Variation in hyperemesis gravidarum definition and outcome reporting in randomised clinical trials: a systematic review

Background: Hyperemesis gravidarum (HG) is a common cause of hospital admission in early pregnancy. There is no international consensus on the definition of HG, or on outcomes that should be reported in trials. Consistency in definition and outcome reporting is important for the interpretation and synthesis of data in meta-analyses. Objective: To identify which HG definitions and outcomes are currently in use in trials. Search strategy: We searched the following sources: (1) Cochrane Central Register of Controlled Trials, (2) Embase and (3) Medline for published trials and the WHO-ICTRP database for ongoing trials (27 October 2017). Selection criteria: All randomised clinical trials reporting on any intervention for HG were eligible. Data collection and analysis: Two reviewers independently assessed trial eligibility and extracted data on HG definition and outcomes. Main results: We included 31 published trials reporting data from 2511 women and three ongoing trials with a planned sample size of 360 participants. We identified 11 definition items. Most commonly used definition items were vomiting (34 trials) and nausea (30 trials). We identified 34 distinct outcomes. Most commonly reported outcomes were vomiting (29 trials), nausea (26 trials), need for hospital treatment (14 trials) and duration of hospital (re)admission(s) (14 trials). Conclusion: There is substantial variation of HG definition and outcome reporting in trials. This hampers meaningful aggregation of trial results in meta-analysis and implementation of evidence in guidelines. To overcome this, international consensus on a definition and a core outcome set for HG trials should be developed. Tweetable abstract: There is a wide variation of definitions and outcomes reported in trials on hyperemesis gravidarum.