Process of study selection, data extraction and risk of bias assessment
The online tool Rayyan QCRI was used for registration of the selection process.10 Duplicate articles were automatically excluded. Two assessors independently screened all titles and abstracts. Eligible articles were retained for full-text reading. Uncertainties at each stage were discussed and resolved by consensus with a low threshold for inclusion.
One reviewer extracted all data on pregnancy course and complications occurring during delivery and the postpartum period, which was thereafter verified by the second reviewer. Uncertainties at each stage were again discussed and resolved by consensus. All data were described in a narrative format.