Monitoring Delirium in Critically Ill Patients
2003; American Association of Critical-Care Nurses; Volume: 23; Issue: 2 Linguagem: Inglês
10.4037/ccn2003.23.2.25
ISSN1940-8250
Autores Tópico(s)Anesthesia and Neurotoxicity Research
ResumoDelirium is one of the most frequent complications experienced in the ICU. Critical care nurses must be able to recognize delirium at the bedside. The CAM-ICU, presented in this article, is a valid, reliable, quick, and easy-to-use serial assessment tool for monitoring delirium in ICU patients.Critical care nurses are well aware that many patients in medical and surgical intensive care units (ICUs) experience some degree of cognitive impairment, which may range from coma to delirium. In order to perform invasive procedures and provide lifesaving, supportive care, patients are given potent psychoactive drugs such as benzodiazepines and opiates, which often lead to increases in cognitive abnormalities. For many healthcare providers, cognitive impairment in the ICU is expected, temporary, and of little consequence; that is, it is part of the "ICU psychosis." However, recent studies indicate that delirium, one of the most frequent complications in the ICU, is actually an independent risk factor for prolonged length of stay1 and that many patients who experience delirium may have prolonged neurocognitive deficits after discharge from the hospital.2,3 In the 2002 clinical practice guidelines of the Society of Critical Care Medicine4 for the sustained use of sedatives and analgesics in critically ill adults, patients' comfort is recognized as a primary goal in the ICU, and pain control, adequate sedation, and minimization of delirium are recommended ways to achieve this goal.Studies5–9 have indicated that delirium goes underrecognized more than two thirds of the time. In this review, we discuss the important features of delirium, its etiology, its risk factors, a new delirium assessment tool for the ICU, and interventions to minimize this complication.Recent reports indicate that delirium develops in more than 8 of 10 patients in the ICU.10,11 In other ICU cohorts12,13 with less severe illness, the delirium rates could be lower. Despite this prevalence, delirium remains unrecognized in 66% to 84% of patients whether they are in an ICU, a hospital ward, or an emergency department.5,6,14,15 Furthermore, the prevalence of delirium is likely to increase as aggressive care of older persons increases.16 Delirium is considered 1 of the top 3 most important target areas for improvement in quality of care in vulnerable older adults.17 In addition, delirium in the non-ICU setting has repeatedly been associated with prolonged hospital stay, medical complications that can increase mortality, greater dependency on care at the time of discharge, and higher rates of discharge to nursing homes.18–24 In patients in medical and coronary ICUs, delirium is an independent predictor of prolonged ICU and hospital lengths of stay and of mortality rates 6 months after discharge,1,25 even after adjustments are made for age, sex, ethnicity, and severity of illness. Delirium may also predispose ICU survivors to prolonged neuropsychological deficits.2,3,26 The high prevalence and seriousness of these adverse outcomes should establish delirium as a major concern for ICU staff.More than 25 different terms have been used to describe the spectrum of cognitive impairment in the ICU, including ICU psychosis, ICU syndrome, acute confusional state, septic encephalopathy, and acute brain failure.26–28 Recently, both the medical and the nursing literature indicate that the signs and symptoms of ICU psychosis are consistent with delirium,26,28–31 a consensus that will allow improved recognition, understanding, and communication concerning this impairment.Delirium is often confused with dementia. Although cognitive impairment occurs in both abnormalities, distinct characteristics differentiate them.29,31,32 Dementia is defined as memory impairment and cognitive disturbance that develops gradually (over months and/or years) and progressively worsens.33 Because many of the features of delirium are distinct from those of dementia, detection of delirium is possible in most patients who have preexisting dementia.10,11,34,35 The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition33 officially defines delirium as a disturbance of consciousness with inattention accompanied by a change in cognition or perceptual disturbance that develops during a short period (hours to days) and fluctuates over time. Changes in cognition may be manifested as memory impairment, disorientation, and rambling or irrelevant speech. Perceptual changes such as hallucinations (usually visual), illusions, and delusions are not required for the diagnosis of delirium and occur less often.Delirium is categorized according to level of alertness and level of psychomotor activity. The clinical subtypes are hyperactive, hypoactive, and mixed29,32,36–38 (Table 1). Hypoactive delirium is often referred to as "quiet" delirium.18,23,39 The mixed subtype is due to the fluctuating nature of delirium. For example, agitated patients with "positive" indications of hyperactive delirium may be given sedatives in the ICU to calm them and then emerge from the sedation in a markedly hypoactive delirious state.Patients with delirium are at risk for various associated outcomes. Patients with hyperactive delirium who are agitated or combative may remove tubes, pull out central venous access catheters, and even fall out of bed.40–42 Patients with hyperactive delirium are often given higher doses of sedatives that commit them to at least another day of mechanical ventilation and place them at risk for being left in a cognitively impaired state and for receiving mechanical ventilation unnecessarily.43In non-ICU settings, the prevalences of the delirium subtypes are 30% hyperactive, 24% hypoactive, and 46% mixed.37 However, in the ICU, because of the magnitude of psychoactive medications given and the elevated severity of illness, hypoactive delirium may be more prevalent than hyperactive delirium. Recent reviews36–38 indicate that in general medicine units, hypoactive delirium is associated with longer duration of delirium, longer hospital stays, and higher mortality than are hyperactive and mixed delirium. Hypoactive delirium may be associated with a worse prognosis because of complications such as aspiration, pulmonary embolism, decubitus ulcers, and other outcomes related to immobility.26 Therefore, all ICU patients should be monitored for delirium at regular intervals.The exact pathophysiological mechanisms involved in the development and progression of delirium are unknown.7 However, they are thought to be related to imbalances in the neurotransmitters that modulate the control of cognitive function, behavior, and mood.29,38 The 3 main neurotransmitter systems involved in the pathophysiology of delirium are dopamine, γ-aminobu-tyric acid, and acetylcholine.28,44 Whereas dopamine increases excitability of neurons, γ-aminobu-tyric acid and acetylcholine decrease it.28 An imbalance in any of these 3 (or a combination of these 3) systems yields neuronal instability and therefore unpredictable and inconsistent neuro-transmission. In addition to these 3 neurotransmitter systems, the following may also be involved in the development of delirium: serotonin imbalance, endorphin hyperfunction, increased central noradrenergic activity, and damaged interneuronal enzyme system.38,44Neurotransmitter imbalance is associated with various causal factors, including reduction in cerebral metabolism, primary intracranial disease, systemic diseases, secondary infection of the brain, exogenous toxic agents, withdrawal from substances of abuse such as alcohol and sedative-hypnotic agents, hypoxemia and metabolic disturbances, and the administration of psychoactive medications such as benzodiazepines and narcotics.45 Because the cerebral concentrations of neurotransmitters are sensitive to many organic and biochemical changes, many conditions can result in an imbalance.44,45 Although environmental factors play a role in the development and escalation of delirium, delirium usually occurs after marked physiological and metabolic changes that lead to organ dysfunction.33Inouye et al46 developed a helpful predictive model for delirium and divided risk factors into 2 categories: predisposing and precipitating. Predisposing factors are present on admission to the hospital and indicate baseline vulnerability; precipitating factors include noxious stimuli or injuries and/or hospital-related factors that contribute to the development of delirium. Patients with high vulnerability may become delirious in response to any precipitating factor; those with low vulnerability become delirious only in response to particularly noxious stimuli.Diseases and disorders that predispose patients to a high level of vulnerability to delirium include those that result in a chronic imbalance in 1 or more of the neurotransmitter systems described earlier. For example, the availability of acetylcholine is reduced in Alzheimer disease,45 and the central cholinergic and dopamine systems are affected by aging.38,45 Similarly, any condition that results in acute imbalances of the neurotransmitters is considered a precipitating factor. Precipitating factors include hypoxia, metabolic disturbances, electrolyte imbalances, withdrawal syndromes, acute infection (intracranial and systemic), seizures, dehydration, hyperthermia, head trauma, vascular disorders, and intracranial space-occupying lesions.28,47–49In addition, medications are a large subclass of precipitating factors. For example, neurotransmitter levels are affected by drugs with anticholinergic properties,28,29 and psychoactive medications are the leading iatrogenic risk factor for delirium.23,46,50 Benzodiazepines, narcotics, and other psychoactive drugs are associated with a 3- to 11-fold increase in relative risk for the development of delirium.23 However, in a recent study,12 although use of opiates was strongly related to the development of delirium, use of benzodiazepines and propofol was not. Clearly, more research is needed to determine the specific agents associated with delirium.Numerous risk factors (both predisposing and precipitating) for delirium have been identified14,18,19,22,26,31,51–55 (Table 2). For non-ICU patients with 3 or more risk factors, the likelihood of developing delirium is 60%.22 Because of the increased severity of illness and the use of catheters, invasive procedures, and psychoactive medications in the ICU, ICU patients are at great risk for delirium. In a recent study,56,57 ICU patients had more than 10 risk factors for delirium. In another study done in an ICU,12 hypertension, smoking history, abnormal bilirubin level, use of morphine, and the presence of an epidural catheter were statistically associated with delirium. Of interest, some of the traditional risk factors found in other studies (eg, age, severity of illness, renal failure, and history of dementia) were not significantly associated with delirium.12Because more than 95% of ICU patients receive sedatives, opiates, and other psychoactive agents, they are at high risk for this complication due to medications alone. The intent of these drugs is to reach a quality of sedation that optimizes patients' cooperativeness, comfort, and ability to tolerate supportive care. However, critically ill patients are often deeply sedated and unresponsive for extended periods, and their clinical outcomes may be worse than those of similar patients cared for with less intense or prolonged sedation.58 As a result of prospective, randomized controlled trials,59,60 daily interruption of bolus doses and continuous infusions of sedatives is recommended to improve neurological assessments of wakefulness and content of consciousness.The guidelines of the Society of Critical Care Medicine recommend that all critically ill patients be simultaneously monitored for level of sedation and for delirium.4 Well-validated, reliable, objective, brief assessment tools are needed to monitor for both components of consciousness (level and content).61 Level of consciousness, measured by sedation scales, only describes patients' level of arousal, whereas assessment of "content of consciousness" describes patients' ability to pay attention and organize thoughts. Patients should be monitored for sedation and delirium by using a 2-step approach.The first step is to assess a patient's level of consciousness or sedation. The recommended standard of care is to use objective assessment scales to avoid oversedation and to promote earlier weaning from mechanical ventilation.4,58,60,62–64 Sedation scales help provide common language for the multidisciplinary team to use when discussing goals and treatments for patients.64–66 Although the Ramsay Scale67 has been the most widely used instrument for decades,68 other recently developed instruments have been better validated and have more objective parameters.69,70 One of these is the Richmond Agitation Sedation Scale (RASS)11,71,72 (Table 3). All patients who are at least minimally responsive to verbal stimuli should be assessed for delirium. In other words, all patients who have RASS scores of -3 or lighter level of sedation (-3 to +4) must progress to step 2.Instruments used to assess delirium should be based on observation, be appropriate for use at the bedside during regular care, be suitable for repetitive use, and not be overly burdensome to patients.7,10,29 A 1990 review73 of instruments for screening cognitive and mental status in older adults does not list any tools for assessment of delirium. Since then, a few tools such as the Delirium Rating Scale,74 the Confusion Assessment Method (CAM),75 and the Cognitive Test for Delirium76 have been developed. Neither the Delirium Rating Scale nor the CAM was designed for use with ICU patients, who are usually unable to speak because they are intubated and receiving mechanical ventilation.The first delirium assessment tool designed specifically for ICU patients, the CAM for the ICU (CAM-ICU), has been validated. The CAM-ICU was adapted for use in nonverbal ICU patients from the original CAM, a well-validated scale for assessment of delirium that is widely used and easy to administer. Two investigations10,77 involving use of geriatric-psychiatric ratings based on the Diagnosis and Statistical Manual of Mental Disorders, Fourth Edition, and more than 750 observations of patients indicated the validity and reliability of the CAM-ICU. This instrument can be used even with difficult populations of patients (ie, patients who may have dementia, patients >65 years old, and patients with very high severity of illness). The CAM-ICU was designed to be a serial assessment tool for use by bedside clinicians (nurses and physicians). Thus, it is easy to use; it takes only 2 minutes to complete and requires minimal training.Assessment of delirium with the CAM-ICU incorporates the same 4 key features as the original CAM:Delirium is present when both features 1 and 2 and either feature 3 or feature 4 are present (Figure 1).For patients with RASS scores of -3 to +4 (or a comparable level on any of the other sedation scales), each bedside assessment should include a detailed assessment of the 4 CAM-ICU features (Table 4). In order to determine if an alteration or a fluctuation in mental status has occurred (feature 1), the patient's current mental status must be compared with the baseline information obtained at the time of admission. Baseline information can be obtained directly from the patient, from his or her family and friends, and/or from recently written medical records. Serial measurements of sedation obtained by using the RASS or other validated sedation scales and the Glasgow Coma Scale are helpful in detecting fluctuations from baseline. These scales are standard tools in most ICUs. A change from baseline or any fluctuation during the previous 24-hour period in the patient's scores on the Glasgow Coma Scale or the RASS indicates the presence of feature 1.Feature 2 of the CAM-ICU is inattention. Attention refers to the ability to attend to a specific stimulus without being distracted by extraneous internal or environmental stimuli.78 Clinically, a patient is considered inattentive when he or she cannot sustain sufficient attention to succeed in the simple tests of attention.78 The Attention Screening Exam (ASE) of the CAM-ICU includes a visual component and an auditory component, which were derived from previously validated tools used to assess attention.10,11 Both components have been validated.11,76,78,79 Both are simple to administer and do not require verbal responses from patients.For the visual component of the ASE, the patient is shown 5 simple pictures at 3-second intervals and asked to remember them (Figure 2). The patient is then immediately shown 10 more pictures and asked to nod yes or no for each picture to indicate whether or not he or she has just seen it. Because 5 of the 10 pictures have been shown to the patient already (correct nod = yes), and the other 5 are new (correct nod = no), a perfect score is 10 correct nods (5 yes nods and 5 no nods). Errors are scored for errors of omission (indicating no for a previously shown picture) and for errors of commission (indicating yes for a picture not previously shown).For the auditory (letter) component of the ASE, the nurse says, "I am going to read you a series of 10 letters. Whenever you hear the letter A, indicate by squeezing my hand." The nurse then reads the 10 letters (S, A, H, E, V, A, A, R, A, T) in a normal tone at a rate of 1 letter per second. The scoring is similar to that used for the visual component. Errors are counted when the patient does squeeze the nurse's hand on the letter A and when the patient squeezes the nurse's hand on any letter other than A. Feature 2 (inattention) is present if the patient has fewer than 8 correct answers in either the visual component or the auditory component of the ASE.Disorganized thinking is by far the hardest area to assess in nonver-bal patients. This feature is the most subjective of the 4. Thought is expressed via words (verbalized or written). Mechanical ventilation and loss of fine motor movement limit this expressive ability in most ICU patients. Therefore, the CAM-ICU uses easy, straightforward yes/no questions and simple commands to assess organization of thought. Disorganized thinking (feature 3) is present if the patient cannot answer at least 3 of the 4 questions correctly and cannot complete simple commands (Table 4). Feature 4, altered level of consciousness, is present if the nurse rates the patient as having any level of consciousness other than "alert" when the standardized definitions presented in Table 4 are used.The following is an illustration of the use of the CAM-ICU in a critically ill patient. The patient, Mr Y., was a 45-year-old man who was admitted to the ICU after being found unresponsive. He had diabetic ketoacidosis and community-acquired pneumonia. After he arrived in the ICU, his oxygen saturation decreased markedly despite supplemental oxygen, and mechanical ventilation was started. On day 2 of his ICU stay, he was awake, but agitated; he pulled at his gown and linens and attempted to remove his central venous catheter. His family was interviewed to provide baseline information about Mr Y. They reported that normally he functioned at a high level and was a practicing lawyer with no visual or hearing deficits. He had no memory or attention problems.On ICU day 2, Mr Y. was hyper-alert, with an RASS score of +3. On the auditory (letter) component of the ASE, he scored 6/10. On the visual (picture) component, he responded yes to all the second-round pictures for a score of 5/10. When asked the 4 questions to assess for disorganized thinking, he did not answer any of them correctly, and he could not determine how many fingers the nurse was holding up. The nurse assessed his level of consciousness as vigilant or hyperalert. According to this assessment, all 4 features of the CAM-ICU were present, and Mr Y. would be considered CAM-ICU positive and in hyperactive delirium.Mr Y. was given lorazepam every 4 hours throughout the night. On day 3 in the ICU, he opened his eyes to verbal stimulus and made eye contact with the nurse, but he did not maintain eye contact for longer than 10 seconds (ie, RASS score -2). He could not keep his eyes open very long and therefore could not complete the visual (picture) component of the ASE. He did attempt to complete the auditory (letter) component; however, his score was 3/10. He responded to only the first of the 4 questions used to detect disorganized thinking and could not follow any commands. Again all 4 features of the CAM-ICU were present, and he was in quiet or hypoactive delirium.The next morning, Mr Y.'s breathing continued to improve. He woke up more and began to breathe on his own, and mechanical ventilation was discontinued. The late evening assessment on day 4 was as follows. He was alert and calm with an RASS score of 0 (previous RASS score was -2). He had a score of 8/10 on the visual component of the ASE and a score of 9/10 on the auditory component. He answered 2 of 4 questions correctly. He said he did not have unclear thinking, and he correctly followed the directions to hold up the fingers. In this assessment, features 1 and 3 were present, but features 2 and 4 were not. Mr Y. was not delirious.The first step in managing patients with delirium is early recognition of the complication. Once delirium is detected, efforts should focus on determining the cause. Often, the etiology can be determined by assessing for the presence of risk factors. Both prevention and treatment should focus on minimizing and/or eliminating predisposing and precipitating factors. The goals of management are to improve patients' mental status and reduce safety risks.Primary prevention is preferred; however, some degree of delirium most likely is inevitable in the ICU. Although no data are available on primary prevention or use of non-pharmacological interventions in the ICU, the data obtained in non-ICU settings focus on minimizing risk factors. Strategies include the following80,81: repeatedly reorienting patients, providing cognitively stimulating activities several times a day, following a nonpharmacological sleep protocol, using early mobilization activities, using range-of-motion exercises, removing catheters and physical restraints in a timely manner, having patients use eye glasses and magnifying lenses, having patients use hearing aids and removing earwax, correcting dehydration soon after its onset, following a scheduled pain protocol, and minimizing unnecessary noise and stimuli. In one study,80 consistent implementation of these interventions resulted in a 40% reduction in the development of delirium. Involvement of their families can be helpful in reorienting and soothing delirious patients. Patients' family members should also be educated about delirium, with emphasis on its temporary nature and likely fluctuation.28,29The first step in pharmacological treatment of delirium is to determine if any of the patient's current medications may be adding to the delirium. Inappropriate drug regimens for sedation or analgesia can exacerbate delirium.4 Patients who are psychotic or delirious may become more mentally dulled and confused when treated with sedatives, causing a paradoxical increase in agitation.82 In fact, benzodiazepines and narcotics that are often used in the ICU to treat "confusion" (delirium) actually worsen cognition and exacerbate the problem. A thorough review of a patient's medications will reveal any sedatives, analgesics, or anticholinergic drugs that can be discontinued or decreased in dosage.Currently, no drugs have been approved by the Food and Drug Administration for the treatment of delirium. The guidelines of the Society of Critical Care Medicine recommend haloperidol for the treatment of delirium, though this recommendation is based on sparse outcomes data from nonrandomized case series and anecdotal reports (ie, level C data).4,83–91 Haloperidol is a dopamine receptor antagonist and inhibits dopamine neurotransmission. It is used to treat positive symptoms (eg, hallucinations, unstructured thought patterns) and often results in a sedative effect (may lead to hypoactive delirium).92 Neither haloperidol nor similar agents (ie, droperidol and chlorpromazine) have been extensively studied in ICU patients.4 Patients receiving haloperidol should be monitored for adverse effects such as QT prolongation, arrhythmias, and extrapyramidal effects.4 Other antipyschotic and neuroleptic agents (eg, risperidone and olanzapine) may also be helpful for delirium.4 However, prospective randomized controlled trials are needed to evaluate the effectiveness and safety of these agents relative to one another.4Critical care nurses are in a unique position to improve patients' quality of care and outcomes by recognizing delirium early, determining the causes, and providing knowledgeable care.30,81 Many patients are at great risk for the development of some type of delirium in the ICU; however, delirium is grossly under-recognized and undertreated. ICU nurses are on the front line for detecting and monitoring delirium. Accurate identification and prompt modification of the risk factors that increase patients' risk for delirium may prevent many adverse outcomes associated with this phenomenon. Likewise, appropriate interventions may decrease the severity and/or length of delirium. Increased recognition of delirium is the first step in decreasing the toll of delirium and maximizing patients' comfort in the ICU.93 More research is needed on the outcomes of delirium and on the use of delirium assessment tools in standard bedside assessments.A free training manual that includes CD videos, FAQs, peer-reviewed publications, and a kit to implement the CAM-ICU is available. For information, contact Brenda Truman at brenda.truman@vanderbilt.edu or (615) 936–1010.Financial support was provided by the Division of Allergy, Pulmonary, and Critical Care Medicine, Vanderbilt University Medical Center, and the Geriatric Research and Education Clinical Center, Veterans Affairs Tennessee Valley Healthcare System. In addition, Dr Ely is a recipient of the Pharmacology in Aging Grant from the American Federation for Aging Research and the Paul Beeson Faculty Scholar Award from the Alliance for Aging Research. He is a recipient of a K23 from the National Institutes of Health (AG01023–01A1).
Referência(s)