Postpartum domperidone use: what is the added value of observational and preclinical data in an assessment of potential benefits versus harm?




Poster session 3


Tuesday 25 October 2016 - 10:30 to 11:00


All authors in correct order:

Puil L1, Mintzes B2
1 Dept Anesthesiology, Pharmacology, and Therapeutics, University of British Columbia and Cochrane Hypertension Review Group, Canada
2 Faculty of Pharmacy and Charles Perkins Centre, University of Sydney, Cochrane Hypertension Review Group, Australia
Presenting author and contact person

Presenting author:

Barbara Mintzes

Contact person:

Abstract text
Background: Domperidone, a dopamine antagonist, is increasingly being used to stimulate postpartum breast milk supply. In 2011, 20% of new mothers in British Columbia, Canada were prescribed domperidone. Recent randomized controlled trials (RCTs) in Thailand and Pakistan suggest widespread use. This is an off-label use for which uncertainty exists about benefits and harm. Domperidone is subject to safety advisories due to risks of serious cardiac arrhythmia and sudden death, mainly observed in the elderly.

Objectives: To investigate the added value of including non-randomised and preclinical studies in a systematic review (SR) on benefits and harm of postpartum domperidone use.

Methods: We are conducting a SR, following Cochrane methods guidance, with three components: 1) for efficacy and common harms, we included RCTs in new mothers (pre-term or full-term births) comparing domperidone with placebo, other galactagogues, non-drug care, or no treatment; 2) to assess cardiac harms, we included RCTs, controlled cohort, case-control and case-cross-over studies; 3) for mechanisms of domperidone’s proarrhythmic action, we synthesised preclinical studies (tissue culture and animal models). We searched CENTRAL, MEDLINE, Embase, CINAHL, and other databases. Screening, data extraction, and risk of bias assessment were conducted by two independent reviewers. Data are stratified by research question (short-term benefits and harm, cardiac harm, mechanisms), and study design, and meta-analyses conducted using a random-effects model. We contacted study authors for unpublished and sex/age disaggregated data.

Results & Conclusions: We identified eight published postpartum RCTs (five pre-term; three full-term), with limited outcomes reported beyond milk volume, and three terminated/unpublished RCTs. For cardiac harms, two RCTs and eight non-randomised studies included women of reproductive age. Mechanistic and observational data suggest cardiac risks extend to younger age groups. Issues related to different study designs, important outcomes, and variations in reporting will be discussed, along with evidence gaps, and results compared with prior SRs.