Summary of studies assessing the effects of low-dose doxepine (DXP) on sleep
| Study | Age | Diagnosis | Design + Number of Participants | Results | Adverse Events | Conclusion |
|---|---|---|---|---|---|---|
| Krystal et al. (2010) | 65+ | Primary insomnia by DSM-IV | 12-wk RCT. DXP 1 mg/day vs. 3 mg/day vs. placebo. N = 240. | LPS was not significantly different from placebo for any dose of DXP. However, WASO was significantly reduced at all time points in the DXP 3 mg/day group (P < 0.001) and on night 1 (P < 0.01) and night 85 (P < 0.05), but not night 29 in the DXP 1 mg/day group. TST was increased significantly in the DXP 3 mg/day group at all time points and in the DXP 1 mg/day group at night 1 (P < 0.05) and night 85 (P < 0.05) but not night 29. | Rates of treatment-emergent AEs were lower in the DXP groups than the placebo group: 52% of placebo subjects reported an AE compared with DXP 1 mg/day (40%) and DXP 3 mg/day (38%). Less discontinuation with DXP as well: placebo discontinuation was 14%, DXP 1 mg/day was 9%, DXP 3 mg/day was 10%. No evidence of next-day sedation. | Low-dose DXP is effective at reducing symptoms of insomnia. |
| Krystal et al. (2011) | 18–64 | Primary insomnia by DSM-IV-TR | 35-day RCT followed by two nights of single-blind placebo treatment to assess discontinuation effects. DXP 3 mg/day vs. 6 mg/day vs. placebo. N = 221. | By polysomnography: LPS was only improved relative to placebo in both DXP 3 and 6 mg/day on night 1, P < 0.0001, and not improved on night 15 or 29. TST was significantly improved for both doses and for all time points except night 15 in the 3 mg/day group. WASO was significantly improved at all time points and in both dosage groups. | Rates of discontinuation were similar between treatment arms: placebo = 12%, DXP 3 mg/day = 12%, DXP 6 mg/day = 11%. No clinically meaningful changes in mean laboratory values, ECGs, body weight, or vital signs. | Low-dose DXP is effective at reducing symptoms of insomnia. DXP seems particularly effective at maintaining sleep. |
| Lankford et al. (2012) | 65+ | Primary insomnia by DSM-IV-TR | 4-wk RCT. DXP 6 mg/day vs. placebo. N = 255. | Subjective TST (minute) increased from 283.1 ± 50.0 to 346.1 ± 66.4 at week 4 in the DXP group vs. an increase from 293.5 ± 49.1 to 336.4 ± 64.7 at week 4 in the placebo group, P < 0.01 (DXP separated from placebo on this measure by week 1). Subjective WASO was reduced relative to placebo at week 1 (P < 0.0001) and week 4 (P < 0.01). | Rates of AEs were similar between groups: 27% for placebo and 31% for DXP 6 mg/day. Rates of discontinuation were lower for DXP: placebo = 10% vs. DXP 6 mg/day = 5%. No changes in laboratory values, ECGs, body weight, neurologic examinations, or vital signs. | Low-dose DXP is effective at reducing symptoms of insomnia. |
| NCT#00755495 | see Ramelteon section | |||||
| Roth et al. (2007) | 18–64 | Chronic primary insomnia by DSM-IV | Four-group crossover RCT with 2 polysomnographic assessment nights per treatment. DXP 1 mg/day vs. 3 mg/day vs. 6 mg/day vs. placebo. N = 67. | For the measurement of LPS (polysomnography), none of the treatments were statistically better than placebo by one analysis; a statistical re-analysis post hoc found P = 0.0139 for DXP 6 mg. All three doses improved total sleep time relative to placebo: P = 0.0005 for DXP 1 mg/day, P < 0.0001 for DXP 3 mg/day, P < 0.0001 for DXP 6 mg/day. | AEs in the DXP groups occurred at a similar rate of those in the placebo group and did not appear to be dose-related. Headache and somnolence were the most common AEs. | Low-dose DXP is effective in the treatment of insomnia. All doses of DXP significantly improved SE during the entire night, each one P < 0.05. Only DXP 3 mg/day and DXP 6 mg/day significantly improved wake time during sleep. |
| Roth et al. (2010) | 25–55 | Healthy participants required to have no history of insomnia and normal sleep patterns during last 3 mo. Transient insomnia was induced in participants. | One night RCT. DXP 6 mg/day vs. placebo. N = 565. | By polysomnography: LPS was 47.7 ± 59.5 min in the placebo group and 31.8 ± 44.0 min in the DXP group, P < 0.0001. TST was increased 51.1 min in the DXP group relative to the placebo group, P < 0.0001. | Incidence of AEs was similar to placebo. On one measure of next-day residual effects (the Digit Symbol Substitution Test) there was no difference between DXP and placebo. However, DXP resulted in statistically significant worsening on two other scales, the Symbol Copying Test and the Visual Analog Scale for sleepiness. | DXP is effective at reducing symptoms of insomnia. |
| Scharf et al. (2008b) | 65+ | Primary insomnia by DSM-IV | Four-period crossover RCT. DXP 1 vs. 3 vs. 6 mg/day vs. placebo. N = 76. | LPS was not significantly different from placebo for any dose of DXP. However, TST and SE were significantly increased relative to placebo for all doses, P < 0.0001. Wake time during sleep were significantly reduced at all doses, P < 0.0001. | Number of AEs was similar between placebo and DXP and were similar between DXP doses. | Low-dose DXP is effective at reducing symptoms of insomnia. It also increases SE. DXP seems particularly effective at maintaining sleep. |
| Yeung et al. (2015) | >18 | Varied | Systematic review and meta-analysis of nine placebo-controlled RCTs. | Pooled results were largely not reported due to heterogeneity; however, the authors were able to conclude the significant efficacy and safety of DXP 3–6 mg for one to two nights. | In young and middle-aged adults, the incidence of any AEs were 27%, 35%, and 32% in placebo, DXP 3 mg/day (RR = 1,27, 95% CI = 0.8, 2.1, P = 0.29), and DXP 6 mg/day (RR = 1.30, 95% CI 0.8, 2.1, P = 0.29). Common AEs were headache, somnolence, and nausea. Results in older adults were similar. | Low-dose DXP is effective at reducing symptoms of insomnia. |