Coronary artery bypass surgery can offer excellent results when performed with cardiopulmonary bypass (on pump) or without cardiopulmonary bypass (off pump). The debate as to which technique is superior remains unanswered. Intra-operative conversion from off- to on-pump coronary surgery is a relatively unexplored phenomenon, which cannot be assessed within randomised controlled trial design. We aimed to assess the effect of off-pump conversion on patient mortality. Medline, Embase, Cochrane and Google Scholar databases were systematically reviewed for studies published between 1980 and 2010 that compared the incidence of mortality between converted and non-converted off-pump patients. Publication bias and heterogeneity were assessed and data were extracted independently by multiple observers. We undertook a meta-analysis of these studies using random effects modelling. A total of 17 studies fulfilled our inclusion criteria, containing data for 18,870 off-pump coronary artery bypass operations spanning a decade (1998-2008), involving 920 cases of conversion. Overall, conversion increased mortality by an odds ratio of 6.18 (95% confidence interval 4.65-8.20), whereas emergency conversion further raised the odds ratio of mortality to 6.99 (95% confidence interval 5.18-9.45). The conversion from off- to on-pump cardiac surgery may significantly increase the chance of an adverse outcome, whereas emergency conversion confers a significant rise in mortality. The risk of conversion should be discussed when obtaining the patient's informed consent and its prevention warrants serious consideration by cardiac surgeons and cardiac surgical training programmes.