Validation of Various Sleep Assessment Tools in SICU

Sleep is a periodic, reversible state of reduced consciousness, and response to external stimuli. A total sleep duration of normal human sleep is approximately 7-8 hours per night. Sleep deprivation is common in critical patients and it can cause impair consolidation of memory, cognitive function, metabolic function, immune, neurological and respiratory system as well as worsen the quality of life after discharge. The impact of sleep deprivation in critically ill patients is gaining attention as it links between sleep loss and delirium. It has been demonstrated that reducing sleep disturbance could attenuate the development of delirium in ICU patients. However, sleep evaluation is only personal perception and the previous study expressed that medical personnel evaluation still showed some error in approximately 56%. There are various methods for sleep monitoring, in which the most commonly mentioned methods include polysomnography, actigraphy, and the Richards-Campbell Sleep Questionnaire (RCSQ). Polysomnography is considered as the gold standard for monitoring the quantity and quality of sleep. Although, it is difficult to apply in critical care as it requires expensive setup and maintenance and is prone to patient dislodgment and electrical interference leading to low quality of signals and missed interpretation.

Nowadays, actigraphy has been used to assess patterns of rest and activity, which can distinguish sleep from wakefulness. The advantage of actigraphy is easy to use and non-invasive, can be used by non-specialists, and allows continuous measurement over days to weeks. On the other hand, its disadvantages are such as neuromuscular weakness and increases risk of overestimating sleep quantity as periods of inactivity scored as sleep. However, the technology has been improved recently leading to more accurate sleep measurement. Although it has been shown that actigraphy is a good alternative sleep measure in non-ICU patients, the accuracy of actigraphy is unknown in ICU patients.

For subjective tool for evaluating sleep, the Richards-Campbell Sleep Questionnaire (RCSQ) is one of the most commonly used questionnaires. It contains a simple five-item questionnaire which are sleep depth, ability to fall asleep, number of awakening, ability to fall asleep when awake and quality of sleep. The RCSQ shows a good correlation with polysomnography in sleep efficacy index. The advantages of RCSQ are easy-to-use, inexpensive and providing good quality of sleep assessment. It has been widely used in the United States and many countries. Moreover, it has been translated to many languages, such as Spanish, Swedish, German, and Japanese; and the translated version of RCSQ can be used as an alternative sleep measurement to polysomnography. Currently, our research group is translating the RCSQ into Thai version by using standard guidelines on the translation process. In spite of this, The Thai version of RCSQ (T-RCSQ) has not been validated with polysomnography before.

The aims of our study are to validate the accuracy of the T-RCSQ for sleep measurement compared to polysomnography, which is considered as the gold-standard and to access the accuracy of actigraphy for monitoring sleep quality and quantity compared to polysomnography in Thai critically ill patients admitted to surgical intensive care unit (SICU).

Inclusion criteria

• Patient whose age ≥ 18 years old.
• Patients who are anticipated to stay in surgical intensive care unit at least one full night (from 8 p.m. to 6 a.m. next day).
• Patients who can answer the questionnaires and can communicate, read and write in Thai.

Exclusion criteria

• Patients with severe cognitive impairment, dementia, Richmond Agitation Sedation Scale (RASS) score of less than -2 or more than +1 or those receiving neuromuscular blocking agents.
• Patients with pathological lesions that preclude applying electrodes and sensors of polysomnography or applying actigraphy.
• Patients with high acuity of illness whose ICU survival is not expected.