From 1989 to 1990, 32 poor surgical risk patients with urinary retention were treated with transurethral microwave hyperthermia at the department of urology, University of Leuven in Belgium. Mean patient age was 73 years (range 58 to 90 years) and mean duration of retention was 4 weeks (range 3 to 12 weeks). Followup ranged from 13 to 82 weeks, with a mean of 31 weeks. Bilobar or trilobar hyperplasia was diagnosed in 25 patients (78%), while 7 (22%) had median lobe or median bar hypertrophy. The mean prostatic volume was 52 cc (range 25 to 150 cc). Transurethral microwave hyperthermia was given with a helical antenna at 915 MHz. once or twice per week. The mean number of transurethral microwave hyperthermia sessions was 8.9 (range 5 to 10). Each session consisted of a 60-minute treatment at a mean maximum temperature of 45.4C (range 43.7 to 47.2C), average temperature 43.9C (range 42.7 to 45.5C) and minimum temperature 42.0C (range 40.2 to 43.0C). The temperature was continuously monitored, including thermal mapping in all patients. Of the 25 patients who presented with bilobar or trilobar hyperplasia 18 (72%) were catheter-free for the duration of followup. Of the 7 median lobe or median bar patients 1 (14%) showed sufficient improvement to warrant catheter removal. This patient, however, had recurrent retention 4 months after transurethral microwave hyperthermia. In patients with bilobar and trilobar hyperplasia a strong correlation was observed among maximum temperature (p = 0.0006), average temperature (p = 0.0033) and treatment response. As expected, no such correlation existed between minimum temperature and response to treatment (p = 0.56). Our study has again demonstrated therapeutic activity in patients with benign prostatic hyperplasia treated with transurethral microwave hyperthermia. A new finding was a strong correlation between temperature and response.