Objectives: To find out whether any of a preset battery of clinical and laboratory parameters has significant correlation with blood loss caused by transurethral prostatectomy (TURP).Patients and Methods: All new patients undergoing a TURP over a 1–year period were included in the study. For each patient the following parameters were documented: (1) Pre–operatively: Age, mode of presentation, blood pressure, complete blood count, coagulation screening, calcification of abdominal blood vessels on plain X–ray, prostate size on digital rectal examination (DRE), pre–operative urine culture, presence of haematuria, ECG changes and drugs taken. (2) Intraoperatively: prostate morphology at cystoscopy, type of anaesthesia, operating surgeon, operating time, weight of resected prostate tissue and any blood transfusions. (3) Postoperatively: Length of irrigation time and prostate histology. All the intra–operative and postoperative blood loss was calculated. Their sum constituted total blood loss which constituted the dependent variable of the study. Statistical methods used included bivariate correlation, Students t test or the χ2 test as appropriate.Results: Of 140 TURPs, 121 were eligible for final analysis. The mean total blood loss was 552 ml (34 ml per gram of tissue) and the mean resected prostate weight was 18 g. Of the continuous variables, the one with strongest correlation with total blood loss was resected prostate weight (r = 0.47, p = 0.000) followed by prostate size on DRE (r = 0.45, p = 0.000) then operating time (r = 0.39, p = 0.000). Of the categorical variables, the one with the strongest correlation with total blood loss was preoperative urine culture (p = 0.01).Conclusions: The only reversible factor associated with blood loss caused by TURP is pre–operative urinary infection. Operating time is partly controllable, as it may improve (i.e. decrease) with experience of the operator.