A computerized study of past infectious events and neurological parameters revealed the existence of a history of repeated respiratory-tract infections (RRI) beginning in childhood, in 52% of 251 multiple sclerosis (MS) patients studied. The 251 MS patients were divided into two groups: those without a past history of RRI were designated as "MS type I", and those with a history of infectious problems before the onset of MS as "MS type II". Significant differences in the neurological symptoms, the treatment received and some general parameters were found between the groups, which suggested a correlation between the evolution of MS and the presence or not of RRI in these patients. When compared to the MS type I group, a significantly higher percentage of MS type II patients reported visual problems (P less than or equal to 0.01), paresthesia (P less than or equal to 0.01), loss of sensitivity (P less than or equal to 0.03), pain (P less than or equal to 0.004), motor problems (P less than or equal to 0.016) and sexual dysfunction in males (P less than or equal to 0.02). The mean number of attacks in the first 5 years of the disease was significantly more frequent in MS type II patients, 6.2 compared to 2.9 (P less than or equal to 0.02). A significantly higher percentage of MS type II patients also received oral corticosteroids (P less than or equal to 0.02) or ACTH (P less than or equal to 0.003). Although the age of onset of MS was the same for both groups, MS type II patients were significantly younger than MS type I patients, the mean age being 36 years compared to 41 years (P less than or equal to 0.001). Only 12% of patients in the MS type II group compared to 30% in the MS type I group had the disease for more than 15 years (P less than or equal to 0.001). As is usual with MS, the majority of the patients in both groups were females, 79.3% in the MS type II compared to 63.4% in the MS type I group. These findings suggest that MS patients with a past history of RRI (MS type II) have a different evolution of their disease from MS type I patients and that in general the disease is more severe. The past infectious history of patients would thus appear of putative value, in addition to neurological criteria, in assessing the probable future evolution of the disease.