Early Goal Directed Sedation, a bridge to better clinical outcomes

Critical care has become a central part of modern medicine and an integral essential to improve the outcomes in acute care. The increasing complexity of surgical procedures, severity of co-morbid diseases, life expectancy and societal expectation is likely to escalate the burden and the global need for critical care resources. While millions of patients are ventilated worldwide every year, projections suggest that the number of ventilated patients is expected to rise by 31% over the following decade.1 Almost all ventilated patients receive sedative agents to promote comfort and tolerance of endotracheal intubation and associated life-sustaining interventions.2 Despite its widespread use, sedation carries a significant risk,3 critically ill patients with impaired liver and renal function are, at an especially increased risk with accumulation sedative medications and its metabolites, leading to prolonged deep sedation (unintended drug-induced coma), respiratory depression, immune suppression, and adverse cardiovascular effects such as hypotension.4,5 Deep sedation contributes to immobility and weakness with prolonged ventilation and extended ICU stay.6 Reports have also linked complications associated with sedative use such as delirium and immobility with poor long-term outcomes, including impaired cognitive function, increased physical dependence and disability, depression and emotional hardship.7,8 Therefore, sedative strategies may be an important factor in determining survival and long-term outcomes after critical illness and mechanical ventilation. The Society of Critical Care Medicine Clinical Practice Guidelines on Pain, Agitation and Delirium has made two important recommendations, to provide analgesia before sedation and to maintain light sedation whenever clinically appropriate.9 This has supported clinicians change in practice towards better pain management, lighter sedation and awareness of delirium as an emerging issue in modern critical care.10 Strategies to achieve light sedation Following a landmark study by Kress et al,11 sedative interruption (DSI) has been, in many countries, the predominant strategy to achieve light sedation. However, this study was a single centre with small number (128 patients) who were recruited 48 hours after initiation of mechanical ventilation. A multicenter randomized trial compared a sedation protocol (to a target light sedation) with and without DSI in 423 critically ill mechanically ventilated patients, showed no difference in the duration of mechanical ventilation, ICU and hospital length of stay between the groups.12 This study also recruited patients up to 4 days after initiation of mechanical ventilation. A strategy of protocolised analgesia and sedation resulted in reduced sedative induced coma rates from 18.1% to 7.1%, with reduced total doses of sedative and opioids, and reduced ventilation and ICU time, in addition it also showed an absolute risk reduction in hospital mortality of 6.5%.13 Although this study was a before and after design, it highlighted the need to overcome the problem with delayed recruitment and late randomisation seen in previous trials. How often is light sedation achieved? While guidelines recommend and clinicians strive to achieve light sedation, it is not clear how often that is achieved. More importantly, it is not known how long it takes clinicians to achieve light sedation following an episode of mechanical ventilation and sedation. Patients in the cohort described by Pandharipande and colleagues spent a median of 3 days in coma in the early phase of critical illness.7 A multicenter multinational longitudinal observational cohort studies in 25 ICU in ANZ (n=251), 11 ICUs in Malaysia (n=259), with a total of more than 7 000 ICU days of follow-up, found that light sedation was only achieved after 72 hours, and deep sedation was very common in the first 48 hours following the initiation of mechanical ventilation. More importantly, early deep sedation independently predicted time to extubation, hospital and 180 days mortality.14,15 The above studies identified two important and potentially modifiable elements of sedation therapy: a high prevalence (76%) of deep sedation especially in the first 48 hoursand a high incidence of delirium (50.7% of all assessable patients) which increased substantially with longer ICU stay.14 The above observations suggest that clinicians pay little attention to sedation in the early phase of critical illness and only achieve light sedation late during an episode of critical illness. Although desirable, the ability of clinicians and bed side carers to achieve light sedation in the first 48 hours following ventilation and critical illness is highly questionable. Early Goal-Directed Sedation Most randomized trials, by virtue of design and need for consent, did not take into account early deep sedation. Amongst many other limitations, this has been a significant drawback of most sedation trials. Therefore, the potential benefits of any strategy which targets light sedation is significantly diminished by late randomization which allows non-systematic non-protocolised interventions and undetermined duration of deep sedation following initiation of mechanical ventilation. The above observations suggest that a strategy that reduces early deep sedation would have the potential to improve relevant and clinically important clinical outcomes such as mortality. Early goal directed sedation (EGDS) is a strategy that aims to maximize the time critically ill patients spend lightly sedated through achieving optimal sedation shortly after initiation of mechanical ventilation. EGDS is an algorithm which combines the tight control of sedation target and frequent assessment, titration and the optimal choice of sedative agents. Key elements of EGDS EGDS should be delivered early and shortly after initiation of sedation and mechanical ventilation, adopts an integrated process of targeting light sedation through frequent assessment of sedation levels, using a validated scale and incorporates the use of sedative agents known to promote wakefulness and arousal and/or agents with favorable pharmacokinetic profile such as short onset and offset time. The 2013 SCCM guidelines recommend early and adequate analgesia, routine use of targeted light sedation and suggested that sedation strategies using non-benzodiazepine sedatives (such as propofol or dexmedetomidine) may be preferred over benzodiazepines. This strengthened the choice of dexmedetomidine and/or propofol as the sedative agents to be used in EGDS. The alpha2 agonist dexmedetomidine has emerged over the last 15 years as a viable alternative to traditional sedatives for mechanically ventilated patients with a growing evidence that dexmedetomidine facilitates arousable sedation, shortens ventilation time, and attenuates the occurrence of delirium.16,17 Based on the above considerations, key elements of any plausible strategy directed at improving patient centered outcomes through reduction of early deep sedation MUST include: 1. Early effective analgesia titrated to effect according to standardized pain assessment; 2. Early delivery of sedative agents, including dexmedetomidine as a primary sedative as needed, shortly after initiating mechanical ventilation. 3. Tight targeting of light sedation early, from the time of initiation, with regular and frequent assessment of patient wakefulness/sedative state and titration of sedative infusions. 4. Avoidance of benzodiazepines and minimization of other sedatives, opioids and antipsychotics. 5. Reduced overall sedation depth with targeted light sedation. The EGDS Algorithm EGDS is designed to be delivered by bed side clinicians and nurses; it is simple, pragmatic and emulates real life sedation management of ventilated critically ill patients.9 While the description of the algorithm follows a step by step flow for clarity, the algorithm is designed to be delivered as a package with many interventions administered simultaneously (Figure 1).Figure 1.: Generic algorithm early goal directed sedation. This algorithm is best achieved with the use of dexmedetomidine as a primary sedative agent supplemented by low dose propofol, see text. There is no data to suggest that EGDS algorithm can be achieved with midazolam or other benzodiazepines.Once a patient is ventilated, pain assessment should be done; the use of a Numeric Reporting Scale 0—10 is an accepted and validated assessment tool for patients who are able to report pain. The Critical Care Pain Observation Tool (CPOT)18 is a validated and effective assessment tool inpatients who are unable to report pain. The provision of adequate analgesia using opiates by an infusion or incremental boluses is the commonest and the most effective pain management in most patients. Agents such as low dose ketamine (4—8 mg/h) and/or paracetamol can also be used. Care should be given not to use opiates or ketamine for the purpose of providing sedation but rather to alleviate pain only. As pain is assessed and treated, sedation level is assessed using any of the validated sedation scales. In our algorithm we use the Richmond Agitation Sedation scale (RASS).19 The target sedation level is light sedation defined as a RASS range between -2 and +1. Pain and sedation assessment is a continuous process that should be done at least every hour and on need basis. This will guide the use and titration of analgesic & sedative agents. During EGDS, we recommend the use of dexmedetomidine as a first line sedative agent starting at 1 mcg·kg-1·h-1 without a loading dose. The onset of the sedative effect for dexmedetomidine is about 45—60 minutes. During this period, if sedation is required, then propofol can be given by infusion at 10—70 mg/h or small boluses of 10—30 mg/h. The use of propofol should be scaled back to the lowest dose at all times to achieve the target RASS of -2 to +1. In more than 75% of patients, dexmedetomidine as a sole agent would be adequate. If the RASS score is -3 to -5 (moderate to deep sedation), propofol should be stopped first, followed by a gradual reduction in dexmedetomidine infusion by 0.2 mcg·kg-1·h-1 every 30 minutes. The process of titration is also a continuous process guided by RASS assessments. Caution is recommended in applying EGDS using the above algorithm in patients with head injury, cardiogenic shock and in patients with acute liver failure. EGDS may not be appropriate in patients needing extracorporeal support or high frequency ventilation, suffering seizure activity, receiving palliative therapy and in patients where death is imminent. Emergent agitation in EGDS A common problem associated with critical illness is breakthrough agitation. With EGDS, the following algorithm to manage agitation and/or delirium is suggested: 1. Increase the dose of dexmedetomidine a maximum dose up to 1.4 mcg·kg-1·h-1. 2. Add small boluses or infusion of propofol, in particular, to control acute emergent agitation, is recommended at a dose of 10—70 mg/h by infusion or blouses of 10—50 mg/h. For persistent agitation, add enteral quetiapine at a starting dose of 12.5 mg twice daily and up to 50 mg twice daily.20 If agitation persists, add haloperidol 2.5—5.0 mg IV blouses every 4—6 hours. For refractory agitation add a benzodiazepine such as midazolam boluses only, 1—3 mg as needed. In this case, further investigations such as electro-encephalogram EEG and/or a CT scan may be warranted. Is early goal directed sedation feasible? The premise of EGDS, is the provision of adequate analgesia and optimal sedation soon after the initiation of mechanical ventilation. The feasibility of EGDS is based on achieving key objectives; First, time to deliver EGDS Second, achieving optimal sedation. Third, can EGDS be achieved with conventional sedatives? Fourth, is EGDS achievable outside high intensity bedside nursing such as Australian based intensive care practice? And finally, is EGDS feasible in the context of randomized clinical trials? We tested the feasibility of delivering a process of care sedative interventions that had the above design features in a pilot randomized trial comparing EGDS with Standard Sedation in Australian ICUs.21 Randomisation occurred within a median (Interquartile range) of 1.1 (0.46—1.90) hours after intubation or ICU admission for out of ICU intubation. A significantly higher proportion of patients were lightly sedated on day 1, 2 and 3 in the EGDS (12/19 (63.2%), 19/21 (90.5%) and 18/20 (90%) compared with 2/14 (14.3%), 8/15 (53.3%) and 9/15 (60%) in standard care (P=0.005, 0.011, 0.036)). More RASS assessments in the EGDS were between (-2 to +1), in the first 48 hours 203/307 (66%) vs. (74/197(38%) (P=0.01) compared to STDS, respectively. EGDS patients received midazolam on 6/173 (3.5%) of study days while STDS patients received dexmedetomidine on 4/114 (3.5%) patient-days. Propofol was given to 16/21 (76%) of EGDS vs. 16/16 (100%) of STDS patients (P=0.04). EGDS patients required significantly less physical restraints 1 (5%) vs. 5 (31%) (P=0.03) than STDS patients. There were no differences in vasopressor use and self-extubation.21 We also conducted a similar pilot in 10 ICUs in Malaysia (n=60) and concluded in May 2013. This pilot was designed to test the feasibility of achieving EGDS using conventional sedatives and in addition test the feasibility of EGDS in ICUs outside an Australian based ICU practice. The study showed that even with targeted protocol, EGDS could not be delivered with conventional sedatives. Furthermore, this pilot proved that EGDS is a concept that can be delivered in tertiary, regional and rural ICUs. The above data supports the concept of EGDS as an innovative strategy to reduce deep sedation that could be delivered in the context of randomised control trial. Despite the comprehensive review undertaken, the 2013 SCCM Guidelines called for definitive studies to inform the practice of sedation in critical illness. The paucity of data on long-term and patient centered outcome is a stark reminder for the need of adequately powered well designed clinical trials into sedation practice. There are many limitations of previous sedation trials22 that limited the external validity and general applicability of the evidence produced in different care settings. In conclusion, although there is an acceptance that light sedation should be targeted whenever clinically safe, deep sedation for up to 72 hours after initiation of mechanical ventilation is common, unrecognized, unjustified and is a strong independent predictor of time to extubation and long-term mortality. The first 48 hours of mechanical ventilation, which represent at least 33% of total ventilation time in most patients.14 A strategy to reduce early deep sedation such as Early Goal Directed Sedation is feasible, effective and safe. Testing the clinical effectiveness of EGDS on long-term and patient centered outcomes is an important clinical priority to inform future sedation practice that improves patients' outcome.