To evaluate the clinical impact on progression and recurrence according to presence and absence of a muscle layer, we conducted a retrospective, multicenter study.We retrospectively reviewed 247 patients who received transurethral resection (TUR) of bladder tumors and were pathologically diagnosed as having T1G3 bladder cancer from 1990 to 2009. We ruled out 8 patients who received immediate cystectomy and analyzed the remaining 239 T1G3 patients. Patients who had invasion to the prostatic urethra and patients who underwent a second TUR were not included.TUR specimens from 194 patients were confirmed to have a definite muscle layer and those from 45 did not. The median follow-up period was 53 months, ranging from 3 to 181 months. The progression-free survival rates at 5 years after TUR were 91.1 % for patients who had a muscle layer in their specimen and 77.3 % for those who did not (p = 0.005, log-rank test). Multivariate analysis indicated that the absence of a muscle layer was a risk factor for progression (p = 0.006, Cox proportional hazards analysis).Patients without a muscle layer in the specimen had high risk for progression. The initial TUR must have a muscle layer in the specimen. Variations of progression rates in previous studies might be due to different proportions of patients who had a muscle layer in TUR specimens.