|| Checking for direct PDF access through Ovid
We investigated predictive factors of failure and performed a resource consumption analysis in patients who underwent active surveillance for nonmuscle invasive bladder cancer.This prospective observational study monitored patients with a history of pathologically confirmed stage pTa (grade 1-2) or pT1a (grade 2) nonmuscle invasive bladder cancer, and recurrent small size and number of tumors without hematuria and positive urine cytology. The primary end point was the failure rate of active surveillance. Assessment of failure predictive variables and per year direct hospital resource consumption analysis were secondary outcomes. Descriptive statistical analysis and Cox regression with univariable and multivariable analysis were done.Of 625 patients with nonmuscle invasive bladder cancer 122 with a total of 146 active surveillance events were included in the protocol. Of the events 59 (40.4%) were deemed to require treatment after entering active surveillance. Median time on active surveillance was 11 months (IQR 5–26). Currently 76 patients (62.3%) remain under observation. On univariable analysis only time from the first transurethral resection to the start of active surveillance seemed to be inversely associated with recurrence-free survival (HR 0.99, 95% CI 0.98–1.00, p = 0.027). Multivariable analysis also revealed an association with age at active surveillance start (HR 0.97, 95% CI 0.94–1.00, p = 0.031) and the size of the lesion at the first transurethral resection (HR 1.55, 95% CI 1.06–2.27, p = 0.025). The average specific annual resource consumption savings for each avoided transurethral bladder tumor resection was €1,378 for each intervention avoided.Active surveillance might be a reasonable clinical and cost-effective strategy in patients who present with small, low grade pTa/pT1a recurrent papillary bladder tumors.