Nine RCTs of donepezil (n=3,250) and two RCTs of rivastigmine (n=1,424) were included. Three open-label extension studies of donepezil (n=976 patients) and one-open label extension study of rivastigmine (463 patients) were also included.
Donepezil.
ADAS-cog as the primary outcome (6 RCTs): the studies showed that donepezil 3 mg (1 RCT), 5 mg (6 RCTs) and 10 mg (3 RCTs) significantly reduced ADAS-cog scores, indicating a significant benefit compared with placebo.
Overall clinical status using the CGIC (2 RCTs) or the CIBIC-plus (4 RCTs): the studies showed that donepezil statistically significantly increased treatment success rates compared with placebo; the success rates ranged from 21 to 89% with donepezil 5 to 10 mg, and from 14 to 80% with placebo. There was no statistically significant difference in effect according to the dose of donepezil.
The one RCT in 290 patients with moderate to severe disease (included in the CIBIC-plus results above) found that patients treated with donepezil had significantly higher mean CIBIC-plus scores compared with patients treated with placebo (P<0.001).
One RCT (431 patients) showed that donepezil increased the median time till functional decline (375 days versus 208 days). Another RCT (286 patients) showed that donepezil improved the mean decline in GBS score over 52 weeks compared with placebo; the change from baseline was 7.3 (95% CI: 3.2, 11.4) with donepezil versus 13.5 (95% CI: 9.4, 17.6) with placebo.
The authors stated that adverse effects were few and generally of mild to moderate severity. The highest rate of adverse effects was 83% with donepezil compared with 80% with placebo in one RCT. The adverse effects included diarrhoea, headache and respiratory tract infection.
Open-label studies providing information on medium-term efficacy: one study (patients from 2 RCTs) found that for patients in one of the RCTs, donepezil (5 to 10 mg) improved CRD-SB scores until week 24 of the open-label study and then the scores declined.
Rivastigmine.
ADAS-cog (2 RCTs, intention-to-treat results reported): high-dose (6 to 12 mg) rivastigmine reduced cognitive decline compared with placebo (the differences compared with placebo were 3.78 and 1.60). The results for low-dose (1 to 4 mg) rivastigmine differed between RCTs: one RCT (233 patients in the rivastigmine group) found low-dose rivastigmine significantly reduced cognitive decline at 7 weeks compared with placebo (difference 1.73, P<0.05), while the other (243 patients in the rivastigmine group) found no significant difference at 12 weeks (difference 0.03). Patients in the placebo group of the RCT showing no difference between treatments did not deteriorate immediately from baseline, whereas patients in the other RCT did.
The RCTs found similar, apparently dose-related rates of withdrawal: 15 and 14% with low-dose rivastigmine, compared with 35 and 33% with high-dose rivastigmine and 16 and 13% with placebo. Adverse effects were generally not severe. The most common adverse effects were nausea, vomiting, diarrhoea, abdominal pain and anorexia. Dizziness, headache, fatigue and malaise tended to occur only at higher drug doses.