203
Journal of Alzheimer’s Disease 41 (2014) 203–211 DOI 10.3233/JAD-132180 IOS Press
Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians Eman Al-khateeba,1,∗ , Ebaa Al-zayadneha , Osama Al-dalahmahb , Zeinab Alawadib , Faisal khatiba , Randa Naffaa and Yanal Shafagoja,1 a Department b Medical
of Physiology & Biochemistry, University of Jordan, Amman, Jordan students, Faculty of Medicine, University of Jordan, Amman, Jordan
Accepted 15 January 2014
Abstract. The purpose of the current study was to examine the association of serum copper and lipid concentrations with changes in cognitive function in elderly Jordanian individuals. The study population consisted of two groups: 52 dementia patients and 50 control subjects. All individuals were screened using the Mini-Mental State Examination and Clock Drawing Test. Serum copper and lipid profile were also assessed. Results were statistically evaluated at p < 0.05 level of significance. The dementia group had 10.1% higher copper level than control subjects that was not statistically significant. No significant differences could be found between the two groups in lipid profile levels. There was no significant correlation between serum copper, lipid profile, and cognitive decline in elderly Jordanians. Demographic variables indicated that educational level less than 12 years and illiterate demonstrated a 3.29 fold (p = 0.026) and 6.29 fold (p = 0.002) increase in risk of developing dementia, respectively. Coffee intake demonstrated a protective effect against cognitive decline with 6.25 fold lower risk with increased coffee intake. Keywords: Alzheimer’s disease, cholesterol, coffee, copper, dementia, lipid profile
Recent increase in the proportion and number of elderly in many developing countries has drawn attention to diseases related to this vulnerable age group [1, 2]. Older people in Jordan will continue to grow at a rate of 4.1% in the next forty years, from 220,000 to an expected 500,000 by 2030 and 1.25 million by 2050 [2]. One of the most significant age-related conditions that pose a challenge to health and welfare services is dementia. Alzheimer’s disease (AD) currently accounts for 60–80% of dementia cases, which is characterized by extracellular fibrillar amyloid- (A). Oxidative stress plays important role in the pathophysiology of AD with metals like zinc, copper, and iron being important pro-oxidant factors. Dyshomeostasis of these metals is thought to play 1 These
authors contributed equally. to: Eman Al-khateeb, MBChB, MSc, PhD, Department of Physiology and Biochemistry, Faculty of Medicine, University of Jordan, P.O. Box 13490, Amman 11942, Jordan. Tel.: +962 6 5355000 23479; Fax: +962 6 5300820; E-mails: [email protected]; [email protected]. ∗ Correspondence
a central role in the formation and neurotoxicity of A and neurofibrillary tangles [3–5]. Copper has been reported to bind to A in the N-terminal copper binding domain (CuBD-II) of the amyloid- protein precursor (APP) and increase its aggregation in vitro. APP reduces Cu(II) to Cu(I), leading to oxidative modification of APP [6]. Immunohistochemical studies have demonstrated that trace amount of copper prompt the APP upregulation leading to A deposition in the brain of cholesterol-fed mice [7]. In vivo, studies demonstrated that elevated serum copper levels differentiate between AD patients and age-matched normal individuals with good selectivity and specificity [8]. In the Chicago Health and Aging Project, high copper intake from food and supplement was linked to a significantly faster rate of cognitive decline in individuals that consumed a diet high in saturated and trans fats [9]. George Brewer suggested that copper toxicity, related to copper plumbing and ingestion of supplements containing copper, was partially causative of cognitive decline, dementia, and AD epidemic in
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
204
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians
developed countries [10]. Other studies reported a correlation between mild copper deficiency and mild to moderate AD [11]. Therefore the role of copper in the pathogenesis of dementia remains unclear. Cholesterol plays central role in promoting the production of A and possibly the progression of AD [12]. The role of elevated cholesterol in dementia remains unclear, with several studies demonstrating differences in the risk profile of hypercholesterolemia in AD patients and high cardiac risk patients [13]. Data from cross-sectional studies suggested an association between atherosclerosis and AD, with elevated serum cholesterol as the common risk factor [14]. Longitudinal studies suggested a relationship between elevated midlife cholesterol levels and late-life cognitive impairment or AD [15]. The Framingham study demonstrated that baseline and long-term average serum total cholesterol levels were not associated with the risk for incident AD [16]. Furthermore, some investigators found no difference in plasma HDL and LDL levels between AD and controls [17]. Higher levels of LDL cholesterol were found to be significantly associated with an increased risk of vascular dementia [18]. Recent research has focused on identifying biological markers of dementia. Such markers may play a key role in the early diagnosis and management of dementia. Moreover, medications that delay the course of dementia are available, further emphasizing the need for diagnostic biomarkers to identify incipient dementia. Measuring serum copper and lipid concentration is affordable, accurate, and minimally invasive thus representing an appealing biological marker for cognitive abnormalities. The aim of this study was to investigate possible associations between lipid profile, serum copper, and dementia in the elderly Jordanians. SUBJECTS AND METHODS Subjects The study population consisted of elderly Jordanians that were recruited from senior homes and Jordan University Hospital. The institutional review board of the University of Jordan/Faculty of Medicine approved the study and a written consent was provided by each subject or legal guardian. Study inclusion required individuals be over 60 years of age with no history of any relevant psychiatric disease and no systemic use of statins or other lipid lowering agents, psychotropic drugs, or substance abuse. Information about risk factors was gathered from a questionnaire completed
by all participants. The questionnaire covered sociodemographic characteristics, lifestyle (fat intake and coffee intake), family and medical history including chronic diseases and medications. Participants were asked whether they use low-calorie, low saturated fat diet comprising whole grains, vegetables, chicken, fish, and olive oil regularly or they use diets high in saturated fat (such as animal meats, dairy products). Regular consumption of coffee was defined as nearly every day. Initially, all subjects were screened with two scales, the Mini-Mental State Examination (MMSE) and the Clock Drawing Test (CDT). Elderly individuals who scored abnormally on these tests underwent further evaluation with a non-structured interview for dementia. The diagnosis of dementia was made according to the criteria of Diagnostic and Statistical Manual IV (DSM-IV). The severity of dementia was assessed by the Clinical Dementia Rating (CDR), dementia patients included were (CDR = 2) [19]. A total 165 individuals agreed to participate in the study, with 72 having abnormal MMSE and CDT scores. After the interview and the application of the DSM-IV criteria, only 52 subjects were eligible for the moderate dementia group. In parallel, a group of 50 controls was randomly selected from different sectors of the same hospital (relatives, caregivers, and visitors). Control group subjects scored normally on both rating scales, were functionally independent and cognitively healthy (CDR = 0). Mental scales The MMSE is widely used screening instrument for assessing dementia in older adults. It tests five areas of cognitive functions: orientation, registration, attention/calculation, recall, and language [20]. Since there is no standardized Arabic form, we translated it to the Arabic language and adapted it to the local Arabic dialogue. Taking into consideration the high rate of illiteracy among the elderly population, the cutoff point for the presence of cognitive impairment was set according to the corresponding levels of education. For the control group, a score at 21/30 was set for those with no formal education, 23/30 for those who received 12 years of education. Moderate dementia group for individuals scored no less than 11 [20]. Executive function and visio-spatial performance was measured by the CDT using a pre-drawn circle. CDT is highly correlated with the MMSE and other tests of cognition dysfunction. Normal clock-drawing ability reasonably excludes cognitive impairment. The
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians
test is easy to administer, takes very little time, and is easy to document graphically in clinical records [21]. Subjects with low education, advanced age, and depression commonly perform poorly [22, 23]. Nevertheless, CDT using a pre-drawn circle has good diagnostic utility for detecting persons with early dementia; educational level appears to influence optimal cut-off scores in comprehensive but not abbreviated CDT methods [24]. Blood samples collection Blood samples were routinely collected in the morning after an overnight fast. Samples were centrifuged within 30 min at 3500 rpm for 10 min and rapidly stored at 4◦ C until analysis.
205
and standard deviations for continuous variables (age, MMSE, CDT, copper, TC, TG, HDL and LDL) or frequency and percentages for categorical variables (gender, marital status, education level, coffee intake, and fat intake). The association between continuous variables and mental status was assessed by comparisons of means using t-test (control and dementia groups). The association between categorical variables and mental status (control and dementia groups) was examined utilizing Chi-square test and binary logistic regression. Bivariate correlations of the MMSE and CDT scores and the continuous variables was analyzed and presented as Pearson correlation coefficients. Statistical significance was set at p < 0.05. Multivarient linear regression was used to measure the predictive value of continuous variables on MMSE and CDT scores.
Lipid profile measurement RESULTS The lipid profile measurements included serum total cholesterol (TC), triglycerides (TG), high density lipoprotein (HDL), and low density lipoprotein (LDL). All samples were measured by Roche Diagnostics COBAS INTEGRA 800 Biochemistry analyzer based on an enzymatic colorimetric method. Copper measurement Serum copper levels were quantified using A Shimadzu AA6800 Atomic Absorption Spectroscopy with Graphite Furnace Atomization System using wizAArd 2.31 software. A standard addition method was applied to construct a calibration curve to quantify copper in the samples. Copper measurement in water samples Eight water samples were analyzed from different sources (tap water, filtered water, and bottled water) and the following areas: Amman (Tla’a al Ali, Juayda), Madaba, Salt, and Irbid. The water sources represented the same region of the recruited subjects. Standard atomic absorption conditions were utilized for copper, inductively coupled plasma mass spectrometry. A Shimadzu AAS6300 was used for analysis of copper concentration from water samples.
Demographic variables Fifty-two dementia patients and 50 control subjects were enrolled in the study. There were no significant difference between dementia and control groups for age (p = 0.215) and gender (p = 0.290), respectively (Table 1). Some demographic variables were studied as a possible risk factors associated with cognitive decline. Only educational level (p = 0.006) and coffee intake (p = 0.000) showed significant association (Table 1). Interestingly, individuals that were illiterate or had educational level less than 12 years demonstrated a 6.29 fold (p = 0.002) and 3.29 fold (p = 0.026) increase in risk of developing dementia respectively. The intake of coffee showed protective effect against cognitive decline by a 6.25 fold decrease in risk. Other variables like marital status and fat intake showed no significant correlation. Mental scales Table 2 demonstrated the results of MMSE and CDT in the two groups. There was a significant correlation between CDT and MMSE results (p = 0.000). CDT in dementia patients was 4.03 ± 1.36 and only 2.22 ± 1.36 in control subjects.
Statistical analysis
Biochemical parameters
Demographic, clinical, and laboratory data were coded and entered into SPSS for windows 16.0 (SPSS Inc., Chicago, IL). Data were summarized as means
Table 3 showed the biochemical parameters in dementia group and control subjects. Most of the parameters measured did not differ between
206
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians Table 1 Demographics characteristics of study participants with dementia versus control subjects
Characteristic Age∗ (y) Age range (y) Gender (%) Women Men Marital status (%) Single/divorced Married Education Level (%) More than 12 Less than 12 Illiterate Coffee intake (%) Yes No Fat intake (%) Olive oil Olive oil + others
Controls (n = 50)
Patients with dementia (n = 52)
d.f
p-value
OR
95% confidence interval
68.9 ± 7.11 (60–88)
70.7 ± 7.63 (60–94)
1, 100
0.215
–
–
1, 100
0.29
–
–
17 (34.0) 33 (66.0)
23 (44.2) 29 (55.8) 1, 99
0.14
0.52
0.214–1.249
11 (22) 39 (78)
18 (35.3) 33 (64.7) 2, 100
0.006†
1, 100
0.026† 0.002† 0.000†
3.29 6.29 0.16
1.151–9.378 2.010–19.654 0.066–0.37
1, 98
0.073
2.08
0.929–4.673
20 (40.0) 20 (40.0) 10 (20.0)
7 (13.5) 23 (44.2) 22 (42.3)
37(74.0) 13(26.0)
16 (30.8) 36 (69.2)
25 (51.0) 24 (49.0)
17 (33.3) 34 (66.7)
OR, odds ratio. † p-value is significant. ∗ Values shown = mean±standard deviation (SD). Table 2 Tests used in this study for participants with dementia versus controls Test MMSE∗ CDT∗
Controls (n = 50)
Patients with dementia (n = 52)
d.f
p-value
95% confidence interval
26.66 ± 2.78 2.22 ± 1.36
17.37 ± 3.49 4.03 ± 1.36
1, 100 1, 80
0.000† 0.000†
8.054–0.535 −2.407 – −1.202
MMSE, Mini-Mental State Examination; CDT, Clock Drawing Test. † p- value is significant. ∗ Values shown = mean±standard deviation (SD). Table 3 Biochemical parameter values of individuals with dementia versus controls Biochemical test
Controls (n = 50)
Patients with CI (n = 52)
d.f
95% confidence interval
p-value
Copper (g/dL) TC∗ (mg/dL) TG∗ (mg/dL) HDL∗ (mg/dL) LDL∗ (mg/dL)
114.55 ± 57.6 187.16 ± 42.29 166.51 ± 80.44 38.47 ± 10.45 111.58 ± 34.39
126.12 ± 71.78 186.39 ± 47.60 165.07 ± 90.43 41.35 ± 10.31 113.74 ± 37.97
1, 98 1, 99 1, 99 1, 99 1, 99
−37.468–14.319 −17.020–18.556 −32.36–35.26 −6.975–1.222 −16.468–12.153
0.377 0.932 0.933 0.167 0.765
TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein; CI, cognitive impairment. † p-value is significant. ∗ Values shown = mean±standard deviation (SD). Table 4 Correlation between biochemical test and MMSE and CDT tests for dementia patients MMSE MMSE
CDT
Pearson Correlation Sig. (2-tailed) n Pearson Correlation Sig. (2-tailed) n
1 52 −0.295 0.077 37
Age −0.380 0.005∗∗ 52 0.221 0.189 37
Copper −0.043 0.765 51 0.057 0.743 36
TC 0.089 0.534 51 −0.175 0.308 36
TG −0.026 0.857 51 −0.015 0.932 36
HDL
LDL
0.119 0.407 51 −0.138 0.423 36
0.038 0.793 51 −0.207 0.226 36
CDT −0.295 0.077 37 1 37
TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein; MMSE, Mini-Mental State Examination; CDT, Clock Drawing Test. ∗∗ Correlation is significant at the 0.01 level (2-tailed).
the two groups. However the value of serum copper was numerically higher in the dementia group (126.12 g ± 71.78) compared to control group (114.55 g ± 57.6). These differences did not reach statistical significance (p = 0.377) and the values
were within the normal physiological reference range between 70 g/dl and 142.7 g/dl [25]. The lipid profile analysis in the two groups, showed slightly higher than normal TG and LDL and lower HDL when compared to the NCEP healthy lipid profile
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians
207
Table 5 Linear regression between biochemical test and MMSE test for dementia patients Coefficientsa Model (Constant) TC TG HDL LDL copper age
Unstandardized coefficients B
Std. Error
29.550 0.037 −0.006 −0.011 −0.034 −0.444 −0.190
5.681 0.029 0.007 0.062 0.033 0.697 0.064
Standardized coefficients
95% Confidence interval for B
Beta
t
Sig.
Lower Bound
0.516 −0.143 −0.032 −0.373 −0.092 −0.416
5.202 1.290 −0.755 −0.176 −1.034 −0.638 −2.973
0.000 0.204 0.454 0.861 0.307 0.527 0.005∗
18.094 −0.021 −0.020 −0.137 −0.100 −1.849 −0.318
Upper Bound 41.006 0.096 0.009 0.115 0.032 0.961 −0.061
a Dependent Variable: MMSE. TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein. ∗ Regression
is significant at the 0.01 level (2-tailed). R square = 0.200.
guidelines (TC 60 mg/dl, LDL
Journal of Alzheimer’s Disease 41 (2014) 203–211 DOI 10.3233/JAD-132180 IOS Press
Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians Eman Al-khateeba,1,∗ , Ebaa Al-zayadneha , Osama Al-dalahmahb , Zeinab Alawadib , Faisal khatiba , Randa Naffaa and Yanal Shafagoja,1 a Department b Medical
of Physiology & Biochemistry, University of Jordan, Amman, Jordan students, Faculty of Medicine, University of Jordan, Amman, Jordan
Accepted 15 January 2014
Abstract. The purpose of the current study was to examine the association of serum copper and lipid concentrations with changes in cognitive function in elderly Jordanian individuals. The study population consisted of two groups: 52 dementia patients and 50 control subjects. All individuals were screened using the Mini-Mental State Examination and Clock Drawing Test. Serum copper and lipid profile were also assessed. Results were statistically evaluated at p < 0.05 level of significance. The dementia group had 10.1% higher copper level than control subjects that was not statistically significant. No significant differences could be found between the two groups in lipid profile levels. There was no significant correlation between serum copper, lipid profile, and cognitive decline in elderly Jordanians. Demographic variables indicated that educational level less than 12 years and illiterate demonstrated a 3.29 fold (p = 0.026) and 6.29 fold (p = 0.002) increase in risk of developing dementia, respectively. Coffee intake demonstrated a protective effect against cognitive decline with 6.25 fold lower risk with increased coffee intake. Keywords: Alzheimer’s disease, cholesterol, coffee, copper, dementia, lipid profile
Recent increase in the proportion and number of elderly in many developing countries has drawn attention to diseases related to this vulnerable age group [1, 2]. Older people in Jordan will continue to grow at a rate of 4.1% in the next forty years, from 220,000 to an expected 500,000 by 2030 and 1.25 million by 2050 [2]. One of the most significant age-related conditions that pose a challenge to health and welfare services is dementia. Alzheimer’s disease (AD) currently accounts for 60–80% of dementia cases, which is characterized by extracellular fibrillar amyloid- (A). Oxidative stress plays important role in the pathophysiology of AD with metals like zinc, copper, and iron being important pro-oxidant factors. Dyshomeostasis of these metals is thought to play 1 These
authors contributed equally. to: Eman Al-khateeb, MBChB, MSc, PhD, Department of Physiology and Biochemistry, Faculty of Medicine, University of Jordan, P.O. Box 13490, Amman 11942, Jordan. Tel.: +962 6 5355000 23479; Fax: +962 6 5300820; E-mails: [email protected]; [email protected]. ∗ Correspondence
a central role in the formation and neurotoxicity of A and neurofibrillary tangles [3–5]. Copper has been reported to bind to A in the N-terminal copper binding domain (CuBD-II) of the amyloid- protein precursor (APP) and increase its aggregation in vitro. APP reduces Cu(II) to Cu(I), leading to oxidative modification of APP [6]. Immunohistochemical studies have demonstrated that trace amount of copper prompt the APP upregulation leading to A deposition in the brain of cholesterol-fed mice [7]. In vivo, studies demonstrated that elevated serum copper levels differentiate between AD patients and age-matched normal individuals with good selectivity and specificity [8]. In the Chicago Health and Aging Project, high copper intake from food and supplement was linked to a significantly faster rate of cognitive decline in individuals that consumed a diet high in saturated and trans fats [9]. George Brewer suggested that copper toxicity, related to copper plumbing and ingestion of supplements containing copper, was partially causative of cognitive decline, dementia, and AD epidemic in
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
204
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians
developed countries [10]. Other studies reported a correlation between mild copper deficiency and mild to moderate AD [11]. Therefore the role of copper in the pathogenesis of dementia remains unclear. Cholesterol plays central role in promoting the production of A and possibly the progression of AD [12]. The role of elevated cholesterol in dementia remains unclear, with several studies demonstrating differences in the risk profile of hypercholesterolemia in AD patients and high cardiac risk patients [13]. Data from cross-sectional studies suggested an association between atherosclerosis and AD, with elevated serum cholesterol as the common risk factor [14]. Longitudinal studies suggested a relationship between elevated midlife cholesterol levels and late-life cognitive impairment or AD [15]. The Framingham study demonstrated that baseline and long-term average serum total cholesterol levels were not associated with the risk for incident AD [16]. Furthermore, some investigators found no difference in plasma HDL and LDL levels between AD and controls [17]. Higher levels of LDL cholesterol were found to be significantly associated with an increased risk of vascular dementia [18]. Recent research has focused on identifying biological markers of dementia. Such markers may play a key role in the early diagnosis and management of dementia. Moreover, medications that delay the course of dementia are available, further emphasizing the need for diagnostic biomarkers to identify incipient dementia. Measuring serum copper and lipid concentration is affordable, accurate, and minimally invasive thus representing an appealing biological marker for cognitive abnormalities. The aim of this study was to investigate possible associations between lipid profile, serum copper, and dementia in the elderly Jordanians. SUBJECTS AND METHODS Subjects The study population consisted of elderly Jordanians that were recruited from senior homes and Jordan University Hospital. The institutional review board of the University of Jordan/Faculty of Medicine approved the study and a written consent was provided by each subject or legal guardian. Study inclusion required individuals be over 60 years of age with no history of any relevant psychiatric disease and no systemic use of statins or other lipid lowering agents, psychotropic drugs, or substance abuse. Information about risk factors was gathered from a questionnaire completed
by all participants. The questionnaire covered sociodemographic characteristics, lifestyle (fat intake and coffee intake), family and medical history including chronic diseases and medications. Participants were asked whether they use low-calorie, low saturated fat diet comprising whole grains, vegetables, chicken, fish, and olive oil regularly or they use diets high in saturated fat (such as animal meats, dairy products). Regular consumption of coffee was defined as nearly every day. Initially, all subjects were screened with two scales, the Mini-Mental State Examination (MMSE) and the Clock Drawing Test (CDT). Elderly individuals who scored abnormally on these tests underwent further evaluation with a non-structured interview for dementia. The diagnosis of dementia was made according to the criteria of Diagnostic and Statistical Manual IV (DSM-IV). The severity of dementia was assessed by the Clinical Dementia Rating (CDR), dementia patients included were (CDR = 2) [19]. A total 165 individuals agreed to participate in the study, with 72 having abnormal MMSE and CDT scores. After the interview and the application of the DSM-IV criteria, only 52 subjects were eligible for the moderate dementia group. In parallel, a group of 50 controls was randomly selected from different sectors of the same hospital (relatives, caregivers, and visitors). Control group subjects scored normally on both rating scales, were functionally independent and cognitively healthy (CDR = 0). Mental scales The MMSE is widely used screening instrument for assessing dementia in older adults. It tests five areas of cognitive functions: orientation, registration, attention/calculation, recall, and language [20]. Since there is no standardized Arabic form, we translated it to the Arabic language and adapted it to the local Arabic dialogue. Taking into consideration the high rate of illiteracy among the elderly population, the cutoff point for the presence of cognitive impairment was set according to the corresponding levels of education. For the control group, a score at 21/30 was set for those with no formal education, 23/30 for those who received 12 years of education. Moderate dementia group for individuals scored no less than 11 [20]. Executive function and visio-spatial performance was measured by the CDT using a pre-drawn circle. CDT is highly correlated with the MMSE and other tests of cognition dysfunction. Normal clock-drawing ability reasonably excludes cognitive impairment. The
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians
test is easy to administer, takes very little time, and is easy to document graphically in clinical records [21]. Subjects with low education, advanced age, and depression commonly perform poorly [22, 23]. Nevertheless, CDT using a pre-drawn circle has good diagnostic utility for detecting persons with early dementia; educational level appears to influence optimal cut-off scores in comprehensive but not abbreviated CDT methods [24]. Blood samples collection Blood samples were routinely collected in the morning after an overnight fast. Samples were centrifuged within 30 min at 3500 rpm for 10 min and rapidly stored at 4◦ C until analysis.
205
and standard deviations for continuous variables (age, MMSE, CDT, copper, TC, TG, HDL and LDL) or frequency and percentages for categorical variables (gender, marital status, education level, coffee intake, and fat intake). The association between continuous variables and mental status was assessed by comparisons of means using t-test (control and dementia groups). The association between categorical variables and mental status (control and dementia groups) was examined utilizing Chi-square test and binary logistic regression. Bivariate correlations of the MMSE and CDT scores and the continuous variables was analyzed and presented as Pearson correlation coefficients. Statistical significance was set at p < 0.05. Multivarient linear regression was used to measure the predictive value of continuous variables on MMSE and CDT scores.
Lipid profile measurement RESULTS The lipid profile measurements included serum total cholesterol (TC), triglycerides (TG), high density lipoprotein (HDL), and low density lipoprotein (LDL). All samples were measured by Roche Diagnostics COBAS INTEGRA 800 Biochemistry analyzer based on an enzymatic colorimetric method. Copper measurement Serum copper levels were quantified using A Shimadzu AA6800 Atomic Absorption Spectroscopy with Graphite Furnace Atomization System using wizAArd 2.31 software. A standard addition method was applied to construct a calibration curve to quantify copper in the samples. Copper measurement in water samples Eight water samples were analyzed from different sources (tap water, filtered water, and bottled water) and the following areas: Amman (Tla’a al Ali, Juayda), Madaba, Salt, and Irbid. The water sources represented the same region of the recruited subjects. Standard atomic absorption conditions were utilized for copper, inductively coupled plasma mass spectrometry. A Shimadzu AAS6300 was used for analysis of copper concentration from water samples.
Demographic variables Fifty-two dementia patients and 50 control subjects were enrolled in the study. There were no significant difference between dementia and control groups for age (p = 0.215) and gender (p = 0.290), respectively (Table 1). Some demographic variables were studied as a possible risk factors associated with cognitive decline. Only educational level (p = 0.006) and coffee intake (p = 0.000) showed significant association (Table 1). Interestingly, individuals that were illiterate or had educational level less than 12 years demonstrated a 6.29 fold (p = 0.002) and 3.29 fold (p = 0.026) increase in risk of developing dementia respectively. The intake of coffee showed protective effect against cognitive decline by a 6.25 fold decrease in risk. Other variables like marital status and fat intake showed no significant correlation. Mental scales Table 2 demonstrated the results of MMSE and CDT in the two groups. There was a significant correlation between CDT and MMSE results (p = 0.000). CDT in dementia patients was 4.03 ± 1.36 and only 2.22 ± 1.36 in control subjects.
Statistical analysis
Biochemical parameters
Demographic, clinical, and laboratory data were coded and entered into SPSS for windows 16.0 (SPSS Inc., Chicago, IL). Data were summarized as means
Table 3 showed the biochemical parameters in dementia group and control subjects. Most of the parameters measured did not differ between
206
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians Table 1 Demographics characteristics of study participants with dementia versus control subjects
Characteristic Age∗ (y) Age range (y) Gender (%) Women Men Marital status (%) Single/divorced Married Education Level (%) More than 12 Less than 12 Illiterate Coffee intake (%) Yes No Fat intake (%) Olive oil Olive oil + others
Controls (n = 50)
Patients with dementia (n = 52)
d.f
p-value
OR
95% confidence interval
68.9 ± 7.11 (60–88)
70.7 ± 7.63 (60–94)
1, 100
0.215
–
–
1, 100
0.29
–
–
17 (34.0) 33 (66.0)
23 (44.2) 29 (55.8) 1, 99
0.14
0.52
0.214–1.249
11 (22) 39 (78)
18 (35.3) 33 (64.7) 2, 100
0.006†
1, 100
0.026† 0.002† 0.000†
3.29 6.29 0.16
1.151–9.378 2.010–19.654 0.066–0.37
1, 98
0.073
2.08
0.929–4.673
20 (40.0) 20 (40.0) 10 (20.0)
7 (13.5) 23 (44.2) 22 (42.3)
37(74.0) 13(26.0)
16 (30.8) 36 (69.2)
25 (51.0) 24 (49.0)
17 (33.3) 34 (66.7)
OR, odds ratio. † p-value is significant. ∗ Values shown = mean±standard deviation (SD). Table 2 Tests used in this study for participants with dementia versus controls Test MMSE∗ CDT∗
Controls (n = 50)
Patients with dementia (n = 52)
d.f
p-value
95% confidence interval
26.66 ± 2.78 2.22 ± 1.36
17.37 ± 3.49 4.03 ± 1.36
1, 100 1, 80
0.000† 0.000†
8.054–0.535 −2.407 – −1.202
MMSE, Mini-Mental State Examination; CDT, Clock Drawing Test. † p- value is significant. ∗ Values shown = mean±standard deviation (SD). Table 3 Biochemical parameter values of individuals with dementia versus controls Biochemical test
Controls (n = 50)
Patients with CI (n = 52)
d.f
95% confidence interval
p-value
Copper (g/dL) TC∗ (mg/dL) TG∗ (mg/dL) HDL∗ (mg/dL) LDL∗ (mg/dL)
114.55 ± 57.6 187.16 ± 42.29 166.51 ± 80.44 38.47 ± 10.45 111.58 ± 34.39
126.12 ± 71.78 186.39 ± 47.60 165.07 ± 90.43 41.35 ± 10.31 113.74 ± 37.97
1, 98 1, 99 1, 99 1, 99 1, 99
−37.468–14.319 −17.020–18.556 −32.36–35.26 −6.975–1.222 −16.468–12.153
0.377 0.932 0.933 0.167 0.765
TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein; CI, cognitive impairment. † p-value is significant. ∗ Values shown = mean±standard deviation (SD). Table 4 Correlation between biochemical test and MMSE and CDT tests for dementia patients MMSE MMSE
CDT
Pearson Correlation Sig. (2-tailed) n Pearson Correlation Sig. (2-tailed) n
1 52 −0.295 0.077 37
Age −0.380 0.005∗∗ 52 0.221 0.189 37
Copper −0.043 0.765 51 0.057 0.743 36
TC 0.089 0.534 51 −0.175 0.308 36
TG −0.026 0.857 51 −0.015 0.932 36
HDL
LDL
0.119 0.407 51 −0.138 0.423 36
0.038 0.793 51 −0.207 0.226 36
CDT −0.295 0.077 37 1 37
TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein; MMSE, Mini-Mental State Examination; CDT, Clock Drawing Test. ∗∗ Correlation is significant at the 0.01 level (2-tailed).
the two groups. However the value of serum copper was numerically higher in the dementia group (126.12 g ± 71.78) compared to control group (114.55 g ± 57.6). These differences did not reach statistical significance (p = 0.377) and the values
were within the normal physiological reference range between 70 g/dl and 142.7 g/dl [25]. The lipid profile analysis in the two groups, showed slightly higher than normal TG and LDL and lower HDL when compared to the NCEP healthy lipid profile
E. Al-khateeb et al. / Relation between Copper, Lipid Profile, and Cognition in Elderly Jordanians
207
Table 5 Linear regression between biochemical test and MMSE test for dementia patients Coefficientsa Model (Constant) TC TG HDL LDL copper age
Unstandardized coefficients B
Std. Error
29.550 0.037 −0.006 −0.011 −0.034 −0.444 −0.190
5.681 0.029 0.007 0.062 0.033 0.697 0.064
Standardized coefficients
95% Confidence interval for B
Beta
t
Sig.
Lower Bound
0.516 −0.143 −0.032 −0.373 −0.092 −0.416
5.202 1.290 −0.755 −0.176 −1.034 −0.638 −2.973
0.000 0.204 0.454 0.861 0.307 0.527 0.005∗
18.094 −0.021 −0.020 −0.137 −0.100 −1.849 −0.318
Upper Bound 41.006 0.096 0.009 0.115 0.032 0.961 −0.061
a Dependent Variable: MMSE. TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein. ∗ Regression
is significant at the 0.01 level (2-tailed). R square = 0.200.
guidelines (TC 60 mg/dl, LDL
