CLINICAL QUESTION

Is mirtazapine or amitriptyline effective for adults with insomnia?

BOTTOM LINE

In the small, pragmatic study, mirtazapine and amitripty-line significantly improved sleep quality but only for the first 6 weeks. Mirtazapine resulted in recovery from insomnia but only for the first 6 weeks. There were no meaningful improvements in daytime fatigue or other measures of sleep impact. (Level of Evidence = 1b)

SYNOPSIS

The trial took place in primary care practices in the Netherlands. The researchers recruited adults aged 18 to 85 years with insomnia who had early morning awakenings or difficulty maintaining sleep and were seeking medication because the patient or their primary care clinician deemed nonpharmacologic treatment insufficient. The authors randomized participants to 16 weeks of one or two identical-appearing tablets of mirtazapine (7.5 mg; n = 27), amitriptyline (10 mg; n = 26), or placebo (n = 27). Initially, patients took one tablet; the clinician or the patient could double the dose at preplanned visits until week 14. At multiple times over the 16 weeks of the study, the researchers evaluated each participant's Insomnia Severity Index (ISI) score and other sleep parameters, including daytime fatigue. At the end of 16 weeks, six, nine, and three participants discontinued the intervention, respectively, primarily due to adverse effects and ineffectiveness. In addition to the mean change in scores, the authors also reported the proportion of adults who improved (at least a 7-point improvement on the ISI) or recovered (an ISI score of 10 or less). In the intention-to-treat analysis, the only period when the participants experienced meaningful improvement or recovery was at 6 weeks. At 6 weeks, more patients improved with mirtazapine (52%) or amitriptyline (40%) than placebo (13.6%; numbers needed to treat = 3 and 4, respectively). However, only participants who took mirtazapine had more recoveries that were statistically significant compared with placebo (56% vs 14.3%; NNT = 3) at 6 weeks. No other outcomes were significantly improved during the follow-up intervals. One participant experienced two serious events that resulted in hospitalization, but the authors report that these were unrelated to the study intervention.