Nine studies with 892 patients were included.
The cohort assembly methods of studies varied: two studies selected consecutive MS suspects from patients referred to EP laboratories; one study was population-based; and the method could not be determined in another two studies. The number of MS suspects enrolled ranged from 21 to 222. The spectrum of the disease varied considerably.
The prospective design meant that the EPs in all studies were interpreted without knowing which patients develop CDMS. However, none of the studies were described as having used techniques to ensure that the EPs were interpreted without knowledge of the patients' clinical presentations. Many articles did not discuss the patients lost to follow-up.
The criteria used to diagnose CDMS appeared similar in most studies. The percentage of MS suspects that developed CDMS during the follow-up period ranged from 17 to 51%. There was a trend for studies with longer follow-up periods to have a higher frequency of CDMS.
Five studies were classed as Grade IV because it was not possible to determine if an acceptable 'gold' standard for diagnosing MS had been used, and/or it was not possible to calculate the strength of the EP-CDMS association and other parameters describing diagnostic accuracy. These studies were not considered in further analyses because of these serious methodological limitations.
Visual EPs (3 studies): an association between abnormal visual EPs and an increased risk of CDMS has been established with moderate clinical certainty.
Somatosensory EPs (4 studies): the evidence describing the relationship between abnormal somatosensory EPs and the development of CDMS was inconclusive and conflicting.
Brainstem auditory EPs (3 studies): the absence of a useful association between abnormal brainstem auditory EPs and an increased risk of CDMS has been established with moderate clinical certainty.
Multimodal EPs (2 studies): a slight gain in sensitivity from using multimodal EPs was offset by a greater loss in specificity.