Externally validated deep learning model to identify prodromal Parkinson's disease from electrocardiogram.

Little is known about electrocardiogram (ECG) markers of Parkinson's disease (PD) during the prodromal stage. The aim of the study was to build a generalizable ECG-based fully automatic artificial intelligence (AI) model to predict PD risk during the prodromal stage, up to 5 years before disease diagnosis. This case-control study included samples from Loyola University Chicago (LUC) and University of Tennessee-Methodist Le Bonheur Healthcare (MLH). Cases and controls were matched according to specific characteristics (date, age, sex and race). Clinical data were available from May, 2014 onward at LUC and from January, 2015 onward at MLH, while the ECG data were available as early as 1990 in both institutes. PD was denoted by at least two primary diagnostic codes (ICD9 332.0; ICD10 G20) at least 30 days apart. PD incidence date was defined as the earliest of first PD diagnostic code or PD-related medication prescription. ECGs obtained at least 6 months before PD incidence date were modeled to predict a subsequent diagnosis of PD within three time windows: 6 months-1 year, 6 months-3 years, and 6 months-5 years. We applied a novel deep neural network using standard 10-s 12-lead ECGs to predict PD risk at the prodromal phase. This model was compared to multiple feature engineering-based models. Subgroup analyses for sex, race and age were also performed. Our primary prediction model was a one-dimensional convolutional neural network (1D-CNN) that was built using 131 cases and 1058 controls from MLH, and externally validated on 29 cases and 165 controls from LUC. The model was trained on 90% of the MLH data, internally validated on the remaining 10% and externally validated on LUC data. The best performing model resulted in an external validation AUC of 0.67 when predicting future PD at any time between 6 months and 5 years after the ECG. Accuracy increased when restricted to ECGs obtained within 6 months to 3 years before PD diagnosis (AUC 0.69) and was highest when predicting future PD within 6 months to 1 year (AUC 0.74). The 1D-CNN model based on raw ECG data outperformed multiple models built using more standard ECG feature engineering approaches. These results demonstrate that a predictive model developed in one cohort using only raw 10-s ECGs can effectively classify individuals with prodromal PD in an independent cohort, particularly closer to disease diagnosis. Standard ECGs may help identify individuals with prodromal PD for cost-effective population-level early detection and inclusion in disease-modifying therapeutic trials.

Predicting Parkinson's Disease and Its Pathology via Simple Clinical Variables.

BACKGROUND: Parkinson's disease (PD) is a chronic, disabling neurodegenerative disorder. OBJECTIVE: To predict a future diagnosis of PD using questionnaires and simple non-invasive clinical tests. METHODS: Participants in the prospective Kuakini Honolulu-Asia Aging Study (HAAS) were evaluated biannually between 1995-2017 by PD experts using standard diagnostic criteria. Autopsies were sought on all deaths. We input simple clinical and risk factor variables into an ensemble-tree based machine learning algorithm and derived models to predict the probability of developing PD. We also investigated relationships of predictive models and neuropathologic features such as nigral neuron density. RESULTS: The study sample included 292 subjects, 25 of whom developed PD within 3 years and 41 by 5 years. 116 (46%) of 251 subjects not diagnosed with PD underwent autopsy. Light Gradient Boosting Machine modeling of 12 predictors correctly classified a high proportion of individuals who developed PD within 3 years (area under the curve (AUC) 0.82, 95%CI 0.76-0.89) or 5 years (AUC 0.77, 95%CI 0.71-0.84). A large proportion of controls who were misclassified as PD had Lewy pathology at autopsy, including 79%of those who died within 3 years. PD probability estimates correlated inversely with nigral neuron density and were strongest in autopsies conducted within 3 years of index date (r = -0.57, p < 0.01). CONCLUSION: Machine learning can identify persons likely to develop PD during the prodromal period using questionnaires and simple non-invasive tests. Correlation with neuropathology suggests that true model accuracy may be considerably higher than estimates based solely on clinical diagnosis.

Genome-wide epigenetic analyses in Japanese immigrant plantation workers with Parkinson's disease and exposure to organochlorines reveal possible involvement of glial genes and pathways involved in neurotoxicity.

BACKGROUND: Parkinson's disease (PD) is a disease of the central nervous system that progressively affects the motor system. Epidemiological studies have provided evidence that exposure to agriculture-related occupations or agrichemicals elevate a person's risk for PD. Here, we sought to examine the possible epigenetic changes associated with working on a plantation on Oahu, HI and/or exposure to organochlorines (OGC) in PD cases. RESULTS: We measured genome-wide DNA methylation using the Illumina Infinium HumanMethylation450K BeadChip array in matched peripheral blood and postmortem brain biospecimens in PD cases (n = 20) assessed for years of plantation work and presence of organochlorines in brain tissue. The comparison of 10+ to 0 years of plantation work exposure detected 7 and 123 differentially methylated loci (DML) in brain and blood DNA, respectively (p < 0.0001). The comparison of cases with 4+ to 0-2 detectable levels of OGCs, identified 8 and 18 DML in brain and blood DNA, respectively (p < 0.0001). Pathway analyses revealed links to key neurotoxic and neuropathologic pathways related to impaired immune and proinflammatory responses as well as impaired clearance of damaged proteins, as found in the predominantly glial cell population in these environmental exposure-related PD cases. CONCLUSIONS: These results suggest that distinct DNA methylation biomarker profiles related to environmental exposures in PD cases with previous exposure can be found in both brain and blood.

Electrocardiographic changes predate Parkinson's disease onset.

Autonomic nervous system involvement precedes the motor features of Parkinson's disease (PD). Our goal was to develop a proof-of-concept model for identifying subjects at high risk of developing PD by analysis of cardiac electrical activity. We used standard 10-s electrocardiogram (ECG) recordings of 60 subjects from the Honolulu Asia Aging Study including 10 with prevalent PD, 25 with prodromal PD, and 25 controls who never developed PD. Various methods were implemented to extract features from ECGs including simple heart rate variability (HRV) metrics, commonly used signal processing methods, and a Probabilistic Symbolic Pattern Recognition (PSPR) method. Extracted features were analyzed via stepwise logistic regression to distinguish between prodromal cases and controls. Stepwise logistic regression selected four features from PSPR as predictors of PD. The final regression model built on the entire dataset provided an area under receiver operating characteristics curve (AUC) with 95% confidence interval of 0.90 [0.80, 0.99]. The five-fold cross-validation process produced an average AUC of 0.835 [0.831, 0.839]. We conclude that cardiac electrical activity provides important information about the likelihood of future PD not captured by classical HRV metrics. Machine learning applied to ECGs may help identify subjects at high risk of having prodromal PD.

Excessive daytime sleepiness and topographic expansion of Lewy pathology.

OBJECTIVE: While excessive daytime sleepiness (EDS) can predate the clinical diagnosis of Parkinson disease (PD), associations with underlying PD pathogenesis are unknown. Our objective is to determine if EDS is related to brain Lewy pathology (LP), a marker of PD pathogenesis, using clinical assessments of EDS with postmortem follow-up. METHODS: Identification of LP was based on staining for α-synuclein in multiple brain regions in a sample of 211 men. Data on EDS were collected at clinical examinations from 1991 to 1999 when participants were aged 72-97 years. RESULTS: Although EDS was more common in the presence vs absence of LP (p = 0.034), the association became stronger in neocortical regions. When LP was limited to the olfactory bulb, brainstem, and basal forebrain (Braak stages 1-4), frequency of EDS was 10% (4/40) vs 17.5% (20/114) in decedents without LP (p = 0.258). In contrast, compared to the absence of LP, EDS frequency doubled (36.7% [11/30], p = 0.023) when LP reached the anterior cingulate gyrus, insula mesocortex, and midfrontal, midtemporal, and inferior parietal neocortex (Braak stage 5). With further infiltration into the primary motor and sensory neocortices (Braak stage 6), EDS frequency increased threefold (51.9% [14/27], p < 0.001). Findings were similar across sleep-related features and persisted after adjustment for age and other covariates, including the removal of PD and dementia with Lewy bodies. CONCLUSIONS: The association between EDS and PD includes relationships with extensive topographic LP expansion. The neocortex could be especially vulnerable to adverse relationships between sleep disorders and aggregation of misfolded α-synuclein and LP formation.

Most cases with Lewy pathology in a population-based cohort adhere to the Braak progression pattern but 'failure to fit' is highly dependent on staging system applied.

Braak et al.'s 2003 paper detailing the caudo-rostral progression of Lewy body pathology (LP) formed the foundation of current understanding of disease spread in Parkinson's disease (PD); however, its methods are difficult to recreate and consequently multiple new staging systems emerged to recapitulate Braak's staging system using standard neuropathological methods and to account for other patterns of LP. Studies using these systems have documented widely variable rates of cases that 'fail to fit' expected patterns of LP spread. This could be due to population differences, features of individual systems, or may constitute under-recognized patterns of disease. We examined 324 neuropathological cases from the Honolulu Asia Aging Study and applied four different LP staging systems to determine the proportion of cases adhering to different staging methodologies and those that 'fail to fit' expected patterns of LP. Of 141 cases with LP (24: PD, 8: Dementia with Lewy bodies (DLB), 109: Incidental Lewy body disease (ILBD)), our application of Braak et al., 2003 classified 83.7%, Müller et al., 2005 classified 87.9%, Beach et al., 2009 classified 100%, and Leverenz et al., 2008 classified 98.6%. There were significant differences in the cases classifiable by the Leverenz and Beach systems versus the Braak and Müller systems (p < 0.001 for each). In this population-based autopsy cohort with a high prevalence of ILBD, the majority of cases were consistent with the progression characterized by the Braak et al. however, the determination of cases as atypical is highly dependent on the staging system applied.

Association of brain heptachlor epoxide and other organochlorine compounds with lewy pathology.

BACKGROUND: Organochlorine pesticides are associated with an increased risk of Parkinson's disease. A preliminary analysis from the Honolulu-Asia Aging Study suggested that heptachlor epoxide, a metabolite from an organochlorine pesticide extensively used in Hawaii, may be especially important. This was a cross sectional analysis to evaluate the association of heptachlor epoxide and other organochlorine compounds with Lewy pathology in an expanded survey of brain organochlorine residues from the longitudinal Honolulu-Asia Aging Study. METHODS: Organochlorines were measured in frozen occipital or temporal lobes in 705 brains using gas chromatography with mass spectrometry. Lewy pathology was identified using hematoxylin and eosin- and α-synuclein immunochemistry-stained sections from multiple brain regions. RESULTS: The prevalence of Lewy pathology was nearly doubled in the presence versus the absence of heptachlor epoxide (30.1% versus 16.3%, P < 0.001). Although associations with other compounds were weaker, hexachlorobenzene (P = 0.003) and α-chlordane (P = 0.007) were also related to Lewy pathology. Most of the latter associations, however, were a result of confounding from heptachlor epoxide. Neither compound was significantly related to Lewy pathology after adjustment for heptachlor epoxide. In contrast, the association of heptachlor epoxide with Lewy pathology remained significant after adjustments for hexachlorobenzene (P = 0.013) or α-chlordane (P = 0.005). Findings were unchanged after removal of cases of PD and adjustment for age and other characteristics. CONCLUSIONS: Organochlorine pesticides are associated with the presence of Lewy pathology in the brain, even after exclusion of PD cases. Although most of the association is through heptachlor epoxide, the role of other organochlorine compounds is in need of clarification. © 2018 International Parkinson and Movement Disorder Society.

Neuropathologic comorbidity and cognitive impairment in the Nun and Honolulu-Asia Aging Studies.

OBJECTIVE: To examine frequencies and relationships of 5 common neuropathologic abnormalities identified at autopsy with late-life cognitive impairment and dementia in 2 different autopsy panels. METHODS: The Nun Study (NS) and the Honolulu-Asia Aging Study (HAAS) are population-based investigations of brain aging that included repeated cognitive assessments and comprehensive brain autopsies. The neuropathologic abnormalities assessed were Alzheimer disease (AD) neuropathologic changes, neocortical Lewy bodies (LBs), hippocampal sclerosis, microinfarcts, and low brain weight. Associations with screening tests for cognitive impairment were examined. RESULTS: Neuropathologic abnormalities occurred at levels ranging from 9.7% to 43%, and were independently associated with cognitive impairment in both studies. Neocortical LBs and AD changes were more frequent among the predominantly Caucasian NS women, while microinfarcts were more common in the Japanese American HAAS men. Comorbidity was usual and very strongly associated with cognitive impairment. Apparent cognitive resilience (no cognitive impairment despite Braak stage V) was strongly associated with minimal or no comorbid abnormalities, with fewer neocortical AD lesions, and weakly with longer interval between final testing and autopsy. CONCLUSIONS: Total burden of comorbid neuropathologic abnormalities, rather than any single lesion type, was the most relevant determinant of cognitive impairment in both cohorts, often despite clinical diagnosis of only AD. These findings emphasize challenges to dementia pathogenesis and intervention research and to accurate diagnoses during life.

Midlife milk consumption and substantia nigra neuron density at death.

OBJECTIVE: To examine the relationship between midlife milk intake and Parkinson disease (PD) incidence through associations with substantia nigra (SN) neuron density and organochlorine pesticide exposure in decedent brains from the Honolulu-Asia Aging Study. METHODS: Milk intake data were collected from 1965 to 1968 in 449 men aged 45-68 years with postmortem examinations from 1992 to 2004. Neuron density (count/mm(2)) was measured in quadrants from a transverse section of the SN. Additional measures included brain residues of heptachlor epoxide, an organochlorine pesticide found at excessively high levels in the milk supply in Hawaii in the early 1980s. RESULTS: Neuron density was lowest in nonsmoking decedents who consumed high amounts of milk (>16 oz/d). After removing cases of PD and dementia with Lewy bodies, adjusted neuron density in all but the dorsomedial quadrant was 41.5% lower for milk intake >16 oz/d vs intake that was less (95% confidence interval 22.7%-55.7%, p < 0.001). Among those who drank the most milk, residues of heptachlor epoxide were found in 9 of 10 brains as compared to 63.4% (26/41) for those who consumed no milk (p = 0.017). For those who were ever smokers, an association between milk intake and neuron density was absent. CONCLUSIONS: Milk intake is associated with SN neuron loss in decedent brains unaffected by PD. Whether contamination of milk with organochlorine pesticides has a role in SN neurodegeneration warrants further study.

Total and differential white blood cell counts in late life predict 8-year incident stroke: the Honolulu Heart Program.

OBJECTIVES: To study the association between total and differential white blood cell (WBC) count and incident stroke in an older Asian population. DESIGN: Prospective population-based study with 8 years of follow-up. SETTING: The Honolulu Heart Program, Oahu, Hawaii. PARTICIPANTS: Japanese-American men aged 71 to 93 who were free of stroke and had baseline WBC counts measured in 1991-93 (N=3,342). MEASUREMENTS: Participants were divided into quartiles of total and differential WBC count for analysis and were followed for incident stroke (thromboembolic and hemorrhagic (hemorrhagic)) for 8 years using data from a comprehensive hospital surveillance system. RESULTS: Age-adjusted incident stroke rates increased significantly with increasing WBC quartile (Q1, 7.68; Q2, 9.04; Q3, 9.26; Q4, 14.10 per 1,000 person-years of follow-up, P=.001). Hazard ratios (HRs) for stroke for each quartile of total and differential WBC count were obtained using Cox regression analysis, with the lowest quartile as the reference group. After full adjustment, including age; cardiovascular risk factors; fibrinogen; prevalent coronary heart disease, cancer, or chronic obstructive pulmonary disease, and nonsteroidal anti-inflammatory drug use, HRs were 1.62 (95% confidence interval (CI)=1.04-2.52, P=.03) in the highest quartile of total WBC and 2.19 (95% CI=1.41-3.39, P<.001) in the highest quartile of neutrophil counts. Significant associations were also seen for thromboembolic but not for hemorrhagic strokes. No significant associations were found between lymphocyte or monocyte counts and incident stroke or subtypes. CONCLUSION: In elderly Japanese-American men, higher total WBC and neutrophil counts were independent predictors of overall stroke, as well as thromboembolic stroke.

Effect of creatine monohydrate on clinical progression in patients with Parkinson disease: a randomized clinical trial.

IMPORTANCE: There are no treatments available to slow or prevent the progression of Parkinson disease, despite its global prevalence and significant health care burden. The National Institute of Neurological Disorders and Stroke Exploratory Trials in Parkinson Disease program was established to promote discovery of potential therapies. OBJECTIVE: To determine whether creatine monohydrate was more effective than placebo in slowing long-term clinical decline in participants with Parkinson disease. DESIGN, SETTING, AND PATIENTS: The Long-term Study 1, a multicenter, double-blind, parallel-group, placebo-controlled, 1:1 randomized efficacy trial. Participants were recruited from 45 investigative sites in the United States and Canada and included 1741 men and women with early (within 5 years of diagnosis) and treated (receiving dopaminergic therapy) Parkinson disease. Participants were enrolled from March 2007 to May 2010 and followed up until September 2013. INTERVENTIONS: Participants were randomized to placebo or creatine (10 g/d) monohydrate for a minimum of 5 years (maximum follow-up, 8 years). MAIN OUTCOMES AND MEASURES: The primary outcome measure was a difference in clinical decline from baseline to 5-year follow-up, compared between the 2 treatment groups using a global statistical test. Clinical status was defined by 5 outcome measures: Modified Rankin Scale, Symbol Digit Modalities Test, PDQ-39 Summary Index, Schwab and England Activities of Daily Living scale, and ambulatory capacity. All outcomes were coded such that higher scores indicated worse outcomes and were analyzed by a global statistical test. Higher summed ranks (range, 5-4775) indicate worse outcomes. RESULTS: The trial was terminated early for futility based on results of a planned interim analysis of participants enrolled at least 5 years prior to the date of the analysis (n = 955). The median follow-up time was 4 years. Of the 955 participants, the mean of the summed ranks for placebo was 2360 (95% CI, 2249-2470) and for creatine was 2414 (95% CI, 2304-2524). The global statistical test yielded t1865.8 = -0.75 (2-sided P = .45). There were no detectable differences (P < .01 to partially adjust for multiple comparisons) in adverse and serious adverse events by body system. CONCLUSIONS AND RELEVANCE: Among patients with early and treated Parkinson disease, treatment with creatine monohydrate for at least 5 years, compared with placebo did not improve clinical outcomes. These findings do not support the use of creatine monohydrate in patients with Parkinson disease. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00449865.

Mid-life proteinuria and late-life cognitive function and dementia in elderly men: the Honolulu-Asia Aging Study.

BACKGROUND: Impaired renal function has been linked to cognitive impairment. We assessed mid-life proteinuria and late-life cognitive function in elderly Asian men. METHODS: The Honolulu Heart Program is a prospective study that began in 1965 with 8006 Japanese-American men aged 45 to 68 years. Mid-life proteinuria was detected by urine dipstick in 1971 to 1974. The Honolulu-Asia Aging Study began 20 years later, with cognitive assessment by the Cognitive Abilities Screening Instrument (CASI) in 3734 men. Standard criteria were used to classify 8-year incident dementia and subtypes. RESULTS: The age-adjusted incidence of dementia increased significantly from 13.8, to 22.8, to 39.7 per 1000 person years follow-up, among those with no, trace, and positive mid-life proteinuria (P=0.004). Using linear regression adjusting for age, education, APOEε4, stroke, hypertension, systolic blood pressure, diabetes, fasting blood glucose, physical activity, and baseline CASI, those with positive proteinuria had significantly higher annual change in CASI over 8 years follow-up (-1.24, P=0.02) (reference=no proteinuria). Multivariate Cox regression found that positive proteinuria had a significant association with incident all-cause dementia (RR=2.66; 95%CI, 1.09-6.53; P=0.03), but no significant associations with incident Alzheimer disease or vascular dementia. CONCLUSION: Mid-life proteinuria was an independent predictor for late-life incident all-cause dementia and cognitive decline over 8 years.

Associations of brain lesions at autopsy with polysomnography features before death.

OBJECTIVE: To determine how sleep-disordered breathing, nocturnal hypoxia, and changes in sleep architecture in the elderly may be related to the development of the neuropathologic correlates of dementia. METHODS: The Honolulu-Asia Aging Study is a prospective cohort study of Japanese American men in Honolulu, HI. We examined brain lesions at autopsy (Braak stage, neurofibrillary tangle and neuritic plaque counts, microinfarcts, generalized brain atrophy, lacunar infarcts, Lewy bodies [LBs], neuronal loss and gliosis in the locus ceruleus) in 167 participants who underwent polysomnography in 1999-2000 (mean age, 84 years) and died through 2010 (mean 6.4 years to death). Polysomnography measures included the apnea-hypopnea index, duration of apnea or hypopnea, duration of hypoxemia, minimum oxygen saturation (SpO2), duration of slow-wave sleep (SWS, non-REM stage N3), and arousals. RESULTS: Sleep duration with SpO2 <95% was associated with higher levels of microinfarcts (adjusted odds ratio [OR] 3.88, 95% confidence interval [CI] 1.10-13.76, comparing the highest to lowest quartiles of %sleep with SpO2 <95%). Greater SWS duration was associated with less generalized atrophy (adjusted OR 0.32, 95% CI 0.10-1.03, comparing highest to lowest quartiles of %sleep in SWS). LBs were less common with greater %sleep with SpO2 <95% (adjusted OR 0.17, 95% CI 0.04-0.78, comparing highest to lowest quartiles). Higher minimum SpO2 during REM sleep was associated with less gliosis and neuronal loss in the locus ceruleus. Cognitive scores declined less among men with greater SWS duration. CONCLUSIONS: The findings support a role for lower nocturnal oxygenation and SWS in the development of microinfarcts and brain atrophy, but not Alzheimer lesions or LBs.

Shorter men live longer: association of height with longevity and FOXO3 genotype in American men of Japanese ancestry.

OBJECTIVES: To determine the relation between height, FOXO3 genotype and age of death in humans. METHODS: Observational study of 8,003 American men of Japanese ancestry from the Honolulu Heart Program/Honolulu-Asia Aging Study (HHP/HAAS), a genetically and culturally homogeneous cohort followed for over 40 years. A Cox regression model with age as the time scale, stratified by year of birth, was used to estimate the effect of baseline height on mortality during follow-up. An analysis of height and longevity-associated variants of the key regulatory gene in the insulin/IGF-1 signaling (IIS) pathway, FOXO3, was performed in a HHP-HAAS subpopulation. A study of fasting insulin level and height was conducted in another HHP-HAAS subpopulation. RESULTS: A positive association was found between baseline height and all-cause mortality (RR = 1.007; 95% CI 1.003-1.011; P = 0.002) over the follow-up period. Adjustments for possible confounding variables reduced this association only slightly (RR = 1.006; 95% CI 1.002-1.010; P = 0.007). In addition, height was positively associated with all cancer mortality and mortality from cancer unrelated to smoking. A Cox regression model with time-dependent covariates showed that relative risk for baseline height on mortality increased as the population aged. Comparison of genotypes of a longevity-associated single nucleotide polymorphism in FOXO3 showed that the longevity allele was inversely associated with height. This finding was consistent with prior findings in model organisms of aging. Height was also positively associated with fasting blood insulin level, a risk factor for mortality. Regression analysis of fasting insulin level (mIU/L) on height (cm) adjusting for the age both data were collected yielded a regression coefficient of 0.26 (95% CI 0.10-0.42; P = 0.001). CONCLUSION: Height in mid-life is positively associated with mortality, with shorter stature predicting longer lifespan. Height was, moreover, associated with fasting insulin level and the longevity genotype of FOXO3, consistent with a mechanistic role for the IIS pathway.

Proteinuria in midlife and 39-year total mortality: the Honolulu Heart Program.

BACKGROUND: Previous population-based studies have shown that proteinuria is an independent predictor of total mortality. However, no studies have examined multiple proteinuria measurements or had a follow-up period longer than two decades. METHODS: Proteinuria was measured by urine dipstick on 6,815 Japanese-American men on two occasions, 6 years apart. Participants were classified into the "no proteinuria" group if both examinations were negative, "transient proteinuria" if either was positive, and "persistent proteinuria" if both were positive and followed for total mortality over 39 years. RESULTS: Prevalence of transient and persistent proteinuria was 6.4% and 1.3%, respectively. Age-adjusted total mortality rates were 41.9, 55.0, and 71.9 per 1000 person-years follow-up for no, transient, and persistent proteinuria groups, respectively (p for trend <.0001). Multivariate Cox proportional hazards models showed increased total mortality risk in a dose-response manner: HR, 1.40; P < .001 and HR, 2.26; P < .001 for transient and persistent proteinuria groups, respectively (using no proteinuria as reference). Stratified analyses showed stronger associations between proteinuria and mortality among those with prevalent cardiovascular diseases compared with those without. CONCLUSIONS: Proteinuria was independently associated with higher total mortality risk over 39 years. This risk was stronger among high-risk populations but also remained significant in low-risk populations. Simple urine dipstick can be a good risk assessment tool in the general population.

Prestroke weight loss is associated with poststroke mortality among men in the Honolulu-Asia Aging Study.

OBJECTIVE: To examine baseline prestroke weight loss and poststroke mortality among men. DESIGN: Longitudinal study of late-life prestroke body mass index (BMI), weight loss, and BMI change (midlife to late life) with up to 8-year incident stroke and mortality follow-up. SETTING: Community-based aging study data. PARTICIPANTS: Japanese-American men (N=3581; age range, 71-93y) who were stroke free at baseline. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE: Poststroke mortality: 30 days poststroke, analyzed with stepwise multivariable logistic regression; and long-term poststroke (up to 8y), analyzed with stepwise multivariable Cox regression. RESULTS: Weight loss (4.5kg decrements) was associated with increased 30-day poststroke mortality (adjusted odds ratio=1.48; 95% confidence interval [CI], 1.14-1.92), long-term mortality after incident stroke (all types, n=225; adjusted hazards ratio (aHR)=1.25; 95% CI, 1.09-1.44), and long-term mortality after incident thromboembolic stroke (n=153; aHR=1.19; 95% CI, 1.01 to 1.40). Men with overweight/obese late-life BMI (≥25kg/m(2), compared with healthy/underweight BMI) had increased long-term mortality after incident hemorrhagic stroke (n=54; aHR=2.27; 95% CI, 1.07-4.82). Neither desirable nor excessive BMI reductions (vs no change/increased BMI) were associated with poststroke mortality. In the overall sample (N=3581), nutrition factors associated with increased long-term mortality included the following: (1) weight loss (10lb decrements; aHR=1.15; 95% CI, 1.09-1.21), (2) underweight BMI (vs healthy BMI; aHR=1.76; 95% CI, 1.40-2.20), and (3) both desirable and excessive BMI reductions (vs no change or gain, separate model from weight loss and BMI; aHR range, 1.36-1.97; P<.001). CONCLUSIONS: Although obesity is a risk factor for stroke incidence, prestroke weight loss was associated with increased poststroke (all types and thromboembolic) mortality. Overweight/obese late-life BMI was associated with increased posthemorrhagic stroke mortality. Desirable and excessive BMI reductions were not associated with poststroke mortality. Weight loss, underweight late-life BMI, and any BMI reduction were all associated with increased long-term mortality in the overall sample.

Prevalence of vitamin D deficiency and association with functional status in newly admitted male veteran nursing home residents.

OBJECTIVES: To provide the first report on prevalence of vitamin D deficiency in newly admitted nursing home (NH) residents and associations with functional disabilities and chronic diseases. DESIGN: Retrospective chart review. SETTING: Nursing home (NH). PARTICIPANTS: Male veterans newly admitted to a NH for rehabilitation, skilled-nursing care, intermediate care, or respite care between January 2011 and June 2012. MEASUREMENTS: Total serum 25-hydroxyvitamin D (25(OH)D) levels were measured on admission. Vitamin D supplement users and those without 25(OH)D measurement within 7 days of admission were excluded, leaving an analytical sample of 104 residents. Vitamin D deficiency was defined as 25(OH)D less than 20 ng/mL. Data were collected on age, ethnicity, season, body mass index (BMI), functional disability in activities of daily living (ADLs) (mobility, bathing, dressing, toileting, continence, and feeding), and prevalent chronic diseases. RESULTS: Prevalence of vitamin D deficiency was 49.0%. In multivariate logistic regression models adjusted for age, ethnicity, and BMI, vitamin D deficiency was significantly associated with number of ADL disabilities (odds ratio (OR) = 1.4 for each 1-point increase in ADL disability score, P = .03) and prevalent diabetes mellitus (OR = 3.0, P = .03). In regression models using each ADL disability as a separate variable, only disability in feeding (OR = 4.7, P = .05) and diabetes mellitus (OR = 2.9, P = .04) remained significant. CONCLUSION: Almost half the individuals entering the NH and not taking vitamin D supplements had vitamin D deficiency. Greater number of ADL disabilities, disability in feeding, and prevalent diabetes mellitus were independently associated with vitamin D deficiency.

Antihypertensive medication use and risk of cognitive impairment: the Honolulu-Asia Aging Study.

OBJECTIVE: To determine the associations between classes of antihypertensive medication use and the risk of cognitive impairment among elderly hypertensive men. METHODS: The Honolulu-Asia Aging Study is a prospective, community-based cohort study of Japanese American men conducted in Honolulu, Hawaii. We examined 2,197 participants (mean age 77 years at cohort entry, 1991-1993, followed through September 2010) with hypertension and without dementia or cognitive impairment at baseline, who provided information on medication use. Cognitive function was assessed at 7 standardized examinations using the Cognitive Abilities Screening Instrument (CASI). Cognitive impairment was defined as a CASI score <74. RESULTS: A total of 854 men developed cognitive impairment (median follow-up, 5.8 years). ß-Blocker use as the sole antihypertensive drug at baseline was consistently associated with a lower risk of cognitive impairment (incidence rate ratio [IRR] 0.69; 95% confidence interval [CI] 0.50-0.94), as compared with men not taking any antihypertensive medications, adjusting for multiple potential confounders. The use of diuretics, calcium channel blockers, angiotensin-converting enzyme inhibitors, or vasodilators alone was not significantly associated with cognitive impairment. Results were similar excluding those with cardiovascular disease or <1 year of follow-up, and additionally adjusting for pulse pressure, heart rate, baseline and midlife systolic blood pressure, and midlife antihypertensive treatment (IRR 0.65; 95% CI 0.45-0.94). The association between ß-blocker use and cognitive impairment was stronger among men with diabetes, men aged >75 years, and those with pulse pressure ≥70 mm Hg. CONCLUSIONS: ß-blocker use is associated with a lower risk of developing cognitive impairment in elderly Japanese American men.