Postoperative delirium: Addressing the rising scourge in healthcare
Abstract
The rising global life expectancy comes with challenges relating to the increasing elderly population, including delirium which those with background cognitive deficits are especially prone to. Advances in anaesthesia and surgery have contributed to a rapidly increasing number of surgeries in the elderly. Postoperative delirium with its worrisome prognosis is a common complication of surgery in the elderly; yet widely under-diagnosed, and under-treated. The use of validated tools to detect delirium, and preoperative cognitive impairment which is a core risk and prognostic factor is key to risk stratification, prevention and treatment. Management of postoperative delirium is anchored on prevention through optimization of modifiable risk factors, early detection using validated tools and care directed at reducing its severity and duration when it occurs. This is best achieved by individualized care through multi-component interventions involving the anaesthesiologist, surgeon, geriatrician, psychiatrist, physiotherapist, nursing services and the patient’s family. However, It is imperative to state that while the key principles of multi-component intervention apply broadly, the specific components of each regimen may vary widely. Despite existing international guidelines on the management of postoperative delirium, wide knowledge and practice gap is still prevalent.
Key words: Elderly; surgery; delirium; cognitive impairment; anaesthesia
Introduction
The United Nations population estimates indicate that by 2050, 16% of the global population will be 65years or older (with Japan and parts of Europe having as high as 40%); compared to 9% in 2019.1 This changing demography is a major social transformation with implications for healthcare. In reality, the demand for healthcare services by the elderly is disproportionately beyond the foregoing statistics, as represented by the 65% acute hospital admissions in England and Wales.2 Postoperative delirium (POD) is acute brain dysfunction that manifests as acute onset of fluctuating altered consciousness, inattention and disorganized thinking following surgery. Postoperative delirium is common in the elderly but widely under-diagnosed, while the diagnosed cases are poorly treated. Prevention through the optimization of modifiable risk factors is key in the management. The reported incidence of POD varies widely and could range from 3.6% to 45%3,4 depending on the sample patient population, method of observation and delirium instrument deployed for the assessment. The Incident rate of 1.5% has been reported following total joint replacement surgery based on retrospective review of routine nursing assessment chart,5 while 51% rate was reported among patients that underwent elective cardiac surgery who were assessed twice daily during the first five postoperative days using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU);6 lending credence to the influence of different methodologies on the quoted incidence rates in the literature. Owing to its fluctuating nature enhanced detection of delirium requires multiple daily assessments.7 However, with the near absence of literature on postoperative delirium in Africa how this scourge is being addressed in the region can only be imagined. Hence the purpose of this review is to acquaint perioperative practitioners with practice recommendations relating to this serious complication, and for which poor awareness and sub-optimal care have been near-universal.
Materials and methods:
Design: This is an unsystematic narrative review of several clinical and educational issues relating to postoperative delirium in the elderly, rather than a narrowly focused review of a research question typical of systematic reviews. It has been synthesized from previously published literature in order to illuminate the various aspects of clinical and knowledge gaps which inform the objective of this review. Consequently, no rating or grading of evidence was used.
Search Strategy: Online search was conducted on two databases; PUBMED and Cochrane Database of Systematic Reviews. The PUBMED database was searched using the term ….. ‘delirium elderly’ for the period spanning 1990 to December 2020; and 9117 articles were turned in; with the observed exponential increase in the articles over the period supposedly reflecting the surge in interest regarding delirium in the elderly within the research community. The search term ‘postoperative delirium studies in Cochrane evidence’ was used to explore the Cochrane Database of Systematic Reviews. Further hand searches of the references of retrieved literature were also conducted.
Eligibility criteria: Only studies of human subjects with available English language full texts, published in peer reviewed journals were considered, and retrieved. The study was limited to adult and elderly populations, as such studies considering delirium in children were excluded.
Consequences of POD
Whereas the episode of postoperative delirium is transient, it is a harbinger of wide-ranging consequences; including an independent association with higher morbidity and mortality rates in diverse surgical settings. The patients with POD have an increased length of stay in the hospital with attendant increase in healthcare costs,8,9 and they are more prone to postoperative complications.10,11 The development of POD presages a cascade of deterioration in physical and cognitive function, with increased deaths. Following POD return to the functional state prior to the surgery is less likely8,12 while both short-term and long-term mortality have been shown to be significantly higher.8,12,13 Patients who developed POD are at increased risk of fresh onset and more rapid deterioration of pre-existing cognitive impairment, with a progressive and irreversible decline in mental functions.14-16
Risk factors for POD
A constellation of preoperative, intraoperative, postoperative, patient and surgical factors have been implicated in the development of postoperative delirium, but among them advanced age and poor preoperative cognitive status rank high.17-20 Other associated factors include; poor functional status, co-morbidities, American society of anesthesiology (ASA) grade, polypharmacy, malnutrition, alcohol use, lower education, perioperative anaemia or blood transfusion, history of diabetes, perioperative use of anticholinergics, opioids and benzodiazepines, perioperative hypotension or hypoxia, perioperative dehydration or electrolyte imbalance, increased surgery duration, perioperative pain or hypoxia, night time surgeries, pain, sleep deprivation and urethral catheter in-situ. Patients admitted into the ICU after surgery are at increased risk of POD due to the patients’ isolation, strange environment, sleep disturbance, device attachments and lack of family support. Mechanical ventilation and the duration of ICU stay have equally been found to be risk factors for delirium.21
Prognostic indicators in POD
Postoperative delirium can be classified based on psychomotor features as; hyperactive, hypoactive, or mixed.22 The hypoactive form which carries the worst prognosis is less dramatic and most often unrecognized, yet is the most common.23,24 Formal assessment and recognition of motor subtypes of delirium in patients with POD aids in creating awareness, and relating better with the respective adverse event profile. Patients with severe delirium, or long-lasting delirium also have worse outcomes.25,26 Another determinant of prognosis in patients with POD is co-morbid dementia. Investigation of the impact of delirium superimposed on dementia among elderly Italian patients by Bellelli et al. found that the mortality in this subgroup was increased over two-fold compared to those with dementia alone, with delirium alone, or with neither dementia nor delirium;27 further highlighting the grave impact of pre-existing cognitive impairment, not only on delirium incidence but as a predictor of worse outcome in delirious patients. Similarly, a Brazilian study conducted to evaluate the combined effect of dementia and delirium on elderly individuals posted in-hospital mortality of 8% for patients without delirium or dementia, 12% for patients with dementia alone, 29% for patients with delirium alone, and 32% for ‘Delirium Superimposed on Dementia’ (DSD) patients.28 Other factors such as advanced age and low baseline functional status have also been implicated as indicators of poorer prognosis in elderly patients with delirium.29 Recent evidence also suggests that pharmacological intervention with antipsychotics alone, or in combination with benzodiazepines is associated with longer length of hospital stay, mortality and institutionalized care post-discharge.30
Preoperative cognitive assessment in older adults
Cognitive deficit is a frequent morbidity in older adults with a recent report indicating a prevalence rate of 33.3% in a community-based study conducted in Cameroon, Central Africa.31 A very high prevalence of cognitive impairment (51.9%) was observed among elderly patients during preoperative evaluation for elective cancer surgery in Romania, with a considerably higher figures being noted with respect to the very elderly segment of this population; reflecting increasing prevalence with age.32 Ageing is associated with a global decrease in physiologic reserve, including a decline in cognitive reserve related to chronic neuroinflammation “inflammaging”33,34 - a mechanism of age-related susceptibility to cognitive decline that has also found expression in animal model.35 Baseline cognitive deficit, frequently present in the elderly has consistently been shown to be a major risk factor for postoperative delirium. We had earlier been informed that delirium is associated with accelerated cognitive decline, long-term functional impairment and other poor outcomes. Other studies have revealed that baseline cognitive status independently predicts mortality following surgery in the elderly, even after adjustment for confounders such as age, ASA grade and functional status; and despite similarity in postoperative complication rates and hospital length of stay.36,37 These considerations provide sufficient ground for routine formal assessment of cognitive status during clinical assessment of elderly patients for anaesthesia and surgery. Current European and American guidelines on the preoperative evaluation of elderly patients have recognized this.38,39 While it is acknowledged that many among the battery of tools currently deployed for detecting cognitive impairment are inclined to cultural and educational bias the Mini Mental Status Exam (MMSE) is still adjudged the criterion-standard, and is most frequently used. This, in spite of its low points in sensitivity40 and very poor performance among elderly patients with poor educational and socio-economic background.41,42 However, the short blessed test (SBT)43 also termed the short Orientation-Memory-Concentration test which is a very brief, rapid and simpler tool without copyright protection has demonstrated comparable performance to the MMSE in detecting cognitive impairment in the elderly.44,45 Thus it presents an attractive alternative to the MMSE and can be administered by non-psychiatry-trained personnel and on less-educated, or visually impaired elderly patients. The challenge posed by the lack of suitable instrument with optimal performance in detecting cognitive impairment in resource-poor countries with limited health personnel and uneducated elderly patients is a major one, as these regions account for the majority of patients with cognitive impairment.46 This concern has also been raised recently by Magklara et al.47 For elderly patients in whom cognitive deficits have been detected current evidence does not support pharmacological interventions in the management.48 Instead, non-pharmacological interventions through lifestyle modification offer better prospects, and encompass physical, mental and social activities tailored to the needs and ability of the individual patient. Among these are physical exercises, mental exercises (cognitive training and rehabilitation), mobility training, sensory stimulation with visual and hearing aids, sleep hygiene, proper nutrition and hydration, smoking and alcohol control.49,50
Delirium detection tools
Psychiatrists’ evaluation based on the DSM-V Diagnostic and Statistical Manual for Mental Disorders is currently regarded as the gold standard method for diagnosing delirium.51 The inherent expertise and time requirements however limit its utility for the ubiquitous and fleeting nature of the scourge that is POD in the clinical setting. Consequently, brief and easy tools are preferred. Among the common tools deployed in detecting POD are; the Confusion Assessment Method (CAM), Confusion Assessment Method for Intensive Care Unit (CAM-ICU), Nursing Delirium Symptom Checklist (NuDESC), the Neelon and Champagne (NEECHAM) Confusion scale, the Delirium Observation Screening (DOS) scale, and the 4 A's test (4AT). However, a comparative study conducted by Neufeld et al. concluded that neither the CAM-ICU nor the NuDESC had satisfactory sensitivity, despite their high specificity for POD.52 The NEECHAM confusion scale and the DOS scale (like the CAM-ICU and the NuDESC), are popular tools for delirium detection among nurse-raters; but time consumption and difficulty of use make the former less attractive for routine clinical application.53,54 The 4AT is a brief screening tool recently developed for delirium screening in acute care, but which has been validated with good test properties for use in POD detection.55 The ‘Confusion assessment method’ (CAM)56 is regarded as the best offering among the bedside tools for detecting POD in the elderly; on account of its simplicity, speed, versatility , validity and reliability.57-59 While the CAM was designed for use by nonpsychiatry-trained physicians, the importance of training in its use has been emphasized in the work of Ryan et al.60
Clinical interventions
Non-pharmacological clinical interventions to prevent and treat POD have been receiving some modest attention and several of these have been successfully implemented in hospitalized patients. While many of these had been isolated uni-component therapies61,62 multi-component measures are preferred in view of the plurality of risk factors implicated in POD. The meta-analysis conducted by Hshieh et al. using 14 interventional studies to evaluate the effectiveness of multi-component non-pharmacologic delirium prevention interventions in reducing delirium attests to the effectiveness of this approach.63 Economic analysis of these prevention strategies has also proven that they are hugely cost-effectiveness in surgical,64 medical65 and ICU settings.66
Enhanced Recovery After Surgery (ERAS) program
Since its introduction for abdominal surgery patients over two decades ago, ERAS has been adapted to surgical specialties that have predominant elderly population such as arthroplasty and oncology. Conceived to enhance recovery, reduce morbidity and length of stay in surgical patients through optimization of perioperative care, ERAS has become increasingly attractive to both patients and care providers. Evaluation of elderly patients undergoing fast-track surgery program has revealed that POD could contribute significantly to increased length of stay in the ERAS program and mitigate against its objective.67 Hospitalization has several inherent attributes that predispose elderly patients to development of delirium.68,69 However there is substantial evidence suggesting that ERAS is associated with less POD than traditional care.70 This should not be a surprise though, since the multidisciplinary, multimodal and individualized patient optimization that is the sine qua non for delirium prevention is implicitly implemented in the enhanced recovery pathway of ERAS. Delirium prevention thus qualifies as one of the quality metrics for ERAS programs in the elderly. ERAS may have also presented a platform for implementing many of the preventive interventions for postoperative delirium. Prehabilitation with its three main pillars of physical, nutritional, and psychological optimization of patients, as distinct from traditional medical optimization of comorbid conditions has evidence base in improving surgical outcome, and has also found accommodation in the ERAS model.71
The Hospital Elder Life Program (HELP)
A model of the multi-component bundle therapy for preventing POD tagged “The Hospital Elder Life Program (HELP)” was structured over two decades ago by Inuoye et al.72 and has received wide acclaim. However it must be emphasized that the multidisciplinary team and the six risk factors targeted in the HELP (cognitive impairment, sleep deprivation, immobility, visual and hearing impairment, and dehydration) are by no means exhaustive and several modifications have been effectively adapted to suit diverse patient populations and settings73-80 (Table 1). Several other preoperative, intraoperative and postoperative risk factors are amenable to environmental and supportive modification by multidisciplinary teams through; medication review, nutrition review, correcting preoperative anaemia and electrolyte derangement, encouraging visits by family, music, avoiding prolonged fasting and unnecessary invasive procedures, optimal opioid-sparing pain management, encouraging daytime surgery, with provision of appropriate lighting and clock to facilitate orientation and sleep.
Anaesthesia perspectives
The anaesthesiologists implement a handful of interventions addressing the modifiable risk factors for POD. In view of the pre-eminence of cognitive deficits in predicting POD and outcome, including mortality in geriatric surgical patients, it is apposite that anaesthesiologists conduct mandatory cognitive assessment in this patient population in order to situate overall preoperative fitness and risk appreciation properly. This is in addition to the traditional American Society of Anesthesiologists (ASA) grading which considers the patients physical status only, but not cognitive status. Other measures include preoperative polypharmacy review, prevention of prolonged fasting and dehydration, perioperative hypoxia, hypothermia and hypotension, avoidance of anticholinergic drugs and dehydration, optimal perioperative pain management with opioid-sparing multimodal analgesia. An anaesthesia-based program implementing some of these multi-component interventions (early surgery, oxygen therapy, and control of peri-operative hypotension) has been credited with markedly reducing POD incidence, duration and severity.81 Titration of anaesthetics and cerebral monitoring of anaesthetic depth in surgical patients undergoing sedation or general anaesthesia facilitated by processed EEG have long been part of anaesthetic practice. Earlier comparative studies using Bispectral index (BIS) or auditory evoked potentials (AEP) monitoring had shown that titration of anaesthetics with these non-invasive devices led to reduced anaesthetic exposure with improved recovery profile and patient satisfaction, compared to routine clinical monitoring.82,83 Subsequently, interest in POD prevention through anaesthetic-sparing has festered, with several meta-analysis studies and Cochrane reviews attesting to POD reduction accruable to cerebral monitoring in patients undergoing general anaesthesia.84-86 However, its approval for this indication has so far remained contentious.87,88 Sedation is extensively used for procedures in the operating room and ICU to improve patient comfort and cooperation. Still, the role of cerebral monitoring in preventing delirium has been over-represented in patients undergoing general anaesthesia. Nevertheless, information from the few available studies on BIS monitoring of sedation suggests that it could also have a beneficial impact on reducing POD.89,90 Thus cerebral monitoring during anaesthesia and sedation could be considered a worthy option for avoiding unduly high exposure to anaesthetics in patients at high risk of delirium. Current guidelines do not recommend any technique of anaesthesia (general versus local anaesthesia), anaesthetic agent or pharmacological prophylaxis for the prevention of POD. While there is preponderance of evidence conferring superiority on regional techniques of anaesthesia over general anaesthesia in respect of deep vein thrombosis, blood loss and respiratory complications, among other outcomes; there is no such evidence to support a technique over the other regarding POD.91 However, large scale, multicentre randomized controlled trials are ongoing and may provide further insight regarding the impact of anaesthetic technique on POD.92 Similarly, the route of administering postoperative analgesia does not seem to impact the development of POD.93 Despite of the popularity of the neuroinflammation hypothesis of POD,94,95 meta-analysis of several works evaluating the role of anti-inflammatory adjuncts such as dexamethasone in preventing POD drew blank.96 While the use of pharmacological prophylaxis for the prevention of POD is not recommended by current guidelines, the recent meta-analysis by Ciu et al. attests to the better safety profile of dexmedetomidine regarding POD, compared to other drugs used for perioperative sedation.97 In the ICU setting where delirium incidence is highest dexmedetomidine-based sedation has displayed similar trends.98,99 Optimization of pain control using multimodal strategies such as acetaminophen, non-steroidal anti-inflammatory drugs (NSAIDs), pregabalin, gabapentin, dexmedetomidine, local infiltration anaesthesia and nerve blocks have also been credited with remarkable opioid-sparing benefits and satisfactory pain control that benefits patients at risk of POD.
From the foregoing, and his pivotal role in the multiple initiatives at enhancing safety and better surgical outcomes for the elderly such as prehabilitation100 and ERAS,101 the anaesthesiologist has substantial potential in influencing the course of POD in older adults.
Pharmacological prophylaxis
Pharmacological prophylaxis per se is currently not recommended for POD prevention. However, following the reported efficacy of dexmedetomidine prophylaxis in several studies.102-104 interest in this drug is increasing despite some contradictions and queries regarding the evidence for it.105,106 The use on low dose of intraoperative ketamine to prevent POD had gained some attention, but a recent meta-analysis which utilized 6 RCTs did not yield convincing evidence.107 Furthermore, the Prevention of Delirium and Complications Associated with Surgical Treatments [PODCAST] study revealed that prophylactic administration of ketamine was not superior to normal saline placebo in preventing POD, rather it was associated with greater incidence of postoperative hallucinations and nightmares.108 Currently, available evidence does not recommend the use of antipsychotics for the prevention or treatment of delirium.109 Comparative investigation of prophylactic use of haloperidol versus normal saline placebo among intensive care unit (ICU) patients with high delirium risk had also found no significant difference between the two regarding the incidence of delirium, and other outcomes associated with it.110 In the systematic review and meta-analysis conducted by Igwe et al. to evaluate pharmacological interventions for reducing POD in elderly patients it was also concluded that the use of haloperidol for this indication was no better than a placebo.111 The risk of harm from antipsychotics is substantial in the elderly despite the questionable benefit of the therapy. Their use has also been implicated as prognostic of worse outcome in delirious older adults. Antipsychotics and other psychotropic medications have instead been implicated with accelerated functional and cognitive decline.112
Delirium treatment
The goal of treatment in patients who have developed POD is to reduce its severity and duration, by identifying and addressing the contributing factors. Non-pharmacological multi-component interventions are equally applicable in the treatment of delirious patients, as they are with the prevention; and have recorded some success.113 Pharmacological intervention may only be indicated to manage delirious patients with severe agitation who are at risk of harming themselves or others. Antipsychotics even for this indication have been regarded as mere chemical restraints by Inuoye et al.114 They may also be considered when delirium persists despite the implementation of all relevant non-pharmacological measures. In such instances titrated low-dose antipsychotics are recommended. Haloperidol is often preferred despite its association with QT interval elongation, but chlorpromazine and atypical antipsychotics such as risperidone, clozapine or olanzapine may be used. Chlorpromazine and atypical antipsychotics are less likely than haloperidol to cause extrapyramidal motor effects. Both chlorpromazine and clozapine have marked anticholinergic effects, with sedation and orthostatic hypotension; and may lead to falls, deep vein thrombosis episodes, constipation, and urinary retention in elderly patients with prostatic hypertrophy. Therapy with clozapine also carries greater risk of agranulocytosis, myocarditis and seizures. Considerations for these side effects and safety in the elderly tend to determine the choice of antipsychotic agent more than efficacy, as most have good efficacy in controlling agitated delirious patients. These adverse effects of antipsychotics in the elderly compel their use at minimal effective doses for brief periods and wide expert opinion recommends tapering off of antipsychotics as soon as the delirium resolves. Despite the caveat on the use of antipsychotics there are still disconcerting reports of their overwhelming use in most delirious geriatric patients, irrespective of the motor subtype,115,116 and in preference to non-pharmacological multi-component interventions. Similar abuse has also been observed with regard to the continuation of antipsychotics at discharge, in patients who had commenced it in the ICU for treatment of delirium.117 Similarly, misuse of benzodiazepines which are known risk factors for the development of POD is widespread in the treatment of POD. This is unfortunate, as they may only be indicated for managing agitation associated with drug withdrawal such as benzodiazepines, and alcohol. The suboptimal care of the elderly patients regarding POD is not surprising with the pervading background of knowledge deficit regarding the scourge.118,119 This, in spite of recent effort at establishing international guidelines on POD.120–122
Limitations
This review is short on elaborating the various pathophysiologic mechanisms that have been hypothesized for the evolution of POD. This was actually intended, in order to keep within the declared objective of the study. It would further be acknowledged that owing to the lack of explicit benchmark for the selection of the source articles used for the synthesis, subjectivity and bias may not be entirely ruled out. In the same vein, the review unlike a systematic review had no focused research question being interrogated, and applied no explicit methodology in the synthesis. These shortcomings are necessarily inherent in this type of study design and guided by the stated objective.
Conclusion
Postoperative delirium is a common complication that exerts enormous burden on patients, their families, and healthcare resources. Clinicians are likely to encounter delirium frequently in elderly surgical patients, and its management should be considered in all stages of the surgical care pathway. Addressing the knowledge and practice gap regarding multi-component preventive and treatment interventions is crucial, while prompt detection of delirium using validated tools is key to changing the narrative of poor care provided to the elderly. The SBT and CAM are valid and reliable easy-to-use tools that every physician irrespective of background can use during routine assessment of elderly patients. Anaesthesia-based multi-component interventions have great potential among the broad options in the prevention and treatment of POD in surgical patients.
References