Transrectal ultrasound-guided intraprostatic injection of absolute ethanol with and without carmustine: a feasibility study in the canine model.
OBJECTIVES: To develop a reliable intraprostatic injection technique and to define the local and systemic toxicity of intraprostatic injection of dehydrated ethanol with and without carmustine. METHODS: Twenty-three random-source male canines were divided into a control group (n = 3), a dehydrated ethanol-alone group (group 1, n = 10), and a dehydrated ethanol-plus-carmustine group (group 2, n = 10). A reliable intraprostatic injection technique was developed with the control animals. The optimal volume of dehydrated ethanol for intraprostatic injection and the local tissue effects of dehydrated ethanol injection were defined with group 1. The local tissue effects of escalating doses of carmustine were defined with group 2. All animals were injected under general anesthesia using transrectal ultrasound (TRUS) guidance. Fourteen days after injection, a repeated TRUS of the prostate was done, the animals were killed, and the bladder, prostate, and periprostatic tissues were excised for pathologic examination. RESULTS: Sonographic changes in the prostate 2 weeks after injection were present in all group 1 and 2 animals. All prostates had varying amounts of hemorrhagic and coagulative necrosis, which correlated with the TRUS findings. There were no differentiating pathologic features between group 1 and group 2 specimens. The relative amount of necrosis varied with the doses of dehydrated ethanol and carmustine injected, but was not predictable on the basis of the doses administered. Subclinical prostatic microabscesses were identified in 6 of 10 group 1 animals and 4 of 10 group 2 animals. Only group 2 animals had alterations in their blood chemistry results, all of which were self-limited. Two had white blood cell nadirs of less than 2000 5 days after injection. No animals developed incontinence, and there were no rectal injuries. CONCLUSIONS: Intraprostatic dehydrated ethanol and carmustine injections were readily controllable under TRUS guidance and resulted in hemorrhagic and coagulative necrosis of prostatic tissue with minimal associated morbidity and no incontinence in the dog model. Hematologic changes observed in the animals that received carmustine were self-limiting.