of therapeutic benefit (i.e. reduction in eclamptic seizures associated with magnesium sulphate therapy in hypertensive pregnancies) was determined by a systematic quantitative overview of controlled clinical trials. Baseline rates of seizures without magnesium sulphate therapy were derived from a recent cohort study. A treatment threshold was generated using estimates of treatment associated morbidities which were weighted against the potential reduction in seizures from magnesium sulphate therapy considering the relative values assigned to these outcomes by obstetricians practising in our hospital.

The number of hypertensive women needed to be treated with magnesium sulphate to prevent a single case of eclamptic seizures varied in a curvilinear fashion dropping from 1000 to 14 as the baseline rate of seizures increased from 0.1% to 10%. The treatment threshold as measured by number needed to treat was 64 (range 57-77). The number needed to treat for nonproteinuric hypertension was 1000 (95% CI 180-40,000), whereas it was 32 (95% CI 20-57) for proteinuric hypertension. Considering the uncertainty in estimation of the numbers needed to treat and treatment threshold, magnesium sulphate therapy may be recommended for women at high risk of eclampsia (e.g. severe pre-eclampsia) while it should be withheld in cases at low risk (e.g. nonproteinuric hypertension and mild pre-eclampsia).

While awaiting further research obstetricians intuitively make decisions about seizure prophylaxis in hypertensive pregnancies. Our decision-making framework generated therapeutic recommendations by explicit consideration of the available evidence.