Sedation and Analgesia in the Critically Ill (Review)
Pain, agitation, and delirium are all extremely common in ICU patients--so much so that they've been termed the "ICU triad." No one knows exactly how common each is, because ICU patients are often too delirious to complain of pain; or their agitation hides their delirium; or their unidentified pain may cause their agitation; or .... you get the picture. Pain, agitation, and delirium can have dozens of causes, frequently coexist interdependently, and (let's be honest) are probably impossible to parse out accurately and consistently in most ICU patients, especially after sedation and analgesia (for pain or agitation) have been applied.
But, we do what we can. The good news is that since pain, agitation, and delirium are tightly linked, reducing one may favorably impact the others. The general principles: try nonpharmacologic methods, followed by the lowest possible doses of a rationally applied blend of analgesics and sedative-hypnotics to maintain patient comfort.
Pain and Analgesia in ICU Patients
Do most patients in the ICU suffer from pain? The NEJM review offers this definitive statement: "prospective studies confirm that the majority of patients who are treated in ICUs have pain." Certainly true if we count needlesticks, but do mechanically ventilated patients experience more pain than medical ward patients? The supporting reference, a qualitative review from 2000 in the American Journal of Critical Care is not available for online viewing, suggesting to me that few of the 147+ citing authors have read it, or evaluated its original sources. They are mostly from 1985-1990 and of the few that are online-accessible, none I saw included a convincing assessment of pain in medical ICU patients by today's standards. One concluded "no correlation to pain was found" with mechanical ventilation.
The NEJM authors modulate their assertion by noting "assessing whether a patient in the ICU is in pain may be difficult ... patients may not be sufficiently interactive to give valid responses ... hypertension and tachycardia correlate poorly with more intuitively valid measures of pain."
Suffice it to say that pain in critically ill medical patients (non-surgical, non-trauma) is hard to study objectively, for the reasons described above. The validated Behavioral Pain Scale and Critical Care Pain Observation Tool are considered the best objective methods for assessing pain in critically ill patients. These have been validated in communicative patients using their pain reports as the gold standard; for noncommunicative patients, their reactions during known painful procedures were used as the defining/validating standard.
Endotracheal tubes and mechanical ventilation are probably better described as noxious or uncomfortable than painful for most patients (since some can calmly tolerate extended intubation without any analgesia or sedation at all). What we know for sure is that patients on ventilators often grimace and writhe when they are moved, or when their limbs, heads or sensitive areas are manipulated. In one review of 171 communicative patients in 2 U.S. ICUs at high risk of dying, 40% reported pain.
In practice, low-dose continuous opioid infusions (e.g., fentanyl) are often employed to treat any pain that may be there in noncommunicative patients, and also for their sedating and calming effects. Morphine is avoided for its metabolites' toxicity in patients with renal failure, which is highly prevalent in ICU patients.
An "analgesia first" approach for delirious or agitated patients has been advocated, including by the Society for Critical Care Medicine. Generally speaking, in cases of uncertainty as to the presence of pain after a patient evaluation, it is recommended to give analgesics (preferably in bolus doses) and reassess.
Sedation in Critically Ill Patients
ICU patients have historically been oversedated, unnecessarily extending ventilator days and ICU stays. The ascendance of sedation-limiting protocols as standard care has seemed to reduce the prevalence of oversedation to some extent.
Only a minority of critically ill patients require deep sedation, for conditions such as severe respiratory failure (e.g. ARDS), intracranial hypertension, refractory status epilepticus, and those receiving neuromuscular blocking agents.
Patients undergoing mechanical ventilation usually receive some degree of pharmacologic sedation, because of the anxiety and discomfort that are widely attributed to the experience. Randomized and observational studies confirm the benefits of limiting sedative-hypnotic agents to the minimum possible dose, as well:
Kress et al (NEJM 2000) showed daily sedation interruptions resulted in less total sedation and fewer days on the ventilator and in the ICU, in a single-center randomized trial.
This was confirmed by Girard et al who paired "sedation vacations" with spontaneous breathing trials in a multicenter randomized trial (Lancet 2008).
Patients receiving morphine PRN only for sedation weaned from the ventilator and left the ICU faster than those receiving continuous sedation with daily interruptions, in a single center randomized trial (Lancet 2010).
A more recent, multicenter RCT showed no advantage of daily sedation interruptions over a protocolized strategy to reduce sedation to the minimum possible (JAMA 2012). Sedation vacations actually resulted in greater total doses of sedatives.
In a prospective observational multicenter trial of 251 patients, use of early deep sedation was independently associated with greater time on the ventilator and with mortality in-hospital and at 6 months (AJRCCM 2012).
In a randomized trial of 127 patients, lighter sedation did not result in greater frequencies of post-traumatic stress disorder or other mental health disturbances.
Which Sedative Should You Use in the ICU?
After more than 90 comparative studies, no single sedative agent has been shown superior to all others for use in the ICU. The wide practice variation in sedation use within and across countries suggest that culture, familiarity, and local purchasing decisions drive the choice of sedatives in the ICU.
Propofol and remifentanil (an opioid with sedating effects) are the shortest acting sedatives, providing a theoretical advantage if the goal is to titrate sedation to the lowest dose possible. The GABA-modulating benzodiazepines midazolam and lorazepam are probably the most commonly used sedative agents in the critically ill, probably owing to tradition and their lower cost per daily dose. Dexmedetomine (Precedex) is a central α2-adrenoceptor agonist that some regard as superior to benzodiazepines for its analgesic properties, reduced respiratory depression, and patients' greater awareness and cooperativeness while sedated.
The newer and more expensive agents, propofol and Precedex, have been tested against benzodiazepines and each other, often in industry-funded randomized trials intended to establish the new agents' place in ICU care:
Dexmedetomidine resulted in less delirium and fewer ventilator-days than midazolam or lorazepam, but did not reduce length of stay in the ICU or hospital, in 3 randomized trials (JAMA 2009, 2007, 2012).
Propofol appeared equivalent to Precedex in sedation efficacy, length of stay and ventilator-days in the 2012 JAMA trial.
A 2013 meta-analysis of 6 trials (including some of the above) suggested nonbenzodiazepines (dexmedetomidine or propofol) reduced ICU length of stay by ~1.5 days, ventilator days by ~2, but had no impact on delirium or short-term mortality rates, as compared to benzodiazepines (midazolam or lorazepam).
Remifentanil has not been compared against other agents in large randomized trials and its use remains relatively uncommon.
Current critical care guidelines (SCCM's) "suggest that sedation strategies using nonbenzodiazepine sedatives (either propofol or dexmedetomidine) may be preferred over sedation with benzodiazepines (either midazolam or lorazepam) to improve clinical outcomes in mechanically ventilated adult ICU patients." (Grade 2B, a weak recommendation / suggestion based on moderate quality evidence).
The potential shorter times on ventilators and in ICUs have been used in cost analyses to justify the increased cost-per-daily-dose of nonbenzodiazepines.
Use a Sedation Scale to Titrate Sedatives
The use of objective sedation scales is essential for proper titration of sedatives, meaningful communication among care team members, and application of clinical trial results to critically ill patients in the community. Use of sedation scales markedly reduced time on mechanical ventilation in a randomized trial of 102 patients.
The Riker Sedation–Agitation Scale (RSAS) (target score: 3-4) and the Richmond Agitation–Sedation Scale (RASS) (target score: 0 to -1) are both validated scales with good interobserver reliability, and both are in common use. The RSAS and RASS correlate well with each other. Both scores' target values indicate a calm and cooperative patient; lower scores (1-2 on RSAS; -5 to -2 on RASS) indicate deeper sedation and higher scores (5-7 on RSAS; +1 to +4 on RASS) indicate worsening agitation.
The 2 sedation scales have been tested head-to-head; it was a tie. Both are good; pick one and stick to it (although hopefully, your hospital and nursing team already has done so).
Michael C. Reade and Simon Finfer. Sedation and Delirium in the Intensive Care Unit. N Engl J Med 2014; 370:444-454.
Barr J 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 Jan;41(1):263-306. doi: 10.1097/CCM.0b013e3182783b72.