© 2017 Royal College of Obstetricians and Gynaecologists Background: Randomised trials and their syntheses in meta-analyses offer a unique opportunity to assess the frequency and severity of adverse reactions. Objective: To assess safety reporting in pre-eclampsia trials. Search strategy: Systematic search using bibliographic databases, including Cochrane Central Register of Controlled Trials, Embase, and MEDLINE, from inception to August 2017. Selection criteria: Randomised trials evaluating anticonvulsant or antihypertensive medication for pre-eclampsia. Data collection and analysis: Descriptive statistics appraising the adequacy of adverse reaction and toxicity reporting. Main results: We included 60 randomised trials. Six trials (10%) were registered with the International Clinical Trials Registry Platform, two registry records referred to adverse reactions, stating ‘safety and toleration’ and ‘possible side effects’ would be collected. Twenty-six trials (43%) stated the frequency of withdrawals within each study arm, and five trials (8%) adequately reported these withdrawals. Adverse reactions were inconsistently reported across eligible trials: 24 (40%) reported no serious adverse reactions and 36 (60%) reported no mild adverse reactions. The methods of definition or measurement of adverse reactions were infrequently reported within published trial reports. Conclusions: Pre-eclampsia trials regularly omit critical information related to safety. Despite the paucity of reporting, randomised trials collect an enormous amount of safety data. Developing and implementing a minimum data set could help to improve safety reporting, permitting a more balanced assessment of interventions by considering the trade-off between the benefits and harms. Tweetable abstract: Developing @coreoutcomes could help to improve safety reporting in #preeclampsia trials. @NIHR_DC.
BJOG: An International Journal of Obstetrics and Gynaecology
795 - 803