Abstract
Background
Presence of significant subjective complaints about cognition (SCD) is considered the first behavioral manifestation of Alzheimer disease (AD). However, SCD has not yet overcome ...the challenge of becoming a reliable preclinical AD marker. Severity indices were proposed to improve the accuracy of complaints when predicting the risk of AD (Jessen et al., 2010).
Our aim was to compare the predictive accuracy of ML algorithms using a more (95%ile) or less (5%ile) restrictive cut‐off point in severity of complaints for classification in Low (LSC) and High (HSC) subjective complaints groups.
Method
One hundred and ninety‐nine participants from the Compostela Aging Study (ComPAS) were identified at baseline as SCDs and completed three follow‐up measurements (54‐72 months). Considering their total scores in the Subjective Memory Questionnaire, participants were classified as LSC and HSC following using two distinct cut‐off points: complaint scores above or below than 5%ile vs 95%ile. Participants were labeled as ‘worsening’ or ‘stable’ based on their progression or not to MCI or AD. ML classifier algorithms (Random Forest, Support Vector Machine, Extra Tree) were applied to forty‐one measures (socio‐demographic, time, health, cognitive, behavioral, cognitive reserve) collected at baseline.
Results
The best performing model was the Random Forest (
95%ile
: PPV =.87; Sensibility =.92; Specificity =.38;
5%ile
: PPV =.66; Sensibility=.61; Specificity =.51). The confusion matrix (Figure 1) showed that: a) both, the more (95%ile) and the less (5%ile) restrictive criteria mostly classified the HSC‐stable participants as LSC‐stable; and b) criterion 5%ile, but not 95%ile, was able to differentially identify progressors based on their complaints. Episodic memory, executive functions, depression, cognitive reserve and progression timing measures assume the highest levels of importance in the ML algorithm that differentially predicts the progression of SCD to MCI and AD according to the level of complaints.
Conclusions
For both the criteria, algorithms failed to successfully identify stable SCD participants. The less restrictive criterion resulted in a better classification than the more lenient one to differentially identify LSC and HSC participants who get worse. Cognitive, affective, cognitive reserve and progression timing were the more prominent variables in conforming the predictive algorithm.
Abstract
Background
Neurobehavioral Symptoms (NPS) have been usually measured with the Neuropsychiatric‐Inventory (NPI‐Q) (Kaufer et al., 2000) in pre‐dementia individuals. However, this instrument ...was designed to dementia states. The Mild Behavioral Impairment Checklist (MBI‐C) (Ismail et al., 2017) is an instrument developed to evaluate NPS in pre‐dementia states, including Mild cognitive impairment (MCI) and people with Subjective Cognitive Complaints (SCC). Evidence of longitudinal behavioral change obtained with the MBI‐C is still scarce. Our objective was to compare NPS scores, measured with the NPI‐Q and the MBI‐C, in MCI and SCC participants, at follow‐up.
Method
Two hundred forty participants from the Compostela Ageing Study recruited from primary care health centers, were classified into two groups, MCI (84) and SCC (166). Socio‐demographic, neuropsychological and NPS measures, including MBI‐C and NPI‐Q total scores on severity, were collected at baseline and at follow‐up (mean interval, 24 months) (Table 1). Instrument (NPI‐Q and MBI‐C) total score differences as a function of measurement Time (baseline vs follow‐up) were analyzed using repeated measure ANOVAs including Group (SCC vs MCI) as inter‐subject factor.
Results
Significant main effects of Time,
F
(1, 251)=7.68,
p
= .006,
ηp
2
=.030,
observed power
=.789, and Group,
F
(1, 251)=9.88,
p
= .002,
ηp
2
=.038,
observed power
=.879, were observed for the NPI‐Q total scores (Figure 1). Time*Group interaction was not found. Mixed ANOVA for the MBI‐C total score showed significant main effect of Group,
F
(1, 248)=13.93,
p
< .001,
ηp
2
=.053,
observed power
=.961, and Time*Group interaction,
F
(1, 248)=4.89,
p
= .028,
ηp
2
=.020,
observed power
=.596. Post hoc Bonferroni tests for the MBI‐C total score showed (Figure 2) significantly higher behavioral impairment in MCI than in SCC group in both time measurements (baseline and follow‐up).
Conclusions
Time main effect in NPI‐Q total scores showed decreases in severity of behavioral symptoms in the follow‐up measurement in spite of the Group (SCC vs MCI). On the contrary, MBI‐C scores pointed‐out slight increases in the follow‐up measurement and, considering the descriptive trends, particularly in MCI group. Only the MBI‐C was able to detect significant differences between MCI and SCC groups both in baseline and follow‐up measurements.
Background
Neurobehavioral Symptoms (NPS) have been usually measured with the Neuropsychiatric‐Inventory (NPI‐Q) (Kaufer et al., 2000) in pre‐dementia individuals. However, this instrument was ...designed to dementia states. The Mild Behavioral Impairment Checklist (MBI‐C) (Ismail et al., 2017) is an instrument developed to evaluate NPS in pre‐dementia states, including Mild cognitive impairment (MCI) and people with Subjective Cognitive Complaints (SCC). Evidence of longitudinal behavioral change obtained with the MBI‐C is still scarce. Our objective was to compare NPS scores, measured with the NPI‐Q and the MBI‐C, in MCI and SCC participants, at follow‐up.
Method
Two hundred forty participants from the Compostela Ageing Study recruited from primary care health centers, were classified into two groups, MCI (84) and SCC (166). Socio‐demographic, neuropsychological and NPS measures, including MBI‐C and NPI‐Q total scores on severity, were collected at baseline and at follow‐up (mean interval, 24 months) (Table 1). Instrument (NPI‐Q and MBI‐C) total score differences as a function of measurement Time (baseline vs follow‐up) were analyzed using repeated measure ANOVAs including Group (SCC vs MCI) as inter‐subject factor.
Results
Significant main effects of Time, F(1, 251)=7.68, p = .006, ηp
2=.030, observed power=.789, and Group, F(1, 251)=9.88, p = .002, ηp
2=.038, observed power=.879, were observed for the NPI‐Q total scores (Figure 1). Time*Group interaction was not found. Mixed ANOVA for the MBI‐C total score showed significant main effect of Group, F(1, 248)=13.93, p < .001, ηp
2=.053, observed power=.961, and Time*Group interaction, F(1, 248)=4.89, p = .028, ηp
2=.020, observed power=.596. Post hoc Bonferroni tests for the MBI‐C total score showed (Figure 2) significantly higher behavioral impairment in MCI than in SCC group in both time measurements (baseline and follow‐up).
Conclusions
Time main effect in NPI‐Q total scores showed decreases in severity of behavioral symptoms in the follow‐up measurement in spite of the Group (SCC vs MCI). On the contrary, MBI‐C scores pointed‐out slight increases in the follow‐up measurement and, considering the descriptive trends, particularly in MCI group. Only the MBI‐C was able to detect significant differences between MCI and SCC groups both in baseline and follow‐up measurements.
Abstract
Background
Characterization of mild cognitive impairment (MCI) for clinical and research purposes involves identification of biological and neuropsychological markers in order to predict ...possible progression to dementia. The main aim of this work was to analyze the correlation of anthropometric measurements and plasmatic levels of leptin, testosterone and estrogens with cognitive, CSF and MRI regional atrophy markers in a sample of MCI participants.
Methods
Eighty‐one patients from the Compostela Aging Study diagnosed as MCI according to the Petersen criteria (Petersen, 2004; Albert et al., 2011) underwent neuropsychological assessment (CAMCOG‐R memory, CVLT short and long delayed free recall, subjective memory complains), CSF (Aβ
42
, t‐Tau and p‐Tau) and plasmatic analyses as well as an MRI study (parahippocampal and entorhinal cortex thickness measurements) (Table 1). Spearman’s correlation was used to evaluate the relationship between anthropometric measurements and blood test results with memory scores, CSF and MRI regional atrophy markers.
Results
Episodic memory scores correlated with CSF measurements (Aβ
42
, t‐Tau, p‐Tau) and parahippocampal and entorhinal cortical thickness (right and left) (Table 2). BMI, suprailiac, pectoral and subscapular skinfolds, circumference of waist, hip, arm and calf were negatively correlated with t‐Tau and p‐Tau levels. BMI, triccipital, thigh and pectoral skin folds as well as calf circumference were positively correlated with parahippocampal and entorhinal cortical thickness. Suprailiac, triccipital and thigh skin folds were positively correlated with episodic memory scores. Leptin plasma levels showed a negative correlation with CSF biomarkers (t‐Tau and p‐Tau) and a positive correlation with parahippocampal cortex thickness. Testosterone and estradiol levels showed a negative correlation with scores on memory tests and enthorhinal cortical thickness.
Conclusions
Our findings suggest that body composition is associated with episodic memory function, CSF biomarkers and cortical atrophy in patients with MCI. Leptin levels and sexual hormones may play a role in this association. These results point to a role of fat tissue and its regulating mechanisms in AD pathology and progression from MCI to dementia.
Background
Characterization of mild cognitive impairment (MCI) for clinical and research purposes involves identification of biological and neuropsychological markers in order to predict possible ...progression to dementia. The main aim of this work was to analyze the correlation of anthropometric measurements and plasmatic levels of leptin, testosterone and estrogens with cognitive, CSF and MRI regional atrophy markers in a sample of MCI participants.
Methods
Eighty‐one patients from the Compostela Aging Study diagnosed as MCI according to the Petersen criteria (Petersen, 2004; Albert et al., 2011) underwent neuropsychological assessment (CAMCOG‐R memory, CVLT short and long delayed free recall, subjective memory complains), CSF (Aβ42, t‐Tau and p‐Tau) and plasmatic analyses as well as an MRI study (parahippocampal and entorhinal cortex thickness measurements) (Table 1). Spearman’s correlation was used to evaluate the relationship between anthropometric measurements and blood test results with memory scores, CSF and MRI regional atrophy markers.
Results
Episodic memory scores correlated with CSF measurements (Aβ42, t‐Tau, p‐Tau) and parahippocampal and entorhinal cortical thickness (right and left) (Table 2). BMI, suprailiac, pectoral and subscapular skinfolds, circumference of waist, hip, arm and calf were negatively correlated with t‐Tau and p‐Tau levels. BMI, triccipital, thigh and pectoral skin folds as well as calf circumference were positively correlated with parahippocampal and entorhinal cortical thickness. Suprailiac, triccipital and thigh skin folds were positively correlated with episodic memory scores. Leptin plasma levels showed a negative correlation with CSF biomarkers (t‐Tau and p‐Tau) and a positive correlation with parahippocampal cortex thickness. Testosterone and estradiol levels showed a negative correlation with scores on memory tests and enthorhinal cortical thickness.
Conclusions
Our findings suggest that body composition is associated with episodic memory function, CSF biomarkers and cortical atrophy in patients with MCI. Leptin levels and sexual hormones may play a role in this association. These results point to a role of fat tissue and its regulating mechanisms in AD pathology and progression from MCI to dementia.
PDF
