Brain Magnetic Resonance Imaging Correlates of Impaired Cognition in Patients With Type 2 Diabetes
http://www.100md.com
糖尿病学杂志 2006年第4期
1 Department of Neurology, Rudolf Magnus Institute of Neuroscience, University Medical Center, Utrecht, the Netherlands
2 Helmholtz Research Institute, University Medical Center, Utrecht, the Netherlands
3 Department of Radiology, University Medical Center, Utrecht, the Netherlands
4 Julius Center for Health Sciences and Primary Care, University Medical Center, Utrecht, the Netherlands
BCR, bicaudate ratio; BFR, bifrontal ratio; DWML, deep WML; FFR, frontal interhemispheric fissure ratio; FLAIR, fluid-attenuating inverse recovery; MRI, magnetic resonance imaging; PWML, periventricular WML; SFR, Sylvian fissure ratio; WML, white matter lesion
ABSTRACT
The structural correlates of impaired cognition in type 2 diabetes are unclear. The present study compared cognition and brain magnetic resonance imaging (MRI) between type 2 diabetic patients and nondiabetic control subjects and assessed the relationship between cognition and MRI findings and blood pressure and metabolic control. The study included 113 patients and 51 control subjects. Brain MRI scans were rated for white matter lesions (WMLs), cortical and subcortical atrophy, and infarcts. Neuropsychological test scores were divided into five cognitive domains and expressed as standardized Z values. Type 2 diabetes was associated with deep WMLs (P = 0.02), cortical (P < 0.001) and subcortical (P < 0.05) atrophy, (silent) infarcts (P = 0.06), and impaired cognitive performance (attention and executive function, information-processing speed, and memory, all P < 0.05). Adjustment for hypertension did not affect the results. Within the type 2 diabetic group, cognitive function was inversely related with WMLs, atrophy, and the presence of infarcts (adjusted for age, sex, and estimated IQ), and there was a modest association with HbA1c and diabetes duration. This association was strongest for age, even more so than in control subjects. We conclude that cognitive impairments in patients with type 2 diabetes are not only associated with subcortical ischemic changes in the brain, but also with increased brain atrophy.
Type 2 diabetes can affect the central nervous system (1). Neuropsychological studies reported moderate degrees of cognitive impairment, particularly in tasks involving verbal memory or complex information processing (2,3). Epidemiological studies demonstrated an association between diabetes and dementia (4,5). It is not clear which factors mediate accelerated cognitive decline in patients with type 2 diabetes. Both comorbid conditions (e.g., hypertension and depression) and diabetes-specific factors (e.g., glycemic control) have been implicated (2,6). Some investigators have suggested that hypertension is an important mediator (2,7,8), but this was not supported by others (9,10).
The structural correlates and pathophysiological mechanisms underlying these cognitive impairments are still uncertain. Previous studies report that modest cortical and subcortical atrophy and symptomatic or asymptomatic infarcts are more common in type 2 diabetic patients than in control subjects (11eC15). Findings from studies on so-called WMLs are less consistent; some report an association with type 2 diabetes (16), but others report no statistically significant effects (11,17). To the best of our knowledge, there are no published studies that combine detailed assessment of cognitive functioning and magnetic resonance imaging (MRI) of the brain in patients with type 2 diabetes.
The present study aimed to compare cognition and brain MRI in type 2 diabetic patients and nondiabetic control subjects and to relate cognitive functioning in the type 2 diabetic patients to MRI findings, as well as to blood pressure and metabolic control.
RESEARCH DESIGN AND METHODS
The UDES (Utrecht Diabetic Encephalopathy Study) is a cross-sectional population-based study on determinants of impaired cognition in type 2 diabetes. Because the study aimed to identify potential risk factors for cognitive impairment in type 2 diabetes, patients were not selected for the presence or absence of diabetes complications, comorbid conditions (e.g., hypertension), or exposure to other risk factors (e.g., smoking). For inclusion, type 2 diabetic patients had to be 55eC80 years of age, have a diabetes duration of at least 1 year, be functionally independent, and speak Dutch. Exclusion criteria for all participants included a psychiatric or neurological disorder (unrelated to diabetes) that could influence cognitive functioning, history of alcohol or substance abuse, or history of dementia, and for control subjects, fasting blood glucose 7.0 (18). Subjects with a history of noninvalidating stroke were included. Twice as many diabetic patients as control subjects were included in order to increase statistical power for within-group analyses in the type 2 diabetic group.
In the UDES, 122 patients with type 2 diabetes (aged 56eC80 years), 40 patients with type 1 diabetes (aged 52eC77 years), and 61 control subjects (aged 53eC78 years) were included between September 2002 and November 2004. General practitioners in the area (ACKNOWLEDGMENTS) were asked to participate in the project and to invite all eligible type 2 diabetic patients from their practice. Control subjects were recruited among the spouses or acquaintances of the diabetic patients. The study was approved by the medical ethics committee of the University Medical Center Utrecht, and each participant signed an informed consent form. Participants attended the clinic on 2 consecutive days and underwent MRI of the brain and neuropsychological and neurological examinations. Medical history and medication use were recorded. Fasting blood samples were collected, blood pressure was recorded, and urine was collected overnight. The study protocol for the type 1 diabetic patients was slightly different and reported separately.
No MRI could be obtained for nine diabetic patients due to MRI contraindications, such as claustrophobia or a pacemaker. The present study includes all type 2 diabetic patients with an MRI (n = 113, aged 56eC80 years) and all control subjects with an MRI that were at least 56 years of age (n = 51, aged 57eC78 years).
MRI scanning protocol.
The MRI investigation (1.5 Tesla; Philips Medical Systems, Best, the Netherlands) consisted of an axial T1 weighted scan and an axial T2 and T2 fluid-attenuating inverse recovery (FLAIR) scan (TR/TE/TI: 6000/100/2000, FOV 230 mm, matrix 180 x 256, slice thickness 4.0 mm, contiguous slices, 38 slices). The images were printed on hard copy with a reduction factor of 2.9. WMLs, atrophy, and number and location of infarcts were rated on hard copies or on digital images on a personal computer.
WMLs.
WMLs were considered present if they were hyperintense on FLAIR images and not hypointense on T1 weighted images. WMLs were distinguished into periventricular and deep (subcortical) lesions and rated according to the Scheltens rating scale (19). Periventricular WMLs (PWMLs) were rated semiquantitatively per region, adjacent to the frontal horns (frontal capping), the lateral wall of the lateral ventricles (bands), and the occipital horns (occipital capping) on a scale ranging from 0 to 2, with 0 = no PWMLs, 1 = PWMLs 5 mm (real size), and 2 = PWMLs >5 mm. The overall degree of PWMLs was calculated by adding up the scores for the three separate categories on the left and right (range 0eC12). This is a slight modification of the original scale, which only counts the side with the highest score (range 0eC6).
For the rating of deep WMLs (DWMLs), the brain was divided into six regions: frontal, parietal, occipital, temporal, basal ganglia, and infratentorial. This is a slight modification of the original scale, which divides the basal ganglia and infratentorial regions into 5 and 4 different smaller subregions, respectively (19). The different brain regions were determined according to anatomical landmarks, namely the central sulcus, the Sylvian fissure, and the parieto-occipital sulcus, and were shown on templates during the rating. The size and number of the DWMLs were rated per region, on a scale ranging from 0 to 6, with 0 = no DWMLs, 1 = DWMLs 3 mm (real size) (n 5), 2 = DWMLs 3 mm (n 6), 3 = DWMLs 4eC10 mm (n 5), 4 = DWMLs 4eC10 mm (n 6), 5 = DWMLs 11 mm (n 1), and 6 = confluent. The overall degree of DWMLs was calculated by adding the scores of all the regions (range 0eC36).
Furthermore, brain infarcts were scored by location (cortical and subcortical), type (lacunar or large), and number. A lesion was considered a lacunar infarct if its score was hypointense on T1 and FLAIR images and if its appearance was unlike a perivascular space.
Atrophy rating scales.
Cortical atrophy was evaluated quantitatively by the frontal interhemispheric fissure ratio (FFR), i.e., the maximal width of the interhemispheric fissure from any of the cuts demonstrating the frontal lobes divided by the transpineal coronal inner table diameter (20), and by the Sylvian fissure ratio (SFR), i.e., the average of the maximal Sylvian fissure widths taken from the cut showing the widest Sylvian fissure divided by the transpineal coronal inner table diameter (20). Subcortical atrophy was evaluated by the bicaudate ratio (BCR) on the cut best showing the caudate nuclei and by the bifrontal ratio (BFR) measured on the same cut as the BCR. BCR and BFR are defined, respectively, as the minimal distance between the caudate indentations of the frontal horn (20) and the distance between the tips of the frontal horns divided by the distance between the inner tables of the skull along the same line (20). To relate cerebral atrophy to cognitive functioning, the raw data were converted into a cortical atrophy Z score (mean Z FFR and Z SFR) and subcortical atrophy Z score (mean Z BCR and Z BFR), based on the pooled mean of the whole study population.
All MRI scans were independently rated by two raters blinded to diabetes status (S.M.M. and G.J.B.). In the case of disagreements of more than one point on the WML scales in a particular region or >5 mm (actual size) on any of the atrophy measurements (2 mm for fissure widths), consensus readings were held (involving no PWML, 4% DWML, and 4% of atrophy ratio readings). In all other cases the readings of both readers were averaged.
Neuropsychological tests.
All participants performed an extensive neuropsychological examination tapping the major cognitive domains in both a verbal and nonverbal manner. Eleven tasks were administered in a fixed order that took 90 min to complete. These tasks were divided into five cognitive domains to reduce the amount of neuropsychological variables and for clinical clarity. This division was made a priori, according to standard neuropsychological practice and cognitive theory, as described in detail in Lezak’s Neuropsychological Assessment (21). The domain "abstract reasoning" was assessed by Raven Advanced Progressive Matrices (12-item short form). The domain "memory," included four subdomains, "working memory," assessed by the forward and backward digit span of the WAIS-III (Wechsler Adult Intelligence Scale-III) and the Corsi Block-Tapping Task; "immediate memory and learning rate," including verbal memory assessed by the Rey Auditory Verbal Learning Test and visual memory assessed by the Location Learning Test; "forgetting rate," assessed by the delayed task of the Rey Auditory Verbal Learning Test and of the Location Learning Test; and "incidental memory," assessed by the delayed trial of the Rey-Osterrieth Complex Figure. The domain "information processing speed" was assessed by the Trail Making Test Part A, the Stroop Color-Word Test (Part I and II), and the subtest Digit Symbol of the WAIS-III. The domain "attention and executive function" was assessed by the Trail Making Test Part B, the Stroop Color-Word Test (Part III), the Brixton Spatial Anticipation Test, a verbal fluency test using the N and A, and category fluency using animal names. The domain of visuoconstruction was assessed by the Rey-Osterrieth Complex Figure (copy trial). A premorbid IQ was tested with the Dutch version of the National Adult Reading Test.
Since depression is more common in patients with type 2 diabetes than in control subjects (22) and depression may influence cognitive functioning, mood was assessed with a Beck Depression Inventory (23). Both the total score on this self-rated depressive symptoms inventory and the percentage of people scoring above the cutoffcriterion of 15 were recorded.
To compare the five different cognitive domains between the two groups and to perform regression analysis within the type 2 diabetic group, the raw scores were standardized into Z scores per domain. These Z scores were calculated on the pooled mean of the whole study population.
Medical history, blood pressure, blood samples, and vascular disease.
In a standardized interview, participants were asked about diabetes duration, height and weight, history of hypertension, stroke or cardiovascular disease, and smoking. Furthermore, all participants measured their blood pressure at home at nine different time points during the day (Omron MX3; Omron, Mannheim, Germany). Fasting glucose, HbA1c (A1C), fasting triglycerides, and fasting cholesterol were determined. BMI was calculated as weight divided by the square of height. Hypertension was defined as an average systolic blood pressure 160 mmHg or diastolic blood pressure 95 mmHg or self-reported use of blood pressureeClowering drugs. Hypercholesterolemia was defined as a fasting cholesterol >6.2 mmol/l or self-reported use of cholesterol-lowering drugs (24).
Microvascular and macrovascular complications were also assessed. Fundus photographs were rated according to the Wisconsin Epidemiologic Study of Diabetic Retinopathy scale (25). A score 1.5 was defined as retinopathy. Albuminuria was defined as microalbuminuria (albumin 30eC250 mg/l) or macroalbuminuria (albumin 250 mg/l or positive protein dip-stick test) in the overnight urine sample. Neuropathy was defined as a score 6 on the Toronto Clinical Neuropathy Scoring System (26). "Any microvascular disease" was defined as retinopathy, albuminuria, or neuropathy. "Any macrovascular event" was defined as a history of myocardial infarction, stroke, or surgery or endovascular treatment for coronary, carotid, or peripheral (legs, abdominal aorta) artery disease. More detailed data on these complications in relation to cognition and brain MRI will be reported separately.
Statistical analysis.
For the population characteristics and cognition and brain MRI findings, between-group differences were analyzed with t test for means, Mann-Whitney U test for nonparametric data, and 2 test for proportions. Between-group differences on cognition and brain MRI were also assessed by regression analyses and expressed as estimated between-group difference with 95% CI. The primary analyses were adjusted for age, sex, and estimated IQ and in additional analyses, also for blood pressure and the depression inventory score.
Within the type 2 diabetic population, associations between cognition, brain MRI findings, and disease variables were assessed by linear or logistic regression analysis, adjusting for age, sex, and estimated IQ. For the between-group comparisons, P < 0.05 was considered statistically significant. For the within group analyses in the type 2 diabetic patients, a significance level of P < 0.01 was used to accommodate the effects of repeated testing.
RESULTS
Clinical data.
The groups were well balanced for age, sex, level of education (seven categories) (27), and estimated IQ (Table 1). Of the 113 type 2 diabetic participants, 11 (10%) subjects had no treatment or only dietary treatment, 68 (60%) received oral antidiabetic drugs, and 34 (30%) received insulin. Patients with type 2 diabetes had more hypertension, whereas their lipid profile was better than that of control subjects.
Differences between type 2 diabetic patients and control subjects on brain MRI and cognitive domains.
Patients with type 2 diabetes had more cortical and subcortical atrophy than control subjects (FFR P < 0.001, SFR P < 0.001, BCR P = 0.003, and BFR P = 0.14.) (Table 2). Patients with type 2 diabetes had a higher DWML score than control subjects (P = 0.02) (Fig. 1), but PWMLs were not different between the groups (P = 0.13). Furthermore, patients with type 2 diabetes had more (silent) cerebral infarcts than control subjects (type 2 diabetes 22 of 113 and control subjects 4 of 54, P = 0.06). From the 22 patients with infarcts, 12 had lacunar infarcts, 6 had other infarcts (e.g., cortical or large subcortical), and 4 had both lacunar and other infarcts. Of the 22 patients with a visible infarct on their MRI, 6 reported a history of stroke. One control subject had a lacunar infarct, and three control subjects showed other infarcts on their MRI, but none of them reported a history of stroke. The effect of diabetes on cortical atrophy persisted after adjusting for the presence of WML and infarcts.
The performance of type 2 diabetic patients on all five cognitive domains was worse compared with that of control subjects, but statistically significant changes were observed only for the domains "attention and executive functioning" (P = 0.01), "information processing speed" (P = 0.01), and "memory" (P = 0.01) (Fig. 2). Effect sizes were in the small to moderate range (0.2eC0.4).
Adjustment for the possible effects of blood pressure (mean arterial pressure) did not affect the difference between the type 2 diabetic and control groups on the MRI measures or cognition (Table 3). Adjustment for the presence or absence of hypertension gave similar results (data not shown). Moreover, when the findings in hypertensive and nonhypertensive subjects were analyzed separately, the magnitude of the effect of diabetes remained essentially the same (Table 3). The exclusion of control subjects (n = 10) with impaired fasting glucose (>6.0 mmol/l) (28) did not have an apparent effect on between-group differences on cognition or brain MRI.
One control subject and eight type 2 diabetic patients scored above the cutoffcriterion of 15 for the depression inventory score. Exclusion of these participants from the analyses, or adjustment for the depression score, did not affect the between-group differences on cognition (data not shown).
Relation between cognition, MRI, and disease variables in the patients with type 2 diabetes.
Within the type 2 diabetic group, statistically significant associations between MRI abnormalities and cognition were noted, even after adjustment for age, sex, and IQ (Table 4). DWMLs, cortical atrophy, and infarcts were related to information processing speed (P < 0.01). PWMLs and subcortical atrophy also tended to be related to information processing speed (P < 0.05). Subcortical atrophy was related to attention and executive function (P < 0.01) and tended to be related to two other cognitive domains. In a multivariate model with age, sex, IQ, and the different MRI measures as the independent variables and information processing speed as the dependent variable, only the associations with age (P = 0.001), IQ (P < 0.001), DWMLs (P < 0.001), and infarcts (P < 0.001) were statistically significant.
Age was evidently related to atrophy and WML severity (Table 5). After adjusting age for diabetes duration, these results did not change notably. No significant (at the P < 0.01 level) relations between MRI parameters and blood pressure, A1C, and diabetes duration were found, although mean arterial blood pressure tended to be related to PWMLs (P < 0.05). Age was also significantly related to cognition in three of the five cognitive domains (Table 5). Adjustment for diabetes duration did not notably change these results. Mean arterial blood pressure tended to be related to improved memory function (P < 0.05), whereas hypertension was associated with lower performance on the other four cognitive domains, albeit not statistically significant. A1C level and diabetes duration tended to be related to information processing speed (P < 0.05). A history of macrovascular events tended to be related to impaired cognition (P < 0.05) and more severe DWMLs (P < 0.05). Although microvasular disease was not related to brain MRI findings, there was an association with lower performance on the five cognitive domains, albeit not statistically significant. There was no relation between sex and cognition, but subcortical atrophy was more pronounced in men. For the domains "information processing speed" and "memory," the association with age appeared to be stronger in the type 2 diabetic than in the control group (B values per 5 years for control subjects vs. type 2 diabetic patients, respectively: information processing speed eC0.09 (P = 0.19) vs. eC0.32 (P < 0.001) and memory eC0.06 (P = 0.25) vs. eC0.21 (P < 0.001). For memory, this interaction between age and group was statistically significant (P = 0.01).
DISCUSSION
On brain MRI, patients with type 2 diabetes had more cortical and subcortical atrophy and more DWMLs and infarcts than control subjects. The performance of the patients with type 2 diabetes on the neuropsychological examination was worse, particularly affecting the domains "attention," "executive functioning," "information processing speed," and "memory." Adjustment for hypertension did not affect the results. Within the type 2 diabetic group, cognitive function was inversely related to WMLs, atrophy, and the presence of infarcts, and there was a modest association between impaired cognition and A1C and diabetes duration. This association was strongest for age, even more so than in the control group.
Cognitive function of patients with type 2 diabetes has been the subject of several studies (rev. in 2,3), generally reporting deficits in verbal memory and information processing speed and less consistently in executive functioning and nonverbal memory. Our results are in line with these findings. Studies that examined relations between different disease variables and cognitive functioning showed that patients with worse glycemic control were more likely to show cognitive deficits (29). Most of these studies used sample sizes smaller than those used in our study. Moreover, although most studies did not use age as an independent predictor, the largest effect of type 2 diabetes on cognitive function was observed in studies in which patients were older (30). When addressing the relation between hypertension and cognition, the results of previous studies are less consistent. One population-based study found that hypertensive patients with type 2 diabetes were at the greatest risk for poor cognitive performance (31), but other longitudinal studies did not observe an evident relation between hypertension and cognitive performance (9,32,33), in line with the present observations.
Thus far, relatively few studies have specifically addressed brain MRI abnormalities in patients with type 2 diabetes. In line with the present observations, these studies indicate that modest cortical and subcortical atrophy and symptomatic or asymptomatic infarcts are more common in type 2 diabetic patients than in control subjects (11eC15). The relation between cerebral atrophy and hypertension in type 2 diabetic patients, however, is less clear. One study reports no effects of adjustment for hypertension (12), which is consistent with our findings, whereas another study indicated that hypertension appeared to be a major determinant of cerebral atrophy in type 2 diabetic patients (11). Results of previous studies on the association between type 2 diabetes and WMLs are inconsistent. The majority of these studies involved selected subgroups of patients with, for example, clinically manifest cardiovascular disease or stroke (16,17) and used relatively insensitive measures to rate WMLs. Some of these studies in patients with vascular disease reported relatively more severe WMLs in patients with type 2 diabetes (16), whereas others reported no statistically significant relation between type 2 diabetes and WMLs (17). The study on WMLs in elderly subjects with diabetes that involved the largest cohort and the most detailed rating method thus far reported no effect of diabetes on PWMLs, although the volume of DWMLs tended to be higher in the diabetic group (11).
There are, to our knowledge, no previous studies that specifically address the relation between brain MRI abnormalities and cognitive functioning in patients with type 2 diabetes. However, previous studies in general populations of elderly subjects that studied brain MRI abnormalities and cognitive function show results that are comparable with ours. A study of 139 healthy adults (50eC81 years of age), for example, observed an association between atrophy of the prefrontal cortex and the volume of WMLs in the prefrontal region and age-related impairments of executive functioning but not with memory (34). Another study involving 68 healthy, nondemented individuals tested at ages 50, 60, 70, and 80 years reported that PWMLs were related to decline in information processing speed and visuoconstruction and that DWMLs were related to visuoconstruction (35).
The present study may provide some clues regarding the causes of cognitive deficits in patients with type 2 diabetes, as cognitive impairments were associated with subcortical ischemic MRI abnormalities (silent infarcts and WMLs). However, nonvascular mechanisms could also be involved because more diffuse cerebral changes, like cortical atrophy, were also related to impaired cognition. Blood pressure showed a relation with PWMLs and might thus to some extent be involved in the cognitive deficits. However, as is clearly indicated in Table 3, adjustment for hypertension had no obvious effect on the differences in cognition and MRI ratings between control subjects and patients with type 2 diabetes. Chronic hyperglycemia could also be involved, although the relation between A1C and diabetes duration and changes in cognition was only modest. In the present study, the strongest determinant of changes in cognition and on MRI was age. Age was strongly related to all MRI parameters and to three of the five cognitive domains in patients with type 2 diabetes, and the interaction term of age and group was significant for the domain "memory." This points to an interaction between type 2 diabetes and ageing. In fact, several processes that have been implicated in brain ageing, including oxidative stress, accumulation of so-called advanced glycosylation end products, microvascular dysfunction, and alterations in cerebral glucose and insulin metabolism may be accelerated by diabetes (36), which may explain part of this interaction.
A strength of our study is that we combined a detailed analysis of both cognitive function and brain imaging by means of MRI, thus allowing an accurate assessment of the relation between these parameters. A possible limitation of our study could be selection bias. In general, individuals who participate in research projects that include a detailed work-up at a hospital tend to be less affected than individuals who refuse participation. Hence, we do not think this selection bias has a large impact on our results, if any; the strength of the associations would be underestimated because of a healthier study population. We specifically decided not to exclude participants with comorbidity such as hypertension or macrovascular events, as this comorbidity is an integral part of the diabetic condition. If we had excluded subjects with these disorders a priori, our findings would not be generalizable to the general population of type 2 diabetic subjects. The characteristics (i.e., A1C, total cholesterol, triglycerides, blood pressure, and BMI) of the patients with type 2 diabetes that were included in the present study are similar to those reported in large population-based surveys of type 2 diabetic subjects in the Netherlands (37,38). Mean A1C was 6.9% in our type 2 diabetic population, which indicates a moderately well-controlled diabetes. Another limitation of the present study might be its cross-sectional design. This can clearly affect the interpretation of data on the relation between impaired cognition and potential mediators such as blood pressure and glycemic control. In elderly individuals, various factors can affect the outcome measures that were assessed in the present study, of which only a proportion are directly related to type 2 diabetes. A follow-up project involving the present study population has been initiated and may provide more detailed information on the potential role of different metabolic and vascular risk factors.
We conclude that cognitive impairments in patients with type 2 diabetes are related to structural changes in the brain. These changes are indicative of a vascular etiology, although the increased cortical brain atrophy and the relation with age are also suggestive of other mechanisms such as accelerated brain ageing.
APPENDIX
Members of the Utrecht Diabetic Encephalopathy Study Group (in alphabetical order by department).
The Department of Clinical Neurophysiology: A.C. van Huffelen; the Department of Internal Medicine: H.W. de Valk; the Julius Center for Health Sciences and Primary Care: A. Algra and G.E.H.M. Rutten; the Department of Medical Pharmacology: W.H. Gispen; the Department of Neurology: A. Algra, G.J. Biessels, L.J. Kappelle, S.M. Manschot, and J. van Gijn; the Department of Neuropsychology and Helmholtz Research Institute: A.M.A. Brands (currently Zuwe Hofpoort, Regionaal Psychiatrisch Centrum, Woerden, the Netherlands), E.H.F. de Haan, R.P.C. Kessels, and E. van den Berg; and the Department of Radiology: J. van der Grond; all part of the University Medical Center, Utrecht, the Netherlands.
ACKNOWLEDGMENTS
This research was financially supported by the Dutch Diabetes Research Foundation (grant 2001.00.023).
Patients were recruited through the community-based "Utrecht Diabetes Programma" (UDP) and the "IJsselstein Diabetes Project" (mentor Ph.L. Salomee). The UDP is a long-running collaborative project between general practitioners and a hospital-based endocrinologist, with the aim to improve the level of care for patients with type 2 diabetes in the community without referral to the hospital.
The work of Jan Groothuis (Utrecht University) on the validation of the MRI scales is gratefully acknowledged.
FOOTNOTES
A complete list of the members of the Utrecht Diabetic Encephalopathy Study Group can be found in the APPENDIX.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Gispen WH, Biessels GJ: Cognition and synaptic plasticity in diabetes mellitus. Trends Neurosci 23:542eC549, 2000
Stewart R, Liolitsa D: Type 2 diabetes mellitus, cognitive impairment and dementia. Diabet Med 16:93eC112, 1999
Awad N, Gagnon M, Messier C: The relationship between impaired glucose tolerance, type 2 diabetes, and cognitive function. J Clin Exp Neuropsychol 26:1044eC1080, 2004
Leibson CL, Rocca WA, Hanson VA, Cha R, Kokmen E, O’Brien PC, Palumbo PJ: Risk of dementia among persons with diabetes mellitus: a population-based cohort study. Am J Epidemiol 145:301eC308, 1997
Ott A, Stolk RP, Van Harskamp F, Pols HA, Hofman A, Breteler MM: Diabetes mellitus and the risk of dementia: the Rotterdam Study. Neurology 53:1937eC1942, 1999
Allen KV, Frier BM, Strachan MW: The relationship between type 2 diabetes and cognitive dysfunction: longitudinal studies and their methodological limitations. Eur J Pharmacol 490:169eC175, 2004
Hassing LB, Hofer SM, Nilsson SE, Berg S, Pedersen NL, McClearn G, Johansson B: Comorbid type 2 diabetes mellitus and hypertension exacerbates cognitive decline: evidence from a longitudinal study. Age Ageing 33:355eC361, 2004
Asimakopoulou KG, Hampson SE, Morrish NJ: Neuropsychological functioning in older people with type 2 diabetes: the effect of controlling for confounding factors. Diabet Med 19:311eC316, 2002
Kanaya AM, Barrett-Connor E, Gildengorin G, Yaffe K: Change in cognitive function by glucose tolerance status in older adults: a 4-year prospective study of the Rancho Bernardo study cohort. Arch Intern Med 164:1327eC1333, 2004
Luchsinger JA, Reitz C, Honig LS, Tang MX, Shea S, Mayeux R: Aggregation of vascular risk factors and risk of incident Alzheimer disease. Neurology 65:545eC551, 2005
Schmidt R, Launer LJ, Nilsson LG, Pajak A, Sans S, Berger K, Breteler MM, de Ridder M, Dufouil C, Fuhrer R, Giampaoli S, Hofman A: Magnetic resonance imaging of the brain in diabetes: the Cardiovascular Determinants of Dementia (CASCADE) study. Diabetes 53:687eC692, 2004
den Heijer T, Vermeer SE, van Dijk EJ, Prins ND, Koudstaal PJ, Hofman A, Breteler MM: Type 2 diabetes and atrophy of medial temporal lobe structures on brain MRI. Diabetologia 46:1604eC1610, 2003
Vermeer SE, den Heijer T, Koudstaal PJ, Oudkerk M, Hofman A, Breteler MM: Incidence and risk factors of silent brain infarcts in the population-based Rotterdam Scan Study. Stroke 34:392eC396, 2003
Longstreth WT Jr, Bernick C, Manolio TA, Bryan N, Jungreis CA, Price TR: Lacunar infarcts defined by magnetic resonance imaging of 3660 elderly people: the Cardiovascular Health Study. Arch Neurol 55:1217eC1225, 1998
Araki Y, Nomura M, Tanaka H, Yamamoto H, Yamamoto T, Tsukaguchi I, Nakamura H: MRI of the brain in diabetes mellitus. Neuroradiology 36:101eC103, 1994
Schmidt R, Fazekas F, Kleinert G, Offenbacher H, Gindl K, Payer F, Freidl W, Niederkorn K, Lechner H: Magnetic resonance imaging signal hyperintensities in the deep and subcortical white matter: a comparative study between stroke patients and normal volunteers. Arch Neurol 49:825eC827, 1992
Manolio TA, Kronmal RA, Burke GL, Poirier V, O’Leary DH, Gardin JM, Fried LP, Steinberg EP, Bryan RN: Magnetic resonance abnormalities and cardiovascular disease in older adults: the Cardiovascular Health Study. Stroke 25:318eC327, 1994
Expert Committee on Diagnosis and Classification of Diabetes Mellitus: Report of the Expert Committee on Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 21 (Suppl. 1):S5eCS19, 1998
Scheltens P, Barkhof F, Leys D, Pruvo JP, Nauta JJ, Vermersch P, Steinling M, Valk J: A semiquantative rating scale for the assessment of signal hyperintensities on magnetic resonance imaging. J Neurol Sci 114:7eC12, 1993
Gomori JM, Steiner I, Melamed E, Cooper G: The assessment of changes in brain volume using combined linear measurements: a CT-scan study. Neuroradiology 26:21eC24, 1984
Lezak MD, Howieson DB, Loring DW: Neuropsychological Assessment. New York, Oxford University Press, 2004
Anderson RJ, Freedland KE, Clouse RE, Lustman PJ: The prevalence of comorbid depression in adults with diabetes: a meta-analysis. Diabetes Care 24:1069eC1078, 2001
Beck A, Ward C, Mendelson M, Mock J, Erbaugh J: An inventory for measuring depression. Arch Gen Psychiatry 4:561eC571, 1961
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III): Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 285:2486eC2497, 2001
Klein R, Klein BE, Magli YL, Brothers RJ, Meuer SM, Moss SE, Davis MD: An alternative method of grading diabetic retinopathy. Ophthalmology 93:1183eC1187, 1986
Bril V, Perkins BA: Validation of the Toronto Clinical Scoring System for diabetic polyneuropathy. Diabetes Care 25:2048eC2052, 2002
Hochstenbach J, Mulder T, van Limbeek J, Donders R, Schoonderwaldt H: Cognitive decline following stroke: a comprehensive study of cognitive decline following stroke. J Clin Exp Neuropsychol 20:503eC517, 1998
American Diabetes Association: Standards of Medical Care for Patients with Diabetes Mellitus (Position Statement). Diabetes Care 25:213eC229, 2002
Strachan MWJ, Deary IJ, Ewing FME, Frier BM: Is type 2 diabetes associated with an increased risk of cognitive dysfunction Diabetes Care 20:438eC445, 1997
Ryan CM, Geckle M: Why is learning and memory dysfunction in type 2 diabetes limited to older adults Diabete Metab Res Rev 16:308eC315, 2000
Elias PK, Elias MF, D’Agostino RB, Cupples LA, Wilson PW, Silbershatz H, Wolf PA: NIDDM and blood pressure as risk factors for poor cognitive performance: the Framingham Study. Diabetes Care 20:1388eC1395, 1997
Gregg EW, Yaffe K, Cauley JA, Rolka DB, Blackwell TL, Narayan KM, Cummings SR: Is diabetes associated with cognitive impairment and cognitive decline among older women Study of Osteoporotic Fractures Research Group. Arch Intern Med 160:174eC180, 2000
Fontbonne A, Berr C, Ducimetiere P, Alperovitch A: Changes in cognitive abilities over a 4-year period are unfavorably affected in elderly diabetic subjects: results of the Epidemiology of Vascular Aging Study. Diabetes Care 24:366eC370, 2001
Gunning-Dixon FM, Raz N: Neuroanatomical correlates of selected executive functions in middle-aged and older adults: a prospective MRI study. Neuropsychologia 41:1929eC1941, 2003
Garde E, Mortensen EL, Krabbe K, Rostrup E, Larsson HB: Relation between age-related decline in intelligence and cerebral white-matter hyperintensities in healthy octogenarians: a longitudinal study. Lancet 356:628eC634, 2000
Biessels GJ, van der Heide LP, Kamal A, Bleys RL, Gispen WH: Ageing and diabetes: implications for brain function. Eur J Pharmacol 441:1eC14, 2002
Goudswaard AN, Stolk RP, Zuithoff P, Rutten GE: Patient characteristics do not predict poor glycaemic control in type 2 diabetes patients treated in primary care. Eur J Epidemiol 19:541eC545, 2004
de Grauw WJ, van Gerwen WH, van de Lisdonk EH, van den Hoogen HJ, van den Bosch WJ, van Weel C: Outcomes of audit-enhanced monitoring of patients with type 2 diabetes. J Fam Pract 51:459eC464, 2002(Sanne M. Manschot, August)
2 Helmholtz Research Institute, University Medical Center, Utrecht, the Netherlands
3 Department of Radiology, University Medical Center, Utrecht, the Netherlands
4 Julius Center for Health Sciences and Primary Care, University Medical Center, Utrecht, the Netherlands
BCR, bicaudate ratio; BFR, bifrontal ratio; DWML, deep WML; FFR, frontal interhemispheric fissure ratio; FLAIR, fluid-attenuating inverse recovery; MRI, magnetic resonance imaging; PWML, periventricular WML; SFR, Sylvian fissure ratio; WML, white matter lesion
ABSTRACT
The structural correlates of impaired cognition in type 2 diabetes are unclear. The present study compared cognition and brain magnetic resonance imaging (MRI) between type 2 diabetic patients and nondiabetic control subjects and assessed the relationship between cognition and MRI findings and blood pressure and metabolic control. The study included 113 patients and 51 control subjects. Brain MRI scans were rated for white matter lesions (WMLs), cortical and subcortical atrophy, and infarcts. Neuropsychological test scores were divided into five cognitive domains and expressed as standardized Z values. Type 2 diabetes was associated with deep WMLs (P = 0.02), cortical (P < 0.001) and subcortical (P < 0.05) atrophy, (silent) infarcts (P = 0.06), and impaired cognitive performance (attention and executive function, information-processing speed, and memory, all P < 0.05). Adjustment for hypertension did not affect the results. Within the type 2 diabetic group, cognitive function was inversely related with WMLs, atrophy, and the presence of infarcts (adjusted for age, sex, and estimated IQ), and there was a modest association with HbA1c and diabetes duration. This association was strongest for age, even more so than in control subjects. We conclude that cognitive impairments in patients with type 2 diabetes are not only associated with subcortical ischemic changes in the brain, but also with increased brain atrophy.
Type 2 diabetes can affect the central nervous system (1). Neuropsychological studies reported moderate degrees of cognitive impairment, particularly in tasks involving verbal memory or complex information processing (2,3). Epidemiological studies demonstrated an association between diabetes and dementia (4,5). It is not clear which factors mediate accelerated cognitive decline in patients with type 2 diabetes. Both comorbid conditions (e.g., hypertension and depression) and diabetes-specific factors (e.g., glycemic control) have been implicated (2,6). Some investigators have suggested that hypertension is an important mediator (2,7,8), but this was not supported by others (9,10).
The structural correlates and pathophysiological mechanisms underlying these cognitive impairments are still uncertain. Previous studies report that modest cortical and subcortical atrophy and symptomatic or asymptomatic infarcts are more common in type 2 diabetic patients than in control subjects (11eC15). Findings from studies on so-called WMLs are less consistent; some report an association with type 2 diabetes (16), but others report no statistically significant effects (11,17). To the best of our knowledge, there are no published studies that combine detailed assessment of cognitive functioning and magnetic resonance imaging (MRI) of the brain in patients with type 2 diabetes.
The present study aimed to compare cognition and brain MRI in type 2 diabetic patients and nondiabetic control subjects and to relate cognitive functioning in the type 2 diabetic patients to MRI findings, as well as to blood pressure and metabolic control.
RESEARCH DESIGN AND METHODS
The UDES (Utrecht Diabetic Encephalopathy Study) is a cross-sectional population-based study on determinants of impaired cognition in type 2 diabetes. Because the study aimed to identify potential risk factors for cognitive impairment in type 2 diabetes, patients were not selected for the presence or absence of diabetes complications, comorbid conditions (e.g., hypertension), or exposure to other risk factors (e.g., smoking). For inclusion, type 2 diabetic patients had to be 55eC80 years of age, have a diabetes duration of at least 1 year, be functionally independent, and speak Dutch. Exclusion criteria for all participants included a psychiatric or neurological disorder (unrelated to diabetes) that could influence cognitive functioning, history of alcohol or substance abuse, or history of dementia, and for control subjects, fasting blood glucose 7.0 (18). Subjects with a history of noninvalidating stroke were included. Twice as many diabetic patients as control subjects were included in order to increase statistical power for within-group analyses in the type 2 diabetic group.
In the UDES, 122 patients with type 2 diabetes (aged 56eC80 years), 40 patients with type 1 diabetes (aged 52eC77 years), and 61 control subjects (aged 53eC78 years) were included between September 2002 and November 2004. General practitioners in the area (ACKNOWLEDGMENTS) were asked to participate in the project and to invite all eligible type 2 diabetic patients from their practice. Control subjects were recruited among the spouses or acquaintances of the diabetic patients. The study was approved by the medical ethics committee of the University Medical Center Utrecht, and each participant signed an informed consent form. Participants attended the clinic on 2 consecutive days and underwent MRI of the brain and neuropsychological and neurological examinations. Medical history and medication use were recorded. Fasting blood samples were collected, blood pressure was recorded, and urine was collected overnight. The study protocol for the type 1 diabetic patients was slightly different and reported separately.
No MRI could be obtained for nine diabetic patients due to MRI contraindications, such as claustrophobia or a pacemaker. The present study includes all type 2 diabetic patients with an MRI (n = 113, aged 56eC80 years) and all control subjects with an MRI that were at least 56 years of age (n = 51, aged 57eC78 years).
MRI scanning protocol.
The MRI investigation (1.5 Tesla; Philips Medical Systems, Best, the Netherlands) consisted of an axial T1 weighted scan and an axial T2 and T2 fluid-attenuating inverse recovery (FLAIR) scan (TR/TE/TI: 6000/100/2000, FOV 230 mm, matrix 180 x 256, slice thickness 4.0 mm, contiguous slices, 38 slices). The images were printed on hard copy with a reduction factor of 2.9. WMLs, atrophy, and number and location of infarcts were rated on hard copies or on digital images on a personal computer.
WMLs.
WMLs were considered present if they were hyperintense on FLAIR images and not hypointense on T1 weighted images. WMLs were distinguished into periventricular and deep (subcortical) lesions and rated according to the Scheltens rating scale (19). Periventricular WMLs (PWMLs) were rated semiquantitatively per region, adjacent to the frontal horns (frontal capping), the lateral wall of the lateral ventricles (bands), and the occipital horns (occipital capping) on a scale ranging from 0 to 2, with 0 = no PWMLs, 1 = PWMLs 5 mm (real size), and 2 = PWMLs >5 mm. The overall degree of PWMLs was calculated by adding up the scores for the three separate categories on the left and right (range 0eC12). This is a slight modification of the original scale, which only counts the side with the highest score (range 0eC6).
For the rating of deep WMLs (DWMLs), the brain was divided into six regions: frontal, parietal, occipital, temporal, basal ganglia, and infratentorial. This is a slight modification of the original scale, which divides the basal ganglia and infratentorial regions into 5 and 4 different smaller subregions, respectively (19). The different brain regions were determined according to anatomical landmarks, namely the central sulcus, the Sylvian fissure, and the parieto-occipital sulcus, and were shown on templates during the rating. The size and number of the DWMLs were rated per region, on a scale ranging from 0 to 6, with 0 = no DWMLs, 1 = DWMLs 3 mm (real size) (n 5), 2 = DWMLs 3 mm (n 6), 3 = DWMLs 4eC10 mm (n 5), 4 = DWMLs 4eC10 mm (n 6), 5 = DWMLs 11 mm (n 1), and 6 = confluent. The overall degree of DWMLs was calculated by adding the scores of all the regions (range 0eC36).
Furthermore, brain infarcts were scored by location (cortical and subcortical), type (lacunar or large), and number. A lesion was considered a lacunar infarct if its score was hypointense on T1 and FLAIR images and if its appearance was unlike a perivascular space.
Atrophy rating scales.
Cortical atrophy was evaluated quantitatively by the frontal interhemispheric fissure ratio (FFR), i.e., the maximal width of the interhemispheric fissure from any of the cuts demonstrating the frontal lobes divided by the transpineal coronal inner table diameter (20), and by the Sylvian fissure ratio (SFR), i.e., the average of the maximal Sylvian fissure widths taken from the cut showing the widest Sylvian fissure divided by the transpineal coronal inner table diameter (20). Subcortical atrophy was evaluated by the bicaudate ratio (BCR) on the cut best showing the caudate nuclei and by the bifrontal ratio (BFR) measured on the same cut as the BCR. BCR and BFR are defined, respectively, as the minimal distance between the caudate indentations of the frontal horn (20) and the distance between the tips of the frontal horns divided by the distance between the inner tables of the skull along the same line (20). To relate cerebral atrophy to cognitive functioning, the raw data were converted into a cortical atrophy Z score (mean Z FFR and Z SFR) and subcortical atrophy Z score (mean Z BCR and Z BFR), based on the pooled mean of the whole study population.
All MRI scans were independently rated by two raters blinded to diabetes status (S.M.M. and G.J.B.). In the case of disagreements of more than one point on the WML scales in a particular region or >5 mm (actual size) on any of the atrophy measurements (2 mm for fissure widths), consensus readings were held (involving no PWML, 4% DWML, and 4% of atrophy ratio readings). In all other cases the readings of both readers were averaged.
Neuropsychological tests.
All participants performed an extensive neuropsychological examination tapping the major cognitive domains in both a verbal and nonverbal manner. Eleven tasks were administered in a fixed order that took 90 min to complete. These tasks were divided into five cognitive domains to reduce the amount of neuropsychological variables and for clinical clarity. This division was made a priori, according to standard neuropsychological practice and cognitive theory, as described in detail in Lezak’s Neuropsychological Assessment (21). The domain "abstract reasoning" was assessed by Raven Advanced Progressive Matrices (12-item short form). The domain "memory," included four subdomains, "working memory," assessed by the forward and backward digit span of the WAIS-III (Wechsler Adult Intelligence Scale-III) and the Corsi Block-Tapping Task; "immediate memory and learning rate," including verbal memory assessed by the Rey Auditory Verbal Learning Test and visual memory assessed by the Location Learning Test; "forgetting rate," assessed by the delayed task of the Rey Auditory Verbal Learning Test and of the Location Learning Test; and "incidental memory," assessed by the delayed trial of the Rey-Osterrieth Complex Figure. The domain "information processing speed" was assessed by the Trail Making Test Part A, the Stroop Color-Word Test (Part I and II), and the subtest Digit Symbol of the WAIS-III. The domain "attention and executive function" was assessed by the Trail Making Test Part B, the Stroop Color-Word Test (Part III), the Brixton Spatial Anticipation Test, a verbal fluency test using the N and A, and category fluency using animal names. The domain of visuoconstruction was assessed by the Rey-Osterrieth Complex Figure (copy trial). A premorbid IQ was tested with the Dutch version of the National Adult Reading Test.
Since depression is more common in patients with type 2 diabetes than in control subjects (22) and depression may influence cognitive functioning, mood was assessed with a Beck Depression Inventory (23). Both the total score on this self-rated depressive symptoms inventory and the percentage of people scoring above the cutoffcriterion of 15 were recorded.
To compare the five different cognitive domains between the two groups and to perform regression analysis within the type 2 diabetic group, the raw scores were standardized into Z scores per domain. These Z scores were calculated on the pooled mean of the whole study population.
Medical history, blood pressure, blood samples, and vascular disease.
In a standardized interview, participants were asked about diabetes duration, height and weight, history of hypertension, stroke or cardiovascular disease, and smoking. Furthermore, all participants measured their blood pressure at home at nine different time points during the day (Omron MX3; Omron, Mannheim, Germany). Fasting glucose, HbA1c (A1C), fasting triglycerides, and fasting cholesterol were determined. BMI was calculated as weight divided by the square of height. Hypertension was defined as an average systolic blood pressure 160 mmHg or diastolic blood pressure 95 mmHg or self-reported use of blood pressureeClowering drugs. Hypercholesterolemia was defined as a fasting cholesterol >6.2 mmol/l or self-reported use of cholesterol-lowering drugs (24).
Microvascular and macrovascular complications were also assessed. Fundus photographs were rated according to the Wisconsin Epidemiologic Study of Diabetic Retinopathy scale (25). A score 1.5 was defined as retinopathy. Albuminuria was defined as microalbuminuria (albumin 30eC250 mg/l) or macroalbuminuria (albumin 250 mg/l or positive protein dip-stick test) in the overnight urine sample. Neuropathy was defined as a score 6 on the Toronto Clinical Neuropathy Scoring System (26). "Any microvascular disease" was defined as retinopathy, albuminuria, or neuropathy. "Any macrovascular event" was defined as a history of myocardial infarction, stroke, or surgery or endovascular treatment for coronary, carotid, or peripheral (legs, abdominal aorta) artery disease. More detailed data on these complications in relation to cognition and brain MRI will be reported separately.
Statistical analysis.
For the population characteristics and cognition and brain MRI findings, between-group differences were analyzed with t test for means, Mann-Whitney U test for nonparametric data, and 2 test for proportions. Between-group differences on cognition and brain MRI were also assessed by regression analyses and expressed as estimated between-group difference with 95% CI. The primary analyses were adjusted for age, sex, and estimated IQ and in additional analyses, also for blood pressure and the depression inventory score.
Within the type 2 diabetic population, associations between cognition, brain MRI findings, and disease variables were assessed by linear or logistic regression analysis, adjusting for age, sex, and estimated IQ. For the between-group comparisons, P < 0.05 was considered statistically significant. For the within group analyses in the type 2 diabetic patients, a significance level of P < 0.01 was used to accommodate the effects of repeated testing.
RESULTS
Clinical data.
The groups were well balanced for age, sex, level of education (seven categories) (27), and estimated IQ (Table 1). Of the 113 type 2 diabetic participants, 11 (10%) subjects had no treatment or only dietary treatment, 68 (60%) received oral antidiabetic drugs, and 34 (30%) received insulin. Patients with type 2 diabetes had more hypertension, whereas their lipid profile was better than that of control subjects.
Differences between type 2 diabetic patients and control subjects on brain MRI and cognitive domains.
Patients with type 2 diabetes had more cortical and subcortical atrophy than control subjects (FFR P < 0.001, SFR P < 0.001, BCR P = 0.003, and BFR P = 0.14.) (Table 2). Patients with type 2 diabetes had a higher DWML score than control subjects (P = 0.02) (Fig. 1), but PWMLs were not different between the groups (P = 0.13). Furthermore, patients with type 2 diabetes had more (silent) cerebral infarcts than control subjects (type 2 diabetes 22 of 113 and control subjects 4 of 54, P = 0.06). From the 22 patients with infarcts, 12 had lacunar infarcts, 6 had other infarcts (e.g., cortical or large subcortical), and 4 had both lacunar and other infarcts. Of the 22 patients with a visible infarct on their MRI, 6 reported a history of stroke. One control subject had a lacunar infarct, and three control subjects showed other infarcts on their MRI, but none of them reported a history of stroke. The effect of diabetes on cortical atrophy persisted after adjusting for the presence of WML and infarcts.
The performance of type 2 diabetic patients on all five cognitive domains was worse compared with that of control subjects, but statistically significant changes were observed only for the domains "attention and executive functioning" (P = 0.01), "information processing speed" (P = 0.01), and "memory" (P = 0.01) (Fig. 2). Effect sizes were in the small to moderate range (0.2eC0.4).
Adjustment for the possible effects of blood pressure (mean arterial pressure) did not affect the difference between the type 2 diabetic and control groups on the MRI measures or cognition (Table 3). Adjustment for the presence or absence of hypertension gave similar results (data not shown). Moreover, when the findings in hypertensive and nonhypertensive subjects were analyzed separately, the magnitude of the effect of diabetes remained essentially the same (Table 3). The exclusion of control subjects (n = 10) with impaired fasting glucose (>6.0 mmol/l) (28) did not have an apparent effect on between-group differences on cognition or brain MRI.
One control subject and eight type 2 diabetic patients scored above the cutoffcriterion of 15 for the depression inventory score. Exclusion of these participants from the analyses, or adjustment for the depression score, did not affect the between-group differences on cognition (data not shown).
Relation between cognition, MRI, and disease variables in the patients with type 2 diabetes.
Within the type 2 diabetic group, statistically significant associations between MRI abnormalities and cognition were noted, even after adjustment for age, sex, and IQ (Table 4). DWMLs, cortical atrophy, and infarcts were related to information processing speed (P < 0.01). PWMLs and subcortical atrophy also tended to be related to information processing speed (P < 0.05). Subcortical atrophy was related to attention and executive function (P < 0.01) and tended to be related to two other cognitive domains. In a multivariate model with age, sex, IQ, and the different MRI measures as the independent variables and information processing speed as the dependent variable, only the associations with age (P = 0.001), IQ (P < 0.001), DWMLs (P < 0.001), and infarcts (P < 0.001) were statistically significant.
Age was evidently related to atrophy and WML severity (Table 5). After adjusting age for diabetes duration, these results did not change notably. No significant (at the P < 0.01 level) relations between MRI parameters and blood pressure, A1C, and diabetes duration were found, although mean arterial blood pressure tended to be related to PWMLs (P < 0.05). Age was also significantly related to cognition in three of the five cognitive domains (Table 5). Adjustment for diabetes duration did not notably change these results. Mean arterial blood pressure tended to be related to improved memory function (P < 0.05), whereas hypertension was associated with lower performance on the other four cognitive domains, albeit not statistically significant. A1C level and diabetes duration tended to be related to information processing speed (P < 0.05). A history of macrovascular events tended to be related to impaired cognition (P < 0.05) and more severe DWMLs (P < 0.05). Although microvasular disease was not related to brain MRI findings, there was an association with lower performance on the five cognitive domains, albeit not statistically significant. There was no relation between sex and cognition, but subcortical atrophy was more pronounced in men. For the domains "information processing speed" and "memory," the association with age appeared to be stronger in the type 2 diabetic than in the control group (B values per 5 years for control subjects vs. type 2 diabetic patients, respectively: information processing speed eC0.09 (P = 0.19) vs. eC0.32 (P < 0.001) and memory eC0.06 (P = 0.25) vs. eC0.21 (P < 0.001). For memory, this interaction between age and group was statistically significant (P = 0.01).
DISCUSSION
On brain MRI, patients with type 2 diabetes had more cortical and subcortical atrophy and more DWMLs and infarcts than control subjects. The performance of the patients with type 2 diabetes on the neuropsychological examination was worse, particularly affecting the domains "attention," "executive functioning," "information processing speed," and "memory." Adjustment for hypertension did not affect the results. Within the type 2 diabetic group, cognitive function was inversely related to WMLs, atrophy, and the presence of infarcts, and there was a modest association between impaired cognition and A1C and diabetes duration. This association was strongest for age, even more so than in the control group.
Cognitive function of patients with type 2 diabetes has been the subject of several studies (rev. in 2,3), generally reporting deficits in verbal memory and information processing speed and less consistently in executive functioning and nonverbal memory. Our results are in line with these findings. Studies that examined relations between different disease variables and cognitive functioning showed that patients with worse glycemic control were more likely to show cognitive deficits (29). Most of these studies used sample sizes smaller than those used in our study. Moreover, although most studies did not use age as an independent predictor, the largest effect of type 2 diabetes on cognitive function was observed in studies in which patients were older (30). When addressing the relation between hypertension and cognition, the results of previous studies are less consistent. One population-based study found that hypertensive patients with type 2 diabetes were at the greatest risk for poor cognitive performance (31), but other longitudinal studies did not observe an evident relation between hypertension and cognitive performance (9,32,33), in line with the present observations.
Thus far, relatively few studies have specifically addressed brain MRI abnormalities in patients with type 2 diabetes. In line with the present observations, these studies indicate that modest cortical and subcortical atrophy and symptomatic or asymptomatic infarcts are more common in type 2 diabetic patients than in control subjects (11eC15). The relation between cerebral atrophy and hypertension in type 2 diabetic patients, however, is less clear. One study reports no effects of adjustment for hypertension (12), which is consistent with our findings, whereas another study indicated that hypertension appeared to be a major determinant of cerebral atrophy in type 2 diabetic patients (11). Results of previous studies on the association between type 2 diabetes and WMLs are inconsistent. The majority of these studies involved selected subgroups of patients with, for example, clinically manifest cardiovascular disease or stroke (16,17) and used relatively insensitive measures to rate WMLs. Some of these studies in patients with vascular disease reported relatively more severe WMLs in patients with type 2 diabetes (16), whereas others reported no statistically significant relation between type 2 diabetes and WMLs (17). The study on WMLs in elderly subjects with diabetes that involved the largest cohort and the most detailed rating method thus far reported no effect of diabetes on PWMLs, although the volume of DWMLs tended to be higher in the diabetic group (11).
There are, to our knowledge, no previous studies that specifically address the relation between brain MRI abnormalities and cognitive functioning in patients with type 2 diabetes. However, previous studies in general populations of elderly subjects that studied brain MRI abnormalities and cognitive function show results that are comparable with ours. A study of 139 healthy adults (50eC81 years of age), for example, observed an association between atrophy of the prefrontal cortex and the volume of WMLs in the prefrontal region and age-related impairments of executive functioning but not with memory (34). Another study involving 68 healthy, nondemented individuals tested at ages 50, 60, 70, and 80 years reported that PWMLs were related to decline in information processing speed and visuoconstruction and that DWMLs were related to visuoconstruction (35).
The present study may provide some clues regarding the causes of cognitive deficits in patients with type 2 diabetes, as cognitive impairments were associated with subcortical ischemic MRI abnormalities (silent infarcts and WMLs). However, nonvascular mechanisms could also be involved because more diffuse cerebral changes, like cortical atrophy, were also related to impaired cognition. Blood pressure showed a relation with PWMLs and might thus to some extent be involved in the cognitive deficits. However, as is clearly indicated in Table 3, adjustment for hypertension had no obvious effect on the differences in cognition and MRI ratings between control subjects and patients with type 2 diabetes. Chronic hyperglycemia could also be involved, although the relation between A1C and diabetes duration and changes in cognition was only modest. In the present study, the strongest determinant of changes in cognition and on MRI was age. Age was strongly related to all MRI parameters and to three of the five cognitive domains in patients with type 2 diabetes, and the interaction term of age and group was significant for the domain "memory." This points to an interaction between type 2 diabetes and ageing. In fact, several processes that have been implicated in brain ageing, including oxidative stress, accumulation of so-called advanced glycosylation end products, microvascular dysfunction, and alterations in cerebral glucose and insulin metabolism may be accelerated by diabetes (36), which may explain part of this interaction.
A strength of our study is that we combined a detailed analysis of both cognitive function and brain imaging by means of MRI, thus allowing an accurate assessment of the relation between these parameters. A possible limitation of our study could be selection bias. In general, individuals who participate in research projects that include a detailed work-up at a hospital tend to be less affected than individuals who refuse participation. Hence, we do not think this selection bias has a large impact on our results, if any; the strength of the associations would be underestimated because of a healthier study population. We specifically decided not to exclude participants with comorbidity such as hypertension or macrovascular events, as this comorbidity is an integral part of the diabetic condition. If we had excluded subjects with these disorders a priori, our findings would not be generalizable to the general population of type 2 diabetic subjects. The characteristics (i.e., A1C, total cholesterol, triglycerides, blood pressure, and BMI) of the patients with type 2 diabetes that were included in the present study are similar to those reported in large population-based surveys of type 2 diabetic subjects in the Netherlands (37,38). Mean A1C was 6.9% in our type 2 diabetic population, which indicates a moderately well-controlled diabetes. Another limitation of the present study might be its cross-sectional design. This can clearly affect the interpretation of data on the relation between impaired cognition and potential mediators such as blood pressure and glycemic control. In elderly individuals, various factors can affect the outcome measures that were assessed in the present study, of which only a proportion are directly related to type 2 diabetes. A follow-up project involving the present study population has been initiated and may provide more detailed information on the potential role of different metabolic and vascular risk factors.
We conclude that cognitive impairments in patients with type 2 diabetes are related to structural changes in the brain. These changes are indicative of a vascular etiology, although the increased cortical brain atrophy and the relation with age are also suggestive of other mechanisms such as accelerated brain ageing.
APPENDIX
Members of the Utrecht Diabetic Encephalopathy Study Group (in alphabetical order by department).
The Department of Clinical Neurophysiology: A.C. van Huffelen; the Department of Internal Medicine: H.W. de Valk; the Julius Center for Health Sciences and Primary Care: A. Algra and G.E.H.M. Rutten; the Department of Medical Pharmacology: W.H. Gispen; the Department of Neurology: A. Algra, G.J. Biessels, L.J. Kappelle, S.M. Manschot, and J. van Gijn; the Department of Neuropsychology and Helmholtz Research Institute: A.M.A. Brands (currently Zuwe Hofpoort, Regionaal Psychiatrisch Centrum, Woerden, the Netherlands), E.H.F. de Haan, R.P.C. Kessels, and E. van den Berg; and the Department of Radiology: J. van der Grond; all part of the University Medical Center, Utrecht, the Netherlands.
ACKNOWLEDGMENTS
This research was financially supported by the Dutch Diabetes Research Foundation (grant 2001.00.023).
Patients were recruited through the community-based "Utrecht Diabetes Programma" (UDP) and the "IJsselstein Diabetes Project" (mentor Ph.L. Salomee). The UDP is a long-running collaborative project between general practitioners and a hospital-based endocrinologist, with the aim to improve the level of care for patients with type 2 diabetes in the community without referral to the hospital.
The work of Jan Groothuis (Utrecht University) on the validation of the MRI scales is gratefully acknowledged.
FOOTNOTES
A complete list of the members of the Utrecht Diabetic Encephalopathy Study Group can be found in the APPENDIX.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Gispen WH, Biessels GJ: Cognition and synaptic plasticity in diabetes mellitus. Trends Neurosci 23:542eC549, 2000
Stewart R, Liolitsa D: Type 2 diabetes mellitus, cognitive impairment and dementia. Diabet Med 16:93eC112, 1999
Awad N, Gagnon M, Messier C: The relationship between impaired glucose tolerance, type 2 diabetes, and cognitive function. J Clin Exp Neuropsychol 26:1044eC1080, 2004
Leibson CL, Rocca WA, Hanson VA, Cha R, Kokmen E, O’Brien PC, Palumbo PJ: Risk of dementia among persons with diabetes mellitus: a population-based cohort study. Am J Epidemiol 145:301eC308, 1997
Ott A, Stolk RP, Van Harskamp F, Pols HA, Hofman A, Breteler MM: Diabetes mellitus and the risk of dementia: the Rotterdam Study. Neurology 53:1937eC1942, 1999
Allen KV, Frier BM, Strachan MW: The relationship between type 2 diabetes and cognitive dysfunction: longitudinal studies and their methodological limitations. Eur J Pharmacol 490:169eC175, 2004
Hassing LB, Hofer SM, Nilsson SE, Berg S, Pedersen NL, McClearn G, Johansson B: Comorbid type 2 diabetes mellitus and hypertension exacerbates cognitive decline: evidence from a longitudinal study. Age Ageing 33:355eC361, 2004
Asimakopoulou KG, Hampson SE, Morrish NJ: Neuropsychological functioning in older people with type 2 diabetes: the effect of controlling for confounding factors. Diabet Med 19:311eC316, 2002
Kanaya AM, Barrett-Connor E, Gildengorin G, Yaffe K: Change in cognitive function by glucose tolerance status in older adults: a 4-year prospective study of the Rancho Bernardo study cohort. Arch Intern Med 164:1327eC1333, 2004
Luchsinger JA, Reitz C, Honig LS, Tang MX, Shea S, Mayeux R: Aggregation of vascular risk factors and risk of incident Alzheimer disease. Neurology 65:545eC551, 2005
Schmidt R, Launer LJ, Nilsson LG, Pajak A, Sans S, Berger K, Breteler MM, de Ridder M, Dufouil C, Fuhrer R, Giampaoli S, Hofman A: Magnetic resonance imaging of the brain in diabetes: the Cardiovascular Determinants of Dementia (CASCADE) study. Diabetes 53:687eC692, 2004
den Heijer T, Vermeer SE, van Dijk EJ, Prins ND, Koudstaal PJ, Hofman A, Breteler MM: Type 2 diabetes and atrophy of medial temporal lobe structures on brain MRI. Diabetologia 46:1604eC1610, 2003
Vermeer SE, den Heijer T, Koudstaal PJ, Oudkerk M, Hofman A, Breteler MM: Incidence and risk factors of silent brain infarcts in the population-based Rotterdam Scan Study. Stroke 34:392eC396, 2003
Longstreth WT Jr, Bernick C, Manolio TA, Bryan N, Jungreis CA, Price TR: Lacunar infarcts defined by magnetic resonance imaging of 3660 elderly people: the Cardiovascular Health Study. Arch Neurol 55:1217eC1225, 1998
Araki Y, Nomura M, Tanaka H, Yamamoto H, Yamamoto T, Tsukaguchi I, Nakamura H: MRI of the brain in diabetes mellitus. Neuroradiology 36:101eC103, 1994
Schmidt R, Fazekas F, Kleinert G, Offenbacher H, Gindl K, Payer F, Freidl W, Niederkorn K, Lechner H: Magnetic resonance imaging signal hyperintensities in the deep and subcortical white matter: a comparative study between stroke patients and normal volunteers. Arch Neurol 49:825eC827, 1992
Manolio TA, Kronmal RA, Burke GL, Poirier V, O’Leary DH, Gardin JM, Fried LP, Steinberg EP, Bryan RN: Magnetic resonance abnormalities and cardiovascular disease in older adults: the Cardiovascular Health Study. Stroke 25:318eC327, 1994
Expert Committee on Diagnosis and Classification of Diabetes Mellitus: Report of the Expert Committee on Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 21 (Suppl. 1):S5eCS19, 1998
Scheltens P, Barkhof F, Leys D, Pruvo JP, Nauta JJ, Vermersch P, Steinling M, Valk J: A semiquantative rating scale for the assessment of signal hyperintensities on magnetic resonance imaging. J Neurol Sci 114:7eC12, 1993
Gomori JM, Steiner I, Melamed E, Cooper G: The assessment of changes in brain volume using combined linear measurements: a CT-scan study. Neuroradiology 26:21eC24, 1984
Lezak MD, Howieson DB, Loring DW: Neuropsychological Assessment. New York, Oxford University Press, 2004
Anderson RJ, Freedland KE, Clouse RE, Lustman PJ: The prevalence of comorbid depression in adults with diabetes: a meta-analysis. Diabetes Care 24:1069eC1078, 2001
Beck A, Ward C, Mendelson M, Mock J, Erbaugh J: An inventory for measuring depression. Arch Gen Psychiatry 4:561eC571, 1961
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III): Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 285:2486eC2497, 2001
Klein R, Klein BE, Magli YL, Brothers RJ, Meuer SM, Moss SE, Davis MD: An alternative method of grading diabetic retinopathy. Ophthalmology 93:1183eC1187, 1986
Bril V, Perkins BA: Validation of the Toronto Clinical Scoring System for diabetic polyneuropathy. Diabetes Care 25:2048eC2052, 2002
Hochstenbach J, Mulder T, van Limbeek J, Donders R, Schoonderwaldt H: Cognitive decline following stroke: a comprehensive study of cognitive decline following stroke. J Clin Exp Neuropsychol 20:503eC517, 1998
American Diabetes Association: Standards of Medical Care for Patients with Diabetes Mellitus (Position Statement). Diabetes Care 25:213eC229, 2002
Strachan MWJ, Deary IJ, Ewing FME, Frier BM: Is type 2 diabetes associated with an increased risk of cognitive dysfunction Diabetes Care 20:438eC445, 1997
Ryan CM, Geckle M: Why is learning and memory dysfunction in type 2 diabetes limited to older adults Diabete Metab Res Rev 16:308eC315, 2000
Elias PK, Elias MF, D’Agostino RB, Cupples LA, Wilson PW, Silbershatz H, Wolf PA: NIDDM and blood pressure as risk factors for poor cognitive performance: the Framingham Study. Diabetes Care 20:1388eC1395, 1997
Gregg EW, Yaffe K, Cauley JA, Rolka DB, Blackwell TL, Narayan KM, Cummings SR: Is diabetes associated with cognitive impairment and cognitive decline among older women Study of Osteoporotic Fractures Research Group. Arch Intern Med 160:174eC180, 2000
Fontbonne A, Berr C, Ducimetiere P, Alperovitch A: Changes in cognitive abilities over a 4-year period are unfavorably affected in elderly diabetic subjects: results of the Epidemiology of Vascular Aging Study. Diabetes Care 24:366eC370, 2001
Gunning-Dixon FM, Raz N: Neuroanatomical correlates of selected executive functions in middle-aged and older adults: a prospective MRI study. Neuropsychologia 41:1929eC1941, 2003
Garde E, Mortensen EL, Krabbe K, Rostrup E, Larsson HB: Relation between age-related decline in intelligence and cerebral white-matter hyperintensities in healthy octogenarians: a longitudinal study. Lancet 356:628eC634, 2000
Biessels GJ, van der Heide LP, Kamal A, Bleys RL, Gispen WH: Ageing and diabetes: implications for brain function. Eur J Pharmacol 441:1eC14, 2002
Goudswaard AN, Stolk RP, Zuithoff P, Rutten GE: Patient characteristics do not predict poor glycaemic control in type 2 diabetes patients treated in primary care. Eur J Epidemiol 19:541eC545, 2004
de Grauw WJ, van Gerwen WH, van de Lisdonk EH, van den Hoogen HJ, van den Bosch WJ, van Weel C: Outcomes of audit-enhanced monitoring of patients with type 2 diabetes. J Fam Pract 51:459eC464, 2002(Sanne M. Manschot, August)