sign in sign up
<<
Home , Dexmedetomidine

IMANCUAGEMENT & PRACTICE INTENSIVE CARE - EMERGENCY MEDICINE - ANAESTHESIOLOGY VOLUME 18 - ISSUE 3 - AUTUMN 2018

SPECIAL SUPPLEMENT in collaboration with CSL Behring

ShockPathophysiology of endotoxic shock, F. Forfori et al. Fluids in shock, M. Cecconi et al. It is time for improved fluid stewardship,M. LNG Malbrain, T.W. Rice, M. Mythen Vasoactive medication and RCTs, J. Gutteling & A.R.J. Girbes Advances in source control in patients with sepsis and septic shock, J.J. De Waele & I. Martin-Loeches Organ cross-talk in shock and critical illness, J.R. Prowle POCUS and SHOCK, A. Wong & J. Wilkinson

PLUS Xenon limits brain damage Cancer patients in the ICU, Humanizing the ICU experience following cardiac arrest, M. Maze I. Prieto del Portillo et al. with enhanced communication, & T. Laitio What should we stop doing in the A. Rocher

What’s new in sepsis in children? ICU? F.G. Zampieri Implementing ECCO2R and E. Esteban et al. Caring for very old patients in the vv-ECMO in non-academic Optimising in the ICU, ICU, H. Flaatten centres, K. Kogelmann M.C. Reade & D. Liu The sepsis box, bag and trolley, Improving access to safe C. Hancock & A. Hermon anaesthesia, J. Mellin-Olsen ©For personal and private use only. Reproduction must be permitted by the copyright holder. Email to [email protected]. must be permitted use only. Reproduction and private ©For personal

icu-management.org @ICU_Management 200 MATRIX

Michael C. Reade Intensivist & Australian Defence Force Optimising sleep in the ICU Professor of Military Medicine and Surgery Faculty of Medicine University of Queensland and Joint Health Command Disordered sleep is common in ICU patients. While many of the reasons for Australian Defence Force this are impossible to modify, and others rely on improvement in the under- [email protected] lying condition, many directly depend on the actions of the treating team: for example, exposure to noise, timing of therapeutic procedures, tapering of sedating drug doses, and daytime mobilisation. Some patients might benefit David Liu Postdoctoral Research Fellow from nocturnal sedation, but there is reasonable evidence that benzodiaz- Burns, Trauma & Critical Care epines and are not the best options. Although unproven in large Research Centre University of Queensland clinical trials, options including dexmedetomidine, (and ramelt- eon), and are all reasonable, especially as their Registrar Department of Intensive Care effect is usually able to be assessed over 1-2 nights, facilitating an “n of 1” Medicine Royal Brisbane and Women’s trial approach to individualised therapy. Hospital Brisbane, Australia

[email protected]

Critical illness reduces normal sleep Cough is distressing to patients and staff, but delusions then hallucinations and psychosis Most critically ill ICU patients report, recall, a is a temporary solution that (Waters et al. 2018). Physical performance or are observed to have disordered sleep. would make many underlying problems worse, (Kirschen et al. 2018) and immune function Specifically, the number of awakenings per while also causing psychological distress. It (Mullington et al. 2010) are also degraded hour is higher than in health (Elliott et al. is much better to make a diagnosis, address by inadequate sleep. 2013; Roche-Campo et al. 2013, Drouot et underlying causes, and use symptomatic Despite their ability to produce the outward al. 2014), daytime somnolence is increased temporising treatments that do not create appearance of sleep, GABA-ergic to as much as 50% of total sleep (White et adverse effects worse than the problem they used to facilitate tolerance of an endotracheal al. 1983; Cordoba-Izquierdo et al. 2013), are designed to treat. tube can have the opposite effect. Benzodi- and patients report sleep quality as worse azepines increase N2 (light) but reduce N3 than baseline (Elliott et al. 2013; Freedman (deep—thought to be the most restorative) et al. 1999; Little et al. 2012). EEG recordings commencing or sleep (Achermann and Borbely 1987; Borbely show a higher than normal proportion of increasing the rate of a et al. 1985), as does propofol (Herregods et light to deep sleep and that sedating drugs al. 1989). Similarly, also reduce N3 are primarily responsible for an “atypical infusion is not a and REM sleep (Kamdar et al. 2012). Benzo- sleep” pattern characterised by disorganised rational strategy to treat diazepines in particular have been associated delta waves and the absence of k complexes with in the ICU (Pandharipande and sleep spindles [summarised in Devlin et in most ICU et al. 2006; 2008), and delirium itself is al. (2018)]. There are many possible reasons patients an independent risk for disordered sleep for disordered sleep, and as the importance of (Devlin et al. 2018). Therefore, commencing each will vary in different patients (Figure 1), Sleep is a physiological state of cognitive or increasing the rate of a sedative infusion so will the optimal approach to management. and sensory disengagement from the envi- is not a rational strategy to treat insomnia in ronment (Kamdar et al. 2012) required in most ICU patients. Rather, specific treatments Sedation is a poor substitute for sleep some form by all mammals. Various cardio- such as those listed below should be tried Virtually every ICU clinician has at some stage vascular, respiratory, gastrointestinal and first. The pain, agitation/sedation, delirium, asked: “My patient didn’t sleep, could we give thermoregulatory effects are observed, the immobility, and sleep (PADIS) guidelines a sedative?” The single most important goal importance of which is not fully understood. recommend against propofol as a strategy to of this paper is to explain why this question However, acute sleep deprivation experiments improve the sleep of critically ill patients, and is analogous to the request “My patient keeps are simple to conduct, revealing perceptual against in general (Devlin coughing, could we give a muscle relaxant?” distortions within 24-48 hours, followed by et al. 2018). Of course, there remain many ©For personal and private use only. Reproduction must be permitted by the copyright holder. Email to [email protected]. must be permitted use only. Reproduction and private ©For personal ICU Management & Practice 3 - 2018 201 MATRIX

17/09/2018 Word Art

17/09/2018 Word Art

1/1

Figure 1. Word clouds illustrating the relative importance of combinations of factors that might contribute to Figure 2. The contribution of each noise category disturbed sleep in two different ICU patients. Clearly the correct response to each will be different. for (A) the acoustic energy and (B) the number of predicted loudness peaks.

Reproduced from (Simons et al. 2014) under the terms of the Creative specific indications for opioids or GABA- the Richards-Campbell Sleep Questionnaire Commons Attribution License.

ergic sedatives in the ICU other than sleep. (RCSQ) has been validated in ICU1/1 patients and correlates well with polysomnography countries and ICUs where it is the default Measuring sleep (Richards et al. 2000), and appears better mode in preference to assist control) was If specific treatments for sleep are to be used, than any technological device at present. An not studied. No recommendation was made it would be logical to measure their effect observational study formally comparing all of on whether adaptive modes of ventilation using a validated instrument. Regrettably, these measurement approaches is in progress are beneficial. It is likely that this question sleep is more difficult to identify than most (Delaney et al. 2018). is suitable for an “n of 1” trial design—that other physiological variables, and current is, in a patient with insomnia, trial of a night tools are so imperfect that recent consensus Non-pharmacological methods to on assist control or SIMV+PSV seems likely guidelines (Devlin et al. 2018) recommend improve sleep to lose little. against routine clinical use. Nonetheless, Ventilator mode technology is advancing rapidly in this area. The 2018 PADIS guidelines recommended Music Polysomnography (electroencephalogram, assist-control ventilation over pressure-support One small randomised trial has tested the effect electromyogram and electrooculogram), ventilation, based on three comparative of music on sleep (Su et al. 2013). Participants the gold standard, is too complex to acquire studies in which sleep had been measured listened to 45 minutes of classical-type music and interpret for anything but research use. as an outcome (Devlin et al. 2018). All three written specifically for the purpose, or no Actigraphy, using motion-sensors on the trials (which together comprised only 61 music. Those played music had significantly wrist, while sufficiently accurate in routine patients) found a significant benefit in sleep lower heart rate, blood pressure, respiratory sleep studies, over-estimates sleep in criti- efficiency (proportion of time meant to be rate, and spent significantly longer in stage cally ill patients who can be immobile for asleep actually spent asleep)(18.3% greater, N3 sleep and had significantly better subjec- reasons other than sleep (Kamdar et al. 95% CI 7.9%-28.8%), and also a small but tive sleep scores. 2012). Compressed EEG signals (primarily significant increase in the proportion of Bispectral Index, BIS) can estimate sleep total sleep time spent in REM sleep. Whether Reduction of ambient noise depth, but poorly define different stages in this would also be true for synchronised Ambient noise levels in the ICU are approxi- sleep architecture and are difficult to use over intermittent mandatory ventilation + pres- mately double that recommended by the many hours. Subjective assessment using sure support ventilation (SIMV+PSV) (in World Health Organization (Darbyshire and ©For personal and private use only. Reproduction must be permitted by the copyright holder. Email to [email protected]. must be permitted use only. Reproduction and private ©For personal ICU Management & Practice 3 - 2018 202 MATRIX

Table 1. Non-pharmacological strategies to improve sleep Table 2. Drugs to improve sleep

Strategy Drug Suggested Effect on sleep Known effects on dose other outcomes Ventilator mode (assist-control in preference to pressure support) Melatonin 3-10mg In normal people and people with No other benefit has been primary insomnia, reduces time to observed in ICU patients Music at sleep time fall asleep, but no clinically signifi- (Devlin et al. 2018) Reduction of ambient noise cant effect on time spent asleep. In patients unable to sleep due to a Earplugs medical cause (“secondary” insom- Reduction of ambient light at night nia), moderate to high quality studies show melatonin has little or no Scheduling of patient care activities during beneficial effect on sleep (Buscemi daytime et al. 2004) Tapering of drugs with sedative effects 8mg Shortens time to fall asleep and Lower incidence and duration Daytime mobilisation increases total duration of sleep of delirium, and fewer night- (Neubauer 2008) time awakenings (Nishikimi et al. 2018) Young, 2013). However, background noise is Dexmedetomidine 0.1mcg/kg/hr Increases total sleep time and Reduced postoperative probably less important for sleep disruption proportion of time spent in N2 (deep- delirium, reduced reported than the frequency and magnitude of peak er) stage of sleep; reduces proportion pain, improved reported sleep of time spent in N1 (lighter) sleep. No (Su et al. 2016) levels, which were above 85dBA up to 16 change in REM sleep (Wu et al. 2016) times an hour. That this is not the inevitable Amitriptyline 10-50mg Shortens time to fall asleep and No benefit has been proven consequence of electronic devices, mechanical increases overall sleep time, but in ICU patients when used for ventilators, etc. was shown by a 2014 Dutch reduces REM sleep (Wilson and this indication observational study (Simons et al. 2014) that Argyropoulos 2005) found the loudness peaks (part B of Figure 2) Mirtazapine 15-30mg Increases total slow wave sleep and No benefit has been proven REM sleep, as well as improving in ICU patients when used for were 60% due to staff activity and 32% due insomnia scores (Shen et al. 2006) this indication to staff speech. Only 6% were due to equip- 50mg Increases total slow wave sleep No benefit has been proven ment alarms. but reduces REM sleep. Improves in ICU patients when used for subjective insomnia. No effect on this indication total sleep duration or time to fall Earplugs asleep (Montgomery et al. 1983) If encouraging clinical staff to be quiet is impossible, another approach could be to use times per night (Gabor et al. 2003; Tamburri patient earplugs. This is surprisingly effective, early mobilisation et al. 2004) really should be questioned. as shown by a 2017 meta-analysis of nine is to date the non-pharma- Often these activities are not done for staff studies/1,455 patients (Litton et al. 2016) convenience but because of the realities of that found an overall relative risk of delirium cological intervention staff rostering; taking this into account at a of 0.59 (95% CI 0.44-0.78), although not associated with the departmental level might alleviate this major all of the included studies measured sleep. Of problem. those that did, two found earplugs did indeed greatest observed improve self-reported sleep quality, while one reduction in delirium Tapering of drugs with sedative effects did not observe any significant difference. Many drugs used in critical care have sedating utility of this intervention alone. Nonetheless, effects, and abrupt withdrawal after a period Light it is difficult to argue against such a low-cost, of habituation leads to a withdrawal state Five studies (summarised in Bion et al. 2018) low risk intervention as turning down the characterised by hyper-alertness and insomnia. have assessed the effect of reducing environ- intensity of the lighting at night. Unless there is a good reason, it is usually mental light at night. However, each has done better to slowly reduce doses of opioids and this as part of a multi-component intervention Scheduling of patient care activities other sedating medications over several days that also reduced noise and other disruptions When asked, patients reported that having (Brown et al. 2000) Doing so can avoid the to patient sleep, and all used subjective sleep their vital signs assessed and having blood need to simply replace one type of sedative assessments. The reviewers concluded that this, taken were more disruptive to sleep than any with another. combined with the different patient populations noise (Freedman et al. 1999). Critical care is studied (from non-ventilated neuro ICU patients a 24-hour activity, but whether medication Mobilisation to restore day-night rhythm to mechanically ventilated ICU patients), made administration, radiographs, wound care, and Early mobilisation is to date the non-phar- it difficult to reach any conclusion about the bathing need to interrupt sleep up to 40-60 macological intervention associated with the ©For personal and private use only. Reproduction must be permitted by the copyright holder. Email to [email protected]. must be permitted use only. Reproduction and private ©For personal ICU Management & Practice 3 - 2018 ©For personal and private use only. Reproduction must be permitted by the copyright holder. Email to [email protected]. tive non-ventilated non-cardiac surgery high dependency unit of N3 sleep (Akeju et al. 2018). Inastudy of 76 postopera benzodiazepines. Unlike thesedrugs, itincreases theproportion illpatientscritically by distinctfrom propofol amechanism and Dexmedetomidine, an alpha-2 , produces sedation in Dexmedetomidine currently limitsutility. CNS orotheradverse effects. ofavailability Lack insomecountries this is a promising intervention, apparently without substantial fewer night-time awakenings (Nishikimi et al. 2018), all suggesting (0.78 vs 1.40 days, p=0.048), and the nonintubated patients had (24.4%vs.of delirium 46.5%, p=0.04)ofnearly halftheduration 45 whoreceived 8mg/dramelteonhadnearly halftheincidence underway. In the published single-centre study of 88 patients, the of Trial/Registration/TrialReview.aspx?id=369434 prevention the units for Care trial (Pro-MEDIC study) ACTRN12616000436471 [ Melatonin Intensive of in Delirium administration Pro-phylactic (MELLOW-1 ( agonist ramelteon was published in 2018, and two larger trials ofthemelatonin-receptorOne smallrandomisedcontrolled trial Ramelteon (MELLOW-1 Patients Ill Critically in [ (MICE Delirium of Population Prevention for Elderly Care Intensive [ the in Use Melatonin (DEMEL Melatonin in ICUpopulations ( 2018). At areleast three planned or alreadymore trials recruiting low quality that enrolled a total of only 60 patients (Devlin et al. melatonin and sleep, based on three identified trials they class of The 2018PADIS guidelines make norecommendation regarding Melatonin that treatment withmedication shouldbeat leastattempted. by treatment tonon-pharmacological insomniaorso refractory above should be considered first, some patients are so distressed cal illness. measures listed While allofthenon-pharmacological sedativesas nocturnal inpatients withorrecovering from criti illness,in critical itwould benzodiazepines seemunwisetochoose Noting the adverse effects when used as sedatives of GABA-ergic drugs Drugs to improve sleep activities are yet illpatients. tobeassessedincritically case. The cognitive andsleepeffects ofenhancingotherdaytime circadian rhythm earlierthanmightotherwisehave beenthe theday,sleep during andhencere-established theirday-night must bethelikelihood that patients were lesslikely tobeable to While there are many for possible this effect, mechanisms one greatest (Schweickert etal. reduction observed indelirium 2009). clinicaltrials.gov/ct2/show/NCT02615340 clinicaltrials.gov/ct2/show/NCT03013790 Melatonin for Prevention of Delirium in Critically Ill Patients Patients Ill Critically in Delirium of Prevention for Melatonin ) [ ) clinicaltrials.gov/ct2/show/NCT02615340 ) [ ) clinicaltrials.gov/ct2/show/NCT03524937 Prevention of Delirium in Intensive Care by by Care Intensive in Delirium of Prevention – a randomized placebo controlled ]. ], and ] are currently anzctr.org.au/ Melatonin Melatonin ] and ] and ], ], ) ) - - 204 MATRIX

(HDU) patients aged ≥ 65 years, very low Amitriptyline benzodiazepines. This practice was recently dose dexmedetomidine (0.1 mcg/kg/hr) Amitriptyline is not covered in the 2018 supported by a systematic review of 45 studies increased the proportion of N2 sleep from PADIS guideline (Devlin et al. 2018) and is (Jaffer et al. 2017). Its off-label use (at 50mg 15.8% (IQR 1.3-62.8%) with placebo to recommended against by some authors on the nocte for at most 7 days) was recommended in 43.5% (16.6%-80.2%), prolonged total grounds that it has “not been studied for use a 2006 guideline from the UK Intensive Care sleep time, and improved subjective sleep in insomnia and has important potential side Society as an alternative to benzodiazepines, quality (Wu et al. 2016). In a larger subse- effects including hypotension, arrhythmias, and although there are no trials in this context quent trial, the same investigators found the anticholinergic syndrome. Use … to promote (Borthwick et al. 2006). same protocol associated with significantly sleep has been discouraged by an NIH consensus improved subjective sleep quality, along with panel on chronic insomnia” (Kamdar et al. (, doxyl- less than half the incidence of postoperative 2012). However, chronic insomnia is quite amine) delirium (9% vs. 23%; odds ratio 0.35, 95% different to brief treatment in ICU, and the Diphenhydramine (25-50mg nocte) and CI 0.22-0.54; p<0.0001) (Su et al. 2016). doses usually prescribed (10-50mg nocte) are (25mg nocte) both reduce sleep Survival rates were higher initially with dexme- most unlikely to cause the listed complications, latency and increase total sleep time (Koski at six months, one year, and 2 especially when patient weight and metabolic 2011). However, a quoted 70% increased risk of years (rate difference 5.2%, 5.3%, and 6.7%, function are considered. Amitriptyline, the cognitive decline in a cohort study comparing respectively; p<0.05), but after three years tricyclic antidepressant most commonly used as hospitalised patients receiving diphenhydr- the difference was not significant (32.6% vs. a nocturnal sedative, is generally recognised to amine to those not receiving it, along with 34.9% mortality; hazard ratio 0.87, 95% CI reduce REM sleep, but to reduce sleep latency more behavioural disturbances (Agostini et al. 0.68-1.13; p=0.303) (Zhang et al. 2018). In and to increase overall sleep time (Wilson and 2001), have led to recommendations against contrast, a study of 100 initially delirium-free Argyropoulos 2005). Whether this provides the use of sedating antihistamines as nocturnal critically-ill patients randomised to 0.2-0.7 benefit in an individual patient is readily sedatives in hospitalised patients. While this mcg/kg/hr dexmedetomidine at night vs. appreciated after only 1-2 nights’ treatment. is not trial evidence and these adverse effects placebo found dexmedetomidine associated While non-pharmacological treatments are could be the result of many cofounding influ- with less delirium (relative risk, 0.44; 95% always better first-line options, given the ences, availability of the alternatives listed CI, 0.23-0.82; p=0.006), but no observable known adverse effects of benzodiazepines and above argue against using antihistamines as difference in sleep quality on a subjective sleep the absence of other good options, some argue first-line options in adults. questionnaire (Skrobik et al. 2018). This led that amitriptyline is a reasonable alternative. the PADIS guideline authors to be circumspect Conclusion in their recommendation, stopping short of Mirtazapine Facilitating sleep at night is likely to have recommending dexmedetomidine for sleep Mirtazapine, an atypical with outcome benefit in many patients, and is alone but noting its potential benefit on sleep a mechanism that includes presynaptic alpha- also likely to address the insomnia that many could be considered when choosing a seda- 2 negative feedback blockade, postsynaptic commonly recall as a particularly distressing

tive if one was indicated (Devlin et al. 2018). serotonergic (5HT2 and 5HT3) blockade, and memory of their ICU stay. The non-pharmaco-

enhanced noradrenergic and 5HT1 neurotrans- logical approaches to insomnia are almost always , , , mission, also causes somnolence as a side effect preferable first-line alternatives in critically ill Conveniently grouped as “z-drugs”, zolpi- that has been used as a primary indication in patients. There is evidence that achieving the dem (an imidazopyridine), zopiclone and some patients. Its effect on sleep in patients outward manifestations of sleep through use eszopiclone (cyclopyrrolones) and zaleplon (a with major depression is more encourag- of benzodiazepines or other GABA-ergic drugs pyrazolopyrimidine) are non- ing than what is known of amitriptyline: it has a net detrimental effect. There are several

of the GABAA receptor. They are increases total slow wave sleep and REM sleep, non-GABA-ergic alternatives that show promise, claimed to have fewer adverse effects than the as well as improving insomnia scores (Shen but none has convincingly shown benefit in commonly-used sedative- benzodi- et al. 2006). As for amitriptyline, its use in randomised controlled trials. In part, this is azepines (typically , and critical illness is essentially not studied, but due to the practical difficulties of objectively ), although there is little evidence doses of 15-30mg should be safe, and n-of- measuring sleep in critically ill patients. “N for this. Perhaps for this reason, there has been 1 trials in individual patients would appear of 1” trials of certain agents until the optimal almost no research on these drugs as ICU to be a reasonable strategy in the absence of approach is found for each individual might sedatives, and they rarely appear in critical large randomised trials. be the best strategy, in anticipation of future care guidelines, including the 2018 PADIS definitive trials. guidelines (Devlin et al. 2018). Perhaps their Trazodone only indication is to continue chronic use (in Trazodone, a tetracyclic antidepressant, is References preference to abrupt withdrawal) in a patient commonly prescribed to outpatients as a For full references, please email editorial@icu-management. planned to stay only briefly in the ICU. treatment for insomnia as an alternative to org or visit https://iii.hm/o28 ©For personal and private use only. Reproduction must be permitted by the copyright holder. Email to [email protected]. must be permitted use only. Reproduction and private ©For personal ICU Management & Practice 3 - 2018