A comprehensive systematic review and meta-analysis of randomized controlled trials has found that sleep-promoting and circadian interventions are associated with a reduced risk of postoperative delirium among adult surgical patients. Conducted by researchers at Changde Hospital of Xiangya School of Medicine at Central South University, the investigation evaluated clinical trials where targeted non-pharmacological and pharmacological sleep strategies were deployed to mitigate acute cognitive complications following surgical procedures. The synthesis, encompassing twenty-nine studies, indicated an overall thirty-eight percent reduction in the relative risk of postoperative delirium across pooled dichotomous trials. However, the study authors emphasize that the underlying evidence base exhibits substantial clinical and methodological heterogeneity, with the majority of certainty ratings judged as low or very low under the GRADE framework.

Physiological disruptions induced by major surgical trauma, acute pain, noisy clinical environments, nocturnal nursing assessments, unfamiliar ward routines, and anesthesia frequently compromise sleep architecture and uncouple endogenous circadian rhythms. This fragmentation of sleep and circadian pacemakers is thought to impair central nervous system recovery, compromise synaptic homeostasis, aggravate systemic neuroinflammatory signaling, and diminish the cognitive reserve necessary to resist acute postoperative mental confusion.

Previous research in perioperative medicine frequently struggled to evaluate sleep-specific benefits because published reviews routinely pooled sedative drugs, melatonergic agents, and expansive multicomponent delirium-prevention bundles. As the study team pointed out, commonly evaluated agents such as dexmedetomidine are sedative alpha-2 adrenergic agonists that lower delirium incidence but cannot be mechanically equated with the restoration of natural, physiological sleep or circadian synchronization. Furthermore, hospital-wide delirium care bundles often implement early physical mobilization, frequent reorientation, nutritional optimization, sensory aids like glasses and hearing aids, and pain protocols simultaneously, which obscures the specific preventive contribution made by sleep targeting alone. The newly published systematic review was intentionally designed to isolate strategies primarily formulated to optimize nighttime sensory environments, reinforce circadian entrainment, or support natural sleep physiology without primary reliance on general anesthetics or sedatives.

To assemble the evidence base, the review team executed comprehensive literature searches across seven major electronic databases through July 31, 2026, comprising PubMed/MEDLINE, Embase, Cochrane CENTRAL, Web of Science, Scopus, CINAHL, and PsycINFO without language restrictions. The investigators applied rigorous inclusion criteria specifying randomized controlled trials and cluster-randomized trials that enrolled adult surgical patients undergoing elective or emergency surgery under general, regional, or combined anesthesia across post-anesthesia care units, general wards, and surgical intensive care environments. Trials evaluating primary anesthetic, sedative, antipsychotic, ketamine, or opioid-centered protocols were systematically excluded, as were therapies specifically managing sleep-disordered breathing like positive airway pressure. From an initial identification of 1,049 database records, the authors screened fifty full-text reports, ultimately identifying twenty-nine eligible study reports, with twenty-seven trials contributing extractable dichotomous delirium event data across 2,788 surgical participants.

Across the complete pooled dataset of twenty-seven dichotomous trials, the meta-analysis documented a statistically significant association between sleep-promoting interventions and a reduced risk of postoperative delirium, yielding a pooled risk ratio of 0.62 with a 95% confidence interval ranging from 0.49 to 0.78, accompanied by moderate heterogeneity represented by an I-squared value of 41.8%. To evaluate the influence of diagnostic quality, the investigators prespecified a Core analysis restricted to eighteen trials involving 2,189 patients that ascertained delirium through validated diagnostic criteria or accepted clinical assessment tools, such as the Confusion Assessment Method, CAM for the Intensive Care Unit, 3D-CAM, 4AT, Nursing Delirium Screening Scale, Delirium Observation Screening Scale, or DSM criteria. In this Core cohort, the association was noticeably attenuated but remained statistically significant, demonstrating a pooled risk ratio of 0.70 with a 95% confidence interval between 0.52 and 0.93 and an I-squared of 49.7%.

The synthesized trials represented five primary therapeutic categories: melatonergic agents, bright light and circadian protocols, sensory-environment strategies, multicomponent or behavioral sleep interventions, and dual orexin receptor antagonists. Melatonergic agents formed the largest clinical subset, comprising seventeen randomized trials and 2,007 participants evaluating melatonin or ramelteon. Although melatonergic interventions demonstrated a significant delirium reduction across the broad analysis, yielding a risk ratio of 0.66 with a 95% confidence interval of 0.50 to 0.88, the association was not robust when subjected to stricter criteria. When limited to the twelve Core melatonergic trials using validated diagnostic tools, the confidence interval crossed the null effect line at a risk ratio of 0.76 with an interval of 0.54 to 1.08. Furthermore, after excluding trials that featured balanced background infusions of dexmedetomidine in both intervention and control arms, the risk ratio attenuated to 0.83 with an interval of 0.57 to 1.22 across ten trials.

In contrast to the pharmacological findings, non-pharmacological interventions evaluated in Core studies demonstrated a more stable association with reduced delirium incidence. When five Core trials comprising 521 patients receiving non-pharmacological protocols were synthesized, the analysis yielded a pooled risk ratio of 0.44 with a 95% confidence interval of 0.21 to 0.93 and low heterogeneity at an I-squared of 14.7%. Interventions focusing on environmental sensory modification, including nocturnal use of earplugs and eye masks, were evaluated in two trials totaling 141 participants and showed a wide, statistically non-significant estimate with a risk ratio of 0.54 and an interval spanning 0.15 to 1.98. Multicomponent or behavioral interventions, encompassing strategies such as music interventions, clustered nighttime nursing care, and brief cognitive behavioral sessions, generated a pooled risk ratio of 0.45 across four trials with a confidence interval between 0.18 and 1.12.

Light-based circadian interventions yielded the most pronounced numerical reduction in delirium risk across three trials involving 94 surgical patients, exhibiting a pooled risk ratio of 0.24 with a 95% confidence interval of 0.09 to 0.64 and zero observed heterogeneity. However, the authors urge substantial caution regarding this finding, noting that the estimate relies entirely on a very small total patient sample subjected to small-subgroup inference limitations. When an exploratory modified Hartung-Knapp variance-safeguard model was applied to this small subgroup, the 95% confidence interval widened dramatically from 0.03 to 2.21, crossing the null threshold and losing statistical significance.

Methodological quality assessments conducted using the Cochrane Risk of Bias 2 tool revealed considerable vulnerability across the included trials, with ten studies judged at low risk of bias, ten raising some concerns, and seven categorized as high risk of bias. High-risk determinations were primarily driven by non-standardized outcome assessment methods, such as reliance on passive chart reviews, general nursing documentation, or abbreviated mental test proxies rather than structured, validated delirium diagnostic instruments. Studies utilizing broad, unvalidated delirium ascertainment reported substantially exaggerated effect sizes, pooling a risk ratio of 0.42 compared to 0.70 in Core studies, a difference that reached statistical significance with a p-value of 0.026. Furthermore, 95% prediction intervals for both the overall analysis, spanning 0.31 to 1.22, and the Core analysis, spanning 0.35 to 1.40, extended beyond the null value of 1.0, indicating that future clinical trials could easily observe neutral or non-beneficial outcomes.

The review authors also conducted an exploratory analysis assessing trials according to whether objective or validated subjective sleep improvements were documented during the postoperative stay. Across eight trials involving 706 patients where sleep quality was demonstrably improved, the pooled risk ratio for delirium was 0.44 with an interval of 0.31 to 0.63, whereas nineteen trials without documented sleep improvements exhibited a risk ratio of 0.69 with an interval of 0.53 to 0.90. While the between-group difference was statistically significant with a p-value of 0.019, the researchers cautioned that this trial-level finding represents exploratory effect modification rather than confirmed mechanistic mediation. Secondary clinical endpoints, including delirium duration, delirium severity, intensive care unit length of stay, total hospital stay, and opioid consumption, could not be quantitatively synthesized due to widespread reporting disparities across studies. The authors concluded that while circadian entrainment and environmental sleep protection remain promising avenues of research, the current evidence does not support routine clinical practice changes, underscoring the urgent need for large-scale, standardized randomized trials.

Subject of Research: Sleep-promoting and circadian interventions for preventing postoperative delirium in surgical patients

Article Title: Sleep-promoting and circadian interventions for prevention of postoperative delirium in adult surgical patients: a systematic review and meta-analysis

Article References: Li, Q., Tang, M., Gao, Y.-X., Nie, Z.-J., Mao, X., & Fei, M. (2026). Sleep-promoting and circadian interventions for prevention of postoperative delirium in adult surgical patients: a systematic review and meta-analysis. Frontiers in Psychiatry, 17, Article 1969855. https://doi.org/10.3389/fpsyt.2026.1969855

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1969855

Keywords: postoperative delirium, circadian interventions, sleep promotion, melatonin, ramelteon, suvorexant, bright light therapy, sensory environment, earplugs and eye masks, systematic review, randomized controlled trials, meta-analysis, CAM-ICU, surgical intensive care, dexmedetomidine, sleep architecture, hospital recovery