Trazodone has been widely prescribed for off-label use as a sleep aid. Identifying how trazodone impacts the performance of polysomnographic sleep architecture in insomnia disorder will provide additional data that can be used to guide clinical application. To assess the efficacy of trazodone in altering the polysomnographic sleep architecture in insomnia disorder so that sleep can be facilitated. PubMed, EMBASE, Web of Science, PsycINFO, Cochrane Library, Chinese Biomedical Literature Database (SinoMed), China National Knowledge Infrastructure, Wanfang Database, and the China Science and Technology Journal Database were searched for articles published between inception and June 2022. RCTs in patients with insomnia disorder applying trazodone in one arm of interventions at least 1 week, and reporting PSG parameters in the outcomes were eligible. RoB 2 was used to evaluate the risk of bias. The results of quality of evidence assessed by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. When I2 < 50%, the fixed effects model was used. When I2 ≥ 50%, the random effects model was used. The mean differences (MD) or standardized mean differences (SMD) and odds ratios (OR) with 95% confidence intervals (CIs) were estimated. Eleven randomized controlled trials were selected and participants were 466. Risk of bias was low in 5 trials (45.5%), and was moderate in 6 (54.5%). Compared with the control group, trazodone significantly increased total sleep time (TST, min) (MD = 39.88, 95% CI 14.44–65.32, P = 0.002) and non-rapid eye movement stage 3 (N3, mixed min and %) (SMD = 1.61, 95% CI 0.69–2.53, P = 0.0006); trazodone significantly decreased latency to onset of persistent sleep (LPS, min) (MD = − 19.30, 95% CI − 37.28 to − 1.32, P = 0.04), non-rapid eye movement stage 1 (N1, mixed min and %) (SMD = − 0.62, 95% CI − 1.13 to − 0.12, P = 0.02), the number of awakenings (NAs, including both arousal times and arousal index) (SMD = − 0.67, 95% CI − 0.91 to − 0.42, P < 0.00001), and waking time after persistent sleep onset (WASO, mixed min and %) (SMD = − 0.42, 95% CI − 0.81, − 0.03, P = 0.04), with no obvious effect on non-rapid eye movement stage 2 (N2, mixed min and %) (SMD = − 0.15, 95% CI − 0.41 to 0.11, P = 0.25), rapid eye movement (REM, mixed min and %) (SMD = 0.22, 95% CI − 0.26 to 0.70, P = 0.37), rapid eye movement latency (REML, min) (MD = 2.33, 95% CI − 27.56 to 32.22, P = 0.88), or apnea–hypopnea index (AHI) (MD = − 4.21, 95% CI − 14.02 to 5.59, P = 0.40). Daytime drowsiness (OR = 2.53, 95% CI 1.14–5.64, P = 0.02) and decreased appetite (OR = 2.81, 95% CI 1.14–6.92, P = 0.02) occurred with greater frequency in the trazodone group as compared to the control group, and the differences were significant. The results of quality of evidence were very low in TST, N3 and AHI, were low in LPS, WASO and REM, and were moderate in N1 and NAs. The sources of heterogeneity in TST and N3 were not found out from sensitive and subgroup analysis and there was no high quality of evidence in outcomes by GRADE Assessment. Trials with combination of other therapy could be a problem in this meta-analysis as the possibility of interactions were found from sungroup analysis. Trazodone could improve sleep by changing the sleep architecture in insomnia disorder, but it should be used with caution due to the adverse events that may occur.PROSPERO registration register name: The effect of trazodone on polysomnography sleep architecture in patients with insomnia: a systematic review and meta-analysis protocol; Registration Number CRD42020215332.