Background

Cognitive functioning is an evolving process encompassing human life and is a continuous endeavor of learning, maintaining short-and long-term memory, and electing to facilitate executive functioning [1]. The term “cognitive dysfunction” refers to difficulties with selective and sustained attention, learning and memory problems, deficits in visual and auditory processing, and processing speed [2]. Cognitive dysfunction can be caused by a direct neurological phenomenon, such as a brain injury, or by a non-direct phenomenon, such as surgery [3].

Post-operative cognitive dysfunction (POCD) is a specific cognitive impairment that involves functional impairment of the nervous system’s activities, such as selective attention, vigilance, perception, learning, memory, executive function, verbal and language abilities, emotion, visuospatial, and visuomotor abilities. POCD occurs in the absence of head trauma or other brain-related injuries [4, 5]. POCD is a serious public health concern, with prevalence rates ranging from 36.6% in young adults to 42.4% in the elderly [6]. It can occur following major invasive procedures such as cardiac [7], non-cardiac [6, 8], and carotid surgery that are lengthy and intensive [9]. POCD is diagnosed using standardized neuropsychological assessment tools that are available to assess cognitive function both pre- and post-operatively. The most clinically used screening tests for evaluating the overall cognitive functioning of patients are the Mini-Mental State Examination (MMSE; [10]) and the Montreal Cognitive Assessment (MoCA; [11]). The most commonly used clinically for assessing specific domains of cognition function rather than the overall assessment of cognitive function, are neuropsychological test batteries. These include the Trail-Making Test A (assessing visual attention, processing speed, and executive function), Trail-Making Test B (assessing cognitive flexibility, set-shifting, and executive functions), Digit Span Forward (assessing short-term auditory memory and attention), Digit Span Backward tests (assessing working memory and cognitive flexibility), Wechsler Memory Scale (assessing short-term memory, episodic memory and visuo-spatial working memory), and the Wechsler Adult Intelligence Scale (assessing vocabulary, comprehension, arithmetic and reasoning skills) [12,13,14]. While this information is valuable, due to POCD’s nature consisting of subjective and objective aspects, the assessment and diagnosis of POCD require approaches that extend beyond sole reliance on standardized test results, such as incorporating patients’ subjective reports alongside the outcomes of neuropsychological tests for a comprehensive evaluation.

Previous studies [6, 7, 15,16,17] have focused on the variations in the timing of symptom manifestation and their influence on the assessment and diagnosis of POCD. As a result, POCD has been categorized into early (acute) and late-onset POCD. These timing distinctions become essential as it recognized that several health-related outcomes are associated with POCD, such as delayed recovery, dependency on financial assistance [18], decreased quality of life [19], and an increased risk of death [20].

Despite the possibility of adverse long-term effects of POCD, the phenomenon has yet to be assigned a Diagnostic and Statistical Manual of mental disorders (DSM 5-TR) category [21, 22] even in the latest fifth edition; and conceptual clarity of POCD needs to be improved. For example, different terms have been used interchangeably to describe POCD, such as post-operative cognitive decline and post-operative cognitive impairment. Even the 1st, 2nd and 3rd International Study Group of Post-Operative Cognitive Dysfunction (ISPOCD), the major scientific and scholarly association focusing on this phenomenon, has not provided a clear definition of POCD [5, 23, 24]. To address the notable gaps in research while advancing the state of the science, our two-fold purpose of this study was to: (1) critically appraise, discuss, and challenge the various meanings of POCD thus providing a befitting definition for both clinical and research purposes; and (2) narratively synthesize the attributes, surrogate or related terms, antecedents including risk factors, and consequences of the concept. Such analysis can potentially lead to the development of a conceptual framework that is testable, and to research that can promote improved patient care and enhanced clinical outcomes.

Methods

Design and data sources

Our concept analysis follows Rodgers’ evolutionary method [25], recommended for analyzing concepts that change over time with increasing research and those that vary across contextual circumstances. The chosen concept for analysis is POCD, for which surrogate terms were identified. The method comprises six steps: (1) identifying the concept of interest (2), selecting the appropriate context and sample (3), identifying antecedents, attributes, and consequences (4), evaluating data about the concept’s characteristics (5), providing an illustrative example, and (6) recognizing implications for its ongoing development. Studies were retrieved through searches in PubMed, Cumulative Index of Nursing and Allied Health Literature (CINAHL), and Web of Science databases, assisted by a master’s-prepared university-affiliated librarian. The keywords used for finding relevant articles were: “post-operative cognitive dysfunction,” “post-operative cognitive decline,” “POCD,” and “post-operative cognitive impairment.” Database searches were limited to publications after 1998, which was the year of the first ISPOCD convention.

Inclusion criteria encompassed English-language quantitative original and review articles focusing on human subjects that used standard cognitive function assessment for POCD (using MoCA, MMSE, battery of tests, etc.), and provided definitions, assessed risk factors, or evaluated consequences (short and long terms) of POCD. Also, Webster’s dictionary, lezak’s neuropsychological assessment textbook, and DSM 5-TR were used to provide more accurate definitions of POCD and its attributes. Exclusion criteria encompassed conference abstracts, dissertations, letters to editors, opinions, commentary papers, and study protocols. Additionally, animal, in vivo and in vitro laboratory studies were excluded. The initial search yielded a total of 5.662 titles and abstracts from three sources of Web of Science, CINAHL, and PubMed. After exclusion for various reasons detailed in Figs. 1 and 86 studies were eligible for inclusion in the concept analysis (Fig. 1).

Fig. 1
figure 1

– PRISMA flow diagram

Full size image

Two independent reviewers (HV and TM) screened all titles, abstracts, and full-text articles. In the case of any discrepancies, the two reviewers discussed the different views on certain points and came to an agreement. The goal was to include an article that defined POCD (attributes) and/or identified risk factors (antecedents) and/or consequences. Lastly, a conceptual and operational definition was formulated. HV and AKPK entered all relevant information related to each area into an Excel 2020 database and discussed the findings with each other. HV and AKPK discussed all findings to ensure consistency and clarity. Four related (surrogate) terms were found in the literature, and they searched separately to find the definition and distinction by the first author.

Results

Of the 84 articles included in this concept analysis (see appendix 1), 33.3% were from the USA (n = 28), 19% from China (n = 16), 8.3% from Denmark (n = 7), 7.1% from Germany (n = 6), and 4.7% from the UK (n = 4). The remaining 27.6% represented various countries (n = 23). In terms of study designs, 52.4% were reviews (n = 44), 38% were cohort studies (n = 32), 7.1% were randomized controlled trials (RCT; n = 6), and 2.5% used other designs (n = 2).

Cognitive function and dysfunction

The DSM 5-TR identifies complex attention, executive function, learning and memory, language, perceptual-motor control, and social cognition as the six essential domains of cognitive function [22].

Attention is the most fundamental element of effective cognitive function [22, 26], and may be categorized into sustained, selective, and sustained selective types. Sustained attention is the ability to maintain sensitivity to stimuli, while selective attention is the ability to process a certain part of sensory input while omitting others. Sustained selective attention is the ability to process certain stimuli at the expense of others in a span of time, which some believe is a framework for executive function development [27]. Executive function refers to the capacity to complete complicated tasks such as planning, reasoning, problem solving, and adaptation to complex situations. It also includes working memory, inhibitory control, and cognitive flexibility. Executive function also controls other cognitive domains such as attention and memory [28]. Memory and learning are mutually interdependent and can be categorized by recent (short-term), rapid, recall, recognition, and long-term memory. Language or semantic memory is reflective of verbal fluency, grammar use, and syntax. Other important aspects of cognition include perceptual-motor control, which refers to the ability to understand visual perceptions. Finally, social cognition reflects emotion recognition and the ability to discern other’s mental status and feeling states. Cognitive processing is viewed in the cognitive psychology field as a series of sequential phases during which sensory data are converted, reduced or simplified, elaborated, stored, retrieved, and used for effective function and survival [29, 30]. According to DSM 5-TR, Neuropsychological testing typically shows cognitive deficits in areas such as executive function, verbal memory, and speed of processing.

In the current literature, three words are primarily used to describe cognitive changes: dysfunction, decline, and impairment. In Webster’s dictionary [30], dysfunction is defined as damaged (or impaired) unhealthy functioning (abnormal) that can be seen in interpersonal interactions or group interactions. Because POCD is often referred to with other deficit terms, such as decline and impairment, these terms are viewed as being related. Per Webster’s dictionary, the word decline (noun) refers to a “gradual physical or mental sinking and wasting away, or the period during which something is deteriorating or approaching its end (downward slope)”. Further, the word impairment is defined as “diminishment or loss of function or ability” [30]. Thus, the words “dysfunction” and “impairment” are considered approximate, except for some minor differences. Cognitive dysfunction typically describes temporary or reversible alterations in cognitive function, often stemming from factors like fatigue, medication side effects, or acute illnesses, and generally improves when the underlying cause is addressed. Cognitive impairment, on the other hand, suggests more enduring and noticeable deficits in cognitive abilities, often linked to conditions such as mild neurocognitive disorder (NCD), where cognitive issues exceed normal age-related changes but do not meet the criteria for dementia. In contrast, cognitive decline signifies a gradual and progressive deterioration in cognitive function over time, often associated with age-related changes but also seen in conditions like Alzheimer’s disease and various dementias. Cognitive decline tends to be an ongoing process and can ultimately become pervasive. These distinctions help clarify the nature and potential reversibility of cognitive changes.

In addressing the subject at hand, numerous studies have reviewed POCD definitions. Newman et al., in their review, defined POCD as a’significant’ change in postoperative cognition compared to the preoperative period, a definition more consistent than POD [4]. Liu et al. [13] portrayed POCD as a reversible form of MCI, while Needham et al. [12] described POCD as a mild neurocognitive disorder developing between 7 days and 1 year from surgery. Other studies have suggested that POCD can last up to 5 years, and even 7.5 years postoperatively. However, a study conducted by Steinmetz et al. [31] on 686 patients undergoing non-cardiac surgery indicated that POCD is largely reversible over time [12, 13, 24, 31, 32].

This uncertainty in the definition of POCD is manifested in the methods used by researchers to evaluate the incidence of POCD. For instance, Steinmetz et al. utilized the Visual Verbal Learning Test, the Concept Shifting Test, the Stroop Color Word Interference Test, and the Letter Digit Coding Test and calculated the change from the preoperative baseline while also comparing the normative results from healthy individuals to count for the learning effect [33]. Although Evered et al. [32] also benefited from the use of results obtained from healthy individuals, their neurocognitive battery measurement consisted of the Consortium to Establish a Registry for Alzheimer’s Disease-Auditory Verbal Learning Test, the Digit-Symbol Substitution Test, Trail Making Tests A and B, the Controlled Oral Word Association Test, the Semantic Fluency Test, the Grooved Pegboard Test, and the National Adult Reading Test, which is different from the Steinmetz study [31]. Newman et al. used a battery of five tests, comprising the Short Story module of the Randt memory test, the Digit Span and Digit Symbol subtests of the Wechsler Adult Intelligence Scale-Revised test, the Benton Revised Visual Retention test, and the Trail-Making test B [34].

Defining attributes

As noted in Table 1, the defining attributes (consistent conceptual characteristics) of POCD in patients post-surgery are as follows: (1) manifestation following after the acute phase (4–6 weeks), surpassing the usual time needed for recovery from the effects of surgery, which differs from surgery to surgery [35, 36] of the post-operative period [12]; (2) subtle- [37,38,39] to-extensive cognitive change (continuum of negative impact); (3) affects single or multiple cognitive domains [4, 31, 33, 37]; (4) reversible nature [12, 38, 31, 41]; and (5) may last for several days to years [5, 23, 32, 34, 37, 41] Implicitly, POCD should be assessed and detected by neuropsychological tests [5, 12,13,14], such as MoCA, MMSE, Digit Span test [34], trail making test [32], and other standardized neuropsychological battery of tests.

Table 1 – Defining attributes (characteristics) of POCD
Full size table

Related concepts/surrogate terms

Post-operative delirium (POD)

POD is an acute state of fluctuating and altered consciousness due to underlying internal and external factors such as anesthetics, pain, and cerebral hypoperfusion [102]. A critical element in managing POD involves appropriately identifying and addressing the condition’s underlying cause(s). Delirium is defined as “a disruption in attention (i.e., diminished capacity to direct, concentrate, sustain, and change in focus of attention) and awareness (reduced orientation to the environment)” or “additional disturbances in cognition,” as noted in the DSM 5-TR [22] manual. The incidence of POD was reported to range from 5 to 40% [22, 103, 104].

Major neurocognitive disorder (dementia)

Formerly known as dementia, major NCD is any progressive and irreversible cognitive disorder characterized by a cognitive deterioration that interrupts a person’s ability to perform various occupational, home, or social tasks. Dementia is seen as pathological changes in the brain with several potential causes, as opposed to a distinct illness or syndrome [105, 106]. Dementia occurs as part of a spectrum of diseases that include Alzheimer’s, vascular (occurs with stroke and diabetes) dementia, dementia with Lewy bodies (DWLB), Parkinson’s, and mixed varieties [107].

Mild neurocognitive disorder

Formerly known as mild cognitive impairment, mild NCD refers to the period between age-related normal cognition changes and the onset of dementia symptoms [108]. Mild NCD is a cognitive impairment with little interference in everyday instrumental tasks, thus different from dementia [109, 110]. Mild NCD can be treated or prevented from progressing to a more severe stage [111], which can be considered a reversible condition. Mild NCD is not associated with surgery per se but can occur as a result of aging (degenerative diseases) and changes in the environment (such as moving to a nursing home) [112].

Diagnosing mild neurocognitive disorder (NCD) is applicable when composite scores in multiple domains deviate 1–2 standard deviation from normal, and major NCD is when the score is 2 or more standard deviation from the mean [22].

Cognitive vitality (CV)

CV refers to an individual’s capacity to adjust to cognitive changes and maintain an optimal level of cognitive functioning, life satisfaction, and independent living by effectively balancing their energy and personal resources. CV serves as a compensatory mechanism to adapt against the cognitive decline of aging. Although CV isn’t measured in a singular manner, its scope can be outlined by five essential domains: (1) physiological and metabolic health; (2) physical capability; (3) cognitive function; (4) psychological well-being; and (5) social well-being [113, 102]. CV is closely related to resilience and may reflect an individual’s capacity to recover from assaults such as surgery; it can increase the quality of life and result in healthy longevity [114, 115].

Antecedents (risk factors)

The risk factors of POCD, as mentioned in Table 1, can be separated in 3 groups. First, lower educational level [45, 46], illiteracy [46, 47], and older age [20, 42,43,44] that are in relation to lower cognitive reserve ; Second is operation-related risk factors such as, surgical types and techniques (like cardiopulmonary bypass) [43, 44, 46, 53, 55,56,57,58,59,60,61,62,63,63], intraoperative management of homeostasis parameters (cerebral oxygenation, body temperature, blood glucose, and blood pressure) [65,66,67,67, 76, 77, 79] and type of anesthesia [48, 66,67,68,69,70,71,72,73,74,75,76,77,78,79,79, 105]. The third group consists of other perioperative risk factors such as alcohol use disorder [48, 49], genetic factors [46, 47, 50], preexisting cognitive impairment [51, 116], comorbidities [80, 81, 84,85,86,87,87, 117], and POD [41, 89,90,90].

Consequences (outcomes)

Outcomes of POCD may include impaired quality of life [8, 24, 34, 92,93,93], which may result from withdrawal from the labor force with increased patients’ dependencies [6, 18]; cognitive decline [32, 34, 94,95,96,97,97]; increased risk of dementia [31, 97]; increasing healthcare costs [98, 99]; and eventual mortality [6, 18, 96, 99,100,101,101].

Conceptual definition of POCD

Figure 2 provides a comprehensive definition of POCD, encompassing modifiable factors (those that can be manipulated by healthcare providers, i.e., surgeon, anesthesiologist, or nurse practitioner, to reduce the risk of incidence) and nonmodifiable factors (those that can’t be changed prior to or during surgery). The definition includes risk factors, defining attributes, and outcomes established based on previously discussed research, offering a thorough insight into POCD.

Fig. 2
figure 2

– POCD risk factors, defining attributes and outcomes diagram

Full size image

POCD manifests following the acute phase (4–6 weeks), surpassing the usual time needed for recovery from the effects of surgery, which differs from surgery to surgery. The post-operative period involves a subtle-to-extensive cognitive change (continuum of negative impact) that affects single or multiple cognitive domains, is reversible, and may last for several days to years. Post-Operative Delirium (POD), distinct from POCD, involves acute confusion and disorientation following surgery, requiring immediate medical attention. Since POD primarily entails dysfunction in attention and awareness, while POCD involves dysfunction in memory and executive function with or without other cognitive function domains, it is suggested to separate POD and POCD. Mild neurocognitive disorder, previously known as mild cognitive impairment, signifies cognitive difficulties that exceed normal age-related changes but fall short of dementia, typically involving memory problems. In contrast, major neurocognitive disorder, commonly referred to as dementia, is a severe and often irreversible condition characterized by significant cognitive decline that impedes daily functioning. This decline can stem from various underlying causes like Alzheimer’s disease and vascular dementia. These terms collectively reflect the diverse aspects of cognitive health, ranging from temporary postsurgical changes to chronic cognitive impairments and the promotion of lifelong cognitive well-being. For a precise diagnosis and management of cognitive issues, consultation with healthcare professionals is essential. Cognitive vitality represents a holistic approach to maintaining and enhancing cognitive well-being through lifestyle choices such as mental stimulation, a balanced diet, exercise, and social engagement.

According to our findings, we provided four case examples (See Fig. 3): (A) a patient with a normal cognitive function who is undergoing major surgery (total knee replacement) and will have no POD and, after hospital discharge, have normal cognitive function similar to before surgery (no POCD). (B) a patient with a normal cognitive function who is undergoing major surgery (cardiac surgery) and will have mild POD, and after a hospital discharge, it takes 6 weeks to go back to normal cognitive function similar to before surgery (presence of reversible POCD). (C) a patient with normal cognitive function who is undergoing major surgery (cardiac surgery) and will have severe POD, and after a hospital discharge, it takes 10 weeks to go back to normal cognitive function similar to before surgery (presence of reversible POCD). (D) a patient with a major neurocognitive disorder who undegoing major surgery (hip replacement) and will have severe POD, and after hospital discharge, it takes 10 weeks to go back to cognitive function but is not similar to before surgery (presence of persistent POCD/ cognitive decline).

Fig. 3
figure 3

Case Examples of Trajectories of Cognitive Function Before and After Surgery

Full size image

Discussion

In this study, an evolutionary approach was used to analyze the concept of POCD. Various studies address POCD, focusing mainly on a specific patient group and the concept’s risk factors. A recent review also noted that the heterogeneity of previous studies may make drawing solid conclusions from the literature difficult [118]. While previous studies fell short of providing an in-depth definition of POCD, incorporating its characteristics, risk factors, outcomes, and most importantly differentiating POCD from surrogate terms commonly mistaken for, the current study provides a unified definition of POCD that can be adopted in future studies.

A few studies conducted in 1998 and afterward did adequately defined POCD [5, 12,13,14, 23, 30,37,38,39,40,40] most of these studies identified the reported risk factors [20, 42,43,44,45,46,47,48, 51, 54,55,56,57,58,59,60,61,62,63,63, 65, 66] and outcomes [8, 31, 32, 34, 91, 92, 95,96,97,98,99,100,101, 119] previously reported in our concept analysis. The problem is that most of these earlier studies were conducted without a clear theoretical framework guiding the research. Consequently, there are significant variations in the foundational principles underlying these studies. As a result, disparities have emerged concerning the precise timing, diagnosis, and even the definition of POCD. This lack of theoretical guidance has led to three potential discrepancies or issues, which, in turn, may explain the divergent incidence rates of POCD that have been reported. First, the three [5, 23, 24] international societies of POCD (ISPOCD) have not provided a transparent, evidence-based, well-established, and consistent definition of POCD. As a result, some studies [17, 109,110,110, 120,121,122] assessed POCD over a short duration postoperatively, while others [111, 123,124,125] examined POCD over a long period. To standardize the timing of assessment across studies, it is better to distinguish between POCD (occur after acute phase) and POD (occur in acute phase), as they have different pathologies and diagnostic criteria. This idea is supported by Monk et al. [6], who categorized the development of POCD as early (at hospital discharge) and late (three months after surgery). Although Monk and colleagues reported the rate of delirium and showed that the incidence of POD was higher in patients with POCD, they did not explain how they distinguished between POD and POCD. Also, POD detection-related scales such as the Richmond Agitation and Sedation Scale (RASS) or Confusion Assessment Method (CAM)-ICU were not reported. It is suggested that future studies focus on POD and report its related variables.

Second, because no universal agreement exists regarding tools to assess POCD [37], previous studies used different measures or scales. MMSE [16, 17, 34, 125, 126] and MoCA [17, 122] are the tests used most frequently in studies. Also, different neuropsychological tests (battery) are employed, such as Trail-Making Test A and B Digit Span Forward and Backward, Wechsler Memory Scale, and the Wechsler Adult Intelligence Scale [122, 124, 126, 127]. When using the same tools (especially tests like the MMSE) repeatedly within a short period, a potential risk of “practice effect” (PE) arises. PE involves observing pseudo-improvement in patients’ cognitive function due to repeated exposure to the same questions [13]. Various tools and scales employed in earlier research make it challenging to pinpoint the specific cognitive domains that surgery is more likely to impact. This occurrence makes the studies too heterogeneous and complex to compare. Future studies should assess at least attention, learning, and memory, which are the fundamentals of cognition and are usually lacking in POD patients [115, 128].

Third, functional status (including quality of life), an important aspect of differentiating mild and major NCD, and symptoms (such as pain and sleep disruption) that may be associated with POCD were usually not evaluated in previous studies [129]. This information is essential to determine if patients’ symptoms or functional status are mediators or moderators in the development of POCD or even if patients’ symptoms lie completely under NCDs. In addition, because POCD can impair quality of life [24, 33, 34, 92,93,93], knowing a patient’s quality of life before surgery and how it could change in the future is also important.

Other than the three previously mentioned discrepancies found in the literature, some minor methodological considerations [130] differed across the studies. Some studies [34, 42, 85, 97] did not include a control (non-surgical) group. A control group is critical because it helps in understanding the magnitude of the intervention’s (surgery’s) effect and adjusting for cognitive change caused by normal aging. Furthermore, it is suggested that patients with preexisting mild NCD be included in future studies [129] and that more advanced statistical analysis (multivariate analysis, structural equation modeling) be performed to better correlate between pre- and postoperative cognitive dysfunction with POCD.

Over the last two decades, a new term, “post-operative cognitive improvement” (POCI) emerged. POCI refers to significant improvement in cognitive function postoperatively that is not affected by a practice effect or natural variability [131]. However, which domains of cognitive function should be assessed before and after surgery is not well understood. POCI is usually reported in carotid endarterectomy (CEA) surgery [127, 15] and coronary artery bypass graft (CABG) surgery [15, 132, 133]. As discussed earlier, these studies did not include non-surgical control groups, so their findings might not be generalizable. Also, it is unclear that POCI is unaffected by practice and/or learning effects. One assumption that may be helpful for future studies is that the majority of POCI reported in patients may have been caused by symptoms such as pain, memory problems, and dyspnea; therefore, after surgery, when certain symptoms are relieved, and patients are no longer suffering, they may be able to focus (pay attention) on answering neurocognitive assessment tests. This justification was also supported by the finding of Kougias et al. [127], who showed that cognitive improvement after surgery can be seen in attention, executive function, learning, and memory. Finally, although blood biomarkers and radiological imaging [134, 135] were considered in different studies along with neurocognitive tests to detect POCD, it is unclear which findings should be considered to support this diagnosis.

Implications for practice

Regarding the aforementioned risk factors, its advisable that strategies be developed for nurses and healthcare workers to prevent POCD: (1) screening all patients before surgery by conducting a cognitive assessment; (2) optimizing chronic conditions preoperatively appropriately; (3) applying best practices in anesthesia management (by a certified registered nurse anesthetist or anesthesiologist); (4) early screening for POD (and treating it if applicable); (5) and follow-up with patients about cognitive functioning (short- and long-term) are highly recommended.

Limitations

Although this study has many notable strengths, two main limitations are noted. First, the literature search was limited to studies published after ISPOCD1 (1998) and written in English, so some important studies may have been missed. Second, the literature search was limited to three online databases, and gray literature was not considered in this study.

Conclusion

POCD is a subtle-to-extensive cognitive change that affects single or multiple cognitive domains. It manifests following the acute phase of surgery, is reversible, and may last for several weeks to years. Implicitly, POCD should be assessed and detected by neuropsychological tests. POD is an acute change in cognition and should be differentiated from POCD. In certain types of surgery, patients may experience POCI after periods of POCD. However, more studies are needed to support this contention and the findings of this concept analysis.