DEATHS AND COMPLICATIONS FOLLOWING PROSTATECTOMY IN 1400 MEN IN THE NORTHERN REGION OF ENGLAND

Objective To determine the degree of variation in mortality and major morbidity following transurethral resection of the prostate (TURF), and to assess intersite variation for mortality and morbidity over 12 sites within the Northern Region. Further, to determine whether the previously observed effects on morbidity of unit size, patient through-put and emergency admission were borne out in contemporary urological practice in the Northern Region. Patients and methods For an 8 month period, 1 April 1991-31 November 1991, an independent audit of TURF was performed on 12 different hospital sites throughout the Northern Region. A constant data set was designed which was collected on each patient before and 3 months after operation by two independent clinical co-ordinators who travelled to each of the sites. All case notes were reviewed at 3 months after operation by the co-ordinators using a standard proforma, rather than depending upon self reporting by medical staff. Data on factors potentially affecting mortality and morbidity were collected, including emergency admission, diagnosis of prostate cancer, American Society of Anesthesiologists' co-morbidity scores, and age and differences in throughput in the 12 sites. The effect of through-put or 'volume' on mortality and morbidity was assessed by comparing morbidity and the number of cases performed. Results The early mean death rate was 13 of 1396 patients (0.9%), with an inter-site variation ranging from 0% to 3.8%. A mean of 2.0% of men were returned to theatre after TURF, 2.4% of patients received a blood transfusion (>2 units) after operation, and 8.0% of patients developed post-operative sepsis: these complications varied sixfold, sevenfold and 17-fold across the different sites respectively. Those units performing less than or equal to 100 operations over the audit period (equivalent to <150 operation per year) had a significantly increased rate of deaths and complications which was not related to population differences, though some low volume units had good results. Elderly men who were admitted as emergencies or with prostate cancer were particularly vulnerable to complications. Conclusions The overall early mortality rate after TURF for benign prostatic hyperplasia across the Region compares well with other reported large series. The significant variation in morbidity rates found in this study suggests that careful attention needs to be paid by Urologists, Purchasers and Providers to morbidity rates after prostatectomy.