We performed a cost-effectiveness analysis to examine the relative efficacy and costs of percutaneous ultrasonic lithotripsy (PUL), extracorporeal shock-wave lithotripsy (ESWL), and surgery for the treatment of upper urinary tract stones. We developed a Markov model with 35 states, cycles of 3 months, and a time frame of 5 years. Probability estimates were derived from a meta-analysis of the published literature. For stones less than or equal to 2 cm, ESWL is preferred to PUL, since it prevents 2 additional days of morbidity and saves $440. For larger stones, PUL is preferable to ESWL, avoiding 4 more days of morbidity, and saving $722. Both ESWL and PUL were superior to surgery. Sensitivity analysis showed that the results are sensitive to ESWL efficacy rates, the stone recurrence rate, and the hospital component of the ESWL cost. Our analysis suggests that although ESWL is preferable, relatively small changes in the efficacy and cost can shift the preferred strategy; in addition, these findings underscore the need for more reliable data.