BACKGROUND: Animal studies have suggested that drugs inhibiting the enzyme histone deacetylase might have a beneficial effect on multiple sclerosis (MS). Valproic acid (VPA), an anti-epileptic drug, is the only widely used human drug with a histone deacetylase inhibitory effect.

OBJECTIVE: The objective of this paper is to examine if VPA use is associated with a reduced risk of MS.

METHODS: We conducted a propensity score-matched cohort study in the period 1997-2011 linking nationwide register data on filled VPA prescriptions, MS cases, and several covariates. The VPA users were matched on propensity scores in a 1:4 ratio with non-users of VPA. Incidence rates of MS were compared among VPA users and non-users of VPA using Cox regression to estimate hazard ratios (HRs).

RESULTS: Among 16 028 ever-users of VPA and 54 172 non-users, 18 and 26 cases of MS were identified, respectively. Neither current VPA users nor recent users of VPA, who had ceased VPA treatment within the last year, were at a reduced risk of MS compared with non-users of VPA (HR = 1.30 (95% confidence interval, 0.44-3.80), n = 4, and HR = 1.22 (0.28-5.32), n = 2, respectively). Similarly, in an intention-to-treat analysis, ever-users of VPA were not at reduced risk of MS (HR = 2.41 (1.32-4.43), n = 18).

CONCLUSION: In the first human study addressing a possible beneficial effect of VPA use on the risk of MS, we found no support for a protective effect. However, given the wide confidence intervals, only large effects can be ruled out with sufficient certainty. Copyright © 2014 John Wiley & Sons, Ltd.