401
Journal of Alzheimer’s Disease 41 (2014) 401–410 DOI 10.3233/JAD-132480 IOS Press
Apolipoprotein E Genotype and Cerebrovascular Alterations Can Influence Conversion to Dementia in Patients with Mild Cognitive Impairment Giovanna Viticchia , Lorenzo Falsettib , Fabrizio Vernieric , Claudia Altamurac , Riccardo Altavillac , Simona Luzzia , Marco Bartolinia , Leandro Provincialia and Mauro Silvestrinia,∗ a Neurological
Clinic, Marche Polytechnic University, Ancona, Italy and Subintensive Medicine, Ospedali Riuniti Ancona, Italy c Neurology Unit, Campus Bio-Medico University, Rome, Italy b Internal
Handling Associate Editor: Jack de la Torre
Accepted 29 January 2014
Abstract. The presence of apolipoprotein E (APOE) ε4 allele is the only recognized genetic risk factor for the sporadic form of Alzheimer’s disease. The aim of this study was to investigate the relationship between APOE genotype and the functional and anatomic status of cerebral vessels in patients with mild cognitive impairment (MCI). Moreover, we explored whether the possible correlation between APOE genotype and cerebrovascular parameters influences the risk of conversion from MCI to dementia. 75 MCI patients underwent a complete neuropsychological battery at baseline and after 24 months to evaluate the possible conversion to dementia. Ultrasound assessment of neck and intracranial vessels was performed to assess common carotid artery intima-media thickness (IMT), plaque index, and cerebrovascular reactivity (breath-holding index, BHI). APOE genotype was determined to classify patients as carriers (APOE ε4+) and non-carriers (APOE ε4−). Pathologic values of BHI and IMT were significantly more common in ε4 carriers than in non-carriers [OR 6.603 (95% CI: 1.678–25.997), p < 0.05 and OR 5.195 (95% CI 1.319–20.464), p < 0.05; logistic regression adjusted model]. The risk of conversion to dementia was significantly higher in APOE ε4+ patients than in APOE ε4− ones (OR: 6.818; 95% CI:1.894–24.545, p = 0.003). A path-analysis model showed that APOE genotype influences the progression to dementia directly and indirectly by increasing the risk of pathologic values of IMT or BHI. Our data, besides confirming an increased susceptibility of MCI patients with APOE ε4 to develop dementia, show an association between functional and anatomic impairment of the cerebral vessels and APOE ε4+ genotype. Keywords: Alzheimer’s disease, apolipoproteins E, cerebrovascular diseases, cognition, ultrasonography
INTRODUCTION Alzheimer’s disease (AD) is the most common form of dementia, affecting millions of people with a ∗ Correspondence to: Mauro Silvestrini, Neurological Clinic, Marche Polytechnic University, Via Conca 1, 60020 Ancona, Italy. Tel.: +39 071 596 5668; Fax: +39 071 887262; E-mail: [email protected].
relevant and increasing cost for public health [1]. In the absence of effective treatments, preventive measures to lower the incidence of AD through the management of potential or actual risk factors are a reasonable clinical strategy. For this reason, identification of preclinical stages of dementia is of paramount importance. In this respect, the condition of mild cognitive impairment (MCI) is of particular interest since 20–40% of MCI patients can gradually develop AD [2].
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
402
G. Viticchi et al. / ApoE, Vascular Impairment, and Dementia
Modifiable and unmodifiable risk factors for AD have been identified. Among the modifiable ones, many studies focus on the relevance of vascular risk factors [3–5]. The real contribution of each factor is under debate for the inconsistent results obtained in different investigations [6]. Furthermore, a large amount of evidence reports a correlation between AD and anatomic and functional impairment of cerebral circulation. In the last few years, the contribution of cerebral macro- and micro-circulation derangement on cognitive impairment has been widely investigated and a narrow relationship between cerebral circulation failure and progression of cognitive deterioration has been described in patients with MCI or AD [7–12]. Among the unmodifiable risk factors, besides amyloid- protein precursor (APP) or presenilin 1 and 2, which are associated with the familiar form of AD, the only recognized genetic predisposition for sporadic AD is the presence of the apolipoprotein E (APOE) ε4 allele [13]. APOE ε4 allele in both homozygosis and heterozygosis increases the risk of developing AD [14]. APOE is synthetized in several different organs, including liver, skin, and brain. The different isoforms modulate amyloid- aggregation and clearance [15]. Moreover, it is implicated in several other functions such as inflammation, neurodevelopment, synaptic plasticity, and, above all, lipid metabolism [15]. In particular, it encodes for a protein component of lipoproteins and is involved in the transport of lipids and employed in neuronal cellular repair [16]. The presence of APOE ε4 allele also predisposes a subject to an increased risk of cardiovascular and cerebrovascular pathologies [11]. In mice, after a brain injury, lipids, cholesterol, and APOE protein production are intensified, and this mechanism seems to be under the control of APOE gene. If ε4 haplotype is present, the whole process of damage repair appears to be less effective [17]. Finally, APOE action is strongly linked to microcirculation derangement. APOE ε4 allele carriers are prone to have higher cholesterol levels [18] and higher intracranial atherosclerosis if compared with noncarriers [19]. Currently, there is no clear evidence of a relationship among APOE, cognitive deterioration, and vascular impairment. In this respect, a definition of the mechanisms of interaction between genetic and anatomic and functional vessel characteristics and the effects on cognitive deterioration would be interesting to better define some pathophysiologic aspects related to the prognosis of patients with cognitive impairment. The aim of this study was to evaluate whether in MCI patients with APOE ε4 allele, the increased suscepti-
bility to develop dementia can be associated with an impairment of the functional and anatomic status of cerebral vessels. MATERIALS AND METHODS Patients were selected from consecutive subjects referred to our dementia outpatient services by general practitioners during a 1-year period (from May 2011) for suspected cognitive impairment. The only inclusion criterion was a diagnosis of MCI according to the National Institute on Aging and the Alzheimer’s Association diagnostic criteria [20]. In particular, to select subjects with the highest risk of conversion to dementia, we only recruited amnestic-MCI patients based on Petersen’s criteria [21]. Exclusion criteria were: a) basal MMSE score 1.5 mm measured from the media-adventitia interface to the intima-lumen interface [25]. The plaque degree was calculated in all arterial vessels. Carotid plaques were defined as a thickening over 1.2 mm which does not include the whole vessel surface. In each arterial segment, the plaque degree was quantified as follows: 0: no plaque, 1: one small plaque 50% of vessel diameter or multiple plaques with at least one medium plaque. The plaque index was calculated by adding the scores of the right and left carotid arteries [26]. The measurements of IMT were performed on the segment of 1.5 cm of the common carotid that precedes the carotid bifurcation [25]. A longitudinal image of the distal segment of common carotid arteries was taken, and the measurement was obtained with an automatic system at the thickest point where there were no plaques on the proximal and distal wall. IMT was defined as the mean of the maximum thickness for proximal and distal wall of the common carotid of both sides. Based on the results of previous investigations about the predictive value of IMT on the risk of cardiovascular disease and cognitive impairment, a cut-off for pathological values was set at 1.0 mm [8, 27]. Intracranial circle was evaluated by means of TCD (Multidop X DWL; ElektronischeSysteme GmbH, Germany). Cerebrovascular reactivity to the hypercapnia (VMR) was calculated with the BHI [28]. This index is obtained by dividing the percent increase in mean flow velocity (MFV) occurring during breathholding by the length of time (s) the subjects hold their breath after a normal inspiration: BHI = [(breathholding MFV - basal MFV/basal MFV) × 100/seconds of breath-holding]. Two transducers placed on the temporal bone window were used to obtain a bilateral continuous measurement of flow velocity of middle cerebral arteries. Subjects were requested to hold their breath for a period of 30 s. Breath-holding was monitored by a capnometer. For each subject, three recordings were obtained, and the BHI considered was the mean of all values obtained. On the basis of our previous results on the predictivity of cerebrovascular impairment on the risk of cerebrovascular disease and cognitive impairment, we defined normative ranges for BHI values and classified cerebral hemodynamic parameters as either impaired (BHI 1 on Wahlund’s criteria, 3 for the presence of serious general diseases, and 4 for relevant cardiopathies). Four subjects did not accept to participate in the study. Finally, we obtained a sample of 75 MCI patients.
G. Viticchi et al. / ApoE, Vascular Impairment, and Dementia Table 1 Baseline characteristics of MCI patients with (+) or without (−) APOE ε4 allele APOE ε4+ (n = 24)
APOE ε4− (n = 51)
Age (y ± SD) 72.91 (±7.68) Gender (males, %) 6 (25.0%) Hypertension (%) 14 (58.3%) Diabetes (%) 5 (20.8%) Dyslipidemia (%) 13 (54.2%) Smoking (%) 6 (25.0%) Antihypertensives 14 (58.3%) Antidiabetics 5 (20.8%) Statins 9 (37.5%) Antiplatelet 7 (29.2%) Education (y ± SD) 7.58 (±3.49) MMSE at T0 (±SD) 24.79 (±2.17)
75.14 (±5.79) 20 (39.2%) 34 (66.7%) 4 (7.8%) 23 (45.1%) 8 (15.7%) 32 (62.7%) 4 (7.8%) 19 (37.3%) 22 (43.1) 8.43 (±3.84) 26.35 (±2.70)
Difference p = 0.168 p = 0.301 p = 0.607 p = 0.135 p = 0.621 p = 0.355 p = 0.801 p = 0.135 p = 1.000 p = 0.313 p = 0.362 p = 0.016
Table 2 Differences between genotype groups in terms of breath-holding index (BHI) and carotid intima-media thickness (IMT) treated as dichotomous variables BHI APOE ε4+ APOE ε4−
Pathologic (1 mm)
Normal (≤1 mm)
Total
13 (50.0%) 13 (50%)
11 (22.4%) 38 (77.6%)
24 (32.0%) 51 (68.0%)
Odds ratio (95% CI)
p
2 × 2 table 3.455 (1.245–9.584) Logistic 5.195 (1.319–20.464) regression
0.020 (χ2 ) 0.018
According with APOE ε4 genotype, 24 subjects were APOE ε4+ and 51 APOE ε4−. Demographic and clinical data of the two groups are shown in Table 1. No significant difference was observed between the groups in terms of demographic characteristics. Prevalence of hypertension, diabetes, smoking habit, and dyslipidemia were not statistically different between APOE ε4+ and APOE ε4− patients, and the two groups turned out to be similar in terms of antihypertensives, antidiabetics, antiplatelet agents, and statins. MMSE mean score was significantly lower in APOE ε4+ with respect to APOE ε4− patients at T0. APOE ε4+ patients had a higher probability of a pathologic BHI than APOE ε4− patients [OR 6.603 (95% CI: 1.678–25.997); p < 0.05]. Logistic regression results, adjusted for covariates and risk factors, are summarized in Table 2. The probability of a pathologic IMT was also significantly higher among APOE ε4+ than APOE
405
Table 3 Difference in the rate of conversion from MCI to dementia based on the patients’ APOE genotype and breath-holding index (BHI) and carotid intima-media thickness (IMT) treated as dichotomous variables APOE ε4 Genotype ε4 carriers ε4 non-carriers IMT Pathologic (>1 mm) Normal (≤1 mm) BHI Pathologic (
Journal of Alzheimer’s Disease 41 (2014) 401–410 DOI 10.3233/JAD-132480 IOS Press
Apolipoprotein E Genotype and Cerebrovascular Alterations Can Influence Conversion to Dementia in Patients with Mild Cognitive Impairment Giovanna Viticchia , Lorenzo Falsettib , Fabrizio Vernieric , Claudia Altamurac , Riccardo Altavillac , Simona Luzzia , Marco Bartolinia , Leandro Provincialia and Mauro Silvestrinia,∗ a Neurological
Clinic, Marche Polytechnic University, Ancona, Italy and Subintensive Medicine, Ospedali Riuniti Ancona, Italy c Neurology Unit, Campus Bio-Medico University, Rome, Italy b Internal
Handling Associate Editor: Jack de la Torre
Accepted 29 January 2014
Abstract. The presence of apolipoprotein E (APOE) ε4 allele is the only recognized genetic risk factor for the sporadic form of Alzheimer’s disease. The aim of this study was to investigate the relationship between APOE genotype and the functional and anatomic status of cerebral vessels in patients with mild cognitive impairment (MCI). Moreover, we explored whether the possible correlation between APOE genotype and cerebrovascular parameters influences the risk of conversion from MCI to dementia. 75 MCI patients underwent a complete neuropsychological battery at baseline and after 24 months to evaluate the possible conversion to dementia. Ultrasound assessment of neck and intracranial vessels was performed to assess common carotid artery intima-media thickness (IMT), plaque index, and cerebrovascular reactivity (breath-holding index, BHI). APOE genotype was determined to classify patients as carriers (APOE ε4+) and non-carriers (APOE ε4−). Pathologic values of BHI and IMT were significantly more common in ε4 carriers than in non-carriers [OR 6.603 (95% CI: 1.678–25.997), p < 0.05 and OR 5.195 (95% CI 1.319–20.464), p < 0.05; logistic regression adjusted model]. The risk of conversion to dementia was significantly higher in APOE ε4+ patients than in APOE ε4− ones (OR: 6.818; 95% CI:1.894–24.545, p = 0.003). A path-analysis model showed that APOE genotype influences the progression to dementia directly and indirectly by increasing the risk of pathologic values of IMT or BHI. Our data, besides confirming an increased susceptibility of MCI patients with APOE ε4 to develop dementia, show an association between functional and anatomic impairment of the cerebral vessels and APOE ε4+ genotype. Keywords: Alzheimer’s disease, apolipoproteins E, cerebrovascular diseases, cognition, ultrasonography
INTRODUCTION Alzheimer’s disease (AD) is the most common form of dementia, affecting millions of people with a ∗ Correspondence to: Mauro Silvestrini, Neurological Clinic, Marche Polytechnic University, Via Conca 1, 60020 Ancona, Italy. Tel.: +39 071 596 5668; Fax: +39 071 887262; E-mail: [email protected].
relevant and increasing cost for public health [1]. In the absence of effective treatments, preventive measures to lower the incidence of AD through the management of potential or actual risk factors are a reasonable clinical strategy. For this reason, identification of preclinical stages of dementia is of paramount importance. In this respect, the condition of mild cognitive impairment (MCI) is of particular interest since 20–40% of MCI patients can gradually develop AD [2].
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
402
G. Viticchi et al. / ApoE, Vascular Impairment, and Dementia
Modifiable and unmodifiable risk factors for AD have been identified. Among the modifiable ones, many studies focus on the relevance of vascular risk factors [3–5]. The real contribution of each factor is under debate for the inconsistent results obtained in different investigations [6]. Furthermore, a large amount of evidence reports a correlation between AD and anatomic and functional impairment of cerebral circulation. In the last few years, the contribution of cerebral macro- and micro-circulation derangement on cognitive impairment has been widely investigated and a narrow relationship between cerebral circulation failure and progression of cognitive deterioration has been described in patients with MCI or AD [7–12]. Among the unmodifiable risk factors, besides amyloid- protein precursor (APP) or presenilin 1 and 2, which are associated with the familiar form of AD, the only recognized genetic predisposition for sporadic AD is the presence of the apolipoprotein E (APOE) ε4 allele [13]. APOE ε4 allele in both homozygosis and heterozygosis increases the risk of developing AD [14]. APOE is synthetized in several different organs, including liver, skin, and brain. The different isoforms modulate amyloid- aggregation and clearance [15]. Moreover, it is implicated in several other functions such as inflammation, neurodevelopment, synaptic plasticity, and, above all, lipid metabolism [15]. In particular, it encodes for a protein component of lipoproteins and is involved in the transport of lipids and employed in neuronal cellular repair [16]. The presence of APOE ε4 allele also predisposes a subject to an increased risk of cardiovascular and cerebrovascular pathologies [11]. In mice, after a brain injury, lipids, cholesterol, and APOE protein production are intensified, and this mechanism seems to be under the control of APOE gene. If ε4 haplotype is present, the whole process of damage repair appears to be less effective [17]. Finally, APOE action is strongly linked to microcirculation derangement. APOE ε4 allele carriers are prone to have higher cholesterol levels [18] and higher intracranial atherosclerosis if compared with noncarriers [19]. Currently, there is no clear evidence of a relationship among APOE, cognitive deterioration, and vascular impairment. In this respect, a definition of the mechanisms of interaction between genetic and anatomic and functional vessel characteristics and the effects on cognitive deterioration would be interesting to better define some pathophysiologic aspects related to the prognosis of patients with cognitive impairment. The aim of this study was to evaluate whether in MCI patients with APOE ε4 allele, the increased suscepti-
bility to develop dementia can be associated with an impairment of the functional and anatomic status of cerebral vessels. MATERIALS AND METHODS Patients were selected from consecutive subjects referred to our dementia outpatient services by general practitioners during a 1-year period (from May 2011) for suspected cognitive impairment. The only inclusion criterion was a diagnosis of MCI according to the National Institute on Aging and the Alzheimer’s Association diagnostic criteria [20]. In particular, to select subjects with the highest risk of conversion to dementia, we only recruited amnestic-MCI patients based on Petersen’s criteria [21]. Exclusion criteria were: a) basal MMSE score 1.5 mm measured from the media-adventitia interface to the intima-lumen interface [25]. The plaque degree was calculated in all arterial vessels. Carotid plaques were defined as a thickening over 1.2 mm which does not include the whole vessel surface. In each arterial segment, the plaque degree was quantified as follows: 0: no plaque, 1: one small plaque 50% of vessel diameter or multiple plaques with at least one medium plaque. The plaque index was calculated by adding the scores of the right and left carotid arteries [26]. The measurements of IMT were performed on the segment of 1.5 cm of the common carotid that precedes the carotid bifurcation [25]. A longitudinal image of the distal segment of common carotid arteries was taken, and the measurement was obtained with an automatic system at the thickest point where there were no plaques on the proximal and distal wall. IMT was defined as the mean of the maximum thickness for proximal and distal wall of the common carotid of both sides. Based on the results of previous investigations about the predictive value of IMT on the risk of cardiovascular disease and cognitive impairment, a cut-off for pathological values was set at 1.0 mm [8, 27]. Intracranial circle was evaluated by means of TCD (Multidop X DWL; ElektronischeSysteme GmbH, Germany). Cerebrovascular reactivity to the hypercapnia (VMR) was calculated with the BHI [28]. This index is obtained by dividing the percent increase in mean flow velocity (MFV) occurring during breathholding by the length of time (s) the subjects hold their breath after a normal inspiration: BHI = [(breathholding MFV - basal MFV/basal MFV) × 100/seconds of breath-holding]. Two transducers placed on the temporal bone window were used to obtain a bilateral continuous measurement of flow velocity of middle cerebral arteries. Subjects were requested to hold their breath for a period of 30 s. Breath-holding was monitored by a capnometer. For each subject, three recordings were obtained, and the BHI considered was the mean of all values obtained. On the basis of our previous results on the predictivity of cerebrovascular impairment on the risk of cerebrovascular disease and cognitive impairment, we defined normative ranges for BHI values and classified cerebral hemodynamic parameters as either impaired (BHI 1 on Wahlund’s criteria, 3 for the presence of serious general diseases, and 4 for relevant cardiopathies). Four subjects did not accept to participate in the study. Finally, we obtained a sample of 75 MCI patients.
G. Viticchi et al. / ApoE, Vascular Impairment, and Dementia Table 1 Baseline characteristics of MCI patients with (+) or without (−) APOE ε4 allele APOE ε4+ (n = 24)
APOE ε4− (n = 51)
Age (y ± SD) 72.91 (±7.68) Gender (males, %) 6 (25.0%) Hypertension (%) 14 (58.3%) Diabetes (%) 5 (20.8%) Dyslipidemia (%) 13 (54.2%) Smoking (%) 6 (25.0%) Antihypertensives 14 (58.3%) Antidiabetics 5 (20.8%) Statins 9 (37.5%) Antiplatelet 7 (29.2%) Education (y ± SD) 7.58 (±3.49) MMSE at T0 (±SD) 24.79 (±2.17)
75.14 (±5.79) 20 (39.2%) 34 (66.7%) 4 (7.8%) 23 (45.1%) 8 (15.7%) 32 (62.7%) 4 (7.8%) 19 (37.3%) 22 (43.1) 8.43 (±3.84) 26.35 (±2.70)
Difference p = 0.168 p = 0.301 p = 0.607 p = 0.135 p = 0.621 p = 0.355 p = 0.801 p = 0.135 p = 1.000 p = 0.313 p = 0.362 p = 0.016
Table 2 Differences between genotype groups in terms of breath-holding index (BHI) and carotid intima-media thickness (IMT) treated as dichotomous variables BHI APOE ε4+ APOE ε4−
Pathologic (1 mm)
Normal (≤1 mm)
Total
13 (50.0%) 13 (50%)
11 (22.4%) 38 (77.6%)
24 (32.0%) 51 (68.0%)
Odds ratio (95% CI)
p
2 × 2 table 3.455 (1.245–9.584) Logistic 5.195 (1.319–20.464) regression
0.020 (χ2 ) 0.018
According with APOE ε4 genotype, 24 subjects were APOE ε4+ and 51 APOE ε4−. Demographic and clinical data of the two groups are shown in Table 1. No significant difference was observed between the groups in terms of demographic characteristics. Prevalence of hypertension, diabetes, smoking habit, and dyslipidemia were not statistically different between APOE ε4+ and APOE ε4− patients, and the two groups turned out to be similar in terms of antihypertensives, antidiabetics, antiplatelet agents, and statins. MMSE mean score was significantly lower in APOE ε4+ with respect to APOE ε4− patients at T0. APOE ε4+ patients had a higher probability of a pathologic BHI than APOE ε4− patients [OR 6.603 (95% CI: 1.678–25.997); p < 0.05]. Logistic regression results, adjusted for covariates and risk factors, are summarized in Table 2. The probability of a pathologic IMT was also significantly higher among APOE ε4+ than APOE
405
Table 3 Difference in the rate of conversion from MCI to dementia based on the patients’ APOE genotype and breath-holding index (BHI) and carotid intima-media thickness (IMT) treated as dichotomous variables APOE ε4 Genotype ε4 carriers ε4 non-carriers IMT Pathologic (>1 mm) Normal (≤1 mm) BHI Pathologic (
