In this study, we employed a sensitive activity assay system to measure cerebrospinal fluid (CSF) and serum levels of active matrix metalloproteinase-9 (MMP-9) in 37 relapsing-remitting (RR), 15 secondary progressive (SP) and nine primary progressive (PP) multiple sclerosis (MS) patients, grouped according to clinical and magnetic resonance imaging (MRI) evidence of disease activity. We also studied, as neurological controls, 48 patients with other inflammatory neurological disorders (OIND) and 48 with non-inflammatory neurological disorders (NIND). To assess active MMP-9/TIMP-1 circuit, CSF and serum levels of MMP-9 tissue inhibitor TIMP-1 were quantified by ELISA in the same patient population. CSF mean levels of active MMP-9, CSF active MMP-9/TIMP-1 ratios and intrathecal active MMP-9 synthesis, as indicated by specific index, were more elevated in MS than in NIND (P < 0.05, < 0.02 and < 0.02, respectively), serum active MMP-9/TIMP-1 ratio was higher in MS (P < 0.01) and OIND (P < 0.02) than in NIND, and serum TIMP-1 concentrations were lower in MS than in NIND (P<0.05). More importantly, serum active MMP-9 mean levels, serum active MMP-9/TIMP-1 ratio and intrathecal production of active MMP-9 were increased in MS patients with clinical (P < 0.001, < 0.001 and < 0.05, respectively) and MRI (P < 0.001, < 0.001 and < 0.02, respectively) disease activity, whereas CSF mean concentrations of active MMP-9 and CSF active MMP-9/TIMP-1 ratio were enhanced only in MS patients with MRI evidence of disease activity (P < 0.02 and < 0.01, respectively). Altogether, these findings suggest that a shift in MMP-9/TIMP-1 balance towards proteolytic activity of MMP-9 could be relevant in MS immune dysregulation. In addition, our results indicate that CSF and serum levels of active MMP-9 may represent a potential surrogate biomarker for monitoring MS disease activity. In particular, serum active MMP-9/TIMP-1 ratio seems to be a very appropriate indicator of ongoing MS inflammation, since it is easily measurable.