Severe sialorrhoea (drooling) in children and young people with chronic neurological disorders: oral glycopyrronium bromide

Clinical effectiveness

This evidence summary is based on 2 double-blind, placebo-controlled RCTs of children and young people with chronic neurological disorders and severe sialorrhoea (defined in Zeller et al. 2012a as drooling in the absence of treatment such that clothing became damp approximately 5 to 7 days per week). Both studies had an 8 week treatment duration, which included a 4 week dose titration phase.

The formulations of glycopyrronium bromide used in the 2 RCTs were not the new licensed formulation of glycopyrronium bromide (Sialanar). However, in line with Article 10(a) of Directive 2001/83/EC, license applications for a medicine that has a well-established use with a recognised efficacy and safety profile, can be supported by a bibliographic application that does not require the manufacturer to carry out new clinical trials with their formulation of the drug (EPAR: glycopyrronium).

Mier et al. (2000)

Mier et al. (2000) was a dose-ranging, crossover RCT in 39 children and young people aged 4 to 19 years with neurological conditions and severe sialorrhoea. Participants were randomised to 8 weeks of oral glycopyrronium bromide capsules or placebo capsules 3 times daily. This was followed by a 1‑week washout period and a second 1‑week observation period, then a crossover to 8 weeks of the alternative intervention.

The initial glycopyrronium bromide dose was based on the participant's weight, and increased weekly over 4 weeks. Doses were increased according to a pre-defined schedule unless adverse effects occurred or desired 'dryness' (defined by the parent or carer) occurred. The maximum tolerated dose was then continued for a further 4 weeks.

Over 8 weeks there was a statistically significant improvement in the primary outcome of mean modified Teacher's Drooling Scale (mTDS) score with glycopyrronium bromide (from a mean baseline score of 7.52 to a maximum [best] mean score of 1.85) compared with placebo (from a score of 7.44 to 6.33), with a difference between groups of 4.48 points (p<0.001, 95% confidence interval [CI] not reported).

Glycopyrronium bromide appeared to have a dose–response relationship, with participants on the lowest doses of glycopyrronium bromide at the end of the study having the highest mTDS score (indicating more severe drooling), while participants on the higher doses had lower mTDS scores.

Zeller et al. (2012a)

Zeller et al. (2012a) was an RCT in 38 people aged 3 to 23 years with cerebral palsy or another neurological condition, and severe sialorrhoea. Participants were randomised to 8 weeks of 3 times daily glycopyrronium bromide oral solution or placebo oral solution. The initial glycopyrronium bromide dose was based on weight and increased weekly for 4 weeks until the optimal tolerated response was achieved. Participants then continued on the same dose for a further 4 weeks.

After 8 weeks there was a statistically significant improvement in the primary outcome of 'responder rate' (the proportion of participants with an improvement of at least 3 points in mTDS score with glycopyrronium bromide (14/19, 73.7%) compared with placebo (3/17, 17.6%, p=0.001). A statistically significant difference in responder rate was also observed after 2 weeks treatment in the glycopyrronium bromide group (52.6%) compared with the placebo group (0%, p=0.0007).

Mean improvements in mTDS score at 8 weeks were statistically significantly greater with glycopyrronium bromide (3.94 points, 95% CI 2.97 to 4.91) compared with placebo (0.71 points, 95% CI −0.43 to 1.84, p<0.0001), with a difference between groups of 3.23 points.

Statistically significant differences were observed for the global assessments of study medication by investigators and parents or carers. In the glycopyrronium bromide group 84.2% of investigators and 100% of parents or carers felt that the treatment was worthwhile, compared with 41.2% of investigators (p=0.014) and 56.3% of parent or carers in the placebo group (p=0.0017).

An overview of the results for clinical effectiveness can be found in results tables.

Safety and tolerability

The SPC advises that glycopyrronium bromide is contraindicated in children and young people with glaucoma, urinary retention, severe renal impairment, a history of intestinal obstruction, ulcerative colitis, paralytic ileus, pyloric stenosis and myasthenia gravis. Concomitant treatment with oral forms of potassium chloride and other anticholinergics is also contraindicated (SPC: glycopyrronium).

The SPC reports that adverse effects are common with glycopyrronium bromide due to its anticholinergic effects. The SPC states that the most common anticholinergic adverse effects in the RCTs were dry mouth, constipation, diarrhoea and vomiting, all occurring at a rate of 15% or higher. Other common anticholinergic symptoms include urinary retention, flushing and nasal congestion.

The SPC advises that anticholinergic effects may be dose dependent and difficult to assess in a child with disabilities. Monitoring by clinicians and carers is required, and the parent or carer should stop treatment and seek advice from the prescriber in the event of:

After evaluation of the adverse effect, a decision should be made about whether glycopyrronium bromide should be discontinued or restarted at a lower dose.

The EPAR notes that central nervous system effects have been reported with glycopyrronium bromide. The clinical significance of this in children with neurological disorders is uncertain. In addition, the EPAR states that there is no information on neurodevelopment or growth in children, which may be affected, particularly when treatment is taken long-term or in repeated episodes.

The EPAR states that it is uncertain whether the plasma concentration of glycopyrronium bromide in children and young people with neurological conditions would have a clinically relevant effect on blood pressure or heart rate. This means there is uncertainty about the cardiovascular safety of glycopyrronium bromide in this population. The SPC advises caution in children with acute myocardial infarction, hypertension, coronary artery disease, cardiac arrhythmias and conditions characterised by tachycardia. The parent or carer should be advised to measure the pulse rate if the child seems unwell and report a very fast or very slow heart rate.

The EPAR also notes that pneumonia appears to be associated with glycopyrronium bromide use, although there is insufficient information on the severity of this adverse effect. Glycopyrronium bromide can dry bronchial secretions and cause the formation of thick mucus, especially when given at inappropriate doses, which may increase the risk of a person developing pneumonia.

The SPC notes that reduced salivation can increase the risk of oral cavities and periodontal diseases, and it is important that people treated with glycopyrronium bromide receive adequate daily dental hygiene and regular dental health checks.

In Mier et al. 2000 adverse effects were reported for 25 out of 36 participants (69%) while taking glycopyrronium bromide compared with 5 out of 30 participants (17%) while taking placebo (no statistical analysis reported). Four of the 7 participants who dropped out while taking glycopyrronium bromide did so before the end of the first week while on the lowest dose. More participants experienced adverse effects as the dose increased, with only 14% of participants reporting adverse events at the lowest dose, compared with 81% at the highest dose. No hospital admissions or deaths were reported.

In Zeller et al. (2012a), adverse effects occurring during treatment were reported in all (20/20; 100%) participants receiving glycopyrronium bromide compared with 15 out of 18 (83.3%) receiving placebo (no statistical analysis reported). Four people (20%) in the glycopyrronium bromide group had at least 1 severe adverse effect occurring during treatment compared with none in the placebo group.

The SPC notes that published safety data are not available beyond 24 weeks treatment duration. Given the limited long-term safety data and the uncertainties around the potential risk of carcinogenicity, the SPC advises that the total treatment duration should be kept as short as possible.

An overview of the results for safety and tolerability can be found in results tables.

Evidence strengths and limitations

Two double-blind, placebo‑controlled RCTs were identified as relevant, evaluating the efficacy of oral glycopyrronium bromide for treating severe sialorrhoea in children and young people with a neurological condition (Mier et al. 2000 and Zeller et al. 2012a). Neither of the RCTs described allocation concealment. In addition, no description of the randomisation method was provided in Mier et al. (2000).

Zeller et al. (2012a) used a modified intention to treat (ITT) analysis, defined as all randomised participants who were within the age range of the final amended protocol and received at least 1 dose of study treatment, with lowest rank observations carried forward for any participants who dropped out of the study.

In Mier et al. (2000), efficacy analyses were not based on the ITT population, but on the population of participants who completed the study. The authors do not describe this population in detail and no account is given for the participants who dropped out. By not including participants who dropped out of the study in the efficacy analysis, the investigators may have introduced bias. For example, if participants had dropped out because of lack of effectiveness, excluding these people from the analysis could result in an overestimate of the effectiveness of the treatment. The EPAR reports that the company presented an analysis of all 39 randomised participants, using a baseline observation carried forward approach for the 12 people who did not complete the study. This analysis reduced the treatment effect from an improvement of 4.56 points in mTDS score to 3.06 points, but it is not clear whether the difference between groups remained statistically significant.

Furthermore, the EPAR states that the statistical tests used in Mier et al. (2000) were sub-optimal for a cross-over study, noting that the standard test would be analysis of variance (ANOVA).

Both trials used the efficacy outcome measure of mTDS, a 9‑point drooling rating scale that was reported by the person, parent or carer as a measure of the severity and frequency of drooling. No objective measure was used to quantify the amount of drooling in any of the trials. The subjective nature of this outcome evaluation is a limitation of all the studies.

The 2 RCTs were short-term, with a treatment duration of 8 weeks. They do not provide evidence for the safety and efficacy of long‑term use of glycopyrronium bromide for managing severe sialorrhoea in children and young people. In a longer open-label non-randomised study (Zeller et al. 2012b) 52.3% (95% CI 43.7% to 60.9%) of participants responded to treatment (improvement of ≥3 points in mTDS score) at 24 weeks. However, as this study did not include a placebo group, firm conclusions on the long-term effectiveness of glycopyrronium bromide cannot be drawn.

In addition, both RCTs were small, including 40 or fewer participants, and neither study reports whether a power calculation was carried out to support the small sample size.

There was also variation in the severity of the different neurological conditions among participants. Most of the participants in the studies had cerebral palsy so it is difficult to assess the effectiveness or safety of glycopyrronium bromide in children and young people with a neurological condition other than cerebral palsy. However, cerebral palsy is the most common cause of physical disability in children and young people in the developed world.

Glycopyrronium bromide has not been compared to other active treatments for severe sialorrhoea in a published study. However, the EPAR advises that the choice of comparator was supported since no other medicine for severe sialorrhoea has been extensively authorised in the European Union.

In both RCTs previous use of glycopyrronium bromide was permitted, which may have introduced bias. However, the EPAR notes that in Zeller et al. 2012a the number of participants who had previously taken glycopyrronium bromide was similar between treatment groups, and in Mier et al. (2000) only 5 people had been treated previously. In Mier et al. (2000), most parents indicated that they knew when their child was receiving glycopyrronium bromide because of the improvement in drooling. This could have biased both the efficacy results and the reporting of adverse effects. In Zeller et al. (2012a), because children and young people receiving placebo would be expected to continue drooling chronically, parents and carers were specifically encouraged to keep them in the study until at least the end of the 4‑week titration period.

An overview of the quality assessment of each included study can be found in evidence tables.