Lumbar spinal surgery is most commonly performed under general anaesthesia. However, spinal anaesthesia has also been used. We aimed to systematically review the comparative evidence. We only included randomised, controlled trials in this meta-analysis and calculated the risk ratio or standardised mean difference for haemodynamics, blood loss, surgical time, analgesic requirement, nausea and/or vomiting, and length of hospital stay. Eight studies with a total of 625 patients were included. These were considered to be at high risk of bias. Compared with general anaesthesia, the risk ratio (95% CI) with spinal anaesthesia for intra-operative hypertension was 0.31 (0.15-0.64), I = 0% (p = 0.002); for intra-operative tachycardia 0.51 (0.30-0.84), I = 0% (p = 0.009); for analgesic requirement in the postanaesthesia care unit 0.32 (0.24-0.43), I = 0% (p < 0.0001); and for nausea/vomiting within 24 h postoperatively 0.29 (0.18-0.46), I = 12% (p < 0.00001). The standardised mean difference (95% CI) for hospital stay was -1.15 (-1.98 to -0.31), I = 89% (p = 0.007). There was no evidence of a difference in intra-operative hypotension and bradycardia, blood loss, surgical time, analgesic requirement within 24 h postoperatively, and nausea/vomiting in the postanaesthesia care unit. We conclude that spinal anaesthesia appears to offer advantages over general anaesthesia for lumbar spine surgery.