Introduction
Evidence suggests that increased age, low education, poor preoperative cognitive performance, perioperative complications, prolonged anesthesia, and diminished general health place individuals at greater risk for cognitive decline after surgery (1–4). But what constitutes improvement after surgery (postoperative cognitive improvement; POCI)? Although less is known regarding this topic, this chapter will highlight key factors indicated in prior studies and will summarize what may constitute cognitive improvements after surgery and anesthesia.
Findings specific to epilepsy, a disease model with substantial research on pre- to postoperative outcomes, will be reviewed. A general working definition of postoperative cognitive improvement (POCI) is presented and examples from a number of surgical procedures are discussed. Next, Engel’s biopsychosocial mode of health (5) is briefly introduced, and proposed as an appropriate model to conceptualize POCI. To illustrate, key factors that protect against perioperative neurocognitive disorders (PND) are identified, and briefly summarized within each domain of the biopsychosocial framework. Importantly, the list is not exhaustive, rather the goal is to highlight factors that have been indicated as important, and illuminate areas warranting further investigation.
Epilepsy: A Model for Pre- to Postoperative Cognitive Changes
Pre- and postoperative neuropsychological assessment is an integral component of clinical care for patients considering epilepsy surgery. Preoperatively, neuropsychological findings help to establish a patient’s decision-making capacity, demarcate functional brain centers to guide surgery, and, when integrated with other findings, help to identify the epileptic focus (6,7). Initially, neuropsychological findings had limited clinical utility in predicting a patient’s unique risk for postoperative decline. Normative data available at the time failed to account for practice effects and regression to the mean (8,9). Subsequently, reliable change index (RCI) and standardized regression-based change (SRB) were calculated to account for these limitations, and are now routinely reported in research (10,11). This expanded the utility of neuropsychological findings, which are now used to inform patient selection (12), facilitate decision-making for patients and families, and guide postoperative interventions (13).
Incorporating RCI and SRB into epilepsy research advanced our understanding about the factors that optimize postoperative outcomes. For example, younger age at the time of temporal lobe resection predicts long-term global cognitive functioning 2 to 10 years after surgery (14). This is not surprising, as aging is associated with physical changes in the brain (e.g., reduced brain volume, increased white matter lesions; 15,16) and cognitive functioning (e.g., fluid intelligence, episodic memory; 17). Surgery type is also a key characteristic. For example, patients undergoing right-temporal lobe resection and anterior temporal lobectomy are less likely to exhibit reductions in verbal memory and expressive language than their counterparts underdoing left-temporal lobe resection and left-anterior temporal lobectomy (10,18,19). Those findings are also expected, given that left-sided surgeries alter cognitive abilities known to be mediated by the left hemisphere (e.g., language, information consolidation).
Cognitive factors also predict postoperative outcomes after epilepsy surgery. Evidence indicates that baseline cognitive functioning explains memory and language scores up to 10 years after surgery, above and beyond seizure characteristics (14). Similarly, memory performance in the hemisphere contralateral to the epileptic focus during Wada testing (intracarotid sodium amobarbital procedure) predicts postoperative cognitive functioning (14,20).
The field of anesthesia can learn a great deal from looking at the history of epilepsy and neurosurgery in order to optimize postsurgical outcomes. Systematically assessing pre- to postcognitive changes via RCI and SRB may improve decision-making for patients and families and increase awareness about the potential cognitive risks associated with surgery. Progress in this area may facilitate clinical practice. That is, clinicians will be better prepared to identify factors needing intervention before, during, and after surgery.
Postoperative Cognitive Improvement: A Working Definition
Postoperative cognitive improvement (POCI) can be conceptualized as an increment in cognitive abilities following surgery that is statistically significant and not attributed to practice effect or normal variability (i.e., regression to the mean). Quantifying POCI is challenging. There is little consensus as to which cognitive domains should be evaluated pre- and postoperatively. Additionally, there are logistical and psychometric considerations when trying to reliably assess cognitive change within fast-paced environments. Nevertheless, some evidence indicates that cognitive increments can be achieved following surgery. For example, Liimatainen et al. (21) explored cognitive functioning in adults undergoing aortic valve replacement (AVR). They found significant improvement in inhibitory control and processing speed 3 months after surgery. Similarly, Knipp et al. (22) evaluated cognitive abilities in adults undergoing valve surgery and coronary artery bypass grafting (CABG) at 3 months, and 3–4 years after surgery. At 3 months, patients who underwent valve surgery exhibited improvements in simple attention, complex reasoning, and visuospatial abilities with respect to baseline functioning. Overall, improvements in simple attention and visuospatial abilities were sustained 3–4 years after valve surgery (22).
In the context of bariatric surgery, robust cognitive gains are often observed postoperatively. In particular, improvements in memory, executive functioning, and language have been reported one year after surgery, and maintained for up to 3 years (23). Similarly, Handley et al. (24) found that rapid weight loss after bariatric surgery is associated with improvements in memory, executive functioning, and cognitive control.
While the findings are still inconclusive, cognitive improvements after kidney transplantation have been found in some groups. Kaya et al. (25) reported significant increments in attention, executive functioning, processing speed, memory, and language 6 to 12 months after renal transplantation, with respect to baseline abilities. In a systematic review, Joshee et al. (26) found that improvements in general cognitive status, psychomotor speed, spatial reasoning, verbal memory, and visual memory are consistently reported in the literature after kidney transplantation.
Following orthopedic surgery, improved postoperative outcomes have been reported among adults who engaged in prerehabilitation prior to surgery (27,28). For example, Topp et al. (29) found that older adults who participated in a total-body prehabilitation program prior to undergoing knee replacement surgery reported increased independence, improved physical functioning 4 months after surgery, and reduced pain when compared to their counterparts who did not participate in the training.
POCI after participation in prerehabilitation programs has been less systematically assessed. Nevertheless, available case reports are promising. Carli et al. (30) described an 88-year-old patient who underwent robotic-assisted total abdominal hysterectomy. She was at high risk for postoperative delirium and completed a 3 week, home-based, prerehabilitation program focused on aerobic exercise, strength training, and nutrition. Postoperatively, the patient did not experience delirium. Moreover, she exhibited improved cognitive performance as well as increased functional capacity that was sustained 8 weeks after surgery (30). While such case reports are encouraging, more research in this area is needed in order to understand whether prerehabilitation programs are generalizable across clinical groups (i.e., patients with variable medical histories and undergoing nonorthopedic surgeries). Additionally, formally assessing different cognitive domains, pre- and postoperatively, will help quantify the extent and nature of any cognitive changes, and clarify whether pre to post cognitive changes are statistically meaningful. Nevertheless, the neuroprotective nature of physical activity is well established; regular participation in physical activity leads to improvements in cognitive functioning (31,32).
Mechanisms for POCI remain unclear and need study. In the context of cardiac surgeries, preserved resting-state functional connectivity in specific brain regions (e.g., posterior cingulate cortex, superior frontal gyrus) and reduced ischemia during the perioperative and postoperative periods have been identified as possible protective factors (33,34). In noncardiac surgery, younger age, more education (3), less depression (35,36), higher scores on executive function measures, and increased brain volume prior to surgery (37) are associated with POCI. It is unclear what processes support POCI in the context of bariatric surgery (38). However, increments in cognitive functioning may be attributed to improved physical and mental health secondary to weight loss. This is a reasonable hypothesis, as increments in executive functioning, language fluency, and memory have been attributed to weight loss, without surgical intervention, and through behavioral and lifestyle modifications (39). In the context of renal transplantation, POCI remains a contentious topic (40). Many studies are under powered and fail to account for other patient-specific characteristics that may explain POCI (e.g., age, general health). Overall, it is unlikely that factors related to surgery itself, an event that induces sympathetic activation and inflammation, promote cognitive improvements (41). Thus, cognitive reserve (42), psychiatric stability, decreased reliance on medications, and increased stamina are collectively viable mechanisms that promote POCI.
In contrast to POCI, PND (largely the focus of this book) is also a multidimensional syndrome, precipitated by several risk factors (see Table 4.1). Briefly, older adults who have reduced brain integrity, cognitive impairment, numerous medical comorbidities, and low education prior to surgery are at increased risk for PND. Thus, PND is well suited for the biopsychosocial framework. Up to now, research in the area seeks to identify risk factors. Less is known about the social and lifestyle processes that protect against PND, and promote POCI.
| Preoperative | Intraoperative | Postoperative | |
|---|---|---|---|
| Biological | Genetic | Surgery type | Pulmonary emboli |
| Age | Anesthetic | Blood loss | |
| Brain status | Induction rate | Oxygenation | |
| Comorbidity | Blood loss | Pain medication | |
| Rewarming rate | Fatigue level | ||
| Emboli | |||
| Psychological | Cognitive status | NA | Mood status |
| Mood status | |||
| Social | Education | NA | Nursing care |
| Lifestyle habits | Caregiver support |
Biopsychosocial Model of Health: Application to POCD/POCI
Under the biopsychosocial model of health, illness is a complex state influenced by an individual’s unique biological, psychological, and social characteristics (5,43,44). The biological level of the model recognizes that there are underlying physiological mechanisms involved in illness. Factors such as age, genetic predisposition, and general health contribute to disease development. In biomedicine, the presence and absence of disease is almost entirely established at the biological level (e.g., biomarker, laboratory results, and scans). The biopsychosocial model of health expands on the biomedical approach, asserting that psychological processes (e.g., hostility, identification with the sick role, and coping abilities) and social factors (e.g., participation in screening behaviors, exercise, and diet) contribute uniquely to the development and maintenance of illness. Within this framework, optimal health is achieved when interventions are comprehensive, and target the biological, psychological, and social factors that precipitate and maintain a condition (5,45).
Since its introduction in 1977, the biopsychosocial model of health has been widely incorporated into medical curricula (45). Supporters of the model describe it as an organizational framework that can guide research design and highlight its role in promoting more integrative, and client-centered, practices. Critics contend that the model is too vague and not suitable for the rigorous scientific endeavor. Others also argue that while it is useful inside the classroom and as a theoretical framework, it is too ambitious and difficult to implement in fast-paced clinical settings (46). Nevertheless, the biopsychosocial model of health has been widely used in a number of medical specialties.
In addiction medicine, the biopsychosocial framework helped to elucidate psychological and social factors that promote participation in risky behaviors (47,48) and has informed the development of client-centered interventions to promote behavioral change (49). In primary care, the biopsychosocial model of health has guided research into diabetes. As a result, psychological mechanisms (e.g., denial, diminished problem solving) and routine behaviors (e.g., sedentary lifestyle, poor nutrition) that promote illness have been identified. More importantly, this knowledge has advanced practice, and informed successful interventions design to optimize treatment compliance (49,50).
Factors That Promote POCI
Biological Level
General health. Good general health can be associated with optimal postoperative recovery (51). For example, healthy pulmonary functioning and normal-range serum creatinine level have been associated with preserved memory functions after cardiac surgery among older adults (52). Similarly, preoperative cardiovascular health also predicts postoperative outcomes, and Le Manach et al. (53) found that general preoperative cardiovascular health was associated with improved cognitive functioning. Reduced inflammation and physical resilience (i.e., more capable of resisting intraoperative complications shall they occur) may be some of the mechanisms that support this association. Katznelson et al. (54) found that administration of statin medication preoperatively is associated with significant reductions in vascular postoperative complications (e.g., myocardial infarction, stroke) and inflammation, which in turn was associated with hastened recovery (55).
Neuroanatomical. White and gray matter integrity are associated with optimal outcome over time from an “insult.” Key aspects to consider are cortical and subcortical gray as well as regional white matter integrity. These variables in combination with whole brain volume and ventricular size are now known to be important predictors of anesthesia response, cognitive decline, and rate of change into dementia, not only for anesthesia and surgery but also general health. In a review paper, Madden et al. (56) found that reduced white matter integrity was consistently correlated with decrements in processing speed, working memory, and executive functioning. Additionally, white matter atrophy has also been implicated in age-related cognitive decline (57) with white matter lesions associated with cognitive and disease patterns (e.g., 36,58,59). Similarly, brain volume loss in medial and lateral temporal lobe structures as well as anterior and posterior cingulate cortex predicted long-term decrements in verbal memory and complex attention (60). Presumably, in each case, preservation of these brain areas should be associated with reduced PND and perhaps, POCI
In the context of postoperative cognition, Sato et al. (61) found that increased postoperative fractional anisotropy was associated with increments in verbal and visually mediated memory after uncomplicated carotid endarterectomy. More recently, Kant et al. (62) found that white matter hyperintensities and cerebral infarcts are better predictors of PND than brain volume and ventricle size, although this is not a consistent finding. Price et al. (36) reported white matter hyperintensities as a risk factor for PND after total knee replacement surgery (TKA).
Additionally, patients with postoperative increases in hemispheric glucose metabolism exhibited significant improvements in measures of verbal and visually mediated memory (63). Similarly, in adults recovering from carotid artery stenting to treat carotid stenosis, Kao-Li et al. (64), found that increased glucose metabolism after surgery was significantly and positively associated with increments in global cognitive abilities, as measured by the MMSE, and verbal fluency. Combined, these findings suggest that patients who exhibit preserved CNS metabolism and have preserved white matter integrity may be more resilient after a procedure and demonstrate POCI.
Neuroinflammation. Chronic neuroinflammation has been implicated in neurodegenerative disorders (65,66), obesity (67), and traumatic brain injury (68). In general, dysregulation in inflammatory responses is associated with increased pathology, implicating it as a possible mechanism for cognitive increments/decrements (38). Literature on the role of inflammatory responses in the development of postoperative cognitive dysfunction is burgeoning. Li et al. (69) examined serum S-100β, interleukin (IL)-1β, IL-6, tumor necrosis factor-α (TNF-α), and C-reactive protein (CRP) in patients who underwent hip replacement surgery. Overall, patients with PND had higher levels of IL-6 and S-100β 1 hour and 6 hours after surgery, with respect to those who did not exhibit PND. In a comprehensive systematic review, Peng et al. (70) reported on the association between PND and peripheral inflammatory markers. Consistent with findings by Li et al. (69), lower plasma concentrations of neuroinflammatory markers, S-100β and IL-6, were often reported in patients who did not develop PND. Studies examining the role of other common inflammatory biomarkers such as CRP and interleukin cytokines (IL-2) have been inconclusive (38). Research examining inflammatory responses in the context of PND is still necessary. It is important to determine which markers are consistently associated with the condition and whether perioperative treatment, that targets specific markers, can successfully promote POCI or, at least, shorten PND.
Pain. Inattention, reduced processing speed, and executive dysfunction are common in adults experiencing chronic pain (71). In the context of postoperative cognitive changes, moderate and severe preoperative pain as well as pain increments, with respect to baseline level, immediately after surgery are associated with increased confusion and disorientation (72). Furthermore, pain throughout the night appears to be negatively and significantly associated with decrements in global cognitive functioning following surgery for femoral neck facture (73). Conversely, older adults with well-controlled pain, via the use of oral opioid analgesics, are less likely to experience delirium than their counterpart using intravenous opioid analgesics and other interventions (72). Nimmo et al. (73) suggests that effective pain management is fundamental to optimal recovery after surgery, as effective analgesia reduces stress, promotes physical independence, and improves the quality of recovery, which are known to support adequate cognition. Additional research in this area is necessary. Future studies should use accepted validated tools for the assessment of PND, pre- and postsurgical pain assessment, and report details of the anesthetic and analgesic techniques used during and after surgery.
Psychosocial Level
Consistent with the biopsychosocial model of health, it is now widely accepted that biological mechanisms underlying PND/POCI do not stand alone. That is, psychological as well as social factors contribute uniquely to changes in cognitive functioning. More explicitly, the evidence suggests that a person with a significant biological disease load (i.e., reduced brain glucose metabolism, diminished neuroanatomical integrity, and/or increased inflammation), who has adaptive coping skills and regularly participates in healthy behaviors, may exhibit less decline than a similar individual with weak social and psychological health.
Cognitive. Changes in cognitive functioning are the hallmark symptoms in PND. Assessing cognitive abilities in a systematic and psychometrically sound manner is necessary to characterize risk and optimize treatment, particularly for high-risk individuals. However, administration of neuropsychological measures is time consuming, and they have limited applicability within fast-paced clinical settings. While several cognitive screens are currently available (e.g., MMSE, MoCA, Mini-Cog) and have shown value for predictive outcome (2) examiners need to be cognizant of practice effects and effort. Existing screening tools do not provide enough details about individual cognitive domains, which limits their utility when providing recommendations or informing interventions. Furthermore, interpretation should involve qualified persons with expertise in understanding effort, practice effects, normative considerations, and neuroanatomical/cognitive profiles.
Efforts are currently under way by experts in the fields of neuropsychology, neurology, biomedical engineering, and anesthesiology to test psychometrically sound screening tools that could alert us about patients at high risk. Additionally, the idea that certain cognitive domains are the building blocks of other higher-order functions is gaining increased attention. In fact, basic attention, working memory, learning, and memory are now being considered cognitive vital signs, necessary to complete activities of daily living and instrumental to a patient’s ability to make informed decisions about their treatment and participate in their care.
Psychological. Emotional distress has been widely reported in patients opting to undergo cardiac (74,75) and bariatric surgeries (23,24). Evidence indicates that active depressive symptomatology may increase the likelihood for new coronary artery events, promote additional hospitalizations, and increase risk for mortality after cardiac surgery. In the context of acute postoperative cognitive changes, a diagnosis of major depressive disorder prior to surgery was associated with increased risk for delirium in adults undergoing cardiac surgery (74). Depressive symptomatology also appears to be a risk factor for long-term cognitive effects. For example, preoperative scores on a depression scale predicted cognitive decline 18 months after cardiac surgery, above and beyond baseline cognitive characteristics (75).
Recent evidence suggests that active mood symptoms may be proxies for general brain health, and active depressive symptomatology has been associated with reduced prefrontal and subcortical activation. Since most anesthetics target frontal-lobe networks, patients experiencing active mood symptoms may experience atypical anesthetic induction (59). Overall, preoperative evaluation of depressive symptoms will help to identify patients who may benefit from additional postoperative support. Future studies should also examine whether preoperative interventions, behavioral or pharmacological, designed to stabilize or improve mood prior to a procedure may help to optimize postoperative outcomes.
Others have examined the role of personality characteristics and perceived stress on postoperative cognitive outcomes. For example, Hudetz et al. (76) found that dispositional optimism prior to surgery significantly predicted reduced POCD following cardiac surgery. More recently, Hudetz (77) also explored the influence of preoperative attitudes and perception on postoperative outcomes. They found that perceived stress prior to surgery significantly correlated with preoperative depression scores, which in turn predicted postoperative decrements in processing speed and working memory (77). Thus, aside from mood, various characteristics of personality may impact the incidence of both PND and POCI.
Social Level
Lifestyle habits. Preoperative physical fitness and nutritional status are associated with surgical complications and recovery time (27,78). These findings have influenced clinical care, and interventions to protect against adverse outcomes have been implemented. In this regard, the concept of prehabilitation has been introduced to describe interventions designed to optimize a patient’s functional and mental capacity prior to an intervention (e.g., surgery; 79). Current evidence suggests that intense prehabilitation programs, focused on physical activity and nutrition, are effective in promoting functional independence. While the focus of these interventions has been functional independence, increments in aspects of cognition have also been reported. Research on the effects of prehabilitation on postoperative cognitive functioning would address an important gap in our current understanding.
Caregiver support. More than 33 million Americans provided unpaid caregiving to an older adult in 2015 (80). Primary informal caregivers (PICs) assist with finances, medication management, and ensure the safety and well-being of their loved one. PICs play an important role in recovery, often facilitating treatment compliance. Evidence indicates that adults with cognitive difficulties are more reliant on their caregivers than those experiencing physical limitations without any cognitive difficulties (81). This is a societal concern since increased caregiver burden is associated with depression and diminished health-related quality of life for the caregiver (80,82). There is increased interest in the role of social support in promoting optimal postoperative recovery (55,76). Additionally, there is growing interest in the ethical considerations that emerge when patients exhibit cognitive deficiencies prior to undergoing surgery, and appropriate caregivers have not been identified (83). Unfortunately, few studies have systematically assessed caregiver burden in adults with PND or whether reliable social support can hasten recovery and POCI.
Conclusion
Research on postoperative cognitive outcomes has been primarily dominated by concerns over the identification of decline (POCD or PND) and the factors that precipitate it. In this context, knowledge about PND has grown exponentially in the past 15 years, and risk factors have been identified. In general we believe the area lacks a clear theoretical framework, which limits our ability to synthesize findings and influence clinical practice. In this chapter, we propose that studying postoperative cognitive outcomes within the biopsychosocial framework may advance scientific inquiry and help to inform clinical practice.
In spite of our growing understanding about PND, research into the converse, POCI, has lagged considerably. It clearly exists and is important to characterize to understand the complex modulators of cognition perioperatively. As such, another goal of this chapter was to provide a working definition of POCI and summarize some of the literature in the area. Additionally, we described some of the biological, psychological, and social characteristics that promote POCI. In the context of POCI, biological factors have been disproportionately studied, and less is known about the psychological as well as social characteristics that may promote postsurgical recovery. As proposed by the biopsychosocial model of health, we argue that in addition to biological optimization (e.g., improved glucose regulation, hydration, oxygenation, and decreased inflammation), there are psychological as well as social characteristics that may be apt for interventions. Future research in these areas is necessary and will benefit from utilizing a theoretical framework that reflects the multifaceted nature of perioperative cognition.
References
Related posts:
Chapter 1 – Emergence Delirium
Chapter 2 – Postoperative Delirium
Chapter 5 – Persistent Perioperative Neurocognitive Disorder
Chapter 12 – Biomarkers of Postoperative Cognitive Dysfunction: Finding the Signal amid the Noise
Chapter 11 – Cognitive Testing for Perioperative Neurocognitive Disorder
Chapter 14 – Preoperative Testing to Identify Vulnerable Subgroups
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
