Journal of Critical Care 30 (2015) 436.e1–436.e6
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Electronic Articles
A web-based survey of United Kingdom sedation practice in the intensive care unit☆,☆☆ Sarah M. Yassin, MRPharmS a,c, Marius Terblanche, FRCA, MBBS b,c, James Yassin, MRCP, FRCA d, Catherine A. McKenzie, PhD, FFPharmS a,c,⁎ a
Institute of Pharmaceutical Sciences, Franklin-Wilkins Building, Kings College, London, SE1 7RT, UK School of Medicine, Guys Campus, King's College, London, SE1 9RT, UK Pharmacy and Critical Care Guys and St Thomas' NHS Foundation Trust (GSTFT), St Thomas' Hospital, London, SE1 7EH, UK d Department of Critical Care, Royal Sussex County Hospital, Brighton, BN2 5BE, UK b c
a r t i c l e
i n f o
Keywords: Sedation Midazolam Propofol Intensive care Morphine Delirium
a b s t r a c t Purpose: The purpose of this work was to obtain a detailed perspective of sedation practice. Sedation included sedative and opioid choice, presence of local guidelines, and use of scoring systems. Methods: A Web-based survey was designed. The aim was to gain sufficient detail of UK sedation while also being succinct enough to complete in 15 minutes. It was composed of relevant demographics, policy, sedative choice, and analgesia. The survey was piloted before launch. The investigators selected the intensive care unit (ICU) pharmacist as the respondent. Results: One hundred fifty-seven ICUs responded. Eighty-nine (59%) reported use of sedation guidelines, 78% undertook sedation holds, and 87% use sedation scores. Only 42% used a daily sedation target. Seventy (43%) assess for delirium; 27 of those use a validated tool. Propofol (89%) use was common, followed by midazolam (49%). Morphine (49%), fentanyl (34%), and alfentanil (34%) were the most frequently used opioids. Conclusion: This survey confirmed expected variation in UK sedation practice. Recognized strategies such as target sedation score and sedation policy are underused. A 43% uptake in delirium screening suggests that larger engagement is required to meet national standards. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Sedation is required for invasively ventilated patients [1]. A combination of sedatives and opioids is commonly used. Opioids are used as analgesics or antitussives and to aid ventilator synchrony [2]. Hypnotics, commonly either propofol or a benzodiazepine, are for amnesia, anxiolysis, and somnolence [2]. The choice of both opioid and sedative remains controversial, and more recently, there has been publication questioning the need for a hypnotic or “comfort” agent in its entirety [3]. There is consensus over sedative selection in specific situations, such as short-acting agents in neurologic assessment [4]. However, there remains equipoise in overall sedation prescribing practice within the intensive care unit (ICU) [5]. ☆ This work was undertaken from a grant awarded from Guys and St Thomas' Charity. ☆☆ The research was also supported by the National Institute for Health Research Biomedical Research Centre based at Guy's and St Thomas' National Health Service Foundation Trust and King's College London. The views expressed are those of the author (s) and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health. ⁎ Corresponding author. Guys and St Thomas' NHS Foundation Trust, Pharmacy and Critical Care, St Thomas' Hospital, Westminster Bridge Road, London SE1 7EH. Tel.: +44 7796193246. E-mail address: [email protected] (C.A. McKenzie). http://dx.doi.org/10.1016/j.jcrc.2014.11.006 0883-9441/© 2014 Elsevier Inc. All rights reserved.
Evidence in sedation has grown over the last 2 to 3 decades. There is now sufficient evidence to suggest that certain agents (hypnotics and opioids) may be harmful in specific patient groups. One of the earliest reports of opioid accumulation occurred in 1984 when Gordon [6] reported accumulation of morphine glucuronide metabolites and prolonged effect in renal failure. The benzodiazepines, lorazepam, and midazolam (MDZ) are also reported to be associated with harm. Evidence suggests that MDZ's complex pharmacokinetic profile may predispose specific patient groups to oversedation [7,8]. This includes those with renal failure and the elderly with or without compromised cardiac function [7]. This resulted in the recommendation to use lorazepam in the United States [1]. Lorazepam remained widely used, particularly in the United States, until Pandharipande et al [9] in 2006 reported it as the highest independent risk factor for development of critical care delirium. In addition, there were reports of metabolic acidosis after prolonged infusions of lorazepam [10]. With continuing emergence of new therapies and specific interventions such as sedation holds [11], there is a need to maintain a perspective of UK wide sedation prescribing. In 2005, a piece of work undertaken by the investigators reported that more than 90% of ICUs still use morphine and MDZ first line for sedation in ventilated patients [7]. This was despite evidence that both agents have active glucuronide metabolites that accumulate in renal failure [7,8].
436.e2
S.M. Yassin et al. / Journal of Critical Care 30 (2014) 436.e1–436.e6
Surveys are useful tools to describe sedation practice and to monitor changes in practice over time. In the UK in 2008, Reschreiter et al [5] published a postal survey of sedation practice. The investigators described a wide variation in practice but noted an increase in uptake of sedation scoring systems from a previous survey in 2000 [5]. Use of validated sedation scales has also increased in America, up to 88% in 2007 [12], and in Germany, use of a scale doubled between 2002 and 2006 [13]. The German survey also showed a reduction in use of MDZ for short-term sedation from 46% to 36% [13]. Changes in practice, however, are known to lag behind both evidence and national guidelines [14]. Most of these surveys were paper-based, postal surveys. They involved contact with the medical consultant lead of the ICU who would either complete the survey on behalf of the ICU [5,15] or delegate to a colleague. One of the advances in the ICU over the last 10 years has been the establishment of the intensive care pharmacist [16,17]. Pharmacists are often involved in the development of medicines related guidelines and could be considered to observe overall unit practice. This and the fact that the UK national group, the United Kingdom Clinical Pharmacy Critical Care Group (UKCPA CCG), has strong links with the investigating team were influential in the decision to select the critical care pharmacist as the survey respondent. 1.1. Aim and objectives The primary aim of this sedation survey was to gain a detailed perspective of UK sedation practice. Objectives were as follows: To establish national trends in prescribing and review applications or techniques to assist in safe prescribing To describe the pharmacy intensive care workforce and demographics in the context of the pharmacist's ability to report prescribing practice in the ICU and wider critical care environment. 2. Materials and methods The survey protocol was submitted to the local research ethics committee at St Thomas' Hospital. Approval was granted on chairman's action, and the survey could proceed without the requirement for a full ethics application. The study was registered with and approved by Guys and St Thomas' NHS Foundation Trust Research and Development (RJ110/N194). 2.1. Survey design The survey was designed using a Web-based provider, www. zoomerang.com. The questionnaire aimed to cover as much sedation practice as possible while being sufficiently brief to complete in 15 minutes. The questionnaire included a wide selection of questions that were deemed essential to gain a picture of sedation practice over the UK. Flexibility was allowed for more than 1 reply per trust to allow for practice within specialist critical care areas to be described separately. The survey questionnaire was split into 4 sections: 1. Workforce and demographics: This section was composed of detail regarding the hospital and its critical care unit(s), including number of level 3/2 beds and patient mix. There was also a subsection on pharmacist workforce, including length of time spent on the ICU, band/grade, and whether they worked in isolation or as part of a team. 1
2. Policy: Most questions in this section were derived from evidence-based recommendations on sedation practice. These included whether the ICU had a formal sedation policy; undertook sedation holds; what type of sedation scoring system was in place; whether sedation was titrated to a score daily; and whether units regularly assessed patients for delirium; and, if so, which (if any) delirium assessment method was used. 3. Sedative choice: This comprised detailed questions on the selection of sedatives in most common clinical scenarios encountered within ICU. This included most commonly used agents, sedative selection for short-term sedation, selection for multiple organ failure, agents for status epilepticus, drug withdrawal or dependence, weaning agents, and raised intracranial pressure. More than 1 answer was allowed per question, for example, sedatives used in the majority of patients. 4. Analgesia: This primarily concerned opioids. Questions included first-line agents, short-term agents, agents when increased pain is anticipated, therapeutic hypothermia, analgesia-based sedation, enhanced respiratory depression agents, and antipyretics in the ICU. 2.2. Mandatory and nonmandatory questions A number of the questions were considered essential to gain an overall perspective of sedation practice. These included questions on whether the ICU had sedation/analgesia guideline. These questions were designated as “mandatory,” which meant the respondent could not proceed unless they were completed. 2.3. Survey prelaunch pilot To ensure that the survey was robust, sensitive, and reliable, it was piloted on a number of professional groups. At the design phase, it was presented to the critical care research group, including intensivists, nurses, physiotherapists, and pharmacists. This group commented extensively on survey content. It was reviewed by 5 senior pharmacists who agreed the overall design of each question. After this review, the survey was considered ready. To test the validity of the survey immediately before launch, a pilot version was emailed to 2 consultant pharmacists for final comment. 2.4. Survey site After design and piloting, the survey was posted via the UKCPA CCG Web site (www.ukcpa.org.uk). To ensure maximum response: the Web link was also posted on the infection, cardiology, and general discussion message boards. Accompanying the survey link was a cover letter, which informed the respondent of the detail required for completion. 2.5. Response rate To maximize response, the investigators compared the list of responding hospitals against those using the Intensive Care National Audit and Research Council case mix program [5]. Where there was a gap in response, the investigators contacted the pharmacist firstly by email and then telephone to request survey completion. A follow-up call was made if the survey was not completed within 7 days. 3. Results 3.1. Response rate
1 Pharmacists in the UK are banded according to “Agenda for Change.” The band ranges from band 5 (preregistration pharmacist) to band 9, which would typically be a chief pharmacist in a tertiary referral institution. Band 8 rages from a to d; band 8a is considered a senior level and would normally be a pharmacist who has completed their foundation clinical training.
A total of 178 pharmacists responded from 157 hospitals from within the UK. This represents approximately 60% of UK ICUs and 70% of those with a clinical pharmacy service. Where an ICU did not complete a survey, telephone follow-up call was made to confirm if there was a
S.M. Yassin et al. / Journal of Critical Care 30 (2015) 436.e1–436.e6
clinical pharmacy service to the ICU. Partially completed surveys were included. 1. Workforce data On average, the pharmacist completing the survey was funded for 0.5 whole time equivalents (WTE) by the ICU. In 14 cases (8%), the service was not financially supported by the ICU. Fifteen ICUs contacted remain without any service. The length of time the pharmacist spent on the ICUs varied from 0.1 to 1.0 WTE. Almost 50% of pharmacists work as part of a team. More than 85% of respondents were band 8a or above. Critical care demographics The number of level 3 beds per ICU varied from 3 to 45 (mean, 11). Twenty-four units reported a split between level 2 and 3 beds with a mean of 10 level 3 beds (overall ratio of 1 level 3 to 0.7 level 2 beds). The mean number of reported annual intensive care admissions was 825. Most of the ICUs were mixed, with more surgical than medical patients. Nine ICUs had at least 90% cardiothoracic patients, and 4 were predominantly neurosurgical. 2. Policy Eighty-nine respondents (55%) report using sedation guidelines, and 125 (78%) of all respondents work in units that undertake sedation holds. Sedation scoring systems are commonly used (87%), although a wide range of tools were reported (Table 1). The most common was the Richmond Agitation Sedation Score (RASS) with 40 (27%) of respondents reporting use. Despite high use of sedation scoring systems, less than half (42%) of ICUs target sedation to a score on the daily ward rounds. Those units who targeted a sedation score were just as likely to us sedation holds. Seventy ICUs (43%) regularly assess for delirium. Seventy (43%) do not. A proportion (14%) of responders did not know. Of those who did, 27 reported use of a formal scoring system. The Confusion Assessment Method for the ICU (CAM-ICU) is used in 20 of these ICUs, and the Intensive Care Delirium Screening Checklist is used in 6. Forty-four responses indicated that although delirium is assessed, a scoring system is not used. 3. Sedative choice The most commonly used sedative was propofol. Propofol is used first line for most patients in 136 ICUs (89%) (Fig. 1, Table 2). It is the preferred sedative for most indications including short-term sedation (93%) and for use in hepatic or renal failure (80%). Midazolam use remains common. It is considered for most patients in 49% of units. The use of MDZ for short-term sedation or patients with hepatic or renal failure is lower at 12% and 26%, respectively. In patients who require frequent neurologic assessment, propofol use is high, with 100 ICUs (75%) reporting use. MDZ is used rarely in such patients, but still, 11 units (8%) describe prescription. Sedative choice did not differ between units that did or did not practice sedation holds. Opioid-based sedation was not addressed directly by this survey; however, 14 ICUs (9%) specifically commented that they commonly use opioid-based sedation, predominantly with fentanyl or alfentanil. This rises to 19 (13%) when short-term sedation is the priority and to 30 units (20%) when patients have renal or hepatic dysfunction. Forty-six ICUs use opioid-based sedation in
Table 1 Range of sedation scoring systems described Sedation scoring system
n (%)
RASS Ramsay SAS (Riker) Scale Don't know Other
40 (27%) 28 (19%) 8 (5%) 48 (33%) 27 (18%)
436.e3 Midazolam Propofol
100 90 80 70 60 50 40 30 20 10 Propofol
0 Used in majority of Used for short term patients sedation (up to 48hours)
Midazolam Used in hepatic or renal failure
Used when frequent Used where sedation neurological is challenging assessment required
Fig. 1. Differing described use of midazolam and propofol in selected indications (separate).
those requiring frequent neurologic assessment. Remifentanil is chosen most frequently (n = 22 reported) in such circumstances. When sedation is more challenging or in patients with a history of alcohol or drug abuse, then choice of sedative is more variable (Table 2). In such cases, MDZ is used more commonly than propofol (73% vs 68%). Clonidine is considered in 44%. Haloperidol, ketamine, lorazepam, or diazepam use is also reported. During status epilepticus, sedative choice is split between propofol (54%) and MDZ (48%), then lorazepam (38%) and thiopentone (28%). Sedatives administered orally comprise haloperidol (54%), diazepam (42%), lorazepam (42%), and clonidine (41%). Clonidine is the most frequently used adjuvant (68%), followed by haloperidol (40%), MDZ (27%), and lorazepam (24%). During longterm ventilatory weaning, propofol (44%) is most commonly used, although this is closely followed by clonidine (40%). MDZ is still being prescribed in 28% of ICUs. Many commented that later in their stay, most patients have tracheostomies in place and require little, if any, sedation. In alcohol withdrawal or a history of alcohol abuse, benzodiazepines feature highly, although propofol or clonidine are also used. Some pharmacists also commented that chlordiazepoxide is prescribed (n = 25). The response rate for questions concerning head injury patients was low. At least 41 respondents do not see this patient group. Fifty-nine pharmacists (56%) that responded describe propofol prescription for patients with raised intracranial pressure. Midazolam (20%) or thiopentone (20%) is less commonly prescribed. Ketamine is more likely to be used for bronchospasm than any other indication and is considered in 79 units (66%), although it remains a rarely prescribed agent with 20% of respondents not seeing it used for at least 5 years and 57% reporting its use less than 5 times a year. Of the drugs listed, the least likely prescribed is thiopentone. Fifty-six percent of ICUs have not used it for 5 years, and 65% of ICUs use it less than 5 times a year. 4. Analgesia Morphine is the most common first-line prescription described in 68 ICUs (49%) (Table 3). Morphine is closely followed by the synthetic opioids, fentanyl (34%) and alfentanil (34%). Paracetamol is also a commonly prescribed analgesic (32%). In patients requiring frequent neurologic assessment or where drug accumulation is of concern, morphine remains a consideration in 15 (12%) and 16 (12%) of units, respectively. Remifentanil is the most popular choice for patients with uncertain neurology (50%), and both alfentanil (53%) and remifentanil (44%) are commonly prescribed when drug accumulation is of concern. The opioid most commonly prescribed for withdrawal or dependence is methadone (60%). Gabapentin (87%) is the most common agent for neuropathic pain followed by amitriptyline
436.e4
S.M. Yassin et al. / Journal of Critical Care 30 (2014) 436.e1–436.e6
Table 2 Sedative choice in described indication Sedative name
Propofol n (%) MDZ
Lorazepam Clonidine Diazepam Thiopentone Ketamine Haloperidol Other
136 (89) 141 (93) 115 (80) 100 (75) 97 (68) 76 (54) 2 (2) 7 (5) 60 (44) 58 (43)
2 (1) 2 (1) 4 (3) 0 13 (9) 53 (38) 56 (42) 35 (24) 11 (8) 27 (20)
11 (7) 11 (7) 12 (8) 2 (2) 63 (44) 1 (1) 54 (41) 98 (68) 54 (40) 57 (42)
0 2 (1) 1 (1) 0 6 (4) 16 (11) 56 (42) 21 (15) 18 (13) 32 (24)
0 0 0 2 (2) 3 (2) 39 (28) 0 0 0 0
1 (1) 2 (1) 1 (1) 1 (1) 13 (9) 0 0 11 (8) 0 0
20 (20) 0 39 (33) 0 1 (1) 10 (13) 5 (4) 33 (27)
0 3 (3) 3 (4) 9 (7)
0 0 18 (23) 36 (29)
20 (20) 1 (1) 43 (56) 80 (65)
1 (1) 79 (66) 15 (20) 70 (57)
Indication Used in majority of patients Used for short-term sedation (up to 48 h) Used in hepatic or renal failure Used when frequent neurologic assessment required Used where sedation is challenging Used for status epilepticus Used for oral or enteral sedation Used as an adjuvant agent Used for weaning from long-term ventilation Used for alcohol withdrawal or known history of alcohol abuse Used for intracranial pressure control in head injury patients Used in bronchospasm Not used in clinical practice for last 5 y Used less frequently than 5 times a year
57 (56) 64 (53) 1 (1) 0
75 (49) 18 (12) 37 (26) 11 (8) 105 (73) 68 (48) 7 (5) 39 (27) 38 (28) 60 (44)
4 (3) 4 (3) n/a 0 22 (15) n/a 72 (54) 58 (40) 20 (15) 12 (9)
15 (10) 20 (13) 29 (20) 46 (33) 30 (21) 12 (9) 22 (17) 33 (23) 25 (19) 37 (27)
0 1 (1) 6 (8) 9 (7)
42 (43) 22 (18) 12 (16) 4 (3)
More than 1 answer allowed per question.
(51%). Pregabalin prescription was described by 9 respondents. Paracetamol is commonly prescribed as an adjuvant agent (96%), although nonsteroidal anti-inflammatory agents (35%) are also used. Paracetamol is commonly prescribed as an antipyretic (84%). In at least 25 ICUs, however, it is not routine practice to reduce temperature. 4. Discussion The investigators have achieved their aim to describe a detailed perspective of UK sedation practice. This detailed perspective confirms that sedation practice remains highly varied and that strategies proven to improve it are not uniformly adopted. More than 55% of respondents use a sedation guideline. This is lower than the 80% reported in the study by Reschreiter et al [5]. Guidelines are a highly effective tool in ensuring consistency of clinical practice. Indeed, Mehta et al [18] in 2012 described that an effective protocolized sedation policy may negate the need for sedation holds when measured in terms of days on mechanical ventilation. We found that sedation scores are commonly used in the UK (87%), confirming the result of Reschreiter et al (89%) [5]. The systems used have changed, with a higher uptake of the RASS reported by our survey at 27%. Reschreiter et al reported just 5.4% of ICUs using RASS [5]. The RASS has gained favor in recent years with evidence for its reliability and validity [19]. It is preferred over the Ramsay sedation scale, as it has more discrete criteria to assess different levels of sedation, hence a more sensitive marker for monitoring effects of sedatives over time
[20]. There remains, however, a large variation in scale used, confirming the lack of a singular score in the UK. This is a finding common to other national surveys where uptake of scoring systems has been slower and variation still exists. For example, Patel et al [12] in America found a split of 37% using the Ramsay scale with 26% using RASS. Only 44% of UK ICU set a daily sedation score target on the medical ward round. One of the core strategies to improve sedation locally in our trust is to set a target RASS daily on the ward round. Our survey confirmed that sedation holds are now a routine part of prescribing practice in the UK with 80% of respondents reporting use of formal sedation holds. Kress et al [11] described, in 2000, a reduction in both duration of mechanical ventilation and length of stay using this technique, but as mentioned previously, use of effective protocols may well reduce the need [18]. The ideal strategy is probably a combination of a unit guideline, a daily target RASS, and sedation hold. The original sedation hold work took place in an ICU that prescribed morphine and MDZ [11]. It may be that sedation holds are of greatest benefit in the presence of longer active sedative and opioid. In this survey, we found no association between MDZ use, targeting of sedation scores, and presence of sedation holds. Delirium is a frequent occurrence in the ICU and can affect up to two thirds of ventilated patients [21]. Affected patients have 3 times the mortality risk and 9 times the risk of long-term cognitive impairment [22]. Screening tools are advocated to identify such patients especially those who are not overtly agitated [21]. Approximately half of respondents who answered this question indicated that their ICU assessed for delirium. Only 27 describe use of a scoring system. This low uptake
Table 3 Analgesic choice in described indication Analgesic name
Alfentanil
Fentanyl
Gabapentin
Methadone
Morphine
NSAIDs
48 (34) 50 (38) 42 (33)
48 (34) 41 (31) 21 (17)
1 (1) 0 0
0 4 (3) 0
68 (49) 59 (45) 15 (12)
0 0 1 (1)
72 (53)
42 (31)
0
0
16 (12)
44 (35)
35 (28)
0
0
43 (31) 25 (20) 1 (1) 3 (2) 0 20 (26) 27 (24)
51 (37) 15 (12) 10 (7) 13 (10) 0 13 (17) 22 (20)
3 (2) 2 (2) 118 (87) 33 (25) 0 3 (4) 11 (10)
0 76 (60) 2 (1) 2 (2) 0 30 (37) 51 (46)
Paracetamol
Remifentanil
Amitriptyline
Other
45 (32) 21 (16) 26 (20)
13 (9) 26 (20) 63 (50)
– – –
3 (2) 7 (5) 15 (12)
0
27 (20)
59 (44)
–
8 (6)
45 (36)
3 (2)
25 (20)
24 (19)
–
4 (3)
98 (72) 60 (48) 4 (3) 15 (11) 0 1 (1) 4 (4)
11 (8) 5 (4) 2 (1) 46 (35) 6 (4) 13 (17) 32 (29)
62 (45) 18 (14) 24 (18) 128 (96) 113 (84) 0 0
22 (16) 11 (9) – 0 0 18 (23) 26 (23)
– – 69 (51) – – – –
9 (7) 22 (17) 24 (18) 9 (7) 4 (3) 8 (10) –
Indication Used as first-line agent Used where respiratory depression is essential Used in patients who require frequent neurologic assessment Used in hepatic or renal failure where drug accumulation is a concern Used post cardiac arrest in patients who are therapeutically cooled Used for analgesia-based sedation Used for opiate withdrawal/dependence Used routinely in patients for neuropathic pain Used as an adjuvant agent Used as an antipyretic Not in clinical practice for last 5 y Less than 5 times a year in clinical practice
More than 1 answer allowed per question. NSAID indicates nonsteroidal anti-inflammatory drug.
S.M. Yassin et al. / Journal of Critical Care 30 (2015) 436.e1–436.e6
of sensitive diagnostic tools has been reported in previous studies in the UK, and it is highly likely that delirium is underdiagnosed [23]. Surveys in other countries show similar findings with use of delirium scoring systems ranging from 3.7% in Canada in 2002 [24], 13% in Brazil in 2008 [25], and up to 33% in America in 2007 [12]. Of note, some of the sedation surveys reviewed did not include questions about delirium [5,13,26]. Perhaps the newly published national standards for ICU, which include recommendations on managing delirium, will ensure a wider uptake of delirium scoring tools [27]. Propofol is now the most commonly used sedative agent. It is short acting and without active metabolites [28]. Midazolam remains a frequently used sedative for invasively ventilated patients. The complex pharmacokinetics and risk of oversedation mean that its use should be targeted at younger patients without cardiac failure or those with single-organ respiratory failure [7]. It is surprising that MDZ is still widely used in patients, even with renal or hepatic failure. Midazolam is used less commonly for short-term sedation now as was also found in the Reschreiter survey [5]. This prescribing trend was also found in Germany [13]. Lorazepam has more favorable kinetics in critical illness; however, it has been associated with delirium, and high doses can cause metabolic acidosis [9,10]. This delirium was reported after much larger intravenous doses, which are commonly used in the United States. There is less prevalence of delirium in doses less than 20 mg/d [9]. Lorazepam should be reserved for use as an adjuvant in small enteral or parenteral doses in critical illness. Clonidine is frequently used in long-term sedation for challenging patients or when weaning. Supporting evidence for its use is lacking, and effective dosing tends to be limited by adverse hemodynamic effects. The recent licensing in the UK of dexmedetomidine has reintroduced the concept of awake sedation and the use of α2 agonists for sedation in the critically ill. This follows publication of the MIDEX and PRODEX studies demonstrating noninferiority with MDZ and propofol [29]. Comparative sedative cost dictates that dexmedetomidine should be reserved for use in those patients where there is most likely to be benefit [30]. Dexmedetomidine was not included in this sedation survey, although there are plans to include it in the next version. Opioid choice is fairly evenly spread between morphine, fentanyl, and alfentanil for most indications. When accumulation is a concern, then shorter acting agents are more popular such as with frequent neurologic examinations or organ failure, as described in Fig. 2. The evidence for shorter acting opioids over morphine is not conclusive; thus, our results are not unexpected, although accumulation of morphine's glucuronide metabolites is well described in renal failure [6]. Given that as much as 50% of ICU patients can develop some form of renal failure the investigators purport that widespread morphine use is perhaps surprising despite the availability and low relative cost of both fentanyl and alfentanil. Up to 30 units in this survey described use of analgesic-based sedation. The advantage of this is obvious, as it aims to keep patients calm rather than asleep and allows the patient to communicate their needs. It can also avoid the deleterious effects of benzodiazepines. Newer agents such as dexmedetomidine further facilitate awake sedation in a proportion of patients [30]. Use of paracetamol as an antipyretic appears to be common despite lack of solid evidence of benefit in sepsis. Almost half of those who responded describe only using paracetamol in temperatures greater than 39.5°C, which concurs with national recommendations. In accordance with National Institute of Clinical Excellence, gabapentin is the preferred agent for neuropathic pain, although 9 ICUs favor pregabalin, with its more linear pharmacokinetics [31]. These are probably preferred agents in the ICU as they have evidence of rapid effectiveness: essential in the critically ill patient with neuropathic pain. In summary, we have demonstrated that an Internet-based survey directed at clinical pharmacists is both a feasible and effective method for data collection on prescribing in the ICU. Given that 85% were senior
436.e5 Alfentanil Fentanyl Morphine
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0% Used as first line agent
Used in patients who require frequent neurological assessment
Used in hepatic or renal failure where drug accumulation is a concern
Fig. 2. Opioid choice in selected indications.
(band 8a or above) and spent a mean of half their time in the ICU, one would expect a sound knowledge of sedation practice. We achieved a response rate of 60% that provided a detailed insight into sedation and analgesic practice. Workforce demographics were recorded, but this was not the primary focus of this survey. There is potential to use these national pharmacy groups to gain a wide perspective of prescribing practice within the ICU and wider critical care environment. Previous UK surveys have used a paper format to conduct their survey [5,15]. Other countries have used electronic surveys successfully, although this can result in convenience samples and their associated bias [25]. This survey targeted pharmacists and had a better response rate than previous surveys that targeted nurses and physicians [26]. This could be attributed to the diligent follow-up of nonresponders rather than the profession itself. There are limitations of the study. The investigators chose the ICU pharmacist as their point of contact, but input varies. Some units do not have a dedicated pharmacist, and therefore, their knowledge of sedation practice may vary. This may be reflected in many responses. In addition, there was also a variation in description of unit specialty, which limited analysis of the data. In future surveys, this will be standardized. The units that did indicate specialism comprised 9 cardiothoracic and 4 neurosurgical ICUs, and it was felt that the number was too small for subgroup analysis. All the responses were, therefore, analyzed together. Although there were sections on delirium scoring, the investigators did not include questions on antipsychotic agents. In addition, there was no question on the use of pain scores. The investigators intend to repeat the survey over 2015 and include a section on antipsychotics and a question on behavioral pain scores. Despite the investigator's very best efforts, an overall response rate of just 60% was attained, leaving the sedation practice for 40% of units unaccounted for. Lastly, selfreported bias may mean that some of the responses did not represent a fully accurate reflection of their unit practice.
5. Conclusions The reported sedation survey has succeeded in its overall aims and has provided a detailed perspective of UK sedation practice. This included sedation guidance and holds, prevalence of delirium assessment, sedative choice, and analgesic choice in a number of clinical scenarios. It has also provided national trends in sedation prescribing practice, and finally, with a response rate of 70% in the presence of a clinical pharmacy service, it has confirmed that the pharmacy workforce can be a source of prescribing practice description in the ICU.
436.e6
S.M. Yassin et al. / Journal of Critical Care 30 (2014) 436.e1–436.e6
Competing interests Dr McKenzie has received honoraria from Orion Pharma and GlaxoSmithKline Kline for educational events and advisory boards.
[13]
[14]
Authors' contributions [15]
Dr McKenzie and Dr Terblanche devised the original survey. Mrs Yassin collected the results and undertook all initial analysis. Dr Yassin reviewed and made extensive comment in preparation of the revised manuscript. All authors contributed to the manuscript preparation. Acknowledgments
[16]
[17]
[18]
The investigators would like to thank the UKCPA CCG for completing the survey and Ms Joyce Lim for assisting in the analysis.
[19]
References
[20]
[1] Jacobi J, Fraser GL, Coursing DB, Riker RR, Fontaine D, Wittered ET, et al. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119–41. [2] Park GR, Ward B. Sedation and analgesia in the critically ill. In: Darrell D, Cox T, Firth J, Benz, editors. Oxford Textbook of Medicine. 4th ed. Oxford University Press; 2003. p. 1250–6. [3] Strøm T, Martinussen T, Toft P. A protocol of no sedation for critically ill patients receiving mechanical ventilation: a randomised trial. Lancet 2010;375(9713):475–80. http://dx.doi.org/10.1016/S0140-6736(09)62072-9 [Epub. 2010 Jan 29]. [4] Riker RR, Fugate JE, the Participants in the International Multi-disciplinary Consensus Conference on Multimodality Monitoring. Clinical monitoring scales in acute brain injury: assessment of coma, pain, agitation, and delirium. Neurocrit Care 2014 [Pub ahead of print]. [5] Reschreiter H, Maiden M, Kapila A. Sedation practice in the intensive care unit: a UK national survey. Crit Care 2008;12:R152. [6] Gordon HL. Morphine intoxication in renal failure: the role of morphine-6glucuronide. Br Med J (Clin Res Ed) 1986;293(6550):818–9. [7] McKenzie CA, McKinnon W, Naughton DP, Treacher D, Davies G, Phillips GJ, et al. Differentiating midazolam over-sedation from neurological damage in the intensive care unit. Crit Care 2005;9:R32–6. [8] Bauer TM, Ritz R, Haberthür C, Ha HR, Hunkeler W, Sleight AJ, et al. Prolonged sedation due to accumulation of conjugated metabolites of midazolam. Lancet 1995;346 (8968):145–7. [9] Pandharipande P, Shintani A, Peterson J, Pun BT, Wilkinson GR, Dittus RS, et al. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anaesthesiology 2006;104(1):21–6. [10] Wilson KC, Reardon C, Theodore AC, Farber HW. Propylene glycol toxicity: a severe iatrogenic illness in ICU patients receiving IV benzodiazepines: a case series and prospective, observational pilot study. Chest 2005;128(3):1674–81. [11] Kress JP, Pohlman AS, O'Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 2000;342: 1471–7. [12] Patel R, Gambrell M, Speroff T, Scott T, Pun B, Okahashi J, et al. Delirium and sedation in the intensive care unit: survey of behaviours and attitudes of 1384 healthcare pro-
[21]
[22]
[23] [24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
fessionals. Crit Care Med 2009;37:825–32. http://dx.doi.org/10.1097/CCM. 0b013e31819b8608. Martin J, Franck M, Sigel S, Weiss M, Spies C. Changes in sedation management in German intensive care units between 2002 and 2006: a national follow-up survey. Crit Care 2007;11:R124. http://dx.doi.org/10.1186/cc6189. Tanios M, de Wit M, Epstein SK, Devlin JW. Perceived barriers to the use of sedation protocols and daily sedation interruption: a multidisciplinary survey. J Crit Care 2009;24:66–73. http://dx.doi.org/10.1016/j.jcrc.2008.03.037. Ramaswamy S, Rangappa D, Wheatley E. National survey on current practice of sedation and analgesia in intensive care in United Kingdom. Intensive Care Med 2009;35(S112) [0342-4642 (abstract only)]. Leape LL, Cullen DJ, Clapp MD, Burdick E, Demonaco H, Erickson JI, et al. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. JAMA 1999;282(3):267–70. Marshall JJ, Finn CA, Theodore AC. Impact of a clinical pharmacist-enforced intensive care unit sedation protocol on duration of mechanical ventilation and hospital stay. Crit Care Med 2008;36(2):427–33. Mehta S, Burry L, Cook D, Fergusson D, Steinberg M, Granton J, et al. Daily sedation interruption in mechanically ventilated critically ill patients cared for with a sedation protocol. JAMA 2012:1985–92. http://dx.doi.org/10.1001/jama.2012.13872. Sessler CN, Gosnell M, Grap MJ, Brophy GT, O'Neal PV, Keane KA, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care patients. Am J Respir Crit Care Med 2002;166:1338–44. Ely EW, Truman B, Shintani A, Thomason JWW, Wheeler AP, Gordon S, et al. Monitoring sedation status over time in ICU patients: the reliability and validity of the Richmond Agitation Sedation Scale (RASS). JAMA 2003;289:2983–91. Ely EW, Inouye SK, Bernard GR, Gordon S, Francis J, May L, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA 2001;286(21):2703–10. Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrell Jr FE, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA 2004;14:1753–62. Mac Sweeney R, Barber V, Page V, Ely EW, Perkins GD, Young JD, et al. A national survey of the management of delirium in UK intensive care units. QJM 2010;103(4):243–51. Mehta S, Burry L, Fischer S, Martinez-Motta C, Hallett D, Bowman D, et al. Canadian survey of the use of sedatives, analgesics, and neuromuscular blocking agents in critically ill patients. Crit Care Med 2006:34374–80. http://dx.doi.org/10.1097/01.CCM. 0000196830.61965.F1. Salluh JIF, Dal-Pizzol F, Mello P, Friedman G, Silva EMD, Teles JMM, et al. Delirium recognition and sedation practices in critically ill patients: a survey on the attitudes of 1015 Brazilian critical care physicians. J Crit Care 2009;24:556–62. O'Connor M, Bucknall T, Manias E. Sedation management in Australian and New Zealand intensive care units: doctors' and nurses' practices and opinions. Am J Crit Care 2010;19:285–95. Intensive Care Society/Faculty of Intensive Care Medicine. Core Standards for the Intensive Care Unit 2013, Edition 1. Available from http://www.ics.ac.uk/icshomepage/guidelines-standards. Astra Zeneca UK ltd. Sedation and analgesia in the critically ill. Summary of product characteristics for Diprivan 1%. Last updated on the eMC; 2012. Jakob SM, Ruokonen E, Grounds RM, Sarapohja T, Garratt C, Pocock SJ, et al. Dexmedetomidine for Long-Term Sedation Investigators: dexmedetomidine vs midazolam or propofol for sedation during prolonged mechanical ventilation: two randomized controlled trials (PRODEX and MIDEX). JAMA 2012;21:1151–60. Pasin L, Greco T, Feltracco, Vittorio A, Neto CN, Cabrini L, Landoni G, Finco G, Zangrillo A. Dexmedetomidine as a Sedative Agent in Critically Ill Patients: A Meta-Analysis of Randomized Controlled Trials. PLoS One 2014 Accessed from http://www.ncbi.nlm.nih.gov/pubmed/24391726. Pfizer UK ltd. Summary of product characteristics for Lyrica. Last updated on the eMC; 2013.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Electronic Articles
A web-based survey of United Kingdom sedation practice in the intensive care unit☆,☆☆ Sarah M. Yassin, MRPharmS a,c, Marius Terblanche, FRCA, MBBS b,c, James Yassin, MRCP, FRCA d, Catherine A. McKenzie, PhD, FFPharmS a,c,⁎ a
Institute of Pharmaceutical Sciences, Franklin-Wilkins Building, Kings College, London, SE1 7RT, UK School of Medicine, Guys Campus, King's College, London, SE1 9RT, UK Pharmacy and Critical Care Guys and St Thomas' NHS Foundation Trust (GSTFT), St Thomas' Hospital, London, SE1 7EH, UK d Department of Critical Care, Royal Sussex County Hospital, Brighton, BN2 5BE, UK b c
a r t i c l e
i n f o
Keywords: Sedation Midazolam Propofol Intensive care Morphine Delirium
a b s t r a c t Purpose: The purpose of this work was to obtain a detailed perspective of sedation practice. Sedation included sedative and opioid choice, presence of local guidelines, and use of scoring systems. Methods: A Web-based survey was designed. The aim was to gain sufficient detail of UK sedation while also being succinct enough to complete in 15 minutes. It was composed of relevant demographics, policy, sedative choice, and analgesia. The survey was piloted before launch. The investigators selected the intensive care unit (ICU) pharmacist as the respondent. Results: One hundred fifty-seven ICUs responded. Eighty-nine (59%) reported use of sedation guidelines, 78% undertook sedation holds, and 87% use sedation scores. Only 42% used a daily sedation target. Seventy (43%) assess for delirium; 27 of those use a validated tool. Propofol (89%) use was common, followed by midazolam (49%). Morphine (49%), fentanyl (34%), and alfentanil (34%) were the most frequently used opioids. Conclusion: This survey confirmed expected variation in UK sedation practice. Recognized strategies such as target sedation score and sedation policy are underused. A 43% uptake in delirium screening suggests that larger engagement is required to meet national standards. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Sedation is required for invasively ventilated patients [1]. A combination of sedatives and opioids is commonly used. Opioids are used as analgesics or antitussives and to aid ventilator synchrony [2]. Hypnotics, commonly either propofol or a benzodiazepine, are for amnesia, anxiolysis, and somnolence [2]. The choice of both opioid and sedative remains controversial, and more recently, there has been publication questioning the need for a hypnotic or “comfort” agent in its entirety [3]. There is consensus over sedative selection in specific situations, such as short-acting agents in neurologic assessment [4]. However, there remains equipoise in overall sedation prescribing practice within the intensive care unit (ICU) [5]. ☆ This work was undertaken from a grant awarded from Guys and St Thomas' Charity. ☆☆ The research was also supported by the National Institute for Health Research Biomedical Research Centre based at Guy's and St Thomas' National Health Service Foundation Trust and King's College London. The views expressed are those of the author (s) and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health. ⁎ Corresponding author. Guys and St Thomas' NHS Foundation Trust, Pharmacy and Critical Care, St Thomas' Hospital, Westminster Bridge Road, London SE1 7EH. Tel.: +44 7796193246. E-mail address: [email protected] (C.A. McKenzie). http://dx.doi.org/10.1016/j.jcrc.2014.11.006 0883-9441/© 2014 Elsevier Inc. All rights reserved.
Evidence in sedation has grown over the last 2 to 3 decades. There is now sufficient evidence to suggest that certain agents (hypnotics and opioids) may be harmful in specific patient groups. One of the earliest reports of opioid accumulation occurred in 1984 when Gordon [6] reported accumulation of morphine glucuronide metabolites and prolonged effect in renal failure. The benzodiazepines, lorazepam, and midazolam (MDZ) are also reported to be associated with harm. Evidence suggests that MDZ's complex pharmacokinetic profile may predispose specific patient groups to oversedation [7,8]. This includes those with renal failure and the elderly with or without compromised cardiac function [7]. This resulted in the recommendation to use lorazepam in the United States [1]. Lorazepam remained widely used, particularly in the United States, until Pandharipande et al [9] in 2006 reported it as the highest independent risk factor for development of critical care delirium. In addition, there were reports of metabolic acidosis after prolonged infusions of lorazepam [10]. With continuing emergence of new therapies and specific interventions such as sedation holds [11], there is a need to maintain a perspective of UK wide sedation prescribing. In 2005, a piece of work undertaken by the investigators reported that more than 90% of ICUs still use morphine and MDZ first line for sedation in ventilated patients [7]. This was despite evidence that both agents have active glucuronide metabolites that accumulate in renal failure [7,8].
436.e2
S.M. Yassin et al. / Journal of Critical Care 30 (2014) 436.e1–436.e6
Surveys are useful tools to describe sedation practice and to monitor changes in practice over time. In the UK in 2008, Reschreiter et al [5] published a postal survey of sedation practice. The investigators described a wide variation in practice but noted an increase in uptake of sedation scoring systems from a previous survey in 2000 [5]. Use of validated sedation scales has also increased in America, up to 88% in 2007 [12], and in Germany, use of a scale doubled between 2002 and 2006 [13]. The German survey also showed a reduction in use of MDZ for short-term sedation from 46% to 36% [13]. Changes in practice, however, are known to lag behind both evidence and national guidelines [14]. Most of these surveys were paper-based, postal surveys. They involved contact with the medical consultant lead of the ICU who would either complete the survey on behalf of the ICU [5,15] or delegate to a colleague. One of the advances in the ICU over the last 10 years has been the establishment of the intensive care pharmacist [16,17]. Pharmacists are often involved in the development of medicines related guidelines and could be considered to observe overall unit practice. This and the fact that the UK national group, the United Kingdom Clinical Pharmacy Critical Care Group (UKCPA CCG), has strong links with the investigating team were influential in the decision to select the critical care pharmacist as the survey respondent. 1.1. Aim and objectives The primary aim of this sedation survey was to gain a detailed perspective of UK sedation practice. Objectives were as follows: To establish national trends in prescribing and review applications or techniques to assist in safe prescribing To describe the pharmacy intensive care workforce and demographics in the context of the pharmacist's ability to report prescribing practice in the ICU and wider critical care environment. 2. Materials and methods The survey protocol was submitted to the local research ethics committee at St Thomas' Hospital. Approval was granted on chairman's action, and the survey could proceed without the requirement for a full ethics application. The study was registered with and approved by Guys and St Thomas' NHS Foundation Trust Research and Development (RJ110/N194). 2.1. Survey design The survey was designed using a Web-based provider, www. zoomerang.com. The questionnaire aimed to cover as much sedation practice as possible while being sufficiently brief to complete in 15 minutes. The questionnaire included a wide selection of questions that were deemed essential to gain a picture of sedation practice over the UK. Flexibility was allowed for more than 1 reply per trust to allow for practice within specialist critical care areas to be described separately. The survey questionnaire was split into 4 sections: 1. Workforce and demographics: This section was composed of detail regarding the hospital and its critical care unit(s), including number of level 3/2 beds and patient mix. There was also a subsection on pharmacist workforce, including length of time spent on the ICU, band/grade, and whether they worked in isolation or as part of a team. 1
2. Policy: Most questions in this section were derived from evidence-based recommendations on sedation practice. These included whether the ICU had a formal sedation policy; undertook sedation holds; what type of sedation scoring system was in place; whether sedation was titrated to a score daily; and whether units regularly assessed patients for delirium; and, if so, which (if any) delirium assessment method was used. 3. Sedative choice: This comprised detailed questions on the selection of sedatives in most common clinical scenarios encountered within ICU. This included most commonly used agents, sedative selection for short-term sedation, selection for multiple organ failure, agents for status epilepticus, drug withdrawal or dependence, weaning agents, and raised intracranial pressure. More than 1 answer was allowed per question, for example, sedatives used in the majority of patients. 4. Analgesia: This primarily concerned opioids. Questions included first-line agents, short-term agents, agents when increased pain is anticipated, therapeutic hypothermia, analgesia-based sedation, enhanced respiratory depression agents, and antipyretics in the ICU. 2.2. Mandatory and nonmandatory questions A number of the questions were considered essential to gain an overall perspective of sedation practice. These included questions on whether the ICU had sedation/analgesia guideline. These questions were designated as “mandatory,” which meant the respondent could not proceed unless they were completed. 2.3. Survey prelaunch pilot To ensure that the survey was robust, sensitive, and reliable, it was piloted on a number of professional groups. At the design phase, it was presented to the critical care research group, including intensivists, nurses, physiotherapists, and pharmacists. This group commented extensively on survey content. It was reviewed by 5 senior pharmacists who agreed the overall design of each question. After this review, the survey was considered ready. To test the validity of the survey immediately before launch, a pilot version was emailed to 2 consultant pharmacists for final comment. 2.4. Survey site After design and piloting, the survey was posted via the UKCPA CCG Web site (www.ukcpa.org.uk). To ensure maximum response: the Web link was also posted on the infection, cardiology, and general discussion message boards. Accompanying the survey link was a cover letter, which informed the respondent of the detail required for completion. 2.5. Response rate To maximize response, the investigators compared the list of responding hospitals against those using the Intensive Care National Audit and Research Council case mix program [5]. Where there was a gap in response, the investigators contacted the pharmacist firstly by email and then telephone to request survey completion. A follow-up call was made if the survey was not completed within 7 days. 3. Results 3.1. Response rate
1 Pharmacists in the UK are banded according to “Agenda for Change.” The band ranges from band 5 (preregistration pharmacist) to band 9, which would typically be a chief pharmacist in a tertiary referral institution. Band 8 rages from a to d; band 8a is considered a senior level and would normally be a pharmacist who has completed their foundation clinical training.
A total of 178 pharmacists responded from 157 hospitals from within the UK. This represents approximately 60% of UK ICUs and 70% of those with a clinical pharmacy service. Where an ICU did not complete a survey, telephone follow-up call was made to confirm if there was a
S.M. Yassin et al. / Journal of Critical Care 30 (2015) 436.e1–436.e6
clinical pharmacy service to the ICU. Partially completed surveys were included. 1. Workforce data On average, the pharmacist completing the survey was funded for 0.5 whole time equivalents (WTE) by the ICU. In 14 cases (8%), the service was not financially supported by the ICU. Fifteen ICUs contacted remain without any service. The length of time the pharmacist spent on the ICUs varied from 0.1 to 1.0 WTE. Almost 50% of pharmacists work as part of a team. More than 85% of respondents were band 8a or above. Critical care demographics The number of level 3 beds per ICU varied from 3 to 45 (mean, 11). Twenty-four units reported a split between level 2 and 3 beds with a mean of 10 level 3 beds (overall ratio of 1 level 3 to 0.7 level 2 beds). The mean number of reported annual intensive care admissions was 825. Most of the ICUs were mixed, with more surgical than medical patients. Nine ICUs had at least 90% cardiothoracic patients, and 4 were predominantly neurosurgical. 2. Policy Eighty-nine respondents (55%) report using sedation guidelines, and 125 (78%) of all respondents work in units that undertake sedation holds. Sedation scoring systems are commonly used (87%), although a wide range of tools were reported (Table 1). The most common was the Richmond Agitation Sedation Score (RASS) with 40 (27%) of respondents reporting use. Despite high use of sedation scoring systems, less than half (42%) of ICUs target sedation to a score on the daily ward rounds. Those units who targeted a sedation score were just as likely to us sedation holds. Seventy ICUs (43%) regularly assess for delirium. Seventy (43%) do not. A proportion (14%) of responders did not know. Of those who did, 27 reported use of a formal scoring system. The Confusion Assessment Method for the ICU (CAM-ICU) is used in 20 of these ICUs, and the Intensive Care Delirium Screening Checklist is used in 6. Forty-four responses indicated that although delirium is assessed, a scoring system is not used. 3. Sedative choice The most commonly used sedative was propofol. Propofol is used first line for most patients in 136 ICUs (89%) (Fig. 1, Table 2). It is the preferred sedative for most indications including short-term sedation (93%) and for use in hepatic or renal failure (80%). Midazolam use remains common. It is considered for most patients in 49% of units. The use of MDZ for short-term sedation or patients with hepatic or renal failure is lower at 12% and 26%, respectively. In patients who require frequent neurologic assessment, propofol use is high, with 100 ICUs (75%) reporting use. MDZ is used rarely in such patients, but still, 11 units (8%) describe prescription. Sedative choice did not differ between units that did or did not practice sedation holds. Opioid-based sedation was not addressed directly by this survey; however, 14 ICUs (9%) specifically commented that they commonly use opioid-based sedation, predominantly with fentanyl or alfentanil. This rises to 19 (13%) when short-term sedation is the priority and to 30 units (20%) when patients have renal or hepatic dysfunction. Forty-six ICUs use opioid-based sedation in
Table 1 Range of sedation scoring systems described Sedation scoring system
n (%)
RASS Ramsay SAS (Riker) Scale Don't know Other
40 (27%) 28 (19%) 8 (5%) 48 (33%) 27 (18%)
436.e3 Midazolam Propofol
100 90 80 70 60 50 40 30 20 10 Propofol
0 Used in majority of Used for short term patients sedation (up to 48hours)
Midazolam Used in hepatic or renal failure
Used when frequent Used where sedation neurological is challenging assessment required
Fig. 1. Differing described use of midazolam and propofol in selected indications (separate).
those requiring frequent neurologic assessment. Remifentanil is chosen most frequently (n = 22 reported) in such circumstances. When sedation is more challenging or in patients with a history of alcohol or drug abuse, then choice of sedative is more variable (Table 2). In such cases, MDZ is used more commonly than propofol (73% vs 68%). Clonidine is considered in 44%. Haloperidol, ketamine, lorazepam, or diazepam use is also reported. During status epilepticus, sedative choice is split between propofol (54%) and MDZ (48%), then lorazepam (38%) and thiopentone (28%). Sedatives administered orally comprise haloperidol (54%), diazepam (42%), lorazepam (42%), and clonidine (41%). Clonidine is the most frequently used adjuvant (68%), followed by haloperidol (40%), MDZ (27%), and lorazepam (24%). During longterm ventilatory weaning, propofol (44%) is most commonly used, although this is closely followed by clonidine (40%). MDZ is still being prescribed in 28% of ICUs. Many commented that later in their stay, most patients have tracheostomies in place and require little, if any, sedation. In alcohol withdrawal or a history of alcohol abuse, benzodiazepines feature highly, although propofol or clonidine are also used. Some pharmacists also commented that chlordiazepoxide is prescribed (n = 25). The response rate for questions concerning head injury patients was low. At least 41 respondents do not see this patient group. Fifty-nine pharmacists (56%) that responded describe propofol prescription for patients with raised intracranial pressure. Midazolam (20%) or thiopentone (20%) is less commonly prescribed. Ketamine is more likely to be used for bronchospasm than any other indication and is considered in 79 units (66%), although it remains a rarely prescribed agent with 20% of respondents not seeing it used for at least 5 years and 57% reporting its use less than 5 times a year. Of the drugs listed, the least likely prescribed is thiopentone. Fifty-six percent of ICUs have not used it for 5 years, and 65% of ICUs use it less than 5 times a year. 4. Analgesia Morphine is the most common first-line prescription described in 68 ICUs (49%) (Table 3). Morphine is closely followed by the synthetic opioids, fentanyl (34%) and alfentanil (34%). Paracetamol is also a commonly prescribed analgesic (32%). In patients requiring frequent neurologic assessment or where drug accumulation is of concern, morphine remains a consideration in 15 (12%) and 16 (12%) of units, respectively. Remifentanil is the most popular choice for patients with uncertain neurology (50%), and both alfentanil (53%) and remifentanil (44%) are commonly prescribed when drug accumulation is of concern. The opioid most commonly prescribed for withdrawal or dependence is methadone (60%). Gabapentin (87%) is the most common agent for neuropathic pain followed by amitriptyline
436.e4
S.M. Yassin et al. / Journal of Critical Care 30 (2014) 436.e1–436.e6
Table 2 Sedative choice in described indication Sedative name
Propofol n (%) MDZ
Lorazepam Clonidine Diazepam Thiopentone Ketamine Haloperidol Other
136 (89) 141 (93) 115 (80) 100 (75) 97 (68) 76 (54) 2 (2) 7 (5) 60 (44) 58 (43)
2 (1) 2 (1) 4 (3) 0 13 (9) 53 (38) 56 (42) 35 (24) 11 (8) 27 (20)
11 (7) 11 (7) 12 (8) 2 (2) 63 (44) 1 (1) 54 (41) 98 (68) 54 (40) 57 (42)
0 2 (1) 1 (1) 0 6 (4) 16 (11) 56 (42) 21 (15) 18 (13) 32 (24)
0 0 0 2 (2) 3 (2) 39 (28) 0 0 0 0
1 (1) 2 (1) 1 (1) 1 (1) 13 (9) 0 0 11 (8) 0 0
20 (20) 0 39 (33) 0 1 (1) 10 (13) 5 (4) 33 (27)
0 3 (3) 3 (4) 9 (7)
0 0 18 (23) 36 (29)
20 (20) 1 (1) 43 (56) 80 (65)
1 (1) 79 (66) 15 (20) 70 (57)
Indication Used in majority of patients Used for short-term sedation (up to 48 h) Used in hepatic or renal failure Used when frequent neurologic assessment required Used where sedation is challenging Used for status epilepticus Used for oral or enteral sedation Used as an adjuvant agent Used for weaning from long-term ventilation Used for alcohol withdrawal or known history of alcohol abuse Used for intracranial pressure control in head injury patients Used in bronchospasm Not used in clinical practice for last 5 y Used less frequently than 5 times a year
57 (56) 64 (53) 1 (1) 0
75 (49) 18 (12) 37 (26) 11 (8) 105 (73) 68 (48) 7 (5) 39 (27) 38 (28) 60 (44)
4 (3) 4 (3) n/a 0 22 (15) n/a 72 (54) 58 (40) 20 (15) 12 (9)
15 (10) 20 (13) 29 (20) 46 (33) 30 (21) 12 (9) 22 (17) 33 (23) 25 (19) 37 (27)
0 1 (1) 6 (8) 9 (7)
42 (43) 22 (18) 12 (16) 4 (3)
More than 1 answer allowed per question.
(51%). Pregabalin prescription was described by 9 respondents. Paracetamol is commonly prescribed as an adjuvant agent (96%), although nonsteroidal anti-inflammatory agents (35%) are also used. Paracetamol is commonly prescribed as an antipyretic (84%). In at least 25 ICUs, however, it is not routine practice to reduce temperature. 4. Discussion The investigators have achieved their aim to describe a detailed perspective of UK sedation practice. This detailed perspective confirms that sedation practice remains highly varied and that strategies proven to improve it are not uniformly adopted. More than 55% of respondents use a sedation guideline. This is lower than the 80% reported in the study by Reschreiter et al [5]. Guidelines are a highly effective tool in ensuring consistency of clinical practice. Indeed, Mehta et al [18] in 2012 described that an effective protocolized sedation policy may negate the need for sedation holds when measured in terms of days on mechanical ventilation. We found that sedation scores are commonly used in the UK (87%), confirming the result of Reschreiter et al (89%) [5]. The systems used have changed, with a higher uptake of the RASS reported by our survey at 27%. Reschreiter et al reported just 5.4% of ICUs using RASS [5]. The RASS has gained favor in recent years with evidence for its reliability and validity [19]. It is preferred over the Ramsay sedation scale, as it has more discrete criteria to assess different levels of sedation, hence a more sensitive marker for monitoring effects of sedatives over time
[20]. There remains, however, a large variation in scale used, confirming the lack of a singular score in the UK. This is a finding common to other national surveys where uptake of scoring systems has been slower and variation still exists. For example, Patel et al [12] in America found a split of 37% using the Ramsay scale with 26% using RASS. Only 44% of UK ICU set a daily sedation score target on the medical ward round. One of the core strategies to improve sedation locally in our trust is to set a target RASS daily on the ward round. Our survey confirmed that sedation holds are now a routine part of prescribing practice in the UK with 80% of respondents reporting use of formal sedation holds. Kress et al [11] described, in 2000, a reduction in both duration of mechanical ventilation and length of stay using this technique, but as mentioned previously, use of effective protocols may well reduce the need [18]. The ideal strategy is probably a combination of a unit guideline, a daily target RASS, and sedation hold. The original sedation hold work took place in an ICU that prescribed morphine and MDZ [11]. It may be that sedation holds are of greatest benefit in the presence of longer active sedative and opioid. In this survey, we found no association between MDZ use, targeting of sedation scores, and presence of sedation holds. Delirium is a frequent occurrence in the ICU and can affect up to two thirds of ventilated patients [21]. Affected patients have 3 times the mortality risk and 9 times the risk of long-term cognitive impairment [22]. Screening tools are advocated to identify such patients especially those who are not overtly agitated [21]. Approximately half of respondents who answered this question indicated that their ICU assessed for delirium. Only 27 describe use of a scoring system. This low uptake
Table 3 Analgesic choice in described indication Analgesic name
Alfentanil
Fentanyl
Gabapentin
Methadone
Morphine
NSAIDs
48 (34) 50 (38) 42 (33)
48 (34) 41 (31) 21 (17)
1 (1) 0 0
0 4 (3) 0
68 (49) 59 (45) 15 (12)
0 0 1 (1)
72 (53)
42 (31)
0
0
16 (12)
44 (35)
35 (28)
0
0
43 (31) 25 (20) 1 (1) 3 (2) 0 20 (26) 27 (24)
51 (37) 15 (12) 10 (7) 13 (10) 0 13 (17) 22 (20)
3 (2) 2 (2) 118 (87) 33 (25) 0 3 (4) 11 (10)
0 76 (60) 2 (1) 2 (2) 0 30 (37) 51 (46)
Paracetamol
Remifentanil
Amitriptyline
Other
45 (32) 21 (16) 26 (20)
13 (9) 26 (20) 63 (50)
– – –
3 (2) 7 (5) 15 (12)
0
27 (20)
59 (44)
–
8 (6)
45 (36)
3 (2)
25 (20)
24 (19)
–
4 (3)
98 (72) 60 (48) 4 (3) 15 (11) 0 1 (1) 4 (4)
11 (8) 5 (4) 2 (1) 46 (35) 6 (4) 13 (17) 32 (29)
62 (45) 18 (14) 24 (18) 128 (96) 113 (84) 0 0
22 (16) 11 (9) – 0 0 18 (23) 26 (23)
– – 69 (51) – – – –
9 (7) 22 (17) 24 (18) 9 (7) 4 (3) 8 (10) –
Indication Used as first-line agent Used where respiratory depression is essential Used in patients who require frequent neurologic assessment Used in hepatic or renal failure where drug accumulation is a concern Used post cardiac arrest in patients who are therapeutically cooled Used for analgesia-based sedation Used for opiate withdrawal/dependence Used routinely in patients for neuropathic pain Used as an adjuvant agent Used as an antipyretic Not in clinical practice for last 5 y Less than 5 times a year in clinical practice
More than 1 answer allowed per question. NSAID indicates nonsteroidal anti-inflammatory drug.
S.M. Yassin et al. / Journal of Critical Care 30 (2015) 436.e1–436.e6
of sensitive diagnostic tools has been reported in previous studies in the UK, and it is highly likely that delirium is underdiagnosed [23]. Surveys in other countries show similar findings with use of delirium scoring systems ranging from 3.7% in Canada in 2002 [24], 13% in Brazil in 2008 [25], and up to 33% in America in 2007 [12]. Of note, some of the sedation surveys reviewed did not include questions about delirium [5,13,26]. Perhaps the newly published national standards for ICU, which include recommendations on managing delirium, will ensure a wider uptake of delirium scoring tools [27]. Propofol is now the most commonly used sedative agent. It is short acting and without active metabolites [28]. Midazolam remains a frequently used sedative for invasively ventilated patients. The complex pharmacokinetics and risk of oversedation mean that its use should be targeted at younger patients without cardiac failure or those with single-organ respiratory failure [7]. It is surprising that MDZ is still widely used in patients, even with renal or hepatic failure. Midazolam is used less commonly for short-term sedation now as was also found in the Reschreiter survey [5]. This prescribing trend was also found in Germany [13]. Lorazepam has more favorable kinetics in critical illness; however, it has been associated with delirium, and high doses can cause metabolic acidosis [9,10]. This delirium was reported after much larger intravenous doses, which are commonly used in the United States. There is less prevalence of delirium in doses less than 20 mg/d [9]. Lorazepam should be reserved for use as an adjuvant in small enteral or parenteral doses in critical illness. Clonidine is frequently used in long-term sedation for challenging patients or when weaning. Supporting evidence for its use is lacking, and effective dosing tends to be limited by adverse hemodynamic effects. The recent licensing in the UK of dexmedetomidine has reintroduced the concept of awake sedation and the use of α2 agonists for sedation in the critically ill. This follows publication of the MIDEX and PRODEX studies demonstrating noninferiority with MDZ and propofol [29]. Comparative sedative cost dictates that dexmedetomidine should be reserved for use in those patients where there is most likely to be benefit [30]. Dexmedetomidine was not included in this sedation survey, although there are plans to include it in the next version. Opioid choice is fairly evenly spread between morphine, fentanyl, and alfentanil for most indications. When accumulation is a concern, then shorter acting agents are more popular such as with frequent neurologic examinations or organ failure, as described in Fig. 2. The evidence for shorter acting opioids over morphine is not conclusive; thus, our results are not unexpected, although accumulation of morphine's glucuronide metabolites is well described in renal failure [6]. Given that as much as 50% of ICU patients can develop some form of renal failure the investigators purport that widespread morphine use is perhaps surprising despite the availability and low relative cost of both fentanyl and alfentanil. Up to 30 units in this survey described use of analgesic-based sedation. The advantage of this is obvious, as it aims to keep patients calm rather than asleep and allows the patient to communicate their needs. It can also avoid the deleterious effects of benzodiazepines. Newer agents such as dexmedetomidine further facilitate awake sedation in a proportion of patients [30]. Use of paracetamol as an antipyretic appears to be common despite lack of solid evidence of benefit in sepsis. Almost half of those who responded describe only using paracetamol in temperatures greater than 39.5°C, which concurs with national recommendations. In accordance with National Institute of Clinical Excellence, gabapentin is the preferred agent for neuropathic pain, although 9 ICUs favor pregabalin, with its more linear pharmacokinetics [31]. These are probably preferred agents in the ICU as they have evidence of rapid effectiveness: essential in the critically ill patient with neuropathic pain. In summary, we have demonstrated that an Internet-based survey directed at clinical pharmacists is both a feasible and effective method for data collection on prescribing in the ICU. Given that 85% were senior
436.e5 Alfentanil Fentanyl Morphine
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0% Used as first line agent
Used in patients who require frequent neurological assessment
Used in hepatic or renal failure where drug accumulation is a concern
Fig. 2. Opioid choice in selected indications.
(band 8a or above) and spent a mean of half their time in the ICU, one would expect a sound knowledge of sedation practice. We achieved a response rate of 60% that provided a detailed insight into sedation and analgesic practice. Workforce demographics were recorded, but this was not the primary focus of this survey. There is potential to use these national pharmacy groups to gain a wide perspective of prescribing practice within the ICU and wider critical care environment. Previous UK surveys have used a paper format to conduct their survey [5,15]. Other countries have used electronic surveys successfully, although this can result in convenience samples and their associated bias [25]. This survey targeted pharmacists and had a better response rate than previous surveys that targeted nurses and physicians [26]. This could be attributed to the diligent follow-up of nonresponders rather than the profession itself. There are limitations of the study. The investigators chose the ICU pharmacist as their point of contact, but input varies. Some units do not have a dedicated pharmacist, and therefore, their knowledge of sedation practice may vary. This may be reflected in many responses. In addition, there was also a variation in description of unit specialty, which limited analysis of the data. In future surveys, this will be standardized. The units that did indicate specialism comprised 9 cardiothoracic and 4 neurosurgical ICUs, and it was felt that the number was too small for subgroup analysis. All the responses were, therefore, analyzed together. Although there were sections on delirium scoring, the investigators did not include questions on antipsychotic agents. In addition, there was no question on the use of pain scores. The investigators intend to repeat the survey over 2015 and include a section on antipsychotics and a question on behavioral pain scores. Despite the investigator's very best efforts, an overall response rate of just 60% was attained, leaving the sedation practice for 40% of units unaccounted for. Lastly, selfreported bias may mean that some of the responses did not represent a fully accurate reflection of their unit practice.
5. Conclusions The reported sedation survey has succeeded in its overall aims and has provided a detailed perspective of UK sedation practice. This included sedation guidance and holds, prevalence of delirium assessment, sedative choice, and analgesic choice in a number of clinical scenarios. It has also provided national trends in sedation prescribing practice, and finally, with a response rate of 70% in the presence of a clinical pharmacy service, it has confirmed that the pharmacy workforce can be a source of prescribing practice description in the ICU.
436.e6
S.M. Yassin et al. / Journal of Critical Care 30 (2014) 436.e1–436.e6
Competing interests Dr McKenzie has received honoraria from Orion Pharma and GlaxoSmithKline Kline for educational events and advisory boards.
[13]
[14]
Authors' contributions [15]
Dr McKenzie and Dr Terblanche devised the original survey. Mrs Yassin collected the results and undertook all initial analysis. Dr Yassin reviewed and made extensive comment in preparation of the revised manuscript. All authors contributed to the manuscript preparation. Acknowledgments
[16]
[17]
[18]
The investigators would like to thank the UKCPA CCG for completing the survey and Ms Joyce Lim for assisting in the analysis.
[19]
References
[20]
[1] Jacobi J, Fraser GL, Coursing DB, Riker RR, Fontaine D, Wittered ET, et al. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119–41. [2] Park GR, Ward B. Sedation and analgesia in the critically ill. In: Darrell D, Cox T, Firth J, Benz, editors. Oxford Textbook of Medicine. 4th ed. Oxford University Press; 2003. p. 1250–6. [3] Strøm T, Martinussen T, Toft P. A protocol of no sedation for critically ill patients receiving mechanical ventilation: a randomised trial. Lancet 2010;375(9713):475–80. http://dx.doi.org/10.1016/S0140-6736(09)62072-9 [Epub. 2010 Jan 29]. [4] Riker RR, Fugate JE, the Participants in the International Multi-disciplinary Consensus Conference on Multimodality Monitoring. Clinical monitoring scales in acute brain injury: assessment of coma, pain, agitation, and delirium. Neurocrit Care 2014 [Pub ahead of print]. [5] Reschreiter H, Maiden M, Kapila A. Sedation practice in the intensive care unit: a UK national survey. Crit Care 2008;12:R152. [6] Gordon HL. Morphine intoxication in renal failure: the role of morphine-6glucuronide. Br Med J (Clin Res Ed) 1986;293(6550):818–9. [7] McKenzie CA, McKinnon W, Naughton DP, Treacher D, Davies G, Phillips GJ, et al. Differentiating midazolam over-sedation from neurological damage in the intensive care unit. Crit Care 2005;9:R32–6. [8] Bauer TM, Ritz R, Haberthür C, Ha HR, Hunkeler W, Sleight AJ, et al. Prolonged sedation due to accumulation of conjugated metabolites of midazolam. Lancet 1995;346 (8968):145–7. [9] Pandharipande P, Shintani A, Peterson J, Pun BT, Wilkinson GR, Dittus RS, et al. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anaesthesiology 2006;104(1):21–6. [10] Wilson KC, Reardon C, Theodore AC, Farber HW. Propylene glycol toxicity: a severe iatrogenic illness in ICU patients receiving IV benzodiazepines: a case series and prospective, observational pilot study. Chest 2005;128(3):1674–81. [11] Kress JP, Pohlman AS, O'Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 2000;342: 1471–7. [12] Patel R, Gambrell M, Speroff T, Scott T, Pun B, Okahashi J, et al. Delirium and sedation in the intensive care unit: survey of behaviours and attitudes of 1384 healthcare pro-
[21]
[22]
[23] [24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
fessionals. Crit Care Med 2009;37:825–32. http://dx.doi.org/10.1097/CCM. 0b013e31819b8608. Martin J, Franck M, Sigel S, Weiss M, Spies C. Changes in sedation management in German intensive care units between 2002 and 2006: a national follow-up survey. Crit Care 2007;11:R124. http://dx.doi.org/10.1186/cc6189. Tanios M, de Wit M, Epstein SK, Devlin JW. Perceived barriers to the use of sedation protocols and daily sedation interruption: a multidisciplinary survey. J Crit Care 2009;24:66–73. http://dx.doi.org/10.1016/j.jcrc.2008.03.037. Ramaswamy S, Rangappa D, Wheatley E. National survey on current practice of sedation and analgesia in intensive care in United Kingdom. Intensive Care Med 2009;35(S112) [0342-4642 (abstract only)]. Leape LL, Cullen DJ, Clapp MD, Burdick E, Demonaco H, Erickson JI, et al. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. JAMA 1999;282(3):267–70. Marshall JJ, Finn CA, Theodore AC. Impact of a clinical pharmacist-enforced intensive care unit sedation protocol on duration of mechanical ventilation and hospital stay. Crit Care Med 2008;36(2):427–33. Mehta S, Burry L, Cook D, Fergusson D, Steinberg M, Granton J, et al. Daily sedation interruption in mechanically ventilated critically ill patients cared for with a sedation protocol. JAMA 2012:1985–92. http://dx.doi.org/10.1001/jama.2012.13872. Sessler CN, Gosnell M, Grap MJ, Brophy GT, O'Neal PV, Keane KA, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care patients. Am J Respir Crit Care Med 2002;166:1338–44. Ely EW, Truman B, Shintani A, Thomason JWW, Wheeler AP, Gordon S, et al. Monitoring sedation status over time in ICU patients: the reliability and validity of the Richmond Agitation Sedation Scale (RASS). JAMA 2003;289:2983–91. Ely EW, Inouye SK, Bernard GR, Gordon S, Francis J, May L, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA 2001;286(21):2703–10. Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrell Jr FE, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA 2004;14:1753–62. Mac Sweeney R, Barber V, Page V, Ely EW, Perkins GD, Young JD, et al. A national survey of the management of delirium in UK intensive care units. QJM 2010;103(4):243–51. Mehta S, Burry L, Fischer S, Martinez-Motta C, Hallett D, Bowman D, et al. Canadian survey of the use of sedatives, analgesics, and neuromuscular blocking agents in critically ill patients. Crit Care Med 2006:34374–80. http://dx.doi.org/10.1097/01.CCM. 0000196830.61965.F1. Salluh JIF, Dal-Pizzol F, Mello P, Friedman G, Silva EMD, Teles JMM, et al. Delirium recognition and sedation practices in critically ill patients: a survey on the attitudes of 1015 Brazilian critical care physicians. J Crit Care 2009;24:556–62. O'Connor M, Bucknall T, Manias E. Sedation management in Australian and New Zealand intensive care units: doctors' and nurses' practices and opinions. Am J Crit Care 2010;19:285–95. Intensive Care Society/Faculty of Intensive Care Medicine. Core Standards for the Intensive Care Unit 2013, Edition 1. Available from http://www.ics.ac.uk/icshomepage/guidelines-standards. Astra Zeneca UK ltd. Sedation and analgesia in the critically ill. Summary of product characteristics for Diprivan 1%. Last updated on the eMC; 2012. Jakob SM, Ruokonen E, Grounds RM, Sarapohja T, Garratt C, Pocock SJ, et al. Dexmedetomidine for Long-Term Sedation Investigators: dexmedetomidine vs midazolam or propofol for sedation during prolonged mechanical ventilation: two randomized controlled trials (PRODEX and MIDEX). JAMA 2012;21:1151–60. Pasin L, Greco T, Feltracco, Vittorio A, Neto CN, Cabrini L, Landoni G, Finco G, Zangrillo A. Dexmedetomidine as a Sedative Agent in Critically Ill Patients: A Meta-Analysis of Randomized Controlled Trials. PLoS One 2014 Accessed from http://www.ncbi.nlm.nih.gov/pubmed/24391726. Pfizer UK ltd. Summary of product characteristics for Lyrica. Last updated on the eMC; 2013.
