Delirium is a psychiatric syndrome produced by an underlying medical aetiology and it is a common complication in the postsurgical setting. It can be difficult to detect and is frequently missed by clinicians. It is important to recognise and prevent delirium, given its association with serious and costly negative outcomes. It is characterised by disturbances of awareness, attention and cognition including perception, thinking, memory, language and orientation. Psychomotor behaviour, emotion and sleep–wake cycle are also observed disturbances in delirium. Often abrupt in onset, delirium tends to fluctuate throughout the day and is variable in severity and duration. It can last from hours to days, and less commonly for weeks to months.
Negative clinical consequences associated with postoperative delirium include prolonged hospitalisation, loss of functional independence, reduced cognitive function, incomplete recovery, delayed rehabilitation and death. The duration of delirium in the intensive care unit setting is correlated with an increased cognitive impairment in ICU survivors, followed as outpatients up to 1 year after discharge. Delirium can also be associated with significant distress in patients and is frightening for family members who stand by and watch the pronounced and often perplexing changes in the behaviour of a loved one. In patients who undergo cardiac surgery, delirium is associated with postoperative complications including respiratory insufficiency and sternum instability. In the USA, the health care costs due to all episodes of delirium in hospitalised patients range from $38 to $153 billion annually, and among older hospitalised medical patients it has been shown to be preventable in up to 40% of cases.
Delirious patients are frequently overlooked or misdiagnosed and prevention strategies are not yet a part of routine care in most surgical units. The growing body of literature demonstrating the negative short-term and long-term impacts associated with delirium highlights the need to improve recognition, treatment and prevention of this syndrome. After describing the epidemiology, this chapter will review prevention strategies, and discuss detection and treatment of postoperative delirium in adult patients.
Delirium presents as a sudden change in ability to think, resulting in faulty attention. Common findings in delirium include:
Inability to attend and concentrate resulting in disrupted short-term recall and disorientation to time and place;
Change in word finding ability and language usage – talk may seem confused and difficult to follow;
Disrupted visuospatial abilities – clock drawing or copying interlocking pentagons;
Changes in emotional tone – uncharacteristic mood lability and disinhibition;
Change in level of arousal/sleep–wake cycle;
Perceptual changes including hallucinations (usually visual);
Unusual beliefs, which are often frightening and not uncommonly include misinterpreting the intensions of the medical staff.
A recognised feature of delirium is that the symptoms can fluctuate in severity throughout the day and night.
Three subtypes of delirium are distinguished by varying levels of psychomotor activity: hyperactive, hypoactive and mixed type.
The hyperactive subtype is associated with:
Restlessness, psychomotor agitation and seemingly purposeless movement;
Aggression on the part of the patient, often arising from fearfulness and misinterpretation of the intentions of the staff and/or disorientation to surroundings;
Easy recognition that something is wrong with the patient;
Agitated behaviour that may be misattributed to a primary psychiatric disturbance;
Alcohol and benzodiazepine withdrawal states can be the underlying aetiology;
Among trauma and surgical patients pure hyperactivity is rare compared to mixed or hypoactive subtypes, making up about 1% of delirium presentations.
The hypoactive subtype is associated with the following:
Psychomotor slowing and somnolence – often misattributed to depression;
Can be difficult to recognise – patients with hypoactive delirium do not demand much attention from staff;
Easy to miss the cognitive changes with superficial interactions –such as asking patients only ‘yes/no’ questions;
Generally associated with worse outcomes and more long-term cognitive effects;
Most prevalent subtype (~88%) in surgical and trauma patients.
The mixed subtype is associated with the following:
Alternating states of hypoactivity and agitation;
Possibly resulting from the administration of sedating medications to a hyperactive delirious patient;
About 11% of delirium presentations in surgical and trauma patients.
Delirium is prevalent in all hospital settings and in the postoperative course ranges from 5 to 50% depending on the comorbidity and type of surgery. The incidence of delirium for the following surgeries is reported as:
Hip fracture (35–65%);
Coronary artery bypass grafting (37–52%);
Peripheral vascular surgery (30–48%);
Infrarenal abdominal aortic aneurysm repair (33–54%).
The risk of delirium increases with the degree of medical urgency. Emergency surgery is associated with much more delirium (both prior to and after surgery) compared with elective and outpatient surgeries. The highest risk ratings on the American Society Anesthesiology (ASA) scoring system are associated with a higher risk of postoperative delirium.
The development of delirium is usually multifactorial involving the interaction of a vulnerable patient with one or more perioperative insults. However, delirium can develop in a young healthy patient if the insult is significant enough. Prediction models have been developed in both cardiac and non-cardiac surgery to help identify those patients most at risk for postoperative delirium.
The most important factors include:
Advanced age; and
Baseline cognitive impairment.
Other additional factors include:
Impaired physical function;
Preoperative depressive symptoms;
Preoperative stroke or transient ischaemic attacks;
Abnormal laboratory values such as low albumin;
History of alcohol and other substance abuse;
Invasiveness and length of the surgery.
Other predisposing factors, such as atrial fibrillation and peripheral vascular disease, and precipitating factors, such as transfusion of aged red blood cells and the use of an intra-aortic balloon pump (IAPB), have been described in other observational studies. Risk factors for the development of delirium can be divided into predisposing and precipitating, and are presented in a timeline related to surgical course in Figure 47.1.
Figure 47.1 Delirium risks, measures and interventions in adult patients undergoing surgery.
Although the pathophysiology is not yet understood, neuroinflammation and neurotransmitter dysregulation have been implicated as possible mechanisms involved in delirium. More specifically, impaired cerebral oxidative metabolism resulting in increases in dopaminergic, noradrenergic, and glutamatergic activity and decreases in cholinergic activity in response to stress and inflammation of surgical procedures have been hypothesised to contribute to delirium.
Given that pre-existing cognitive impairment is the most common independent risk factor across studies, assessing for baseline cognitive impairment is beneficial for two reasons:
It identifies those patients at highest risk for the development of delirium.
It documents a baseline to which postoperative performance can be compared, aiding in the detection of delirium.
Standardised screening tests are the most efficient way to test cognition. Many brief tests are available. Casual conversation with a patient is not sufficient to assess for preoperative cognitive deficits. The American College of Surgeons’ National Surgical Quality Improvement Program suggests using the Mini-Cog as a cognitive screen of the preoperative geriatric surgical patient to establish a cognitive baseline and identify high risk patients. Other tools can also be used and are listed in Table 47.1. Preoperative functional status is also an independent risk factor for the development of delirium. Assessing the functional status of a patient by using a screening tool assessing activities of daily living and frailty can provide risk stratification for delirium and can also provide information about the expected course of recovery postoperatively.
|Cognitive screening tools||Delirium screening tools|
|Clinical Dementia Rating (CDR)||Bedside Confusion Scale|
|Clock drawing test||Clinical Assessment of Confusion (CAC)|
|Days of the week backwards||Confusion Assessment Method (CAM), short form|
|Digit cancellation test||Confusion Assessment Method-ICU version (CAM-ICU)*|
|Digit span – forward and backward||Confusion Rating Scale (CRS)|
|Mini-Cog||Delirium Observation Screening (DOS)|
|Months of the year backward||Delirium Symptom Interview (DSI)|
|Montreal Cognitive Assessment (MOCA)||Intensive Care Delirium Screening Checklist (ICDSC)*|
|Trailmaking Test A||Nursing Delirium Screening Scale (Nu-DESC)|
|Trailmaking Test B||Single Question in Delirium (SQiD)|
* Tools best used for mechanically ventilated, critically ill patients.
Adapted from AGS 2014.
Preoperative Screening for Comorbid Disease and Substance Use
Abnormal laboratory values, including glucose, sodium, potassium and albumin, which represent underlying disease or organ system dysfunction, are risk factors for delirium:
Hypoalbuminaemia appears to be of particular importance due to its association with fluid management, drug binding and malnutrition;
A blood urea nitrogen (BUN) to creatinine ratio of greater than or equal to 18 (index of dehydration).
Delirium is associated with preoperative alcohol, benzodiazepine and other substance abuse such as opioids, and prior stroke or transient ischaemic attack. In general, collection of historical information from the patient or collateral informant is sufficient and routine cerebral imaging is not required for complete assessment of delirium risk.
Depression has been identified as a risk factor for postoperative delirium in several studies and has been associated with incomplete recovery to independent functioning after surgery. The pathophysiology of this relationship has not been determined. There are many available tools that can be used to screen for depression. Assessment of depression may also help predict the patient’s motivation for recovery and can help in postoperative planning.
In addition to baseline patient vulnerability, perioperative insults contribute to the development of delirium and several interventions to reduce the risk have been described. These include pharmacological as well as procedural interventions. A recent meta-analysis indicated that several interventions might reduce postoperative delirium.
Light sedation versus deep sedation may reduce the prevalence of postoperative delirium, particularly in patients receiving spinal anaesthesia for hip replacements.
Although deeper levels of sedation are associated with increased delirium, use of an electroencephalographic (EEG) monitor to measure depth of sedation is not yet routinely recommended.
Type of general anaesthetic agent does not appear to affect the development of delirium but administering additional ketamine during anaesthesia induction may result in less postoperative delirium.
Regional anaesthesia, specifically femoral block and fascia iliaca block, for lower extremity procedures has been shown to reduce postoperative delirium in the elderly.
Additional areas of inquiry for delirium reduction and improved postoperative outcomes in cardiac surgery include personalised management of blood pressure within the zone of cerebral autoregulation, as predicted through near infrared spectroscopy (NIRS) and Doppler monitoring.
Medications probably contribute to the complex pathophysiology of delirium through opioid and gamma-aminobutyric acid (GABA) receptor agonism. Moderate quality evidence suggests that opioid medication use is associated with delirium in medical and surgical patients. It is important to weigh the risks of opioid use with the benefits of treating acute severe pain, particularly since uncontrolled pain is also cited as a risk factor for delirium. It is best to use the lowest effective dose for pain control.
Evidence also suggests that benzodiazepine medication use increases delirium in a mixed surgical group of patients. This was demonstrated with higher doses of benzodiazepines over a 24 hour period and with longer acting agents. Studies investigating the risk of delirium with antihistamine medications such as diphenhydramine noted a trend toward increased delirium as well.
Adequate postoperative analgesia has been associated with decreased delirium incidence. Increased levels of pain in the postoperative period have been independently associated with the development of delirium. Non-opioid medications should be considered when possible: gabapentin and non-steroidal anti-inflammatory drugs are associated in some studies with reduced incidence and severity of postoperative delirium.
Preoperative and postoperative sedation regimens may have an impact on incidence of delirium. Significantly longer time to recovery in the PACU with more cognitive dysfunction following anaesthesia has been demonstrated in 1062 patients less than 70 years of age, receiving a premedication of lorazepam (2.5 mg) compared to placebo. This suggests that such premedication is best avoided in older more vulnerable patients. Dexmedetomidine has been shown in some studies to be more effective than other sedatives (benzodiazepines and propofol) in preventing postoperative delirium, but the results have been mixed. Benzodiazepine use is independently associated with delirium in the cardiac ICU and the Society for Critical Care Medicine (SCCM) Guidelines suggest that sedation reduction and discontinuation should be the goal for the management of the critically ill patient.
Some studies suggest that the prophylactic use of typical and atypical antipsychotics can reduce postoperative delirium; however the trials have been small, heterogeneous and of variable quality. At least one study bears replication in larger trials: a randomised controlled trial of on-pump cardiac surgery patients with subsyndromal delirium upon immediate recovery following surgery, treated with risperidone versus placebo, demonstrated a significant decrease in incident postoperative delirium. While there is more research to be done, the recommendations arising from a recent meta-analysis suggest that there is no clear evidence for routine prophylactic use of antipsychotics for delirium prevention at this time.
Several other interventions have not demonstrated a reduction in the prevalence or duration of postoperative delirium. These include:
Intraoperative controlled hypotension;
Neuraxial versus general anaesthesia;
Long acting morphine postoperatively;
Postoperative sedation using alpha-2 adrenoreceptor;
Use of acetylcholinesterase inhibitors;
Use of anticonvulsants;
Use of histamine (H2) blockers.