Sample Dissertation on Postoperative Cognitive Dysfunction
Investigations of effects of surgical treatment on physical and mental status have attracted increasing interest. Advancements in surgical and anesthetic techniques have improved overall morbidity and mortality in patients undergoing major general and cardiac surgery. However, postoperative cognitive dysfunction (POCD) and decline remain as significant complications, particularly in older patients. POCD is defined as deterioration of intellectual function presenting as impaired memory or concentration. POCD is a common complication in older patients undergoing cardiac and non-cardiac surgery under general and local anesthesia. Postoperative cognitive deficits and decline result in prolonged hospitalizations, increased morbidity and mortality, and increased costs, and has an adverse impact on quality of life (Stockton, Cohen-Mansfield, and Billig., 2000; Canet et al., 2003; Hanning, 2005).
Research has progressed from case-study vignettes to epidemiologic analysis of morbidity and mortality. In order to achieve comparability across cases, some studies have focused on specific procedures, including varicose vein stripping, prostatectomy, cataract surgery, and hip replacement, whereas others have assessed outcomes related to more than one type of surgical intervention. Some investigators measured recovery time, as determined by cognitive functioning after the administration of different anesthetic agents (Ancelin, Roquefeuil, Ledesert, et al., 2001). It has frequently been speculated that POCD might be avoided by performing appropriate surgical procedures under regional anesthesia (Makensen, G. and Gelb, A., 2004). Numerous comparative studies using various assessment techniques have been conducted to test this hypothesis but no significant difference has yet been found. Some studies have compared outcome of surgery using different techniques, for example, spinal vs. general anesthesia for procedures including hip surgery and prostatectomy, and general vs. local anesthesia for cataract extraction. Follow up times have varied from 24 hours or less, to 1 month, 3 months, 10 months, and 1 year (Abildstrom, Rasmussen, Rentowl et al., 2001; 2004)
However, with the exception of open heart surgery, where a risk of prolonged neurological and neuropsychological deficits has consistently been identified, not merely in older patients, only two studies of other elective procedures reported significant mental changes that persisted and adversely affected the quality of life several months later. One study of other elective procedures reported significant mental changes that persisted and adversely affected the quality of life several months later. Hole et al. (1983) identified these changes in participants randomly assigned to general as compared with spinal anesthesia for hip replacement, but no standardized pre or post operative testing was conducted to validate the report.
Subsequent studies in various surgical groups have attempted to relate postoperative cognitive changes to 1) specific parameters of mental status, 2) time at which these parameters were studied, 3) type of surgery, 4) type of anesthesia, 5) demographic variables, and 6) pre-, intra, – and postoperative medical conditions (Vingerhoets, VanNooten, and Vermassen, 1997; Collie, Darby, Falleti et al. ,2002; Canet, Raeder, Rasmussen et al., 2003) . The most recent studies have been carried out preoperatively, postoperatively, and at short and long term follow-up. Moller and colleagues (1998) studied mental status changes pre and postoperatively using various non-cardiac surgical groups (including elective orthopedic patients) in an international multicenter program. Although they did find POCD in their patient group, it is not known if POCD exists for purely elective orthopedic surgical patients.
Preoperative and immediate postoperative cognitive comorbidities as well as physical comorbidities as predictors of outcome have been included in more recent studies in various surgical groups (Billing, Stockton, Cohen-Mansfield, 1996; Goldstein, Fogel, Young, 1996; Goldstein,Young, Fogel, and Benedict,1998). Several studies have also focused on the effects of anesthesia type on postoperative mental status (Williams-Russo, Urquhart, Sharrock et al., 1992; Crul, Hulstijn, and Burger, 1992; Goldstein et al., 1998). Despite these advances in methodology, all except for one of these studies have failed to find significant mental status changes among patients undergoing elective orthopedic surgery.
Advances in anaesthetic and surgical techniques have led to the conclusion that postoperative cognitive decline is currently less common than previously thought (Abilstrom et al., 2000; Rasmussen, Larsen, Houx, et al, 2001; Rasmussen, Johnson, Kuipers et al., 2003) . The benefits of such technological advances, however, may have been offset by the inclusion of older patients with more comorbidity. The extent to which postoperative cognitive dysfunction is detected will depend on measurement techniques, timing of the assessment, and statistical methods. One issue concerns the selection of neuropsychological tests. Because the cognitive changes may arise from more than one etiological mechanism, a procedure that assesses all major cognitive domains is preferable. If the test battery does not include assessment of, for example, frontal-lobe functions such as planning and abstraction, or parietal-lobe functions such as spatial and constructional abilities, abnormalities in these areas will be missed. Because of the limited time available for neuropsychological testing preoperatively, few studies have included tests that cover all major cognitive domains.
A second issue is preoperative baseline performance. There is substantial variability in neuropsychological performance at baseline, with some patients performing at expected age-adjusted and education-adjusted levels and others performing significantly below expected levels. This variability has been ascribed to emotional distress and preexisting cognitive impairment in some patients (Shaw, Bates, Cartlidge et al., 1986; Stump, Newman, Coker, Phipps, Miller, 1990; Moller, Cluitmans, Rasmussen, 1998), These findings continue to complicate interpretation of studies of cognitive outcome. A decline secondary to surgical procedures in patients who are already impaired at baseline may be underestimated. On the other hand, surgical procedures itself may be associated with cognitive dysfunction. The most common definition of cognitive change is a decline in performance by 1 SD on two or more tests (Rasmussen, 1998). An alternative is to assess change in specific cognitive domains, by examining the proportion of patients who show change in memory, language, and other areas (Rasmussen, 1998).
This approach may shed light on the pathophysiology of cognitive changes after surgical procedures. For example, if some cognitive domains are more susceptible to the effects of pain as a result of surgical procedures, changes might occur in certain cognitive domains but not in others. This domain specific approach also has the advantage that the pattern of cognitive changes may be differentiated from expected changes in normal aging, or with other causes of cognitive decline in the elderly.
Cognition is defined as the mental processes of perception, memory, and information processing, which allows the individual to acquire knowledge, solve problems, and plan for the future. It comprises the mental processes required for everyday adaptive living. Cognitive dysfunction is therefore the impairment of these processes. It is usually expressed by patients in terms of failure to perform certain tasks, or inability to complete mental tasks that were previously attainable. More specifically, deficits are generally characterized as limitations in attention, cognition, recognition, orientation, memory, and learning (Rasmussen, 2001). According to the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV), cognitive disorders can be divided into several conditions that are characterized by clinically significant deficits in cognition and memory, and representing a significant change from a previous level of functioning:
Delirium is an acute disturbance of consciousness and a change in cognition that tends to fluctuate during the course of the day.
Dementia is characterized by multiple cognitive deficits including memory impairment. The condition causes impairment in occupational or social functioning.
Amnestic Disorders are characterized by memory impairment in the absence of other significant cognitive impairments.
Mild neurocognitive disorder is one of the cognitive disorders not otherwise specified and for which research criteria have been suggested. This presentation is characterized by cognitive dysfunction presumed to be caused by either a general medical condition or substance use that does not meet criteria for any other disorders. An essential feature is that impairment in cognitive functioning is evidenced by neuropsychological testing.
Several researchers have attempted to categorize these different entities of cognitive dysfunction. The central anticholinergic syndrome is a complication observed following general anesthesia with manifestations ranging from excitatory symptoms such as agitation to central nervous system depression such as stupor, coma, and respiratory depression. The incidence varies between 1 and 40%. It is assumed that drugs used for general anesthesia are blocking central cholinergic transmission result in a relative lack of acetylcholine in the brain, which is essential for learning and memory (i.e. benzodiazepines, opioids) (Link, Papadopoulos, Dopjans, Guggenmoos-Holzmann, and Eyrich,1997). POCD has to be differentiated from delirium, which is a separate clinical syndrome. It is characterized by an acute decline in attention and cognition. Following surgery and anesthesia the incidence varies between 0 and 73% dependent on the type of surgery and studied patient population. The incidence is highest following cardiac and orthopedic surgery and in elderly patients. The symptoms of delirium usually develop within the first 4 to 5 days of the postoperative period with an acute onset and with a peak on the 2nd or 3rd postoperative day (Dyer, Ashton, and Teasdale, 1995). These two were harbingers for longer lasting postoperative cognitive deficits (POCD).Postoperative cognitive dysfunction (POCD) varies enormously depending on the definition, the composition of the test battery, and the time of the postoperative assessment. The incidence is reported to be 30-80% a few weeks after cardiac surgery and 10-60% after 3-6 months.
In the vast majority of patients with POCD, it must be regarded as a mild neurocognitive disorder and it requires neuropsychological testing for detection. More recent investigations have began to select neuropsychological tests that are more sensitive to cognitive impairment. Neuropsychological testing evaluates several aspects of cerebral function, such as problem solving, speed of information processing, flexibility and memory. POCD is statically and conventionally defined by comparing the preoperative performance to the postoperative performance at different moments in time.
There are several ways to define POCD. You can compare the individual cognitive performance of the patients before and after surgery, in which the patients also acts as his “own control.” This method has the advantage of revealing the individual magnitude of the cognitive dysfunction. This definition is conventional and what is considered as sufficient decline is defined different ways by various authors. Another method is applying pre and postoperative neuropsychological tests between different groups of patients. Rasmussen (2001) states that if an incidence of POCD is to be calculated, it is necessary to define diagnostic criteria based on the test battery used. This is usually done by considering deficits in the single tests. The degree of change in a test deemed to constitute POCD has varied widely between studies ranging from a deterioration of 1 SD in one or more tests in a battery to a more rigorous z-score (Rasmussen, 2001). The amplitude of POCD can be arbitrarily quantified as mild, moderate or severe, depending on the SD change score from baseline (J, 1.5, or 2 SD’s respectively). The computation of a z-score for each neuropsychological test; most authors agree that a Z-score >1.96 in more than 2 tests or a value of the composite Z score >1.96. Lastly, the earliest test point should be about one week after surgery once centrally acting analgesics are no longer required and any active metabolites have been eliminated.
Risk of POCD
The etiology leading to POCD is still poorly understood and is most likely multifactorial. Risk factors for POCD are either present preoperatively, including patient comorbidities and baseline cognitive function, are generated during surgery or are developed in the postoperative period mainly as complications. There have been several risk factors identified in the literature. Majority of studies to date have reported that advanced age as a risk factor for POCD. Particularly older patients with a history of alcohol abuse presented an increased risk for POCD compared to a control group (ISPOCD studies, 1998, 2000). Several studies point out that lower education level is associated with adyerse^mtcomes following cardiac as .well-as noncardiac surgery. Preexisting cognitive impairment and depression are often excluded from POCD studies, although they are at high risk to aggravate cognitive dysfunction Ancelin, Roquefeuil, Ledesert et al., 2001).
Many studies directed at POCD have studied patients immediately after surgery or within 1-2 days following surgery. Clearly, patients who are still under the influence of anaesthetic or analgesic drugs may be affected. This is of significance if they are expected to remember instructions or carry out complex tasks, such as driving a motor vehicle but this does not constitute POCD (Rasmussen et al., 2003). Rasmussen and colleagues (1998) state that the earliest test point should be about one week after surgery once centrally acting analgesics are no longer required and any active metabolites have been eliminated. Motivation may also be a problem in the early postoperative period. False negatives were encountered in a study of mid latency auditory evoked responses to awareness with patients who heard the command (and thus were aware) but did not respond (Wu, Hsu, Richman et al, 2004).
Duration and type of surgery have been determined to play a crucial role in the development of POCD. While elderly patients undergoing minor surgery with a mean duration of 33 minutes have not demonstrated a greater risk of developing POCD than a nonoperated control group. Older patients undergoing major noncardiac surgery with a mean duration of 190 minutes showed an increased incidence of POCD. The type of surgery presents an important risk factor for POCD with an increased risk for vascular, orthopedic and cardiac surgery (ISPOCD studies, 1998, 2000; Canet, Raeder, Rasmussen et al., 2003). A recent study has shown that older patients after non-cardiac surgery who experienced POCD were more likely to demonstrate moderate pain compared to patients without any cognitive deficits. Additionally, adequate pain management in this study appears to play an important role as orally administered analgesia was related to a lower risk for POCD compared to patients with intravenous patient controlled analgesia. However, the treatment was not randomized in this study so oral analgesia may only be a marker for a less painful state and the assessment of cognitive function was restricted to the 1st and 2nd postoperative day and might have been compromised by the sedating effects of the opioids (Wang et al., 2007). Hypothermia during the early postoperative period has also been associated with cognitive decline six weeks following cardiac surgery (Grocott et al., 2002).
The role of anesthesia was examined in a randomized comparison of 364 elderly patients undergoing major, predominantly orthopedic surgery, under either general anesthesia (GA) or regional anesthesia (RA) (Rasmussenet al., 2003). Cognitive function was assessed using five neuropsychological tests undertaken preoperatively and 7 days and three months postoperatively. POCD was defined as a combined Z score >1.96 or a Z score >1.96 in two or more test parameters. At 7 days, POCD was found 19.7% of patients after GA and 12.5% in RA. After 3 months, POCD was present in 14% after GA and 14% after RA. The incidence of POCD after one week was significantly greater after GA when they excluded patients who did not receive the allocated anesthetic. There was no significant difference in the frequency of POCD between the groups receiving GA and those receiving RA after three months, suggesting that anesthesia was not a risk factor. The investigators concluded that there was no causative relationship between general anesthesia and long term POCD. Regional anesthesia may decrease the incidence of POCD early after surgery. However, as most patients in the RA group received a sedative infusion of propofol, this conclusion may be questioned.
Harming and colleagues (2003) investigated the effects in rats of repeated anesthesia throughout life with pentobarbital, compared with a control group. Central cholinergic function was estimated by radiolabeled alpha-bungarotoxin and epibatidine binding in the cortex, striatum and hippocampus when the rats were.26 months old.
There was a highly significant reduction in the bungarotoxin binding in the superior cortex. Aipha-Bungarotoxin binds to the alpha-seven subunit of the nicotinic receptor, which is also one of sites for anaesthetic binding (Backman, Fiset, and Plourde, 2004)., and is often most reduced in patients with Alzheimer’s Disease (Culley, Baxter, Yukhanov, and Crosby, G., 2003; 2004). This was a small study with an agent no longer used in human anaesthetic practice and the results should be interpreted with caution. There is however, some support for an effect of long term administration of cholinergic drugs on cognitive function from other fields. For example, patients with Parkinson’s disease treated with anti-muscarinic drugs are more likely to show Alzheimer pathology at post-mortem examination (Perry, Kilford, Lees et al., 2003). Nicotine has been shown to be protective of nicotine cholinergic receptors and low level exposure to organophosphorus esters may cause neurotoxicity (Jamal, Hansen, and Julu, 2002).
Greater impulsivity in behavioral task performance was noted between the older rats that had been subject to repeated anesthesia throughout life and the control animals mentioned above. Culley and colleagues (2003) have reported long term effects of anesthesia on cognitive function in rats with agents commonly used in human practice.
The apolipoprotein (APOE) genotype has been shown to be one of the first genetic variants to be associated with POCD. The role of APOE genotype was investigated in 976 patients undergoing major surgery in the ISPOCD2 studies (Abildstrom, Rasmussen, Christiansen et al, 2000; 2004) In this study, the e4 allele was not considered a risk factor for POCD. This does not rule out the possibility of a genetic proprensity for POCD but suggest that other candidate genes should be sought. Other recent studies have questioned the impact of APOE on the development of POCD, speculated that the APOE is only one of multiple gene variants altering cognitive outcomes.
Hypercortisolaemia has been known for some time to impair cognitive function. It was hypothesized that hypothalamic cell loss in the elderly impaired the normal mechanisms that damp down the increased Cortisol secretion that follows stress and the normal hypercortisolaemia of surgery would be enhanced and prolonged. Morning and afternoon salivary Cortisol concentrations were measured in patients receiving a general anesthetic in the randomized study of major surgery in the elderly, which formed part of the ISPOCD2 studies. Other stress markers such as IL-6 were also measured. There was no evidence of greater or prolonged Cortisol release in participants with POCD although there was a loss of circadian effect in those patients (Rasmussen et al., 2003).
It is still unclear if patients undergoing noncardiac surgery and anesthesia have long term or permanent decline in cognitive function, at least after the first week in older patients. There is a body of evidence suggesting that POCD does occur in these patients, but further research is necessary.
The evidence for POCD following cardiopulmonary bypass (CPB) is much stronger and the incidence is much greater than noncardiac surgery. However, two studies used the same test battery and methodology in both cardiac and major abdominal surgical patients and reported a similar or greater incidence of POCD in the non-cardiac group as recorded in Appendix A. Before the 1990’s, most reports of POCD in noncardiac surgery were anecdotal and were generally felt to be a response to some perioperative catastrophe that was not caught by the medical staff. Many studies demonstrated marked hypoxemia at night which was at its worst on the second to fourth night after surgery (DeKosky, Ikonomovic, Paulin et al., 2000). Other studies showed that this was a result of rebound slow wave and rapid eye movement on those nights, following their suppression on the first postoperative night coupled with the parallel decline in lung function. This previously unreported hypoxemia seemed to be the cause of several postoperative complications including myocardial ischemia and infarction and cognitive dysfunction.
Moller and colleagues (1998), conducted a major international multicenter study on the benefits of pulse oximetry in anesthetic practice, co-ordinated an international group of investigators (ISPOCD) to determine the prevalence or absence of POCD in the elderly. Thirteen hospitals in eight European countries and the United States recruited patients to same study by the same protocol. Eligible patients had presented for major abdominal, non-cardiac thoracic, or orthopedic surgery under general anesthesia and a hospital stay of at least 4 days. They gave priority to patients presenting for major abdominal and thoracic surgery. Centers were asked to recruit no more than 25% of their patients from those admitted for major orthopedic surgery (hip and knee arthroplasty). The neuropsychological test battery included a visual verbal learning test (based on Rey Auditory Test), concept shifting test (based on trail making test from Halstead and Reitan’s neuropsychological test battery). It also included the Stroop Color-Word Interference test, a paper and pencil memory scanning test, letter-digit coding test (adopted from Weschsler Adult Intelligence Scale), and the four boxes test. Patients IQ was measured preoperatively by part III of the Cattell culture fair IQ test. Mood was assessed with Zung depression scale, and patients self assessed cognitive decline from the short Cognitive -failures questionnaire.
The neuropsychological test battery and continuous physiological monitoring were done on 1218 patients, aged over 60 years old, before and one week and 3 months after major surgery. A subgroup of 336 patients was studied again 1-2 years later. Forty seven normal participants were studied with the same test battery at the same time intervals. POCD was defined as a z-score of more than 2.0 in at least two tests or a composite z-score above 2.0. At 7 days postoperatively, investigators found cognitive dysfunction in 266 (25.8%) patients. The second postoperative test was done in 3 months and found cognitive dysfunction in 94 patients (9.9%). The investigators also found a significant relation between early POCD and increasing age, increasing duration of anesthesia, less education, second operation, postoperative infections, and respiratory complications. The investigators concluded that POCD existed and age was a major risk factor. Neither hypoxemia nor hypotension, nor the combination, were risk factors for POCD.
The same group conducted, in a second multicenter collaborative program of research (ISPOCD2) to investigate further whether POCD followed minor surgery in the elderly and major surgery in the middle aged. Inpatient surgery was characterized by a maximum expected stay of 1 night and 1 preoperative night’s stay. Outpatient surgery was characterized by expected discharge from the hospital on the day of surgery, with no preoperative night’s stay. Subjective assessment of cognitive decline was evaluated using the Subjective Cognitive Functioning questionnaire (SCF). Mood was evaluated with the Geriatric Depression Scale (GDS) and degree of independence in activities of daily living was evaluated with the Instrumental Activity of Daily Living questionnaire (IADL). Twenty- two of 323 patients undergoing minor surgery displayed POCD (6.8%) 7 days postoperatively. At 3 months, the incidence of POCD was 6.6%. They concluded that POCD was present to a very small degree in the elderly after in-patient minor surgery after one week, but not at three months. The same was true for the middle aged undergoing major surgery (Rasmussen et al., 2001).
Abildstrom and colleagues (2000) examined if POCD persisted 1-2 years after surgery. Investigators recruited 336 patients and subjected them to the same battery as the ISPOCD studies at 12 months and 24 months. A control group of 47 non-hospitalized volunteers of similar age were tested with the same test battery at the same intervals. They found 1-2 years after surgery, 35 out of 336 patients (10.4%) had cognitive dysfunction. Three patients showed POCD at all three postoperative test sessions.
Logistic regression analysis identified age, early POCD, and infection within the. first three months as significant risk factors for long term cognitive dysfunction. While the ISPOCD studies remain the largest and best controlled of the studies conducted to date, it can properly be questioned whether the psychometric test battery was sufficiently sensitive and robust. Rasmussen (2004) reanalyzed the data from all patients that participated in the ISPOCD studies to examine the effects of test-retest variability. He concluded that by comparing the ratio of POCD with postoperative cognitive improvement (POCI), he could be confident of cognitive decline only in older patients one week after major surgery. Only 30-48% of patients with POCD at 3 months also had POCD at one week. POCD may be progressive and only become apparent several months after surgery.
The greatest incidence of POCD and the greatest number of studies is in patients undergoing cardiopulmonary bypass surgery (CPB). Many studies do not meet the stringent criteria set by Rasmussen, but there are sufficient large scale studies using appropriate test batteries and control groups to suggest that POCD, both early and late, does commonly occur in these patients. Phillips-Bute, Matthew, Blumenthal et al. (2006) examined POCD on the quality of fife after coronary artery bypass grafting surgery (CABG) from the patients’ perspective. Surgical patients from Duke University Hospital (N=551) were assessed at baseline, 6 weeks, and 1 year following surgery. The cognitive battery consisted of a short story module of the Randt Memory Test, the Digit Span subtest from the Weschsler Adult Intelligence Scale -Revised (WAIS-R), Modified Visual Reproduction Test from the Weschsler Memory Scale (WMS), Digit Symbol Coding subtest from the WAIS-R, and the Trail Making Test (Part B). Neurocognitive deficits, defined as 1 SD decline in one or more domains, occurred in 41% of patients at the 6 week follow up. At one year, 36.8% of patients demonstrated POCD.
Newman (2001) used the same battery and found an incidence of 53% at discharge from the hospital and an incidence of 36,24, and 42% six weeks, six months, and five years respectively, after surgery. Early decline predicted later decline in this patient group. Similar findings have been reported by Stygall and colleagues (2003). However, neither study used a rigorous definition of POCD or a control group to control for learning effects. Such studies raise the possibility that operation is a risk factor for early cognitive decline.
Further evidence comes from studies by Collie and colleagues (2002). The development of Alzheimer’s 5-6 years after surgery was determined in 5216 patients who had undergone coronary bypass grafting (CABG) and compared with 3954 patients who had undergone percutaneous transluminal corornary angioplasty (PTCA). Assessment data was collected postoperatively only. The adjusted risk of CABG vs. PTCA was 1.71 (P=.04). However, in a further study, the same group compared the incidence of Alzheimer’s patients who had undergone either a prostatectomy or a herniography under general (GA) or loco-regional anesthesia (LA). The adjusted risk of GA vs. LA was .65 and .71 respectively. The authors suggest that this may indicate that GA may delay the onset of AD. As the patients were not randomly allocated to LA or GA however, it may indicate also that frailer patients had their operations under LA.
There have been several studies (listed in Appendix B) examining the mental status of orthopedic patients after surgery, although not purporting to study POCD. Sosa Rex, Worland, and Blanco (2004) investigated the incidence of mental status change following total joint arthroplasty. They examined cognitive status in thirty total knee arthroplasty patients pre and postoperatively using the MMSE. There was no reported change in MMSE scores from baseline to third day postoperatively and after discharge six weeks later.
Kagansky, Rimon, Simona and colleagues (2004) sought to determine the incidence of delirium and its precipitating factors in older hip fracture patients. A group of 137 hip fracture patients, over the age of 75, had their cognitive status, measured pre and post operatively using the MMSE. Cognitive assessment was taken on admission, one week after surgery, and one month after surgery. Delirium participants were assessed even further at three months after surgery. Non delirious patients (N=90) MMSE scores were higher than the delirious patients and their scores gradually improved from baseline, one week, to one month after surgery.
Milisen, Foreman, Abraham et al. (2001) developed an intervention program for older hip fracture patients that experience delirium during their hospital stay. They assessed cognitive status in 60 patients using the MMSE at admission and on the first, third, fifth, eighth, and twelfth postoperative day. Investigators found that the nondelirious patient group’s MMSE scores gradually improved on each postoperative day.
Williams-Russo et al. (1992) compared the effects of postoperative analgesia using epidural vs. intravenous infusions on the incidence of delirium in bilateral knee replacement patients. Fifty one patients, age 65 years and older, were candidates for orthopedic surgery. They used the Mattis Dementia Rating Scale (DRS) as the assessment of cognitive status at baseline preoperatively and daily postoperatively until discharge (days 1 -7). All participants received regional anesthesia. Non delirious surgical group (N~30) showed general practice effects for the MMSE showing daily incremental improvement approaching the maximum possible score.
Two orthopedic studies to date have studied postoperative cognitive decline and its influence on other patient postoperative factors such as delirium. Gruber-Baldini, Zimmerman, and Morrison (2003) studied cognitive impairment in hip fracture patients and its relationship to delirium and long-term cognitive decline. They recruited 674 hip fracture patients ages 65 and older from hospitals in Baltimore, Maryland. They used the MMSE as the measure of cognitive status. They administered assessments at admission, postoperatively, and again at 2 and 12 month follow-ups. Cognitive impairment first detected in the hospital (preoperatively, postoperatively) persisted over 2 and 12 months in more than 40% of patients with delirium. Fifty percent of non-delirious patients (N-263), with no baseline cognitive impairment showed gradual improvement in their cognitive function from admission to two and twelve month time periods.
Dolan et al. (2000) studied delirium and it relationship to long-term cognitive status in 682 hip fractured patients post-operatively only and at 2, 6, 12, 18 and 24-month follow-ups. Cognitive status was assessed with the MMSE. Non-delirious patients (N=590) MMSE scores stayed relatively stable between two and eighteen months. After eighteen months, non-delirious patients began to show a slight decrease in their MMSE scores by the 2 year assessment period. However, baseline cognitive status was not assessed in this group.
This review of the orthopedic literature brings up another issue in common with numerous other clinical and epidemiological investigations of the mental status of older people for more than a decade. The Mini-Mental State Exam (MMSE) has been the primary instrument used in examining cognition in surgical studies. However, methodologies have varied, with differences in the age range of study participants and other inclusion criteria, and also in outcome measures. For example, a score of 23/30 has been used to define and/or exclude participants with preoperative “cognitive impairment” in some investigations, but not in others. Evaluation criteria have included between group changes in mean MMSE score, within-subjects variation above and below the 23/30 impairment level, and a 2- point decrease in MMSE performance. The results of studies using the MMSE have largely replicated those in which other instruments were used; that is cognitive decline detected in the early postoperative period has resolved within hours or days.
Results from the review on POCD suggest that surgical procedures increase the risk of short and long-term neuropsychological decline. Unfortunately, the strength of the conclusion is limited by the methodological differences and limitations of the investigations, including the use of the Mini Mental State Examination (MMSE) specifically in orthopedic literature and in the methodological approaches used to identify cognitive impairment. The MMSE is not sensitive enough to the level of cognitive change in various patient groups as well as identify particular domains with impaired performance (Sosa Rex et al., 2004; Kagansky et al, 2004). Most normal adult participants of any age, score at, or very close to, the maximum score with ease. Subtle degrees of cognitive decline will therefore not be detected because of a ceiling effect. In addition, it has no parallel versions and consequently the same questions are administered with each application. This permits learning effects in participants who score just less than the maximum but retain sufficient mental capacity to learn, but not in those who score either very low or the maximum. Such differences make it difficult to generalize findings across studies and raise questions regarding the conclusions of specific individual studies, particularly as they relate to the incidence of cognitive impairment.
There are several reasons for studying elective joint replacement patients. Elective orthopedic surgery is commonly performed in older persons. The elective nature allows for more thorough and accurate determination of potential risk factors through preoperative assessment of medical, mental, and functional status. It also allows for more standardized presentation and delivery of surgery, anesthesia, postoperative nursing care, and surveillance. Lastly, it avoids the confounding effects of fracture-related pain, stress, and analgesia (due to use of local anesthetics as opposed to general anesthesia). However, this group of surgical patients have still shown to have one of the highest incidences of postoperative delirium which is believed to be a marker for postoperative cognitive decline.
Patients with mood disorders in general and depression in particular to show cognitive impairment in various cognitive domains. The more severe their condition, the more apt they are to be cognitively impaired. Depressed individuals may have severe, global cognitive deficits or focal, discrete cognitive deficits, or they may be cognitively intact (Sabbe, Hulstijn,VanHoof et al., 1999; Ravnkilde, Videbech,Clemmensen et al.,2002; Rinck and Becker, 2003). Their cognitive status is dependent of age, depression severity, premorbid cognitive state, or whether they have a comorbid condition, such as stroke or early dementia. However, even mildly depressed patients without complicating factors can show more impairment, as a group than normals. Patients with bipolar disorder or depression with psychosis are more impaired than patients with nonpsychotic unipolar depression (Brown, Scott, Bench et a!., 1994; Jeste, Heaton, Paulsen et al.,1996; Barch, Sheline, and Csernansky, 2003).
Researchers have established that cognitive impairment is comorbid with affective illness, and deficits have been elicited in every domain. The focus of recent research has been to detect a pattern of specific cognitive functions that might be selectively impaired in depressed patients. Specific cognitive deficits have been demonstrated in tests of sustained and/or selective attention. Attention problems have been demonstrated in mixed groups of depressed patients, including young patients that are drug free, as well as geriatric patients after medical surgeries (Brown, Scott, Bench et al, 1994; Porter, Gallagher, Thompson et al., 2003; Kaiser, Unger, Kiefer et al., 2003). Impairments in working memory, a cognitive function that requires effortful attention, also occur in patients with depression. The attentional deficits of depressed patients are more likely to be evident in effortful tasks (Ancelin, Roquefeuil, Ledesert et al., 2001).
Studies of attentional impairment in depressed patients highlight the delicate nature of their subjective responses and the interaction between cognition and emotional response. Depressed patients tend to overreact to the mistakes they make. Farrin, Hull and Unwin et al. (2003) found that depressed men made more errors on a sustained attention task than nondepressed men, but they reported much higher incidences of cognitive failures on a standardized questionnaire. It was concluded that depressed patients responded “catastrophically” to errors. Making mistakes, even on a simple task of sustained attention, seemed to heighten their subjective sense of failure.
Another specific area of cognitive impairment observed in depressed patients is on tests of memory. Depressed patients are aware of memory impairment in their everyday lives. Studies have suggested that memory systems reliant on medial temporal lobe structures are impaired in patients with depression. The relationship between depression and some specific component of the memory system, however, is ambiguous. To date, studies have demonstrated problems with encoding as well as retrieval, recall as well as recognition. Depressed patients have particular difficulties with memory tasks requiring sustained effort, such as list learning and free recall, which are qualitatively different from tasks carried out automatically (i.e. memory for spatial tasks) (Barch et al., 2003; Goldberg, T.E., Gold, J. M, Greenberg, R., et al. (1999).
Psychomotor retardation is not necessarily pathognomonic of depression. However, it is one of the most persistent symptoms of major depression and one that is often demonstrable on cognitive tests. Psychomotor retardation can be demonstrated in depressed patients numerous ways including reaction time measures, information processing speed, writing and drawing tasks, and other fine and gross motor measures. Older depressed patients are also more likely to evidence psychomotor retardation than younger depressed patients (Houx and Jolles, 1993; Sabbe et al., 1999).
Language functions tend to be preserved in various forms of depression, although impairments in fluency have been noted. Fossati, Guillaume, Ergis et al. (2003) found verbal fluency impairments in patients with depression to be associated with reduced ability to shift mental set on card sorting tests, suggesting that language deficits were not primary but reflective of general executive functioning problems in depression.
Performance on measures of executive control functions tend to be impaired in depressed patients. Researchers have proposed that even if the impairment in brain function is global and diffuse, there is “particular involvement of the frontal lobes in nonpsychotic unipolar depression” (Fossati and Ergis, 2002). Moreover, executive dysfunction can be demonstrated in old as well as young depressed patients. Unipolar depressed patients exhibit executive deficits in tests of inhibition, problem-solving and planning. Cognitive inhibition deficits in depressed patients can lead to inefficient allocation of cognitive resources. They can cause the depressed patient to process information that is either irrelevant or counterproductive and thus reduce his or her capacity to deal effectively with depressive thinking and mood control (Fossati and Ergis, 2002).
Several studies have found evidence of problem-solving impairments in depressed patients. In card sorting tasks, depressed participants have difficulty with hypothesis testing and cognitive flexibility. This state of cognitive rigidity can prevent patients from coping with life events, thus perpetuating depressed mood by prolonging stress exposure. Planning tasks, such as Tower of London tests, also demonstrate that depressed patients fail to use negative feedback as a motivational boost to improve their performance (Fossati and Ergis, 2002; Fossati, Coyette, and Ergis, 2002). In addition, executive functioning deficits may predict a poorer outcome in depression. Thus initiation and perseveration scores – measure of cognitive flexibility – are associated with relapse and recurrence of depression and residual depressive symptoms (Brown et al,, 1994; Fossati et al, 2003; Bhalla, Butters, Zmuda et al, 2005).
Brain imaging studies show that reduced blood flow, particularly in the medial prefrontal cortex and dorsal anterior cingulate cortex, subserves executive impairments in depression. Neuroimaging studies also underscore the importance of mood-cognitive interactions in depression. A recent working model of depression implicates the failure of the coordinated interaction of distributed cortical-limbic pathways in the pathology of depression. According to this model, neocortical (prefrontal and parietal regions) and superior limbic elements (dorsal anterior cingulate) are postulated to mediate impaired attention and executive function, whereas ventral limbic regions (ventral anterior cingulate, subcortical structures) are postulated to mediate circadian and vegetative aspects of depression (Bell-McGinty, Butters, Meltzer et al. ,2002).
Several reports indicate an increased prevalence of depressive illness among general hospital inpatients compared to a non-inpatient population. Veroy and colleagues (2003) studied the prevalence of depression in general surgical patients between the ages of 18-65. These patients were undergoing orthopedic, gastrointestinal, or pulmonary surgery and were required to have an inpatient stay of more than two days. Current depression and dysthymia was diagnosed using the Structured Clinical Inte Depression is also known to influence attention and short term memory and it appears partly and variably involved in the decline in cognitive performance after anesthesia (Ancelin et al., 2001; Burns, Banerjee, Morris et al., 2007). Yesavage et al. (1983) alluded to the problem of discriminating between dementia and depression in older patients, since depressed mood may be manifest as passive refusal to respond appropriately to cognitive tests, and may also be accompanied by subjective experience of cognitive impairment, including memory loss and difficulty concentrating.
Chodosh, Kado, Seeman and Karlamangla (2007) studied 1,189 older surgical patients (age 70-79 at baseline) to look at the relationship between depression and long term cognitive decline. Cognitive performance was measured at baseline and at a seven year follow up with the Short Portable Mental Status Questionnaire (SPMSQ). Summary scores from standard tests of naming, construction, spatial recognition, abstraction, and delayed recall were studied. Depressive symptoms were also assessed at same time periods with the Hopkins Symptoms Checklist. After adjusting for several confounds such as age, education, and chronic health conditions such as diabetes and hypertension, a higher number of baseline depressive symptoms were strongly associated with greater seven-year decline in cognitive performance with higher odds of cognitive impairment (decline in SPMSQ score to <6). They concluded that depressive symptomatology independently predicts cognitive decline and cognitive impairment postoperatively in previously high functioning older adults.
As research has broadened in recent years, including data from neuroimaging technology, investigators have aimed to ascribe a pattern to the cognitive deficits that occur in depression. There are at least three theories that ascribe to a specific pattern to the neuropsychology of depression. One theory is the effort hypothesis, which states that performance on effortful tasks is disproportionately impaired in depressives comparedwith the performance on automatic tasks. The second, the cognitive speed hypothesis, states that depression is characterized by cognitive slowness and that slowing may be at the root of other cognitive impairments. Research indicates that cognitive functioning in depression is characterized by a reduced speed of information processing. Researchers who favor the cognitive speed hypothesis tend to dismiss the effort hypothesis, although the two are by no means mutually exclusive. The third hypothesis, is that impairment in executive control functions is central to the cognition in depressed patients. Because at least some degree of neuropsychological impairment is a trait marker for depression “localizing” the deficits of depressed patients to one particular functional system would be a signal advance. In contrast to theories of specific impairment is what could be referred to as the global impairment hypothesis; that depressed patients suffer from diffuse cognitive impairments – that their test performance is heterogeneous and that group analysis does not reveal any coherent pattern of dysfunction. Ravnkilde et al. (2002) concluded that “the large range of existing neuropsychological, neuropsychiatry, and more recently, neuroimaging investigations have not yet given a consistent picture of the psychological disturbances involved in depression”. Their research indicated that 1) depressed patients suffer from widespread cognitive impairments, 2) test performance was heterogeneous, and 3) group analysis did not allow any hypothesis on a possible pattern to the dysfunction.
Pertinent literature is compromised by studies of heterogeneous patients and the use of different tests that render comparisons across studies extremely difficult, or they administer test batteries that address performance in only one or two cognitive domains. Nevertheless, the literature is clear in showing that patients with depression are, as a group, subject to neuropsychological deficits in attention, memory, psychomotor speed, processing speed, and executive dysfunction. Although many prospective studies support a causal relationship between depression and cognitive dysfunction and decline, they may be confounded by preexisting participant comorbity (i.e. pain) that contribute to the development of both depression and cognitive dysfunction.
The neuroanatomical pathways associated with pain perception include complex mechanisms at the level of the spinal cord and complex supraspinal neural networks involved in processing acute and chronic pain. There are two systems involved in pain perception that include the lateral and medial lemniscus pain system which is made on the basis of the divergence of the spinothalamic or trigeminothalamic projections in the thalamus. The lateral system refers to the projections to the ventral posterolateral and the ventral posteromedial thalamic nuclei which in turn project to somatosensory cortex. The medial system involves projections to the medial thalamic nuclei and from there to limbic cortices including the anterior cingulated cortex, orbitofrontal cortex, amygdala, and other structures. In addition, the medial system involves connections with the periaqueductal gray matter that is involved in mediating nocioceptive inhibition as well as integrating behavioral response to potentially threatening or stressful stimuli (Nicholson and Martelli, 2004).
In most affected patients, postoperative chronic pain closely resembles neuropathic pain (Jung, Ahrendt, Oaklander, Dworkin, 2003; Mikkelsen, Werner, Lassen, Kehlet, 2004). Major nerves trespass the surgical field of most of the surgical procedures associated with chronic pain, and damage to these nerves is probably a prerequisite for the development of chronic pain. An ideal model for studying chronic pain in surgical patients, and establishing predictive factors for the condition, would include preoperative and postoperative assessment of psychological and neurophysiological factors, as well as thorough clinical investigation to exclude other causes of the chronic pain state.
Physical disability is one of most feared consequences of aging, and can be contributed to by a wide range of factors, including cognitive impairments, psychosocial disruption, and physical disorders that directly impact function, such as cerebrovascular accidents, hip fractures, and arthritis. Because pathology in each of these domains (i.e., cognitive, psychosocial, and physical) can independently lead to functional decline, disorders that impact all three raise a red flag in the minds of geriatric practitioners.
The American Geriatric Society (2002) defines persistent pain as a painful experience that continues for a prolonged period of time that may or may not be associated with a recognizable disease process. Given their longer life expectancy and greater likelihood of having multiple disabling health problems, older women are more likely than their male counterparts to experience pain as part of their daily lives. Because of the physical discomfort of pain, older women often develop a range of undesirable consequences, including impaired mobility, decreased socialization, depression, and sleep disturbances (Peat, Thomas, Handy, and Croft, 2004; Roberto and Reynolds, 2002). Further, less obviously related phenomena may be worsened by pain, such as gait disturbances, falls, malnutrition, slow rehabilitation, and cognitive dysfunction (Williams, 1995) more so than their male counterparts.
Neuropsychological testing is an important tool to assess cognitive performance and its impact on physical disability. However, a number of potentially confounding factors also can interfere with performance on these tasks. Among the many medical and psychiatric conditions that are known to be particularly prevalent in older persons is the experience of chronic pain. Epidemiological evidence suggests that chronic pain may be present in a range of one quarter to more than one half of persons over age 65 (American Geriatrics Society, 2002). Pain has been shown to be related to many cognitive difficulties including areas of memory and concentration, problem solving, abstract thought, and cognitive efficiency (Karp, Reynolds, Butters et al., 2006). Examples of difficulties reported by chronic pain patients include problems keeping scheduled appointments, keeping track of medications, following through on an exercise program, being able to perform previous work tasks, following conversations in interpersonal interactions, and making simple decisions in daily living.
Chronic pain in older adults is also related to sleep difficulties, increased medication usage, depression, decreased mobility, and lower self-perceived quality of health. In addition, pain and medications such as opioids used in its treatment have been reported to have adverse effects on cognition in mixed age groups (Eccleston, 1994b). Declines across the adult lifespan have been observed in both cross-sectional and longitudinal studies of the performance of tasks that require different perceptual or cognitive processes. Older adults may be at greater risk of cognitive impairment from both pain and its treatment with opioids, which may further degrade cognitive abilities that are already stressed by normal age-related cognitive decline, depression and polypharmacy, and in some cases preclinical dementia (Eccleston, 1994a).
Weiner, Rudy, Morrow et al. (2006) studied 323 older adults to examine the relationship between chronic lower back pain and neuropsychological performance (NP). All participants (pain vs. pain free) had neuropsychological testing with the Repeatable Battery for the Assessment of Neuropsychological Status, Trail Making Test, and the Grooved Pegboard Test. Pain intensity was measured using the McGill Pain Questionnaire Short Form. Investigators found that older adults with chronic lower back pain demonstrated impaired NP performance as compared with the pain-free older adults. The differences were noted on five scales including immediate memory, language, delayed memory (RBANS domain scores), mental flexibility (Trails B), and fine motor function (Grooved Pegboard). Further, pain severity was inversely correlated with NP performance, and NP function mediated the relationship between pain and physical performance.
Neuropsychological testing has often been used to evaluate outcome after surgery. In some surgeries, pain is not a significant factor, for example, in coronary artery bypass surgery or carotid endarterectomy. However, in other surgeries such as spine surgery and orthopedic surgery, pain is present both before and after surgery. Therefore, pain may interfere with cognitive test performance. The etiology behind the development of chronic pain after surgery is not fully known, but several risk factors have been identified. Preoperative pain and acute postoperative pain have been shown to increase the risk of postoperative pain. Intra-operative events, such as intra-operative nerve damage may play a role in chronic pain, as well as genetic and various psychosocial factors. Nikolajsen and colleagues (2006) examined chronic post surgical pain after orthopedic surgery. Moderate to severe pre-operative pain in the primary hip joint was indication for a total hip arthroplasty (THA). Therefore, investigators examined the prevalence of chronic pain after THA in relation to pre-operative pain and early postoperative pain. Investigators sent out a questionnaire to 1231 patients who had undergone THA 12-18 months previously. They found that 294 patients (28%) had chronic ipsilateral hip pain at the time of completion of the questionnaire, and pain limited daily activities to a moderate, severe or very severe degree in 12.1% of those patients. The chronic pain state was related to the recalled intensity of early postoperative pain and pain complaints from other sites of the body, but not the pre-operative intensity of pain. However, it is important to note that 29% – 53% of patients described moderate to severe pain preoperatively, and 15% of patients reported disabling pain postoperatively.
Clinical studies examining the relationship between pain and cognitive function in older adults have been performed in the postoperative setting. While a number of postoperative factors have been shown to contribute to cognitive dysfunction, such as delirium, infection, electrolyte disturbances, and hypoxia, multiple studies have demonstrated the importance of pain itself as a cause of postoperative cognitive dysfunction. Morrison, Magaziner, Gilbert et al. (2003) demonstrated a cause and effect relationship between pain and delirium in older surgical patients and between higher doses of morphine and improved neuropsychological performance. Dugleby and Lander (1994) suggested that pain following total hip replacement in patients aged 50-80 years was a strong predictor of mental status decline in the postoperative period. Lynch, Lazor, Gellis et al. (1998), in a study of 361 patients (mean age 67 years) undergoing elective noncardiac surgeries, found that pain was an independent risk factor for the development of delirium. Heyer, Sharma,Winfree et al et al.(2000) found that postoperative pain, not duration of surgery or dose/type of anesthetic, predicted impaired neuropsychological performance in postoperative spinal surgery patients over the age of 60 years old.
There have been several studies on the effects of postoperative pain, postoperative pain treatment and its effect on postoperative cognitive dysfunction. Two prior studies investigated the association of postoperative analgesia and development of postoperative cognitive dysfunction, but they revealed mixed results with one study showing that postoperative pain treatment had no effect on the development of change in cognitive status and another showing that the use of epidural analgesia was associated with the onset of postoperative cognitive dysfunction. Neither of these studies considered the contribution of preoperative pain to the development of postoperative cognitive dysfunction (Goldstein, Fogel, Young, 1993; 1996). Wang et al. (2007) examined the effects of postoperative pain in 225 older adults over the age of 65 after noncardiac surgery on change in cognitive status. They found 15% of patients who experienced postoperative cognitive dysfunction (POCD) reported moderate pain postoperatively compared to those who did not experience postoperative cognitive changes. They also found patients that experienced POCD were more likely to have experienced hip and knee surgery as opposed to other noncardiac surgeries. One of their suggestions for future research was to study preoperative pain experience and its effects on POCD to ascertain how physical discomfort preoperatively can be a predisposing factor to cognitive decline.