Botulinum toxin type B for cervical dystonia or involuntary positioning of the head

Studies were short term (16 weeks) employing a single BtB injection session. All were multicentre and conducted in the US. All patients included had previously received BtA. The trials differed with respect to whether or not the patients were still responding to BtA but other entry criteria were similar. All studies used a dose of 10,000 Units of BtB in one group and the technique of administration was the same.
Meta-analysis of three trials enrolling 308 participants showed statistically and clinically significant improvements in the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) total score at week four with a Peto odds ratio (OR) for the number of patients who had at least a 20% improvement of 4.69 (95% CI 2.06 to 10.69) and a weighted mean difference of -5.92 (95% CI -9.61 to -2.23). Subjective rating scales (Patient Global Assessment of Change, Investigator Global Assessment of Change, and Patient Analog Pain Assessment) also improved. Adverse events clearly associated with the mechanism of action of BtB included dysphagia and dry mouth and the number of patients with any adverse event were more frequent in BtB treatment groups. Subgroup analyses showed a clear dose-response relationship for subjective and objective benefit, for frequency and severity of adverse events, and a greater benefit for BtA resistant patients than BtA responders in the primary outcome. The duration of effect was about 16 weeks.

We found three eligible studies enrolling 308 participants. Studies were short term (16 weeks) employing a single BtB injection session. All were multicentre and conducted in the US. All patients included had previously received BtA. The trials differed with respect to whether or not the patients were still responding to BtA but other entry criteria were similar. Patient groups were appropriately selected and well matched. From the methodological point of view these trials were probably not subjected to important selection, performance or attrition bias and all studies used an intention-to-treat analysis.

The dose varied significantly between studies although all used 10,000 Units of BtB in one group and the technique of administration was the same. The primary outcome in all trials was change in TWSTRS total score at week four and other efficacy outcomes were similar between studies. The number of dropouts was small and balanced in all trials. Reasons for withdrawals were given. One randomized double-blind placebo-controlled study was excluded because data couldn't be extracted for the outcomes.
Meta-analysis showed statistically and clinically significant improvements with a Peto odds ratio (OR) of 20% in TWSTRS total score at week four (OR 4.69; 95% CI 2.06 to 10.69) and a weighted mean difference of -5.92 (95% CI -9.61 to -2.23). Subjective rating scales (Patient Global Assessment of Change, Investigator Global Assessment of Change, and Patient Analog Pain Assessment) also improved. The weighted mean difference for changes in these subjective scales varied between -13% to -21%. However, for many of the outcomes, we could not combine data from all studies. Only adverse events clearly associated with the mechanism of action of BtB were more frequent in the treatment group. These included dysphagia and dry mouth. The number of patients with any adverse event was more frequent with BtB.
Subgroup analyses showed a clear dose-response relationship for subjective and objective benefit and for frequency and severity of adverse events. Subgroup analyses showed a greater benefit for the BtA resistant patients than BtA responders in the primary outcome. The duration of effect was about 16 weeks. These trials did not measure quality of life nor did they establish the long term duration of effect or immunogenicity