Trends in 1-Year Recurrent Ischemic Stroke in the US Medicare Fee-for-Service Population

BACKGROUND

There have been important advances in secondary stroke prevention and a focus on healthcare delivery over the past decades. Yet, data on US trends in recurrent stroke are limited. We examined national and regional patterns in 1-year recurrence among Medicare beneficiaries hospitalized for ischemic stroke from 2001 to 2017.

METHODS

This cohort study included all fee-for-service Medicare beneficiaries aged ≥65 years who were discharged alive with a principal diagnosis of ischemic stroke from 2001 to 2017. Follow-up was up to 1 year through 2018. Cox models were used to assess temporal trends in 1-year recurrent ischemic stroke, adjusting for demographic and clinical characteristics. We mapped recurrence rates and identified persistently high-recurrence counties as those with rates in the highest sextile for stroke recurrence in ≥5 of the following periods: 2001-2003, 2004-2006, 2007-2009, 2010-2012, 2013-2015, and 2016-2017.

RESULTS

There were 3 638 346 unique beneficiaries discharged with stroke (mean age 79.0±8.1 years, 55.2% women, 85.3% White). The national 1-year recurrent stroke rate decreased from 11.3% in 2001-2003 to 7.6% in 2016-2017 (relative reduction, 33.5% [95% CI, 32.5%-34.5%]). There was a 2.3% (95% CI, 2.2%-2.4%) adjusted annual decrease in recurrence from 2001 to 2017 that included reductions in all age, sex, and race subgroups. County-level recurrence rates ranged from 5.5% to 14.0% in 2001-2003 and from 0.2% to 8.9% in 2016-2017. There were 76 counties, concentrated in the South-Central United States, that had the highest recurrence throughout the study. These counties had populations with a higher proportion of Black residents and uninsured adults, greater wealth inequity, poorer general health, and reduced preventive testing rates as compared with other counties.

CONCLUSIONS

Recurrent ischemic strokes decreased over time overall and across demographic subgroups; however, there were geographic areas with persistently higher recurrence rates. These findings can inform secondary prevention intervention opportunities for high-risk populations and communities.

Association of Long-term Statin Use With the Risk of Intracerebral Hemorrhage: A Danish Nationwide Case-Control Study

BACKGROUND AND OBJECTIVES

A causal relationship between long-term statin use and the risk of intracerebral hemorrhage (ICH) remains uncertain. We investigated the association with statin use before hospital admission for ICH in a Danish population-based, nationwide case-control study.

METHODS

We used the Danish Stroke Registry to identify all patients aged 45 years or older with a first-ever ICH between 2005 and 2018. Patients with ICH were matched for age, sex, and calendar year to controls selected from the general population. A medication registry with information on all dispensed prescriptions at community pharmacies in Denmark since 1995 was used to ascertain previous statin exposure that was classified for recency, duration, and intensity. Using conditional regression and adjusting for potential confounders, we calculated adjusted odds ratios (aORs) and corresponding 95% CIs for the risk of ICH.

RESULTS

The study population consisted of 16,235 patients with ICH and 640,943 controls. Current statin use (cases 25.9% vs controls 24.5%; aOR 0.74, 95% CI, 0.71-0.78) and a longer duration of current statin use (<1 year: aOR 0.86; 95% CI, 0.81-0.92; ≥1 to <5 years: aOR 0.72; 95% CI, 0.68-0.76; ≥5 to <10 years: aOR 0.65; 95% CI, 0.60-0.71; ≥10 years of use, 0.53; 95% CI 0.45-0.62; p for trend <0.001) were associated with a lower risk of ICH. Similar treatment duration relationships were found in analyses stratified by statin use intensity (high-intensity therapy: <1 year of use: aOR 0.78; 95% CI, 0.66-0.93; ≥10 years of use: aOR 0.46; 95% CI 0.33-0.65; p for trend 0.001).

DISCUSSION

We found that a longer duration of statin use was associated with a lower risk of ICH.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence that current statin use and a longer duration of statin use are each associated with a lower risk of ICH.

Rural Residence and Antihypertensive Medication Use in US Stroke Survivors

Background Relatively greater increases in hypertension prevalence among US rural residents may contribute to geographic disparities in recurrent stroke. There is limited US information on poststroke antihypertensive medication use by rural/urban residence. We assessed antihypertensive use and lifestyle characteristics for US rural compared with urban stroke survivors and residence-based trends in use between 2005 and 2019. Methods and Results US stroke survivors with hypertension were identified in the 2005 to 2019 national Behavioral Risk Factor Surveillance System surveys. We ascertained the survey-weighted prevalence of reported antihypertensive use and lifestyle characteristics (ie, physical activity, diabetes, cholesterol, body mass index, and smoking) among respondents with hypertension in odd years over this period by rural/urban residence. Separate trend analyses were used to detect changes in use over time. Survey-weighted logistic regression was used to calculate unadjusted and adjusted (sociodemographic and lifestyle factors) odds ratios for antihypertensive use by year. Our study included 82 175 individuals (36.4% rural residents). Lifestyle characteristics were similar between rural and urban residents except for higher smoking prevalence among rural residents. Antihypertensive use was similar between rural and urban stroke survivors in unadjusted and adjusted analyses (>90% in both populations). Trend analyses showed a small but significant increase in antihypertensive use over time among urban (P=0.033) but not rural stroke survivors (P=0.587). Conclusions Our findings indicate that poststroke antihypertensive use is comparable in rural and urban residents with a reported history of hypertension, but additional work is merited to identify reasons for a trend for increased use of these drugs among urban residents.

Committee on High-quality Alzheimer's Disease Studies (CHADS) consensus report

BACKGROUND

Consensus guidance for the development and identification of high-quality Alzheimer's disease clinical trials is needed for protocol development and conduct of clinical trials.

METHODS

An ad hoc consensus committee was convened in conjunction with the Alzheimer's Association to develop consensus recommendations.

RESULTS

Consensus was readily reached for the need to provide scientific justification, registration of trials, institutional review board oversight, conflict of interest disclosure, funding source disclosure, defined trial population, recruitment resources, definition of the intervention, specification of trial duration, appropriate payment for participant engagement, risk-benefit disclosure as part of the consent process, and the requirement to disseminate and/or publish trial results even if the study is negative.

CONCLUSIONS

This consensus guidance should prove useful for the protocol development and conduct of clinical trials, and may further provide a platform for the development of education materials that may help guide appropriate clinical trial participation decisions for potential trial participants and the general public.

Stroke Prevention in Symptomatic Large Artery Intracranial Atherosclerosis Practice Advisory: Report of the AAN Guideline Subcommittee

BACKGROUND AND OBJECTIVES

To review treatments for reducing the risk of recurrent stroke or death in patients with symptomatic intracranial atherosclerotic arterial stenosis (sICAS).

METHODS

The development of this practice advisory followed the process outlined in the American Academy of Neurology Clinical Practice Guideline Process Manual, 2011 Edition, as amended. The systematic review included studies through November 2020. Recommendations were based on evidence, related evidence, principles of care, and inferences.

MAJOR RECOMMENDATIONS

Clinicians should recommend aspirin 325 mg/d for long-term prevention of stroke and death and should recommend adding clopidogrel 75 mg/d to aspirin for up to 90 days to further reduce stroke risk in patients with severe (70%-99%) sICAS who have low risk of hemorrhagic transformation. Clinicians should recommend high-intensity statin therapy to achieve a goal low-density lipoprotein cholesterol level <70 mg/dL, a long-term blood pressure target of <140/90 mm Hg, at least moderate physical activity, and treatment of other modifiable vascular risk factors for patients with sICAS. Clinicians should not recommend percutaneous transluminal angioplasty and stenting for stroke prevention in patients with moderate (50%-69%) sICAS or as the initial treatment for stroke prevention in patients with severe sICAS. Clinicians should not routinely recommend angioplasty alone or indirect bypass for stroke prevention in patients with sICAS outside clinical trials. Clinicians should not recommend direct bypass for stroke prevention in patients with sICAS. Clinicians should counsel patients about the risks of percutaneous transluminal angioplasty and stenting and alternative treatments if one of these procedures is being contemplated.

Cardiovascular Risk Factor Profiles, Emergency Department Visits, and Hospitalizations for Women and Men with a History of Stroke or Transient Ischemic Attack: A Cross-Sectional Study

Background: The relationship between cardiovascular disease risk factors (CVD-RFs) and health care utilization may differ by sex. We determined whether having more CVD-RFs was associated with all-cause emergency department (ED) visits and all-cause hospitalizations for women and men with prior stroke/transient ischemic attack (TIA). Materials and Methods: In this cross-sectional study, we used nationally representative Medical Expenditure Panel Survey (2012-2015) data for persons aged ≥18 years with a prior stroke/TIA. CVD-RF summary scores include six self-reported factors (hypertension, diabetes, high cholesterol, physical inactivity, smoking, and obesity). Sex-specific covariate-adjusted logistic regression models assessed associations between CVD-RF scores and having one or more all-cause ED visits and one or more all-cause hospitalizations. Results: The weighted sample represents 9.1 million individuals (mean age 66.6 years; 54.3% women). Prevalence of low (0-1 risk factors), intermediate (2-3), and high (4-6) CVD-RF scores was 19.4%, 60.5%, and 20.1% for women and 14.6%, 60.2%, and 25.2% for men, respectively. Women having intermediate and high scores had a 1.58-fold (95% confidence interval [CI], 1.14-2.18) and 2.21-fold (95% CI, 1.50-3.25) increased odds of ED visits compared with women with low scores. Women with high CVD-RF scores had a 2.18-fold (95% CI, 1.42-3.34) increased odds of hospitalizations, but there was no association for women with intermediate CVD-RF profiles. There was no association between CVD-RF scores and either outcome for men. Conclusions: Women, but not men, with high and intermediate CVD-RF profiles had increased odds of all-cause ED visits; women with high CVD-RF profiles had increased odds of all-cause hospitalizations. The burden of CVD-RFs may be a sex-specific predictor of higher health care utilization in women with a history of stroke/TIA.

Disparities in Internet Use Among US Stroke Survivors: Implications for Telerehabilitation During COVID-19 and Beyond

Despite evidence-based guidelines,¹ stroke rehabilitation remains underutilized, particularly among women and minorities.² Telerehabilitation is a promising alternative to traditional in-person rehabilitation and offers a novel strategy to overcome access barriers,³ which intensified during the COVID-19 pandemic.⁴ A broadband connection is a prerequisite for its wide adoption but its availability varies across the United States (https://broadbandnow.com/national-broadband-map). Little is known about demographic and geographic variation in internet use among stroke survivors. In this study, we sought to compare internet use in a nationally representative sample of individuals with and without stroke.

Paradoxical Cerebral Air Embolism after Cardiac Ablation in Williams-Beuren Syndrome: A Clinico-Pathological Correlation

Non-traumatic neurological deterioration is a medical emergency that may arise from diverse causes, to include cerebral infarction or intracranial hemorrhage, meningoencephalitis, seizure, hypoxic-ischemic or toxic/metabolic encephalopathy, poisoning, or drug intoxication. We describe the abrupt onset of neurological deterioration in a 53-year-old man with Williams-Beuren syndrome, a sporadically occurring genetic disorder caused by chromosomal microdeletion at 7q11.23. The clinical phenotype of Williams-Beuren syndrome is suggested by distinctive elfin facies, limited intellect, unique personality features, growth abnormalities, and endocrinopathies. The causative microdeletion of chromosomal material will frequently involve loss of the elastin gene, ELN, with resulting arteriopathy, supravalvular aortic stenosis, non-ischemic cardiopathy, and atrial fibrillation. Our patient sustained acute neurological decline within one month after undergoing a cardiac ablative procedure to convert atrial fibrillation to sinus rhythm. We present our findings in the setting of a clinico-pathological correlation, in which we reveal the cause of the abrupt neurological deterioration and discuss how our patient was affected by an uncommon stroke disorder.

Abstract TP179: Candidate Genes For Small Vessel Ischemic Stroke: A Gwas Pilot Study

Background: The genetic basis of small vessel stroke (SVS) is poorly understood. Few genetic loci have been identified that may provide insights into SVS pathogenesis. We sought to identify genetic associations with incident SVS.

Methods: A GWAS analysis was performed on a prospective cohort of patients enrolled in the American Stroke Association- Bugher Small Vessel Intracranial Disease Whole Genome Association Study which included patients from 4 hospitals in North Carolina recruited between December 2007-August 2012. Patients with SVS in the prior 2 years confirmed clinically and on neuroimaging were compared to a control group who had no previous history of stroke. A logistic regression model with SVS as the outcome and single nucleotide polymorphisms (SNPs) as the main predictor was performed and adjusted for baseline demographics and clinical characteristics, including age, race/ethnicity, sex, tobacco use, history of hypertension, diabetes, hyperlipidemia, atrial fibrillation, history of stroke/TIA, and principal components. Because of the relatively small sample size, SNPs with p-values <1E-4 were considered significant.

Results: The cohort included 139 patients who had SVS and 64 controls. Among this cohort, 50% were men and 38% African-American. The mean age of the study group was 63-years. Sixteen SNPs in 9 genes met the statistical cutoff for association with SVS. These were involved in extracellular matrix integrity (CHD23, CLDN14, SPTBN1), neuronal differentiation (DSCAML1), cholesterol transport (GRAMD1B), transcriptional regulation (JARID2, LINC02111, LINC01993) and voltage-gated potassium channel (KCND2).

Conclusion: In this pilot GWAS study, several loci variations were associated with SVS. Although this study is hypothesis generating, previous studies have also shown an association of SVS with genes involved in extracellular matrix integrity. In addition, we found it interesting that SNPs in both CLDN14 and CDH23 genes, which have been linked to non-syndromic sensorineural deafness, are also more common in SVS. Some forms of sensorineural deafness have been associated with cognitive impairment and stroke, and if a genetic association is confirmed, it would suggest a common pathophysiology may link the two conditions.

Abstract 31: One-year Mortality And Stroke Readmissions After Ischemic Stroke In Critical Access Hospitals

Background: Critical access hospitals (CAHs) provide emergency and inpatient care in rural communities. CAHs have higher 30-day mortality after stroke, but little is known about long-term outcomes. We compared 1-year outcomes after ischemic stroke for patients treated at CAHs versus other hospitals.

Methods: We identified all Medicare fee-for-service beneficiaries aged ≥65 years discharged alive from US hospitals with a principal diagnosis of ischemic stroke in 2015. Patients were followed 1 year for death or stroke recurrence, accounting for competing risks. We balanced characteristics between CAH and non-CAH patients using stabilized inverse probability weights (IPW) based on patient demographic and clinical characteristics. We created adjusted Kaplan-Meier curves based on the IPW and fit Cox models to assess differences in 1-year mortality and recurrent stroke weighted by the IPW.

Results: There were 4,487 patients discharged with stroke from CAHs and 202,502 from non-CAHs. CAH vs non-CAH patients were older (mean age 82.8y vs 78.6y) and more often women (61.8% vs 53.9%), white (94.3% vs 83.7%), and dual Medicare-Medicaid eligible (21.6% vs 17.1%). Discharge to home (29.6% vs 36.8%) and inpatient rehabilitation (4.2% vs 18.9%) was less common for CAH patients, whereas discharge to an intermediate care/skilled nursing facility was more common (26.7% vs 23.9%). For CAHs and non-CAHs, respectively, 1-year mortality rates were 27.8% (95% CI 26.5-29.0) and 22.2% (22.0-22.4), and 1-year recurrence rates were 4.3% (3.6-4.9) and 4.6% (4.5-4.7) (Figure). In IPW-adjusted analyses, stroke patients treated at CAHs vs non-CAHs had higher risk of 1-year mortality (HR 1.29, 95% CI 1.22-1.37) but not recurrent stroke (0.91, 0.78-1.06).

Conclusions: Stroke patients discharged from CAHs vs non-CAHs had greater risk of 1-year mortality but not recurrence. Further work is needed to understand the observed disparity, potentially with a focus on post-acute care services.

Mechanical Thrombectomy Access for All? Challenges in Increasing Endovascular Treatment for Acute Ischemic Stroke in the United States

Mechanical thrombectomy (MT) is the most effective treatment for selected patients with an acute ischemic stroke due to emergent large vessel occlusions (LVOs). There is an urgent need to identify and address challenges in access to MT to maximize the numbers of patients who can benefit from this treatment. Barriers in access to MT include delays in evaluation and accurate diagnosis of LVO leading to inappropriate triage, logistical delays related to availability of facilities and trained interventionalists, and financial hurdles that affect treatment reimbursement. Collection of regional data related to these barriers is critical to better understand current access gaps and a measurable access score to thrombectomy could be useful to plan local public health intervention.

Contemporary Neuroscience Core Curriculum for Medical Schools

Medical students need to understand core neuroscience principles as a foundation for their required clinical experiences in neurology. In fact, they need a solid neuroscience foundation for their clinical experiences in all other medical disciplines also because the nervous system plays such a critical role in the function of every organ system. Because of the rapid pace of neuroscience discoveries, it is unrealistic to expect students to master the entire field. It is also unnecessary, as students can expect to have ready access to electronic reference sources no matter where they practice. In the preclerkship phase of medical school, the focus should be on providing students with the foundational knowledge to use those resources effectively and interpret them correctly. This article describes an organizational framework for teaching the essential neuroscience background needed by all physicians. This is particularly germane at a time when many medical schools are reassessing traditional practices and instituting curricular changes such as competency-based approaches, earlier clinical immersion, and increased emphasis on active learning. This article reviews factors that should be considered when developing the preclerkship neuroscience curriculum, including goals and objectives for the curriculum, the general topics to include, teaching and assessment methodology, who should direct the course, and the areas of expertise of faculty who might be enlisted as teachers or content experts. These guidelines were developed by a work group of experienced educators appointed by the Undergraduate Education Subcommittee (UES) of the American Academy of Neurology (AAN). They were then successively reviewed, edited, and approved by the entire UES, the AAN Education Committee, and the AAN Board of Directors.

Amylin Dyshomeostasis Hypothesis: Small Vessel-Type Ischemic Stroke in the Setting of Type-2 Diabetes

Recent histological analyses of human brains show that small vessel-type injuries in the setting of type-2 diabetes colocalize with deposits of amylin, an amyloid-forming hormone secreted by the pancreas. Amylin inclusions are also identified in circulating red blood cells in people with type-2 diabetes and stroke or cardiovascular disease. In laboratory models of type-2 diabetes, accumulation of aggregated amylin in blood and the cerebral microvasculature induces brain microhemorrhages and reduces cerebral blood flow leading to white matter ischemia and neurological deficits. At the cellular level, aggregated amylin causes cell membrane lipid peroxidation injury, downregulation of tight junction proteins, and activation of proinflammatory signaling pathways which, in turn, induces macrophage activation and macrophage infiltration in vascular areas positive for amylin deposition. We review each step of this cascade based on experimental and clinical evidence and propose the hypothesis that systemic amylin dyshomeostasis may underlie the disparity between glycemic control and stroke risk and may be a therapeutic target to reduce the risk of small vessel ischemic stroke in patients with type-2 diabetes.

Stroke Care during the COVID-19 Pandemic: International Expert Panel Review

BACKGROUND

Coronavirus disease 2019 (COVID-19) has placed a tremendous strain on healthcare services. This study, prepared by a large international panel of stroke experts, assesses the rapidly growing research and personal experience with COVID-19 stroke and offers recommendations for stroke management in this challenging new setting: modifications needed for prehospital emergency rescue and hyperacute care; inpatient intensive or stroke units; posthospitalization rehabilitation; follow-up including at-risk family and community; and multispecialty departmental developments in the allied professions.

SUMMARY

The severe acute respiratory syndrome coronavirus 2 uses spike proteins binding to tissue angiotensin-converting enzyme (ACE)-2 receptors, most often through the respiratory system by virus inhalation and thence to other susceptible organ systems, leading to COVID-19. Clinicians facing the many etiologies for stroke have been sobered by the unusual incidence of combined etiologies and presentations, prominent among them are vasculitis, cardiomyopathy, hypercoagulable state, and endothelial dysfunction. International standards of acute stroke management remain in force, but COVID-19 adds the burdens of personal protections for the patient, rescue, and hospital staff and for some even into the postdischarge phase. For pending COVID-19 determination and also for those shown to be COVID-19 affected, strict infection control is needed at all times to reduce spread of infection and to protect healthcare staff, using the wealth of well-described methods. For COVID-19 patients with stroke, thrombolysis and thrombectomy should be continued, and the usual early management of hypertension applies, save that recent work suggests continuing ACE inhibitors and ARBs. Prothrombotic states, some acute and severe, encourage prophylactic LMWH unless bleeding risk is high. COVID-19-related cardiomyopathy adds risk of cardioembolic stroke, where heparin or warfarin may be preferable, with experience accumulating with DOACs. As ever, arteritis can prove a difficult diagnosis, especially if not obvious on the acute angiogram done for clot extraction. This field is under rapid development and may generate management recommendations which are as yet unsettled, even undiscovered. Beyond the acute management phase, COVID-19-related stroke also forces rehabilitation services to use protective precautions. As with all stroke patients, health workers should be aware of symptoms of depression, anxiety, insomnia, and/or distress developing in their patients and caregivers. Postdischarge outpatient care currently includes continued secondary prevention measures. Although hoping a COVID-19 stroke patient can be considered cured of the virus, those concerned for contact safety can take comfort in the increasing use of telemedicine, which is itself a growing source of patient-physician contacts. Many online resources are available to patients and physicians. Like prior challenges, stroke care teams will also overcome this one. Key Messages: Evidence-based stroke management should continue to be provided throughout the patient care journey, while strict infection control measures are enforced.

Abstract P644: Average Daily Temperature Fluctuation and Hospitalizations and 30-Day Mortality for Stroke and AMI

Background: Daily average temperature is associated with increased hospitalizations and mortality for vascular conditions, but it is unclear if daily temperature variation is also associated with these outcomes. We assessed the relationship of daily temperature fluctuations with stroke and AMI hospitalizations and mortality in the elderly.

Methods: We identified fee-for-service Medicare beneficiaries aged ≥65 y with a primary discharge diagnosis of ischemic stroke or AMI in 2014-2015. Daily temperature data from the National Centers for Environmental Information were linked with Medicare beneficiary data by county and admission date. We fit a Poisson model for the relationship between daily temperature range (county daily maximum minus minimum) and 30-day hospitalizations, adjusted for season and patient demographics. Logistic regression assessed 30-day mortality, adjusted for season, patient demographics, and clinical characteristics. Overall and NOAA climate region-stratified relationships were assessed.

Results: There were 311,213 unique stroke hospitalizations (mean age 78.8 y, 53% women, 84% White) and 274,703 for AMI (mean age 77.6 y, 45.4% women, 86% White). The national hospitalization rate per 100,000 beneficiary-years was 735 for stroke and 639 for AMI. Thirty-day mortality was 12.0% for stroke and 12.8% for AMI. Each 1 o F increase in daily temperature range was associated with a 1.26 percentage point (95% CI 1.09-1.44) increase in stroke and a 1.48 percentage point (95% CI 1.43-1.53%) increase in AMI hospitalizations and varied by climate region (figure). Daily temperature range had little influence on stroke or AMI mortality (both OR 1.00, 95% CI 1.00-1.00).

Conclusions: Daily temperature fluctuations were associated with increased hospitalizations for stroke and AMI. Additional research is needed to understand meteorological effects on vascular events to inform prevention efforts for vulnerable populations.

Abstract P67: Role of Genetic Variants in Predicting Cognitive Outcomes Following Small Vessel Ischemic Stroke

Background: About 20% of patients with small vessel ischemic stroke (SVS) have cognitive impairment; however, the role of genetic factors in predicting cognitive outcomes following SVS has not been fully explored. APOE and ABCC9 have been associated with Alzheimer’s disease and hippocampal sclerosis respectively and play an important role in the neurovascular unit. We evaluated whether allelic variants in these genes influence cognitive outcomes following SVS.

Methods: We conducted a retrospective analysis of a prospective cohort of patients enrolled in the ASA-Bugher Small Vessel Intracranial Disease Whole Genome Association Studies. Patients with SVS were categorized by APOE (presence or absence of ε4 allele) and ABCC9 SNP rs704180 (presence or absence of A allele) status. The primary outcomes were total score on the short form of the MoCA, which assesses global cognition, and time to complete Trails B, which is a measure of executive function that can be affected by stroke. Linear regression analyses were performed using the genetic exposures of interest, adjusting for age, education, sex, race/ethnicity, NIHSS score, burden of white matter disease (WMD; using the CHS validated score 0-9), and time between stroke and the cognitive assessment.

Results: The sample included 145 patients who had SVS and available APOE and ABCC9 data. Among this cohort, 51.4% were men and 27.6% African American. The median age of the study participants was 63.4 years, the median years of education was 12, the median NIHSS was 2, and the median WMD burden score was 2. The mean time between stroke and the cognitive assessment was 75 days. The APOE ε4 allele was present in 35.0% and ABCC9 A allele in 74.8%. The presence of APOE ε4 allele was not associated with post-stroke MoCA scores (p=0.31) or Trails B (p=0.86). ABCC9 A allele was also not associated with post-stroke MoCA scores (p=0.34) or Trails B (p=0.31). Older age, higher NIHSS score, and greater burden of WMD were independently associated with longer times to complete Trails B (p<0.0001), but not with the MoCA score.

Conclusion: Following SVS, several patient characteristics, including age, stroke severity, and WMD burden, rather than their APOE and ABCC9 allelic statuses, were associated with post-stroke measures of executive function.

Abstract P245: Sex Differences in the Association Between Cardiovascular Risk Profiles and Hospitalizations/ED Visits Among Patients With a History of Stroke/TIA

Background: Secondary prevention addressing cardiovascular risk factors is a key care component for reducing the risk of vascular events among those who had a stroke/TIA. We assessed cardiovascular risk factor (CVD-RF) profiles and their impact on hospitalizations and Emergency Department (ED) visits and whether the association differs for men and women with prior stroke/TIA.

Methods: We used data from the nationally representative Medical Expenditure Panel Survey (2012-2015) for persons aged ≥ 18 years with a prior stroke/TIA. CVD-RF score included 6 self-reported factors (hypertension, diabetes, high cholesterol, lack of exercise, smoking, obesity), categorized as low (0-1 factors), intermediate (2-3), or high (4-6). Outcomes included ≥ 1 hospitalization discharges or ED visits during the participant’s survey year. Multivariable logistic regression models assessed the association between CVD-RF scores and outcomes, stratified by sex. We tested for interaction by sex in a combined model with men and women.

Results: The weighted sample represents 9.9 million individuals (mean age 65.1 years; 54.3% women). Overall, 16.7%, 59.3%, and 24.1% of men had low, intermediate, and high CVD-RF scores compared to 21.6%, 59.5%, and 18.9% of women. Among men, there was no significant association between CVD-RF score and the two outcomes ( Table ). Among women, even after adjustment for covariates, those with high scores had 1.89 and 2.06-fold increases in the odds of hospitalizations and ED visits, compared to those with low CVD-RF scores ( P <0.05). Furthermore, women with intermediate CVD-RF scores had a 1.68-fold increase in the odds of ED visits compared to those with low scores ( P <0.05). The combined model showed a significant interaction by sex.

Conclusion: Women with increased CVD-RF scores had increased odds of ED visits and hospitalizations, which could be due to less effective secondary prevention. Further research is needed to explore reasons for this sex disparity.

Abstract P655: Race Differences in 10-Year Mortality After Ischemic Stroke

Background: Few studies report 10-year mortality outcomes after stroke in the US by race. We assessed long-term survivorship by race among elderly ischemic stroke patients.

Methods: We identified fee-for-service Medicare beneficiaries aged ≥65 years discharged alive from US acute-care hospitals with a principal diagnosis of ischemic stroke from 2005 to 2007. Patients were followed through 2016 to calculate 10-year all-cause mortality, censoring for change in Medicare enrollment. Inverse probability weighting (IPW) was used to assess race differences in mortality. We used logistic regression to calculate the probability of a patient being Black as a function of age, Medicaid eligibility, comorbidities, in-hospital complications, discharge disposition, length of stay, and Medicare payment. We then fit a Cox regression model for the relationship between race and 10-year mortality that adjusted for sex and the inverse probability of being Black.

Results: There were 744,044 patients discharged alive with stroke (mean age 78.7y, 54.7% women, 85.6% White, 9.9% Black, and 4.5% other race). Black patients tended to be younger and were more often women. There were race differences in comorbidities, with renal failure, dementia, and diabetes more common in Blacks; atherosclerosis and COPD were more common in Whites. The 10-year mortality rate was 75.3% (95% CI 75.2–75.4%) for the overall population, with Blacks having the highest mortality (76.4%, 76.1–76.7%), followed by Whites (75.4%, 75.3–75.5%) and those of other race (70.3%, 69.8–70.8%; Figure). In the IPW analysis, the risk of death within 10 years of stroke was higher for Blacks (RR 1.04, 95% CI 1.03–1.04) but lower for other races (RR 0.92, 95% CI 0.90–0.93) when compared with Whites.

Conclusions: More than 75% of stroke patients died within 10 years. The 10-year stroke mortality risk was higher for Black stroke patients even after accounting for sociodemographic and index hospitalization factors.

Abstract 60: National Temporal Patterns in Recurrent Stroke by Demographic Characteristics and Geographic Regions: 2001-2016

Background: There have been important advances in secondary stroke prevention and a focus on healthcare delivery in the US over the past two decades. Yet, little is known about temporal patterns of recurrent stroke in the US. We examined temporal trends in recurrent stroke by sociodemographic characteristics and geographic areas using national Medicare data.

Methods: We included fee-for-service Medicare beneficiaries aged ≥65y with a primary discharge diagnosis of ischemic stroke from 2001 to 2016. We fit a Cox proportional hazards model that censored for change in Medicare enrollment and accounted for death to evaluate the temporal trend in 1-year recurrent stroke, adjusting for demographic and clinical factors. Models were repeated for subgroups defined by age, sex, race, and state. We mapped smoothed rates of 1-year recurrent stroke by county to assess geographic variation over time.

Results: There were 3,485,618 unique beneficiaries discharged with stroke during the study period. Demographic and clinical characteristics remained relatively stable over time, but the proportions discharged with home health services and inpatient rehabilitation increased. The observed 1-year recurrent stroke rate decreased from 11.2% in 2001-2004 to 9.3% in 2013-2016, with an adjusted annual reduction in recurrence from 2001-2016 of 1.49% (95% CI 1.40%-1.58%). There were significant reductions for all age, sex, and race groups (A). Geographic areas with persistently high rates were identified over time (B). In state-stratified analysis, the annual percentage reduction in recurrence ranged from -1.2% to 2.5% and was significant for all but 12 states.

Conclusions: Recurrent strokes decreased over time overall and by sociodemographic subgroups; however, we identified geographic areas with persistently high recurrence rates. Such findings can target secondary prevention intervention opportunities for high-risk populations and communities.

Abstract 61: Community Factors Associated With Persistently High 1-Year Recurrent Stroke Rates in the United States

Background: There is known geographic variation in recurrent stroke rates in the US; however, the contributions of socioeconomic status (SES), healthcare access/utilization, and community resources to these disparities are uncertain. We assessed community characteristics in counties having persistently higher recurrent stroke rates over a 16-year period.

Methods: We included 3,485,618 fee-for-service Medicare beneficiaries aged ≥65y discharged with ischemic stroke from 2001-2016, grouped into four 4-year periods. We categorized 3221 US counties or equivalents into 6 groups based on the % of stroke patients with a recurrent stroke within 1 year. Persistently high-recurrence counties were those in the highest sextile for each 4-year period. We integrated county-specific demographic, geographic, SES, general health, care availability, health behavior, and environmental data from the US Census Bureau, USDA Economic Research Service, and Dartmouth Atlas. We calculated mean standardized differences in county characteristics between high-recurrence and other counties and used logistic regression to model high-recurrence counties as a function of 12 potentially modifiable county characteristics.

Results: There were 133 persistently high-recurrence counties that were concentrated in the South Central US and included 140,144 stroke patients during the study (A; mean age 78.3y, 57% women, 82% White, 11.5% stroke recurrence vs 79.0 y, 55% women, 86% White, 10.5% stroke recurrence in other counties). Compared with the rest of the US, these counties had populations with lower SES, poorer health, more limited access to care providers and recreation/fitness, and reduced rates of preventive testing (B). The model including 12 potentially actionable characteristics had a c statistic of 0.84.

Conclusions: Our findings highlight the value of identifying potentially modifiable community characteristics that, if improved, might reduce recurrent stroke rates.