Title: Mental slowness and executive dysfunctions in patients with Metabolic Syndrome

Title: Mental slowness and executive dysfunctions in patients with Metabolic Syndrome. Authors: BÀRBARA SEGURA1, MARIA ANGELES JURADO1,5, NÚRIA FREIXENET2, CARLOTA ALBUIN3, JESÚS MUNIESA4, CARME JUNQUÉ1,6 Department of Psychiatry and Clinical Psychobiology. University of Barcelona. Spain. Diabetes, Endocrinology and Nutrition Service. Hospital de Sabadell. Corporació Sanitària Parc Taulí. Sabadell. Spain. 3 Centre d’atenció primària Canaletes. Cerdanyola del Vallés. 4 Centre d’atenció primària de Sant Just Desvern. 5 Institute for Brain, Cognition and Behavior (IR3C) Barcelona, Spain Institute of Biomedical Research August Pi i Sunyer (IDIBAPS), Barcelona, Spain


INTRODUCTION
At the beginning of the twentieth century, Kylin described a syndrome which included hypertension, hyperglycemia and hyperuricemia (1). Over the last twenty years the syndrome has been studied by various disciplines and has received different names: the "deadly quartet" (2), hypertriglyceridemic waist (3), X syndrome, or insulin resistance (4). At present the most widely-used name is Metabolic Syndrome (MetSd), which has been defined by the National Cholesterol Education Program (NCEP) (5) as the presence of at least three of the following vascular risk factors: hypertension, hyperglycemia, hypertriglyceridemia, low levels of high-density lipoprotein (HDL) cholesterol, and obesity as measured by waist circumference (Supplementary data 1).
To date, several epidemiological studies have highlighted the effects of vascular risk factors on cognition (6), and there is some evidence of the neuropsychological effects of obesity (7), impaired fasting glucose (8), dyslipemia (9) and hypertension (10), as individual vascular risk factors of the syndrome. Yaffe et al. (2004) reported that MetSd patients obtained lower Mini Mental State Examination (MMSE) scores than healthy people. MetSd patients also showed a decline in memory and MMSE scores in a three-year follow-up (11). A prospective cohort study of cardiovascular disease performed in 1965 and in 1991 (12) reported an association between the presence of vascular risk factors at baseline with evolution to dementia after a 25-year period in MetSd patients. More recently, studies have linked MetSd to a higher risk of Alzheimer's disease (13), frontosubcortical symptoms, and poorer neuropsychological performance associated with ageing (14,15). However, some authors have reported better cognitive performance in very old people with MetSd (16,17), To our knowledge, most studies of MetSd are cross-sectional (29,32) and use regression analysis to evaluate the association between the syndrome and cognitive decline. Only a few of them include an extensive neuropsychological assessment of MetSd patients.
The aim of the present study was to investigate the neuropsychological impairment of MetSd using a case-control design in which the control group did not manifest any of the risk factors included in the syndrome. We hypothesized that MetSd patients would show a specific cognitive profile, due to the combined effects of vascular risk factors on the brain. Specifically, this profile would probably reflect the impact of the syndrome on white matter (WM) integrity. Therefore, MetSd patients may show differences in processing speed and in correct performance of executive functions.

Research participants
A medical team reviewed the medical records of 600 patients from two public medical centers of Barcelona province (Centres d'atenció primària) in Cerdanyola del Vallés and Sant Just Desvern, Barcelona (Spain). All the participants were volunteers, right handed, aged between 50 and 80 years old, and literate. The exclusion criteria were: uncorrected visual or auditory deficits, drug abuse or alcoholism (more than 28 standard drink units per week for men and more than 17 standard drink units per week for women (18)), a history of developmental pathology, and current neurological, hematological, hormonal, nutritional pathology and/or neoplasm. All selected participants also completed a screening interview in order to check the medical information from their records, to measure their general cognitive capacity according to MMSE score (19) and to estimate their intelligence coefficient using the Vocabulary subtest of the Wechsler Adult Intelligence Scale-3rd edition (WAIS-III) (19). In the same session participants also underwent a modified interview based on the Structured Clinical Interview for DSM-IV Axis I disorders (20) to exclude major psychiatric diseases. Subjects with an MMSE score below 26 and estimated intelligence coefficients below 85 were excluded. Praxis skills were preserved in the whole sample and were assessed by bilateral imitation of hand positions (19). Hand dominance was assessed by Edinburgh test (19). Years of education and occupational level (low, medium or high) were also recorded.

Neuropsychological assessment
All participants underwent a comprehensive neuropsychological evaluation. The tests were administered by a trained neuropsychologist in a fixed order during two 90-min sessions.
The RAVLT variables computed were total learning (sum of correct responses from trial I to trial V), delayed memory recall after 20 min (VI) and retroactive interference, which was defined as the influence of new learning (list b) on the previously achieved learning (list a) (19). Visual memory was assessed by the Faces I subtest of the Wechsler Memory Scale III (WMS-III) (19).
Immediate memory span was evaluated with the forward digit scores of the Digit subtest of the WAIS-III, while measures of working memory were based on backward digit scores of the Digit subtest and the Arithmetic subtest of the WAIS III (19).
Visuoperceptual and visuoconstructive functions were evaluated by the number of correct items on the Benton Facial Recognition Test-Short Form (19)  Verbal fluency was measured by the number of words produced within two restricted categories: semantic (animals) and phonemic (beginning with the letter "p"), within a 1-min limit. The Stroop test (19) was administered to obtain a measure of interference response.
The processing speed variables used were total correct responses in 90 s on the orally-administered Symbol Digit Modalities Test (SDMT) (19) number of seconds taken to solve the Trail Making Test Part A, time taken to complete the Grooved Pegboard Test (19) with the dominant and non-dominant hands, and reaction time in CPT-II (Continuous Performance Test II) (19).
Finally, the total number of correct responses on the short form of the Boston Naming Test (Spanish version) was used to assess language (19).

Statistical analysis
Demographic and neuropsychological differences between groups were compared using the Statistical Package for the Social Sciences (SPSS WIN; v.14.0). The Student's t test or its non-parametric equivalent, the Mann-Whitney U test, were used for quantitative variables when required, while the χ 2 test was used to analyze qualitative variables. Effects of education and gender were controlled by an analysis of covariance (ANCOVA) of significant variables; some of them were converted to guarantee the ANCOVA assumptions. The magnitude of the observed effect was calculated by the r effect size. The discriminative power of the neuropsychological test between MetSd patients and controls was determined by a discriminant analysis (21) in which we included the significant tests obtained in the group comparison analysis, using a stepwise method.

RESULTS
There were no differences between the groups in age (t=1. The group comparison analyses showed significant differences in some cognitive functions (Table 1)  format. This test assesses the ability to organize the elements of the problems as well as working memory and attention resources. (19) The differences between patients and controls remained after controlling for the possible effect of education and gender. In contrast, several cognitive functions remained unaffected by the syndrome. We found non-significant differences in: values compared to controls. WM changes were not related to any isolated vascular risk factor (28). Although the specific mechanism of this damage is not clear, the use of this non-invasive method provides information on subtle WM changes. Smallvessel disease could be responsible for the microstructural alteration, due to the chronic state of vascular dysregulation in the brain. Therefore, the consequences of MetSd in vascular dysregulation may be taken as representing a specific neuropsychological profile. Indeed, DTI results of previous studies are in agreement with our neuropsychological findings: WM degeneration associated to aging, specifically in the frontal lobe, is related to slower processing and working memory deficits (29). Therefore, the anterior-predominance in the deterioration agrees with the neuropsychological deficit observed in our sample of patients.
Similarly, the preservation of other cognitive domains assessed indicated the integrity of posterior parts of the brain and their connections.
Our results suggest that MetSd is a prodromal state for mild vascular cognitive impairment, in the sense that it implies a specific cognitive profile with attentional and executive function deficits (30) as well as psychomotor slowness and immediate memory deficits. showed that the CPT-II reaction time and the Blocks subtest of WAIS-III are the variables that best discriminate between patients and controls. We propose that these measures should be taken in account in clinical protocols, because they can provide information about the evolution of the neuropsychological profile in MetSd patients and the differences between this profile and that of normal ageing.
While our findings are suggestive, the study is limited in several ways. The most important limitation is the sample size, due to the high comorbidity of MetSd and ageing with some of the exclusion criteria such as vascular pathology. Similarly, due to the sample size we were unable to study the effect of gender in depth. It would be particularly interesting to study the interaction effect of gender and age, and the associated risk of MetSd.
Future studies with groups of intermediate subjects with one or two risk factors and larger sample sizes may establish whether MetSd really causes a syndromic effect in the central nervous system that is greater than the sum of its parts.
In conclusion, MetSd patients show a specific neuropsychological profile. The detection of its specific manifestations appears to be the next step in preventing cognitive decline. We therefore propose that MetSd be included as a variable in research studies on cognitive impairment associated with ageing, in order to study its role in the differences between normal and successful ageing.