Published articles reported controversial results about the association of breast cancer survival with smoking. Hence, a meta-analysis was performed to investigate this association.A comprehensive search was performed to identify relevant cohort studies (up to May 31st, 2016). In the current smoking and former smoking v. never smoking analyses, the fixed- or random-effect model was selected based on the heterogeneity test among studies. And the heterogeneity was measured using Q test and I2 statistic. Publication bias was estimated using Egger's regression asymmetry test.Thirteen articles with 44 studies were included. Compared with never smokers, current smokers have a higher breast cancer-specific mortality and all-cause mortality, with pooled hazard ratio (HR) (HR = 1.30 95%CI: 1.16–1.45; I2 = 52.4%) and (HR = 1.59, 95%CI: 1.41–1.78; I2 = 87.1%), respectively. While former smokers tend to have a moderately increased all-cause mortality (HR = 1.10, 95%CI: 1.07–1.12; I2 = 0.0%), but there was no significant association between former smoking and breast cancer-specific mortality (HR = 0.95, 95%CI: 0.90–1.02; I2 = 0.0%).The present evidence indicates that current smoking leads to higher breast cancer-specific mortality and all-cause mortality than never smoking in breast cancer patients. However former smoking just causes a mild increase in all-cause morality, but not breast cancer-specific mortality.