BACKGROUND Stones in the urinary tract are a common medical problem in the general population. At present, the great expansion in minimally invasive techniques has led to the decrease in open surgery. Extracorporeal shock wave lithotripsy (ESWL) has been introduced as an alternative approach which disintegrates stones in the kidney and upper urinary tract through the use of shock waves. Nevertheless, as there are limitations with the success rate in ESWL, other minimally invasive modalities for kidney stones such as percutaneous nephrolithotomy (PCNL) and retrograde intrarenal surgery (RIRS) are also widely applied. OBJECTIVE To evaluate the effectiveness and complications of ESWL compared with PCNL or RIRS for managing kidney stones. METHODS We searched the Cochrane Central Register of Controlled Trials (CENTRAL in The Cochrane Library), MEDLINE, EMBASE and reference lists of articles without language restriction. METHODS Randomised controlled trials (RCTs) assessing the use of ESWL compared to PCNL or RIRS for kidney stone management. METHODS Two authors independently assessed all the studies for inclusion. Statistical analyses were performed using the random effects model and the results expressed as risk ratio (RR) for dichotomous outcomes or mean difference (MD) for continuous data with 95% confidence intervals (CI). RESULTS Three studies (214 patients) were included, however results could not be pooled. Two RCTs compared ESWL to PCNL. The success rate at three months for lower pole kidney stones was statistically higher for PCNL (RR 0.39, 95% CI 0.27 to 0.56). Re-treatment (RR 1.81, 95% CI 0.66 to 4.99) and using auxiliary procedures (RR 9.06, 95% CI 1.20 to 68.64) after PCNL were less compared to ESWL. The efficiency quotient (EQ) in PCNL was higher than ESWL. Hospital stay (MD -3.30 days, 95% CI -5.45 to -1.15), duration of treatment (MD -36.00 minutes, 95% CI -54.10 to -17.90) and complications were less for ESWL. One RCT compared ESWL versus RIRS for lower pole kidney stones. The success rate was not significantly different at the end of the third month (RR 0.91, 95% CI 0.64 to 1.30). CONCLUSIONS Results from three small studies, with low methodological quality, indicated ESWL is less effective for lower pole kidney stones than PCNL but not significantly different from RIRS. Hospital stay and duration of treatment was less with ESWL. More RCTs are required to investigate the effectiveness and complications of ESWL for kidney stones compared to PCNL or RIRS.