Results
We identified 59 patients that were suitable for inclusion in our study. Their baseline demographics and characteristics are summarised in Table 1. Analysis of a resting 12-lead ECG prior to commencing donepezil treatment was compared to the ECG obtained during their most recent acute hospital admission and demonstrated evidence of significant prolongation of the PR, QRS and QT intervals (Table 2). The QT corrected using Bazett, Fredericia, Framingham and Hodges formulae were also compared, and all confirmed significant QT prolongation during donepezil therapy. Analysis using the QT nomogram revealed that 8 (13.6%) patients (3 male and 5 female) went from low arrhythmogenic risk to high while on donepezil therapy (Figure 1). No significant change in mean HR was noted; although one patient had donepezil ceased on admission due to bradycardia and intermittent complete heart block. Cessation of donepezil resulted in normalisation of this patient’s heart rate and resolution of their heart block.
Assessment of the effect of donepezil on ECG parameters demonstrated that males experienced a significantly greater increase in their corrected QT intervals compared to females (Table 3). There were no significant differences in all ECG parameters when comparing donepezil doses (Table 4). Although there was a significant difference in QT interval between the different treatment duration groups, no significant difference was found once QT intervals were corrected. Duration of donepezil therapy did not affect the HR, PR interval or QRS interval (Table 5).
Point-biserial correlation found that concomitant use of rate-limiting calcium channel blockers was associated with significant PR prolongation (r pb = 0.314, P=0.030), while beta-blockers in combination with donepezil were found to significantly reduce the HR (r pb =0.256, P=0.050). All corrected QT intervals whilst on donepezil were significantly prolonged by the use of tricyclic antidepressants (QTcB: r pb = 0.344, P=0.008, QTcFred: r pb = 0.382, P=0.003, QTcFram:r pb = 0.379, P=0.003, QTcH:r pb = 0.352, P=0.006). There was no significant correlation between cardiac comorbidities and changes in ECG parameters.