Objective To systematically review the evidence of pre-operative exercise, known as 'prehabilitation', on peri-and postoperative outcomes in adult surgical populations. Design Systematic review and meta-analysis. Data sources CENTRAL, Medline, EMBASE, CINAHL, PsycINFO and PEDro were searched from 1950 to 2011. Methods Two reviewers independently examined relevant, English-language articles that examined the effects of pre-operative total-body exercise with peri-and postoperative outcome analysis. Given the nascence of this field, controlled and uncontrolled trials were included. Risk of bias was assessed using the Cochrane Risk of Bias Assessment tool. Only data on length of stay were considered eligible for meta-analysis due to the heterogeneity of measures and methodologies for assessing other outcomes. Results In total, 4597 citations were identified b y t he s earch s trategy, o f w hich 2 1 s tudies w ere i ncluded. Trials w ere g enerally small (median = 54 participants) and of moderate to poor methodological quality. Compared with standard care, the majority of studies found that total-body prehabilitation improved postoperative pain, length of stay and physical function, but it was not consistently effective in improving health-related quality of life or aerobic fitness in the studies that examined these outcomes. The meta-analysis indicated that prehabilitation reduced postoperative length of stay with a small to moderate effect size (Hedges' g = −0.39, P = 0.033). Intervention-related adverse events were reported in two of 669 exercising participants. Conclusion The literature provides early evidence that prehabilitation may reduce length of stay and possibly provide postoperative physical benefits. Cautious interpretation of these findings is warranted given modest methodological quality and significant risk of bias.