Early Goal-Directed Sedation Versus Standard Sedation in Mechanically Ventilated Critically Ill Patients: A Pilot Study*

Transcription

1 Early Goal-Directed Sedation Versus Standard Sedation in Mechanically Ventilated Critically Ill Patients: A Pilot Study* Yahya Shehabi, FCICM, FANZCA, EMBA 1,2,3 ; Rinaldo Bellomo, MD, FCICM, FRACP 2,3 ; Michael C. Reade, MBBS, MPH, DPhil, FCICM 4 ; Michael Bailey, PhD 3 ; Frances Bass, RN, BN, GDipICU 5 ; Belinda Howe, RN, BN 3 ; Colin McArthur, FANZCA, FCICM 3,6 ; Lynne Murray, FAIMS 3 ; Ian M. Seppelt, MBBS, FANZCA, FCICM 7 ; Steve Webb, MPH, PhD, FCICM 3,8 ; Leonie Weisbrodt, RN, BN, MN(Hons) 9 ; for the Sedation Practice in Intensive Care Evaluation (SPICE) Study Investigators and the Australian and New Zealand Intensive Care Society (ANZICS) Clinical Trials Group *See also p University New South Wales, Clinical School of Medicine, Director Intensive Care Research, Prince of Wales Hospital, Randwick, New South Wales, Australia. 2 University of Melbourne, Faculty of Medicine, School of Public Health and Preventive Medicine, Monash University, Director Intensive Care Research, Austin Hospital, Heidelberg, Australia. 3 Australian New Zealand Intensive Care Research Centre, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia. 4 Australian Defence Force and Burns, Trauma and Critical Care Research Centre, Royal Brisbane and Women s Hospital, University of Queensland, Queensland, Australia. 5 Department of Intensive Care Services, Prince of Wales Hospital, Randwick, Australia. 6 Department of Critical Care Medicine, Auckland City Hospital, Auckland, New Zealand. 7 Department of Intensive Care Medicine, University of Sydney, Sydney Medical School Nepean, New South Wales, Australia. 8 Royal Perth Hospital, School of Medicine and Pharmacology and School of Population Health, University of Western Australia, Perth, Australia. 9 Department of Intensive Care, Sydney Nursing School, University of Sydney, Nepean Hospital, New South Wales, Australia. Drs. Shehabi, Bellomo, Reade, Bailey, and Ms. Howe had full access to the data and participated in the analysis and interpretation of the data. Drs. Shehabi, Bellomo, Reade, Webb, Seppelt, McArthur, Bass, and Weisbrodt participated in the study concept and design. Drs. Bailey, Howe, Bass, Weisbrodt, Murray participated in the acquisition of data and data management. Drs. Bailey, Shehabi, Bellomo, Reade, Seppelt, Webb, Murray, and McArthur participated in analysis and interpretation of data. Drs. Shehabi, Bellomo, Seppelt, Reade, Webb and McArthur participated in drafting of the manuscript. Drs. Shehabi, Bellomo, Reade, Bailey, Seppelt, Webb, McArthur, Bass, Murray and Weisbrodt participated in critical review and intellectual input. This study was conducted in collaboration with the Australian New Zealand Intensive Care Research Centre (ANZIC RC) and the department of Copyright 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: /CCM.0b013e31828a437d epidemiology and preventive medicine of the faculty of medicine at Monash University, Melbourne, Australia. Hospira and its employees had no input into the design, protocol, study conduct, data collection, data analysis, manuscript preparation, review and submission. However, Hospira, Melbourne, Australia, provided the study drug dexmedetomidine at no cost to study sites. Supported, in part, by an unrestricted Grant-In-Aid from Hospira, Lake Forest, IL. Dr. Shehabi has received unrestricted Grant-In-Aid research grants from Hospira Inc. (Lake Forest, IL); research grants from Roche Diagnostics and Thermofisher Scientific; competitive research funding grants from National Health and Medical Research Council, Australia. Dr. Shehabi's research department has received payment for article preparation for being part of SEDCOM delirium manuscript review 2009; and speakers honoraria and consulting fee from Hospira and Roche Diagnostics. He was on an advisory boards for Hospira and GSK and has received payment from GSK for the development of educational material approved by College of Intensive Care Medicine of ANZ. Dr. Reade has received a consulting fee and research grants from Hospira. Dr. McArthur has received grant support, travel reimbursements, and provisions for writing assistance from Hospira Australia. Dr. Seppelt was on an advisory board in Intensive Care supported by Hospira; has received competitive research funding grants from National Health and Medical Research Council, Australia; and has received payment for the development of educational presentations from Asklepios Medical Education. Dr. Webb has consulted for Aalix Healthcare Services Consulting, Ibis Biosciences, Astra Zeneca, Jansen Cilag, and has received grant support from Fresenius Kabi. The remaining authors have not disclosed any potential conflicts of interest. Participating centers and principal investigators are listed in the Appendix 1. For information regarding this article, y.shehabi@unsw.edu.au Objective: To assess the feasibility and safety of delivering early goal-directed sedation compared with standard sedation. Design: Pilot prospective, multicenter, randomized, controlled trial. Setting: Six ICUs. Patients: Critically ill adults mechanically ventilated for greater than 24 hours. Interventions: Patients randomized to early goal-directed sedation received a dexmedetomidine-based algorithm targeted to Critical Care Medicine

2 Shehabi et al light sedation (Richmond Agitation Sedation Score of 2 to 1). Patients randomized to standard sedation received propofol and/ or midazolam-based sedation as clinically appropriate. Measurements and Main Results: The main feasibility outcomes were time to randomization and proportion of Richmond Agitation Sedation Score assessments in the first 48 hours in the light and deep sedation range. Safety outcomes were delirium-free days, vasopressor and physical restraints use, and device removal. Randomization occurred within a median (interquartile range) of 1.1 hours ( ) after intubation or ICU admission for out of ICU intubation. Patients in the early goal-directed sedation (n = 21) mean (sd) Acute Physiology and Chronic Health Evaluation II score was 20.2 (6.2) versus 18.6 (8.8; p = 0.53) in the standard sedation (n = 16). A significantly higher proportion of patients was lightly sedated on days 1, 2, and 3 (12/19 [63.2%], 19/21 [90.5%], and 18/20 [90%] vs 2/14 [14.3%], 8/15 [53.3%], and 9/15 [60%]; p = 0.005, 0.011, 0.036) and more Richmond Agitation Sedation Scale assessments between ( 2 and 1), in the first 48 hours (203/307 [66%] versus (74/197 [38%]; p = 0.01) in the early goal-directed sedation versus standard sedation, respectively. Early goal-directed sedation patients received midazolam on 6 of 173 (3.5%) versus 4 of 114 (3.5%) standard sedation patient-days when dexmedetomidine was given. Propofol was given to 16 of 21 (76%) of early goal-directed sedation versus 16 of 16 (100%) of standard sedation patients (p = 0.04). Early goal-directed sedation patients had 101 of 175 (58%) versus 54 of 114 (47%; p = 0.27) delirium-free days and required significantly less physical restraints 1 (5%) versus 5 (31%; p = 0.03) than standard sedation patients. There were no differences in vasopressor use and self-extubation. Conclusions: Delivery of early goal-directed sedation was feasible, appeared safe, achieved early light sedation, minimized benzodiazepines and propofol, and decreased the need for physical restraints. The findings of this pilot study justify further investigation of early goal-directed sedation. (Crit Care Med 2013; 41: ) Key Words: critical illness; delirium; dexmedetomidine; goal directed; intensive care; mechanical ventilation; sedation (11 13); variable use of monitoring the depth of sedation, pain, and delirium (14 19); late randomization after 72 to 96 hours of mechanical ventilation and, therefore, delayed intervention (3, 4) and administration of the intervention by research staff rather than clinical staff (20). Furthermore, randomized trials of protocolized algorithms did not control for the choice of sedative agents (21, 22) and did not assess the contribution of different agents to overall sedation quality and outcome. Finally, many studies addressing sedation strategies were highly dependent on the practice model of ICU therapy, including nurse-topatient ratios and weaning protocols (22 24). A recent observational study of sedation practice in Australia and New Zealand (Sedation Practice in Intensive Care Evaluation) found that clinicians used midazolam and propofol as the primary sedative with comparable frequency (9). It also identified a high prevalence of deep sedation in the first 48 hours after initiation of mechanical ventilation which was found to independently predict delayed time to extubation and increased long-term mortality (9). This observation suggests that future interventions should be delivered early with a strategy to facilitate light sedation. In contrast to traditional sedatives, dexmedetomidine is an α 2 -adrenoreceptor agonist that has been increasingly used for ICU sedation (25, 26). Phase II trials have shown that dexmedetomidine might facilitate the delivery of a sedation algorithm to provide arousable sedation, minimize the use of benzodiazepines and propofol sedatives, and modify emergent delirium and/or shorten ventilation time (3, 4, 13, 27). In response to such emerging evidence, we designed a pilot study to compare the effect of a process of care sedative algorithm (termed early goal-directed sedation) that is implemented early after initiation of mechanical ventilation, is goal-directed to target light sedation whenever possible, and uses dexmedetomidine as the primary sedative agent and minimizes the use of benzodiazepines. The aim of this study was to assess whether early goal-directed sedation (EGDS) is feasible, safe, can be delivered in a timely fashion, and can achieve early light sedation more effectively than standard sedation. The use of IV sedatives in mechanically ventilated critically ill patients is ubiquitous (1). The revised Society for Critical Care Medicine Clinical Practice Guidelines (2013) for management of Pain, Agitation, and Delirium (2) made strong recommendations for routine monitoring of pain, sedation, and delirium and also recommended light sedation. However, it did not make any endorsement of any agent or combinations of agents owing to the lack of high level evidence. Very few randomized clinical trials of sedation practice have been either adequately powered (3, 4) or used a patient-centered outcome, such as mortality, as the primary outcome (5). Thus, evidence for practice change has been limited (6 9). In addition, many randomized trials have shared several potential limitations: poor alignment of control group with current practice (10); focus on comparison of two study drugs, despite the fact that most patients are sedated with a combination of drugs MATERIALS AND METHODS Design and Patients This was a pilot, prospective, feasibility and safety, randomized, unblinded, parallel group, controlled trial conducted in six tertiary and regional ICUs in Australia and New Zealand. The institutional human research ethics committees at participating sites gave permission for deferred consent (intervention needed immediately and is an accepted standard of care) or approved the study with prospective consent from a patient surrogate (intervention considered a clinical trial of experimental agent). The study was registered (ACTRN ) and conducted between July 2011 and December Patients were included if they had been intubated within the previous 12 hours, were expected to need mechanical ventilation for longer than 24 hours, and required immediate and ongoing sedation. Exclusion criteria were age less than August 2013 Volume 41 Number 8

3 Clinical Investigations years; pregnancy proven or suspected primary neurological injury; a diagnosis likely to result in prolonged weakness, drug overdose, burn injury, acute liver failure, dementia, or psychiatric illness; need for ongoing neuromuscular blockade, palliative care, or treatment limitations; inability to communicate in English; a mean blood pressure less than 55 mm Hg; a heart rate less than 55/min; or a high-grade AV block in the absence of a functioning pacemaker. Study Outcome Measures The main feasibility outcomes were time to randomization and time spent in the light sedation range ( 2 to 1) of the Richmond Agitation Sedation Scale (RASS) in the 48 hours after randomization (16). Other outcome measures included the proportion of patients treated with dexmedetomidine, propofol, and midazolam; the number of days given and the cumulative dose of sedative, analgesic, and antipsychotic agents received; the proportion of RASS assessments in the deep sedation ( 3 to 5) and agitation (> 1) RASS range; the proportion of patients with delirium (identified by the positive Confusion Assessment Method for the ICU [CAM-ICU]) (18); the number of days spent alive and free of delirium; the number of patients extubated and alive within 7 days of randomization; the number of ventilator-free days at 28 days; and mortality at discharge from hospital and 90 days after randomization. Other feasibility outcomes included the average recruitment rate. Safety outcomes included device removal and selfextubation, the use of physical restraints, major serious adverse events, vasopressor therapy, and hemodynamic instability. Study Process Block randomization was undertaken with concealed envelopes. Data collected were centrally entered into a database using optical recognition software. The study was monitored and managed by the Australian New Zealand Intensive Care Research Centre, Monash University, Melbourne, Australia. Dexmedetomidine (200 μg/2 ml; Hospira Pty, Melbourne, Australia) was diluted in 0.9% saline so that every ml/hr infused = 0.1 μg/kg/hr and given through a dedicated intravenous catheter. All subjects had assessment of RASS score every 4 hours recorded, along with pain assessment (by the bedside nurse) and a CAM-ICU assessment daily by research coordinators, only when the RASS score was greater than 3. The default sedation target for all enrolled patients in both study arms was light sedation as defined by a RASS range of 2 to 1 unless otherwise clinically indicated. Sedative infusions were titrated to the lowest dose needed to achieve target sedation by bedside nurses. In all patients, analgesia was provided by infusion or boluses of an opioid (fentanyl or morphine) or other agents such as ketamine, as determined by the treating clinician. The EGDS Algorithm EGDS (Fig. 1) had to start within 12 hours of endotracheal intubation or within 12 hours of ICU admission for patients intubated outside the ICU (early intervention). The primary sedative agent in this arm was dexmedetomidine infusion at a starting dose of 1 μg/kg/hr without a loading dose. Bolus administration of dexmedetomidine was strictly prohibited owing to the risk of severe bradycardia and sinus arrest. If required, sedation could be supplemented with propofol. Sedatives were administered to achieve the desired level of light sedation whenever possible. Dexmedetomidine infusion was administered between a minimum of zero and maximum of 1.5 μg/kg/hr specified by the treating clinician and titrated by the bedside nurses to achieve the desired level of sedation. Supplemental propofol could be used, at the lowest effective dose, to: a) provide sedation during commencement and initial titration of dexmedetomidine infusion; b) optimize sedation and achieve the level of sedation specified by the treating clinician at any time when dexmedetomidine alone and at maximum dose was deemed insufficient to provide patient comfort and safety; c) provide rescue sedation for immediate control of sudden agitation at any time. Clonidine and remifentanil could not be administered to any patient and benzodiazepines (such as midazolam, diazepam, and clonazepam) could not be administered to any patient in this arm, unless deemed essential by the treating clinician, for example, for the management of convulsions, palliation, procedural anesthesia, or refractory agitation. The Standard Sedation Algorithm The primary sedative agent in this group was at the discretion of the treating clinician. This could be midazolam and/ or propofol or other agents deemed necessary but not dexmedetomidine. Clonidine and remifentanil could not be administered. Selected agents could be given by infusion or boluses and titrated by bedside nurses, including cessation when necessary, to achieve the default light sedation or the level of sedation deemed clinically appropriate and specified by the treating clinician. Breakthrough Agitation In either treatment arm, for patients who developed breakthrough agitation, the default treatment was to optimize sedative prescription according to randomization status. Second-line therapy included intravenous haloperidol boluses 2.5 to 5 mg as clinically required or a nonbenzodiazepine antipsychotic agent, as chosen by the treating physician, such as quetiapine 25 to 100 mg given enterally (28). If significant agitation continued despite these measures, then dexmedetomidine could be used in the standard sedation and midazolam could be used in the EGDS arm. Discontinuation of Therapy Primary sedative infusion in both arms of the study continued until sedation/analgesia was no longer required or up to 28 days of therapy. Sedative/analgesic infusion could continue after extubation if clinically required. If sedation was deemed necessary beyond 28 days after enrollment, the choice of sedative regimen was determined solely by the treating clinician. Critical Care Medicine

4 Shehabi et al test where numbers were small, while continuous variables have been compared using Wilcoxon rank-sum tests. To account for within-patient correlation for repeated assessments, comparisons have been made with generalized linear modeling for repeated measures using a binomial distribution. Analysis was performed using SAS version 9.2 (SAS Institute, Cary, NC) and a two-sided p value of 0.05 was considered to be statistically significant. Figure 1. Early goal-directed sedation algorithm. Once randomized, adequate analgesia is provided after which dexmedetomidine sedative infusion is commenced to achieve a Richmond Agitation Sedation Scale (RASS) target of 2 to 1. Low-dose propofol is added if needed. Statistical Analysis This was a feasibility trial of two sedation protocols. In addition to time to randomization, the main feasibility outcome was the difference in light sedation achieved in the first 48 hours in each arm. The main efficacy outcome included the difference in the number of patients receiving the primary sedative agents, number of days, and cumulative dose of each agent given. A sample size of greater than or equal to 1,000 hours of protocolized sedation in each group was estimated to allow meaningful assessment of feasibility, efficacy, and safety. We reported values as means with standard deviation for normally distributed variables and medians with interquartile ranges (IQRs) for nonnormally distributed continuous variables and proportions for categorical variables. Comparisons between groups have been made using a chi-square test for equal proportion or Fisher exact RESULTS Study Population Of the 154 patients who met all inclusion criteria, 37 patients were randomized, of whom 21 were allocated to EGDS and 16 were allocated to standard sedation (Fig. 2). Baseline patient demographics and characteristics are shown in Table 1. Most patients were admitted with a medical (nonoperative) diagnosis (75% of all patients). Of these, more than one-third had respiratory failure. Most patients received vasopressors during the study period. Time to Randomization Randomization occurred within a median (IQR) of 2.17 ( ) hours in EGDS or 1.05 ( ) hours in standard sedation (p = 0.43). The first RASS assessment was performed within a median (IQR) of 0.40 ( ) hours after randomization. Sedation Depth During the first 48 hours, 203 of 307 (66%) versus 74 of 197 (38%; p = 0.01) of all RASS assessments were in the light sedation range and 93 of 307 (30%) versus 112 of 197 (57%; p = 0.02) were in the deep sedation range in EGDS compared with standard sedation patients, respectively. The distribution of RASS scores between the EGDS and standard sedation in the first 48 hours is shown in Figure 3. Throughout the study period, the proportion of RASS assessments in the light sedation range ( 2 to 1) was significantly higher in the EGDS than in standard sedation patients 707 of 893 (77%) versus 348 of 556 (63%; p = 0.05). Conversely, the proportions RASS scores recorded in the August 2013 Volume 41 Number 8

5 Clinical Investigations There was a significant difference in the number of patients who achieved one or more RASS score in the light sedation range on days 1, 2, and 3 in the EGDS group 12 of 19 (63.2%), 19 of 21 (90.5%), and 18 of 20 (90%) versus 2 of 14 (14.3%), 8 of 15 (53.3%), and 9 of 15 (60%) (p = 0.005, 0.011, 0.036) in the standard sedation group, respectively. Two patients in each group had a short day 1 chart-day and had no RASS assessment after randomization on the calendar day of study participation. Beyond the first 72 hours, the number of patients achieving light sedation was not different between the two groups (Fig. 4). Figure 2. Patient flow diagram. Of screened patients, 198 (56.2%) were not expected to be ventilated for longer than 24 hr. Nearly half of patients who met all inclusion criteria but were excluded 55 (47%) were because of primary brain pathology. BP = blood pressure, EGDS = early goaldirected sedation, NDMR = nondepolarizing muscle relaxants. deep sedation range ( 3 to 5) 168 of 893 (19%) versus 150 of 556 (27%; p = 0.26) and in the agitated RASS range (2 4) was lower (18/893 [2%] vs 58/556 [10%]; p = 0.01) in the EGDS compared with standard sedation patients, respectively. TABLE 1. Patient Characteristics and Demographic Data Patient Characteristics Standard Sedation (n = 16) Early Goal- Directed Sedation (n = 21) Age, yr; mean (sd) 61.6 (17) 65 (15) Male, % (n) 56 (9) 52 (11) Weight, kg; mean (sd) 87 (28.1) 83.9 (24.7) Acute Physiology Evaluation and Chronic Health Evaluation II (33), mean (sd) 18.6 (8.8) 20.2 (6.2) Operative elective, % (n) 13 (2) 5 (1) Operative emergency, % (n) 13 (2) 19 (4) Medical admissions, % (n) 69 (11) 76 (16) Respiratory failure, % (n) 31 (5) 43 (9) Gastrointestinal disorder, % (n) 13 (2) 14 (3) Cardiovascular failure, % (n) 31 (5) 14 (3) Sepsis, % (n) 6 (1) 14 (3) Miscellaneous, % (n) 19 (3) 10 (2) Admission emergency department % (n) 44 (7) 24 (5) Admission hospital ward, % (n) 25 (4) 33 (7) Admission other hospital, % (n) 6 (1) 19 (4) Vasopressor (during study), % (n) 88 (14) 90 (19) Dialysis (during study), % (n) 19 (3) 10 (2) Sedative, Analgesic, and Antidelirium Agents Dexmedetomidine was given to 20 patients (95%) in the EGDS and one (6%) in the standard sedation (p < 0.001). Throughout the study period, midazolam was given for only 6 of 173 (3.5%) EGDS patient-days, whereas dexmedetomidine was given for only 4 of 114 (3.5%) standard sedation patient-days. Sedation in the targeted range was achieved with dexmedetomidine alone in eight EGDS patients (40%) in the first 24 hours postrandomization. There was a significant difference in the number (%) of patients who received propofol and midazolam between the two groups (Table 2). The duration and the cumulative dose of sedative and analgesic agents given are shown in Table 2. In the EGDS arm, three patients received midazolam for seizures (one patient), palliation (one patient), or increased sedation during neuromuscular blockade (one patient). There were differences in the administered dose of dexmedetomidine, propofol, and midazolam, with more midazolam and propofol given to patients allocated to standard sedation. Although propofol was permitted in the EGDS algorithm, the average (range) amount of propofol used in the first 48 hours postrandomization was significantly less in the EGDS arm at 354 (50 1,330) versus 1,013 (210 2,940) mg on day 1 (p = 0.011) and 618 (30 2,640) versus 2,617 (50 6,110) mg on day 2 (p = 0.008). The cumulative doses of both morphine and fentanyl were comparable in both arms (Table 2). The proportion of patients in the EGDS algorithm who received morphine (10 [48%] vs 10 [63%]) or fentanyl (11 [52%] vs 7 [44%]) was not different to standard sedation (p = 0.39 and 0.62), respectively. A small number of patients in both groups received haloperidol (1 [5%] vs 2 [13%]) and/or diazepam (2 [10%] vs 1 [6%]) in the EGDS versus standard sedation arm, respectively. Delirium, Ventilation Time, and Physical Restraints An equal proportion of patients experienced one or more positive CAM-ICU assessments during the study period (38%; p = 0.97). However, the number of patient-days with a negative CAM-ICU assessment in the EGDS group was 58% versus 47% (p = 0.27) (Table 3). Patients in both groups had a median (IQR) ventilation-free days at 28 days that was not different (p = 0.72). However, 20 EGDS patients (95%) were extubated alive by study day 7 compared with 12 (75%; p = 0.09) in the standard sedation group. Device removal or self-extubation Critical Care Medicine

6 Shehabi et al randomization, five patients were excluded owing to inability to obtain consent within 12 hours. Figure 3. Richmond Agitation Sedation Scale (RASS) scores in the first 48 hr in the early goal-directed sedation (EGDS) versus standard sedation (STDS). In the first 48 hr, a total of 307 RASS assessments were performed in the EGDS versus 197 in the STDS. Most of the RASS scores in the EGDS 203 (66%) were in the light sedation range (RASS 2 to 1), whereas most of the RASS scores in the STDS 112 (57%) were in the deep sedation range ( 3). There was a statistically significant difference between the two groups at every RASS level less than 1. DISCUSSION Key Findings We conducted a pilot feasibility study of an approach to sedation based on early intervention targeted to achieve light sedation by the primary use of dexmedetomidine (EGDS). We found that EGDS could be delivered early after initiation of mechanical ventilation (within 2 hr) and safely. In addition, this pilot confirmed the efficacy of EGDS in achieving target sedation, patients spent significantly more time lightly sedated and free of physical restraints compared with patients receiving standard sedation. was reported on two occasions in the EGDS but none in the standard sedation (p = 0.49). Five patients in the standard sedation group, however, needed physical restraints compared with one patient in the EGDS group (p = 0.03) (Table 3). There was no significant difference in the number of patients who required vasopressors on any study day. There was no serious adverse event related to study drug in either arm. Feasibility of Recruitment and Consent Process The average rate of recruitment during the study period was 0.75 patients per site per week. No deferred consent was revoked, and in centers where prospective consent was required prior to Comparisons With Previous Studies There are unique and novel aspects of this study. We addressed potential limitations associated with previous sedation trials. First, we achieved the shortest time to randomization of any trial so far. Second, in addition, the choices of sedative agents were protocolized in the EGDS to facilitate light sedation. Third, EGDS achieved a significant decline in the early deep sedation pattern identified in a previously conducted multicenter observational study (9), achieving early light sedation in 90% of patients within 24 hours. The role of dexmedetomidine as an alternative sedative in mechanically ventilated ICU patients has been investigated TABLE 2. Cumulative Dose/kg Body Weight and Duration of Treatment With Sedatives and Analgesic Agents Throughout the Study Period Drugs Given EGDS n = 21 Standard Sedation n = 16 No. Rx EGDS vs Standard Sedation p Dexmedetomidine (μg) ( ) ( ) 20 vs 1 < Time on dexmedetomidine (d) 4 (3 6) 0 (0 0) < Midazolam (mg) 0.06 (0.02 1) 0.3 ( ) 3 vs Time on midazolam (d) 0 (0 0) 0.5 (0 2) Propofol (mg) 9.89 ( ) ( ) 16 vs Time on propofol (d) 1 (1 5) 2.5 (2 4.5) 0.18 Morphine (mg) 2.42 ( ) 1.43 ( ) 10 vs Fentanyl (μg) 3.36 ( ) ( ) 11 vs EGDS = early goal-directed sedation. All drug doses are expressed in cumulative per kg median (interquartile range) dose August 2013 Volume 41 Number 8

7 Figure 4. Number and percentage of patients achieving light sedation in the early goal-directed sedation (EGDS) versus standard sedation (STDS; days 1 7). This graph shows that the percentage of patients achieving one or more Richmond Agitation Sedation Scale (RASS) in the light sedation range was significantly higher on days 1, 2, and 3 (p = 0.005, 0.011, and 0.036) but are not different thereafter. in three randomized controlled trials (RCTs): The Safety and Efficacy of Dexmedetomidine Compared with Midazolam (SEDCOM) trial (3) and the Dexmedetomidine versus Midazolam or Propofol for Sedation During Prolonged Mechanical Ventilation (MIDEX and PRODEX) trials (4). TABLE 3. Feasibility and Clinical Outcomes Clinical Investigations Although EGDS incorporated the use of dexmedetomidine, there are distinct differences from the above RCTs. The most striking difference is the time to randomization which was up to 96 hours in SEDCOM and 72 hours in MIDEX and PRODEX, allowing nonprotocol sedatives to be used unchecked in the first 3 to 4 days. In addition, target sedation depth in PRODEX and MIDEX was a RASS range of 3 to 0 compared with 2 to 1 with EGDS and in SEDCOM. SEDCOM and PRODEX allowed the use of midazolam as rescue therapy in the intervention arm. In contrast, midazolam was not allowed in the EGDS algorithm. Sedation interruption was used in nearly 90% of patients in SEDCOM, MIDEX, and PRODEX, whereas the EGDS algorithm did not incorporate sedation interruption as a strategy. A recent multicenter RCT evaluating sedation interruption failed to show benefit of such practice and suggested harm in a similar context (23). Thus, both groups Clinical Outcome Early Goal-Directed Sedation n = 21 Standard Sedation n = 16 p Time to randomization (hr) 2.2 [ ] 1.1 [ ] 0.43 CAM-ICU +ve, % (n) 38 (8) 38 (6) 0.97 Days with ve CAM-ICU 58% (101/175) 47% (54/114) 0.27 RASS 2 to 1 first 48 hr (%) 66 (203/307) 38 (74/197) 0.01 RASS 3 to 5 first 48 hr (%) 30 (93/307) 57 (112/197) 0.02 Mobilization, % (n) 43 (9) 50 (8) 0.67 Neuromuscular blockade, % (n) 10 (2) 0 (0) 0.50 Physical restraints, % (n) 5 (1) 31 (5) 0.03 Extubated within 7 days, % (n) 95 (20) 75 (12) 0.09 Vent free days at day 28 a 21.3 [9.2] 20.1 [10.1] 0.72 ICU LOS days b 5.5 [ ] 7.0 [ ] 0.44 Hospital LOS days b 16.1 [ ] 17 [ ] 0.49 Hospital mortality, % (n) 14.3 (3/21) 12.5 (2/16) day mortality, % (n) 23.8 (5/21) 12.5 (2/16) 0.38 CAM-ICU = Confusion Assessment Method for the ICU, RASS = Richmond Agitation Sedation Scale, LOS = length of stay. a Values are expressed as mean (sd). b Values are expressed as median [interquartile range]. Critical Care Medicine

8 Shehabi et al aligned with current practice of sedation in Australia and New Zealand (9). The significant reduction in the application of physical restraints is consistent with findings in SEDCOM, MIDEX, and PRODEX of greater ability of patients to communicate and cooperate with bedside nurses during dexmedetomidine infusion. This pilot study confirmed that EGDS delivered early provided the desired sedation level with clear process and outcome separation from standard sedation. In addition, the rate of recruitment and deferred consent process confirmed the feasibility of conducting an adequately powered randomized trial with a patient-centered primary outcome. The patients recruited in this study are representative of a general mixed medical-surgical adult ICU population with high severity of illness. This population is similar to that recruited in the Sedation Practice in Intensive Care Evaluation observational cohort (9) providing confidence that study inclusion and exclusion criteria are correct for the purpose of conducting a large RCT and mimic populations recruited in other phase II sedation trials (3, 4, 14, 21, 24). Protocolized ICU management of analgesia, sedation, and delirium has been shown to improve clinical outcomes including subsyndromal delirium and mortality (29) using ICDSC (19) to diagnose delirium. Our study, however, used the CAM- ICU despite recent reports of reduced sensitivity (30, 31). Light sedation was also shown to reduce ventilation time when compared with deep sedation in a randomized trial; it also reduced posttraumatic stress 4 weeks after hospital discharge (32). These reports highlight the importance of early delivery of sedation algorithms and strongly support the concept of EGDS. Strengths and Limitations This study was designed to test feasibility and separation between two intervention algorithms. Although major limitations relate to the small sample size and lack of blinding, this study has the strengths of being randomized and multicenter with short time to randomization, thus mimicking real life. The study design followed substantial background work to define current practice and the rationale for EGDS (9). It was centrally coordinated and monitored with systematic assessment of RASS and CAM-ICU by experienced staff. Although blinding in RCTs is highly desirable, this study was a process of care sedation trial with multiple interventions; therefore, blinding was not possible in this pilot and would not be feasible in our planned future process of care sedation RCT. The unblinded nature of the trial, however, may have introduced subjective bias by bedside clinicians. The trial size inevitably has a greater chance of type I and type II error. However, the feasibility outcomes were achieved consistently and strikingly, making a type I error unlikely. CONCLUSIONS A sedation algorithm (EGDS) comprising early intervention, targeting light sedation, and using dexmedetomidine as a primary sedative administered soon after intubation was feasible and effective. EGDS reduced the use of other sedatives and the need for restraints when compared with standard sedation. This pilot study confirmed the feasibility of early randomization and suggests that EGDS can be effective and safe. These findings justify further investigation of EGDS. REFERENCES 1. Patel SB, Kress JP: Sedation and analgesia in the mechanically ventilated patient. Am J Respir Crit Care Med 2012; 185: Barr J, Fraser GL, Puntillo K, et al: Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med 2013; 41: Riker RR, Shehabi Y, Bokesch PM, et al; SEDCOM (Safety and Efficacy of Dexmedetomidine Compared With Midazolam) Study Group: Dexmedetomidine vs midazolam for sedation of critically ill patients: A randomized trial. JAMA 2009; 301: Jakob SM, Ruokonen E, Grounds RM, et al; Dexmedetomidine for Long-Term Sedation Investigators: Dexmedetomidine vs midazolam or propofol for sedation during prolonged mechanical ventilation: Two randomized controlled trials. JAMA 2012; 307: Girard TD, Kress JP, Fuchs BD, et al: Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): A randomised controlled trial. Lancet 2008; 371: Wunsch H: Weighing the costs and benefits of a sedative. JAMA 2012; 307: Jackson DL, Proudfoot CW, Cann KF, et al: A systematic review of the impact of sedation practice in the ICU on resource use, costs and patient safety. Crit Care 2010; 14:R59 8. Mehta S, McCullagh I, Burry L: Current sedation practices: Lessons learned from international surveys. Crit Care Clin 2009; 25: , vii 9. Shehabi Y, Bellomo R, Reade M, et al; Sedation Practice in Intensive Care Evaluation (SPICE) Study Group and the ANZICS CTG: Early intensive care sedation predicts long-term mortality in mechanically ventilated critically ill patients. Am J Resp Crit Care Med 2012; 186: Deans KJ, Minneci PC, Danner RL, et al: Practice misalignments in randomized controlled trials: Identification, impact, and potential solutions. Anesth Analg 2010; 111: Kress JP, O Connor MF, Pohlman AS, et al: Sedation of critically ill patients during mechanical ventilation. A comparison of propofol and midazolam. Am J Respir Crit Care Med 1996; 153: Chamorro C, de Latorre FJ, Montero A, et al: Comparative study of propofol versus midazolam in the sedation of critically ill patients: Results of a prospective, randomized, multicenter trial. Crit Care Med 1996; 24: Pandharipande PP, Pun BT, Herr DL, et al: Effect of sedation with dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients: The MENDS randomized controlled trial. JAMA 2007; 298: Ramsay MA, Savege TM, Simpson BR, et al: Controlled sedation with alphaxalone-alphadolone. Br Med J 1974; 2: Riker RR, Picard JT, Fraser GL: Prospective evaluation of the Sedation-Agitation Scale for adult critically ill patients. Crit Care Med 1999; 27: Sessler CN, Gosnell MS, Grap MJ, et al: The Richmond Agitation- Sedation Scale: Validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med 2002; 166: Gélinas C, Fillion L, Puntillo KA, et al: Validation of the critical-care pain observation tool in adult patients. Am J Crit Care 2006; 15: Ely EW, Inouye SK, Bernard GR, 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: Bergeron N, Dubois MJ, Dumont M, et al: Intensive Care Delirium Screening Checklist: Evaluation of a new screening tool. Intensive Care Med 2001; 27: August 2013 Volume 41 Number 8

9 Clinical Investigations 20. Strøm T, Martinussen T, Toft P: A protocol of no sedation for critically ill patients receiving mechanical ventilation: A randomised trial. Lancet 2010; 375: Kress JP, Pohlman AS, O Connor MF, et al: Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 2000; 342: Bucknall TK, Manias E, Presneill JJ: A randomized trial of protocoldirected sedation management for mechanical ventilation in an Australian intensive care unit. Crit Care Med 2008; 36: Mehta S, Burry L, Cook D, et al; SLEAP Investigators and Canadian Critical Care Trials Group: Daily sedation interruption in mechanically ventilated critically ill patients cared for with a sedation protocol. A randomized controlled trial. JAMA 2012; 308: de Wit M, Gennings C, Jenvey WI, et al: Randomized trial comparing daily interruption of sedation and nursing-implemented sedation algorithm in medical intensive care unit patients. Crit Care 2008; 12:R Gerlach AT, Dasta JF: Dexmedetomidine: An updated review. Ann Pharmacother 2007; 41: Tan JA, Ho KM: Use of dexmedetomidine as a sedative and analgesic agent in critically ill adult patients: A meta-analysis. Intensive Care Med 2010; 36: Shehabi Y, Grant P, Wolfenden H, et al: Prevalence of delirium with dexmedetomidine compared with morphine based therapy after cardiac surgery: A randomized controlled trial (DEXmedetomidine COmpared to Morphine-DEXCOM Study). Anesthesiology 2009; 111: Devlin JW, Roberts RJ, Fong JJ, et al: Efficacy and safety of quetiapine in critically ill patients with delirium: A prospective, multicenter, randomized, double-blind, placebo-controlled pilot study. Crit Care Med 2010; 38: Skrobik Y, Ahern S, Leblanc M, et al: Protocolized intensive care unit management of analgesia, sedation, and delirium improves analgesia and subsyndromal delirium rates. Anesth Analg 2010; 111: Reade MC, Eastwood GM, Peck L, et al: Routine use of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) by bedside nurses may underdiagnose delirium. Crit Care Resusc 2011; 13: Shehabi Y: The search for the holy grail: Diagnosing delirium in the intensive care unit. Crit Care Resusc 2011; 13: Treggiari MM, Romand JA, Yanez ND, et al: Randomized trial of light versus deep sedation on mental health after critical illness. Crit Care Med 2009; 37: Knaus WA, Draper EA, Wagner DP, et al: APACHE II: A severity of disease classification system. Crit Care Med 1985; 13: APPENDIX 1 The SPICE Pilot Study site investigators are as follows (in alphabetical order all in Australia unless specified to New Zealand): Albury Base Hospital, Albury C. Maher, C. Mashonganyika; Auckland City Hospital DCCM, Auckland, New Zealand K. Benefield, C. McArthur, L. Newby, C. Simmonds; Austin Health, Melbourne R. Bellomo, G. Eastwood, L. Peck, M. Reade, H. Young; Calvary Health Care ACT, Canberra G. Ding, P. Hatfield, K. Smith; Nepean Hospital, Sydney C. Cuzner, I. Seppelt, L. Weisbrodt; St. Vincent s Hospital, Sydney P. Nair, C. Reynolds. Critical Care Medicine