부산대학교 도서관

Aims: To investigate the features which are most predictive of IVIg response in patients with CIDP. Methods: We included 38 consecutive CIDP diagnosed according to Rotta et al. (2000). Anti-MAG antibodies (Ab) assay and serum immunofixation were performed in all. IVIg (0.4 g/Kg/day) were the first treatment in all patients. Response to IVIg was defined as increase by at least 1 grade of the Rankin scale score at 1 month. The following features were considered: 1) proximal as well as distal weakness, 2) pure clinical sensory pattern, 3) disease duration lower than 12 months, 4) conduction block in at least 1 motor nerve, 5) mean lower/upper limbs motor nerve CMAP amplitude lower than 50% of lower limit of normal, 6)M monoclonal gammopathy, and 7) anti-MAG Ab. Likelihood ratio (LR) of the features significantly associated with treatment response was evaluated in order to determine the increase or decrease of response to IVIg probability. Results: Twenty-six patients were classified as responders and 12 as non-responders. Age (60.6 ± 9.7 vs. 63.6 ± 9.4 years), disease duration (24 vs. 42 months) and sex ratio (18 vs. 9 male) were not different among the groups. Presence of conduction block in at least one motor nerve was associated significantly with treatment response (responders 50% vs. non-responders 8.3%, p: 0.02) and presence of anti-MAG Ab was associated with treatment failure (non-responders 50% vs. responders 7.6%, p: 0.007). The remaining features were not significantly associated with treatment response. LR of presence on conduction block was 6, whereas LR of anti-MAG Ab was 0.15. Considering 70% as mean percentage of responders, the finding of conduction blocks increase the probability of response to IVIg to 89%, whereas the finding of anti-MAG antibody decreases this probability to 25%. Conclusion: Presence of conduction blocks and absence of anti-MAG antibodies seem to be good predictors of response to IVIg in patients with CIDP. [ABSTRACT FROM AUTHOR]