PURPOSE: To estimate the perturbation of seed position and urethral dose, subsequent to withdrawal of urethral catheters. METHODS AND MATERIALS: A mathematical model based on the volume incompressibility of tissues was used to compute seed positions and doses following removal of the Foley. The model assumed that the central axis of the urethra remains stationary, and that prostate tissue and seeds move radially toward the center of the urethra to fill the void left by the catheter. Seed motion has also been measured using transrectal ultrasound. RESULTS: Based on the computations, seeds located originally close to the urethra travel relatively large distances toward the urethra upon Foley removal, whereas seeds located further away move substantially less. This seed motion leads to higher urethral doses than shown in a standard treatment plan. Dose enhancements increase with catheter size, decrease with increasing prostate volume, are more pronounced for (103)Pd than for (125)I, and range between 3.5% and 32.4%. Postimplant dosimetry is equally affected if images are taken with urethral catheters in place, showing lower urethral doses than actually delivered. Preliminary ultrasound based measurements of seed motion agree with the theory. CONCLUSION: During the implantation procedure, 12 fr or smaller urethral catheters are preferable to larger diameter catheters if urine drainage is sufficient. Treatment planners should avoid planning seeds at 5 mm or closer from the urethra. Special caution is indicated in prostates having about 20 cm(3) or smaller volumes, and when (103)Pd is used. Postimplant dosimetry is susceptible to the same errors.