Abstract WP40: Development Of The System For National Statistics Of Stroke And Acute Myocardial Infarction Using Claims-based Algorithms In Korea

Background: The epidemiology of stroke and acute myocardial infarction (AMI) in Korea is limited by inaccurate methods for estimating incidence. Therefore, this study aimed to build the system for national statistics of stroke and AMI in Korea using claims-based identification algorithms.

Methods: We identified stroke and AMI using the claims-based algorithms based on the 2018 National Health Insurance Service (NHIS) data. The identification algorithms were validated using investigation of medical records of the sampled cases including the patient groups and the control groups based on a 2-stage stratified sampling method. The sampled cases were divided into 6 strata according to regions (capital and non-capital) and types of centers (tertiary hospitals, general hospitals, and hospitals). Based on the medical records results, we calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), and weighted PPV and weighted NPV were calculated by applying strata-specific sampling rates. The incidence rate of stroke and AMI is the number of new cases of stroke and AMI, including recurrent events, which was calculated by applying the algorithms using weighted PPV and NPV to the claims data.

Results: In total, 2,200 cases (1,086 cases in stroke [578 patients and 508 controls] and 1,114 cases in AMI [520 patients and 594 controls]) were sampled after applying algorithms for review of hospital records. In hospital record reviews, the stroke algorithm had 95.2% sensitivity, 99.6% specificity, 89.3% PPV, and 99.8% NPV, and AMI algorithms showed 97.7% sensitivity, 99.9% specificity, 90.0% PPV, and 99.9% NPV. We identified 94,994 stroke and 33,834 AMI cases including recurrent events using validated identification algorithms in 2018. The age- and sex-standardized incidence rate of stroke was 175.7 cases per 100,000 person-years and the incidence rate of AMI was 49.3 cases per 100,000 person-years in 2018.

Conclusion: We developed the national statistical system to estimate the incidence of stroke and AMI using validated claims-based algorithms in Korea. By using this system, we expect that it will be possible to conduct nationwide epidemiological research and improve acute cardiovascular care in Korea.

Abstract WMP81: Dynamic Changes In Intracranial Atherosclerotic Stenosis In Serial Follow-up Of High-resolution Magnetic Resonance Imaging

Background: Pathophysiology of intracranial atherosclerotic stenosis (ICAS) development and subsequent stroke occurrence is diverse, including cholesterol deposition, arterial dissection, and intrinsic vasculopathies. To differentiate these specific etiologies, performing high-resolution MRI (HR-MRI) has increased. However, the information on serial change of ICAS on HR-MRI was limited.

Methods: Patients hospitalized at a tertiary university hospital for AIS and who took HR-MRI more than twice between 2015 and 2019 were collected. Two specialists manually segmented the contour of the culprit and reference vessel's inner lumen, outer wall, and plaque. The stenotic degree, remodeling index, and enhancement signal were measured for the culprit lesion at each examination.

Results: A total of 202 HR-MRI examinations from 93 patients were analyzed. The Median follow-up was 270 days (118-390). The ranges of the serial change in stenotic degree (-86% to 41%), remodeling index (-83% to 266%), and enhancement signal (-85% to 71%) were very diverse. Changes in stenotic degree and enhancement according to the initial stenotic degree were insignificant. On the other hand, the change in enhancement signal was greater in the initially more enhanced lesions (0.1±23.0; mild vs. -23.1±22.6; moderate vs. -35.5±28; severe, p difference <.001) and in the other determined etiology (-14.6±22.8; large atherosclerosis vs. -30.7±30.8; other determined vs. -0.8±30.2; etc., p difference =.005).

Conclusions: ICAS showed very dynamic changes in the follow-up HR-MRIs. To identify the underlying etiology, such as arterial dissection and Moyamoya disease, serial HR-MRI will be helpful.

Abstract WP98: Utilization Of Perfusion Imaging For Acute Ischemic Stroke: Secular Trends And Effects On Outcomes Of Endovascular Treatment

Introduction: Perfusion imaging (PI) could guide decision-making for endovascular treatment (EVT) of acute ischemic stroke (AIS). However, PI was underused even in the US before the pivotal EVT trials proved its usefulness in 2018. This study aimed to describe the secular trends of PI utilization and investigate the effectiveness of PI-based EVT in real-world practice.

Methods: Using a prospective multicenter (n=17) stroke registry in South Korea, we identified patients with AIS who presented within 24 hours from onset between 2011 and 2021. The study period was divided into 3 epochs: 2011-2014, 2015-2017, and 2018-2021. The study population was divided into the early (arrival within 6 hours) and late window (6-24 hours) groups.

Results: A total of 51,650 patients (15,654 patients in 2011-2014, 14,432 patients in 2015-2017, and 21,564 patients in 2018-2021) were analyzed. Utilization of PI decreased in the overall population and early window group ( P trend <0.001); 43.3% and 54.1% in 2011-2014, 40.1% and 44.1% in 2015-2017, and 38.4% and 40.2% in 2018-2021, respectively; but increased in the late window group ( P trend <0.001); 31.3% in 2011-2014, 35.7% in 2015-2017, and 36.5% in 2018-2021. Of 10,872 patients with anterior large-vessel occlusion (aLVO), the EVT rate was not different between patients with and without PI (48.7% vs. 46.6%, P =0.08) in the early window but higher in those with PI than without PI in the late window (29.8% vs. 18.7%, P <0.001). The EVT outcome (3-month mRS 0-2) was not different between patients with and without PI in the early window (44.1% vs. 41.8%, P =0.21) and late window (38.4% vs. 39.2%, P =0.81). Propensity score analysis and instrumental variable analysis with PI rate per center as an instrument will be performed to adjust imbalances between patients with and without PI.

Conclusion: Between 2011 and 2021 in South Korea, PI utilization has decreased in patients arriving within 6 hours from onset but has increased in those arriving between 6 and 24 hours. Among patients with aLVO, PI likely increased the EVT rate in the late window but did not in the early window. PI utilization did not seem to affect the EVT outcomes, but in-depth analysis is required.

Abstract 44: Impact Of Novel Oral Anticoagulant Use On Secular Trend Of Vascular Events Or Death After Atrial Fibrillation-related Acute Ischemic Stroke

Introduction: Now novel oral anticoagulants (NOAC) are strongly recommended for secondary stroke prevention in patients with atrial fibrillation (AF). However, it remains unclear to what extent the introduction of NOACs improved clinical outcomes in real-world practice.

Methods: Using a nationwide prospective multi-center stroke registry database, we identified consecutive AIS patients with AF enrolled between Jan 2011 and Dec 2019, and analyzed one-year clinical events and NOAC prescription at discharge. The primary outcome was the composite of stroke, myocardial infarction, and all-cause death. To assess the mediation effect of NOAC on the outcomes, we performed natural effect models according to the calendar year. The exposure-mediator analysis, exposure-outcome analysis, and mediator-outcome analysis were performed using multivariate regression analysis according to the characteristics of the variables.

Results: We analyzed 12,500 patients (mean age, 74.4 years; 51.3% male; median NIHSS at presentation, 8). From 2011 to 2019, the cumulative one-year incidence of the primary composite outcome (28.3% to 22.1%), all-cause mortality (23.8% to 17.9%), and stroke recurrence (8.3% to 5.1%) significantly decreased, while the NOAC prescription rate at discharge increased (0% to 75.6%). One-year increase in the calendar year was independently associated with a delayed occurrence of primary composite outcomes (Step 1: adjusted Time Ratio (aTR), 1.10; 95% confidence interval, 1.07-1.14) and with an increased NOAC prescription rate (Step 2: adjusted odds ratio, 2.20; 2.14-2.27). Increase in the NOAC prescription rate was significantly associated with the delayed occurrence of primary composite outcome (Step 3: aTR, 3.80; 3.15-4.58). However, after controlling for the NOAC prescription rate (mediator), the calendar year was no longer associated with the primary composite outcomes. (Step 4: aTR, 0.78; 95% CI 0.60-1.03). Thus, our results indicate full mediation of NOAC prescription in the association between the calendar year and primary composite outcomes.

Conclusion: The reduced risk of major vascular events or death over time in AIS patients with AF was fully mediated by the increase in NOAC use.

Abstract TP180: Secular Trends For Demographics, Characteristics, Management, And Clinical Outcome Of Young Adults With Acute Ischemic Stroke

Background: The incidence of stroke in young adults is increasing with the growing prevalence of stroke risk factors among young adults over the last decades. However, studies on secular trends in the young-age stroke population are scarce.

Methods: We extracted data of acute ischemic stroke (AIS) patients aged 18-50 (n=7053; age, 43±6; male, 72%) from the multicenter stroke registry in South Korea. Demographics, risk factors, stroke characteristics, management, and clinical outcomes in young adults with AIS were summarized according to calendar year (2008-2019). Trends analysis was conducted using a linear regression model for continuous variables and the Mann-Kendall trend test for categorical variables.

Results: There was no significant change between 2008 and 2019 in the proportions of young adults among female patients (7.1% to 7.0%, p=0.44). However, the proportions of young adults within male patients tended to decrease (12.4% to 10.7%; p=0.001) The prevalence, and awareness of risk factors were unchanged during the study period. Regarding stroke subtypes, there was a significant increase in other determined etiology (8% to 20%; p<0.001) and a tendency to decrease in large artery atherosclerosis (26% in 2008; 34% in 2009 to 26% in 2019; p=0.28). The endovascular thrombectomy (EVT) rate increased (2% to 8%; p<0.001), but the intravenous thrombolysis rate (8.8% to 10.6%) and door-to-needle time (38 to 36 min) did not change. Use of dual antiplatelet (DAPT; 29% to 55%; p=<0.001) increased and direct oral anticoagulants (DOACs) in patients with atrial fibrillation tended to increase (11.7% in 2013 to 44.8% in 2019; p=0.07; DOAC began to be used in 2013 in Korea). However, the use of anticoagulants among young adult patients having atrial fibrillation tended to decrease (80% to 66%; p=0.054). Clinical outcomes, including a 3-month modified ranking scale 0-2 (84.6% to 85.2%), 1-year mortality (2.8% to 2.4%), and stroke recurrence rate (4.5 to 5.6%), did not improve during the study period.

Conclusions: Among young adults with ischemic stroke, the EVT rate and use of DAPT or DOACs increased. Still, awareness of risk factors, door-to-needle time, and anticoagulation for atrial fibrillation required more attention to improve clinical outcomes.

Abstract WP184: The Risk Of Stroke Recurrence According To Neuroimaging Parameters In Patients With Acute Ischemic Stroke And Atrial Fibrillation

Purpose: The usefulness of the existing risk stratification tools for atrial fibrillation (AF) is limited in predicting stroke recurrence in patients with acute ischemic stroke (AIS). Neuroimaging parameters obtained from diagnostic work-up of AIS could offer more elaborate prediction.

Methods: A multicenter prospective cohort of AIS patients with AF recruited from 14 university hospitals or regional stroke centers were followed up for recurrent ischemic stroke (RIS) and a composite of all stroke and TIA. Neuroimaging features were derived from acute and chronic infarction patterns, and SVD markers such as lacunes, CMBs, and WMH. Cumulative incidences according to each neuroimaging parameter were estimated and compared using the Kaplan-Meier method with log-rank test and multivariable cause-specific hazard models with death as a competing risk.

Results: A total of 2,270 patients were followed up for 431 days (IQR, 365-735), during which 111 RISs and 130 composite outcomes occurred. In unadjusted analysis, lesion multiplicity among acute infarction patterns, the presence of chronic non-lacunar infarction, and the presence of lacunes among SVD markers increased the risk of RIS significantly (Table). Other neuroimaging features such as territory multiplicity and location, confluency, topography, and size of acute lesions, lesion multiplicity, territory multiplicity, confluency, topography, and size of chronic infarction, number of lacunes, presence of CMBs, and WMH did not affect the incidence of RIS. The adjusted hazard ratios of lesion multiplicity of acute infarction, chronic infarction and lacunes were 1.45 (95% CI, 0.99-2.11), 1.57 (1.06-2.34) and 1.97 (1.30-2.98) for RIS, respectively. Similar findings were obtained for the composite outcome.

Conclusions: Several neuroimaging markers were associated with recurrent ischemic stroke in AIS with AF. This could pave the way to a new stratification scheme for AF including neuroimaging parameters.

Abstract WP186: Long-term Incidence Of Gastrointestinal Bleeding After Ischemic Stroke

Background: Incidence of gastrointestinal (GI) bleeding after acute ischemic stroke (AIS) was reported as 1.5% during hospitalization, one-thirds of which required blood transfusion. However, it is not known about the long-term incidence and the incidence rates by period after AIS.

Methods: AIS patients who were admitted to the 14 participating hospitals between 2011 and 2013 were identified using a nationwide multicenter prospective stroke registry database. GI bleeding was captured with related diagnosis codes by International Classification of Diseases-10th Revision through the linkage between the registry database and the claims data. Bleeding requiring at least 2 packs of blood transfusion was defined as major GI bleeding. Incidence rates were calculated for each period as follow; 0-30 days, 31-90 days, 91-180 days, 181-365 days, 1-2 years, 2-3years, after 3 years.

Results: Of 10,818 AIS patients, 59.0% were male and mean age was 67.5 ± 12.9 years. The median follow-up duration was 3.1 (interquartile range 2.3 to 4.0) years. During 31,208 person-years, 947 patients (8.8%) had 1,224 episodes of major GI bleeding. Annual incidence rate was 3.92 per 100 person-years. The incidence rates by periods were the highest at 19.21 per 100 person-years in the first month of AIS, gradually decreased to 9.02 in one to three months, 6.18 in three to six months, and 3.48 in six to twelve months. After three years, it remained at about 2.62 events per 100 person-years. During the observation period, only one major GI bleeding occurred without recurrence in about 80% of patients, about 13% recurred twice, and about 6% of patients had three or more recurrences. In the multivariable recurrent event analysis, anemia at admission, lower eGFR below 60, and mRS at 3 months ≥4 were independently associated with higher risk of major GI bleeding during the most of the observation period above 3 years.

Conclusions: Major GI bleeding, requiring transfusion, seems to occur frequently after AIS, and the risk was gradually decreased after stroke. The efforts are needed to prevent it, especially in stroke patients with anemia and decreased renal function.

Abstract WP196: Tricuspid Regurgitation And Its Association With Cardioembolic Stroke

Introduction: Tricuspid regurgitation (TR) is a common valvular heart disease and traditionally known to develop secondary to cardiopulmonary comorbidities. Since isolated TR without structural heart disease are known to be related with poor outcome, clinical course of TR itself has been focused recently. However, relationship between TR and ischemic stroke had not been investigated. In this study, we aimed to explore the association between characteristics of TR and subtypes of ischemic stroke.

Methods: Patients with ischemic stroke who were admitted to a single tertiary hospital from 2011 to 2020 were screened. Subtypes of stroke were divided into two groups (cardioembolic (CE) and non-CE). Echocardiographic indices including E/e’ ratio, left ventricular (LV) ejection fraction, left atrial volume index (LAVI), right ventricular systolic pressure, severity of valvular heart diseases (no/trivial, mild, moderate/severe) were assessed. Logistic regression models were constructed to determine the association of severity, subtypes of TR and CE stroke. Modifying effects of demographic characteristics and echocardiographic indices were also tested.

Results: Among 6,472 consecutive patients with ischemic stroke, 12.5% of patients had TR and CE stroke was diagnosed in 24.7%. Patients with CE stroke were related to impaired LV systolic and diastolic function, pulmonary hypertension, and elevated LAVI. TR was more frequently observed in CE stroke group (6.2% vs. 32.7%, p -value < 0.01). After adjustment for covariates which were selected from bivariate analysis, TR shown significant association with CE stroke (OR [95% CI], 3.24 [2.61 - 4.02]). This association persisted across models which subdivided TR according to severity (mild TR, 2.91 [2.32 - 3.64]; moderate/severe TR, 5.58 [3.62 - 8.61]) and presence of cardiopulmonary comorbidities (isolated TR 2.04 [1.59 - 2.63]; non-isolated TR 2.15 [1.57 - 2.96]). There were no significant interactions for any of the predefined potential effect modifiers.

Conclusions: Our study demonstrated that TR has significant association with CE stroke. TR related hemodynamic change within cardiac chamber and accompanying atrial cardiopathy might be an underestimated risk factor for CE stroke.

Abstract P648: The Presence and the Features of Silent Brain Infarction Are Associated With Stroke Recurrence in Acute Ischemic Stroke Patients With Atrial Fibrillation

Background: The significance of silent brain infarction (SBI) for stroke recurrence in acute ischemic stroke (AIS) patients with atrial fibrillation (AF) has yet to be elucidated. This study aims to evaluate SBI as an independent predictor and which characteristics of SBI are associated with stroke recurrence in AIS patients with AF.

Methods: A multicenter prospective cohort recruited AIS patients with non-valvular AF from 14 centers from Oct 2017 to Dec 2018, and followed for ischemic stroke recurrence, all types of stroke and TIA, and all-cause mortality. Three patient groups; stroke patients with prior stroke history (PS), first-ever stroke with SBI [F-SBI(+)] and first-ever stroke without SBI [F-SBI(-)] were compared with Cox frailty model according to predetermined covariates. SBI subtypes; embolic-appearing pattern (EAP) and non-EAP, and SBI characteristics; size, numbers, and vascular territory involvements were assessed.

Results: A total of 978 AF-AIS patients [27.5% PS, 29.1% F-SBI(+), 43.4% F-SBI(-)] were followed for 365 [348-374] days (median). Incidence of ischemic stroke recurrence in F-SBI(+) was higher than F-SBI(-), however, there was no significant difference compared to PS (p=0.860). Adjusted hazards for ischemic stroke recurrence and all kinds of stroke and TIA in F-SBI(+) were shown to be elevated [HR 3.87 (95% CI 1.53-9.16) and 2.60 (1.21-5.56)], and similar to PS [4.20 (1.73-10.24) and 2.90 (1.36-6.18)] when compared to F-SBI(-), respectively. Despite irrelevance in non-EAP SBI, a 4-fold increase of hazards in EAP SBI was observed [4.07 (1.63-10.13)]. Other SBI characteristics were not associated with outcomes. SBI and SBI features did not increase all-cause mortality.

Conclusions: SBI and specifically, EAP SBI elevated stroke recurrence in AF-AIS patients as much prior stroke has increased the risk. Considering SBI to predict recurrence is suggested likewise prior stroke history is scored in AF thromboembolic risk estimation tools.