Propofol for the promotion of sleep in adults in the intensive care unit

Sharon R Lewis; Oliver J Schofield‐Robinson; Phil Alderson; Andrew F Smith

doi:10.1002/14651858.CD012454.pub2

. 2018 Jan 8;2018(1):CD012454. doi: 10.1002/14651858.CD012454.pub2

Propofol for the promotion of sleep in adults in the intensive care unit

Sharon R Lewis ^1,^✉, Oliver J Schofield‐Robinson ¹, Phil Alderson ², Andrew F Smith ³

Editor: Cochrane Emergency and Critical Care Group

PMCID: PMC6353271 PMID: 29308828

Abstract

Background

People in the intensive care unit (ICU) experience sleep deprivation caused by environmental disruption, such as high noise levels and 24‐hour lighting, as well as increased patient care activities and invasive monitoring as part of their care. Sleep deprivation affects physical and psychological health, and people perceive the quality of their sleep to be poor whilst in the ICU. Propofol is an anaesthetic agent which can be used in the ICU to maintain patient sedation and some studies suggest it may be a suitable agent to replicate normal sleep.

Objectives

To assess whether the quantity and quality of sleep may be improved by administration of propofol to adults in the ICU and to assess whether propofol given for sleep promotion improves both physical and psychological patient outcomes.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 10), MEDLINE (1946 to October 2017), Embase (1974 to October 2017), the Cumulative Index to Nursing and Allied Health Literature (CINAHL) (1937 to October 2017) and PsycINFO (1806 to October 2017). We searched clinical trials registers for ongoing studies, and conducted backward and forward citation searching of relevant articles.

Selection criteria

We included randomized and quasi‐randomized controlled trials with adults, over the age of 16 years, admitted to the ICU with any diagnoses, given propofol versus a comparator to promote overnight sleep. We included participants who were and were not mechanically ventilated. We included studies that compared the use of propofol, given at an appropriate clinical dose with the intention of promoting night‐time sleep, against: no agent; propofol at a different rate or dose; or another agent, administered specifically to promote sleep. We included only studies in which propofol was given during 'normal' sleeping hours (i.e. between 10 pm and 7 am) to promote a sleep‐like state with a diurnal rhythm.

Data collection and analysis

Two review authors independently assessed studies for inclusion, extracted data, assessed risk of bias and synthesized findings.

Main results

We included four studies with 149 randomized participants. We identified two studies awaiting classification for which we were unable to assess eligibility and one ongoing study.

Participants differed in severity of illness as assessed by APACHE II scores in three studies and further differences existed between comparisons and methods. One study compared propofol versus no agent, one study compared different doses of propofol and two studies compared propofol versus a benzodiazepine (flunitrazepam, one study; midazolam, one study). All studies reported randomization and allocation concealment inadequately. We judged all studies to have high risk of performance bias from personnel who were unblinded. We noted that some study authors had blinded study outcome assessors and participants for relevant outcomes.

It was not appropriate to combine data owing to high levels of methodological heterogeneity.

One study comparing propofol with no agent (13 participants) measured quantity and quality of sleep using polysomnography; study authors reported no evidence of a difference in duration of sleep or sleep efficiency, and reported disruption to usual REM (rapid eye movement sleep) with propofol.

One study comparing different doses of propofol (30 participants) measured quantity and quality of sleep by personnel using the Ramsay Sedation Scale; study authors reported that more participants who were given a higher dose of propofol had a successful diurnal rhythm, and achieved a greater sedation rhythmicity.

Two studies comparing propofol with a different agent (106 participants) measured quantity and quality of sleep using the Pittsburgh Sleep Diary and the Hospital Anxiety and Depression Scale; one study reported fewer awakenings of reduced duration with propofol, and similar total sleep time between groups, and one study reported no evidence of a difference in sleep quality. One study comparing propofol with another agent (66 participants) measured quantity and quality of sleep with the Bispectral Index and reported longer time in deep sleep, with fewer arousals. One study comparing propofol with another agent (40 participants) reported higher levels of anxiety and depression in both groups, and no evidence of a difference when participants were given propofol.

No studies reported adverse events.

We used the GRADE approach to downgrade the certainty of the evidence for each outcome to very low. We identified sparse data with few participants, and methodological differences in study designs and comparative agents introduced inconsistency, and we noted that measurement tools were imprecise or not valid for purpose.

Authors' conclusions

We found insufficient evidence to determine whether administration of propofol would improve the quality and quantity of sleep in adults in the ICU. We noted differences in study designs, methodology, comparative agents and illness severity amongst study participants. We did not pool data and we used the GRADE approach to downgrade the certainty of our evidence to very low.

Plain language summary

Propofol at night to improve sleep in the intensive care unit

Background

Lack of sleep affects a person's physical and mental health and, for people who are critically ill, sleep is thought to improve healing and survival. People in the intensive care unit (ICU) experience poor sleep. Many factors contribute to poor sleep including high noise levels, 24‐hour lighting and intrusive patient care activities. Propofol is an anaesthetic agent given by infusion into a vein that is sometimes used to sedate people who are in the ICU. In this review, we looked at studies in which propofol was given to adults at night‐time to improve the quality and quantity of sleep.

Study characteristics

The evidence is current to October 2017. We included four randomized controlled studies (clinical studies where people are randomly put into one of two or more treatment groups) with 149 participants in the review. Two studies are awaiting classification (because we could not assess their eligibility) and one study is ongoing. All participants were critically ill and were in the ICU.

Key results

We did not combine the results from the studies because of differences in comparison (called control) treatments and study design. One study compared propofol with no agent. This study used polysomnography (which records brain waves, oxygen level in blood, heart rate, breathing, and eye and leg movements) to measure sleep quality and quantity. It reported no improvement in duration of sleep with propofol but participants woke up less often and for shorter lengths of time and described their sleep quality as being improved with propofol. One study compared a higher dose of propofol at night described as additional night sedation, with a constant day‐time and night‐time dose. This study used the Ramsay Sedation Scale (which is normally used by anaesthetists to assess how easily a person is roused) and reported that participants appeared to have an improved sleep rhythm. Two studies compared propofol with benzodiazepines (a tranquilizing medicine; flunitrazepam in one study and midazolam in one study). These studies used the Pittsburgh Sleep Diary and the Hospital Anxiety and Depression Scale to measure quantity and quality of sleep. The study with flunitrazepam reported fewer awakenings of reduced duration with propofol but similar total sleep time in each group and the study with midazolam reported no difference in sleep quality. The study with flunitrazepam also measured sleep with Bispectral Index (used by anaesthetists to assess depth of anaesthesia) and reported longer time in deep sleep, with fewer awakenings. The study with midazolam reported higher levels of anxiety and depression in both groups, and no difference when participants were given propofol. No study reported on side effects.

Quality of evidence

We judged the evidence to be very low quality. We found only four small randomized controlled studies and the results of the studies were not consistent. We noted differences in illness severity of participants and the medicines that were compared with propofol in the included studies. Measuring quality of sleep using diaries, questionnaires and scoring systems is based on, or is influenced by, personal feelings or opinions, and we were concerned that staff and participants were aware which medicine they had been given; we believed that this could have influenced the results. Only one study used polysomnography, which is the most appropriate unbiased measurement tool for sleep.

Conclusions

We were unable to collect sufficient evidence to determine whether propofol given at night to adults in the ICU improves quality and quantity of their sleep, as a way of helping recovery.

Summary of findings

Background

Description of the condition

It is accepted that sleep is an essential requirement for good health. Sleep has a restorative function and for critically ill people is thought to improve healing and survival (Tembo 2009).

Sleep naturally follows a circadian rhythm of approximately 24 hours, in a pattern of once a day. Each period of sleep consists of phases lasting 90 minutes, with typical sleep 'architecture' involving a period of rapid eye movement (REM) and non‐rapid eye movement (NREM). The NREM stage is subdivided into three phases, now labelled as N1, N2 and N3 (also described as 'slow wave patterns'). The REM stage accounts for 15% to 20% of sleep time, and N2 of NREM for approximately 50% (Matthews 2011; Schupp 2003; Silber 2007).

It is likely that people in hospital will be subject to sleep disturbances; this is particularly likely for people in the intensive care unit (ICU). These people are critically ill with diagnoses such as respiratory insufficiency or failure, need for postoperative management, ischaemic heart disorder, sepsis and heart failure (Society of Critical Care Medicine). People may require specialist support after elective surgery or may be emergency admissions following medical events or trauma (e.g. with multiple injuries after a road traffic accident) (Intensive Care Foundation).

In the ICU, the ratio of staff to patients is higher than in general wards, and the environment typically includes 24‐hour lighting, a constant level of noise and more frequent patient care activities (measuring blood pressure, pulse and temperature; taking blood samples; administering medications, etc.) than in general wards. Many people (up to 40%; Esteban 2000) are mechanically ventilated and are subject to invasive procedures such as tracheal intubation and use of nasogastric tubes (Esteban 2000). In addition, people in the ICU have critical conditions that involve pain, anxiety and stress (Kamdar 2012a).

Patients are often prescribed drugs that further contribute to sleep loss. For example, drugs such as benzodiazepines are given for essential sedation (particularly to people who are mechanically ventilated) to relieve discomfort and stress. These agents alter sleep architecture, such that N2 of NREM is longer than normal, which may increase sleep time. However, they also reduce essential REM and N3 phases of sleep (Bourne 2004). Similarly, opioids are commonly used for analgesia in the ICU, and studies report that these, even when given in low doses and to healthy volunteers, reduce the amount of deep sleep by up to 50% (Dimsdale 2007; Grounds 2014).

People perceive that the quality of their sleep in the ICU is disrupted by frequent awakenings and increased daytime sleep (Freedman 1999). This perception is supported by trials that assessed sleep by using objective measures. Polysomnography (PSG) readings, which use a variety of channels to measure electrical activity of the heart, as well as muscle tension, airflow and eye movement, can be used to assess sleep. People in the ICU have alterations to their circadian rhythm, with up to 50% of sleep occurring during the day and with sleep arousals occurring as often as 39 times per hour in people who are mechanically ventilated (Parthasarathy 2004). Changes to sleep 'architecture' are significant, with reductions in both REM and N3 sleep (Cooper 2000; Drouot 2008).

Empirical evidence on immediate and long‐term physical consequences of sleep deprivation for people in the ICU is limited, but data suggest that sleep loss in healthy study participants can result in physical alterations to the immune system, as well as changes in metabolism, nitrogen balance, and the ventilatory and cardiovascular systems (Kamdar 2012a; Matthews 2011; Pisani 2015; Weinhouse 2006). For example, after loss of only one night's sleep, biomarkers are released that are present in people with coronary artery disease (Sauvet 2010), although no longitudinal studies have demonstrated that sleep disturbance in the ICU results in increased cardiovascular mortality (Kamdar 2012a). Psychological consequences associated with sleep loss, such as cognitive dysfunction, depression and poor quality of life measures, affect hospital length of stay, morbidity and mortality in critically ill people (e.g. Ely 2004). Psychological symptoms, such as post‐traumatic stress, may persist for several months after ICU discharge (Eddleston 2000; Figueroa‐Ramos 2009; Kamdar 2012a; Matthews 2011).

Description of the intervention

Non‐pharmacological interventions, such as noise and light reduction strategies (e.g. earplugs, eye masks; Richardson 2007), have been studied specifically in the ICU, and some have been shown to improve the quality of sleep. One Cochrane systematic review has explored the effectiveness of these various strategies (Hu 2015).

The present review aimed to look at the anaesthetic agent propofol as administered to adults in the ICU to promote sleep. This agent has the benefit of a rapid onset and offset, with a duration of action of 10 to 15 minutes (Brown 2001). It is given as a continuous intravenous infusion to induce and then maintain a state of general anaesthesia. Given at different doses, it can initiate different levels of consciousness on a continuum. It is currently among agents used in the ICU to maintain sedation (Grounds 2014).

How the intervention might work

Murphy 2011 used a high‐density electroencephalogram (EEG) on healthy participants anaesthetized with propofol and noted similarities to readings from PSG assessments of normal sleep. In particular, the slow waves produced after administration of propofol were similar (although not identical) in appearance to those seen during NREM of normal sleep. Results of animal studies suggested that recovery from sleep deprivation after propofol use is similar to naturally occurring sleep patterns (Tung 2004).

These study results suggest that propofol may be a suitable agent for replicating normal sleep. Given its rapid onset and offset, propofol can be administered at time points to maintain a person's usual diurnal rhythm and is a familiar drug in the ICU setting.

Why it is important to do this review

In 2014, the James Lind Alliance, a priority setting organization that works with patients, carers and clinicians to establish research priorities for health care, published its top 10 priorities for research in the intensive care setting (www.jla.nihr.ac.uk/priority‐setting‐partnerships/intensive‐care/top‐10‐priorities/). Among these are research topics relevant to enhancing patient comfort in the ICU (including minimizing pain, discomfort, agitation and anxiety), preventing physical consequences of critical illness and providing psychological support for patients.

Although sleep disruption may affect many people in hospital, those in the ICU are particularly vulnerable to disturbances that may subsequently lead to physical and psychological consequences, such as those identified above. Uncertainties remain about the benefits of pharmacological agents given to promote patient sleep in the ICU. It is unknown whether effectively promoting sleep will improve patient outcomes and provide immediate and long‐term clinical benefits. This review addressed the James Lind Alliance priority targets by assessing the effectiveness of propofol for sleep promotion in the ICU setting.

Objectives

Methods

Criteria for considering studies for this review

Types of studies

We included randomized controlled trials (RCTs) and quasi‐randomized controlled trials (e.g. studies in which participants were assigned by alternation, date of birth or medical record number).

Types of participants

We included adults, over 16 years of age, who were admitted to any ICU as an elective or emergency patient, with medical, trauma or surgical primary diagnosis. We did not limit types of participants by severity of the condition. We included participants who were and were not mechanically ventilated.

Types of interventions

We included studies that compared the use of propofol, given at an appropriate clinical dose with the intention of promoting night‐time sleep, against:

no agent;
propofol at a different rate or dose; or
another agent, administered specifically to promote sleep.

We anticipated difficulties with terms used by study authors in this review, in particular, the terms 'sleep' and 'sedation.' Participants may be given propofol during the day with the aim of reducing consciousness as part of routine treatment in the ICU; this reduction in consciousness may be referred to as 'sedation.' We included only studies in which propofol was given during 'normal' sleeping hours (i.e. between 10 pm and 7 am) to promote a sleep‐like state with a diurnal rhythm.

Participants given propofol for daytime sedation may also be given propofol as an intervention agent to promote sleep. We included only studies in which study authors specified that propofol was given at night‐time at a higher dose and with a different stated aim, for example, to promote a diurnal rhythm. We did not specify the dose or rate of administration; a 'lower' daytime dose of propofol given to achieve sedation may differ between participant groups, depending on the severity of the condition of participants and whether participants were mechanically ventilated.

We anticipated that study authors may have used scales to determine levels of participant consciousness, and we planned to report study authors' interpretations of these scales. For example, participants may be sedated at level 3 on the Ramsay Sedation Scale ("Patient responds to commands only") and may be in a sleep‐like state at level 4 or 5 on this scale ("Patient exhibits brisk response to light glabellar tap or loud auditory stimulus" or "Patient exhibits a sluggish response to light glabellar tap or loud auditory stimulus") (Ramsay Sedation Scale).

It is reasonable that study participants may routinely be given other agents that have a sedative effect, for example, benzodiazepines or dexmedetomidine. We planned to record this information at the data collection stage and assess whether it was balanced between study arms. We planned to include these agents in the comparison 'propofol versus another agent, administered specifically to promote sleep' only if they had been given to all participants randomized to receive the comparative agent as part of the study protocol, rather than routinely to all participants. We planned to include multi‐arm studies, for example, those that compared different doses of propofol versus a different agent, only if each agent was used as a comparative agent by randomized participants as part of the study protocol.

Types of outcome measures

We were interested in quantity and quality of sleep. The experience of sleep may not always be representative of objectively measured sleep, and given that people perceive that they have disrupted sleep while in the ICU (Freedman 1999), we included this outcome, regardless of whether validated scales had been used for measurement. An added issue for this outcome is that critically ill people may have limited or no ability to communicate. Therefore, we were equally interested in the perceptions of carers and family members, who may have had an impression of sleep from the bedside, and in subjective measures used by personnel to assess sleep. We included assessments of sleep that had been performed at the end of follow‐up, as defined by study authors, for example, in the morning or during the daytime that followed administration of the intervention. We included assessments that were completed using scales, whether validated or not and modified for each user, such as the Richards‐Campbell Sleep Questionnaire (Richards 2000), or through compilation of sleep diaries, such as the Pittsburgh Sleep Diary (Monk 1994).

We planned also to report the quantity and quality of sleep as measured by objective equipment. In particular, PSG is considered the most accurate and objective tool that can be used to measure sleep and identify sleep disorders (Beecroft 2008). PSG measurements can be analysed to detect sleep onset, sleep efficiency and length of sleep stages, as well as irregularities, such as apnoea and interrupted sleep. However, we planned also to accept measurements obtained when other tools had been used to record sleep activity (Beecroft 2008; Benini 2005; Elliott 2013); 'actigraphy' is a wristband‐style tool that measures gross motor activity and is analysed to score total sleep time, sleep efficiency and awakenings; Bispectral Index (BIS) is typically used to calculate depth of anaesthesia through interpretation of an EEG reading; and measures of EEG alone can be used to interpret electrical brain activity as sleep time. We acknowledge that collection and interpretation of data using objective equipment may be problematic in critically ill people because manual methods having poor inter‐rater reliability (Ambrogio 2008). We planned to use interpretations of these measurements as reported by each study author to define the quantity and quality of sleep.

Our aim was to establish not only whether propofol improves sleep but whether improvement in sleep leads to better patient outcomes. This is reflected in our secondary outcome measure, which considered potential physical and psychological consequences of sleep loss, although we acknowledge that it may not be possible to ascertain whether a reduction in such events is directly attributable to improved sleep. We planned to assess physical consequences of sleep loss by collecting data from studies that reported the number of participants who had experienced an adverse event during follow‐up, as defined by study authors. We planned to include all adverse events as defined by study authors and diagnosed by clinicians at study level, such as cardiovascular or respiratory events and illness resulting from immune deficiency. We assessed psychological consequences of sleep loss by collecting data from studies that reported the number of participants who had been given a diagnosis of anxiety or depression, or both, by using validated assessment tools during follow‐up, as defined by study authors.

Primary outcomes

Quantity and quality of sleep as measured through reports of participants or family members or by personnel assessments.
Quantity and quality of sleep as measured by PSG, actigraphy, BIS or EEG.

Secondary outcomes

Anxiety or depression, or both, as measured using validated tools, such as the Hospital Anxiety and Depression Scale (HADS) (Zigmond 1983).
Adverse events (such as cardiovascular events, respiratory events or illness resulting from immune deficiency).

Search methods for identification of studies

Electronic searches

We identified RCTs through literature searching with systematic and sensitive search strategies as outlined in Chapter 6.4 of the Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2011). We applied no restrictions to language or publication status.

We searched the following databases for relevant trials:

Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 10);
MEDLINE (OvidSP, 1946 to 3 October 2017);
Embase (OvidSP, 1974 to 3 October 2017);
Cumulative Index to Nursing and Allied Health Literature (CINAHL) (EBSCO, 1937 to 3 October 2017);
PsycINFO (EBSCO, 1887 to 3 October 2017).

We developed a subject‐specific search strategy in MEDLINE and used that as the basis for the search strategies in the other listed databases. The search strategy was developed in consultation with the Information Specialist. Search strategies can be found in Appendix 1; Appendix 2; Appendix 3; Appendix 4; and Appendix 5.

We scanned the following trial registries for ongoing and unpublished trials (2 June 2017):

World Health Organization International Clinical Trials Registry Platform (WHO ICTRP; www.who.int/ictrp/en/);
International Standard Randomised Controlled Trial Number (ISRCTN; www.isrctn.com/);
ClinicalTrials.gov (clinicaltrials.gov/).

Searching other resources

We carried out citation searching of identified included studies in Web of Science (apps.webofknowledge.com) on 13 June 2017 and conducted a search of grey literature through 'Opengrey' (www.opengrey.eu./) on 2 June 2017. We carried out backward citation searching of key reviews identified from the searches. We did not need to contact study authors or organizations.

Data collection and analysis

Two review authors (SL and OSR) independently carried out all data collection and analyses before comparing results and reaching consensus. We did not require use of a third review author to resolve conflicts.

Selection of studies

We used reference management software to collate search results and to remove duplicates (Endnote).

We used Covidence software to screen results of the search from titles and abstracts (Covidence), and identified potentially relevant studies by using this information alone. We sourced the full texts of all potentially relevant studies and considered whether they met the inclusion criteria (see Criteria for considering studies for this review). We planned to include abstracts at this stage only if they provided sufficient information and relevant results that included denominator figures for each intervention or comparison group.

We recorded the number of papers retrieved at each stage and reported this information using a PRISMA flow chart (Liberati 2009). We reported brief details of closely related papers excluded from the review.

Data extraction and management

We used a Cochrane template data extraction form to extract the following data from individual studies (Appendix 6).

Methods: type of study design; setting; dates of study and funding sources.
Participants: number of participants randomized to each group; baseline characteristics (to include Acute Physiology and Chronic Health Evaluation (APACHE) II scores, mechanical ventilation status and mode of ventilation, length of time in ICU before study commencement, and concomitant medications).
Interventions: details of intervention and comparison agents (to include dose and timing).
Outcomes: study outcomes as measured and reported by study authors (to include types of assessment tools, methods of data synthesis, units of measure and length of follow‐up).
Outcome data: results of outcome measures.

We considered the applicability of information obtained from individual studies and the generalizability of data to our intended study population (i.e. the potential for indirectness in our review).

If we identified associated publications from the same study, we planned to create a composite dataset from all eligible publications.

Assessment of risk of bias in included studies

We assessed study quality, study limitations and extent of potential bias by using the Cochrane 'Risk of bias' tool (Higgins 2011). We considered the following domains.

Random sequence generation (selection bias).
Allocation concealment (selection bias).
Blinding of participants and personnel (performance bias).
Blinding of outcome assessment (detection bias).
Incomplete outcome data (attrition bias).
Selective outcome reporting (reporting bias).
Other sources of bias (e.g. use of concomitant drugs).

Blinding to intervention and control agents may be feasible if agents are prepared in coded containers, for example, by an independent pharmacist; lack of blinding of personnel may introduce risk of bias. Successful blinding of participants to group allocation would be possible and would reduce performance and detection blinding. It is also feasible that outcome assessors could be blinded to group allocation to reduce bias. As participants are critically ill, rates of mortality and withdrawal of consent may be higher in the studies included in this review. Therefore, we paid particular attention to reasons given for losses, whether losses were related to the intervention or to chance alone and whether losses were comparable between groups. If participants received concomitant medication (e.g. morphine), we considered whether that medication could affect sleep and whether the concomitant medication was comparable between study groups. We addressed other potential biases in the included studies on an individual basis.

For each domain, two review authors (SL and OSR) independently used one of three measures (low, high or unclear) to judge whether study authors made sufficient attempts to reduce bias. We recorded this information in the 'Risk of bias' tables and presented a summary 'Risk of bias' figure.

Measures of treatment effect

We planned to collect information on adverse events as dichotomous data (number of participant events per group). We reported the psychological consequences of sleep deprivation as dichotomous or continuous data (e.g. number of participant events per group, mean scores per group on a scale measuring anxiety). We anticipated that measures of participant‐reported outcomes may differ for each study, depending on the scales used, as may objective measures of quantity and quality of sleep.

Unit of analysis issues

If multi‐arm studies compared more than one relevant intervention versus a control (e.g. no agent), we planned to include both intervention groups but split the data for the comparison or control group (using a 'halving' method), as recommended by Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2011).

Dealing with missing data

We planned to assess whether all measured outcomes had been reported by study authors by comparing, wherever possible, published reports with protocols or clinical trial registration documents that had been prospectively published.

We assessed whether all randomized participants were included in outcome data. We planned to contact authors to request any missing outcome data that were not explained. If we were unable to obtain these data, we planned to report the data only as they were presented in the published report, to include intention‐to‐treat data. We planned not to combine unexplained incomplete data in meta‐analysis but to report these data narratively.

We planned to discuss the potential impact of missing data on the finding of our review in the 'Discussion' section.

Assessment of heterogeneity

We assessed whether our results showed evidence of inconsistency by considering heterogeneity. We planned to include all adult ICU participants with variation in severity of illness, as well as the potential for three different comparisons to propofol. Therefore, we anticipated likely heterogeneity between studies and assessed clinical and methodological heterogeneity by comparing similarities between participants, interventions and outcomes in our included studies, using information collected during the data extraction phase. We planned to complete meta‐analyses only for studies that were clinically and methodologically similar.

We planned to assess statistical heterogeneity by calculating the Chi² or the I² statistics. We planned to judge an I² statistic above 60% and a Chi² P value of 0.05 or less to indicate moderate to substantial statistical heterogeneity (Higgins 2011).

As well as looking at statistical results, we planned to consider point estimates and overlap of confidence intervals (CIs). If CIs overlap, then results are more consistent. However, combined studies may show a large consistent effect with significant heterogeneity. Therefore, we planned to interpret heterogeneity with caution (Guyatt 2011a).

Assessment of reporting biases

We attempted to source published protocols for each of our included studies by using clinical trials registers. We compared the published protocol of one study with published study results to assess the risk of selective reporting bias (Kondili 2012).

If we identified sufficient studies (i.e. more than 10 studies) (Higgins 2011), we planned to generate a funnel plot to assess risk of publication bias in the review; an asymmetrical funnel plot may indicate potential publication of only positive results (Egger 1997).

Data synthesis

We planned to complete meta‐analyses of outcomes when comparable effect measures were available from more than one study, and only when measures of clinical and statistical heterogeneity indicated that pooling of results was appropriate. We planned to use the statistical calculator in Review Manager 5 (RevMan 2014).

We anticipated that our primary outcome of subjective sleep measures would collect data from different sources (participants, carers and personnel). As some evidence suggests that nurses' assessment of sleep may differ from that of patients (Kamdar 2012b), we planned not to combine participant‐, family‐ and personnel‐reported assessments of sleep quality but to report these separately. In addition, we anticipated that subjective sleep assessment tools may not be comparable between studies. If sleep assessment tools included categories of sleep assessment, such as no sleep, minimal sleep, moderate sleep and majority sleep, we planned to split the data into dichotomous results by comparing the number of people reporting moderate and majority sleep versus the number reporting minimal and no sleep. We planned to combine data across assessment tools if data could be split into equivalent categories; otherwise, we presented a descriptive summary of the results of each study.

Similarly, we anticipated that our primary outcome of objective sleep assessment may be measured using different tools that were not comparable. Unless data were reported with the same tools and with the same measurements, such as mean score for BIS or mean length of each sleep stage for PSG, we presented a descriptive summary of the results of each study.

For dichotomous outcomes, we planned to calculate the odds ratio by using summary data presented for each trial. We planned to use the Mantel‐Haenszel effects model, unless events were extremely rare (one per 1000), in which case we planned to use Peto (Higgins 2011). For continuous outcomes, for example, PSG readings, we planned to use mean differences. We planned to use a random‐effects statistical model, which allows for the assumption that included studies may estimate different, but related, intervention effects.

We planned to conduct separate analyses for each comparison type (i.e. propofol versus no agent, propofol versus a different dose of the same agent and propofol versus a different agent).

We planned to calculate 95% CIs and to use a P value of 0.05 or below to judge whether a result was statistically significant.

We planned to consider whether results of analyses were imprecise by assessing the CI around an effect measure; a wide CI would suggest a higher level of imprecision in the results. Inclusion of a small number of studies may also reduce precision (Guyatt 2011b).

Subgroup analysis and investigation of heterogeneity

We planned to assess possible reasons for heterogeneity by performing subgroup analyses. We planned to consider the severity of the condition of participants in each study; participants with a more severe condition may already be subject to increased sleep disruption. Similarly, we planned to consider whether participants had a different outcome according to mechanical ventilation status during the study period. We planned to consider the age of participants; older people (aged over 65 years) have an altered sleep pattern, which includes increased difficulty falling asleep with more awakenings and shorter total sleep time (Ancoli‐Israel 2009); therefore, data on sleep outcomes may be different for members of this age group than for younger people. In summary, planned subgroups were:

severity of the health condition based on APACHE II scores (Knaus 1985) (or comparable severity measures): APACHE II scores less than 25, 25 to 35, greater than 35;
mechanically ventilated participants versus participants not mechanically ventilated;
participants aged 65 years or older versus participants younger than 65 years; and
participants in a surgical ICU versus participants in a medical ICU.

Sensitivity analysis

We planned to explore potential effects of decisions made as part of the review process as follows.

Exclusion of studies that we judged to be at high or unclear risk of selection bias.
Use of the alternate meta‐analytical effects (fixed‐effect or random‐effects) model.

We planned to compare effect estimates from the above results versus effect estimates from the main analysis. We planned to report differences that altered interpretation of effects.

We planned to perform sensitivity analyses on all outcomes.

'Summary of findings' table and GRADE

The GRADE approach incorporates assessment of indirectness, study limitations, inconsistency, publication bias and imprecision. We used assessments made during our analysis to inform the GRADE process (see Data extraction and management, Assessment of risk of bias in included studies, Assessment of heterogeneity, Assessment of reporting biases and Data synthesis). This approach provides an overall measure of how confident we can be that our estimate of effect is correct (Guyatt 2008).

We used the principles of the GRADE system to provide an overall assessment of evidence related to each of the following outcomes.

Quantity and quality of sleep as measured through reports of participants or family members or by personnel assessments.
Quantity and quality of sleep as measured by PSG, actigraphy, BIS or EEG.
Anxiety or depression, or both.
Adverse events (such as cardiovascular events, respiratory events or illness resulting from immune deficiency).

Two review authors (SL and AS) independently used GRADEpro software (available at www.guidelinedevelopment.org/) to create a 'Summary of findings' table for each comparison. We planned to reach consensus and resolve disagreements by consulting with a third review author (PA) if required.

Results

Description of studies

Results of the search

We screened 17,003 titles and abstracts from database searches, results from clinical trials register searches, grey literature searches, and forward and backward citation searches. We carried out full‐text review of 17 articles. We excluded 10 studies, and reported details of two of these excluded studies. We identified four eligible studies, and two studies are awaiting classification. We identified one ongoing study. See Figure 1.

Included studies

We included four studies with 149 participants (Engelmann 2014; Kondili 2012; McLeod 1997; Treggiari‐Venzi 1996).

Three of the four studies used a parallel design (Engelmann 2014; McLeod 1997; Treggiari‐Venzi 1996). The fourth study used a cross‐over design (Kondili 2012).

Study population and setting

The four included studies enrolled adults admitted to the ICU.

One study specified inclusion of only participants that were mechanically ventilated (Kondili 2012), one study included only participants that were not mechanically ventilated (Engelmann 2014), and the remaining two studies did not clearly specify whether participants were mechanically ventilated. Three studies used the APACHE II scoring system to classify the severity of participants; Kondili 2012 reported a median (range) APACHE II of 23 (16 to 34), McLeod 1997 reported a median (range) APACHE II in the intervention group of 18 (6 to 35) and in the comparison group of 17 (6 to 37), Treggiari‐Venzi 1996 reported a mean (standard deviation; range) APACHE II in the intervention group of 13.5 (± 4.5; 8 to 20) and in the comparison group of 14.5 (± 4.1; 9 to 21). Engelmann 2014 reported no scoring system to classify the severity of participant conditions. We noted differences in these scores which could reflect severity of participant conditions between studies.

Participant admission diagnoses included trauma, and postsurgical patients (McLeod 1997; Treggiari‐Venzi 1996), and only surgical patients Engelmann 2014). One study did not report admission diagnoses (Kondili 2012). We noted differences in study exclusion criteria; two studies excluded people with renal impairment (Engelmann 2014; McLeod 1997), two studies excluded people with neurological disorders (Kondili 2012; Treggiari‐Venzi 1996), one study excluded people with head trauma (Treggiari‐Venzi 1996), and one study excluded people with sepsis (Kondili 2012).

Two studies reported use of morphine, if required, during the study period (McLeod 1997; Treggiari‐Venzi 1996); we noted that morphine use was balanced between participants in each group. One study reported no opioid use during the study period, with use of non‐steroidal anti‐inflammatory drugs (NSAIDs) if required (Kondili 2012). One study reported prior use of benzodiazepines and propofol as part of general anaesthesia management in surgical patients, and reported no use of antidepressants or neuroleptics (Engelmann 2014); opioid use was not reported in this study.

Interventions

All studies reported administration of propofol in the evening to promote sleep. Times of administration were: 11 pm to 6 am (Engelmann 2014); 10 pm to 7 am Kondili 2012); and 10 pm to 6 am (McLeod 1997; Treggiari‐Venzi 1996).

Study authors reported administration of a starting bolus of 0.01 mg/kg to 0.05 mg/kg over two minutes (Kondili 2012), and 0.2 mg/kg to 0.3 mg/kg over two minutes (Treggiari‐Venzi 1996). Engelmann 2014 provided a specific dose for the infusion (2 mg/kg/hour) and other reports stated that propofol was given at a rate to maintain a given level of sedation (0.3 mg/kg/hour to 3 mg/kg/hour in Treggiari‐Venzi 1996 but dose was not reported in Kondili 2012 or McLeod 1997).

One study used a cross‐over design over two consecutive nights with participants given propofol or no agent on each night (Kondili 2012). Other studies were for a duration of one night (Engelmann 2014), two nights (McLeod 1997), and five nights (Treggiari‐Venzi 1996).

Comparisons

One study compared propofol with no agent (Kondili 2012). One study compared a higher dose of propofol at night (described as "additional night sedation") versus a lower dose given throughout the day and maintained overnight (described as "constant light sedation") (McLeod 1997). Two studies compared propofol versus a benzodiazepine; flunitrazepam (Engelmann 2014), and midazolam (Treggiari‐Venzi 1996).

Measured outcomes

Sleep quality was assessed using standardized patient sleep diaries (Engelmann 2014) and questionnaires (Treggiari‐Venzi 1996). McLeod 1997 reported on quantity and quality of sleep by comparison of Ramsay sedation scores by personnel. Two studies reported on the quantity and quality of sleep as measured using BIS and PSG (Engelmann 2014; Kondili 2012). Treggiari‐Venzi 1996 assessed psychological consequences of sleep deprivation by considering participant anxiety and depression.

Excluded studies

We assessed the full‐texts of 10 studies that did not meet the review criteria. We reported two articles (of the 10 studies) in the review (Seymour 2012; Wu 2016). One study was a cohort study of mechanically ventilated patients in the ICU that studied whether doses of benzodiazepine and propofol were increased at night‐time and the effect of daytime and night‐time use of delirium, coma and duration of mechanical ventilation (Seymour 2012). One study was an RCT of non‐mechanically ventilated elderly (aged 65 years or older) patients in the ICU that studied the effect of dexmedetomidine versus a placebo on sleep architecture (Wu 2016). See Characteristics of excluded studies table.

Studies awaiting classification

Two studies are awaiting classification (Guo 2012; NCT00826553).

We identified one study in database searches which requires translation; we await translation of the full report before assessing eligibility (Guo 2012). We identified one study in searches of clinical trials registers which had been terminated early due to low accrual rates; we await publication of the results of this study before assessing eligibility (NCT00826553).

See Characteristics of studies awaiting classification table.

Ongoing studies

We identified one ongoing study (Hollinger 2017). This study aims to compare propofol with dexmedetomidine administered at night‐time to promote a circadian rhythm. Participants are in the ICU with a diagnosis of delirium, and the study hypothesis is that reinstituting a normal day‐night cycle will decrease the duration of delirium. See Characteristics of ongoing studies table.

Risk of bias in included studies

See Characteristics of included studies table. See 'Risk of bias' graph and 'Risk of bias' summary (Figure 2; Figure 3).

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study. Note: blank spaces indicate outcome measures that are not reported by study authors

Allocation

All studies reported that participants were randomly assigned to groups. No studies provided sufficient details of randomization or methods to conceal allocation from personnel, and we judged all studies to have unclear risk of selection bias.

Blinding

It was not possible to blind personnel in Engelmann 2014 (agents given by a different method), or in Kondili 2012 (no agent as comparator). Blinding of personnel was possible but problematic in Treggiari‐Venzi 1996 and McLeod 1997; the drugs were given by the same method (i.e. intravenous infusion over the same time length), but doses and sedative aims differed between the intervention and comparison group. Attempts to blind personnel would require deliberate manipulation of study methodology which was not reported in either study; we assumed that study investigators made no attempts to blind personnel. We judged all studies to have high risk of performance bias amongst personnel.

Two studies included participant‐reported outcomes (Engelmann 2014; Treggiari‐Venzi 1996), and blinding of participants to the intervention group was particularly relevant. Treggiari‐Venzi 1996 described an "open" study and therefore participants were not blinded to their treatment group. We judged this study to have high risk of performance bias amongst participants. Engelmann 2014 reported that participants were unaware of which agents they were given, and we judged this to have low risk of performance bias amongst participants.

Two studies reported participant‐reported outcomes using subjective measures (Engelmann 2014; Treggiari‐Venzi 1996). Participants in Treggiari‐Venzi 1996 were aware of group allocation and we judged this study to have high risk of detection bias for these subjective measures. Participants in Engelmann 2014 were blinded to the intervention and we judged this study to have low risk of detection bias for the participant‐reported outcome.

Three studies reported measures of sleep using validated measurement tools, which rely on objective measures (Engelmann 2014; Kondili 2012; McLeod 1997). We judged Kondili 2012 to have low risk of detection bias because outcome assessments were not physicians who had administered propofol. We judged McLeod 1997, in which assessors made judgements of participants' diurnal sleep pattern against the Ramsay Sedation Scale, to have low risk of detection bias because study authors reported that all investigators were blinded to group allocation. Study authors provided no detail in Engelmann 2014 and it was not possible to detection bias for objective measures and we judged this to have an unclear risk of bias.

We did not assess performance bias for participants in Kondili 2012 and McLeod 1997 because participant‐reported outcomes were not assessed in these studies and we did not believe that blinding would influence sleep as measured with objective measures. Similarly, we did not assess risk of detection bias for objective measures in Treggiari‐Venzi 1996, as this study did not report such outcomes.

Incomplete outcome data

We noted a large number of losses in Engelmann 2014, and technical difficulties during the study also reduced available data; we judged this study to have high risk of attrition bias. Study authors reported a large number of losses in Treggiari‐Venzi 1996 and included explanations; we were concerned that reasons for losses could be described as 'not missing at random' (i.e. some participants lost due to reactions against midazolam) and lack of data for these participants could have influenced results. We judged Treggiari‐Venzi 1996 to have high risk of attrition bias.

We noted no risk of bias in the loss of one participant in McLeod 1997, and in Kondili 2012; we judged both studies to have low risk of attrition bias.

Selective reporting

We identified prospective clinical trials registration for one study (Kondili 2012); outcomes in the study report reflected outcomes listed in the registration documents and we judged this study to have low risk of reporting bias. We identified no clinical trials registration documents for the remaining studies and it was not feasible to judge risk of reporting bias.

Other potential sources of bias

Two studies reported additional administration of morphine to participants if required (McLeod 1997; Treggiari‐Venzi 1996). Use of morphine between intervention groups was comparable and we believed it introduced no risk of bias. One study reported that additional opioids were administered to no participants (Kondili 2012) and we judged this to have low risk of bias. We noted no details in use of concomitant drugs in Engelmann 2014 and were unable to assess if additional bias had been introduced through concomitant drug use.

We noted no differences in baseline characteristics between groups in parallel design studies (Engelmann 2014; McLeod 1997; Treggiari‐Venzi 1996).

Effects of interventions

See: Table 1; Table 2; Table 3

Summary of findings for the main comparison. Propofol for the promotion of sleep in the intensive care unit versus no agent.

Propofol for the promotion of sleep in the intensive care unit versus no agent
Patient or population: critically ill adults in the intensive care unit Settings: intensive care unit, Greece Intervention: propofol given to promote overnight sleep Comparison: no agent
Outcomes	Impacts	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
Quantity and quality of sleep as measured through reports of participants or family members or by personnel assessments Data collected at end of study follow‐up	Not reported.	‐	‐	‐
Quantity and quality of sleep as measured by PSG, actigraphy, BIS or EEG Data collected at end of study follow‐up	Outcome measured by PSG. Study authors reported no evidence of a difference in duration of sleep or sleep efficiency, and reported disruption to usual REM sleep with propofol.	13 (1 study)	⊕⊝⊝⊝  Very low¹	We identified only 1 study and could not conduct a meta‐analysis.
Anxiety or depression, or both, as measured using validated tools Data collected at end of study follow‐up	Not reported.	‐	‐	‐
Adverse events (such as cardiovascular events, respiratory events or illness resulting from immune deficiency)	Not reported.	‐	‐	‐
BIS: Bispectral Index; EEG: electroencephalogram; PSG: polysomnography; REM: rapid eye movement.
GRADE Working Group grades of evidence  High quality: further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: we are very uncertain about the estimate.

Outcome: quantity and quality of sleep as measured by PSG, actigraphy, BIS or EEG
Study ID	Interventions	Measurement tool	Narrative results as reported by study authors	Data; median (IQR)
Kondili 2012	Cross‐over design over 2 nights. Propofol vs no propofol	PSG	No statistically significant difference in sleep time.	Total sleep time (minutes) Propofol: 260 (113‐417) No propofol: 214 (40‐285)
			No statistically significant difference in sleep efficiency (note: study authors do not give a definition of sleep efficiency).	Total sleep time. Sleep efficiency (%) Propofol: 76.3 (28.4‐96.9) No propofol: 62.6 (13.1‐85.9) P = 0.37
			No statistically significant difference in Stage 1 sleep.	Total sleep time (%) Propofol: 20.8 (5.6‐80.6) No propofol: 30.7 (4.6‐66.7) P = 1.00
			No statistically significant difference in Stage 2 sleep.	Total sleep time (%) Propofol: 48.9 (4.8‐84.0) No propofol: 46.1 (3.0‐80.4) P = 0.66
			No statistically significant difference in SWS.	Total sleep time (%) Propofol: 0 (0 – 5.8) No propofol: 0 (0 – 0) P = 0.75
			Statistically less REM sleep in propofol group.	Total sleep time (%) Propofol: 0 (0‐0) No propofol: 1.4 (0‐13.0) P = 0.04

Date form completed (dd/mm/yyyy)
Report title (title of paper/ abstract/ report that data extracted from)
Reference details
Report author contact details
Publication type (e.g. full report, abstract, letter)
Study funding sources (including role of funders)
Possible conflicts of interest (for study authors)

Study Characteristics	Eligibility criteria	Yes No Unclear	Location in text (pg /fig / table)
Type of study	Randomized controlled trial Quasi‐randomized controlled trial
Participants	Adult >16yrs. Intensive Care Unit admission
Intervention	Propofol
Comparisons	No agent; or A different dose of propofol; or Another agent, specifically administered for the promotion of sleep
INCLUDE EXCLUDE
Reason for exclusion

	Description (include comparative information for each group (i.e. intervention and controls) if available	Location in text (pg /fig / table)
Population and description (from which study participants are drawn)
Inclusion criteria
Exclusion criteria
Method/s of recruitment of participants
Informed consent obtained	Yes/No/Unclear

	Descriptions as stated in report/paper		Location in text (pg /fig / table
Aim of study
Design(e.g. parallel, crossover, cluster)
Unit of allocation (by individuals, cluster /groups or body parts)
Ethical approval needed/obtained for study	Yes/No/Unclear
Registered with clinical trial registry	Yes/No	Clinical trials ID number:

Domain	Risk of bias Low/ High/ Unclear	Support for judgement	Location in text (pg /fig / table
Sequence generation (selection bias)
Allocation concealment (selection bias)
Baseline Imbalances
Blinding of participants and personnel (performance bias)
Blinding of outcome assessment (detection bias)
Incomplete outcome data (attrition bias)
Selective outcome reporting (reporting bias)
Other bias

	Description as stated in report/paper		Location in text(pg /fig / table
Total no. randomized
Withdrawals and exclusions (if not provided below by outcome)
Age age range (mean)	Intervention	Comparison

Sex Male/Female	Intervention	Comparison

Other baseline characteristics
Post‐operative admission or emergency admission
Co‐morbidities
Time in ICU prior to intervention

	Description as stated in report/paper	Location in text (pg /fig / table
Comparison group type (placebo, no treatment, different drug)
No. randomized to group
Description (type, dose, method of administration)
Duration of treatment period
Time of administration
Concomitant agents (type, dose, method of administration etc)

	Description as stated in report/paper	Location in text (pg /fig / table
Outcome name
Time points measured
Time points reported
Outcome definition
Person measuring/reporting
Unit of measurement
Scales: upper and lower limits (indicate whether high or low score is good)
Is outcome tool validated?	Yes/No/Unclear
Imputation of missing data (e.g. assumptions made for ITT analysis)
Assumed risk estimate (e.g. baseline or population risk noted in Background)
Power

	Description as stated in report/paper		Location in text (pg & ¶ /fig / table
Comparison
Outcome
Results	Intervention	Comparison
No. missing participants and reasons
Any other results reported
Unit of analysis
Statistical methods used & appropriateness of these methods

Methods	RCT. Parallel design.
Participants	Inclusion criteria: adult patients admitted to the medical ICU who required mechanical ventilation and were sedated with a GABA agonist, expectation of being mechanically ventilated for > 24 hours. Exclusion criteria: aged < 18 years; pregnant; inability to obtain informed consent from the patient or his or her surrogate; physiologically benzodiazepine dependent; at risk of withdrawal syndrome, anoxic brain injuries, stroke or neurotrauma; medical team following patient unwilling to change sedation regimen; moribund and not expected to survive 24 hours or actively withdrawing medical support; documented allergy to study medications; advanced heart block at time of screening; prisoners; RASS target ≤ ‐4 at the time of screening; PSG equipment unavailable.
Interventions	Dexmedetomidine vs GABA agonist (e.g. propofol, benzodiazepines).
Outcomes	Time spent in standard sleep stages (N1, N2, N3, REM), time spent in atypical sleep, presence of burst suppression.
Notes	Study terminated due to poor recruitment. We await publication of study report to assess eligibility.

Trial name or title	Comparison of propofol and dexmedetomidine infused overnight to treat hyperactive and mixed ICU delirium: a protocol for the Basel ProDex clinical trial
Methods	RCT. Parallel design.
Participants	Target number of randomized participants: 316. Setting: adult ICU admitting medical or surgical patients, Switzerland. Inclusion criteria: adult patients aged > 18 years, current delirium (hyperactive or mixed type) detected by a specialized assessment method (ICDSC) in 1 of participating ICUs. Exclusion criteria: delirium prior to ICU admission, egg and soy allergy, hypersensitivity to the active substances, advanced heart block (grade 2 or 3) unless paced, bradycardia of different origin, uncontrolled hypotension, acute cerebrovascular conditions, severe cardiac dysfunction, aged < 18 years, terminal state, pregnancy, status epilepticus or postictal states following seizures on EEG, active psychosis, delirium tremens, substance abuse with experience of acute withdrawal.
Interventions	Propofol group: Propofol 1% (1 g/100 mL) 1‐4 mg/kg/hour administered by continuous IV infusion from 8 pm to 6 am beginning the evening after diagnosis of hyperactive or mixed delirium. Dexmedetomidine group: Details: dexmedetomidine (200 µg/2 mL) 0.7 µg/kg bolus, following by 1.4 µg/kg/hour by continuous IV infusion from 8 pm to 6 am beginning the evening after diagnosis of hyperactive or mixed delirium. Participants in both groups were given haloperidol as a rescue medication if required.
Outcomes	Delirium duration in hours, delirium‐free days at 28 days, death (28 days), severity of ICU delirium, number of ventilator‐free days, need for rescue sedation (haloperidol), amount of oral quetiapine, total costs of medication, length of ICU stay (hours), length of hospital stay (days), depth of sedation (using RASS, or SAS), depth of sedation in study group (using EEG).
Starting date	January 2017.
Contact information	alexa.hollinger@usb.ch.
Notes	Clinical trials registration: NCT02807467.

Have important population groups been excluded from the study?	Yes No Unclear
Does the study directly address the review question? (any issues of partial or indirect applicability)	Yes No Unclear
Are there any limitations in the design of the study?	Yes No Unclear

	Description as stated in report/paper	Location in text (pg & ¶ /fig / table
Key conclusion of study authors
References to other relevant studies
Correspondence required for further study information (from whom, what and when)

Methods	RCT. Cross‐over design.
Participants	Total number of randomized participants: 13. N = 13; 1 participant was excluded because of severe acidosis; number analysed = 12. Setting: ICU in Greece. Inclusion criteria: critically ill patients who had been mechanically ventilated for ≥ 48 hours and anticipated to be on assisted modes for ≥ 2 consecutive days; patients not requiring sedation or analgesia with opioids. Exclusion criteria: GCS < 11; APACHE II > 15; presence of delirium; administration of any sedative drugs or opioids in previous 24 hours; detectable plasma levels of sedative drugs (e.g. benzodiazepine, propofol) or opioids (e.g. morphine); history of epilepsy or other neurological disease that may affect quality of sleep; history of sleep apnoea; ongoing sepsis. Baseline characteristics (for 12 participants, not divided by group due to cross‐over design) Gender M/F: 6/6. Age median (IQR): 73 (63‐75) years. APACHE II at admission median (IQR): 23 (16‐34).
Interventions	Propofol bolus of 0.01‐0.05 mg/kg over 2 minutes at 10 pm, followed by IV continuously until 7 am to maintain Ramsay Sedation Scale level 3. Each participant studied for 2 consecutive nights, with or without propofol, in random order.
Outcomes	Quality of sleep measured using PSG (total sleep fragmentation – sum of arousals and awakenings per hour of sleep; breathing variables and asynchrony).
Notes	Funding/declarations of interest: Cretan Critical Care Society (details taken from published protocol). Study dates: October 2009 to October 2011. Notes: attempts made to reduce other confounders to sleep disruption (single ICU room, window blinds closed, noise, nursing and other interventions minimized during study night. Light decreased to minimum level). Attempts to maintain same conditions on both nights of study. Participants administered NSAIDs for analgesia of required.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Study authors reported that participants were randomly assigned to groups but provided no additional details of randomization methods.
Allocation concealment (selection bias)	Unclear risk	No details.
Blinding of personnel (performance bias)	High risk	Propofol compared to 'no agent' and therefore not possible to blind personnel.
Blinding: objective measures (detection bias)	Low risk	Physicians who administered study drug were not given access to EEG readings during study.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Loss of 1 participant who required change of mode of ventilation due to severe respiratory acidosis; unlikely to affect overall outcome data.
Selective reporting (reporting bias)	Low risk	Registered trial ID: ISRCTN72847182. Study appeared to report all outcomes as presented in protocol.
Other bias	Low risk	Study authors reported no opioid use and no sedative use for any participant, except intervention drugs for overnight sleep promotion, during the study period. We considered whether bias was introduced by use of a cross‐over design and whether there was a sufficient washout period after propofol use; propofol has a rapid offset and we judged that this design introduced no bias.

Study	Reason for exclusion
Seymour 2012	Mechanically ventilated adults in the ICU. Aimed to assess whether doses of benzodiazepines and propofol were increased at night. Cohort study design was ineligible.
Wu 2016	RCT. Non‐mechanically ventilated elderly (aged 65 years or older) patients in the ICU. Dexmedetomidine given to promote sleep. Did not include a comparison with propofol.

Methods	RCT. Multi‐arm parallel design.
Participants	Mechanically ventilated patients in the ICU.
Interventions	Circadian rhythm vs daily interruption vs continuous sedation or demand sedation.
Outcomes	Recovering natural circadian rhythm, duration of mechanical ventilation.
Notes	Limited detail in English abstract. Report requires translation from Chinese before eligibility assessment.

PERMALINK

Propofol for the promotion of sleep in adults in the intensive care unit

Sharon R Lewis

Oliver J Schofield‐Robinson

Phil Alderson

Andrew F Smith

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

'Summary of findings' table and GRADE

Results

Description of studies

Results of the search

1.

Included studies

Study population and setting

Interventions

Comparisons

Measured outcomes

Excluded studies

Studies awaiting classification

Ongoing studies

Risk of bias in included studies

2.

3.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Summary of findings for the main comparison. Propofol for the promotion of sleep in the intensive care unit versus no agent.

Summary of findings 2. Propofol for the promotion of sleep in the intensive care unit versus propofol at a different rate or dose.

Summary of findings 3. Propofol for the promotion of sleep in the intensive care unit versus another agent specifically administered to promote sleep in the intensive care unit.

Comparison 1: propofol versus no agent

Primary outcomes

1. Quantity and quality of sleep as measured through reports of participants or family members or by personnel assessments

2. Quantity and quality of sleep as measured by PSG, actigraphy, BIS or EEG

1. Comparison 1: propofol versus no agent.

Secondary outcomes

1. Anxiety or depression, or both, as measured using validated tools

2. Adverse events (such as cardiovascular events, respiratory events or illness resulting from immune deficiency)

Subgroup analysis

Sensitivity analysis