Twin study of multiple sclerosis: an epidemiologic inquiry.

The National Research Council Twin Registry comprises 16,000 pairs of white male twins, both members of which had been in military service, mainly in World War II. All their available military and Veterans Administration records and their responses to a 1965 to 1970 NRC questionnaire have been coded as to disease. Upon review we found 16 cases of multiple sclerosis (MS) among 15 pairs of twins, for an age-specific prevalence rate of 51 per 100,000 veterans aged 43 to 53--about half the expected frequency. Of the 15 sets, three sets refused cooperation and three were unavailable for study. Nine sets were examined and interviewed together with the mother. One of five monozygotic twin pairs was concordant for MS and in another the co-twin of an MS case had had a solitary episode of retrobulbar neuritis; all others were discordant. There were more definable environmental events (as noted below) among the affected twins than among the unaffected co-twins. The greatest excess was within the 20 years before onset. Summing events across the four 5-year periods before onset, among the 10 MS versus the eight not-MS individuals, there were 5:1 instances of trauma, 8:2 or operation, 7:1 of ether anesthesia, 7:1 of allergy, 10:5 of infection, and 9:0 of animal exposure. Summing these same events within each 5-year period, the MS:control ratios were 9:1, 10:2, 12:3, and 15:4, respectively, for 0 to 4, 5 to 9, 10 to 14, and 15 to 19 years before onset.