| Popis: |
To analyze if there is a learning curve to get to perform high quality ultrasound guided prostate biopsies that enable to effectively perform the diagnosis of prostate cancer, taking the pathological result of the samples into consideration.We retrospectively reviewed data from 790 ultrasound-guided biopsies performed consecutively in our center between May 2009 and December 2012 by four surgeons. We reviewed the following data: Surgeon, date of intervention, patient age, PSA, rectal digital examination (DRE), ultrasound, prostate volume, number of cores obtained, pathology result of the biopsy. The surgeons were residents in Urology in their first year in the department. For statistical analysis the biopsies were grouped in blocks of 25 and 75 cases performed by a surgeon, in a chronological order. To evaluate the influence of the learning curve on diagnosis we performed univariate and multivariate analysis between the different variables and pathological report.Median age was 67 years (42-90). 300 biopsies (38%) were positive for prostate cancer. On univariate analysis, regarding cancer detection rate, the variables surgeon, PSA, DRE, ultrasound result, prostate volume, number of cores and groups of 75 showed statistically significant differences, this latter showing 32% in the first 75 biopsies and 43.2% in the last group. The variables age (p=0.11) and groups of 25 patients (p=0.07) did not show differences. Nevertheless, on multivariate analysis only PSA, DRE, prostate volume and groups of 75 patients were statistically significant, this latter with an OR of 1.35 (95%CI 1.09-1.66).The results obtained in our study show that the number of procedures performed, namely surgeon experience, does have an influence to achieve an optimal diagnostic yield in ultrasound guided prostatic biopsies. |
