Study selection and data extraction
Two investigators (A.H. and N.T.) independently identified and extracted
articles for potential inclusion, using the Rayyan QCRI web application
for systematic reviews. Disagreements were resolved by referral to a
third reviewer (B.H.R.). The full texts of the resulting references were
retrieved and analysed. If more than one study published data from the
same cohort, we included the report with the higher quality according to
the Risk of bias In Non-randomized Studies of Interventions (ROBINS-I)
tool (Newcastle-Ottawa Quality Assessment Scale- NOS) to avoid overlap.
Exposure to statins during pregnancy was defined as exposure to any dose
and in any trimester of pregnancy. Pregnancies that were defined as high
risk for future development of uteroplacental insufficiency disorder and
those that had developed uteroplacental insufficiency disorders were
included. The spectrum of uteroplacental insufficiency disorders
included: preeclampsia, early-onset IUGR (<28w), and
obstetrical antiphospholipid syndrome. Multiple pregnancies were
excluded in patient-level data to avoid bias. The primary outcome
included prolongation of pregnancy from study entry to delivery. The
secondary outcomes included neonatal outcomes such as admission to the
neonatal intensive care unit (NICU), birth weight and neonatal death,
and the maternal occurrence of a new diagnosis of preeclampsia. Neonatal
death was defined as stillbirth or death in the first month of life.
Data were extracted from the included studies by a single reviewer and
subsequently evaluated by the second reviewer. For studies that did not
report the outcomes, we contacted the authors and requested the missing
data.