The present randomized study evaluated the physiological strain in transurethral (TURP) versus transvesical (TVP) prostatectomy in 75 patients with clinically benign, medium-sized prostatic enlargement. In both groups 5 patients had coincidental prostatic carcinoma. During TURP a suprapubic trocar was inserted for irrigation with 1.5% glycine. Antibiotics were given to patients with preoperative urinary tract infection and postoperatively tranexamic acid was administered to all patients. Coexisting diseases which might increase the operative risk were equally present in TURP and TVP patients. No patient died in connection with the operative procedures. Median duration of anaesthesia for TVP was 95 min and for TURP 60 min (p less than 0.001). Almost identical operative blood loss was seen in the two groups. However, more blood transfusion were given to TVP patients. Following discharge, more episodes of bleeding were registered in the TURP group, but no significant difference was observed in number of secondary operations for bleeding. In the TVP group, more patients had pulmonary complications, elevated rectal temperature as well as antibiotic treatment and 34% of these patients developed wound infection. Secondary operations before discharge were also more frequently undertaken in TVP patients. No differences between the two groups were seen pre- or postoperatively with respect to electrolyte, protein, creatinine and haemoglobin values. In the patients with incidental prostatic carcinoma a TURP was the most gentle procedure. With respect to physiological strain a TURP is clearly to be preferred over a TVP for obstructive clinically benign, medium-sized prostatic hyperplasia.