The use of quality indicators for assessing radical prostatectomy specimens

The information contained in pathology reports of radical prostatectomy spe cimens is critically important to treating physicians for selecting adjuvan t therapy, evaluating therapy, estimating prognosis, and analyzing outcomes , This information is also of importance to patients and their families. In recent years, the Cancer Committee of the College of American Pathologists and the Association of Directors of Anatomic and Surgical Pathology develo ped suggested protocols for reporting the findings on radical prostatectomy specimens. The objectives of this study were to assess radical prostatecto my-specimen reports by using quality indicators derived from existing sugge sted protocols and to thereby assist pathologists in improving the quality of their reports on such specimens. A retrospective chart review of 554 cas es for the second 6-month period of 1996 focused on 10 quality indicators: submission of a frozen section; location of the adenocarcinoma; proportion of the specimen involved by adenocarcinoma; perineural involvement; vascula r involvement; seminal vesicle involvement; periprostatic fat status; numbe r of nodes submitted; status of nodes; and prostate intraepithelial neoplas ia (PIN). The findings of this study were shared with the pathology departm ents in all hospitals in New York State, In addition, the 113 hospitals fro m which the 554 cases were drawn were given their institution-specific data . Teleconferences were held with the 37 hospitals that accounted for 72.4% of all cases. These conferences included directors of pathology and laborat ories and focused on the aggregate statewide findings. The presence of qual ity indicators in reports varied from a mean of 14.8% (periprostatic fat) t o a mean of 85.9% (seminal vesicle involvement). For all hospitals, 4 indic ators (proportion of the specimen involved by adenocarcinoma, vascular invo lvement, periprostatic fat status, and PIN) were included in fewer than 50% of cases. These 4 quality indicators and an additional 3 others (submissio n of a frozen section, perineural involvement, and the number of nodes subm itted) were included in fewer than 70% of cases. Only 3 indicators (locatio n of the adenocarcinoma, seminal vesicle involvement, and the status of nod es) were found in more than 70% of cases. Although the mean level of qualit y indicator inclusion ranged from 14.8% to 85.9% for all cases examined, th e absolute range for any individual indicator was 0% to 100%. Thus, some ho spitals included a given indicator 100% of the time; others never included it. This pattern held true for all 10 indicators. High-volume hospitals (10 or more cases) performed significantly better than low-volume hospitals (1 -4 cases) on 5 indicators (P < .05), and better, but not significantly so, for an additional 2 indicators. Overall, the mean inclusion levels for all 10 indicators were 10% higher for high-volume hospitals compared with low-v olume and medium-volume hospitals (5-9 cases). This study demonstrated wide variations in the inclusion of quality indicators by pathologists in their radical prostatectomy-pathology reports. Whereas some hospitals always inc lude given indicators, others never mentioned them. These marked disparitie s point to the need for standardized reporting for radical prostatectomy sp ecimens.