Study selection process
We selected cohort studies and case control studies with patients who had liver cirrhosis and were pregnant. Diagnoses of liver cirrhosis were considered confirmed when based on biopsy findings or based on imaging combined with laboratory findings. Cohorts or case control studies had to include a minimum series of five patients with liver cirrhosis. Results had to contain at least one item of one of the designated outcome domains (see heading ‘outcome measurements’). We included studies in the meta-analysis if they contained a control group. When our search found papers which analyzed identical patient populations multiple times, and which reported on identical outcomes, we selected the most recent or most relevant study, and excluded the other studies reporting on the same population. Other exclusion criteria were non-human studies, studies outside pregnancy and case reports, case series, opinions or reviews.
Two reviewers (LLS and RBT) independently screened the titles and abstracts of studies retrieved by the database searches. We requested the full text if a study was potentially relevant and performed independent screening to assess eligibility in duplo on full text. Disagreement about including a study were resolved through discussion or by consulting a third reviewer (RCP).