A Cochrane review included 8 RCTs with a total of 332 participants. Five studies (n=235) compared IVIg with placebo, one trial (n=20) with plasma exchange, one trial (n=32) with prednisolone, and one trial (n=46) with iv. methylprednisolone. A total of 2 g/kg of IVIg was administered in each trial over 2 to 5 days. A significantly higher proportion of participants improved in disability within one month after IVIg treatment as compared with placebo (RR 2.40, 95% CI 1.72 to 3.36; NNT for an additional beneficial outcome 3.03 (95% CI 2.33 to 4.55), 5 trials, n=235). Whether all these improvements are equally clinically relevant cannot be deduced because each trial used different disability scales and definitions of significant improvement. In 3 trials (n=84), the disability score could be transformed to the modified Rankin score, on which improvement of one point after IVIg treatment compared to placebo was not significant (RR 2.40, 95% CI 0.98 to 5.83). One placebo-controlled study (n=208) with a long-term follow-up showed that IVIg improves disability more than placebo over 24 and 48 weeks. The mean disability score revealed no significant difference between IVIg and plasma exchange at 6 weeks. There was no significant difference in improvement in disability on prednisolone compared with IVIg after 2 or 6 weeks, or on methylprednisolone compared to IVIg after 2 weeks or 6 months. There were no statistically significant differences in frequencies of side effects between the 3 types of treatment for which data were available (IVIg vs. placebo or steroids). Mild and transient adverse events were found in 49% of participants treated with IVIg, while serious adverse events were found in 6%.
Comment: The quality of evidence is downgraded by indirectness (short follow-up time) and imprecise results (limited study size for each comparison).