Aims/hypothesis:  

The aim of this study was to evaluate the relation between erectile dysfunction and endothelial functions, coagulation activation, peripheral and autonomic neuropathy in men with Type II (non-insulin-dependent) diabetes mellitus.

Methods:  

We studied 30 Type II diabetic patients with symptomatic erectile dysfunction and 30 potent diabetic patients matched for age and disease. Endothelial functions were assessed with the l-arginine test, plasma thrombomodulin and cell adhesion molecules circulating concentrations. Haemostasis was evaluated with markers of thrombin activation and fibrinolysis. Quantitative sensory testing (vibratory, warming, and heat-pain thresholds), cardiovascular reflex tests and 24-h blood pressure monitoring were used to assess peripheral or autonomic neuropathy.

Results:  

Mean erectile score and HbA_{1 c} were 10.5 ± 5.8 and 8.3 ± 1.6 % in patients with erectile dysfunction, and 24.0 ± 0.7 and 6.8 ± 1.4 % in those without erectile dysfunction, respectively (p < 0.001); there was a significant relation between HbA_{1 c} and erectile function score in patients with erectile dysfunction (r = –0.45, p = 0.02). The decrease in blood pressure and platelet aggregation in response to l-arginine was lower (p < 0.05–0.02) in patients with erectile dysfunction, whereas soluble thrombomodulin, P-selectin and intercellular cell ahhesion molecule-1 concentrations were higher (p < 0.05–0.02). Indices of coagulation activation (F1 + 2 and d-dimers) and reduced fibrinolysis (PAI-1) were also found to be higher in erectile dysfunction patients. Heat-pain and warm perception thresholds, as well as cardiovascular reflex tests, were most commonly abnormal in patients with erectile dysfunction (p < 0.05). In multivariate analysis, HbA_{1 c}, MBP response to l-arginine, P-selectin, indices of coagulation, and quantitative sensory testing were independent predictors of erectile function score.

Conclusion/interpretation:  

Erectile dysfunction in diabetic men correlates with endothelial dysfunction. A reduced nitric oxide activity might provide a unifying explanation. [Diabetologia (2001) 44: 1155–1160]