Abstract TP174: Identifying Predictors for Final Diagnosis of Ischemic Events in an Emergency Department Observation Unit

Response

Transient Ischemic Attacks—Populations and Prognosis

Objectives: Patients with transient ischemic attack (TIA) are at high risk of early stroke, indicating a need for urgent diagnostic workup and treatment. Atrial fibrillation (AF) is one of important causes for TIA. The purpose of this study was to investigate characteristics of TIA patients with AF, using data of a multicenter retrospective study. Methods: The subjects of this study were TIA patients admitted to 13 stroke centers within 7 days after onset between 2008 and 2009. The definition of TIA used in this study was that of clinical symptoms lasting less than 24 hours, regardless of imaging findings. We compared baseline characteristics and diffusion weighted-MRI (DWI) findings between patients with AF and those without AF. Results: A total of 464 patients (69 ± 13 years in age, 172 women) were enrolled. We diagnosed 79 patients (17%) as having AF. TIA patients with AF were older (74 ± 9 vs. 67 ± 14 years old, P = 0.0001) and more likely to have a longer length of hospital stay (15 ± 9 vs. 14 ± 20 days, P = 0.046) than those without AF. Of 464 TIA patients, 458 patients (99%) underwent head MRI. DWI lesion was detected in 96 (21%) patients. There was no difference of DWI positivity rate between patients with or without AF (28% vs. 20%, P = 0.10). Compared to patients without AF, patients with AF were more likely to have a single lesion on DWI (86% vs. 51%, P = 0.005). DWI lesion of ≥15 mm was found more frequently in TIA patients with AF than in those without AF (45% vs. 22%, P = 0.04). Conclusions: This study demonstrated that TIA patients with AF were older and had a longer length of hospital stay than those without AF. In addition, characteristics of DWI findings differed between TIA patient with AF and those without AF; TIA patients with AF were more likely to have a single lesion with a size of ≥15 mm.

062 ECG Predictors of Adverse Outcomes in Patients With Atrial Fibrillation: A Combined Affirm and AF-CHF Sub-Study

Approximately 10% of all ischaemic strokes occur before the age of 50 years [1]. Their incidence is currently increasing [1, 2] because of a raising prevalence of vascular risk factors in this age category [1]. The specificities of ischaemic strokes and transient ischaemic attacks (TIAs) in young people are that they occur in patients who have familial and professional responsibilities and an otherwise long life-expectancy that includes possible pregnancies in women. Many studies evaluated their outcomes with a follow-up of 3–5 years and considered risks of recurrent vascular events (stroke, myocardial infarction, vascular death) as low [3, 4]. This risk was so low in a Norwegian study that the authors questioned the benefit of aspirin in young patients who have no vascular risk factors and no recurrence after 3 years [4]. However, 3–5 years of follow-up might not be long enough for young patients, and outcomes cannot be restricted only to the risks of recurrence and death. In this issue of the European Journal of Neurology Broman et al. [5] included 396 consecutive patients aged 18−55 years who were admitted in three European centres between 2007 and 2010 for ischaemic strokes or TIAs and were followed up over 11.8 years (interquartile range 10.4–12.7 years). They found that 27 (6.8%) of them had died at the end of the follow-up and 89 (22.5%) had a new recurrent vascular event of any type (any type of cerebrovascular event in 62, 15.7%; any other type of vascular event in 34, 8.6%; both in 5). The cumulative 10-year incidence rate per 1000 person-years was 21.6 (95% confidence interval [CI] 17.1–26.9) for any recurrent vascular event and 14.9 (95% CI 11.3–19.3) for any cerebrovascular event. The prevalence of vascular risk factors increased over time during the follow-up, but 22 (13.5%) patients were left without any secondary preventive medication at the last in-person follow-up visit. Atrial fibrillation at baseline was significantly associated with recurrent vascular events, even after adjustment for demographic characteristics and comorbidities. The authors conclude that risks of recurrent vascular events are high in young patients after ischaemic strokes or TIAs. They suggest also that these risks might be reduced by a better adherence of patients to modern strategies of secondary prevention [5]. In this study, patients were recruited up to 55 years, meaning that they could be 65 years old at the end of the follow-up, an age where atrial fibrillation becomes a frequent cause of cerebral ischaemia. However, the authors performed a subgroup analysis in the 153 patients aged 45 years or less at inclusion (38.6% of the whole study population). In this subgroup, 25 patients (16.3%) had any recurrent vascular event, 18 (11.8%) had any cerebrovascular event and 7 (4.6%) died during the follow-up, leading to cumulative 10-year incidence rates per 1000 person-years of 15.6 (95% CI 10.9–21.9) for any recurrent vascular event, 10.5 (95% CI 6.7–15.7) for any cerebrovascular event and 4.0 (95% CI 2.0–7.5) for death. In those patients who were still younger than 55 years at the end of the follow-up, atrial fibrillation remained significantly associated with a higher hazard for recurrent vascular events (hazard ratio [HR] 9.2, 95% CI 2.3–36.5), together with arterial hypertension (HR 3.5, 95% CI 1.3–9.0) and diabetes mellitus (HR 3.0, 95% CI 1.0–8.5). Despite the possible lack of information concerning patients who could not attend face-to-face follow-up visits and some differences in assessing recurrent events and death between the three participating centres, this study provides important information on the 10-year outcome of young patients who had ischaemic strokes or TIAs. It shows that the risk of new vascular event and death is much higher than previously thought [3, 4], and that one patient in seven did not use any secondary preventive medication, especially those who had TIA as the index event. This study showed also that atrial fibrillation should not be neglected as a potential source of ischaemic event in this age category and is a major predictor of new vascular event. More research is needed concerning the long-term outcome of young patients who have ischaemic strokes or TIAs. Other domains should also be explored beyond death and recurrence of vascular events, such as risks of seizures and epilepsy, cognitive impairment, post-stroke depression, fatigue and social aspects such as difficulties at work and in family life. Large multicentre multinational studies conducted with public funds are needed, meaning that public institutions should also fund studies with long follow-up, which is currently difficult. The author has no conflict of interest. Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Insertable cardiac monitors (ICMs) are often implanted after cryptogenic stroke (CS) to detect atrial fibrillation (AF) and guide anticoagulation. However, the impact of this practice on stroke recurrence remains unclear. This study sought to compare stroke recurrence between CS patients with and without ICM-detected AF and identify predictors of recurrent stroke. A retrospective analysis was conducted of consecutive patients admitted to a stroke center with no history of AF who received an ICM for CS or transient ischemic attack. Among 840 patients (median follow-up 990 days), AF was detected in 235. Recurrent strokes occurred in 112 patients, of which 70.5% occurred in patients without AF ever being detected. Patients with AF were observed to have increased stroke recurrence when analyzing AF as a time-varying exposure (adjusted HR: 2.73; 95% CI: 1.33-5.59), despite most patients with AF being started on anticoagulation (97.9%). Other covariates associated with increased risk of recurrent stroke in multivariable analyses included an elevated CHA2DS2-VASc (Congestive heart failure, Hypertension, Age ≥75 years, Diabetes, history of Stroke or transient ischemic attack, VAscular disease, Age 65-74 years, and Sex category [female]) score ≥4 and male sex. In exploratory subgroup analysis of patients with a CHA2DS2-VASc score <4 and age of <65 years (n = 107), only 1 patient with a recurrent stroke had AF detected (<1%). Despite the high rate of detection of AF following CS by ICMs, most strokes occurring in this population may not be due to AF. Further randomized trials are needed to verify these observations and clarify the benefit of routine ICM implantation for secondary CS prevention.

Is rivaroxaban safer and more effective than warfarin in patients with atrial fibrillation and stroke or TIA?

Background: Large scale data on atrial fibrillation (AF) prevalence in acute ischemic stroke (AIS) is sparse since approval of dabigatran for non-valvular AF in 2010. We studied recent trends in prevalence of AF in AIS and transient ischemic attack (TIA) in the United States (US) and association of AF with in-hospital mortality, cost and length of stay (LOS) in AIS. Methods: Adults admitted to US hospitals from 2007-2012 with diagnosis of AIS (n=3,427,806) and TIA (n=502,820) were identified from the Nationwide Inpatient Sample. Weighted prevalence of AF in AIS and TIA by demographics and region was computed. Multivariate logistic regression was used to evaluate association of AF with other clinical factors and mortality in AIS. Association of AF with LOS and cost was assessed using generalized linear models. Results: AF prevalence increased by 11.5% in AIS (22%-24.5%, p&lt;0.001) and by 29% in TIA (13.5%-17.4%, p&lt;0.001) from 2007-2012. AF prevalence varied by age (AIS: 7% in 50-59yo vs 38% in &gt;80yo; TIA: 5% in 50-59yo vs 27% in &gt;80yo), sex (AIS: 20% in M vs 25% in F); TIA: 16% in M vs 15% in F), race (AIS: whites 26% vs blacks 12%) and region (AIS: Northeast 25% vs South 21%). AF prevalence increased in all subgroups over time (p&lt;0.001) except AIS &lt;40yo and TIA&lt;50yo (Figure 1). Advancing age, female sex, white race, high income, Medicare insurance, CHA 2 DS 2 -VASc score and large hospital size were associated with increased odds of AF in AIS. AF was positively associated with death (OR=1.60, 95%CI 1.56-1.64) but mortality in AIS with AF decreased from 13.2% in 2007 to 10.7% in 2012 (p&lt;0.001). AF was associated with increased cost of $2,631 and LOS 1.1 days in AIS. Conclusion: Prevalence of AF in AIS and TIA has continued to increase. Disparity in AF prevalence in AIS and TIA exists by patient and hospital factors. AF is associated with increased mortality, LOS and hospital cost in AIS but mortality in AIS with AF is decreasing. More AIS preventive efforts are needed in AF patients especially in the elderly.

Background —The role of surgical closure of patent foramen ovale (PFO) for cerebral infarction (CI) or transient ischemic attack (TIA) resulting from paradoxical embolism is unclear, and its effect on recurrence is unknown. Our objective was to determine the outcome of surgical closure of PFO in patients with a prior ischemic neurological event, define the rate of CI or TIA recurrence after PFO closure, and identify risk factors for these recurrences. Methods and Results —We retrospectively analyzed 91 patients (58 men, 33 women) with ≥1 previous cerebrovascular ischemic events who underwent surgical PFO closure between April 1982 and March 1998. The presence of a PFO with a right-to-left shunt was confirmed with transesophageal echocardiography. Mean age was 44.2±12.2 years. The index event was a CI in 59 and a TIA in 32; a Valsalva-like episode preceded the event in 15 patients. Deep venous thrombosis was documented in 9 patients, and a hypercoagulable state was identified in 10. Surgical closure was performed with extracorporeal circulation by either direct suture (n=82) or patch closure (n=9). Limited incisions were used in 18.7% of patients. There was no operative mortality. Morbidity included transient atrial fibrillation (n=11), pericardial drainage for effusion (n=4), exploration for bleeding (n=3), and superficial wound infection (n=1). Follow-up totaled 176.3 patient-years, and mean follow-up was 2.0 years. No one had a CI, and 8 had a TIA during follow-up, with 1 caused by temporal arteritis. Transesophageal echocardiography demonstrated all closures to be intact in these patients. The overall freedom from TIA recurrence during follow-up was 92.5±3.2% at 1 year and 83.4±6.0% at 4 years. Having multiple neurological events before PFO closure was the only significant risk factor for TIA or CI recurrence after closure by univariate analysis ( P =0.05); the small number of post-PFO closure cerebral ischemic events precluded multivariate analysis. Conclusions —Surgical closure of PFO can be performed with minimal morbidity and mortality. PFO closure may decrease the risk of recurrent stroke or TIA and may avoid lifelong anticoagulation in the young adult if there is no other indication. Recurrent cerebrovascular ischemic events after surgery should prompt further evaluation to identify causes other than paradoxical embolism.

Whether statins, as upstream therapy for atrial fibrillation (AF), can reduce the incidence of AF in elderly individuals remains unclear. This study aimed to examine whether statin use is linked to a lower risk of incident AF in older adults from Shanghai, China. To provide important implications for optimizing the prevention and management strategies of AF in the elderly. Data for this retrospective cohort study were collected from a community health service center in Shanghai. Individuals had no history of AF in 2018 and before were included, and new cases of AF were documented in 2019, 2020, and 2021. A Poisson generalized linear model was used to evaluate the relationship between statin treatment and AF incidence. Both standard covariate adjustment and propensity score matching (PSM) to control for confounding factors. To evaluate the risk factors for new-onset AF among the elderly, univariate logistic analysis was conducted. Variables with statistical significance in univariate analysis (p < 0.05), such as sex, age, SBP, BMI, WHR, Scr and HbA1c were selected as covariates for inclusion in the multivariable logistic regression model. Out of the 5675 individuals in the study group (43.5% male; median age of 68years), 456 participants (8.0%) received statin treatment. The propensity score unmatched and matched cohorts of 453 participants (stain users and nou-users) were evaluated, in each group. There was no reduction in AF incidence with statin use (95%CI, 0.948 to 1.018) in the unmatched cohort and in the matched cohort (95%CI, 0.459 to 1.512). The multivariate regression analysis results showed that age, SBP, BMI, HbA1c and Scr were factors independently linked to the occurrence of new-onset AF. Despite extensive analysis, our study did not find evidence that statin use reduces the risk of new-onset AF in this population. For older people, taking statins did not reduce risk of new on-set AF. Independent predictors of AF onset including age, SBP, BMI, HbA1c and Scr. This is a retrospective cohort study, making it difficult to completely control for all confounding factors, and the short follow-up period may have resulted in insufficient statistical power to detect a meaningful effect of statins on atrial fibrillation. Future studies may use randomized controlled studies to explore the effects of different types and doses of statins on new-onset atrial fibrillation in older adults.

The nonrandom occurrence of visits to the emergency department

Background and purpose: To date, there have been no thorough investigations on the possible role of atrial fibrillation (AF) pattern in stroke recurrence during early-phase post stroke. We investigated for differences in ischemic outcome between patients with paroxysmal AF (PAF) and those with sustained AF (SAF), within 90 days from acute ischemic stroke (AIS). Methods: Patients from the RAF and RAF-DOAC study databases were categorized into 2 groups: PAF and SAF (persistent + permanent AF). Early recurrence was defined as the composite of ischemic stroke, transient ischemic attack and symptomatic systemic embolism within 90 days from AIS. Results: A total of 2,150 patients (1,155 females, 53.7%) were enrolled: 930 (43.3%) had PAF and 1220 (56.7%) SAF. Patients with SAF were on average older, were more likely to have: diabetes mellitus, hypertension, a history of stroke/TIA, congestive heart failure, atrial enlargement, pacemaker and a high NIHSS. During the 90-day follow-up, 107 recurrent ischemic events were recorded from index stroke. Multivariate analysis for predictive factors of early ischemic recurrence indicated that SAF was associated with a non-significant increase in ischemic events at 90 days, compared to PAF (HR 1.23; 95% CI 0.74-2.04; p = 0.418) (Figure). Conclusions: PAF and SAF patients with acute stroke resulted having similar risks of ischemic recurrence within 90 days from index event. Therein, these findings suggest that AF pattern should not influence decision making concerning the timing of anticoagulant administration after an acute stroke.

Short and Long-Term Risk of Major Cardiovascular Events after Ischemic Stroke or Transient Ischemic Attack in Myeloproliferative Neoplasms

Background: Several clinical scales have been developed for predicting stroke recurrence. These clinical scores could be extremely useful to guide triage decisions. Our goal was to compare the very early predictive accuracy of the most relevant clinical scores [age, blood pressure, clinical features and duration of symptoms (ABCD) score, ABCD and diabetes (ABCD2) score, ABCD and brain infarction on imaging score, ABCD2 and brain infarction on imaging score, ABCD and prior TIA within 1 week of the index event (ABCD3) score, California Risk Score, Essen Stroke Risk Score and Stroke Prognosis Instrument II] in consecutive transient ischemic attack (TIA) patients. Methods: Between April 2008 and December 2009, we included 1,255 consecutive TIA patients from 30 Spanish stroke centers (PROMAPA study). A neurologist treated all patients within the first 48 h after symptom onset. The duration and typology of clinical symptoms, vascular risk factors and etiological work-ups were prospectively recorded in a case report form in order to calculate established prognostic scores. We determined the early short-term risk of stroke (at 7 and 90 days). To evaluate the performance of each model, we calculated the area under the receiver operating characteristic curve. Cox proportional hazards multivariate analyses determining independent predictors of stroke recurrence using the different components of all clinical scores were calculated. Results: We calculated clinical scales for 1,137 patients (90.6%). Seven-day and 90-day stroke risks were 2.6 and 3.8%, respectively. Large-artery atherosclerosis (LAA) was observed in 190 patients (16.7%). We could confirm the predictive value of the ABCD3 score for stroke recurrence at the 7-day follow-up [0.66, 95% confidence interval (CI) 0.54–0.77] and 90-day follow-up (0.61, 95% CI 0.52–0.70), which improved when we added vascular imaging information and derived ABCD3V scores by assigning 2 points for at least 50% symptomatic stenosis on carotid or intracranial imaging (0.69, 95% CI 0.57–0.81, and 0.63, 95% CI 0.51–0.69, respectively). When we evaluated each component of all clinical scores using Cox regression analyses, we observed that prior TIA and LAA were independent predictors of stroke recurrence at the 7-day follow-up [hazard ratio (HR) 3.97, 95% CI 1.91–8.26, p < 0.001, and HR 3.11, 95% CI 1.47–6.58, p = 0.003, respectively] and 90-day follow-up (HR 2.35, 95% CI 1.28–4.31, p = 0.006, and HR 2.20, 95% CI 1.15–4.21, p = 0.018, respectively). Conclusion: All published scores that do not take into account vascular imaging or prior TIA when identifying stroke risk after TIA failed to predict risk when applied by neurologists. Clinical scores were not able to replace extensive emergent diagnostic evaluations such as vascular imaging, and they should take into account unstable patients with recent prior transient episodes.

Paroxysmal atrial fibrillation is often suspected as a probable cause of cryptogenic stroke. Continuous long-term ECG monitoring using insertable cardiac monitors is a clinically effective technique to screen for atrial fibrillation and superior to conventional follow-up in cryptogenic stroke. However, more studies are needed to identify factors which can help selecting patients with the highest possibility of detecting atrial fibrillation with prolonged rhythm monitoring. The clinical relevance of short-term atrial fibrillation, the need for medical intervention and the evaluation as to whether intervention results in improved clinical outcomes should be assessed. The Nordic Atrial Fibrillation and Stroke Study is an international, multicentre, prospective, observational trial evaluating the occurrence of occult atrial fibrillation in cryptogenic stroke and transient ischaemic attack. Patients with cryptogenic stroke or transient ischaemic attack from the Nordic countries are included and will have the Reveal LINQ® Insertable cardiac monitor system implanted for 12 months for atrial fibrillation detection. Biomarkers which can be used as predictors for atrial fibrillation and may identify patients, who could derive the most clinical benefit from the detection of atrial fibrillation by prolonged monitoring, are being studied. The primary endpoint is atrial fibrillation burden within 12 months of continuous rhythm monitoring. Secondary endpoints are atrial fibrillation burden within six months, levels of biomarkers predicting atrial fibrillation, CHA2DS2-VASc score, incidence of recurrent stroke or transient ischaemic attack, use of anticoagulation and antiarrhythmic drugs, and quality of life measurements. The clinical follow-up period is 12 months. The study started in 2017 and the completion is expected at the end of 2020.