To determine whether the transhepatic or transperitoneal approach is the optimal percutaneous cholecystostomy approach. A systematic review and meta-analysis was undertaken in which the Medline, EMBASE, and PubMed databases were searched for studies that compared both approaches in patients undergoing percutaneous cholecystostomy. Statistical analysis of dichotomous variables was carried out using odds ratio as the summary statistic. Four studies totalling 684 patients (396 [58%] males, mean age 74 years) who had undergone percutaneous cholecystostomy via the transhepatic (n=367) and transperitoneal (n=317) approach were analysed. Although the overall risk of bleeding was low (4.1%), it was significantly higher in the transhepatic approach compared with the transperitoneal approach (6.3% versus 1.6% respectively, odds ratio = 4.02 [1.56, 10.38]; p=0.004). There were no significant differences in pain, bile leak, tube-related complications, wound infection, or abscess formation between the approaches. Percutaneous cholecystostomy can be performed safely and successfully via the transhepatic and transperitoneal approaches. Although the overall rate of bleeding was significantly higher with the transhepatic approach, there were confounding factors due to technical differences between the studies. The small number of the included studies, in addition to variability of the definitions of outcomes, imposed other limitations. Further large-volume cases series and ideally a randomised trial with well-defined outcomes are required to confirm these findings.