The most common surgical treatment for lumbar degenerative disc disease (LDDD) is lumbar decompression and fusion. However, the fused vertebrae may increase the motion of adjacent level and may induce the adjacent disease. Lumbar total disc replacement (TDR) provides an alternative treatment for LDDD to preserve the motion of surgical level. The aim of our study is to conduct an updated systematic review and meta-analysis to compare the efficacy and safety of TDR versus lumbar fusion. We comprehensively searched the meta-analysis comparing TDR with fusion through Pubmed, Embase and Cochrane database. Only the randomized controlled trials (RCT) were selected and collected. Retrieval time was searched until June 2017. Two authors independently extracted the data from the studies after being accessed the quality The statistical soft STATA12.0 was used to analysis the data. A total of 7 RCTs (1706 patients) were met the inclusion requirements and included in our analysis. Patients in TDR group had significant improvements in ODI,VAS scores, complication rates and had a greater percentage of being satisfied with the surgery. In addition, the clinical success in TDR group was higher than fusion group. Meanwhile, TDR treated patients had shorter operating time and shorter duration of hospital stay. However, there was no clinical significance between two groups at blood loss, work status and reoperation rate. Our meta-analysis showed that TDR proved superiorities in improved clinical success, reduced pain, patients’ satisfaction, shortened hospital stay and operating time and lessened complication rate. But there were no benefits in blood loss,