Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Intravenous immunoglobulin for MS

Randomised trial [2]

Intravenous immunoglobulin for multiple sclerosis has excited interest because early, uncontrolled, studies seemed to show benefit. Because of the nature of the condition and the dearth of effective therapies, interest was natural, and led to small but inconsistent randomised trials. Bandolier looked at those trials in early 2002, and concluded that the evidence then available was insufficient. Large trials, of sufficient duration to prove the point were needed, and since 2002 two new trials have reported [1, 2].

One [1] was small, with 49 patients, comparing two doses of immunoglobulin with placebo over a year. It found benefits for immunoglobulin over placebo, but the numbers were very small, with no more than 17 patients in a group. A larger (318 patient) high quality study [2] now shows definitively the lack of benefit. It is worth visiting in some detail.

Randomised trial [2]

The trial recruited patients aged 18-55 years with secondary progressive multiple sclerosis, active disease, disease duration of at least three years, and an EDSS score of 3.0 to 6.5. The Expanded Standard Disability Scale (EDSS) at level 3 is moderate disability in one functional score or mild disability in three or four functional scores, though fully ambulatory, and at 6 is intermittent or unilateral constant assistance (walking stick, crutch, brace) required to walk about 100 metres with or without resting.

Secondary progression was defined as continued deterioration of disability for at least 12 months with or without interposed relapses after an initial relapsing-remitting course. Excluded were patients with an attack within 30 days of starting date, patients without MRI brain lesions consistent with multiple sclerosis, and various previous treatments, at least without a treatment-free period.

Recruited patients were randomised to intravenous immunoglobulin at 1 gram/kg per month up to a maximum of 80 grams, or 0.1% albumin in maltose. The last infusion was given at month 26. Randomisation was performed separately, and active and placebo treatments appeared identical. Concomitant treatments with immunomodulating or immunosuppressing drugs were not allowed.

Assessments were made every three months to 27 months, and MRI scans were taken at baseline and 12 and 24 months. Assessments were made by a neurologist who was not involved with treatment administration, and who was not allowed to discuss therapy or adverse events with the patient. The primary outcome was time to start of a confirmed treatment failure. This was a deterioration of 1 point on the EDSS score if it was initially below 6, or 0.5 points if it was 6 or higher. Other outcomes included a 20% worsening of a nine-hole peg test of arm functioning, and various functions of MRI imaging.


Treatment and placebo groups were well matched at baseline. They had an average EDSS score of 5.2, and an average time of five years since the beginning of the secondary progressive phase of the condition. About half had had a relapse during the preceding 24 months.


There was no difference between the groups for any efficacy measure. The time to event analysis between immunoglobulin and placebo was identical, as was the total number with treatment failures, 20% deterioration on the peg test, or both, or clinical relapses (Table 1). The annual relapse rate was 0.5 in both groups. There were no differences between the groups on MRI imaging, though the total lesion load in the brain remained almost unchanged over two years in both groups.

Table 1: Efficacy and harm outcomes for IVIG in MS

Treatment failure (EDSS criterion)
77 (48%)
70 (44%)
Deterioration on peg test by 20%
55 (35%)
53 (33%)
Deterioration on EDSS and peg test
98 (62%)
92 (58%)
77 (48%)
83 (52%)
All-cause discontinuation
39 (25%)
19 (12%)
Adverse effect discontinuation
10 (6%)
5 (3%)
Drug-related adverse events
113 (71%)
91 (57%)
Deep vein thrombosis
6 (4%)
1 (0.6%)
Pulmonary embolism
4 (3%)
0 (0%)


More patients discontinued for any reason, or because of adverse events on immunoglobulin, than on placebo. Drug-related adverse events were common, but more so with immunoglobulin. These included infusion-related adverse events of fever, chills and headache, as well as rashes and eczema. Serious adverse events included deep vein thrombosis and pulmonary embolism, both of which occurred more frequently with immunoglobulin.


Strong negative trials are incredibly helpful because they give us a very positive signal that an intervention does not work in a particular condition. Strength means fulfilling the criteria of quality, validity, and size. This trial did just about everything to randomise and blind the study, and it told us what happened to all the patients. It was large. It was valid, both by measuring a range of important outcomes, including objective outcomes obtained by brain scanning, and by being long. Its two-year duration ensured that it was trustworthy.

It tells us that there were no important benefits. It also tells us that there may be important harm, with serious adverse events of venous thrombosis and pulmonary embolism in one in 14 patients given immunoglobulin.

Most of the trials of immunoglobulin in multiple sclerosis have been in relapsing remitting disease. Their results have been mixed. Now all we need is an equivalent large and valid trial of immunoglobulin to settle the matter.


  1. M Lewanska et al. No difference in efficacy of two different doses of intravenous immunoglobulins in MS: clinical and MRI assessment. European Journal of Neurology 2002 9: 565-572.
  2. OR Hommes et al. Intravenous immunoglobulin in secondary progressive multiple sclerosis: randomised placebo-controlled trial. Lancet 2004 364: 1149-1156.

previous or next story