We performed a meta-analysis to identify the association between systemic lupus erythematosus (SLE) and preterm birth. In this study, we studied the effects of SLE, SLE disease activity, a history of nephritis and active nephritis on preterm birth. Searches were conducted before 20 May 2016 of PubMed, Embase, Medline and Cochrane Library of literature and article reference lists. Eleven observational case–control studies and thirteen cohort studies met the inclusion criteria. The pooled relative risk (RR) for the risk of preterm birth in SLE patients versus controls was 2.05 (95% confidence interval (CI): 1.72–3.32); for active SLE patients versus inactive was 2.98 (95% CI: 2.32–3.83); for SLE patients with a history of lupus nephritis versus those without nephritis it was 1.62 (95% CI: 1.35–1.95); and for SLE patients with active nephritis versus those with quiescent nephritis it was 1.78 (95% CI: 1.17–2.70). In summary, this study identified a significant association in the above results. This association was more significant in active SLE patients versus inactive. With respect to SLE itself, active inflammation (such as disease activity) may be more hazardous for the management of the pregnancy. This suggests that it is essential to control disease activity in order to achieve a better outcome of SLE pregnancy.