Materials and methods Three groups of six rats each were assessed; two groups (1 and 2) were castrated and the third (group 3) underwent a sham operation (control). Eight weeks after castration, group 2 received a subcutaneous injection with testosterone. At 8 weeks, the rats in group 1 and 3 underwent a final functional analysis while those in group 2 did so at 12 weeks. The evaluation included a subcutaneous injection with apomorphine to study centrally mediated erection, and cavernosal nerve electrostimulation and papaverine injection to study peripherally mediated erection. At death a penile mid-shaft specimen was taken for NADPH-diaphorase staining.

Results In the apomorphine study, castration resulted in significantly fewer yawns and erections than in the control, and those in group 2 significantly better central erectile function than in the controls. The mean (sem) number of nitric oxide synthase (NOS)-containing nerve fibres in the corpora cavernosa and both dorsal nerves of castrated rats, at 46.2 (9.1) and 203 (32.1), respectively, were significantly lower than in rats in group 2, at 84.1 (11.2) and 300.6 (17.1), and than in the controls, at 88.6 (10.9) and 306.3 (22.9), respectively. The intracavernosal pressure decreased significantly in the absence of testosterone, both after electrostimulation and intracavernosal papaverine injection. However, there was no difference between the control and group 2 rats in either the number of NOS-containing nerve fibres or in the peripheral erectile functional study.

Conclusions Testosterone acts on the nervous system to mediate erection; when it is absent there may be down-regulation of both the production and activity of NO, thereby decreasing the response to peripheral stimulation via the NO pathway. The restoration of erectile function seen in rats in group 2 supports this phenomenon. Delayed testosterone replacement has no detrimental effect on the restoration of the erectile mechanism after castration.