| |
Albuminuria and Dementia in the Elderly: A Community Study
|
| |
| |
Download the PDF here
"our results suggest the possibility that treatments that decrease albuminuria, such as angiotensin-converting enzyme inhibitors/blockers, may have a salutary impact on the development of dementia. Results of several prospective studies will test this hypothesis.25
.......the association of albuminuria with dementia may help explain in part why conditions associated with albuminuria, such as diabetes mellitus, also are associated with increased risk of dementia.26
......nephrologists often are called upon to manage patients with albuminuria. To date, the focus of dementia evaluation and management for the nephrology community has been in dialysis patients. Recent work from the CHCS showed that cognitive decrease can appear with moderate renal decrease.27 Our study extends these findings to an earlier stage of renal disease because albuminuria often is present before the decrease in glomerular function."
"In this study of older adults, a statistically significant cross-sectional association between increasing albuminuria and dementia was found. This association remained significant after adjustment for factors that associate with dementia, such as hypertension, diabetes, and prevalent cardiovascular disease. The association of albuminuria and dementia may be explained by the many anatomic microvascular similarities found in brains of people with dementia and kidneys of people with albuminuria. Aside from these similarities, functional factors common to both conditions also may explain why these conditions coexist. For example, people with albuminuria have impaired autoregulation of glomerular filtration,21 whereas people with dementia have impaired regulation of cerebrovascular flow.1 Also, albuminuria and dementia may be related through factors not measured in this study, such as increased levels of oxidative stress.22, 23"
"As a continuous variable, there was an unadjusted odds ratio (OR) of 1.22 (95% confidence interval [CI], 1.15 to 1.29) of dementia with every doubling of albuminuria.....here was a borderline association of albuminuria with Alzheimer disease (OR, 1.10......doubling of albuminuria was associated significantly with an increased OR of increased white matter score.....Albuminuria, the excessive excretion of protein in urine, is a marker of renal microvascular disease. It occurs most often in the presence of hypertension and diabetes mellitus and is associated with increasing age, increased systolic blood pressure, and increased levels of inflammation factors.5 Many of these factors are present in people with dementia.2In addition, pathological examination of kidneys in people with albuminuria showed many of the same capillary findings found in brain specimens of people with dementia and retinal vascular disease.6 These findings led us to test the hypothesis that albuminuria is associated on cross section with increased odds of dementia....When albuminuria was categorically defined, a statistically significant fully adjusted relationship with dementia likewise was found (OR, 1.58; 95% CI, 1.09 to 2.30). When these analyses were repeated in participants without diabetes or without hypertension, similar results were obtained, although the association of albuminuria with dementia and MCI was strongest in those without hypertension.....MCI, a transitional phase between normal cognitive function and dementia,15 was associated with increasing albuminuria on unadjusted analysis and with adjustment for demographic factors. Further adjustment for cardiovascular disease, cardiovascular risk factors, and the ApoE-4 genotype made this association statistically nonsignificant."
----------
Albuminuria and Dementia in the Elderly: A Community Study
American Jnl of Kidney Disease 2008
"Of the 73 prevalent cases of dementia removed from analysis, 18 (24.7%) had albuminuria on subsequent testing in 1996 to 1997.....The distribution of participant baseline characteristics by albuminuria status (presence of albumin excretion ≥ 30 mg/g creatinine) is listed in Table 2. Compared with individuals without albuminuria, participants with albuminuria were older and were more likely to be men; have a history of diabetes, hypertension, or both; have lower renal function; and use angiotensin-converting enzyme inhibitors. They also were more likely to have MCI and dementia.....Compared with subjects without cognitive impairment (Table 3), participants with dementia were older, were more likely to be nonwhite, had lower attained educational level, and had more diabetes and atherosclerotic vascular disease."
Background
Dementia is associated with microvascular disease of the retina. In this study, we examine whether cognitive status (normal cognition, mild cognitive impairment, and dementia) is associated with albuminuria, a microvascular disorder of the kidney.
Study Design
Cross-sectional analysis.
Setting & Participants
2,316 participants from the Cardiovascular Health Cognition Study who underwent brain magnetic resonance imaging and testing for albuminuria.
Predictor
Doubling of albuminuria.
Outcome
Dementia defined according to neuropsychological and clinical evaluation.
Measurements
Multinomial logistic modeling was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) of dementia and mild cognitive impairment with doubling of albuminuria compared with the odds with normal cognition.
Results
283 participants (12.2%) had dementia, 344 (14.9%) had mild cognitive impairment, and 1,689 (72.9%) had normal cognition. Compared with participants with normal cognition, doubling of albuminuria was associated with increased odds of dementia (OR, 1.22; 95% CI, 1.15 to 1.29). Adjustment for prevalent cardiovascular disease and cardiovascular risk factors, lipid levels, C-reactive protein level, estimated glomerular filtration rate, and apolipoprotein E-4 genotype attenuated this association, but it remained statistically significant (OR, 1.12; 95% CI, 1.03 to 1.22). Mild cognitive impairment was associated with albuminuria on unadjusted analysis, but not with adjustment for other factors.
Limitations
Results are cross-sectional; causality cannot be imputed.
Conclusions
The odds of dementia increased in the presence of albuminuria. These findings suggest a role of shared susceptibility for microvascular disease in the brain and kidney in older adults.
Microvascular abnormalities are found in the brains of people who die of dementia (reviewed in1, 2). In patients with Alzheimer disease, the leading cause of dementia, abnormalities include basement membrane thickening, luminal narrowing, loss of pericytes, and increased permeability. In patients with vascular dementia, the second leading cause of dementia, many of the same findings also are present. Recently, the Atherosclerosis Risk in Communities (ARIC) Study and the Cardiovascular Health Study (CHS) reported that retinal microvascular findings, eg, arteriovenous nicking, focal arteriolar narrowing, microaneurysms, and exudates, were associated with cognitive decline and dementia.3, 4 All these findings support a role for cerebral microvascular disease in the pathogenesis of dementia.
Albuminuria, the excessive excretion of protein in urine, is a marker of renal microvascular disease. It occurs most often in the presence of hypertension and diabetes mellitus and is associated with increasing age, increased systolic blood pressure, and increased levels of inflammation factors.5 Many of these factors are present in people with dementia.2 In addition, pathological examination of kidneys in people with albuminuria showed many of the same capillary findings found in brain specimens of people with dementia and retinal vascular disease.6 These findings led us to test the hypothesis that albuminuria is associated on cross section with increased odds of dementia.
Methods
Participants in this study were from the CHS, an observational study of cardiovascular risk factors in adults aged 65 years. Recruitment methods have been published.7 In brief, a random sample of community-living individuals derived from Medicare eligibility lists were invited to participate at 4 field centers. A total of 5,201 participants were recruited in 1989 to 1990. In 1992 to 1993, the fifth year of the study, 687 African Americans were added to the study in the same manner in 3 of the 4 centers. All participants gave informed consent on study entry.
All participants were examined annually at their clinic sites through 1998 to 1999 (year 11 of the study) by using the Modified Mini–Mental State Examination (3MSE)8 (Table 1). For individuals who did not come to the clinic, interval cognitive information was obtained by using the Telephone Interview for Cognitive Status (TICS).9 Additional information about cognition was obtained from proxies using the Informant Questionnaire for Cognitive Decline in the Elderly (IQ CODE)10 and from physicians for participants who died or were unable to fill out the 3MSE or TICS. Information regarding functional status was obtained by using activities of daily living and instrumental activities of daily living by means of standardized questionnaires. The presence of dementia was documented during medical record review of all deaths and cardiovascular events.
Participants also underwent baseline blood testing, cardiac and carotid artery ultrasound testing, electrocardiography, ankle-brachial index measurement, and completion of medical history and clinical examination, as previously described.7 Magnetic resonance imaging (MRI) of the brain was completed in 1991 to 1994 and again in 1997 to 1999 for participants willing and able to undergo scanning. Apolipoprotein E-4 (ApoE-4) genotype was determined for participants providing consent for use of DNA. ApoE-4 is a genotype heterozygote or homozygote for at least one copy of ApoE-4.
The Cognition Cohort and Definition of Dementia
In 1998 to 1999, the Cardiovascular Health Cognition Study (CHCS) was assembled. It identified subjects with prevalent dementia at the time of brain MRI in 1991 to 1994 and those who subsequently developed dementia through 1998 to 1999. Inclusion in the CHCS required completion of a 3MSE at the time of MRI and ApoE-4 genotyping. The cohort consisted of 3,608 individuals (Fig 1). Of these individuals, 777 (21.6%) had died by 1998 to 1999. There were 1,741 white and 357 African American women, 1,282 white and 212 African American men, and 16 of other races; 62% of white and 62% of African American subjects completed the MRI examination. Participants who did not undergo scanning (mainly because of refusal, inability to complete MRI, or MRI contraindications) had lower 3MSE scores, were less educated, and had more clinical cardiovascular disease than those who underwent scanning.11 Groups did not differ with regard to prevalence of ApoE-4 genotype.
The 3,608 participants were divided into groups at high and low likelihood of having dementia based on cognitive testing, changes in cognitive scores, nursing home admission, alive or dead status, and history of stroke. If a participant was alive in 1998 to 1999, high risk was defined as 3MSE score less than 80 at 1 of the last 2 clinic visits, a 5-point decrease in 3MSE score from the time of MRI to the time of last visit, TICS score less than 28, IQ CODE score greater than 3.6, having an incident stroke, medical chart review with dementia as a diagnosis, or currently residing in a nursing home. If a participant died before 1998 to 1999, he or she was considered to have a high likelihood of having had dementia if they had at least 1 of the following: 3MSE score less than 80 within 2 years of death, greater than 5-point decrease in 3MSE score from the year of MRI to the year closest to death, TICS score less than 28 or IQ CODE score greater than 3.6 within 2 years of death, or diagnosis of dementia in a medical record or history of incident stroke. Prior analysis showed that few cases of dementia are missed by using this classification system of high versus low risk.12, 13
In 3 clinic sites (in 1998 to 1999), all living high-risk white participants and all African American participants (because of their small sample size) were neuropsychologically evaluated for dementia (1,192 [44%] were classified as high risk, including minorities; 1,492 [56%] of all whites, as low risk). In 1 center (Pittsburgh, PA; n = 927), all participants underwent evaluation for dementia regardless of whether they were considered at high probability for having dementia. This was done to estimate "misses" in low-risk participants at the other 3 centers. Neuropsychological evaluation consisted of tests of intelligence, memory, immediate and delayed recall, language, visual perception and construction, and executive functioning (tests are listed in Table S1, which is provided as supplementary material available with this article at www.ajkd.org). Results were classified as normal or abnormal.13
The diagnosis of dementia was based on a deficit in performance in 2 or more cognitive domains that were of sufficient severity to affect the participant's activities of daily living and a history of normal intellectual function before the cognitive decrease. An abnormal domain was present when results of at least 2 tests of the same domain were abnormal. A memory deficit was not required for the diagnosis of dementia. Diagnosis was made by an adjudication committee of neurologists with expertise in dementia. Year of onset for dementia was determined by using all available data. Mild cognitive impairment (MCI) was defined as cognitive impairment, especially memory deficit, without dementia.14, 15
After the decision of whether dementia was present, classification of the type of dementia was made before and after review of brain MRI results. There was a high degree of correlation of classifications of type of dementia both before and after reviewing MRI scans.14 Classification of dementia type was based on several classification systems (see Table 1 in13 for definitions of each set of criteria). The diagnosis of probable Alzheimer disease was made following the National Institute of Neurological and Communicative Diseases and Stroke (NINDS)/Alzheimer's Disease Related Disorders Association classification system16 and required the participant to show a gradual cognitive decrease without history or evidence of another illness that could cause mental impairment. There could be no evidence of a focal central nervous system lesion based on clinical or radiological examinations. Possible Alzheimer disease used the same criteria as probable Alzheimer disease, but with evidence of other concomitant diseases that may have caused cognitive decrease (eg, depression, hypothyroidism, head injury, alcoholism, central nervous system infection, and cerebrovascular disease). Vascular dementia was diagnosed based on clinical and/or radiological evidence of cerebral infarctions contributing to dementia. The State of California Alzheimer's Disease Diagnostic and Treatment Centers17 and the NINDS Association Internationale pour la Recherche et Enseignement en Neurosciences18 criteria primarily were used for these diagnoses. Classification based on Alzheimer's Disease Diagnostic and Treatment Centers criteria includes more vascular dementia than the NINDS-Association Internationale pour la Recherche et Enseignement en Neurosciences criteria because MRI data are relied on more heavily. MRI infarcts were defined as an area of abnormal signal in a vascular distribution that lacked a mass effect.19 An infarct was considered "silent" if there was no history of stroke or transient ischemic attack at baseline or on follow-up. Only infarcts of 3 mm or greater were included in analyses (reproducibility was low for lesions < 3 mm). Infarcts of the cortical gray and deep nuclear regions had to be brighter than on spin density and T2-weighted images than normal gray matter. Infarcts in white matter were similarly defined, except they had to be hypointense on T1 images to distinguish from diffuse white matter disease. White matter changes were estimated by using the total extent of periventricular and subcortical white matter signal abnormality on spin-weighted images, graded from none/barely present (0 or 1) to almost all white matter involved (grade 9). Prior CHS analyses showed that a high white matter score was an independent predictor of incident clinical stroke.20 When elements of both Alzheimer disease and vascular dementia were present, the diagnosis of mixed dementia was made. For these analyses, individuals with mixed dementia were included in the vascular dementia group.
Albuminuria Testing
Of CHCS participants alive in 1996 to 1997, a total of 2,389 had albuminuria tested on a random morning urine sample. Albuminuria testing was not done at the time of entry into the main CHS study. Urinary albumin was measured by means of rate nephelometry using the Array 360 CE Protein Analyzer (Beckman Instruments, Fullerton, CA). Urinary creatinine was measured on a Kodak Ektachem 700 Analyzer (Kodak, Rochester, NY). Participants with less than 30 mg of albumin/g creatinine were defined as having normoalbuminuria. Those with 30 mg of albumin/g creatinine or greater were defined as having albuminuria.
Statistical Methods
The χ2, t-test, and nonparametric methods were used, as appropriate, to compare baseline parameters between those with and without albuminuria and among those with normal cognition, MCI, and dementia. Multinomial logistic regression was used to test the association of doubling of albuminuria in those with mild cognitive impairment and dementia compared with those with normal cognition. Models were adjusted for: (1) age, sex, race, and education; (2) history of coronary heart disease, stroke, hypertension, diabetes, and smoking; serum cholesterol, low-density lipoprotein cholesterol, and C-reactive protein levels; and estimated glomerular filtration rate; and (3) ApoE-4 genotype. Cox proportional hazards regression models were constructed to assess the association of albuminuria with incident cases of dementia after urine collection in 1997 through follow-up in 1999. Analyses were performed using SPSS, version 14.0 (SPSS Inc, Chicago, IL).
|
|
| |
| |
|
|
|
