Botulinum toxin type B for sialorrhoea in children with cerebral palsy: a randomized trial comparing three doses

Aim The aim of the present study was to evaluate the efficacy and safety of three doses of botulinum toxin type B (BoNT‐B) in reducing persistent sialorrhoea in children with cerebral palsy (CP). Method Children with CP and refractory sialorrhoea were randomized to one of four groups: a control group and three experimental groups receiving a low (1500 mouse units MU]), medium (3000MU), or high (5000MU) dose of BoNT‐B respectively, into bilateral salivary glands. Drooling was measured using the Thomas‐Stonell rating scale, and the weight and the number of bibs used per day were counted in all children at baseline, 4, and 12 weeks after BoNT‐B injection. Results Twenty‐seven children (15 males, 12 females; mean age 7y 10mo, SD 1y 6mo; range 5–15y) were randomized into a control (seven children: four males, three females) and experimental groups receiving low (six children: four males, two females), medium (seven children: four males, three females), and high (seven children: three males, four females) doses of BoNT‐B respectively. All children had mixed neurological disorders consisting of spastic paraparesis, tetraparesis, dystonic movements, and ataxia. Gross Motor Function Classification System levels ranged from III to V, and all children had moderate or severe intellectual disability. Estimated means with their standard errors (SEM) of drooling were at baseline, 4, and 12 weeks respectively, as follows: control group, 12.1 (2.1), 11.9 (2.1), 11.8 (2.2), p for trend 0.992; low dose group, 13.8 (2.3), 11.4 (2.3), 13.9 (2.3), p for trend 0.952; medium dose group, 13.9 (2.1), 6.7 (2.1), 7.1 (2.1) p for trend 0.008; and for the high dose group 14.4 (2.1), 5.0 (2.1), 5.6 (2.1), p for trend 0.002. Side effects included dense saliva, xerostomia, and difficulty in swallowing, and were more frequent in the high‐dose group. Interpretation A 3000MU injection of BoNT‐B into the salivary glands significantly improved the frequency and severity of sialorrhoea in children with CP. The lower dose was ineffective, and the higher dose produced no greater benefit and more side effects.