The National Prostatectomy Audit: the clinical management of patients during hospital admission

OBJECTIVE: To determine everyday practice in the hospital management of men undergoing prostatectomy and the extent of its variation. PATIENTS AND METHODS: A total of 5361 patients, who represented 89% of all those undergoing prostatic procedures in four health regions (Mersey, Wessex, Northern and South West Thames) and one test site (within Trent) were recruited by 103 (97%) surgeons. Clinical information was collected on a pre-coded data collection form which was completed during the hospital stay by the principal operator. Patient identification occurred at the time of surgery. RESULTS: Important findings included: (i) both older men and those of higher social class were more likely to undergo prostatectomy with fewer symptoms; (ii) men who waited longer for surgery had worse symptoms by the time of their operation; (iii) there were unexplained differences in routine pre- and post-operative investigation and treatment. Half the men had their flow rate or residual urine measured as part of their pre-operative assessment. About half the men received prophylactic antibiotics; (iv) when large groups were analysed, a consistent proportion of men throughout the study (12%) were undergoing the operation for a second time. The clinical course of men having a repeat operation differed in many ways from those having a first time procedure; (v) the larger proportion of men (62%) had surgery for strong indications as opposed to symptoms alone; (vi) although most operations were performed by consultants, emergency admissions, though symptomatically more severe and sicker, were more likely to be operated on by trainee surgeons; (vii) significant variation in mean pre-operative symptom severity and bother scores were seen between surgeons. CONCLUSION: The clinical management of prostatectomy has been defined in a large and representative UK sample. In some circumstances consistent variations have been identified. It is not yet clear whether these variations influence outcome. These data can be used by surgeons wishing to compare their own patient management with that described here.