BACKGROUND:Medications, including dual orexin receptor antagonists (DORAs), benzodiazepines (BZDs), Z-drugs and melatonin receptor agonists, are common medications for insomnia disorder, but holistic comparisons of their efficacy and safety are not quite clear.
OBJECTIVE:To investigate the efficacy and safety profiles of agents for treating insomnia disorder and further establish a clinical algorithm in terms of type of insomnia based on the "time window" generated from adjusting for certain confounding factors.
METHODS:Relevant randomized controlled trials (RCTs) were retrieved from PubMed, Embase, Scopus, the Cochrane Library, Web of Science, and ClinicalTrials.gov from inception to April 15, 2025. The standard mean difference (SMD) was generated for consecutive variants, including the wake after sleep onset(WASO), latency to persistent sleep(LPS), total sleep time(TST), and sleep efficiency(SE), for pairwise comparisons via Bayesian network meta-regression (NMR) analyses adjusted for the follow-up period and age by RStudio 4.4.2. Pharmacovigilance (PV) was investigated by leveraging the FAERS database, and odds ratios (ORs) were generated for dichotomous and ordinal variants for pairwise comparisons via STATA 18.0 MP.
RESULT:A total of 15 studies evaluating 10 treatment regimens involving 2408 patients were included in the final analysis. For the WASO, compared with the placebo group, dimdazenil 2.5 mg/d (SMD = -0.388, 95 % CI: -0.608 to -0.166), lemborexant 10 mg/d (SMD = -0.624, 95 % CI: -0.894 to -0.355), lemborexant 5 mg/d (SMD = -0.612, 95 % CI: -0.88 to -0.342), daridorexant 25 mg/d (SMD = -0.957, 95 % CI: -1.436 to -0.479), melatonin 6 mg/d (SMD = -0.741, 95 % CI: -1.423 to -0.044), zolpidem 10 mg/d (SMD = -0.348, 95 % CI: -0.61 to -0.068), doxepin 3 mg/d (SMD = -0.497, 95 % CI: -0.713 to -0.282) were significantly superior. Among the effective regimens, lemborexant 10 mg/d, lemborexant 5 mg/d, daridorexant 25 mg/d, melatonin 6 mg/d, and doxepin 3 mg/d were comparable. However, dimdazenil 2.5 mg/d (SMD = 0.568, 95 % CI: 0.046 to 1.096) and zolpidem 10 mg/d (SMD = 0.608, 95 % CI: 0.074 to 1.171) were significantly inferior to daridorexant 25 mg/d.After adjusting for the follow-up period, the results resembled those of the raw analysis except for the loss of significant superiority of melatonin 6 mg/d (SMD = -0.727, 95 % CI: -1.48 to 0.01) over the placebo. However, melatonin 6 mg/d demonstrated a "time window" of significant superiority over the control group from the 10th to 40th weeks. After adjusting for age, dimdazenil 2.5 mg/d (SMD = -0.355, 95 % CI: -0.652 to -0.1), lemborexant 10 mg/d (SMD = -0.508, 95 % CI: -0.9 to -0.114), zolpidem 10 mg/d (SMD = -0.526, 95 % CI: -0.84 to -0.158) demonstrated significant superiority over placebo. In terms of safety, with respect to nervous system disorders, safety signals were detected in suvorexant (IC025 = 0.212; 95 % CI: 1.214 to 1.334), lemborexant (IC025 = 0.221; 95 % CI: 1.236 to 1.567), daridorexant (IC025 = 0.205; 95 % CI: 1.21 to 1.427) and doxepin (IC025 = 0.066; 95 % CI: 1.091 to 1.411). In terms of dyspnoea, eszopiclone (OR ranging from 0.556 to 0.669) had significantly lower constituent ratios than daridorexant, melatonin and zolpidem did. Melatonin (OR = 1.568, 95 % CI = 1.192 to 2.061, p = 0.001) and zolpidem (OR = 1.302, 95 % CI = 1.026 to 1.653, p = 0.03) had a significantly higher constituent ratio than suvorexant. The proportion of patients with severe dyspnoea caused by daridorexant (OR = 0.256, 95 % CI = 0.096 to 0.678, p = 0.006) was significantly lower than that caused by suvorexant and lemborexant. For adverse reaction outcomes, zaleplon (OR = 9.888, 95 % CI = 1.124 to 86.944, p = 0.039) had a significantly higher effect than daridorexant on severe dyspnoea.
CONCLUSION:Comprehensively considering the efficacy effect size, time windows (follow-up period, age, and types of insomnia), PV, severity of imperative adverse events, we propose prioritizing the use of daridorexant 25 mg/d for insomnia characterized by difficulty maintaining sleep and insufficient sleep duration. For insomnia characterized by difficulty falling asleep, we recommend prioritizing the use of lemborexant 10 mg/day or zolpidem 10 mg/day. For overall poor sleep efficiency, we recommend using lemborexant. Drug selection should be based on the types of insomnia and drug safety. More head-to-head clinical trials are needed to confirm those findings.