Roanna J. Hall

Delirium and long-term cognitive impairment

Delirium is a severe, acute neuropsychiatric syndrome that is highly prevalent in acute hospital populations. Delirium has noticeable effects on length of hospitalization, cost of care, mortality and morbidity. In addition to these well-established adverse consequences, there is increasing evidence linking delirium and a higher risk of long-term cognitive impairment (LTCI), including dementia. A prior review (Jackson, Gordon, Hart, Hopkins, & Ely, 2004), in which nine studies (total N = 1,885, years 1989–2003) were considered, concluded that there was evidence for an association between delirium and LTCI. Here we provide a review of studies published since Jacksons review. We included nine reports, with a total of 2,025 patients. The studies show diverse sample sizes, methodologies, designs and patient populations. However, taken together, the results of these new studies broadly confirm that there is a link between delirium and LTCI. We go on to discuss putative mechanisms and explanations. These include (1) delirium as a marker of chronic progressive pathology, but unrelated to any progression, (2) delirium as a consequence of acute brain damage which is also responsible for a ‘single hit’ or triggering of active processes causing LTCI, (3) delirium itself as a cause of LTCI, and (4) drug treatment of delirium or other conditions as a cause of LTCI. We conclude with suggestions for future research.

New horizons in the pathogenesis, assessment and management of delirium

Delirium is one of the foremost unmet medical needs in healthcare. It affects one in eight hospitalised patients and is associated with multiple adverse outcomes including increased length of stay, new institutionalisation, and considerable patient distress. Recent studies also show that delirium strongly predicts future new-onset dementia, as well as accelerating existing dementia. The importance of delirium is now increasingly being recognised, with a growing research base, new professional international organisations, increased interest from policymakers, and greater prominence of delirium in educational and audit programmes. Nevertheless, the field faces several complex research and clinical challenges. In this article we focus on selected areas of recent progress and/or uncertainty in delirium research and practice. (i) Pathogenesis: recent studies in animal models using peripheral inflammatory stimuli have begun to suggest mechanisms underlying the delirium syndrome as well as its link with dementia. A growing body of blood and cerebrospinal fluid studies in humans have implicated inflammatory and stress mediators. (ii) Prevention: delirium prevention is effective in the context of research studies, but there are several unresolved issues, including what components should be included, the role of prophylactic drugs, and the overlap with general best care for hospitalised older people. (iii) Assessment: though there are several instruments for delirium screening and assessment, detection rates remain dismal. There are no clear solutions but routine screening embedded into clinical practice, and the development of new rapid screening instruments, offer potential. (iv) Management: studies are difficult given the heterogeneity of delirium and currently expert and comprehensive clinical care remains the main recommendation. Future studies may address the role of drugs for specific elements of delirium. In summary, though facing many challenges, the field continues to make progress, with several promising lines of enquiry and an expanding base of interest among researchers, clinicians and policymakers.

Abnormal level of arousal as a predictor of delirium and inattention: An exploratory study

OBJECTIVE Abnormal level of arousal (LoA) and inattention are key features of delirium. However, the extent to which abnormal LoA alone might predict delirium and inattention is unclear. Here we tested the hypotheses that (1) patients with abnormal LoA have delirium, and (2) abnormal LoA is associated with worse performance on tests of attention. METHODS Thirty acute hip fracture patients aged 64-97 years underwent assessments of LoA, delirium status, and attentional functioning in the 24 hours before surgery and at 2-4 and 7-10 days after surgery. The Observational Scale of Level of Arousal (OSLA) and the Richmond Agitation-Sedation Scale (RASS) were used to assess LoA. Sustained attention was measured with the Edinburgh Delirium Test Box. Delirium was assessed with the Confusion Assessment Method and the Delirium Rating Scale-Revised-98. RESULTS Ten patients (33%) were diagnosed with delirium. Abnormal LoA as measured by the OSLA was strongly associated with the presence of delirium. The area under the receiver operating characteristic curve was 0.89 (95% confidence interval: 0.81-0.97), with a sensitivity of 0.87 and a specificity of 0.81. Area under the curve, sensitivity, and specificity for the RASS were 0.81 (95% confidence interval: 0.68-0.94), 0.80, and 0.79, respectively. Abnormal LoA was associated with worse attentional deficits preoperatively and at postoperative days 2-4 (p <0.01). CONCLUSION These exploratory findings suggest that abnormal LoA is a strong indicator of delirium. Also, abnormal LoA is strongly associated with inattention as measured by an objective cognitive test. These findings suggest that acute-onset abnormal LoA could be used as a trigger for delirium assessment in routine clinical practice. Future work will help to clarify further the interrelationships among abnormal LoA, inattention, and delirium.

CEREBROSPINAL FLUID INTERLEUKIN‐8 LEVELS ARE HIGHER IN PEOPLE WITH HIP FRACTURE WITH PERIOPERATIVE DELIRIUM THAN IN CONTROLS

To the Editor Delirium is often precipitated by peripheral infection or injury, but the causal pathways remain unclear. One hypothesis is that the central nervous system (CNS) changes resulting in delirium are triggered by peripheral inflammation induced by such insults.1;2 Indeed, several studies have linked altered serum inflammatory markers with delirium. Higher serum levels of interleukin (IL)-6 and IL-8 were reported in hip fracture patients with delirium compared to controls,3;4 and other studies have reported elevated inflammatory markers such as C-reactive protein and interferon gamma, and reduced anti-inflammatory markers such as insulin-like growth factor 1 and IL-1ra in delirium.5;6 With ageing, particularly where there is neurodegeneration, CNS immune cells show exaggerated production of pro-inflammatory cytokines in response to peripheral stimulation, providing a possible causal pathway from the periphery to CNS dysfunction and consequent delirium.2 This has been demonstrated in several animal studies1;2 but whether there are elevations in CNS pro-inflammatory cytokines in patients with delirium not caused by primary CNS disorders is unknown. Here we compared levels of interleukin (IL)1β, IL-6, IL-8, IL-10 and IL-12p70, and tumor necrosis factor alpha (TNF-α) in the cerebrospinal fluid (CSF) and serum of older hip fracture patients with and without peri-operative delirium. We hypothesized that cases would have higher levels of pro-inflammatory cytokines. Thirty-six patients (28 female) in two university-affiliated hospitals (Edinburgh, Scotland and Amsterdam, the Netherlands) aged 62-93 years with hip fracture and awaiting surgery were assessed for delirium before and 3-4 days after surgery. Delirium was assessed with the Confusion Assessment Method; patients with delirium at any stage were considered cases. CSF in all patients, and serum in the 16 Edinburgh patients, were obtained at the onset of spinal anesthesia as previously described.7 CSF and serum samples were spun at 1000g for 10 minutes at 4°C; supernatants were stored at −80°C. Cytokine levels were measured with a cytometric bead array immunoassay (Human Inflammatory Cytokine Kit, BD Biosciences) with a detection limit of 20.0 pg/mL; levels below this limit were considered to be zero. Case versus control comparisons were analyzed using the Mann-Whitney U-test. The study was approved by the local ethics committees. Delirium was diagnosed in 15 patients. Nine had delirium pre-operatively, and seven further patients developed delirium post-operatively; one patient who had pre-operative delirium recovered after surgery. Mean ages and Charlson Comorbidity Index scores did not differ between the cases and controls. A history of dementia (N=7) was associated with a higher incidence of delirium (p=0.008). Only IL-8 and IL-6 were detected in CSF with levels above the assay detection limit: IL-8 in 33/36 samples, and IL-6 in 3/36 samples (IL-6 levels were not further analyzed). Delirium cases (N=15) had higher CSF IL-8, with median (IQR) of 69.8 pg/mL (47.9 – 125.6) compared with 39.6 pg/mL (28.0 – 64.5) for controls (N=21; Mann-Whitney U=68, p=0.003) (Figure 1). In the 16 serum samples the following cytokines had levels above the detection limit: IL-8 (4/16 samples), IL-6 (12/16 samples) and TNF-α (2/16 samples). Serum IL-8 and TNF-α levels were not further analyzed. Delirium cases had higher serum IL-6 levels, with median (IQR) of 42.4 pg/mL (28.9 – 438.4) compared with 24.3 pg/mL (0 – 217.2) for controls (U=6; p=0.013). Exclusion of the seven patients with dementia did not change the pattern of results, with both CSF IL-8 and serum IL-6 levels remaining significantly higher in patients with delirium. Dementia was not associated with higher levels of CSF IL-8 or serum IL-6 (p>0.10). Figure 1 ‘Cerebrospinal fluid interleukin-8 levels in patients with and without peri-operative delirium’ The present study suggests the possibility that delirium is associated with increased CNS production of IL-8. Our findings are also consistent with some other studies which have found higher serum IL-6 levels in delirium. IL-8 and its rodent homolog CXCL1 (or CINC-1) are produced systemically but also by astrocytes, microglia, brain endothelial cells and infiltrated neutrophils in response to peripheral injury or infection.8 IL-8 and CXCL1 have generally pro-inflammatory actions.8 Though peripheral CXCL1 can cross the blood brain barrier, synthesis by brain endothelium in response to systemic inflammation is significant.9 CNS production of IL-8 is supported by another recent study of hip fracture patients, in which mental status was not assessed, where median CSF IL-8 levels were 63 pg/mL (range 40-115) versus median serum IL-8 levels of 0 pg/mL (range 0-78).10 Taken together, these findings suggest that CNS production of IL-8 induced by peripheral inflammatory stimuli may be linked with delirium, but further research is required. Future studies should confirm these results in larger samples and other populations, employ more sensitive assays, examine relationships with other putative CSF biomarkers of delirium,7 and investigate the role of other cytokines.

A Systematic Literature Review of Cerebrospinal Fluid Biomarkers in Delirium

Background: Cerebrospinal fluid (CSF) analysis has great potential to advance understanding of delirium pathophysiology. Methods: A systematic literature review of CSF studies of DSM or ICD delirium was performed. Results: In 8 studies of 235 patients, delirium was associated with: elevated serotonin metabolites, interleukin-8, cortisol, lactate and protein, and reduced somatostatin, β-endorphin and neuron-specific enolase. Elevated acetylcholinesterase predicted poor outcome after delirium and higher dopamine metabolites were associated with psychotic features. Conclusions: No clear conclusions emerged, but the current literature suggests multiple areas for further investigation with more detailed studies.

Cerebrospinal fluid markers of neuroinflammation in delirium: A role for interleukin-1β in delirium after hip fracture

Objective Exaggerated central nervous system (CNS) inflammatory responses to peripheral stressors may be implicated in delirium. This study hypothesised that the IL-1β family is involved in delirium, predicting increased levels of interleukin-1β (IL-1β) and decreased IL-1 receptor antagonist (IL-1ra) in the cerebrospinal fluid (CSF) of elderly patients with acute hip fracture. We also hypothesised that Glial Fibrillary Acidic Protein (GFAP) and interferon-γ (IFN-γ) would be increased, and insulin-like growth factor 1 (IGF-1) would be decreased. Methods Participants with acute hip fracture aged > 60 (N = 43) were assessed for delirium before and 3–4 days after surgery. CSF samples were taken at induction of spinal anaesthesia. Enzyme-linked immunosorbent assays (ELISA) were used for protein concentrations. Results Prevalent delirium was diagnosed in eight patients and incident delirium in 17 patients. CSF IL-1β was higher in patients with incident delirium compared to never delirium (incident delirium 1.74 pg/ml (1.02–1.74) vs. prevalent 0.84 pg/ml (0.49–1.57) vs. never 0.66 pg/ml (0–1.02), Kruskal–Wallis p = 0.03). CSF:serum IL-1β ratios were higher in delirious than non-delirious patients. CSF IL-1ra was higher in prevalent delirium compared to incident delirium (prevalent delirium 70.75 pg/ml (65.63–73.01) vs. incident 31.06 pg/ml (28.12–35.15) vs. never 33.98 pg/ml (28.71–43.28), Kruskal–Wallis p = 0.04). GFAP was not increased in delirium. IFN-γ and IGF-1 were below the detection limit in CSF. Conclusion This study provides novel evidence of CNS inflammation involving the IL-1β family in delirium and suggests a rise in CSF IL-1β early in delirium pathogenesis. Future larger CSF studies should examine the role of CNS inflammation in delirium and its sequelae.

Predictors and correlates of edentulism in healthy older people.

PURPOSE OF REVIEW To review peer-reviewed, original research studies published in 2008-2009 that present data relating to the predictors and correlates of edentulism and tooth loss in older adults. RECENT FINDINGS Edentulism rates vary markedly between countries and between urban and rural settings within countries. Rates are generally falling over time, but this reduction largely reflects a cohort effect on tooth loss in childhood and young adulthood. Socioeconomic factors, along with accompanying lifestyles and health behaviours remain strong predictors of edentulism, many of these factors relate to peak prior intelligence. Immunological mechanisms of tooth loss are becoming elucidated. Edentulism, itself, predicts mortality and correlates with a wide range of health outcomes, but these, in turn, also correlate with predictors of tooth loss such as peak prior intelligence. Edentulism correlates separately from these lifelong traits with measures of self-esteem and quality of life. SUMMARY Edentulism is important as a correlate of self-esteem and quality of life in older adults. It is also a useful marker of socioeconomic status earlier in life.

Delirium detection and monitoring outside the ICU

Delirium affects many patients in hospital settings but is under-detected and associated with a range of adverse health-care outcomes, including institutionalisation and elevated mortality. Detection is essential because it leads to identification and management of precipitants and assessment and management of distress caused by hallucinations and delusions. Moreover, delirium may affect communication and, thus, assessment of pain. This is important because inadequate analgesia may cause agitation and prolong the delirium. Here, we provide an overview of the main features of delirium. Informal and formal methods of assessment of the features are covered. We describe some of the main rating scales used in delirium screening and severity grading. Incorporating formal and systematic screening and assessment into everyday clinical practice can substantially improve delirium diagnosis and treatment.

Who understands delirium

This New York Times article on Pulitzer Prize-winning historian Justin Kaplan’s experience of delirium [1] provided welcome publicity of this enormously impactful but historically neglected syndrome. Nevertheless, the remarkable scarcity of such exposure in the mass media speaks to what is perhaps the most significant obstacle to progress in delirium research and practice: its invisibility. Why has delirium remained so obscure? There would appear to be several reasons. Patients rarely speak of their experiences of delirium, perhaps because of embarrassment or bewilderment; currently, there are no specific charities or patient advocacy groups. Healthcare staff use a wide variety of informal words and phrases to describe delirium. This diagnostic ambivalence and imprecision greatly hinders the implementation of formal methods of improving care. Another likely factor is that delirium is very challenging: it is heterogeneous in its precipitants, symptomatology, severity and course. The mixture of mental status abnormalities and complex medicine means that healthcare professionals require multiple skills to manage delirium. Indeed, these skills may be lacking: a recent survey of UK junior doctors suggested that there are serious deficiencies in knowledge of the diagnostic criteria [2]. However, there have been major advances over the last three or four decades that ought to be more widely disseminated [3]. We know that delirium is common and that it has several adverse consequences, including loss of independence, acceleration of dementia and death [3–5]. The clinical predisposing and precipitating factors are now well documented [3]. Recent work has shown that delirium persists for months in around 20% of cases [6]. The significance of delirium in critically ill patients and in palliative care settings is now far clearer [7–9]. Crucially, it is now established that multi-component prevention strategies are effective, and should be introduced into standard health-care practice [3]. What might be the current priorities for research? Direct brain insults such as oxygen deprivation, hyponatraemia and adverse drug effects are common causes of delirium, and in many such cases, the routes to disruption of cognitive processes are clear. In contrast, the mechanisms by which peripheral illness such as urinary tract infection can cause acute, severe mental deterioration over periods of as little as a few hours are poorly understood. Recent experiments in animal models and humans have found that in older individuals peripheral stimuli can induce adverse inflammatory and stress system responses in the central nervous system, with resultant maladaptive behavioural change including delirium [10–12]. Delirium could thus represent an exaggeration of the sickness behaviour syndrome (a set of normally adaptive behavioural responses to illness including sleepiness, fatigue, anhedonia and impaired concentration), but may also simply be due to the effects of pathologically sustained high cortisol levels [10]. These leads in dissecting the ‘indirect’ causes of delirium may suggest potential new treatments. The role of stress mediators is also relevant to the patient experience of delirium: studies in animal models suggest that chronic unpredictable psychological stress over periods of days can cause marked brain damage [13]. Patients with delirium are often in a comparable position, suffering pain, fear, an unpredictable environment and strangers undertaking invasive activities. All of this is exacerbated by the inability of the patient’s mind to fully comprehend what is happening to them. The implications here are that effective psychological care of patients with delirium may be particularly important, not simply to relieve distress, but to protect the brain. Detection is a pre-requisite for good care, but the problem of achieving acceptable levels of recognition of delirium has not yet been solved. This partly reflects the lack of validated brief screening tools that do not require special training and that discriminate delirium from dementia [14, 15]. Another need is for methods of recording the status of conscious but untestable patients, that is, those too drowsy to undergo cognitive testing or even a brief interview. Such patients are overwhelmingly likely to have a diagnosis of hypoactive delirium, but are often left without any label and resultant management plan. More broadly, further research on the neuropsychology of delirium is essential to the development of better testing instruments for screening and diagnosis in clinical practice and in research. There are many challenges here, not least because of the range of mental status change, from stupor to relatively mild deficits in attention. Subsyndromal delirium, where there are one or more features of delirium but in which full DSM-IV criteria are not met, has recently been shown to have prognostic significance [16]. Despite this, its neuropsychology is all but unexplored. The clinical role of other methods of assessment, such as neuroimaging, plasma biomarkers and lumbar puncture, also remain greatly under-researched. A recent systematic review of the literature on neuroimaging findings in delirium found that though there may be some relationships between white

Predictors and correlates of edentulism in the healthy old people in Edinburgh (HOPE) study

OBJECTIVES To determine the extent to which correlates of edentulism are explained by an association between tooth loss and cognitive ability. METHODS Participants in the Healthy Old People in Edinburgh (HOPE) study aged 70 or more at baseline were assessed and health, cognitive, socio-economic and socio-environmental data collected on four consecutive occasions. It was noted whether the participant had any retained teeth and if not, the age when the last tooth was lost. Prior determinants of edentulism were investigated with binary logistic regression models. At the 9-year follow-up, associations with edentulism were examined using general linear models with edentulism as an independent factor. RESULTS 201 participants were adequately tested, of whom 104 (51.7%) were edentulous. A logistic regression model that considered age, sex, education, social class, deprivation index of residence, objective distance from dentist, participants estimate of distance from dentist and NART-estimated IQ (NARTIQ) found age (p = 0.032), occupational class (p = 0.019) and NARTIQ (p = 0.027) as significant predictors of edentulism. Coxs proportional hazards modelling found only NARTIQ (p = 0.050) to be correlated. Being edentulous was associated with poorer respiratory function but not hand grip strength (p = 0.23). Edentulous participants had lower self esteem scores (p = 0.020) and poorer dietary assessment scores (p = 0.028). Being edentulous was also associated with significantly lower mean scores on all cognitive testing, although these associations became non-significant after adjustment for NARTIQ and age. CONCLUSIONS In healthy older people, edentulism is associated with relative impairment of cognitive ability, although this association is explained by the fact that lower original intelligence predisposes to edentulism and poorer performance on cognitive tests in old age. Once original intelligence is adjusted for, tooth loss is not related to cognitive ability. Tooth loss is, however, associated with poorer status across a wide range of health measures: physical health, nutrition, disability and self-esteem. Establishing the degree to which these health outcomes are causally related to edentulism could usefully be factored into cost-benefit analyses of programmes designed to prevent tooth loss.