Twenty-five reports of 21 studies were included in the review. These included one controlled trial (270 participants), 13 prospective cohort studies (2,818 participants), six retrospective cohort studies (1,260 participants) and one case control study (6,307 infant participants with or without congenital malformations).
Six prospective cohort studies and one retrospective cohort study were assessed as having a low risk of bias. Three prospective cohort studies, three retrospective cohort studies and one case-control study were assessed as having a medium risk of bias. Four prospective cohort studies, two retrospective cohort studies and one controlled trial were assessed as having a high risk of bias. The most common methodological flaw among the cohort studies (the only studies that were included in the meta-analyses) was a lack of adjustment for confounding factors in analyses.
Compared with controls, pre-pregnancy care participants had statistically significant lower risks of congenital malformations (RR 0.25, 95% CI 0.16 to 0.37; Ι²=30%; 13 cohort studies) and perinatal mortality (RR 0.34, 95% CI 0.15 to 0.75; Ι²=0%; six cohort studies). Numbers needed to treat were 19 for congenital malformations (95% CI 14 to 24) and 46 for perinatal mortality (95% CI 28 to 115). Risk of maternal hypoglycaemia was statistically significantly higher in women who received pre-pregnancy care (RR 1.51, 95% CI 1.15 to 1.99; two cohort studies; substantial heterogeneity Ι²=85%) compared with controls.
Pre-pregnancy care was associated with statistically significant reductions in HbA1c levels in the first trimester of pregnancy (MD -1.92, 95% CI -2.05 to -1,79; five cohort studies) compared with controls. This pooled result indicated substantial heterogeneity between the trials (Ι²=98%); the authors stated the between-study variation was in the size of the effect rather than the direction.