Sexual dysfunction is a very important but often overlooked symptom of multiple sclerosis. To investigate the type and frequency of symptoms of sexual dysfunction in patients suffering from multiple sclerosis, we performed a case-control study comparing 108 unselected patients with definite multiple sclerosis, 97 patients with chronic disease and 110 healthy individuals with regard to sexual function, sphincteric function, physical disorders impeding sexual activity and the impact of sexual dysfunction on social life. Information has been collected from a face-to-face structured interview performed by a doctor of the same gender as the patient. The disability, the cognitive performances, the psychiatric conditions and the psychological profile of patients and controls have been assessed. Sexual dysfunction was present in 73.1% of cases, in 39.2% of chronic disease controls and in 12.7% of healthy controls (P<0.0001). Male cases reported symptoms of sexual dysfunction more frequently than female cases (P<0.002). Symptoms of sexual dysfunction more commonly reported in patients with multiple sclerosis were anorgasmia or hyporgasmia (37.1%), decreased vaginal lubrication (35.7%) and reduced libido (31.4%) in women, and impotence or erectile dysfunction (63.2%), ejaculatory dysfunction and/or orgasmic dysfunction (50%) and reduced libido (39.5%) in men. Seventy-five per cent of cases, 51.5% of chronic disease controls and 28.2% of healthy controls (P<0.0001) experienced symptoms of sphincteric dysfunction. In conclusion, a substantial part of our sample of patients with multiple sclerosis reported symptoms of sexual and sphincteric dysfunction. Both sexual and sphincteric dysfunction were significantly more common in patients with multiple sclerosis than in either control group. Our findings suggest that a peculiar damage of the structures involved in sexual function is responsible for the dysfunction in patients with multiple sclerosis, but the highly significant lower frequency of symptoms of depression and anxiety in healthy controls may also imply a possible causative role of psychological factors.