Sleep and Circadian Rhythms in ICU

Clinical Overview

Sleep is a fundamental physiological process essential for cognitive function, immune modulation, tissue repair, and metabolic homeostasis. In the intensive care unit (ICU), sleep is profoundly disrupted, with patients experiencing severe fragmentation, loss of normal sleep architecture, and circadian rhythm desynchronisation. Understanding normal sleep physiology and the mechanisms of disruption in critical illness is essential for the intensive care physician.

ICU patients typically achieve only 2-4 hours of fragmented sleep per 24 hours, compared to the 7-9 hours recommended for healthy adults. Sleep is characterised by frequent arousals (>20 per hour), absence of normal sleep cycling, and marked reduction in restorative slow-wave and REM sleep. Up to 50% of "sleep" in ICU patients may occur during daytime hours, reflecting loss of circadian entrainment. [1,2,3]

Sleep deprivation and circadian disruption are increasingly recognised as modifiable risk factors for ICU delirium, prolonged mechanical ventilation, immune dysfunction, and poor long-term outcomes. The 2018 PADIS Guidelines (Pain, Agitation/sedation, Delirium, Immobility, and Sleep disruption) include sleep promotion as a core component of ICU care bundles. [4,5]

Epidemiology

Prevalence of Sleep Disruption in ICU:

• 50-100% of ICU patients experience significant sleep disruption
• Mean total sleep time: 2-4 hours per 24 hours (vs 7-9 hours normal)
• Sleep efficiency: 30-50% (vs >85% normal)
• Up to 20-57 arousals per hour (vs <10 normal)
• 30-50% of sleep occurs during daytime hours [1,2,6]

Circadian Rhythm Disruption:

• Loss of normal diurnal melatonin rhythm in 70-90% of ICU patients
• Abnormal cortisol rhythms in up to 80% of septic patients
• Persistent circadian dysregulation for weeks after ICU discharge [7,8]

Clinical Consequences:

• Delirium: Sleep deprivation increases delirium incidence by 2-3 fold
• Mechanical ventilation: Poor sleep quality associated with prolonged weaning
• Immune dysfunction: Increased infection susceptibility
• Mortality: Some studies suggest association with increased mortality [3,9,10]

---

Normal Sleep Physiology

Sleep Architecture

Normal sleep consists of cyclical alternation between two fundamental states: Non-REM (NREM) sleep and REM (Rapid Eye Movement) sleep. A complete sleep cycle lasts approximately 90-120 minutes, with 4-6 cycles occurring during a normal 8-hour sleep period. [11,12]

Non-REM Sleep Stages

Stage N1 (Light Sleep):

• Transition from wakefulness to sleep
• Duration: 1-7 minutes (2-5% of total sleep time)
• EEG: Low-amplitude, mixed-frequency activity (4-7 Hz theta waves)
• Characteristics: Drowsy, easily aroused, slow rolling eye movements
• Decreased muscle tone, reduced responsiveness to environment
• Hypnic jerks may occur (sudden muscle contractions)

Stage N2 (Light Sleep):

• Consolidation of sleep, majority of sleep time
• Duration: 45-55% of total sleep time in adults
• EEG: Sleep spindles (12-14 Hz, 0.5-1.5 seconds) and K-complexes
• Sleep spindles: Bursts of rhythmic activity from thalamocortical circuits
• K-complexes: High-amplitude negative-positive waves, thought to protect sleep
• Moderate threshold for arousal, decreased metabolic rate

Stage N3 (Deep Sleep / Slow-Wave Sleep):

• Most restorative sleep stage
• Duration: 15-25% of total sleep time (decreases with age)
• EEG: High-amplitude delta waves (0.5-2 Hz, >75 μV)
• Functions:
  • Memory consolidation (declarative memory)
  • Growth hormone release (>70% occurs during N3)
  • Tissue repair and regeneration
  • Immune system restoration
  • Metabolic waste clearance via glymphatic system
• High arousal threshold, disorientation if awakened
• Most prominent in first third of the night [11,12,13]

REM Sleep

Characteristics:

• Paradoxical sleep: EEG resembles wakefulness (low-amplitude, mixed-frequency)
• Rapid conjugate eye movements
• Muscle atonia (except diaphragm and ocular muscles)
• Duration: 20-25% of total sleep time
• Predominantly occurs in the latter half of the night
• REM periods progressively lengthen through the night (10→60 minutes)

Physiological Features:

• Irregular respiratory pattern
• Variable heart rate and blood pressure
• Increased cerebral blood flow and metabolism
• Dreaming occurs predominantly during REM
• Thermoregulation impaired (poikilothermic state)
• Penile/clitoral erections occur

Functions:

• Procedural and emotional memory consolidation
• Emotional regulation and processing
• Synaptic pruning and plasticity
• Brain development (particularly in neonates, who spend 50% in REM) [11,12,14]

Sleep Cycle Distribution

Sleep-Wake Regulation

Sleep is regulated by two fundamental processes that interact to determine the timing, duration, and quality of sleep:

Process S (Homeostatic Sleep Drive)

Mechanism:

• Sleep pressure accumulates during wakefulness
• Mediated primarily by adenosine accumulation in the basal forebrain
• Adenosine is a by-product of ATP hydrolysis (neuronal metabolism)
• Acts on A1 and A2A receptors to inhibit wake-promoting neurons
• Dissipates during sleep, particularly during slow-wave sleep

Clinical Relevance:

• Caffeine is an adenosine receptor antagonist (delays sleep)
• Prolonged wakefulness increases adenosine → increased sleep pressure
• Sleep deprivation leads to "rebound" slow-wave sleep
• Process S is preserved in many ICU patients but cannot be expressed due to environmental and physiological disruption [15,16]

Process C (Circadian Rhythm)

Suprachiasmatic Nucleus (SCN):

• Master circadian pacemaker located in the anterior hypothalamus
• Contains approximately 20,000 neurons
• Intrinsic period: Approximately 24.2 hours (circadian = "about a day")
• Generates autonomous rhythms even without external cues
• Entrains to light-dark cycle via retinohypothalamic tract
• Projects to multiple brain regions to coordinate peripheral clocks

Molecular Clock Mechanism:

• Transcription-translation feedback loops (24-hour period):
  • "CLOCK/BMAL1: Positive regulators (activate transcription)"
  • "PER/CRY: Negative regulators (inhibit CLOCK/BMAL1)"
  • "Cycle: CLOCK/BMAL1 activate PER/CRY → PER/CRY accumulate → inhibit CLOCK/BMAL1 → PER/CRY degrade → cycle restarts"
• Post-translational modifications (phosphorylation) fine-tune timing
• REV-ERB and ROR proteins provide additional regulation [17,18]

Circadian Output Signals:

• Melatonin: Pineal gland, darkness-induced, promotes sleep
• Cortisol: Adrenal gland, morning peak, promotes wakefulness
• Core body temperature: Nadir at 04:00-05:00, peak at 18:00-20:00
• Autonomic tone: Parasympathetic dominance during sleep

Melatonin Physiology

Synthesis Pathway:

Tryptophan → 5-Hydroxytryptophan → Serotonin → N-acetylserotonin → Melatonin

• Synthesised in the pineal gland
• Rate-limiting enzyme: Arylalkylamine N-acetyltransferase (AANAT)
• Suppressed by light (retina → SCN → pineal inhibition via sympathetic pathway)
• Peak secretion: 02:00-04:00 (10-80 pg/mL)
• Daytime levels: <10 pg/mL
• Half-life: 30-60 minutes (hepatic metabolism)

Receptors:

• MT1: High affinity, G-protein coupled, inhibits neuronal firing
• MT2: High affinity, phase-shifting properties
• Both located in SCN, cardiovascular system, immune cells

Functions:

• Sleep promotion (decreases sleep latency, increases sleep efficiency)
• Circadian phase entrainment (exogenous melatonin can shift circadian phase)
• Antioxidant and free radical scavenger
• Immunomodulation
• Core body temperature reduction [19,20,21]

Cortisol Circadian Rhythm

Normal Pattern:

• Nadir: 00:00-02:00 (lowest cortisol)
• Cortisol awakening response (CAR): 50-100% increase within 30-45 minutes of waking
• Peak: 08:00-09:00 (morning peak)
• Gradual decline throughout day
• Ultradian pulsatility superimposed on circadian rhythm

Regulation:

• Hypothalamic-pituitary-adrenal (HPA) axis
• CRH → ACTH → Cortisol
• SCN projects to paraventricular nucleus (PVN) to modulate CRH release
• Negative feedback at pituitary and hypothalamus
• Stress can override circadian control [22,23]

Neurobiology of Sleep-Wake States

Wake-Promoting Systems (Arousal)

Orexin/Hypocretin System:

• Critical for sleep-wake stability
• Orexin neurons in lateral hypothalamus
• Project to all major wake-promoting nuclei
• Loss of orexin → Narcolepsy (cataplexy, sleep attacks)
• Active during wakefulness, silent during sleep [24,25]

Sleep-Promoting Systems

Flip-Flop Switch Model (Saper):

• VLPO (sleep) and arousal nuclei are mutually inhibitory
• Creates bistable states (awake or asleep)
• Orexin stabilises the "wake" state
• Transition between states is rapid (not gradual)
• Dysfunction → unstable state transitions [25]

---

Sleep Disruption in ICU

Causes of Sleep Disruption

Sleep disruption in the ICU is multifactorial, involving environmental, physiological, and iatrogenic factors:

Environmental Factors

Noise:

• ICU noise levels: 50-80 dB (peak 85-100 dB)
• WHO recommends <35 dB for hospitals
• Sources: Alarms (60-80 dB), staff conversations (50-65 dB), equipment
• Noise accounts for only 11-17% of arousals (other factors dominant)
• Peak noise (alarms) more disruptive than ambient noise [26,27]

Light:

• Continuous or inappropriate light exposure disrupts melatonin
• Night-time light: 5-1400 lux in ICU (vs <0.5 lux in bedroom)
• Daytime light often inadequate for circadian entrainment
• Blue-enriched light (460-480 nm) maximally suppresses melatonin
• Light exposure timing critical for circadian phase [28,29]

Temperature:

• ICU ambient temperature often outside thermoneutral zone
• Core body temperature rhythm disrupted
• Thermoregulation impaired by sedation and critical illness

Clinical Interventions

Patient Care Activities:

• Nursing interventions: Vital signs, repositioning, medication administration
• Average 50+ patient care activities per night
• Clustering care may reduce but not eliminate disruption
• Necessary interventions difficult to eliminate [30,31]

Mechanical Ventilation:

• Patient-ventilator dyssynchrony disrupts sleep
• Pressure support may increase arousals vs assist-control
• Inappropriate ventilator settings cause discomfort
• Sedation for ventilation further disrupts architecture [32,33]

Monitoring Alarms:

• False alarms: 72-99% of alarms are non-actionable
• Alarm fatigue in staff leads to delayed response
• Contributes to arousal and sleep fragmentation

Medications

Sedation Effects:

• Sedative-induced "sleep" lacks normal architecture
• EEG patterns differ from physiological sleep
• Does not provide restorative benefits of natural sleep
• Deep sedation associated with worse outcomes [34,35,36]

Pathophysiological Factors

Critical Illness:

• Inflammatory cytokines alter sleep (IL-1β, TNF-α, IL-6)
• Sepsis: Profound circadian disruption
• Pain and discomfort
• Metabolic disturbances
• Hypoxia and hypercapnia
• Delirium (bidirectional relationship) [37,38]

Circadian Disruption in Critical Illness:

• Loss of normal melatonin rhythm
• Abnormal cortisol patterns (elevated, loss of diurnal variation)
• Altered core body temperature rhythms
• Dysfunction of peripheral clock genes
• May persist for weeks after ICU discharge [7,8,39]

Consequences of Sleep Deprivation

Cognitive Effects

Acute Sleep Deprivation:

• Attention deficits (proportional to duration)
• Working memory impairment
• Executive function decline
• Decision-making impaired
• 24 hours without sleep ≈ blood alcohol 0.10%
• Microsleeps and lapses in attention [40,41]

ICU-Relevant Cognitive Effects:

• Delirium: Strong bidirectional relationship with sleep deprivation
• Confusion and disorientation
• Difficulty participating in rehabilitation
• Communication impairment

Immune System Effects

Innate Immunity:

• Decreased NK cell activity (35-45% reduction after one night)
• Impaired phagocytic function
• Increased inflammatory cytokines (IL-6, TNF-α, CRP)
• Shift toward pro-inflammatory state [42,43]

Adaptive Immunity:

• Reduced antibody response to vaccination
• Impaired T-cell function
• Decreased IL-2 and IFN-γ production
• Increased susceptibility to infection

Clinical Implications:

• Increased risk of nosocomial infections
• Delayed wound healing
• Potential impact on sepsis outcomes

Metabolic Effects

Glucose Metabolism:

• Decreased glucose tolerance
• Increased insulin resistance
• Elevated cortisol and catecholamines
• Glycemic variability increased [44,45]

Hormonal Changes:

• Decreased growth hormone secretion
• Elevated cortisol (loss of diurnal variation)
• Altered thyroid hormone metabolism
• Reduced testosterone
• Increased ghrelin (hunger), decreased leptin (satiety)

Cardiovascular Effects

• Increased sympathetic tone
• Elevated blood pressure
• Tachycardia
• Endothelial dysfunction
• Pro-inflammatory, pro-thrombotic state
• Increased arrhythmia risk [46]

Respiratory Effects

• Decreased respiratory muscle endurance
• Impaired ventilatory responses to hypoxia/hypercapnia
• Possible contribution to weaning failure
• Upper airway instability [47]

Sleep-Delirium Connection

Bidirectional Relationship:

• Sleep deprivation is an independent risk factor for delirium
• Delirium disrupts sleep architecture
• Shared neurobiological pathways [48,49,50]

Evidence:

• Inouye et al.: Sleep deprivation included in delirium risk factor models
• Weinhouse et al.: Abnormal sleep-wake patterns precede delirium onset
• Watson et al.: REM sleep deprivation associated with delirium in ICU
• Knauert et al.: Sleep promotion may reduce delirium incidence [3,9,10]

Mechanisms:

• Adenosine accumulation → cognitive dysfunction
• Impaired glymphatic clearance of metabolic waste
• Inflammatory mediators affect both sleep and cognition
• Cholinergic dysfunction common to both
• Circadian disruption impairs cognitive processing

---

Assessment of Sleep in ICU

Polysomnography (PSG)

Gold Standard for sleep measurement in research settings:

Components:

• EEG (electroencephalography): 6+ channels for sleep staging
• EOG (electrooculography): Eye movement detection
• EMG (electromyography): Chin and limb muscle activity
• ECG: Heart rhythm monitoring
• Respiratory monitoring: Airflow, effort, oximetry

ICU-Specific Challenges:

• Technical expertise required for setup and interpretation
• Artifact from patient movement, interventions, electrical interference
• Modified montages may be needed (fewer channels)
• ICU sleep architecture differs from normal (atypical EEG patterns)
• Up to 30-50% of ICU recordings uninterpretable
• Expensive, labour-intensive [51,52]

Findings in ICU Patients:

• Severe sleep fragmentation (arousals/awakenings)
• Absent or minimal N3 and REM sleep
• Loss of normal sleep cycling
• Atypical EEG patterns (not fitting standard criteria)
• Pathological wakefulness (EEG suggesting sleep despite behavioural wakefulness)

Actigraphy

Method:

• Wrist-worn accelerometer (watch-like device)
• Infers sleep/wake from movement patterns
• Non-invasive, well-tolerated, continuous monitoring

Advantages:

• Simple, low cost, non-invasive
• Multi-day continuous monitoring feasible
• No specialised personnel required
• Circadian rhythm assessment possible

Limitations in ICU:

• Overestimates sleep (immobile but awake patients)
• Poor accuracy in sedated or paralysed patients
• Cannot assess sleep architecture
• Validated primarily in healthy populations
• Moderate correlation with PSG (60-80% epoch-by-epoch agreement) [53,54]

Bispectral Index (BIS)

Method:

• Processed EEG algorithm developed for anaesthesia depth
• Scale: 0-100 (0 = isoelectric, 40-60 = general anaesthesia, 80-100 = awake)

ICU Sleep Assessment:

• Some studies suggest BIS 60-80 correlates with sleep
• Not validated for physiological sleep staging
• Affected by sedatives and muscle relaxants
• May indicate "unconsciousness" but not equivalent to restorative sleep [55]

Subjective Assessment Tools

Richards-Campbell Sleep Questionnaire (RCSQ):

• 5 visual analogue scales (0-100 mm)
• Domains: Sleep depth, latency, awakenings, return to sleep, quality
• Validated for ICU use
• Requires patient recall and ability to respond
• Score <50 indicates poor sleep perception [56]

Nursing Sleep Assessment:

• Nurse observation of patient sleep
• Limited correlation with PSG
• Overestimates sleep quantity
• Misses arousals and architecture disruption
• Practical but unreliable

Sleep Questionnaires (Pre-ICU):

• Pittsburgh Sleep Quality Index (PSQI)
• Epworth Sleepiness Scale (ESS)
• Berlin Questionnaire (sleep apnoea risk)
• Useful for baseline assessment if patient can communicate

Biomarkers

Melatonin/6-Sulfatoxymelatonin (Urinary Metabolite):

• Assesses circadian rhythm integrity
• Serial sampling required (6-24 hour collection)
• Research tool, not routine clinical use
• Absent or blunted rhythm indicates circadian disruption [7,8]

Cortisol Rhythm:

• Serial measurements assess diurnal variation
• Sepsis often shows elevated, flat pattern
• Research application

---

Interventions to Improve Sleep

PADIS 2018 Guidelines Recommendations

The Society of Critical Care Medicine's PADIS Guidelines provide the following recommendations for sleep promotion: [4]

Recommendations:

1. Use a multicomponent sleep-promotion protocol (ungraded good practice statement)
2. Promote sleep by optimising patients' environments (reducing light, noise, clustering care) (ungraded good practice statement)
3. Nocturnal assist-control ventilation should be considered over pressure support to improve sleep (conditional recommendation, low quality evidence)
4. No recommendation for pharmacological sleep aids due to insufficient evidence

Non-Pharmacological Interventions

Environmental Modifications

Noise Reduction:

• Quiet time protocols (designated quiet periods, especially 22:00-06:00)
• Alarm management (reduce false alarms, adjust alarm parameters)
• Behavioural interventions (staff education, lowered voices)
• Structural changes (single rooms, sound-absorbing materials)
• Earplugs: RCTs show improved sleep quality and reduced delirium [57,58]

Light Optimisation:

• Reduce nocturnal light exposure (<10 lux)
• Eye masks: Evidence supports improved sleep quality [57,59]
• Bright light therapy during daytime (2500-10000 lux for 30-60 minutes)
• Blue-enriched light in morning to reinforce circadian rhythm
• Avoid blue light at night (use amber/red spectrum if illumination needed)

Temperature Control:

• Maintain thermoneutral zone
• Adjust bedding and ambient temperature

Care Organisation

Clustering of Care Activities:

• Combine interventions to reduce interruption frequency
• Designate protected sleep periods (avoid non-urgent interruptions)
• Coordinate with other services (radiology, laboratory draws)

Reduced Nocturnal Interventions:

• Question necessity of overnight vital signs
• Use continuous monitoring to reduce episodic checks
• Delay non-urgent medications to daytime

Sleep Promotion Bundles

Example Bundle Components:

1. Earplugs and eye masks offered nightly
2. Quiet time protocol (22:00-06:00)
3. Lights dimmed at 22:00
4. Clustered care to minimise interruptions
5. Daytime mobilisation and activity
6. Avoidance of daytime sleep >2 hours
7. Natural light exposure during day

Evidence:

• Kamdar et al.: Multi-component bundle improved sleep perception (RCSQ) by 20-30% [60]
• Patel et al.: Sleep bundle reduced delirium incidence [61]
• Van Rompaey et al.: Earplugs reduced confusion risk [62]

Pharmacological Interventions

Melatonin

Mechanism:

• Exogenous supplementation to restore circadian signal
• MT1/MT2 receptor agonism
• Antioxidant properties

Dosing:

• Typical dose: 3-10 mg PO at 21:00-22:00
• Immediate-release preferred for sleep initiation
• Extended-release may help sleep maintenance
• Bioavailability highly variable (10-50%)

Evidence in ICU:

Limitations:

• Variable bioavailability
• Timing of administration critical
• Heterogeneity in ICU populations
• Current evidence does not support routine use for delirium prevention
• May be considered for sleep promotion in selected patients [4,68]

Ramelteon

Mechanism:

• Selective MT1/MT2 receptor agonist
• Higher affinity than melatonin
• No GABAergic activity (no abuse potential)
• FDA-approved for insomnia (sleep onset)

Pharmacology:

• Dose: 8 mg PO at bedtime
• Half-life: 1-2.6 hours (active metabolite M-II: 2-5 hours)
• Hepatic metabolism (CYP1A2 major)
• No respiratory depression
• Contraindicated with fluvoxamine (CYP1A2 inhibitor)

Evidence in ICU:

Limitations:

• Only available in oral formulation (requires enteral access)
• Expensive compared to melatonin
• Evidence limited to specific populations
• Not recommended for routine use [4]

Other Pharmacological Agents

Dexmedetomidine:

• Alpha-2 agonist with unique sleep-promoting properties
• Produces sedation resembling natural sleep (N2 stage)
• Preserves REM to some degree
• SPICE III (Australia/NZ): No mortality benefit but possibly less delirium
• May have role in sleep-focused sedation protocols [72,73]

Agents to Avoid for Sleep Promotion:

• Benzodiazepines: Suppress N3 and REM, worsen delirium, not for sleep
• Propofol: Drug-induced unconsciousness, not restorative sleep
• Diphenhydramine: Anticholinergic, worsens delirium
• Antipsychotics: MIND-USA showed no benefit for delirium
• Zolpidem/Zopiclone: Limited ICU data, potential for paradoxical reactions [4,36]

Ventilator Optimisation

Mode Selection:

• Assist-control ventilation may improve sleep vs pressure support
• Pressure support: High drive can cause central apnoeas during sleep
• Adaptive support ventilation: May improve patient-ventilator synchrony
• PAV+ (Proportional Assist Ventilation Plus): Promising but limited data [32,33]

Settings Optimisation:

• Adequate support to reduce work of breathing
• Appropriate trigger sensitivity
• Minimise dyssynchrony (auto-PEEP, double triggering)
• Consider increasing support at night if WOB increases

---

Australian/NZ Context

ANZICS-CORE Considerations

• Sleep promotion increasingly emphasised in Australian ICU quality improvement initiatives
• ANZICS-CORE APD (Adult Patient Database) does not specifically capture sleep metrics
• Integration into ABCDEF bundle implementation across Australian ICUs
• PADIS guidelines adopted with local adaptation [74]

Indigenous Health Considerations

Aboriginal and Torres Strait Islander Peoples:

• Higher rates of sleep-disordered breathing in community
• Cultural factors affecting sleep patterns and preferences
• Family presence important for comfort and sleep
• Consider flexible visiting arrangements
• Aboriginal Health Worker/Liaison Officer engagement for communication
• Acknowledge that sleep deprivation adds to overall distress burden [75]

Māori Health (New Zealand):

• Whānau (extended family) involvement in care
• Tikanga (cultural protocols) may influence care timing
• Te Whare Tapa Whā model: Sleep relates to taha hinengaro (mental/emotional wellbeing)
• Cultural safety in sleep promotion interventions
• Māori Health Workers for communication and support [76]

---

Clinical Application to ICU

ICU Scenario 1: Post-Cardiac Surgery Patient

Presentation: Day 2 post-CABG, extubated, complaining of inability to sleep despite being exhausted.

Pathophysiology:

• Post-operative pain disrupting sleep
• Inflammatory response to surgery
• ICU environmental factors (noise, light)
• Anxiety and unfamiliar environment
• Medication effects (opioids, sedative withdrawal)

Management Based on Physiology:

1. Optimise analgesia (multimodal, avoid excessive opioids)
2. Offer earplugs and eye mask
3. Dim lights at 22:00, reduce nocturnal interruptions
4. Consider melatonin 3-5 mg at 21:00
5. Avoid benzodiazepines (delirium risk)
6. Encourage daytime mobilisation and activity
7. Re-establish day-night routine

ICU Scenario 2: Septic Patient on Mechanical Ventilation

Presentation: Day 5, sedated on propofol and fentanyl, delirious (CAM-ICU positive), irregular sleep-wake pattern observed.

Pathophysiology:

• Sepsis-induced circadian disruption (inflammatory cytokines)
• Sedative-induced sleep architecture abnormalities
• Continuous ICU environment (no day-night differentiation)
• Delirium-sleep bidirectional interaction

Management Based on Physiology:

1. Lighten sedation (target RASS 0 to -1)
2. Consider dexmedetomidine if continued sedation needed
3. Implement sleep bundle (noise reduction, light optimisation)
4. Daytime bright light exposure, nocturnal darkness
5. Mobility and physical therapy during day
6. Avoid exacerbating factors (anticholinergics, benzodiazepines)
7. Treat underlying sepsis
8. Consider melatonin (though evidence weak)

ICU Scenario 3: Weaning Failure with Sleep Deprivation

Presentation: Day 10, failed SBT ×3, observed to have fragmented sleep with frequent arousals.

Pathophysiology:

• Sleep deprivation impairs respiratory muscle recovery
• Decreased diaphragm endurance
• Cognitive dysfunction affecting cooperation
• Increased sympathetic tone affecting cardiovascular stability

Management Based on Physiology:

1. Implement sleep promotion bundle aggressively
2. Uninterrupted sleep period (protected 22:00-06:00)
3. Optimise ventilator settings for nocturnal comfort
4. Consider assist-control mode overnight
5. Avoid daytime naps >2 hours
6. Maximise daytime awakening and activity
7. Reassess SBT after 48-72 hours of improved sleep

---

Assessment

SAQ Practice Questions

SAQ 1: Sleep Architecture and Circadian Rhythm Regulation (15 marks)

Time: 18 minutes

Question:

A 55-year-old man is Day 3 post-emergency laparotomy for perforated diverticulitis. He is mechanically ventilated, receiving propofol sedation, and appears to have minimal restorative sleep despite periods of apparent unconsciousness.

(a) Describe the normal stages of sleep, including their EEG characteristics and functions. (6 marks)

(b) Explain the two-process model of sleep regulation and the role of the suprachiasmatic nucleus. (5 marks)

(c) Outline how sedation with propofol differs from physiological sleep at the neurobiological level. (4 marks)

---

Model Answer:

(a) Normal Sleep Stages (6 marks)

Sleep alternates between Non-REM (NREM) and REM sleep in 90-120 minute cycles (0.5 marks):

NREM Sleep:

• N1 (Light sleep, 2-5% TST): Low-amplitude mixed-frequency EEG (theta waves 4-7 Hz), drowsy, easily aroused, slow eye movements (1 mark)

• N2 (Light sleep, 45-55% TST): Sleep spindles (12-14 Hz bursts) and K-complexes on EEG, moderate arousal threshold, constitutes majority of sleep (1 mark)

• N3 (Slow-wave/deep sleep, 15-25% TST): High-amplitude delta waves (0.5-2 Hz, >75 μV), highest arousal threshold, functions include growth hormone release, tissue repair, memory consolidation, immune restoration, glymphatic clearance (1.5 marks)

REM Sleep (20-25% TST): Low-amplitude mixed-frequency EEG (paradoxical - resembles wakefulness), rapid conjugate eye movements, muscle atonia except diaphragm, dreaming occurs, functions include procedural and emotional memory consolidation, emotional regulation (1.5 marks)

Normal adult: 7-9 hours total sleep, >85% efficiency, 4-6 complete cycles per night (0.5 marks)

(b) Two-Process Model and SCN (5 marks)

Two-Process Model (Borbély) (2.5 marks):

• Process S (Homeostatic sleep drive): Sleep pressure accumulates during wakefulness, mediated by adenosine accumulation in basal forebrain from ATP metabolism, dissipates during sleep (especially N3), caffeine antagonises adenosine receptors to delay sleep

• Process C (Circadian rhythm): Approximately 24-hour cycle generated by suprachiasmatic nucleus, promotes wakefulness during day and sleep at night regardless of prior sleep, interacts with Process S to determine sleep timing and quality

Suprachiasmatic Nucleus (SCN) (2.5 marks):

• Located in anterior hypothalamus, ~20,000 neurons
• Master circadian pacemaker with intrinsic ~24.2-hour rhythm
• Entrained to light via retinohypothalamic tract (melanopsin-containing retinal ganglion cells)
• Molecular clock: CLOCK/BMAL1 positive regulators activate PER/CRY, which accumulate and inhibit CLOCK/BMAL1, creating 24-hour transcription-translation feedback loop
• Projects to multiple brain regions coordinating peripheral clocks
• Controls melatonin (pineal), cortisol (HPA axis), core temperature rhythms

(c) Propofol vs Physiological Sleep (4 marks)

Propofol-induced unconsciousness differs from physiological sleep (4 marks):

• Mechanism: Propofol enhances GABA-A receptor activity causing generalised cortical depression, whereas physiological sleep involves coordinated activation/inhibition of specific nuclei (VLPO sleep-promoting, orexin wake-promoting) in "flip-flop" switch (1 mark)

• EEG pattern: Propofol produces dose-dependent changes (alpha dominance → burst suppression → isoelectric) that do not follow normal N1→N2→N3→REM cycling; lacks sleep spindles and K-complexes of natural N2, lacks organised delta waves of N3 (1 mark)

• Sleep architecture: Propofol suppresses REM sleep and prevents normal sleep cycling; sedation cannot substitute for restorative functions of physiological sleep (memory consolidation, growth hormone release, immune restoration) (1 mark)

• Neurobiology: Natural sleep involves active processes (VLPO inhibition of arousal centres, orexin withdrawal), while propofol causes passive pharmacological depression of cortical activity without engaging normal sleep circuitry; lacks circadian integration (1 mark)

---

SAQ 2: Sleep Disruption and Interventions in ICU (15 marks)

Time: 18 minutes

Question:

A 62-year-old woman is Day 7 in ICU following severe community-acquired pneumonia requiring mechanical ventilation. She is now awake on minimal sedation but appears exhausted, with an irregular sleep-wake pattern and daytime somnolence. She is CAM-ICU negative.

(a) List the causes of sleep disruption in ICU, categorised by environmental, clinical, and pharmacological factors. (5 marks)

(b) Describe methods for assessing sleep quality in ICU patients, including their limitations. (4 marks)

(c) Outline a comprehensive sleep promotion strategy for this patient, referencing evidence for interventions. (6 marks)

---

Model Answer:

(a) Causes of Sleep Disruption in ICU (5 marks)

Environmental Factors (1.5 marks):

• Noise: ICU levels 50-80 dB (WHO recommends <35 dB); alarms 60-80 dB, conversations 50-65 dB; accounts for 11-17% of arousals
• Light: Continuous/inappropriate light (5-1400 lux at night vs <0.5 lux normal), suppresses melatonin, blue light most disruptive
• Temperature: Often outside thermoneutral zone

Clinical/Intervention Factors (2 marks):

• Nursing care activities: >50 interruptions per night (vital signs, repositioning, medications)
• Patient-ventilator dyssynchrony and ventilator settings
• Monitoring alarms (72-99% are false/non-actionable)
• Pain and discomfort
• Critical illness: Inflammatory cytokines (IL-1β, TNF-α, IL-6), sepsis, metabolic disturbances
• Circadian disruption from critical illness itself

Pharmacological Factors (1.5 marks):

• Benzodiazepines: Suppress N3 and REM, increase N2
• Propofol: Abnormal EEG, no normal sleep architecture
• Opioids: Suppress N3 and REM, cause central apnoea
• Corticosteroids: Disrupt sleep, suppress REM
• Vasopressors: Sympathetic activation
• Beta-blockers: Suppress melatonin, reduce REM

(b) Sleep Assessment Methods (4 marks)

Polysomnography (PSG) (1.5 marks):

• Gold standard: EEG, EOG, EMG, respiratory monitoring
• Limitations in ICU: Technical expertise required, electrical artifact, 30-50% uninterpretable, expensive, labour-intensive, atypical EEG patterns in critical illness

Actigraphy (1 mark):

• Wrist-worn accelerometer, infers sleep from movement
• Advantages: Non-invasive, continuous, multi-day monitoring
• Limitations: Overestimates sleep in immobile/sedated patients (60-80% PSG agreement), cannot assess architecture

Subjective Tools (1 mark):

• Richards-Campbell Sleep Questionnaire (RCSQ): 5 VAS scales, validated for ICU, <50/100 indicates poor sleep
• Limitations: Requires patient recall and ability to respond

Biomarkers and Other (0.5 marks):

• BIS monitor: Not validated for physiological sleep staging
• Melatonin/cortisol rhythms: Research use, serial sampling needed

(c) Comprehensive Sleep Promotion Strategy (6 marks)

Non-Pharmacological - Environmental (2 marks):

• Offer earplugs and eye mask nightly (RCT evidence: improved sleep quality, reduced delirium - Van Rompaey 2012, Hu 2015)
• Reduce nocturnal light to <10 lux from 22:00, use dim amber lighting if needed
• Quiet time protocol (22:00-06:00): Lower voice volume, reduce alarm volume, postpone non-urgent activities
• Cluster care activities to minimise interruptions
• Daytime bright light exposure (natural light or 2500+ lux phototherapy) to reinforce circadian rhythm

Non-Pharmacological - Clinical (2 marks):

• Optimise analgesia (multimodal, avoid excessive opioids that suppress REM)
• Early mobility and physical activity during daytime (part of ABCDEF bundle)
• Avoid daytime sleep >2 hours to maintain night-time sleep pressure
• Encourage day-night routine (awake during day, sleep at night)
• Ventilator optimisation: Consider assist-control mode overnight (PADIS conditional recommendation vs pressure support)

Pharmacological (2 marks):

• Consider melatonin 3-5 mg PO at 21:00-22:00 (Bourne 2008, Huang 2014 showed improved sleep quality; however, Pro-MEDIC 2020 showed no delirium benefit - PADIS: insufficient evidence for recommendation)
• Avoid benzodiazepines and Z-drugs (worsen architecture, delirium risk)
• If sedation needed, consider dexmedetomidine (SPICE III; produces more natural sleep-like state than propofol)
• Ramelteon 8 mg (selective MT1/MT2 agonist) may be considered if melatonin unavailable (Hatta 2014 showed delirium reduction in post-operative patients)
• Do NOT use diphenhydramine or antipsychotics for sleep (anticholinergic, no benefit for sleep)

Evidence: Kamdar et al. 2015 demonstrated multi-component bundle improved RCSQ scores by 20-30%

---

Viva Practice Questions

Viva Scenario 1: Normal Sleep Physiology

Stem: "Let's discuss sleep physiology. Tell me about the normal stages of sleep."

Expected Discussion (10 minutes):

---

Examiner: "Tell me about the normal stages of sleep."

Candidate: "Sleep consists of two fundamental states: Non-REM sleep, which has three stages (N1, N2, N3), and REM sleep. These alternate in approximately 90-120 minute cycles, with 4-6 cycles per night.

Stage N1 is light transitional sleep, comprising 2-5% of total sleep time, characterised by low-amplitude theta waves on EEG, slow rolling eye movements, and easy arousability.

Stage N2 is consolidated light sleep, comprising 45-55% of total sleep time, with characteristic sleep spindles (12-14 Hz bursts) and K-complexes on EEG.

Stage N3 is slow-wave or deep sleep, comprising 15-25% of total sleep time, with high-amplitude delta waves on EEG. This is the most restorative stage, responsible for growth hormone release, tissue repair, memory consolidation, and immune restoration.

REM sleep comprises 20-25% of total sleep time, with paradoxical low-amplitude mixed-frequency EEG resembling wakefulness, rapid eye movements, and muscle atonia. REM is important for procedural and emotional memory consolidation."

---

Examiner: "What regulates the timing of sleep?"

Candidate: "Sleep timing is regulated by the two-process model proposed by Borbély.

Process S is the homeostatic sleep drive that accumulates during wakefulness, mediated primarily by adenosine accumulation in the basal forebrain. Adenosine is a by-product of neuronal ATP metabolism and acts on A1 and A2A receptors to inhibit wake-promoting neurons. This pressure dissipates during sleep, particularly during N3.

Process C is the circadian rhythm generated by the suprachiasmatic nucleus in the anterior hypothalamus. The SCN has an intrinsic period of approximately 24.2 hours and is entrained to the light-dark cycle via the retinohypothalamic tract. It generates output signals including melatonin from the pineal gland and cortisol rhythms that promote sleep at night and wakefulness during the day.

These two processes interact to determine sleep timing, with highest sleep propensity when high sleep pressure coincides with the circadian trough."

---

Examiner: "Describe melatonin physiology."

Candidate: "Melatonin is synthesised in the pineal gland from tryptophan via serotonin. The rate-limiting enzyme is arylalkylamine N-acetyltransferase, which is inhibited by light.

Light detected by melanopsin-containing retinal ganglion cells signals to the SCN, which then inhibits the pineal via a sympathetic pathway, suppressing melatonin release. In darkness, this inhibition is removed and melatonin is synthesised.

Melatonin peaks at 02:00-04:00, reaching levels of 10-80 pg/mL, with daytime levels below 10 pg/mL. Its half-life is 30-60 minutes with hepatic metabolism.

Melatonin acts on MT1 and MT2 receptors in the SCN and other tissues. It decreases sleep latency, increases sleep efficiency, and acts as a circadian phase-entraining signal. It also has antioxidant and immunomodulatory properties."

---

Examiner: "How is sleep disrupted in ICU patients?"

Candidate: "ICU patients typically achieve only 2-4 hours of fragmented sleep per 24 hours, with up to 50% occurring during daytime hours. Sleep architecture is severely abnormal with predominantly N1/N2 sleep, minimal N3 or REM, and 20-57 arousals per hour.

Causes include environmental factors such as noise (50-80 dB in ICU vs recommended <35 dB), light exposure suppressing melatonin, and temperature dysregulation.

Clinical interventions cause disruption through patient care activities (>50 per night), mechanical ventilation and patient-ventilator dyssynchrony, and monitoring alarms.

Medications significantly affect sleep architecture: benzodiazepines suppress N3 and REM, propofol produces pharmacological unconsciousness rather than restorative sleep, and opioids suppress REM and cause central apnoea.

The underlying critical illness itself causes circadian disruption through inflammatory cytokines affecting sleep-wake regulation and the loss of normal melatonin and cortisol rhythms."

---

Examiner: "What are the consequences of sleep deprivation in critically ill patients?"

Candidate: "Sleep deprivation has significant multi-system effects in critically ill patients.

Cognitively, it impairs attention, working memory, and executive function. Importantly, there is a strong bidirectional relationship with delirium - sleep deprivation increases delirium risk 2-3 fold, and delirium further disrupts sleep.

Immunologically, sleep deprivation decreases NK cell activity by 35-45%, impairs phagocyte function, and shifts toward a pro-inflammatory state with elevated IL-6, TNF-α, and CRP. This may increase susceptibility to nosocomial infections and impair wound healing.

Metabolically, it decreases glucose tolerance, increases insulin resistance, and impairs growth hormone secretion which is important for tissue repair.

Cardiovascularly, it increases sympathetic tone, blood pressure, and heart rate, and promotes a pro-inflammatory, pro-thrombotic state.

For respiratory function, sleep deprivation may contribute to weaning failure through impaired respiratory muscle endurance and decreased ventilatory responses to hypoxia and hypercapnia."

---

Viva Scenario 2: Sleep Promotion in ICU

Stem: "A patient on Day 6 post-major surgery has irregular sleep patterns and appears exhausted. How would you approach sleep promotion?"

Expected Discussion (10 minutes):

---

Examiner: "How would you assess sleep in this patient?"

Candidate: "I would use multiple assessment methods recognising their limitations.

First, clinical observation and patient interview if possible - asking about subjective sleep quality, though recognising patients may be unable to accurately recall sleep.

The Richards-Campbell Sleep Questionnaire is validated for ICU and assesses five domains including sleep depth, latency, awakenings, ability to return to sleep, and overall quality. A score below 50/100 indicates poor sleep perception.

Actigraphy using a wrist-worn accelerometer can provide continuous monitoring but may overestimate sleep in immobile patients.

Polysomnography is the gold standard but is resource-intensive and often impractical for clinical use. Additionally, up to 30-50% of ICU recordings may be uninterpretable due to artifact and atypical EEG patterns.

I would also assess for circadian rhythm disruption by observing the patient's 24-hour pattern - is there consolidation of sleep at night or is it scattered throughout the day?"

---

Examiner: "Describe a comprehensive sleep promotion strategy."

Candidate: "I would implement a multi-component sleep bundle based on PADIS guidelines and available evidence.

For environmental modifications: offer earplugs and eye masks nightly, which have randomised controlled trial evidence showing improved sleep quality and reduced delirium. Implement a quiet time protocol from 22:00-06:00 with reduced alarm volumes, lowered voices, and postponement of non-urgent activities. Reduce nocturnal light to below 10 lux and use dim amber lighting if illumination is needed. During daytime, ensure adequate bright light exposure to reinforce circadian rhythms.

For clinical care organisation: cluster care activities to create uninterrupted sleep periods, question necessity of overnight vital signs if continuous monitoring is available, and delay non-urgent medications to daytime. Optimise ventilator settings - PADIS provides a conditional recommendation for assist-control ventilation over pressure support at night. Ensure adequate analgesia using multimodal approaches.

For daytime activity: promote mobility and physical activity, discourage daytime naps longer than 2 hours, and establish a clear day-night routine.

For pharmacological interventions: consider melatonin 3-5 mg at 21:00, though evidence is limited. PADIS does not recommend routine pharmacological sleep aids due to insufficient evidence. Avoid benzodiazepines and anticholinergics which worsen sleep architecture and increase delirium risk."

---

Examiner: "What is the evidence for melatonin and ramelteon in ICU?"

Candidate: "The evidence for melatonin in ICU is mixed and currently does not support routine use.

Bourne et al. in 2008 showed that 10 mg melatonin improved sleep quality measured by BIS in a small RCT of 24 tracheostomised patients. Huang et al. in 2014 showed improved subjective sleep with 5 mg melatonin in 60 ICU patients.

However, larger trials have been disappointing. Nishikimi et al. in 2018 found no effect on delirium or sleep in sepsis patients. The Pro-MEDIC trial by Wibrow et al. in 2020, a well-designed Australian RCT of 210 patients, showed no benefit for delirium prevention with melatonin 4 mg.

Meta-analyses suggest a trend toward improved sleep quality but no consistent effect on delirium. The PADIS guidelines state there is insufficient evidence to recommend melatonin for sleep promotion.

For ramelteon, a selective MT1/MT2 agonist, Hatta et al. in 2014 showed remarkable reduction in delirium from 32% to 3% in elderly post-operative patients. However, Nishikimi et al. in 2018 found no benefit in sepsis. The variable results may relate to patient population, with elderly post-surgical patients potentially benefiting more than septic patients.

Current recommendations are that melatonin or ramelteon may be considered in selected patients but should not be routine. Neither addresses the underlying causes of sleep disruption in ICU."

---

Examiner: "How does dexmedetomidine affect sleep?"

Candidate: "Dexmedetomidine is an alpha-2 agonist with unique effects on sleep that distinguish it from other sedatives.

Unlike benzodiazepines and propofol which work through GABA-A receptor enhancement causing generalised cortical depression, dexmedetomidine acts on alpha-2 receptors in the locus coeruleus. This inhibits noradrenergic wake-promoting neurons while allowing natural sleep circuitry to function.

EEG studies show that dexmedetomidine produces activity resembling Stage N2 sleep with preserved sleep spindles. It preserves some REM sleep, unlike propofol and benzodiazepines which suppress REM. Patients sedated with dexmedetomidine are more arousable and can cooperate with commands - often called 'cooperative sedation'.

The SPICE III trial, an Australian/New Zealand-led multicentre RCT, compared early dexmedetomidine-based sedation to usual care. While there was no mortality benefit, there was a signal toward reduced delirium.

When sedation is needed, dexmedetomidine may be preferred for its more physiological sleep-like state compared to propofol or benzodiazepines. However, it should be recognised that even dexmedetomidine-induced sedation does not fully replicate natural restorative sleep."