Effects of smoking during pregnancy - Five meta-analyses

Background: The purpose of this study was to estimate, using meta-analysis, pooled odds ratios for the effects of smoking on five pregnancy complicati ons: placenta previa, abruptio placenta, ectopic pregnancy, preterm prematu re rupture of the membrane (PPROM), and preeclampsia. Methods: Published articles were identified through computer search and lit erature review. Five criteria were applied to those studies initially ident ified to determine tl-lose eligible for the meta-analysis. A random effects model was applied to derive pooled odds ratios for the eligible studies fo r each pregnancy complication. Meta-analyses were repeated on subsets of th e studies to confirm the overall results. Results: Smoking Tvas found to be strongly associated with an elevated risk of placenta previa, abruptio placenta, ectopic pregnancy and PPROM, and a decreased risk of pre-eclampsia. All pooled odds ratios were statistically significant. The pooled ratios ranged from 1.58 for placenta previa to 1.77 for ectopic pregnancy. The pooled odds ratio for pre-eclampsia was 0.51 an d all subset analyses confirmed this seemingly protective effect. Conclusion: Smoking during pregnancy is a significant and preventable facto r affecting ectopic pregnancy, placental abruption, placenta previa, and PP ROM. The findings of smoking's apparently protective effect on pre-eclampsi a should be balanced with these harmful effects. In addition, the biologica l linkage between smoking and pre-eclampsia is not yet well understood. Pre gnant women should be advised to stop smoking in order to reduce the overal l risk of pregnancy complications as well as any risk of adverse impact on the unborn child. (C) 1999 American Journal of Preventive Medicine.