Multidisciplinary approaches including nursing care to improve sleep in critically ill patients: a review

Introduction

The biological clock plays a key role in regulating circadian rhythms in both humans and animals. The discovery of clock genes and molecular mechanisms in Drosophila melanogaster – awarded the Nobel Prize in 2017 – laid the foundation for understanding the mammalian circadian system [1, 2]. In humans, the biological clock is located in the suprachiasmatic nuclei of the hypothalamus and is synchronized by external cues known as Zeitgebers, primarily light. Other Zeitgebers include fluctuations in temperature, food availability, and social activity. Circadian regulation influences not only sleep–wake cycles but also hormonal secretion, metabolism, and cardiovascular function [3, 4]. Sleep, which occupies nearly one-third of life, is essential for recovery, immune function, and nervous system regulation. It consists of REM and NREM phases, both crucial for cognitive and emotional processing. Sleep disorders – defined as reduced quantity or quality of sleep – can impair daily functioning, mood, and immunity. Long-term disturbances may cause disorientation and hallucinations, and increase vulnerability to infections. They also affect respiratory function, glucose regulation, pain sensitivity, and autonomic balance [5–13].
In intensive care unit (ICU) settings, sleep disorders are particularly common and result from a complex interplay of environmental, therapeutic, and patientrelated factors. ICU patients sleep on average less than two hours per day [9]. Key disruptive elements include environmental noise (ventilators, infusion pumps, staff conversations, other patients), constant lighting that disrupts the light–dark cycle, and ICU architecture that often lacks quiet zones or sound insulation [5, 7, 10–15]. Frequent nighttime procedures such as medication administration or nursing care also interfere with rest [5, 10]. Patient-specific, non-modifiable factors further complicate sleep. These include underlying health conditions (e.g. heart failure, chronic obstructive pulmonary disease – COPD), comorbid mental health disorders, pre-existing sleep disturbances, and physical discomfort. Pain, surgical wounds, catheters, and mechanical ventilation are frequently cited as major contributors to sleep disruption [4, 5, 8, 10, 15–17]. The ventilation mode itself affects sleep: pressure support ventilation (PSV) is associated with more fragmentation and asynchrony than assist-control ventilation (ACV), whereas proportional assist ventilation (PAV), which adjusts to patient effort, may improve sleep quality [7].
In addition to these clinical factors, appropriate assessment of sedation and sleep quality in ICU patients is critical for optimizing care. A patient’s pharmacological sedation can be assessed using the Richmond Agitation-Sedation Scale (RASS). It evaluates the patient’s level of sedation and agitation on a 10-point scale ranging from –5 to +4. A score of –5 indicates that the patient is unresponsive (unarousable), 0 corresponds to an alert and calm state, while a score of +4 means the patient is combative. Negative scores (–1 to –5) reflect increasing levels of sedation, and positive scores (+1 to +4) indicate escalating agitation. The use of the RASS allows healthcare teams to accurately monitor sedation depth and adjust treatment accordingly to ensure patient safety and comfort [18].
Pharmacologic agents commonly used in the ICU, including sedatives and opioids, also affect sleep architecture. Propofol and benzodiazepines reduce restorative REM and N3 phases, often increasing total sleep time but impairing its quality. Benzodiazepines are additionally linked to delirium [12, 19, 20]. Opioids may induce drowsiness but cause shallow, fragmented sleep and influence neurotransmission and melatonin pathways, potentially reversing circadian rhythms [20, 21]. Some nonsteroidal anti-inflammatory drugs (NSAIDs) may reduce melatonin levels and disturb sleep, while inotropic drugs affect cerebral perfusion. Conversely, dexmedetomidine has shown potential in promoting sleep resembling natural patterns by enhancing N2 and N3 stages [22–25].
Clinical experience confirms that ICU patients often suffer from fragmented, non-restorative sleep, with frequent awakenings and prolonged light sleep phases. Though total sleep time may appear adequate, up to half of it occurs during the day, and circadian disruptions often persist after ICU discharge [8, 10–13, 15, 16]. Given the multifactorial nature of sleep disturbances in critical care – spanning iatrogenic, environmental, and pharmacological factors – efforts are underway to mitigate their impact. These include both pharmacologic and non-pharmacologic strategies, such as dimming lights, reducing nighttime noise, and limiting bedside interventions.
The aim of this review was to analyze these factors and evaluate methods for improving sleep parameters in intensive care patients.

Material and methods

We analyzed the effectiveness of methods influencing the quality and/or length of sleep and optimizing circadian rhythm in patients of the Intensive Care Unit. We formulated the research question using the PICO(TS) strategy, based on the following assumptions:
• (P)opulation: adult (> 18 years of age) patients requiring intensive medical care or hospitalization in the ICU, regardless of the level of sedation;
• (I)ntervention: pharmacological and non-pharmacological interventions/actions improving the qualitative and quantitative parameters of sleep or reducing sleep deprivation associated with hospitalization and the use of medical procedures in the ICU;
• (C)omparison or control intervention: no intervention, comparison of two methods/interventions (e.g. pharmacological with non-pharmacological);
• (O)utcome: the main result will be an improvement in the quality and quantity of sleep and a reduction in the negative impact of medical procedures on the circadian rhythm of patients. Measurement tools: objective (polysomnography, electroencephalography – EEG, actigraphy), subjective (scales/standardized questionnaires assessing sleep quality).
Secondary outcomes may include other parameters indirectly affecting sleep, e.g. hormone levels in blood/urine;
• (T)ime: patient follow-up time – not specified (not limited). Publications from the period of 10 years – 2015–2024;
• (S)etting – inpatient healthcare facilities – intensive care units.

Inclusion criteria

We included full-text articles in Polish or English, found in the PubMed, EBSCO, Polish Medical Bibliography, Cochrane databases and directly from the websites of Polish medical publishers (Termedia, ViaMedica, other academic publishing houses) that met the adopted PICO criteria. We used a combination of the following key words to search the databases: “circadian rhythm” or “circadian rhythm disorders” or “circadian dysregulation” and “ICU” or “intensive care unit” – according to MeSH. We included articles from the last 10 years, i.e. from 2015-2024.

Exclusion criteria

Animal studies, review articles, case reports, articles available in the form of abstracts or conference reports without access to full texts. We reviewed the titles and abstracts of all articles from the search and made an initial selection of data that met the criteria. We then downloaded the full texts of the articles, and two authors independently analyzed and collected the data into a preliminary form. We resolved discrepancies by consensus, which was used to establish the final version of the results table.

Results

The initial search yielded 442 citations, which were reduced to 345 after removing duplicates. Following the screening of titles and abstracts, 31 articles were selected for full-text analysis. Most of the excluded studies focused on the incidence and risk factors of delirium in ICU patients without specifically assessing sleep parameters or sleep disorders. They evaluated the impact of ICU hospitalization on sleep, rather than the effects of interventions aimed at improving sleep or circadian rhythms. Ultimately, 12 studies published between 2014 and 2024 met the eligibility criteria and were included in the review [4, 14, 23, 25–33]. Just prior to manuscript submission, we updated the search and screened recent studies for relevance (Fig. 1).
Of the included studies, 8 were randomized controlled trials (RCTs) [14, 26, 27, 29, 31–33], and the others were observational studies [4, 23, 24, 30]. Eight studies assessed the impact of non-pharmacological interventions [4, 14, 26–31], while four evaluated pharmacological strategies [23, 24, 32, 33]. The total number of patients included across the studies was 577, with 443 undergoing non-pharmacological interventions and 134 receiving pharmacological treatments.
Non-pharmacological methods used to support sleep included exposure to light [12], the use of eye masks and/or earplugs [27–29, 31], listening to relaxing music [28], and environmental modifications aimed at minimizing sleep-disruptive stimuli [4, 14, 30]. Some studies implemented combined approaches, such as environmental changes together with oral melatonin administration [29]. Pharmacological interventions involved the intravenous administration of ramelteon [32] or dexmedetomidine [21, 23, 32, 33]. The effectiveness of these interventions was evaluated using various tools, including objective sleep measures such as polysomnography (PSG) [24, 27, 33], actigraphy [31–33], or EEG [29]. Subjective measures of sleep quality and daytime functioning were also used, obtained through psychological questionnaires completed by awake patients or clinical staff [4, 23, 28, 30, 31, 33]. In several studies, sleep assessments were supplemented by environmental analyses, such as measurements of ambient noise or light levels [14, 30].
One study [25] explored the feasibility of a sleep-promoting intervention for modifying sleep disturbances but did not assess sleep parameters or the presence of sleep-disrupting factors in ICU settings. While this study was not included in the final analysis, its large sample and promising results were considered worth citing. A summary of the findings from the included studies is presented in Table 1.

Discussion

In this analysis, no single intervention emerged as clearly superior in improving circadian rhythm and sleep disorders in ICU patients. The studies varied in methodology, and small sample sizes in several trials limited the strength of conclusions. However, complex, multicomponent interventions addressing multiple factors showed promising therapeutic and preventive effects [4, 27, 28, 30]. Most aimed either to eliminate sleep-disrupting influences or to enhance elements supporting natural circadian rhythms. Adverse effects on sleep in the ICU result from both environmental and clinical factors, including medication side effects, pain, and mechanical ventilation, which cannot be avoided [25, 34]. Environmental contributors include frequent disruptions related to hospital care – especially noise, lighting, and nighttime procedures [25, 34, 35]. Noise levels generated by ICU equipment, such as monitors, ventilators, and dialysis machines, often exceed the 55 dB limit recommended by the World Health Organization [34–38]. Studies have demonstrated a correlation between excessive noise and poor sleep outcomes [35].
Mechanical ventilation also plays a significant role in sleep disturbance. Though essential for respiratory support, it contributes to sleep fragmentation due to device noise, patient discomfort from endotracheal tubes, and frequent interventions. Intubation and continuous monitoring can increase cortisol levels, impairing restorative sleep. Light exposure in ICU settings rarely mirrors natural circadian patterns – daylight is insufficient, while nighttime lighting is excessive [4, 24, 35].
Simple interventions such as earplugs and eye masks have proven effective. When used together, they improved subjective measures of sleep quality, depth, and continuity [4, 30]. Polysomnography confirmed longer N3 sleep and fewer prolonged awakenings [27], although no improvements were found in overall clinical status or melatonin/cortisol levels [4, 27]. Some studies reported no objective improvement in sleep, but worsening sleep parameters in control groups suggested a protective effect [30]. A meta-analysis also found reduced delirium risk with the use of these tools [39, 40]. Despite limited strength of evidence, such measures – especially when integrated into a broader sleep hygiene approach – are recommended for conscious and cooperative ICU patients. Conversely, they may be contraindicated in agitated or claustrophobic individuals [35]. Since some sleep-disrupting factors in ICU settings are inherent to necessary therapies, procedures, and pharmacological treatments, monitoring sleep quality becomes crucial for timely and individualized interventions. Both subjective and objective tools are available to assess sleep disturbances in ICU patients. Among subjective measures, the Richards-Campbell Sleep Questionnaire (RCSQ) is the most widely used instrument. It enables patients capable of communication to self-report sleep depth, efficiency, latency, number of awakenings, and overall quality, providing quick and clinically useful feedback. Objective methods include actigraphy, which tracks patient movement to infer sleep patterns, and polysomnography (PSG), considered the gold standard for detailed assessment of sleep architecture. However, PSG remains impractical for routine use in the ICU due to its complexity and resource demands. Actigraphy offers a less invasive alternative but is limited by reduced accuracy in critically ill, immobilized patients. Regular use of validated sleep assessment tools can support clinical decision-making by identifying patients with significant sleep disturbances who may benefit from intensified nonpharmacologic strategies or, when necessary, adjunctive pharmacologic therapies such as melatonin supplementation [41–43].
Building on this, experts and organizations such as the American Thoracic Society recommend implementing behavioral interventions to optimize the ICU environment, including lowering voices, closing doors, dimming lights, and limiting nighttime procedures [34]. These changes require staff education and may be difficult in open-plan ICUs [24, 29, 35]. Cyclic lighting systems simulating natural day-night transitions [4] and morning light therapy [26] have been associated with improved subjective sleep scores and patient satisfaction, including feelings of safety and better communication with staff [4].
Multicomponent, interdisciplinary interventions combined with staff education have proven safe, effective, and practical for ICU use, enhancing both sleep duration and quality [24, 29]. Experts emphasize the importance of maintaining a natural rhythm of light and dark, activity and rest, feeding and sleep – even in critically ill patients [4, 34, 35, 39, 40, 44, 45]. A summary of practical recommendations that can be implemented in the ICU is presented in Table 2.
Melatonin remains the most frequently used pharmacological treatment for ICU sleep disorders, followed by ramelteon and quetiapine [34, 46, 47]. Melatonin was trialed with non-pharmacologic methods, but the study was limited (n = 6) and inconclusive [28]. Ramelteon showed no improvement in sleep duration or circadian alignment in one study [31]. Several studies investigated dexmedetomidine, a sedative considered closest to natural sleep in the ICU [23]. Two polysomnography studies and RASS found longer sleep durations and better sleep efficiency with dexmedetomidine, though no effect on REM sleep or delirium was seen [23, 32]. Another study reported increased sleep efficiency, more stage N2, reduced fragmentation, and restored nocturnal sleep patterns. One trial using only RASS noted reduced anxiety and improved sleep conditions, without direct sleep quality assessment [33]. Thus, if clinically appropriate, dexmedetomidine is worth considering as an alternative to propofol and benzodiazepines for sedation in mechanically ventilated ICU patients, especially for light to moderate sedation. It is noninferior to propofol and midazolam in sedation maintenance and offers benefits such as reduced delirium, shorter ventilation duration, and better patient communication. However, it may increase bradycardia and hypotension risk and is generally less suitable for deep sedation [48–50].

Conclusions

In summary, the analysis confirmed that both environmental and pharmacological interventions can improve objective sleep parameters and subjective experiences of ICU patients; however, there is no strong evidence supporting the superiority of one approach over the other. Sleep disturbances in the ICU are multifactorial, resulting from a combination of environmental, individual, medical, and pharmacological factors. While mechanical ventilation, sedation, and necessary medical procedures significantly fragment sleep and alter its architecture, optimizing the ICU environment, appropriately adjusting pharmacotherapy and ventilation modes, as well as adoption of a holistic approach by the care team, are crucial for improving sleep quality in patients. Complex protocols that address multiple factors simultaneously have shown high effectiveness, but further research is required to confirm these findings.
Implementing some interventions in the ICU can be challenging, as many life-saving measures often take precedence over promoting patients’ natural circadian rhythms. Nevertheless, there is a current trend toward maintaining greater patient activity and minimizing sedation in the ICU. Given the severity of circadian rhythm disturbances and the consequences of sleep disorders in critically ill patients, ICU nurses face new challenges and increased responsibility in promoting and protecting their patients’ sleep.

Disclosures

1. Institutional review board statement: Not applicable.
2. Detailed description of the use of AI: Proofreading of the English translation (verification of the translation made by the authors of the work).
3. Financial support and sponsorship: None.
4. Conflicts of interest: None.

References