1. Översiktlig projektbeskrivning

Engelsk titel

Novel simple indicators for cognitive impairment after stroke

Sammanfattning av projektet

Background – Cognitive impairment is an important but under-recognised consequence of stroke. We investigated whether a subset of items from the NIH Stroke Scale (NIHSS) could yield valid information on cognitive status in a group of stroke patients. Methods – 149 stroke patients from the Gothenburg 70+ Stroke Study were investigated after 18 months. We extracted four items corresponding to the NIHSS items on orientation, executive function, language and inattention. Scores on this subset of 4 NIHSS items (Cog-4) and the Mini-Mental Status Examination (MMSE) were evaluated against a reference diagnosis of severe cognitive impairment. Results – The area under the receiver-operator curve (AUC) plotted for the Cog-4 scale against the diagnosis of severe cognitive impairment was 0.78; the MMSE had slightly better diagnostic precision, with an AUC of 0.84. Making the executive task more difficult increased the precision of the Cog-4, raising the AUC to 0.81. Conclusions – A composite score based on four NIHSS items was almost as good as the MMSE in determining severe cognitive impairment. Ideally, dedicated measures of cognition should be employed as a matter of course after stroke, but in their absence the Cog-4 subscale provides an indication of cognitive functioning.

Typ av projekt

Forskningsprojekt

MeSH-termer för att beskriva ämnesområdet

Inlagda MeSH-termer

Stroke

A group of pathological conditions characterized by sudden, non-convulsive loss of neurological function due to BRAIN ISCHEMIA or INTRACRANIAL HEMORRHAGES. Stroke is classified by the type of tissue NECROSIS, such as the anatomic location, vasculature involved, etiology, age of the affected individual, and hemorrhagic vs. non-hemorrhagic nature. (From Adams et al., Principles of Neurology, 6th ed, pp777-810)

Cognition

Intellectual or mental process whereby an organism becomes aware of or obtains knowledge.

Delirium, Dementia, Amnestic, Cognitive Disorders

Cognitive disorders including delirium, dementia, and other cognitive disorders. These may be the result of substance use, trauma, or other causes.

Rehabilitation

Restoration of human functions to the maximum degree possible in a person or persons suffering from disease or injury.

Psychometrics

Assessment of psychological variables by the application of mathematical procedures.

Projektets delaktighet i utbildning

Avhandling
D-uppsats / Magisterexamen
C-uppsats / Kandidatexamen
ST-läkarutbildning
Annan utbildning
Ej del i utbildning

3. Processen och projektets redovisning

Pågående aktiviteter

Planering och förberedelse före datainsamling
Datainsamling pågår
Analys av insamlade data pågår
Författande av skriftlig redovisning / publikation pågår
En eller flera publikationer från projektet är publicerade
Slutfört och inget mer görs inom ramen för detta projekt

Projektstart (när planeringen påbörjas och börjar dokumenteras skriftligt)

2010-10-02

Publikationer från detta projekt

1. The NIH stroke scale can establish cognitive function after stroke.
   Cumming Toby B, Blomstrand Christian, Bernhardt Julie, Linden Thomas.
   Cerebrovasc Dis. 2010:30(1):7-14.[Source: PubMed®]
   [Links: PMID: 20424439 | DOI länk]
2. Establishing cognitive function after stroke: Can the NIH stroke scale be used?
   Lindén Thomas, Cummings Toby, Bernhardt Julie, Blomstrand Christian.
   Sydney, Australia: 19th Australasian Stroke Society Conference; 2008. [Source: User]

Tillämpning av resultat - tidsaspekt (projektledarens bedömning)

Resultaten kommer sannolikt att tillämpas inom 5 år från projektslut.

Tillämpning av resultat - genomslag (projektledarens bedömning)

Internationellt (i flera länder)

Tillämpning av resultat - beskrivning

Använding i studier på strokepatienter som nu kan redovisa kognitiv funktion hos patienter i en studie även om den från början inte var designad för att göra det.

4. Detaljerad projektbeskrivning

Bakgrundsbeskrivning

The execution of cognitive tasks (e.g., perceiving, attending, organising thoughts, remembering, learning, communicating) is often compromised in stroke patients. Marked cognitive impairments have been documented from the acute phase of stroke [1] up to 3 months [2] and out to 2 years post-stroke [3,4]. Not only are cognitive deficits after stroke common, they are strongly linked to negative outcomes. Cognitive impairment has been independently associated with death and disability at 4 years post-stroke [5]. Severity of cognitive impairment was found to be an independent predictor of institutionalisation in the 3 years following stroke, having more importance than severity of physical disability [6]. Furthermore, cognitive impairment has been related to incomplete recovery in activities of daily living [7], poorer inpatient rehabilitation outcome [8], post-stroke depression [9], lower quality of life [10] and higher health care costs [11].

Awareness of the importance of post-stroke cognitive impairments, however, is typically low. Problems often remain unidentified in the clinic with the result that support to patients and their families is neglected. In addition, many stroke research studies that are otherwise well-designed and would provide excellent platforms for investigating the nature of cognitive problems do not include any evaluation of cognition. This may be due to lack of awareness of cognition’s importance, disbelief in the ability of an intervention to affect cognitive function or simply logistical reasons such as lack of qualified assessors or concerns over including time-consuming and costly assessment procedures. Undoubtedly, part of the reason for the reluctance is due to the lack of appropriate assessment tools.

In those stroke studies that do employ a screening tool for cognitive function, the Mini-Mental State Examination (MMSE) remains the most widely used measure [5]. There is now convincing evidence that the MMSE is not an ideal screen for cognitive impairment after stroke [12-14]. It is weighted heavily towards language and memory ability, and away from frontal and executive functions. More extended cognitive assessment is clearly beneficial. Desmond et al. [15] reported that using a neuropsychological battery to diagnose dementia was superior to using either the MMSE or clinical judgement. Recent studies have employed similar testing batteries to evaluate cognitive impairment following stroke [1,4]. It is not practical, however, for many stroke studies or intervention trials to incorporate 60-90 minutes of cognitive testing. There is a need for cognitive assessment instruments that are applicable to stroke and are easy and fast to administer. Several instruments have been proposed for this purpose, such as the BNI Screen for Higher Cerebral Functions [16]. Recent ‘harmonisation standards’ were developed that not only proposed a 60-minute testing protocol but also shorter 30-minute and 5-minute variants [17]. So far, however, even these short assessments are not routinely included in stroke studies or in clinical stroke services.

One assessment that is almost routinely completed in stroke research is evaluation of impairment at a range of time points using the NIH Stroke Scale (NIHSS) [18]. Given that several of the NIHSS items relate to cognition, it is possible that we may be able to extract some information on cognitive function in stroke studies that fail to include any dedicated cognitive assessments but that do incorporate the NIHSS. Other researchers have attempted to isolate a subset of items from the NIHSS that can best predict physical outcome after stroke [19].

Syfte

The aim of the current study is to determine whether a subset of NIHSS items can yield valid information on cognitive status in groups of stroke patients.

Frågeställning / Hypoteser

A subset of the NIHSS can with clinically relevant accuracy indicate post-stroke cognitive impairment

Metod: Databearbetning

We assessed the distribution of test results corresponding to individual NIHSS items in patients with and without severe cognitive impairment. The indicative value of the presence of deficit on any of the 4 NIHSS items was assessed by calculating kappa. A score between 0 and 9 was calculated for the 4 relevant NIHSS items for each patient. These scores were compared against the reference diagnosis of severe cognitive impairment, with receiver operating curves plotted and area under the curve (AUC) reported. For comparison purposes, the predictive value of each of the 4 NIHSS items in isolation as well as different combinations of 2 or 3 of the 4 NIHSS items were plotted against the reference and the AUC reported. The predictive validity of the MMSE was also assessed against the reference diagnosis of severe cognitive impairment. Finally, scores on the 3-, 4- and 5-step command task and the star-cancellation task were used to investigate whether more detailed assessments could improve the diagnostic validity of the Cog-4 subscale of the NIHSS.

Resultat

The majority of patients had data for each item: 96% (N=142) for the orientation item, 91% (N=135) for the executive function item, 95% (N=141) for the language item and 91% (N=135) for the attention item. A deficit on any of the NIHSS cognitive items showed a moderate association with severe cognitive impairment (kappa = 0.55). Of 142 patients, 110 were correctly classified and 32 misclassified; seventeen as false positives and 15 as false negatives (see Table 3). The positive predictive value was 74% and the negative predictive value was 81%.

One-hundred and nineteen patients had data for each item that allowed calculation of a full NIHSS Cog-4 subscale score. Only two patients scored above 5 on the Cog-4, even though about 40% of patients were classified as severely cognitively impaired. The area under the receiver-operator curve (AUC) plotted for the NIHSS Cog-4 score against the reference diagnosis of severe cognitive impairment was 0.78. Using MMSE scores gave slightly better diagnostic precision, with an AUC of 0.84 when plotted against the reference diagnosis.

Omitting the executive variable gave a very similar AUC value – not surprising as almost all patients successfully completed the NIHSS executive item that assesses the patient’s ability to perform a one-step and a two-step command. Even of the severely cognitively impaired patients, 48/55 (87%) completed the two-step command that is included in the NIHSS. In contrast, when the command consisted of four or more steps, it was completed by only 34/55 (62%) patients with severe cognitive impairment as compared to 79/80 (99%) patients without severe cognitive impairment. A model incorporating this more discriminative cut-off value of a four-step command increased the diagnostic value of the NIHSS Cog-4, as indicated by an AUC value of 0.81.

Changing the cut-off value for inattention using results on the Star-Cancellation Test did not, however, strengthen the model.

Diskussion

The central finding of this study was that a composite score based on four NIHSS items had almost the same diagnostic value as the MMSE in predicting severe cognitive impairment (AUC of 0.78 and 0.84 respectively). This result may open the way for obtaining an indication of cognitive functioning in stroke patients in studies that were not originally designed to investigate cognition. Evaluating cognition is important, as most stroke patients will suffer from cognitive impairments [3,24]. Not only is cognitive impairment a major adverse outcome after stroke in itself, it is also associated with incomplete recovery in activities of daily living [7], poorer inpatient rehabilitation outcome [8], post-stroke depression [9], lower quality of life [10] and higher health care costs [11]. The NIHSS has rapidly become a de facto standard in assessing stroke severity in the acute stage, both in research and in clinical practice, and is used for evaluation of stroke patients prior to administering thrombolysis.

Calculation of the Cog-4 composite score was kept as simple as possible: standard NIHSS scoring was applied and scores on the four items were added to derive a zero to nine NIHSS short-form cognitive score. The poorest of the four items in terms of diagnostic value was the executive item, which consisted of a one- and two-step command. Results indicated that the AUC remained at 0.78 when this item was omitted from the model. The problem was that the one- and two-step commands were too easy, even for patients with severe cognitive impairment. Making the task more difficult by extending the number of steps in the command up to five improved the diagnostic precision of this item and of the Cog-4 composite score (AUC increased from 0.78 to 0.81). This simple addition has the potential to increase the usefulness of the NIHSS as a long term outcome measure, a purpose for which it is poorly suited in its current form.

Hemispatial neglect is a cognitive impairment that differs from the others. It is strongly associated with right-sided cortical strokes, present in about 15% of stroke patients and is related to overall cognitive function [25]. A factor analysis (data not shown) indicated that the inattention item was representative of a different underlying explanatory factor from the other three items. This is consistent with a previous factor analysis of the NIHSS, which found that the items on orientation, executive function and language loaded onto a ‘left brain’ factor whereas inattention loaded onto a ‘right brain’ factor [26]. Deleting the inattention item from the algorithm reduced diagnostic precision, so it was retained in the NIHSS Cog-4 subscale.

It must be noted that we are not arguing for the NIHSS Cog-4 as a miracle indicator of cognitive function. Performing almost as well as the MMSE may be considered unimpressive given the shortcomings that have been demonstrated in using the MMSE to screen for cognitive impairment after stroke [12-14]. While an AUC of 0.78 is far from perfect, it is comparable with the finding of Tirschwell et al. [19], who reported an AUC value of 0.76 for a 5-item NIHSS physical subscale against 3-month functional outcome. It is generally considered that an AUC of 0.70 is fair, 0.80 is good and 0.90 is excellent in terms of diagnostic value. Any non-perfect diagnostic instrument will imply some degree of misclassification. In fact, an instrument with an AUC value of 0.90 would still imply just below 20% of cases misclassified. Using a test like the MMSE (with an AUC value of 0.84) would increase misclassification to 24% and employing a scale like the Cog-4 would increase it to just below 30%. Of course, the ideal situation would be better assessment of cognitive impairment in all stroke patients, both in research trials and in the clinic. Valid assessment methods are available, even in situations where there are constraints on time and resources. For example, a North American committee for standardisation in assessment of vascular cognitive impairment recently proposed a simple five-minute assessment that could be employed in the absence of any other cognitive measures [17]. While these and other more detailed assessments should be used wherever possible, extracting information on cognitive status from the widely used NIHSS provides an option that could be used in the absence of any dedicated cognitive assessment. Despite our proposed value in assessing the frequency of cognitive impairment in groups of stroke patients, the Cog-4 does not, however, offer enough discriminatory value to ever become a diagnostic tool in individual patients.

Some limitations of the study must be mentioned. First, the NIHSS was designed to be a measure of stroke severity in acute stroke patients. Our study was undertaken at 1½ years post-stroke, and our results would benefit from validation in an acute stroke cohort. It is unlikely that the one- and two-step command task, for example, would be successfully completed by so many patients in the acute phase of stroke, when effects of disturbed wakefulness and attention are prominent. The setting might also be important. The data reported here were collected as part of a research study and data collected in a busy clinical environment may be different. Also, in order to specifically study the elderly stroke population we included only patients 70 years of age or older, counteracting the tendency of stroke studies to study patients of lower age than we see in the clinic. The diagnostic profile may have been different in younger patients. Further, the study base reflects the underlying population of the geographical area of study and the results should be extrapolated with caution to populations with different ethnic and cultural backgrounds.

Nevertheless, we believe the NIHSS Cog-4 subscale proposed here can be of value in cases when information on cognitive status in a group of stroke patients is sought but no specific assessment is available. This information can be obtained using a measure that in many cases is already at hand and with a diagnostic precision just a little less than that of the MMSE.

Slutsats

Although we advocate the use of dedicated cognitive assessment batteries for obtaining information on these important functions, we argue that the NIHSS Cog-4 subscale is a useful addition to the existing tools for establishing cognitive function in patients who have suffered a stroke.

Referenser

1.Nys GM, van Zandvoort MJ, de Kort PL, Jansen BP, de Haan EH, Kappelle LJ. Cognitive disorders in acute stroke: Prevalence and clinical determinants. Cerebrovasc Dis 2007;23:408-416.
2.Srikanth VK, Thrift AG, Saling MM, Anderson JF, Dewey HM, Macdonell RA, Donnan GA. Increased risk of cognitive impairment 3 months after mild to moderate first-ever stroke: A community-based prospective study of nonaphasic English-speaking survivors. Stroke 2003;34:1136-1143.
3.Linden T, Skoog I, Fagerberg B, Steen B, Blomstrand C. Cognitive impairment and dementia 20 months after stroke. Neuroepidemiology 2004;23:45-52.
4.Srikanth VK, Quinn SJ, Donnan GA, Saling MM, Thrift AG. Long-term cognitive transitions, rates of cognitive change, and predictors of incident dementia in a population-based first-ever stroke cohort. Stroke 2006;37:2479-2483.
5.Patel MD, Coshall C, Rudd AG, Wolfe CD. Cognitive impairment after stroke: Clinical determinants and its associations with long-term stroke outcomes. J Am Geriatr Soc 2002;50:700-706.
6.Pasquini M, Leys D, Rousseaux M, Pasquier F, Henon H. Influence of cognitive impairment on the institutionalisation rate 3 years after a stroke. J Neurol Neurosurg Psychiatry 2007;78:56-59.
7.Nys GM, van Zandvoort MJ, de Kort PL, van der Worp HB, Jansen BP, Algra A, de Haan EH, Kappelle LJ. The prognostic value of domain-specific cognitive abilities in acute first-ever stroke. Neurology 2005;64:821-827.
8.Paolucci S, Antonucci G, Gialloreti LE, Traballesi M, Lubich S, Pratesi L, Palombi L. Predicting stroke inpatient rehabilitation outcome: The prominent role of neuropsychological disorders. Eur Neurol 1996;36:385-390.
9.Kauhanen M, Korpelainen JT, Hiltunen P, Brusin E, Mononen H, Maatta R, Nieminen P, Sotaniemi KA, Myllyla VV. Poststroke depression correlates with cognitive impairment and neurological deficits. Stroke 1999;30:1875-1880.
10.Hochstenbach JB, Anderson PG, van Limbeek J, Mulder TT. Is there a relation between neuropsychologic variables and quality of life after stroke? Arch Phys Med Rehabil 2001;82:1360-1366.
11.Claesson L, Linden T, Skoog I, Blomstrand C. Cognitive impairment after stroke – Impact on activities of daily living and costs of care for elderly people. The Göteborg 70+ Stroke Study. Cerebrovasc Dis 2005;19:102-109.
12.Blake H, McKinney M, Treece K, Lee E, Lincoln NB. An evaluation of screening measures for cognitive impairment after stroke. Age Ageing 2002;31:451-456.
13.Nys GM, van Zandvoort MJ, de Kort PL, Jansen BP, Kappelle LJ, de Haan EH. Restrictions of the Mini-Mental State Examination in acute stroke. Arch Clin Neuropsychol 2005;20:623-629.
14.Srikanth V, Thrift AG, Fryer JL, Saling MM, Dewey HM, Sturm JW, Donnan GA. The validity of brief screening cognitive instruments in the diagnosis of cognitive impairment and dementia after first-ever stroke. Int Psychogeriatr 2006;18:295-305.
15.Desmond DW, Moroney JT, Bagiella E, Sano M, Stern Y. Dementia as a predictor of adverse outcomes following stroke: An evaluation of diagnostic methods. Stroke 1998;29:69-74.
16.Prigatano GP, Amin K, Rosenstein LD. Manual for the BNI Screen for Higher Cerebral Functions. Phoenix, Arizona: Barrow Neurological Institute, 1995.
17.Hachinski V, Iadecola C, Petersen RC, Breteler MM, Nyenhuis DL, Black SE, Powers WJ, DeCarli C, Merino JG, Kalaria RN, Vinters HV, Holtzman DM, Rosenberg GA, Wallin A, Dichgans M, Marler JR, Leblanc GG. National Institute of Neurological Disorders and Stroke-Canadian Stroke Network vascular cognitive impairment harmonization standards. Stroke 2006;37:2220-2241.
18.Brott T, Adams HP, Olinger CP, Marler JR, Barson WG, Biller J, Spilker J, Holleran R, Eberle R, Hertzberg V, Rorick M, Moomaw CJ, Walker M. Measurements of acute cerebral infarction: A clinical examination scale. Stroke 1989;20:864-870.
19.Tirschwell DL, Longstreth WT, Jr., Becker KJ, Gammans RE, Sr., Sabounjian LA, Hamilton S, Morgenstern LB. Shortening the NIH Stroke scale for use in the prehospital setting. Stroke 2002;33:2801-2806.
20.Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R). Washington: American Psychiatric Association, 1987.
21.Bowler JV, Hachinski V. Vascular cognitive impairment: A new approach to vascular dementia. Baillieres Clin Neurol 1995;4:357-376.
22.Folstein MF, Folstein SE, McHugh PR. Mini-mental state: A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189-198.
23.Halligan PW, Marshall JC, Wade DT. Visuospatial neglect: Underlying factors and test sensitivity. Lancet 1989;2:908-911.
24.Leys D, Henon H, Mackowiak-Cordoliani MA, Pasquier F. Poststroke dementia. Lancet Neurol 2005;4:752-759.
25.Linden T, Samuelsson H, Skoog I, Blomstrand C. Visual neglect and cognitive impairment in elderly patients late after stroke. Acta Neurol Scand 2005;111:163-168.
26.Lyden P, Lu M, Jackson C, Marler J, Kothari R, Brott T, Zivin J. Underlying structure of the National Institutes of Health Stroke Scale: Results of a factor analysis. NINDS tPA Stroke Trial Investigators. Stroke 1999;30:2347-54.