Delirium and sleep in intensive care I – epidemiology, risk factors and outcomes

Overview

Poor quality of sleep and delirium are frequent complications of intensive care. The incidence of both complications is high, and evidence-based medicine has significantly demonstrated serious consequences in both cases. More data are available on delirium. While there is significant room for further research on sleep quality impairment, there are also technical limitations of monitoring and diagnosis. This article summarises known data on the epidemiology and risk factors of decreased quality of sleep and delirium in the intensive care setting.

Keywords:

intensive care – sleep – delirium – risk factors – critical care outcomes

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

 

Introduction

Delirium is a common complication in intensive care units (ICUs) with a prevalence of up to approximately 31% [1]. Poor sleep quality in intensive care has been reported in approximately one-third of unsedated patients and up to 60-97% of sedated patients [2,3]. Despite the high prevalence, identifying risk factors, short-term and long-term consequences is a clinical challenge [4,5]. A large recent study of 5,936 patients demonstrated an association of delirium with a higher risk of mortality in the first 30 days after hospital discharge, while the occurrence of delirium during ICU hospitalization increased the risk of needing rehospitalization or requiring treatment in the emergency department or urgent admission [6]. The reality of clinical practice in the diagnosis of delirium is reflected in a recent questionnaire study conducted at clinics in the Czech Republic. The estimated prevalence of delirium in patients with ictus varies widely between 5-60%, which the authors compare with data from the world literature (10-48%) [7].

 

Sleep - in general

Patients with acute sleep deprivation (SD) exhibit reduced psychomotor performance and impaired short-term memory [8]. SD in a study on healthy adults led to mood disturbances, development of anxiety, depression, irritability, and paranoia [9]. According to an available systematic review, after only one day of SD, psychopathological symptoms begin to manifest: after 24-48 h, distorted perception (specifically in the visual modality in 90%, somatosensory in 52% and auditory in 33%), anxiety, irritability, temporal disorientation and depersonalization. After 48-90 h, complex hallucinations and thought disorders. After 72 h, delusions become apparent and the clinical picture resembles acute psychosis or delirium [10]. Delirium is a qualitative disorder of consciousness that manifests as an acute disturbance of attention, fluctuates over time, and is accompanied by changes in cognitive performance and changes in psychomotor activity [11].

During sleep, there is a cyclical alternation of several stages of sleep. This cycle lasts approximately 90-110 min [12] and repeats four to five times during the night. The alternation of sleep and wakefulness is controlled by the circadian regulatory system and the homeostatic process of sleep. The main pacemaker of the circadian system is located in the suprachiasmatic nucleus in the hypothalamus. The circadian regulatory system is responsible for a number of biological processes including endocrine regulation of certain hormones such as cortisol, thyrotropic hormone and growth hormone. Cortisol levels are lowest at midnight, 2-3 h after the onset of sleep its level gradually increases to its morning maximum. The largest pulse of growth hormone is during the first N3 phase of sleep [13]. However, the main hormone involved in the control of the sleep-wake cycle is melatonin. Moreover, since the production of this hormone decreases with age, it is assumed that melatonin deficiency is at least partly responsible for sleep disturbances. Treating this age-related deficit therefore appears to be a natural way to restore sleep quality [14]. The homeostatic process of sleep, on the other hand, is influenced by the length of previous wakefulness. A long duration of wakefulness produces a strong signal and an increased need for sleep [15].

In healthy volunteers, SD affects immunity and the inflammatory response. For example, it modifies the circadian oscillations of cortisol concentration, which adapts within about 5 days after SD [16]. Several days of SD also reduces thyrotropic hormone levels, although paradoxically the first night thyrotropic axis levels are elevated [17].

Furthermore, SD increases the number of leukocytes, especially neutrophils, increases cortisol levels, insulin resistance and increases sympathetic activity [18]. By these mechanisms, SD increases the risk of developing arterial hypertension and induces the development of metabolic syndrome. Elevated evening cortisol levels during SD may induce memory impairment. Analogous changes in memory occur in old age.

 

Stages and parameters of sleep

Using polysomnography (PSG) during sleep in a healthy adult, we distinguish between non-rapid eye movement (NREM) and rapid eye movement (REM) sleep based on changes in brain activity, muscle tone and eye movement. NREM is further divided into three stages, referred to as N1-3. N1 is light sleep with slow eye movement and higher muscle tone. At N2, eye movement stops and brain activity on the EEG slows down, and muscle tone decreases.

For scoring a sleep epoch as N2, the presence of at least one sleep spindle and/or at least one K-complex is a prerequisite. A sleep spindle is defined as an activity of 11-16 Hz (most commonly 12-14 Hz) of spindle-like (sinusoidal) character with a duration of at least 0.5 s, characteristically most pronounced in the central regions. A K-complex is defined as a negative sharply contoured wave with a subsequent positive component exceeding 0.5 s. A person is least awake during the first NREM approximately one hour after falling asleep. Within one sleep cycle, a person is less awake at N3. Slow wave sleep is defined by the American Academy of Sleep Medicine (AASM) by the presence of waves of 0.5-2.0 Hz with an amplitude exceeding 75 µV on the EEG. Their frequency corresponds to delta waves (defined as 0-4 Hz), which is why we sometimes use the term delta sleep in the Czech environment. An N3 stage epoch is defined by the AASM as the presence of slow wave activity (slow waves as defined above) in ≥ 20% of a given epoch. The presence of transients (characteristic of N2) during N3 is common [19]. During N3, low tonic muscle activity (no movements occur) may still be noted by EMG. During REM sleep, electrooculography (EOG) records rapid eye movements, there may be minor twitching of the mimic muscles and respiratory and heart rates increase, but there is full muscle atonia [20]. In addition to the EEG, EOG and EMG already mentioned, PSG includes ECG monitoring, pulse oximetry, measurement of respiratory effort, nasal airflow, snoring and recording of body movement and sound including a night vision camera.

 

Sleep in the ICU

Although the total sleep time in the ICU is the same or only slightly reduced, its distribution varies considerably among ICU inpatients. ICU patients have sleep fragmented into short periods and up to 50% of sleep time occurs during the day [21]. This creates a mismatch between the circadian regulatory system and the homeostatic process of sleep [22]. Furthermore, sleep disturbances in the ICU setting are characterized by an increase in the length of light sleep (N1 and N2 stages) and a decrease in the length of N3 stage and REM sleep [23].

Associations have been demonstrated between sleep disturbance and the development of delirium [24], between sleep disturbance and prolonged time on artificial pulmonary ventilation (UPV) [25] and the prediction of non-invasive ventilation (NIV) failure in patients admitted with global respiratory insufficiency [26]. Sleep deprivation reduces ventilatory response to hypoxemia and hypercapnia [25]. Severe hypercapnia exerts a hypnotic effect, exacerbating hypoxia and further adversely affecting circadian rhythm.

 

Epidemiology of sleep in the ICU

In patients without sedation or with only mild sedation, the prevalence of abnormal PSG recordings is approximately 23-31% and in sedated patients the prevalence of abnormality is 60-97% [2,27]. Abnormality in the PSG refers to deviation from the standard PSG recording, the characteristics of a normal polysomnogram, and the evaluation of abnormalities can be assessed according to standard scoring tools such as the AASM Scoring Manual [28].

The variability in the prevalence of abnormal PSG recordings is explained by the presence of factors (e.g., sedation, sepsis, and delirium) known to generate abnormal sleep electroencephalogram patterns [23]. When other sleep monitoring methods such as actigraphy and the Richards-Campbell Sleep Questionnaire (RCSQ) were used, the prevalence of poor quality sleep was 47% (RCSQ < 50), reaching up to 66% in the subgroup of patients on UPV compared to 33% in the subgroup without UPV [3].

 

Epidemiology of delirium in the ICU

A recent meta-analysis including 48 studies reported an overall prevalence of delirium of 31%. The prevalence of each motor subtype of delirium was as follows: hyperactive (4%), hypoactive (17%) and mixed (10%) [1].

In the Czech setting, we have limited data from a study in a mixed population of trauma, surgical and internal medicine patients admitted to the ICU [29], where the incidence of ICU delirium was 31.2%. Patients with primary internal medicine and post-trauma are at higher risk of developing delirium compared to surgical patients. There are also Czech data from studies by Brno authors on the incidence of other subtypes of delirium, namely postoperative delirium [30] and delirium after stroke [31].

 

Risk factors for poor sleep quality and the development of delirium

The risk factors for delirium based on research conducted and their associations with sleep quality are often heterogeneous. Nevertheless, multiple studies suggest the following risk factors for poor sleep quality and risk of developing delirium: loss of day-night routine, together with the effect of higher light intensity and noise during night hours, inadequate analgesia, immobilization of the patient, limited ability to communicate, fear of unfamiliar surroundings and concern for one's future, influence of medications, and others. A summary of the findings of the individual studies and the determination of risk factors for sleep disturbances in the ICU was made by Honarmand et al [4] (Table 1). The risk factors for delirium are listed in Table 2 [23,32-36]. These are augmented by 33 predisposing factors in the recently published work of Ormseth et al. [37] including 101,114 patients from 315 studies; this work also identified factors associated with a reduced risk of developing delirium.

The strongest protective factor was cognitive reserve, which is an individual's ability to withstand changes or stresses in brain function. Patients with dementia, where cognitive reserve is already reduced prior to ICU admission, have a higher incidence of delirium and sleep disturbances compared to the healthy population.

 

Consequences of poor sleep quality in the ICU

In critically ill patients with a more severe degree of encephalopathy, a shorter period to extinction of sleep in the N2 phase has been demonstrated. And at the same time changes of N2 stage in comparison with sleep of healthy subjects (disappearance of K-complexes and sleep spindles on EEG). Lower representation of the N2 stage and its alteration correlates negatively with a higher probability of death and may have an important prognostic value [38]. In an intervention study comparing sleep quality by questionnaire methods and the incidence of delirium after the introduction of multimodal non-pharmacological sleep support, there was a significant reduction in the proportion of days on which patients suffered from delirium [39]. There is an association between poor sleep quality, the occurrence of delirium and increased mortality. Given the wide variability in sleep monitoring options and study results, the consequences of poor sleep quality are summarized in the current PADIS (Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption) guidelines as an unclear association between sleep quality and the length of need for mechanical ventilatory support, length of ICU stay, and ICU mortality in critically ill adults. Similarly, the effect of sleep quality and circadian rhythm changes on outcomes of critically ill patients after discharge from the ICU needs further investigation [23].

 

Consequences of delirium

The current PADIS guideline published in 2018 and summarising the results of the studies states the following: delirium is strongly associated with cognitive impairment at follow-up intervals, i.e. 3 and 12 months after ICU discharge, and an association of delirium with longer overall length of hospital stay is possible. In contrast, according to the conclusions of these guidelines, delirium does not prolong ICU stay, is not a risk factor for depression, does not affect where the patient is discharged, does not worsen dependency on others, and does not increase mortality [23]. However, these findings do not agree with the findings of a previous review, where patients with delirium had significantly higher mortality, longer UPV time and longer hospital and intensive care hospital stays [5]. A recent study comparing two cohorts (each cohort comprising 2,968 patients) found an association of delirium with a risk of higher mortality in the first 30 days after hospital discharge, and furthermore, the occurrence of ICU delirium during hospitalization increased the risk of needing rehospitalization or a stay in the emergency department or urgent care [6]. Nor is a coherent conclusion established in the fundamental implications of the occurrence of ICU delirium.

 

Sleep and delirium

The relationship between sleep and delirium in the ICU is complex and interrelated. Sleep deprivation causes or exacerbates delirium, and the occurrence of delirium may contribute to sleep disturbances, creating a potential circulus vitiosus in critically ill patients [15]. Overall duration of delirium in the ICU has been found to be significantly associated with increased sleep disturbance at long-term follow-up (mean 5 months after hospital discharge) [40].

A limitation in current knowledge is that hypotheses about the relationship between sleep and delirium have largely been derived from studies of healthy individuals or patients outside the ICU [15]. Sleep monitoring in critically ill ICU patients is difficult compared to the controlled environment of a sleep laboratory, and PSG monitoring alone may influence sleep.

 

Conclusion

During intensive care hospitalization, most patients have reduced sleep quantity and quality, with an associated increased risk of ICU delirium. Developed ICU delirium also impacts patients' sleep quality. Thus, poor sleep quality and ICU delirium affect each other and cannot be isolated in their approach. Although poor sleep quality and ICU delirium tend to be common complications, the effects, especially the long-term ones, on the patient's condition after discharge from acute care have not been well described.

 

Grant support

Supported by the Ministry of Health of the Czech Republic - RVO (FNBr, 65269705), the Ministry of Education and Science of the Czech Republic through the VVI CZECRIN project (LM2023049) and the MU Brno specific research project (MUNI/A/1186/2022, MUNI/A/1109/2022 and MUNI/A/1105/2022).

 

Conflict of interest

The authors declare that they have no conflict of interest in relation to the topic of the paper.

 

Tables

Table 1. Risk factors for poor sleep quality.

Risk factors present before ICU admission	Risk factors associated with the disease	Risk factors associated with ICU
polymorbidity poor sleep before ICU admission use of hypnotics prior to ICU admission	delirium	environmental influence (noise, lighting) psychological factors and needs (pain, fear for self and close family members, anxiety, nervousness, disorientation over time, thirst and hunger) use of benzodiazepines nursing, diagnostic and therapeutic interventions (including the need for immobilization, monitoring of vital functions) respiratory insufficiency and support (dyspnoea, cough, invasive ventilation: support regimens)

ICU - Intensive Care Unit

 

 

Table 2. Risk factors for delirium.

Risk factors present before ICU admission	Risk factors acquired during ICU stay
older age, Clinical Frailty Scale, APACHE II [32,33] previous compromise of consciousness (dementia, previous coma) [33,34] arterial hypertension [33] acute surgery or trauma before ICU admission [23,34]	benzodiazepine use and inadequate deep sedation [35] use of anticholinergic drugs and steroids [36] use of opioids with the possibility of accumulation (morphine) [34] inadequate pain management [34]  metabolic acidosis [33,34] artificial pulmonary ventilation [33,34] use of restraints [35] length of ICU stay [35] blood transfusion [23]

APACHE - Acute Physiology and Chronic Health Evaluation, ICU - Intensive  Unit