Tool
Clinical application
Additional information
Confusion Assessment Method (CAM)
Sensitivity: 94–100%
Specificity: 90–95%
General use in at-risk patients
CAM Long: comprehensive questionnaire for nonpsychiatrist clinicians
CAM Short: Assesses only the first four components of the CAM Long, used by clinicians and nursing
Can have a false-positive rate of up to 10%
Confusion Assessment Method for the Intensive Care Unit (CAM-ICU)
Sensitivity: 95–100%
Specificity: 89–93%
[17]
Adapted for quick administration by nurses and clinicians in the ICU to allow delirium assessment in critically ill patients with potential communication difficulties (i.e., mechanical ventilation, psychoactive medication, orogastric tubes)
Questionable effectiveness in demented patients
Delirium symptom interview (DSI)
Sensitivity: 90%
Specificity: 80%
[18]
General use in at-risk patients, administered by clinicians and nurses
Information is gathered only from the patient. Acute onset and possible etiology are not part of the assessment
Nursing Delirium Screening Scale (Nu-DESC)
Sensitivity: 85%
Specificity: 86%
[19]
General use in at-risk patients, used by nurses
Assesses psychomotor retardation (not agitation), 13% false-positive rate
Intensive Care Delirium Screening Checklist (ICDSC)
Sensitivity: 99%
Specificity: 64%
[20]
Used for critically ill patients, used by clinicians and nurses
Does not focus on cognitive tasks
Perhaps the most well-known delirium test is the Confusion Assessment Method (CAM). Inouye et al. developed a bedside assessment for delirium that could be administered quickly by nonpsychiatrists. The CAM exists as a diagnostic questionnaire (CAM Long) and a shortened screening algorithm (CAM Short). The CAM Long identifies acute onset and fluctuating course, inattention, disorganized thinking, altered level of consciousness, disorientation, memory impairment, perceptual disturbance, abnormal psychomotor activity, and altered sleep-wake cycle. The CAM Short assesses only the first four components of the CAM Long, requiring presence of acute onset or fluctuating course and inattention, plus either disorganized thinking or altered level of consciousness for delirium identification. The CAM is validated as a sensitive, specific, reliable, and easy to use tool for delirium identification [13, 16]. Similar to the CAM Short, the CAM-ICU is a four-part assessment modified for ICU patients with difficulty communicating (on mechanical ventilation, presence of orogastric tubes, psychoactive medication, etc.) [17]. Additional validated delirium screening tools include the delirium symptom interview (DSI), NEECHAM Confusion Scale, Intensive Care Delirium Screening Checklist (ICDSC), and Nu-DESC (Nursing Delirium Screening Scale) (Table 30.1).
The DSM V diagnosis of delirium requires that symptoms develop quickly (typically hours to days) relative to baseline and classically fluctuate in severity over time. Hallmark features of delirium are (1) altered awareness (e.g., reduced orientation to the environment), (2) additional cognitive disturbances (e.g., memory deficit, altered language, visuospatial ability, or perception), and (3) inattention (e.g., reduced ability to sustain, shift, direct, and focus attention). To diagnose delirium, these symptoms must not be better explained by another preexisting neurocognitive disorder (e.g., dementia). Likewise, symptoms must not be a direct physiological consequence of another medical condition. Examples of conditions which contribute to acute confusion and may mimic postoperative delirium are listed in Table 30.2. Arguably hypoxia and ischemia are the two most important of these ten to rule out in a timely manner. In addition to evaluation of DSM V criteria (ICD-10, CAM), tools like the Delirium Rating Scale Revised 1998 (DRS-98R) and Memorial Delirium Assessment Scale (MDAS) can also be used for delirium diagnosis [13].
Table 30.2
Differential diagnosis for postoperative delirium
Emergence from anesthesia drugs (polypharmacy, withdraw, anticholinergics, antihistamines, barbiturates, and benzodiazepines) |
Endocrine and metabolic disturbances (hypoglycemia, hypothyroidism, hyponatremia, hyperammonemia, etc.) |
Mental disorders (dementia, depression, and anxiety) |
Hypoxia and ventilation disturbances |
Infection |
Sensory deprivation or overload |
Ischemia (TIA, CVA) |
Intracranial neoplasm |
Seizure disorder (postictal state) |
Classification of Delirium Subtypes
PD is classified by duration and symptomatology. PD can be either acute or persistent according to its duration and time course (hours to days versus weeks to months, respectively) and can be further categorized as either hyperactive, hypoactive, or mixed based upon motor activity. Hyperactive delirium is characterized by increased psychomotor activity such as mood lability, agitation, and/or refusal to cooperate with medical care. Hypoactive delirium is characterized by decreased psychomotor activity such as sluggishness or lethargy approaching stupor. In mixed delirium, individuals may have features of both hyper- and hypoactive delirium at different points in their clinical course [21]. Longitudinal assessment of activity subtypes (no subtype, hyperactive throughout, hypoactive throughout, mixed throughout, variable over course) has shown that the majority of delirious patients are stable in their course of delirium, with less than 40% demonstrating subtype variability. The hypoactive subtype is associated with the worst overall prognosis [22].
Diagnosing POCD
The range of abilities associated with cognition is broad, including learning and memory, verbal abilities, perception, attention, executive functions, and abstract thinking [2], and there has been a lack of uniformity in the terminology and methodology used to report on POCD in the literature [23]. Self-reporting of cognitive problems correlates poorly with objective testing, so the diagnosis of POCD requires pre- and postoperative cognitive testing [24]. Generally a group of neurocognitive tests have been combined and administered as a battery, with verbal learning and working memory, episodic memory, processing speed, and set shifting emerging as the most sensitive cognitive testing domains. Some examples of these tests include the logical memory test, CERAD word list memory, Boston Naming Test, category fluency test, digit span test, trail-making test, and digit symbol substitution [2]. A change score using baseline cognition and postoperative performance may be used to identify POCD. POCD may be defined on the basis of significant decline on greater than or equal to two tests or a more subtle decline across the neuropsychological test battery [25].
In order to compare across studies and fit into the greater literature regarding cognitive disorders, a multidisciplinary work group including anesthesiologists, surgeons, geriatricians, neuropsychologists, neurologists, and psychiatrists has been assembled to provide consensus definitions and nomenclature for POCD by late 2016. The subjective component is not sufficient to define POCD; however, it may have some weight in future POCD diagnostic criteria [26].
Pathophysiology of Postoperative Cognitive Impairment
The pathophysiology of postoperative cognitive impairment is being investigated with basic, translational, and clinical research. There is emerging evidence that delirium may be the prodrome of long-term cognitive impairment [27]. Proposed mechanisms for PD and POCD have significant overlap, and current research implicates general health status, inflammation, oxidative stress, and disruption of the circadian clock. Ultimately altered neurotransmission and loss of cellular and regional communication within the CNS are likely responsible for the functional disturbances characteristic of PD and POCD [28].
Systemic inflammation manifesting with a cascade of pro-inflammatory events results from surgical trauma and/or infection. Baseline levels of circulating inflammatory mediators, including cytokines and acute phase proteins, increase several-fold with aging. Likewise, microglia in the aging CNS assume a “primed” phenotype, resulting in an exaggerated and pathologic response to stress or an immune challenge [29, 30]. Markers of immune activation, elevated levels of C-reactive protein (CRP), interleukin (IL)-6, IL-1RA, IL-10, IL-8, neopterin, S-100 beta (S-100β), tumor necrosis factor-α (TNF-α), and cortisol have been reported in delirious patients and can be measured in a variety of tissues including plasma, urine, and CSF [29–31]. During a pro-inflammatory state, development of fever, sickness behavior, and activation of the hypothalamic-pituitary-adrenal (HPA) axis occur [32]. Immune activation ultimately results in CNS dysfunction secondary to altered blood-brain barrier, oxidative stress, and some degree of compromised neuronal and glial function [30].
Outside of the immune system, oxidative stress can occur with any condition where the body’s ability to metabolize reactive oxygen species is overwhelmed. Fundamentally, reactive oxygen species are associated with energy imbalances and local ischemia that leads to excitotoxicity, apoptosis, and escalation of local inflammation. Poor tissue oxygenation has been associated with PD, and an intervention to remedy cerebral oxygen desaturations during major cardiac surgery resulted in decreased PD occurrence [33]. The use of near-infrared spectroscopy (NIRS) in other elderly surgical populations has shown significant position-associated changes in cerebral oxygenation (prone versus supine), but the impact of this on PD and POCD needs further study [34]. Preoperative identification of patients with regional cerebral desaturation prior to noncardiac surgery may identify those at high risk for PD, though trials showing benefit of intraoperative cerebral oxygenation monitoring are lacking [35, 36].
Disruption in circadian rhythms has been reported after minor and major surgery, which affects postoperative sleep quality and recovery. Sleep deprivation leads to decreased cognitive function and may predispose to postoperative delirium [37]; therefore, pharmacologic and non-pharmacologic maintenance of normal circadian rhythms may decrease or ameliorate PD. Melatonin is one drug investigated because of its sleep-wake cycle regulatory effects and also because of its anti-inflammatory and antioxidant properties [30]. Disruption in the endogenous rhythm of plasma melatonin and excretion of the urine metabolite on the first postoperative day have correlated with the duration of major surgery [38], and low postoperative melatonin has been reported in patients who develop PD [39]. However, a recent trial failed to demonstrate benefit of postoperative melatonin supplementation for prevention of PD in ICU patients after major elective surgery, and furthermore the rates of delirium subtypes (hypoactive versus hyperactive) were not altered by melatonin administration [40].
Decreased acetylcholine availability, excess of dopamine, norepinephrine, and/or glutamate release, and variable alterations in serotonin, histamine, and/or g-aminobutyric acid (GABA) may be implicated in PD. Neuronal network connectivity and receptor availability and function may also be implicated [29, 30]. Acetylcholine neurotransmission is vulnerable to dysfunction during immunologic stress and periods of altered synthesis and metabolism (e.g., surgery, ischemia, dehydration, severe illness). Exposure to anesthetics can alter cholinergic neurotransmission [41]. Based upon the pathophysiology of Alzheimer’s disease, a cholinergic mechanism may contribute to increased risk of postoperative cognitive problems in patients with preexisting dementia [42, 43] (see Chap. 10).
Surgical stress, inflammation, medications, and altered perioperative hormonal regulation may play a role in both PD and POCD. More research is required to elucidate various mechanisms, which could vary with severity and type of cognitive compromise postoperatively.
Risk Factors for Postoperative Cognitive Impairment
Risk factors associated with the development of cognitive problems after surgery can be categorized as preoperative, intraoperative, or postoperative in accordance with the point at which they are first introduced during a patient’s perioperative care. These factors are also often divided into predisposing factors (i.e., those factors that are present at baseline) and precipitating factors (i.e., those factors that occur during the patient’s clinical course). Many predisposing factors, such as demographics and health history, are not modifiable, but other factors may improve with treatment or intervention.
Preoperative Factors
Certain unifying themes relate a number of reported preoperative risk factors: demographics, decreased “cognitive reserve,” burden of illness, use of certain substances/medications, psychosocial factors, and poor functional status. Advanced age, previous history of delirium, depression, multiple comorbidities, alcohol abuse, and preoperative ASA score (an assessment of systemic impact of comorbid disease) are the most consistently reported (see Table 30.3).
Table 30.3
Preoperative risk factors for PD
Risk factor | Study | Population |
|---|---|---|
Advanced age | Katznelson et al. [44] | Cardiac surgery patients |
Krzych et al. [45] | Cardiac surgery patients | |
Norkiene et al. [46] | Cardiac surgery patients (CABG) | |
Gao et al. [47] | Spinal surgery patients | |
Böhner et al. [48] | Vascular surgery patients | |
Fineberg et al. [49] | Spinal surgery patients (lumbar) | |
Ushida et al. [50] | Spinal surgery patients (cervical) | |
Miyazaki et al. [51] | Cardiac surgery (CABG) | |
Smulter et al. [52] | Cardiac surgery | |
History of stroke, TIA, or dementia | Shah et al. [53] | Major head and neck cancer surgery |
Subjective reporting of memory complaints | Veliz-Reissmüller et al. [55] | Cardiac surgery (elective) |
MMSE score | Kazmierski et al. [56] | Cardiac surgery |
Rudolph et al. [57] | Cardiac surgery | |
Saczynski et al. [95] | Cardiac surgery | |
Osse et al. [58] | Cardiac surgery | |
Veliz-Reissmüller et al. [55] | Elective cardiac surgery | |
Schoen et al. [74] | Cardiac surgery | |
Cognitive impairment per IQCODE-SF | Juliebø et al. [59] | Hip fracture repair surgery |
Preexisting cognitive impairment | Litaker et al. [54] | Major elective surgery |
Kazmierski et al., the use of DSM-IV and ICD-10 criteria and diagnostic scales for delirium among cardiac surgery patients: results from the IPDACS study [56] | Cardiac surgery patients | |
Shah et al. [53] | Major head and neck cancer surgery | |
Freter et al. [60] | Orthopedic surgery (elective) | |
Greene et al. [61] | Major, elective noncardiac surgery | |
Böhner et al. [48] | Vascular surgery | |
History of delirium | Litaker et al. [54] | Major elective surgery |
Poor sleep/sleep disruption | Leung et al. [37] | Major noncardiac surgery |
Preexisting diabetes | Kazmierski et al. [56] | Cardiac surgery |
Smulter et al. [52] | Cardiac surgery | |
Peripheral artery disease | Kazmierski et al. [56] | Cardiac surgery |
Otomo et al. [63] | Cardiac surgery (CABG) | |
Cerebrovascular disease | Kazmierski et al. [56] | Cardiac surgery |
Loponen et al. [64] | Cardiac surgery (CABG) | |
Atrial fibrillation | Bucerius et al. [65] | Cardiac surgery |
Miyazaki et al. [51] | Cardiac surgery (CABG) | |
Heart failure | Loponen et al. [64] | Cardiac surgery (CABG) |
Katznelson et al. [44] | Cardiac surgery | |
Obstructive sleep apnea | Flink et al. [66] | Knee replacement surgery |
Renal failure | Sasajima et al. [67] | Arteriosclerosis obliterans with lower limb ischemia patients undergoing bypass surgery |
Carotid stenosis of 50% or greater | Miyazaki et al. [51] | Cardiac surgery patients |
Atherosclerosis in the ascending aorta | Otomo et al. [63] | Cardiac surgery patients |
Increased number of medical comorbidities, often measured by the Charlson comorbidity index (CCI) | Robinson et al. [68] | Noncardiac, non-neurological major surgery requiring post-op ICU |
Guenther et al. [69] | Cardiac surgery | |
Tan et al. [70] | Cardiac surgery | |
Pol et al. [71] | Vascular surgery | |
Lee et al. [72] | Hip fracture repair | |
Higher preoperative pain scores | Smulter et al. [52] | Cardiac surgery |
Tan et al. [70] | Cardiac surgery | |
Behrends et al. [73] | Noncardiac major surgery | |
Lower regional oxygen saturation levels in the brain | Schoen et al. [74] | Cardiac surgery |
Morimoto et al. [35] | Abdominal surgery | |
Depression (presenting with ongoing depressive episode) | Kazmierski et al. [56] | Cardiac surgery |
Depression (presenting with depressive symptoms) | Böhner et al. [48] | Vascular surgery |
Leung et al. [75] | Noncardiac elective surgery | |
History of depression | Stransky et al. [76] | Cardiac surgery |
Alcohol use | Litaker et al. [54] | Major elective surgery |
Shah et al. [53] | Major head and neck cancer surgery | |
Patti et al. [77] | Colorectal surgery for carcinoma | |
Drug abuse | Fineberg et al. [49] | Spine surgery (lumbar) |
Smoking history | Benoit et al. [78] | Abdominal aortic aneurysm repair surgery |
Miyazaki et al. [51] | Cardiac surgery (CABG) | |
Decreased functional capacity/preoperative frailty | Juliebø et al. [59]. | Hip fracture repair surgery |
Pol et al. [71] | Vascular surgery | |
Brown et al. [82] | Cardiac surgery patients | |
Increased ADL dependence/reduction in ADLs | Leung et al. [83] | Noncardiac surgery |
Hattori et al. [84] | Vascular, orthopedic, and GI surgery | |
Poor preoperative nutritional status | Ganai et al. [85] | Abdominal surgery |
Tei et al. [86] | Colorectal cancer surgery | |
Dehydration | Harasawa & Mizuno [87] | Cerebrovascular surgery |
Fluid fasting | Radtke et al. [88] | Surgery |
Low BMI | Lee et al. [72] | Hip fracture repair surgery |
Juliebø et al. [59] | Hip fracture repair surgery | |
Benzodiazepine use | Do et al. [79] | Orthopedic surgery |
Psychoactive medications | Benoit et al. [78] | Abdominal aortic aneurysm repair surgery |
Polypharmacy | Goldenberg et al. [93] | Hip fracture repair surgery |
McAlpine et al. [94] | Gynecologic malignancy surgery |
It has been suggested that the risk for PD in adults increases with each increasing year of life, and as a threshold, after 60 years of age, surgical patients are more likely to develop PD [44–52]. Decreased preoperative “cognitive reserve” and/or previous neurological insult is a major risk factor for PD as indicated by history of stroke, transient ischemic attack (TIA), dementia [53], delirium [54], subjective reporting of memory complaints [55], or performance below a pre-identified standard reference score on tests such as the MMSE [56, 58] or the Informant Questionnaire on Cognitive Decline in the Elderly Short Form (IQCODE-SF) [48, 53, 54, 59–61]. Despite the strong connection between preoperative cognitive deficits and risk of postoperative cognitive impairment, genetic markers of Alzheimer’s disease have not been predictive of PD or POCD risk [62].
Burden of illness, as indicated by presence of various comorbidities such as diabetes [52, 56], peripheral artery disease (PAD) [56, 63], cerebrovascular disease (CVD) [56, 64], atrial fibrillation [65, 51], heart failure [44, 64], obstructive sleep apnea [66], and renal failure [67], has been associated with an increased risk of PD development. Research has also shown certain preoperative vascular factors, including preoperative carotid stenosis of 50% or greater [51] and atherosclerosis in the ascending aorta [63], to be significant predictors of PD in the cardiac surgery population. In general, a greater number of medical comorbidities, e.g., a higher Charlson comorbidity index (CCI), is widely recognized as a PD risk factor [68–72]. Higher preoperative pain scores have been associated with increased likelihood of developing PD [52, 70, 73], as have lower baseline regional oxygen saturation levels in the brain [35, 74].
Psychosocial factors also appear to play a role in development of postoperative cognitive deficits. Depression has been demonstrated as a risk factor for PD, whether the patient is presenting with an ongoing depressive episode [56], depressive symptoms [48, 75], or a history of depression [76]. Alcohol use has been associated with risk of PD [53, 54, 77], as well as drug abuse [49] and a history of smoking [51, 78]. One study indicated that patients unsatisfied with their level of social support were more likely to develop PD [79]. It has also been shown that patients with a greater amount of dispositional optimism (a behavior trait characterized by the tendency to react to situations with positive outcome expectations) are less likely to develop PD [80].
Decreased functional capacity and preoperative frailty are risk factors for PD [59, 71, 81]. A recent study in cardiac surgery patients over the age of 55 years reported that the prevalence of frailty was approximately 31%, and frail patients had significantly increased risk of PD compared to non-frail patients [82]. Preoperatively, increased dependence with respect to performing activities of daily living (ADLs) [83] and lower overall quality of life increase risk of developing PD [84]. Poor preoperative nutritional status [85, 86], dehydration [87] and fluid fasting [88], and low BMI have all been associated with PD [59, 72].
Avoidance of polypharmacy and appropriate medication use in elderly patients may decrease the incidence of PD. The AGS Beers Criteria List medications have been deemed inappropriate for geriatric patients for a variety of reasons, some of which are associated with cognitive issues [89, 90]. Beers Criteria medications which may be commonly administered to surgical patients include benzodiazepines, nonsteroidal anti-inflammatories (NSAIDS), antihypertensives, and sliding scale insulin. Anticholinergic medications are another example of Beers Criteria medications which are commonly used for their antihistamine, antispasmodic, and antiemetic properties [91, 92]. Likewise, prescribers should refrain from administration of corticosteroids and meperidine due to increased risk of PD [13, 90]. Polypharmacy is associated with PD, demonstrating importance of assessing a patient’s medication exposure globally, in addition to avoidance of specific medications [93, 94] (see Chap. 21).
Many risk factors for PD have also been identified as risk factors for POCD, including advancing age, preexisting cognitive impairment (PD, MCI, dementia), diminished functional status, multi-morbidity, low level of education, history of alcohol abuse, coronary artery bypass grafting (CABG) surgery, and exposure to psychoactive medications [5, 43, 95–99].
Intraoperative Factors
Characteristics of intraoperative course and patient management contributing to risk for postoperative cognitive impairment include surgical variables, medication-specific risks, and hemodynamic stability (Table 30.4).
Table 30.4
Intraoperative risk factors for PD
Risk factor | Study | Population |
|---|---|---|
Emergency surgery | Krzych et al. [45] | Cardiac surgery |
Kalisvaart et al. [100] | Hip surgery | |
Koebrugge et al. [101] | Endovascular aortoiliac surgery | |
Longer duration of surgery | Shah et al. [53] | Major head and neck cancer surgery |
Norkienė et al. [102] | Cardiac surgery | |
Lee et al. [72] | Hip fracture repair surgery | |
Invasive surgery | Fineberg et al. [49] | Spine surgery (lumbar) |
Koebrugge et al. [101] | Endovascular aortoiliac surgery | |
Salata et al. [103] | Aortic aneurysm repair surgery | |
Hudetz et al. [104] | Cardiac surgery | |
Osse et al. [58] | Cardiac surgery | |
Fentanyl use | Radtke et al. [88] | Surgery |
Andrejaitiene & Sirvinskas [110] | Cardiac surgery | |
Burkhart et al. [111] | Cardiac surgery | |
Midazolam use | Do et al. [79] | Orthopedic surgery |
Greater intraoperative volume loads | Smulter et al. [52] | Cardiac surgery |
Low intraoperative body temperature | Detroyer et al. [119] | Cardiac surgery |
Blood loss
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|