Doxepin and cognitive behavioural therapy for insomnia in patients with Parkinson's disease -- a randomized study

Although a variety of pharmacologic and non-pharmacologic treatments are effective for insomnia in the general population, insomnia in Parkinson's disease differs in important ways and may need different treatments. No studies have conclusively demonstrated effective insomnia treatments in Parkinson's disease.

We conducted a three-arm six-week randomized pilot study assessing non-pharmacologic treatment (cognitive behavioural therapy with bright light therapy) or doxepin (10 mg daily), compared to an inactive placebo in Parkinson's patients with insomnia. Sleep outcomes included insomnia scales, clinical global impression, sleep diaries and actigraphy. Secondary outcomes included motor severity, fatigue, depression and quality of life.

18 patients were randomized, 6 to each group. Compared to placebo, doxepin improved the Insomnia Severity Index (-9 ± 5.4 vs. -2 ± 3.9, p = 0.03), the SCOPA-night score (-5.2 ± 1.5 vs. -2.3 ± 2.8, p = 0.049), the Pittsburgh Sleep Quality Index-sleep disturbances subscale (-0.5 ± 0.5 vs 0.2 ± 0.4, p = 0.02), and both patient and examiner-rated clinical global impression of change (1.7 ± 0.8 vs. 0.5 ± 0.8, p = 0.03 and 1.4 ± 0.5 vs. 0.3 ± 0.5, p = 0.003). On secondary outcomes doxepin reduced the fatigue severity scale (p = 0.02) and improved scores on the Montreal Cognitive Assessment (p = 0.007). Non-pharmacological treatment reduced the Insomnia Severity Index (-7.8 ± 3.8 vs. -2.0 ± 3.9, p = 0.03), and the examiner-reported clinical global impression of change (p = 0.006), but was associated with decline in Parkinson Disease Questionnaire-39. There were no changes in other primary and secondary outcomes, including actigraphy outcomes. Adverse events were comparable in all groups.

Doxepin and non-pharmacologic treatment substantially improved insomnia in Parkinson's disease. These potential benefits must be replicated in a full confirmatory randomized controlled trial.