Abstract P890: Associating Cryptogenic Ischemic Stroke In The Young With Cardiovascular Risk Factor Phenotypes In A Diverse Urban Community

Introduction: Acute Ischemic Stroke (AIS) in the young is increasing in prevalence and the largest sub-type within this cohort is cryptogenic. The risk factors and etiologies for these strokes likely differ by socioeconomic, racial, and ethnic background. To curb this trend, new ways of defining cryptogenic stroke and its risk factors are needed that can be applied to different populations. We aimed to create such a framework using patients in one of the poorest and most diverse urban counties in the country: Bronx, NY.

Methods: We conducted a retrospective cohort study of AIS patients aged 18-49 who presented to an urban tertiary care center. Stroke risk factor phenotypes were determined by multivariate analysis and resultant models were applied to cryptogenic stroke cases.

Results: A total 449 patients met inclusion criteria. The mean age was 41, 49% were women, 39% were Black, and 32% were Hispanic. 133 patients had strokes due to small and large vessel disease (vascular phenotype); these patients had higher rates of hypertension, intracranial atherosclerosis, and diabetes mellitus, and higher admission glucose, HbA1c, admission blood pressure, and cholesterol compared to the patients with cardioembolic AIS. The 69 patients with strokes due to cardioembolism (cardiac phenotype) had significantly higher rates of congestive heart failure (CHF), rheumatic heart disease, atrial fibrillation, clotting disorders, left ventricular hypertrophy, larger left atrial sizes, lower ejection fractions, and higher B-type natriuretic peptide and troponin levels. There were no differences in stroke subtype by race or ethnicity. Adjusted multivariate analysis produced 6 variables independently associated with the vascular phenotype (age, male sex, hemoglobin A1c, EF, LDL cholesterol, and family history of AIS) and 5 independently associated with the cardiac phenotype (age, female sex, decreased EF, CHF, and absence of intracranial atherosclerosis). Applying these models to 97 cryptogenic stroke cases yielded that 51.5% fit the vascular phenotype and 3.1% fit the cardiac phenotype.

Conclusion: In our cohort of young patients in a low-resource, diverse urban community, half of cryptogenic cases fit the risk factor phenotype of small and large vessel strokes.

Abstract P387: Stroke Despite Anticoagulation Therapy Predicts Early Recurrence After Cardioembolic Stroke: The IAC-Study

Background and purpose: A proportion of patients with ischemic stroke and atrial fibrillation (AF) have an ischemic stroke despite being prescribed anticoagulation therapy. In this study of patients with ischemic stroke in the setting of AF, we aim to determine the association between prior anticoagulant therapy and 90-day recurrent ischemic events and delayed symptomatic intracranial hemorrhage (sICH).

Methods: We included consecutive patients with acute ischemic stroke and AF from the Initiation of Anticoagulation after Cardioembolic stroke (IAC) study from 8 comprehensive stroke centers in the United States. We compared recurrent ischemic events and delayed sICH risk using unadjusted and adjusted cox-regression analyses between patients who were prescribed anticoagulation (AC p ) vs. were naïve to anticoagulation therapy prior to the ischemic stroke (AC n ). For ischemic events, we adjusted for CHA 2 DS 2 -Vasc, anticoagulation initiation, and switching anticoagulant (DOAC to Warfarin or Warfarin to DOAC). For delayed sICH, we adjusted for age, sex, NIHSS score, and early hemorrhagic transformation.

Results: 2070 patients had home anticoagulation treatment status recorded. When compared to the AC n group, the AC p group were more likely to have higher median (IQR) CHA 2 DS 2 -Vasc score [5 (4-6) vs. 5 (3-6), p = 0.001], lower NIHSS score [8 (3 - 18) vs. 10 (4 - 18), p = 0.015], severe left atrial enlargement (43.5% vs. 34.5%, p < 0.001), and less likely to receive alteplase (14.4% vs. 36.2%, p < 0.001). In the adjusted cox hazard model, AC p was associated with increased risk of 90-day recurrent ischemic events (adjusted HR 1.52 95% CI 1.01 - 2.29, p = 0.047) but not increased risk of 90-day sICH (adjusted HR 1.10 95% CI 0.46 - 2.61, p = 0.838). In a sensitivity analysis, with severe left atrial enlargement added to the model, the association between AC p and recurrent ischemic event risk did not meaningfully change (adjusted HR 1.41 95% CI 0.87 - 2.28, p = 0.162).

Conclusion: Patients with AF and ischemic stroke despite being prescribed anticoagulation therapy are at higher risk of recurrent events. Studies are needed to understand mechanisms of ischemic stroke in these patients to improve stroke prevention strategies.

Abstract P389: External Validation of the 2CAN Score for Inpatient Stroke Detection

Introduction: Up to 15% of all strokes occur in patients who are already hospitalized for other conditions. A validated clinical tool to help rapidly discriminate between mimics and stroke among inpatients could greatly improve acute stroke care. Recently, the 2CAN score was developed and validated at a single Midwest academic medical center to identify inpatient strokes; a score of ≥2 was highly sensitive and specific for stroke. We sought to externally validate the 2CAN score at our institution.

Methods: We conducted a retrospective cohort study of consecutive inpatient stroke codes at a single Northeast academic medical center from 7/1/2018 to 11/1/2019. Pre-specified variables, including patient demographics, vascular risk factors, and clinical features (neurological examination, vital signs, laboratory values, and final diagnoses), were abstracted from the electronic medical record. We determined the sensitivity, specificity, positive and negative predictive value of a 2CAN score ≥2 for stroke (ischemic stroke, hemorrhagic stroke, or TIA) in our cohort. The 2CAN score consists of clinical deficit score (0-3 points), recent cardiac procedure (1 point), atrial fibrillation (1 point), and code called within 24 hours of admission (1 point). We used multivariate logistic regression to identify additional determinants of stroke.

Results: We identified 111 inpatient stroke codes on 110 patients, mean age 67 ± 1 year, 46.8% women, and 73.8% Black or Hispanic. Final diagnosis was stroke for 54 codes (48.6%) and mimic for 57 codes (51.3%), most commonly toxic-metabolic encephalopathy. 2CAN score ≥2 had 96.3% sensitivity, 45.6% specificity, 62.7% positive predictive value, and 92.3% negative predictive value for stroke. In a multivariable logistic regression model, only recent cardiac procedure (OR: 5.5; 95% CI: 1.1-27.5) and high clinical deficit score (OR: 3.9; 95% CI: 1.9-6.1) predicted stroke.

Conclusion: The 2CAN score is externally valid and helps distinguish stroke from mimic in inpatients; having a score of <2 makes stroke very unlikely.

Abstract P10: Posterior Circulation Strokes Are Less Likely to Receive Alteplase or Mechanical Thrombectomy: Analysis From the IAC Study

Background: Emergent treatment with intravenous thrombolysis and mechanical thrombectomy improved outcomes in patients with acute ischemic stroke. We aim to identify differences in acute stroke treatment trends between strokes occurring in the anterior versus posterior circulation.

Methods: The IAC (Initiation of Anticoagulation after Cardioembolic stroke) study represents pooled data registry of 8 comprehensive stroke centers across the United States and included patients with cardioembolic stroke in the setting of AFib. In a post hoc analysis, we identified and separated patients into posterior circulation stroke (PCS) and anterior circulation stroke (ACS) groups based on imaging. Patients without infarct locations or those with multi-circulation infarcts were excluded. We compared baseline characteristics, stroke severity and the treatment trends with alteplase (tPA) and mechanical thrombectomy (MT) in PCS vs ACS using Fisher exact test, t-test and non-parametric tests. We then performed multivariable logistic regression adjusted for baseline differences to determine the associations between PCS and tPA or MT.

Results: Of the 2084 patients in IAC cohort, 1589 met inclusion criteria for this study, in which 294 (22.7%) had PCS. Mean age was 76.8 years, 29.3% received tPA and 26.9% had MT. When compared to ACS, patients with PCS were more likely to be men (55.4% vs 45.6%, p=0.003), have diabetes (42.8% vs 29.8, p< 0.001) and lower median NIHSS score on admission (4 vs 8, p<0.001). Patients with PCS were less likely to receive tPA (16.3% vs 32.3%, p<0.001) or MT (10.9% vs 30.6%, p<0.001). Other variables were not significantly different. When adjusted for baseline differences, patients with PCS remained less likely to be treated with tPA (adjusted OR 0.49, 95%CI 0.35-0.70, p<0.001) or MT (adjusted OR 0.38, 95%CI 0.25-0.58, p<0.001).

Conclusion: Posterior circulation strokes are half as likely to receive thrombolytic therapy and almost a third as likely to have thrombectomy, even after adjusting for baseline stroke severity scores. This is possibly due to difficulty in timely identification and diagnostic delays. There is need for better tools incorporating posterior circulation stroke signs and symptoms to allow for early detection and treatment.

Associating cryptogenic ischemic stroke in the young with cardiovascular risk factor phenotypes

Acute Ischemic Stroke (AIS) in the young is increasing in prevalence and the largest subtype within this cohort is cryptogenic. To curb this trend, new ways of defining cryptogenic stroke and associated risk factors are needed. We aimed to gain insights into the presence or absence of cardiovascular risk factors in cases of cryptogenic stroke. We conducted a retrospective cohort study of patients aged 18-49 who presented to an urban tertiary care center with AIS. We manually collected predefined demographic, clinical, laboratory and radiological variables. Clinical risk phenotypes were determined using these variables through multivariate analysis of patients with the small and large vessel disease subtypes (vascular phenotype) and cardioembolic subtype (cardiac phenotype). The resultant phenotype models were applied to cases deemed cryptogenic. Within the 449 patients who met criteria, patients with small and large vessel disease (vascular phenotype) had higher rates of hypertension, intracranial atherosclerosis, and diabetes mellitus, and higher admission glucose, HbA1c, admission blood pressure, and cholesterol compared to the patients with cardioembolic AIS. The cardioembolic subgroup (cardiac phenotype) had significantly higher rates of congestive heart failure (CHF), rheumatic heart disease, atrial fibrillation, clotting disorders, left ventricular hypertrophy, larger left atrial sizes, lower ejection fractions, and higher B-type natriuretic peptide and troponin levels. Adjusted multivariate analysis produced six variables independently associated with the vascular phenotype (age, male sex, hemoglobin A1c, ejection fraction (EF), low-density lipoprotein (LDL) cholesterol, and family history of AIS) and five independently associated with the cardiac phenotype (age, female sex, decreased EF, CHF, and absence of intracranial atherosclerosis). Applying these models to cryptogenic stroke cases yielded that 51.5% fit the vascular phenotype and 3.1% fit the cardiac phenotype. In our cohort, half of young patients with cryptogenic stroke fit the risk factor phenotype of small and large vessel strokes.

Prevalence of Cervical Artery Dissection Among Hospitalized Patients With Stroke by Age in a Nationally Representative Sample From the United States

OBJECTIVE

To test the hypothesis that the prevalence of cervical artery dissection remains constant across age groups, we evaluated the relationship between age and cervical artery dissection in patients with stroke using a nationally representative sample from the United States.

METHODS

We used inpatient claims data included in the 2012-2015 releases of the National Inpatient Sample (NIS). We used validated ICD-9-CM codes to identify adults hospitalized with ischemic stroke and a concomitant diagnosis of carotid or vertebral artery dissection. Survey weights provided by the NIS and population estimates from the US census were used to calculate nationally representative estimates. The χ² test for trend was used to compare the prevalence of concomitant dissection among stroke hospitalizations across patient subgroups defined by age. Poisson regression and the Wald test for trend were used to evaluate whether the prevalence of hospitalizations for stroke and concomitant dissection per million person-years varied by age groups.

RESULTS

There were 17,320 (95% confidence interval [CI], 15,614-19,026) hospitalizations involving ischemic stroke and a concomitant dissection. The prevalence of dissection among stroke hospitalizations decreased across 10-year age groups from 7.2% (95% CI, 6.2%-8.1%) among persons younger than 30 years to 0.2% (95% CI, 0.1%-0.2%) among persons older than 80 years (p value for trend <0.001). However, the prevalence of hospitalizations for stroke and concomitant dissection increased from 5.4 (95% CI, 4.6-6.2) hospitalizations per million person-years among adults younger than 30 to 24.4 (95% CI, 21.0-27.9) hospitalizations per million person-years among adults older than age 80 (p value for trend <0.01).

CONCLUSION

In a nationally representative sample, the prevalence of hospitalizations for dissection-related stroke increased with age.

Diagnostic Challenges in Outpatient Stroke: Stroke Chameleons and Atypical Stroke Syndromes

Failure to diagnose transient ischemic attack (TIA) or stroke in a timely fashion is associated with significant patient morbidity and mortality. In the outpatient or clinic setting, we suspect that patients with minor, transient, and atypical manifestations of cerebrovascular disease are most prone to missed or delayed diagnosis. We therefore detail common stroke chameleon symptoms as well as atypical stroke presentations, broadly review new developments in the study of diagnostic error in the outpatient setting, suggest practical clinical strategies for diagnostic error reduction, and emphasize the need for rapid consultation of stroke specialists when appropriate. We also address the role of psychiatric disease and vascular risk factors in the diagnostic evaluation and treatment of suspected stroke/TIA patients. We advocate incorporating diagnostic time-outs into clinical practice to assure that the diagnosis of TIA or stroke is considered in all relevant patient encounters after a detailed history and examination are conducted in the outpatient setting.

Breakdowns in the initial patient-provider encounter are a frequent source of diagnostic error among ischemic stroke cases included in a large medical malpractice claims database

Background Misdiagnosis of dangerous cerebrovascular disease is a substantial public health problem. We sought to identify and describe breakdowns in the diagnostic process among patients with ischemic stroke to facilitate future improvements in diagnostic accuracy. Methods We performed a retrospective, descriptive study of medical malpractice claims housed in the Controlled Risk Insurance Company (CRICO) Strategies Comparative Benchmarking System (CBS) database from 1/1/2006 to 1/1/2016 involving ischemic stroke patients. Baseline claimant demographics, clinical setting, primary allegation category, and outcomes were abstracted. Among cases with a primary diagnosis-related allegation, we detail presenting symptoms and diagnostic breakdowns using CRICO's proprietary taxonomy. Results A total of 478 claims met inclusion criteria; 235 (49.2%) with diagnostic error. Diagnostic errors originated in the emergency department (ED) in 46.4% (n = 109) of cases, outpatient clinic in 27.7% (n = 65), and inpatient setting in 25.1% (n = 59). Across care-settings, the most frequent process breakdown was in the initial patient-provider encounter [76.2% (n = 179 cases)]. Failure to assess, communicate, and respond to ongoing symptoms was the component of the patient-provider encounter most frequently identified as a source of misdiagnosis in the ED. Exclusively non-traditional presenting symptoms occurred in 35.7% (n = 84), mixed traditional and non-traditional symptoms in 30.6% (n = 72), and exclusively traditional in 23.8% (n = 56) of diagnostic error cases. Conclusions Among ischemic stroke patients, breakdowns in the initial patient-provider encounter were the most frequent source of diagnostic error. Targeted interventions should focus on the initial diagnostic encounter, particularly for ischemic stroke patients with atypical symptoms.

Factors associated with hospitalization for ischemic stroke and TIA following an emergency department headache visit

BACKGROUND

Misdiagnosis of cerebrovascular disease among Emergency Department (ED) patients with headache has been reported. We hypothesized that markers of substandard diagnostic processes would be associated with subsequent ischemic cerebrovascular events among patients discharged from the ED with a headache diagnosis even after adjusting for demographic variables and medical history.

METHODS

We conducted a case-control study of adult ED patients diagnosed with a primary headache disorder at Montefiore Medical Center from 9/1/2013-9/1/2018. Cases were defined as patients hospitalized for an ischemic stroke or TIA within 365 days of their index ED visit. Control patients were defined as those who lacked a subsequent hospitalization for cerebrovascular disease. Pre-specified demographic, clinical, and diagnostic process factors were compared between groups; conditional logistic regression was used to assess the separate and joint influence of baseline features on risk of cerebral ischemia.

RESULTS

A total of 93 consecutive headache patients with a subsequent ischemic stroke/TIA hospitalization were matched to 93 controls (n = 186). Cases were older than controls and more likely to have traditional cerebrovascular risk factors. Neurological consultation was obtained more often for cases (13% vs. 4%; P = 0.03), cases were in the ED for longer (6 vs. 5 h, P = 0.03), and more frequently received neuroimaging (80% vs. 48%; P < 0.0001). Rates of neurological examination, documented differential diagnoses, and clear discharge follow up plans were similar between cases and controls. In our conditional logistic regression model, only history of prior stroke/TIA was associated with increased odds of subsequent cerebral ischemia.

CONCLUSION

Factors associated with diagnostic process failures did not increase the odds of subsequent ischemic stroke/TIA hospitalization following ED headache visit in our study.

Anticoagulation Type and Early Recurrence in Cardioembolic Stroke: The IAC Study

BACKGROUND AND PURPOSE

In patients with acute ischemic stroke and atrial fibrillation, treatment with low molecular weight heparin increases early hemorrhagic risk without reducing early recurrence, and there is limited data comparing warfarin to direct oral anticoagulant (DOAC) therapy. We aim to compare the effects of the treatments above on the risk of 90-day recurrent ischemic events and delayed symptomatic intracranial hemorrhage.

METHODS

We included consecutive patients with acute ischemic stroke and atrial fibrillation from the IAC (Initiation of Anticoagulation after Cardioembolic) stroke study pooling data from stroke registries of 8 comprehensive stroke centers across the United States. We compared recurrent ischemic events and delayed symptomatic intracranial hemorrhage between each of the following groups in separate Cox-regression analyses: (1) DOAC versus warfarin and (2) bridging with heparin/low molecular weight heparin versus no bridging, adjusting for pertinent confounders to test these associations.

RESULTS

We identified 1289 patients who met the bridging versus no bridging analysis inclusion criteria and 1251 patients who met the DOAC versus warfarin analysis inclusion criteria. In adjusted Cox-regression models, bridging (versus no bridging) treatment was associated with a high risk of delayed symptomatic intracranial hemorrhage (hazard ratio, 2.74 [95% CI, 1.01-7.42]) but a similar rate of recurrent ischemic events (hazard ratio, 1.23 [95% CI, 0.63-2.40]). Furthermore, DOAC (versus warfarin) treatment was associated with a lower risk of recurrent ischemic events (hazard ratio, 0.51 [95% CI, 0.29-0.87]) but not delayed symptomatic intracranial hemorrhage (hazard ratio, 0.57 [95% CI, 0.22-1.48]).

CONCLUSIONS

Our study suggests that patients with ischemic stroke and atrial fibrillation would benefit from the initiation of a DOAC without bridging therapy. Due to our study limitations, these findings should be interpreted with caution pending confirmation from large prospective studies.

Anticoagulation Timing in Cardioembolic Stroke and Recurrent Event Risk

OBJECTIVE

Guidelines recommend initiating anticoagulation within 4 to 14 days after cardioembolic stroke. Data supporting this did not account for key factors potentially affecting the decision to initiate anticoagulation, such as infarct size, hemorrhagic transformation, or high-risk features on echocardiography.

METHODS

We pooled data from stroke registries of 8 comprehensive stroke centers across the United States. We included consecutive patients admitted with ischemic stroke and atrial fibrillation. The primary predictor was timing of initiating anticoagulation (0-3 days, 4-14 days, or >14 days), and outcomes were recurrent stroke/transient ischemic attack/systemic embolism, symptomatic intracerebral hemorrhage (sICH), and major extracranial hemorrhage (ECH) within 90 days.

RESULTS

Among 2,084 patients, 1,289 met the inclusion criteria. The combined endpoint occurred in 10.1% (n = 130) subjects (87 ischemic events, 20 sICH, and 29 ECH). Overall, there was no significant difference in the composite endpoint between the 3 groups (0-3 days: 10.3%, 64/617; 4-14 days: 9.7%, 52/535; >14 days: 10.2%, 14/137; p = 0.933). In adjusted models, patients started on anticoagulation between 4 and 14 days did not have a lower rate of sICH (vs 0-3 days; odds ratio [OR] = 1.49, 95% confidence interval [CI] = 0.50-4.43), nor did they have a lower rate of recurrent ischemic events (vs >14 days; OR = 0.76, 95% CI = 0.36-1.62, p = 0.482).

INTERPRETATION

In this multicenter real-world cohort, the recommended (4-14 days) time frame to start oral anticoagulation was not associated with reduced ischemic and hemorrhagic outcomes. Randomized trials are required to determine the optimal timing of anticoagulation initiation. ANN NEUROL 2020;88:807-816.

WITHDRAWN: Neurological examination, rather than vascular risk factor assessment, serves to distinguish strokes from stroke mimics in a population with high prevalence of vascular risk factors

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at https://www.elsevier.com/about/our-business/policies/article-withdrawal.

Early ischaemic and haemorrhagic complications after atrial fibrillation-related ischaemic stroke: analysis of the IAC study

INTRODUCTION

Predictors of long-term ischaemic and haemorrhagic complications in atrial fibrillation (AF) have been studied, but there are limited data on predictors of early ischaemic and haemorrhagic complications after AF-associated ischaemic stroke. We sought to determine these predictors.

METHODS

The Initiation of Anticoagulation after Cardioembolic stroke study is a multicentre retrospective study across that pooled data from consecutive patients with ischaemic stroke in the setting of AF from stroke registries across eight comprehensive stroke centres in the USA. The coprimary outcomes were recurrent ischaemic event (stroke/TIA/systemic arterial embolism) and delayed symptomatic intracranial haemorrhage (d-sICH) within 90 days. We performed univariate analyses and Cox regression analyses including important predictors on univariate analyses to determine independent predictors of early ischaemic events (stroke/TIA/systemic embolism) and d-sICH.

RESULTS

Out of 2084 patients, 1520 patients qualified; 104 patients (6.8%) had recurrent ischaemic events and 23 patients (1.5%) had d-sICH within 90 days from the index event. In Cox regression models, factors associated with a trend for recurrent ischaemic events were prior stroke or transient ischemic attack (TIA) (HR 1.42, 95% CI 0.96 to 2.10) and ipsilateral arterial stenosis with 50%-99% narrowing (HR 1.54, 95% CI 0.98 to 2.43). Those associated with sICH were male sex (HR 2.68, 95% CI 1.06 to 6.83), history of hyperlipidaemia (HR 2.91, 95% CI 1.08 to 7.84) and early haemorrhagic transformation (HR 5.35, 95% CI 2.22 to 12.92).

CONCLUSION

In patients with ischaemic stroke and AF, predictors of d-sICH are different than those of recurrent ischaemic events; therefore, recognising these predictors may help inform early stroke versus d-sICH prevention strategies.

Factors associated with therapeutic anticoagulation status in patients with ischemic stroke and atrial fibrillation

BACKGROUND AND PURPOSE

Understanding factors associated with ischemic stroke despite therapeutic anticoagulation is an important goal to improve stroke prevention strategies in patients with atrial fibrillation (AF). We aim to determine factors associated with therapeutic or supratherapeutic anticoagulation status at the time of ischemic stroke in patients with AF.

METHODS

The Initiation of Anticoagulation after Cardioembolic stroke (IAC) study is a multicenter study pooling data from stroke registries of eight comprehensive stroke centers across the United States. Consecutive patients hospitalized with acute ischemic stroke in the setting of AF were included in the IAC cohort. For this study, we only included patients who reported taking warfarin at the time of the ischemic stroke. Patients not on anticoagulation and patients who reported use of a direct oral anticoagulant were excluded. Analyses were stratified based on therapeutic (INR ≥2) versus subtherapeutic (INR <2) anticoagulation status. We used binary logistic regression models to determine factors independently associated with anticoagulation status after adjustment for pertinent confounders. In particular, we sought to determine whether atherosclerosis with 50% or more luminal narrowing in an artery supplying the infarct (a marker for a competing atherosclerotic mechanism) and small stroke size (≤ 10 mL; implying a competing small vessel disease mechanism) related to anticoagulant status.

RESULTS

Of the 2084 patients enrolled in the IAC study, 382 patients met the inclusion criteria. The mean age was 77.4 ± 10.9 years and 52.4% (200/382) were women. A total of 222 (58.1%) subjects presented with subtherapeutic INR. In adjusted models, small stroke size (OR 1.74 95% CI 1.10-2.76, p = 0.019) and atherosclerosis with 50% or more narrowing in an artery supplying the infarct (OR 1.96 95% CI 1.06-3.63, p = 0.031) were independently associated with INR ≥2 at the time of their index stroke.

CONCLUSION

Small stroke size (≤ 10 ml) and ipsilateral atherosclerosis with 50% or more narrowing may indicate a competing stroke mechanism. There may be important opportunities to improve stroke prevention strategies for patients with AF by targeting additional ischemic stroke mechanisms to improve patient outcomes.

Misdiagnosis of Cervicocephalic Artery Dissection in the Emergency Department

Background and Purpose- Cervicocephalic artery dissection is an important cause of stroke. The clinical presentation of dissection can resemble that of benign neurological conditions leading to delayed or missed diagnosis. Methods- We performed a retrospective cohort study using statewide administrative claims data from all Emergency Department visits and admissions at nonfederal hospitals in Florida from 2005 to 2015 and New York from 2006 to 2015. Using validated International Classification of Diseases, Ninth Revision, CM codes, we identified adult patients hospitalized for cervicocephalic artery dissection. We defined probable misdiagnosis of dissection as having an Emergency Department treat-and-release visit for symptoms or signs of dissection, including headache, neck pain, and focal neurological deficits in the 14 days before dissection diagnosis. Multivariable logistic regression was used to compare adverse clinical outcomes in patients with and without probable misdiagnosis. Results- Among 7090 patients diagnosed with a dissection (mean age 52.7 years, 44.9% women), 218 (3.1% [95% CI, 2.7%-3.5%]) had a preceding probable Emergency Department misdiagnosis. After adjustment for demographics and vascular risk factors, there were no differences in rates of stroke (odds ratio, 0.82 [95% CI, 0.62-1.09]) or in-hospital death (odds ratio, 0.26 [95% CI, 0.07-1.08]) between dissection patients with and without a probable misdiagnosis at index hospitalization. Conclusions- We found that ≈1 in 30 dissection patients was probably misdiagnosed in the 2 weeks before their diagnosis.

Assessing diagnostic error in cerebral venous thrombosis via detailed chart review

Background Diagnostic error in cerebral venous thrombosis (CVT) has been understudied despite the harm associated with misdiagnosis of other cerebrovascular diseases as well as the known challenges of evaluating non-specific neurological symptoms in clinical practice. Methods We conducted a retrospective cohort study of CVT patients hospitalized at a single center. Two independent reviewers used a medical record review tool, the Safer Dx Instrument, to identify diagnostic errors. Demographic and clinical factors were abstracted. We compared subjects with and without a diagnostic error using the t-test for continuous variables and the chi-square (χ2) test or Fisher's exact test for categorical variables; an alpha of 0.05 was the cutoff for significance. Results A total of 72 CVT patients initially met study inclusion criteria; 19 were excluded due to incomplete medical records. Of the 53 patients included in the final analysis, the mean age was 48 years and 32 (60.4%) were women. Diagnostic error occurred in 11 cases [20.8%; 95% confidence interval (CI) 11.8-33.6%]. Subjects with diagnostic errors were younger (42 vs. 49 years, p = 0.13), more often women (81.8% vs. 54.8%, p = 0.17), and were significantly more likely to have a past medical history of a headache disorder prior to the index CVT visit (7.1% vs. 36.4%, p = 0.03). Conclusions Nearly one in five patients with complete medical records experienced a diagnostic error. Prior history of headache was the only evaluated clinical factor that was more common among those with an error in diagnosis. Future work on distinguishing primary from secondary headaches to improve diagnostic accuracy in acute neurological disease is warranted.

Redefined Measure of Early Neurological Improvement Shows Treatment Benefit of Alteplase Over Placebo

Background and Purpose- The first of the 2 NINDS (National Institute of Neurological Disorders and Stroke) Study trials did not show a significant increase in early neurological improvement, defined as National Institutes of Health Stroke Scale (NIHSS) improvement by ≥4, with alteplase treatment. We hypothesized that early neurological improvement defined as a percentage change in NIHSS (percent change NIHSS) at 24 hours is superior to other definitions in predicting 3-month functional outcomes and using this definition there would be treatment benefit of alteplase over placebo at 24 hours. Methods- We analyzed the NINDS rt-PA Stroke Study (Parts 1 and 2) trial data. Percent change NIHSS was defined as ([admission NIHSS score-24-hour NIHSS score]×100/admission NIHSS score] and delta NIHSS as (admission NIHSS score-24-hour NIHSS score). We compared early neurological improvement using these definitions between alteplase versus placebo patients. We also used receiver operating characteristic curve to determine the predictive association of early neurological improvement with excellent 3-month functional outcomes (Barthel Index score of 95-100 and modified Rankin Scale score of 0-1), good 3-month functional outcome (modified Rankin Scale score of 0-2), and 3-month infarct volume. Results- There was a significantly greater improvement in the 24-hour median percent change NIHSS among patients treated with alteplase compared with the placebo group (28% versus 15%; P=0.045) but not median delta NIHSS (3 versus 2; P=0.471). Receiver operating characteristic curve comparison showed that percent change NIHSS (ROC_percent) was better than delta NIHSS (ROC_delta) and admission NIHSS (ROC_admission) with regards to excellent 3-month Barthel Index (ROC_percent, 0.83; ROC_delta, 0.76; ROC_admission, 0.75), excellent 3-month modified Rankin Scale (ROC_percent, 0.83; ROC_delta, 0.74; ROC_admission, 0.78), and good 3-month modified Rankin Scale (ROC_percent, 0.83; ROC_delta, 0.76; ROC_admission, 0.78). Conclusions- In the NINDS rt-PA trial, alteplase was associated with a significant percent change improvement in NIHSS at 24 hours. Percent change in NIHSS may be a better surrogate marker of thrombolytic activity and 3-month outcomes.

Abstract WP106: Redefined Measure of Early Neurological Improvement Shows Treatment Benefit of Intravenous Tissue Plasminogen Activator Treatment in NINDS Rt-PA Acute Stroke Trial at 24 Hours

Background and Purpose: The first of the 2 NINDS Stroke Study trials did not show a significant increase in early neurological improvement (ENI), defined as NIHSS improvement by ≥ 4, with alteplase treatment. We hypothesized that ENI defined as a percentage change in NIHSS (percent change NIHSS) at 24 hours is superior to other definitions in predicting 3-month functional outcomes and using this definition there would be treatment benefit of alteplase over placebo at 24 hours.

Methods: We analyzed the NINDS rt-PA Stroke Study (Parts 1 and 2) trial data. Percent change NIHSS was defined as [(admission NIHSS score–24-hour NIHSS score)x100/admission NIHSS score] and delta NIHSS as (admission NIHSS score–24-hour NIHSS score). We compared ENI using these definitions between alteplase vs. placebo patients. We also used receiver operating characteristic (ROC) curve to determine the predictive association of ENI with excellent 3-month functional outcomes [Barthel Index (BI) score 95 – 100 and modified Rankin scale (mRS) 0-1], good 3-month functional outcome (mRS 0-2) and 3-month infarct volume.

Results: There was a significantly greater improvement in the 24-hour median percent change NIHSS among patients treated with alteplase compared to the placebo group (28% vs. 15%, p = 0.045) but not median delta NIHSS (3 vs. 2, p = 0.471). ROC curve comparison showed that percent change NIHSS (ROC percent ) was better than delta NIHSS (ROC delta ) and admission NIHSS (ROC admission ) with regards to excellent 3-month BI (ROC percent 0.83, ROC delta 0.76, ROS admission 0.75), excellent 3-month mRS (ROC percent 0.83, ROC delta 0.74, ROS admission 0.78), and good 3-month mRS (ROC percent 0.83, ROC delta 0.76, ROS admission 0.78). Percentage change had a stronger association with 90-day infarct volume than delta NIHSS score and both delta NIHSS and percent change in NIHSS were more pronounced with faster treatment times.

Conclusion: In the NINDS rt-PA trial, alteplase was associated with a significant percent change improvement in NIHSS at 24 hours. Percent change in NIHSS may be a better surrogate marker of thrombolytic activity and 3-month outcomes.

Abstract WMP91: Head CT at Emergency Department Treat-and-Release Visit for Headache is Associated With Increased Risk of Subsequent Hospitalization for Cerebrovascular Disease

Background: Headache is a common presenting symptom among patients with misdiagnosed cerebrovascular (CVD) disease. Recent studies report that the occurrence of a head CT (HCT) during emergency department (ED) visit for non-specific neurological symptoms, such as vertigo, is associated with increased risk of stroke after ED discharge. We sought to evaluate whether the occurrence of HCT at ED visit for headache is associated with increased CVD risk.

Hypothesis: Patients with headache complaints who have a HCT performed in the ED (exposed) are more likely to have a CVD admission within 365 days of ED discharge than those who do not receive a HCT (unexposed).

Methods: We conducted a retrospective study of consecutive adult patients with headache complaints discharged to home (treat-and-release visit) from all four of the EDs affiliated with a single academic institution. Patients with headache complaints were defined as those with primary ICD-9/10-CM discharge diagnosis of benign headache; patients were included in our study at their first ED treat-and-release visit from 2013 to 2018. Subsequent CVD hospitalizations were identified using ICD-9/10-CM inpatient discharge codes and confirmed via chart review. We matched exposed to unexposed patients in a one-to-one fashion using propensity score methods. Standard descriptive statistics and relative risk (RR) with 95% CIs are reported.

Results: Among 28,121 patients with an ED treat-and-release visit for headache complaints, 45.6% (n=12,812) underwent HCT during ED visit. A total of 0.4% (n=112) patients had a subsequent CVD admission within 365 days of ED visit. Using propensity score matching, 80.3% (n=10,295) of patients with HCT (exposed) were able to be matched. In the matched sample, exposed patients had a nearly two-fold increased relative risk of CVD at 365 days (RR: 1.83; 95% CI: 1.12-3.01). In secondary analysis, CVD risk at 180 days was also higher in exposed (RR: 2.06; 95% CI: 1.13-3.74).

Conclusion: Having a HCT performed at index ED treat-and-release visit among those with headache complaints establishes a clinically meaningful risk gradient for subsequent CVD hospitalization. Some ED patients given a headache diagnosis may have been misdiagnosed at index visit.

Multiple Administrations of Intravenous Thrombolytic Therapy to a Stroke Mimic

BACKGROUND

Patients who present emergently with focal neurological deficits concerning for acute ischemic stroke can be extremely challenging to diagnose and treat. Unnecessary administration of thrombolytics to potential stroke patients whose symptoms are not caused by an acute ischemic stroke-stroke mimics-may result in patient harm, although the overall risk of hemorrhagic complications among stroke mimics is low.

CASE REPORT

We present a case of a stroke mimic patient with underlying psychiatric disease who was treated with intravenous alteplase on four separate occasions in four different emergency departments in the same city. Although he did not suffer hemorrhagic complications, this case highlights the importance of rapid exchange of health information across institutions to improve diagnostic quality and safety. WHY SHOULD AN EMERGENCY PHYSICIAN BE AWARE OF THIS?: Increased awareness of stroke mimics by emergency physicians may improve diagnostic safety for a subset of high-risk patients. Establishing rapid cross-institutional communication pathways that are integrated into provider's workflows to convey essential patient health information has potential to improve stroke diagnostic decision-making and thus represents an important topic for health systems research in emergency medicine.

Is the Cost-Effectiveness of Stroke Thrombolysis Affected by Proportion of Stroke Mimics?

Background and Purpose- Differentiating ischemic stroke patients from stroke mimics (SM), nonvascular conditions which simulate stroke, can be challenging in the acute setting. We sought to model the cost-effectiveness of treating suspected acute ischemic stroke patients before a definitive diagnosis could be made. We hypothesized that we would identify threshold proportions of SM among suspected stroke patients arriving to an emergency department above which administration of intravenous thrombolysis was no longer cost-effective. Methods- We constructed a decision-analytic model to examine various emergency department thrombolytic treatment scenarios. The main variables were proportion of SM to true stroke patients, time from symptom onset to treatment, and complication rates. Costs, reimbursement rates, and expected clinical outcomes of ischemic stroke and SM patients were estimated from published data. We report the 90-day incremental cost-effectiveness ratio of administering intravenous thrombolysis compared with no acute treatment from a healthcare sector perspective, as well as the cost-reimbursement ratio from a hospital-level perspective. Cost-effectiveness was defined as a willingness to pay <$100 000 USD per quality adjusted life year gained and high cost-reimbursement ratio was defined as >1.5. Results- There was an increase in incremental cost-effectiveness ratios as the proportion of SM cases increased in the 3-hour time window. The threshold proportion of SM above which the decision to administer thrombolysis was no longer cost-effective was 30%. The threshold proportion of SM above which the decision to administer thrombolysis resulted in high cost-reimbursement ratio was 75%. Results were similar for patients arriving within 0 to 90 minutes of symptom onset as compared with 91 to 180 minutes but were significantly affected by cost of alteplase in sensitivity analyses. Conclusions- We identified thresholds of SM above which thrombolysis was no longer cost-effective from 2 analytic perspectives. Hospitals should monitor SM rates and establish performance metrics to prevent rising acute stroke care costs and avoid potential patient harms.

Varicella Zoster Virus Vasculitis and Adult Cerebrovascular Disease

The role of Varicella zoster virus (VZV) in neurological illness, particularly cerebrovascular disease, has been increasingly recognized. Primary infection by VZV causes varicella (chickenpox), after which the virus remains latent in neuronal ganglia. Later, during aging or immunosuppression, the virus can reactivate causing zoster (shingles). Virus reactivation can also spread to cerebral arteries causing vasculitis and stroke. Zoster is a recognized risk factor for stroke, but stroke can occur without preceding zoster rash. The diagnosis of VZV cerebral vasculitis is established by abnormal brain imaging and confirmed by presence of viral DNA or anti-VZV antibodies in cerebrospinal fluid. Treatment with acyclovir with or without prednisone is usually recommended. VZV vasculitis is a unique and uncommon stroke mechanism that has been under recognized. Careful diagnostic investigation may be warranted in a subgroup of patients with ischemic stroke to detect VZV vasculitis and initiate appropriate therapy. In the following review, we detail the clinical presentation of VZV vasculitis, diagnostic challenges in VZV detection, and suggest the ways to enhance recognition and treatment of this uncommon disease.

Migraine as a Stroke Mimic and as a Stroke Chameleon

PURPOSE OF REVIEW

This review details the frequency of and ways in which migraine can be both an ischemic stroke/transient ischemic attack mimic (false positive) and chameleon (false negative). We additionally seek to clarify the complex relationships between migraine and cerebrovascular diseases with regard to diagnostic error.

RECENT FINDINGS

Nearly 2% of all patients evaluated emergently for possible stroke have an ultimate diagnosis of migraine; approximately 18% of all stroke mimic patients treated with intravenous thrombolysis have a final diagnosis of migraine. Though the treatment of a patient with migraine with thrombolytics confers a low risk of complication, symptomatic intracerebral hemorrhage may occur. Three clinical prediction scores with high sensitivity and specificity exist that can aid in the diagnosis of acute cerebral ischemia. Differentiating between migraine aura and transient ischemic attacks remains challenging. On the other hand, migraine is a common incorrect diagnosis initially given to patients with stroke. Among patients discharged from an emergency visit to home with a diagnosis of a non-specific headache disorder, 0.5% were misdiagnosed. Further development of tools to quantify and understand sources of stroke misdiagnosis among patients who present with headache is warranted. Both failure to identify cerebral ischemia among patients with headache and overdiagnosis of ischemia can lead to patient harms. While some tools exist to help with acute diagnostic decision-making, additional strategies to improve diagnostic safety among patients with migraine and/or cerebral ischemia are needed.

Abstract WP390: Misdiagnosis of Cervicocephalic Artery Dissection in the Emergency Department

Introduction: Cervicocephalic artery dissection is an important cause of stroke, particularly in the young. The initial presentation of cervicocephalic artery dissection can resemble benign neurological conditions resulting in delayed or incorrect diagnosis. We therefore sought to quantify the rate of possible cervicocephalic artery dissection misdiagnosis and to assess the impact of misdiagnosis on patient outcomes in a large heterogeneous cohort.

Methods: We performed a retrospective cohort study using administrative claims data from all ED visits and nonfederal hospitalizations in New York, California, and Florida from 2005-2015. Using previously validated ICD-9-CM codes, we identified patients hospitalized with a diagnosis of cervicocephalic artery dissection and no prior stroke diagnosis. Possible misdiagnosis of cervicocephalic artery dissection was defined as having a treat-and-release ED visit for headache, facial pain, neck pain, or Horner syndrome in the 14 days prior to a hospitalization for cervicocephalic artery dissection. Multivariable logistic regression was used to compare adverse clinical outcomes (stroke or death at the time of cervicocephalic artery dissection diagnosis) in patients with and without a possible misdiagnosis.

Results: Among 8,874 patients diagnosed with cervicocephalic artery dissection (mean age 52.7 years, 44.1% women), 300 (3.4%; 95% CI, 3.0-3.8%) had a possible ED misdiagnosis. Patients with a possible misdiagnosis of cervicocephalic artery dissection were younger and more often women than those not misdiagnosed. Stroke occurred in 128 (42.7%; 95% CI, 37.0-48.5%) patients with a possible misdiagnosis of dissection and in 3,908 (44.0%; 95% CI, 43.0-45.1%, P =0.63) patients without a possible misdiagnosis of cervicocephalic artery dissection. After adjustment for demographics and vascular risk factors, there were no differences in rates of hospital death (OR 0.48; 95% CI, 0.20-1.9) or stroke (OR 0.94; 95% CI, 0.74-1.2).

Conclusion: We found a low rate of possible misdiagnosis of cervicocephalic artery dissection in the ED. Possible misdiagnosis was not associated with stroke or death at the time of subsequent hospitalization for cervicocephalic artery dissection.

Clinical Decision-Making for Thrombolysis of Acute Minor Stroke Using Adaptive Conjoint Analysis

INTRODUCTION

There is practice variability in the treatment of patients with minor ischemic stroke with thrombolysis. We sought to determine which clinical factors physicians prioritize in thrombolysis decision-making for minor stroke using adaptive conjoint analysis.

METHODS

We conducted our conjoint analysis using the Potentially All Pairwise RanKings of all possible Alternatives methodology via the 1000Minds platform to design an online preference survey and circulated it to US physicians involved in stroke care. We evaluated 6 clinical attributes: language/speech deficits, motor deficits, other neurological deficits, history suggestive of increased risk of complication from thrombolysis, age, and premorbid disability. Survey participants were asked to choose between pairs of treatment scenarios with various clinical attributes; scenarios automatically adapted based on participants' prior responses. Preference weights representing the relative importance of each attribute were compared using unadjusted paired t tests. Statistical significance was set at α = .05.

RESULTS

Fifty-four participants completed the survey; 61% were vascular neurologists and 93% worked in academic centers. All neurological deficits were ranked higher than age, premorbid status, or potential contraindications to thrombolysis. Differences between each successive mean preference weight were significant: motor (31.7%, standard deviation [SD]: 9.5), language/speech (24.1%, SD: 9.6), other neurological deficits (16.6%, SD: 6.4), premorbid status (12.9%, SD: 6.6), age (10.1%, SD: 6.3), and potential thrombolysis contraindication (4.7%, SD: 4.4).

CONCLUSION

In a conjoint analysis, surveyed US physicians in academic practice assigned greater weight to motor and speech/language deficits than other neurological deficits, patient age, relative contraindications to thrombolysis, and premorbid disability when deciding to thrombolyse patients with minor stroke.

Misdiagnosis of Cerebral Vein Thrombosis in the Emergency Department

BACKGROUND AND PURPOSE

Rates of cerebral venous thrombosis (CVT) misdiagnosis in the emergency department and outcomes associated with misdiagnosis have been underexplored.

METHODS

Using administrative data, we identified adults with CVT at New York, California, and Florida hospitals from 2005 to 2013. Our primary outcome was probable misdiagnosis of CVT, defined as a treat-and-release emergency department visit for headache or seizure within 14 days before CVT. In addition, logistic regression was used to compare rates of clinical outcomes in patients with and without probable CVT misdiagnosis. We performed a confirmatory study at 2 tertiary care centers.

RESULTS

We identified 5966 patients with CVT in whom 216 (3.6%; 95% confidence interval [CI], 1.1%-4.1%) had a probable misdiagnosis of CVT. After adjusting for demographics, risk factors for CVT, and the Elixhauser comorbidity index, probable CVT misdiagnosis was not associated with in-hospital mortality (odds ratio, 0.14; 95% CI, 0.02-1.05), intracerebral hemorrhage (odds ratio, 0.97; 95% CI, 0.57-1.65), or unfavorable discharge disposition (odds ratio, 0.90; 95% CI, 0.61-1.32); a longer length of hospital stay was seen among misdiagnosed patients with CVT (odds ratio, 1.62; 95% CI, 1.04-2.50). In our confirmatory cohort, probable CVT misdiagnosis occurred in 8 of 134 patients with CVT (6.0%; 95% CI, 2.6%-11.4%).

CONCLUSIONS

In a large, heterogeneous multistate cohort, probable misdiagnosis of CVT occurred in 1 of 30 patients but was not associated with the adverse clinical outcomes included in our study.

Abstract WP294: Randomized Trial of an Attending Nurse Model of Care for Acute Stroke

Introduction: Poor medication adherence and insufficient stroke related knowledge may contribute to worse outcomes.

Hypothesis: A relationship-based attending nurse (AN) model of inpatient care for stroke patients will enhance medication adherence, stroke-related knowledge, and QOL after hospital discharge.

Methods: We performed a pseudo-randomized trial of AN + standard care vs. standard care at a single comprehensive stroke center (CSC). We enrolled patients with ischemic stroke, TIA, or ICH. The AN intervention consisted of a dedicated nurse focused on individual patient goals, expectations, and disease related knowledge deficits. After discharge, subjects were consented and assessed via structured telephone interviews using the Morisky Medication Adherence Scale (MMAS-4), Stroke Patient Education Retention tool (SPER), and Stroke Impact Scale (SIS).

Results: We randomly allocated 278 subjects to AN and 392 to standard models of care; 47% and 43% consented to participate, respectively. Patients in the AN group tended to have more severe strokes and lower baseline health literacy on a single Health Read question (Table). Overall, no significant differences were observed in the MMAS-4, SPER, or SIS. However, among those with low baseline health literacy, medication adherence was nearly doubled in the AN group, but no significant effect was observed on other outcomes.

Conclusion: Compared to standard nursing care alone, an attending nurse model did not improve medication adherence, stroke knowledge, or QOL in patients treated at a CSC. However, this intervention may hold promise and warrant further study in those with low baseline health literacy.

Abstract WP284: Modelling Hospital-Level Costs of Thrombolysis to Strokes and Stroke Mimics

Introduction: When the diagnosis of stroke is not completely certain, the benefits of rapid treatment with tPA are widely thought to outweigh the minimal risk of complication associated with thrombolysis of stroke mimics (SM). However, acceptable rates of hospital-level SM treatment have not been well established.

Hypothesis: A cost-reimbursement (CR) analysis will identify a threshold of SM treatment with tPA above which the CR ratio is unacceptably high.

Methods: We used stochastic modeling with probability distributions to mathematically examine various tPA treatment scenarios. The main hospital-level variables used were: (1) case-mix or rate of SM and (2) percent of all eligible cases treated with tPA. For each case, a hospital cost was assigned based on whether or not tPA was given, onset to treatment time (OTT), length of stay based on severity distribution and expected clinical outcome, and whether the case was a true stroke or SM. The costs associated with index hospitalization, treatment complications, and non-treatment of true stroke patients were estimated from published rates and data. Reimbursement for stroke with and without thrombolysis as well as treated SM were estimated from Medicare rates. We calculated a CR ratio and identified thresholds of hospital SM treatment rates above which CR ratios are >1.0 and >1.5.

Results: Assuming that 75% of true eligible stroke patients are treated with tPA with median OTT of 120 minutes, we found an increase in CR ratio from 0.97 (0% SM rate) to 1.61 (100% SM rate). The SM rate of <10% was associated with CR of <1; the SM rate of >10% was associated with CR ratio >1.0 while SM rate >80% was required for a CR ratio >1.5. For each OTT interval (0-90 minutes, 90-180 minutes, 180-270 minutes), the CR ratio was on average 0.1 lower for a hospital with a 5% SM rate compared to a hospital with a 20% SM rate.

Conclusions: In a simulation study, we found the CR ratio increased with increasing OTT and hospital SM treatment rate. As other researchers have suggested, hospitals may need to carefully monitor rates of SM patients treated with tPA and establish performance metrics to lower costs and reduce patient harm.

Abstract 139: Low Rate of Probable Misdiagnosis of Cerebral Vein Thrombosis in a Multi-State Cohort

Introduction: Cerebral vein thrombosis (CVT) is a cerebrovascular disease that affects young adults, is difficult to diagnose, and can result in significant disability and death. Studies aimed to evaluate rates of CVT misdiagnosis are limited.

Hypothesis: Misdiagnosis of CVT in the emergency department (ED) would be common and associated with adverse outcomes.

Methods: We performed a retrospective cohort study using ED visits and hospitalizations in California, New York and Florida from 2005 to 2013. We identified patients with CVT using previously validated International Classification of Diseases, Ninth Revision, Clinical Modification ( ICD-9-CM) codes. Primary outcome was a probable misdiagnosis of CVT, defined as an ED visit for benign headache or seizure within 14 days of CVT. In secondary analyses, we assessed whether a probable misdiagnosis of CVT was associated with an increased risk of intracerebral hemorrhage or inpatient mortality. Descriptive statistics was used to calculate crude rates and multivariable logistic regression to compare rates of adverse outcomes. A confirmatory cohort analysis of CVT patients from two tertiary care centers was conducted via detailed chart review.

Results: We identified 5,966 patients with CVT. Mean age was 44.2 ± 18.4 years and 71.7% were women. A total of 216 patients (3.6%; 95% CI, 3.1-4.1%) had a probable misdiagnosis of CVT. Misdiagnosed patients were younger and less likely to have congestive heart failure, atrial fibrillation, cancer, or trauma, but more likely to have a hypercoagulable state. No pregnant patients with CVT (n=758, 12.5%) had a probable misdiagnosis. In confirmatory cohorts, probable CVT misdiagnosis occurred in 8.1% (6 of 74) and 2.6% (2 of 76) of patients. Misdiagnosis of CVT was not associated with an increased risk of death (OR, 0.1; 95% CI, 0.0-1.1), intracerebral hemorrhage (OR, 1.0; 95% CI, 0.6-1.6), unfavorable discharge disposition (OR, 0.9; 95% CI, 0.6-1.3), but was associated with an increased length of stay (OR 1.7; 95% CI, 1.2-2.3).

Conclusion: In a large, heterogeneous group of patients, a probable misdiagnosis of CVT occurred in 1 out of 30 patients. No misdiagnosis occurred in a pregnant patient and misdiagnosis was not associated with adverse clinical outcomes.

Symptom-Disease Pair Analysis of Diagnostic Error (SPADE): a conceptual framework and methodological approach for unearthing misdiagnosis-related harms using big data

BACKGROUND

The public health burden associated with diagnostic errors is likely enormous, with some estimates suggesting millions of individuals are harmed each year in the USA, and presumably many more worldwide. According to the US National Academy of Medicine, improving diagnosis in healthcare is now considered 'a moral, professional, and public health imperative.' Unfortunately, well-established, valid and readily available operational measures of diagnostic performance and misdiagnosis-related harms are lacking, hampering progress. Existing methods often rely on judging errors through labour-intensive human reviews of medical records that are constrained by poor clinical documentation, low reliability and hindsight bias.

METHODS

Key gaps in operational measurement might be filled via thoughtful statistical analysis of existing large clinical, billing, administrative claims or similar data sets. In this manuscript, we describe a method to quantify and monitor diagnostic errors using an approach we call 'Symptom-Disease Pair Analysis of Diagnostic Error' (SPADE).

RESULTS

We first offer a conceptual framework for establishing valid symptom-disease pairs illustrated using the well-known diagnostic error dyad of dizziness-stroke. We then describe analytical methods for both look-back (case-control) and look-forward (cohort) measures of diagnostic error and misdiagnosis-related harms using 'big data'. After discussing the strengths and limitations of the SPADE approach by comparing it to other strategies for detecting diagnostic errors, we identify the sources of validity and reliability that undergird our approach.

CONCLUSION

SPADE-derived metrics could eventually be used for operational diagnostic performance dashboards and national benchmarking. This approach has the potential to transform diagnostic quality and safety across a broad range of clinical problems and settings.

Abstract WP271: Incidence of Diagnostic Error in Cerebral Vein Thrombosis at a Single Urban Center

Introduction: An estimated 150,000 cerebrovascular events are misdiagnosed in US emergency departments each year. Cerebral vein thrombosis (CVT) is a rare cerebrovascular disease caused by a thrombus formed in the cerebral sinuses or veins that can result in intracerebral hemorrhage and ischemia. Early treatment of CVT may improve patient outcomes, but diagnostic errors leading to delayed or missed CVT diagnosis have received limited attention.

Hypothesis: CVT misdiagnosis rate in emergency department (ED) and outpatient settings would be high and associated with adverse outcome.

Methods: We performed a retrospective cohort study at Montefiore Medical Center using electronic medical records. We identified all CVT patients hospitalized between 9/1/2005 to 9/1/2015 using validated International Classification of Diseases, Ninth Revision codes and detailed chart review. Misdiagnosis of CVT was defined as any instance when a patient sought medical attention for headache or seizure within 30 days of initial CVT diagnosis in any care setting that resulted in discharge to home. We tested whether patients’ demographic and clinical features were associated with CVT misdiagnosis. Outcomes at hospital discharge were reported. Categorical variables were compared using chi-squared and continuous variables using student t-test. P-value of <0.05 was considered statistically significance.

Results: We identified 74 CVT patients. Mean age was 46 years (SD: 18) and 44 were female (67%). A total of 15 patients (20%) had a CVT misdiagnosis; 12 initially presented to the ED and 3 to outpatient clinics. Patients with CVT misdiagnosis were younger (35 vs. 48 years of age, P=0.02), and more often Spanish speaking (n=3 vs. n=0, P = 0.02). There was no difference in clinical characteristics, radiographic features, or outcomes between misdiagnosed and accurately diagnosed CVT patients.

Conclusion: Misdiagnosis of CVT occurred in a fifth of CVT patients and was more common in younger and non-English speaking patients. Interventions to improve diagnostic accuracy among patients with CVT should be explored, particularly in high-risk patient subgroups.

Abstract TP246: Evaluating Thrombolysis Decision Making in Minor Stroke Using Adaptive Discrete Choice Experimentation

Introduction: Historically, physicians have been less likely to give intravenous tPA to patients with minor symptoms but evidence that mild stroke carries significant risk of permanent disability may be influencing practice.

Hypothesis: Important factors in minor stroke thrombolytic treatment decision making will be type of neurological deficit, patient age, and risk of complication.

Methods: We circulated an online survey to physicians treating acute stroke patients. We used the potentially all pairwise rankings of all possible alternatives (PAPRIKA) methodology via the 1,000Minds platform to calculate point values for six clinical criteria: communication, motor, and sensory deficits from index stroke as well as relevant history (non-ST-segment myocardial infarction within 3 months and genitourinary bleeding within 21 days), age, and premorbid disability. Survey participants were asked to choose between multiple pairs of clinical attributes in two thrombolysis treatment scenarios (e.g., 92 year old with mild hemiparesis versus 75 year old with mild aphasia) that adapted based on their prior responses. We report participant characteristics and mean part-worth utilities representing the relative importance of each criterion.

Results: Fifty-three participants completed our survey. Most participants were male (32/53), in practice >5 years (37/53), and were vascular neurologists (37/53). Overall (Figure), motor deficits were weighted most heavily (31.8%, SD 9.5) followed by communication (24.1%, SD 9.6) and sensory (16.6%, SD 6.4) deficits. Relevant history was the least heavily weighted of all criteria (4.5%, SD 4.2) while age (10.2%, SD 6.3) was assigned a similar weight as premorbid status (12.8%, SD 6.7).

Conclusions: In thrombolysis decision-making for patients with minor stroke, participants assigned more weight to patients’ motor and communication deficits than sensory deficits; age, premorbid status, and history did not play a major role.

Stroke Chameleons and Stroke Mimics in the Emergency Department

PURPOSE OF REVIEW

We discuss the frequency of stroke misdiagnosis in the emergency department (ED), identify common diagnostic pitfalls, describe strategies to reduce diagnostic error, and detail ongoing research.

RECENT FINDINGS

The National Academy of Medicine has re-defined and highlighted the importance of diagnostic errors for patient safety. Recent rates of stroke under-diagnosis (false-negative cases, "stroke chameleons") range from 2-26% and 30-43% for stroke over-diagnosis (false-positive cases, "stroke mimics"). Failure to diagnosis stroke can preclude time-sensitive treatments and has been associated with poor outcomes. Strategies have been developed to improve detection of posterior circulation stroke syndromes, but ongoing work is needed to reduce under-diagnosis in other atypical stroke presentations. The published rates of harm associated with stroke over-diagnosis, particularly thrombolysis of stroke mimics, remain low. Additional strategies to improve the accuracy of stroke diagnosis should focus on rapid clinical reasoning in the time-sensitive setting of acute ischemic stroke and identifying imperfections in the healthcare system which may contribute to diagnostic error.

Door-to-Needle Delays in Minor Stroke: A Causal Inference Approach

BACKGROUND AND PURPOSE

Thrombolysis rates among minor stroke (MS) patients are increasing because of increased recognition of disability in this group and guideline changes regarding treatment indications. We examined the association of delays in door-to-needle (DTN) time with stroke severity.

METHODS

We performed a retrospective analysis of all stroke patients who received intravenous tissue-type plasminogen activator in our emergency department between July 1, 2011, and February 29, 2016. Baseline characteristics and DTN were compared between MS (National Institutes of Health Stroke Scale score ≤5) and nonminor strokes (National Institutes of Health Stroke Scale score >5). We applied causal inference methodology to estimate the magnitude and mechanisms of the causal effect of stroke severity on DTN.

RESULTS

Of 315 patients, 133 patients (42.2%) had National Institutes of Health Stroke Scale score ≤5. Median DTN was longer in MS than nonminor strokes (58 versus 53 minutes; P=0.01); fewer MS patients had DTN ≤45 minutes (19.5% versus 32.4%; P=0.01). MS patients were less likely to use emergency medical services (EMS; 62.6% versus 89.6%, P<0.01) and to receive EMS prenotification (43.9% versus 72.4%; P<0.01). Causal analyses estimated MS increased average DTN by 6 minutes, partly through mode of arrival. EMS prenotification decreased average DTN by 10 minutes in MS patients.

CONCLUSIONS

MS had longer DTN times, an effect partly explained by patterns of EMS prenotification. Interventions to improve EMS recognition of MS may accelerate care.

A resident boot camp for reducing door-to-needle times at academic medical centers

BACKGROUND

We sought to determine if a structured educational program for neurology residents can lower door-to-needle (DTN) times at an academic institution.

METHODS

A neurology resident educational stroke boot camp was developed and implemented in April 2013. Using a prospective database of 170 consecutive acute ischemic stroke (AIS) patients treated with IV tissue plasminogen activator (tPA) in our emergency department (ED), we evaluated the effect of the intervention on DTN times. We compared DTN times and other process measures preintervention and postintervention. p Values < 0.05 were considered significant.

RESULTS

The proportion of AIS patients treated with tPA within 60 minutes of arrival to our ED tripled from 18.1% preintervention to 61.2% postintervention (p < 0.001) with concomitant reduction in DTN time (median 79 minutes vs 58 minutes, p < 0.001). The resident-delegated task (stroke code to tPA) was reduced (75 minutes vs 44 minutes, p < 0.001), while there was no difference in ED-delegated tasks (door to stroke code [7 minutes vs 6 minutes, p = 0.631], door to CT [18 minutes in both groups, p = 0.547]). There was an increase in stroke mimics treated (6.9% vs 18.4%, p = 0.031), which did not lead to an increase in adverse outcomes.

CONCLUSIONS

DTN times were reduced after the implementation of a stroke boot camp and were driven primarily by efficient resident stroke code management. Educational programs should be developed for health care providers involved in acute stroke patient care to improve rapid access to IV tPA at academic institutions.

Cardiac magnetic resonance imaging has limited additional yield in cryptogenic stroke evaluation after transesophageal echocardiography

Background

The use of cardiac magnetic resonance imaging is increasing, but its role in the diagnostic work-up following ischemic stroke has received limited study. We aimed to explore the added yield of cardiac magnetic resonance imaging to identify cardio-aortic sources not detected by transesophageal echocardiography among patients with cryptogenic stroke.

Methods

A retrospective single-center cohort study was performed from 01 January 2009 to 01 March 2013. Consecutive patients who had both a stroke protocol cardiac magnetic resonance imaging and a transesophageal echocardiography preformed during a single hospitalization were included. All cardiac magnetic resonance imaging studies underwent independent, blinded review by two investigators. We applied the causative classification system for ischemic stroke to all patients, first blinded to cardiac magnetic resonance imaging results; we then reapplied the causative classification system using cardiac magnetic resonance imaging. Standard statistical tests to evaluate stroke subtype reclassification rates were used.

Results

Ninety-three patients were included in the final analysis; 68.8% were classified as cryptogenic stroke after initial diagnostic evaluation. Among patients with cryptogenic stroke, five (7.8%) were reclassified due to cardiac magnetic resonance imaging findings: one was reclassified as “cardio-aortic embolism evident” due to the presence of a patent foramen ovale and focal cardiac infarct and four were reclassified as “cardio-aortic embolism possible” due to mitral valve thickening (n = 1) or hypertensive cardiomyopathy (n = 3). Overall, findings on cardiac magnetic resonance imaging reduced the percentage of patients with cryptogenic stroke by slightly more than 1%.

Conclusion

Our stroke subtype reclassification rate after the addition of cardiac magnetic resonance imaging results to a diagnostic work-up which includes transesophageal echocardiography was very low. Prospective studies evaluating the role of cardiac magnetic resonance imaging and transesophageal echocardiography among patients with cryptogenic stroke should be considered.

Risk of Pulmonary Embolism After Cerebral Venous Thrombosis

BACKGROUND AND PURPOSE

Cerebral vein thrombosis (CVT) is a type of venous thromboembolism. Whether the risk of pulmonary embolism (PE) after CVT is similar to the risk after deep venous thrombosis (DVT) is unknown.

METHODS

We performed a retrospective cohort study using administrative data from all emergency department visits and hospitalizations in California, New York, and Florida from 2005 to 2013. We identified patients with CVT or DVT and the outcome of PE using previously validated International Classification of Diseases, Ninth Revision, Clinical Modification codes. Kaplan-Meier survival statistics and Cox proportional hazards models were used to compare the risk of PE after CVT versus PE after DVT.

RESULTS

We identified 4754 patients with CVT and 241 276 with DVT. During a mean follow-up of 3.4 (±2.4) years, 138 patients with CVT and 23 063 with DVT developed PE. CVT patients were younger, more often female, and had fewer risk factors for thromboembolism than patients with DVT. During the index hospitalization, the rate of PE was 1.4% (95% confidence interval [CI], 1.1%-1.8%) in patients with CVT and 6.6% (95% CI, 6.5%-6.7%) in patients with DVT. By 5 years, the cumulative rate of PE after CVT was 3.4% (95% CI, 2.9%-4.0%) compared with 10.9% (95% CI, 10.8%-11.0%; P<0.001) after DVT. CVT was associated with a lower adjusted hazard of PE than DVT (hazard ratio, 0.26; 95% CI, 0.22-0.31).

CONCLUSION

The risk of PE after CVT was significantly lower than the risk after DVT. Among patients with CVT, the greatest risk for PE was during the index hospitalization.

Abstract WP179: Rate of Pulmonary Embolism After Cerebral Venous Thrombosis

Introduction: Cerebral vein thrombosis (CVT) is associated with an increased risk of subsequent venous thromboembolism. It is unknown whether the risk of pulmonary embolism (PE) after CVT is similar to that of PE after deep venous thrombosis (DVT).

Methods: We performed a retrospective cohort study using administrative claims data from all emergency department visits and hospitalizations in California from 2005-2011, New York from 2006-2013, and Florida from 2005-2013. We identified patients with CVT or DVT as well as the primary outcome of PE using previously validated International Classification of Diseases, Ninth Revision, Clinical Modification ( ICD-9-CM ) codes. In order to minimize misclassification error, patients with both CVT and DVT during the same index hospitalization were excluded and patients with CVT were censored at the time of development of DVT and vice versa. Kaplan-Meier survival statistics and Cox proportional hazards models were used to compare the risk of PE after CVT versus after DVT while adjusting for demographics, vascular risk factors, and the Elixhauser comorbidity index.

Results: We identified 4,450 patients with CVT and 217,589 patients with DVT. During a mean follow-up of 2.0 (±1.7) years, 124 patients with DVT developed a PE and 18,698 patients with DVT developed a PE. Patients with CVT were younger (mean age 45 vs 63), more often female (71% vs 52%), more often pregnant, and had fewer vascular risk factors than patients with DVT. During the index hospitalization, the rate of PE was 1.5% (95% confidence interval [CI], 1.1-1.8%) in patients with CVT and 6.2% (95% CI, 6.1-6.3%, p<0.001) in patients with DVT. By 5 years, the cumulative rate of PE after CVT was 3.7% (95% CI, 3.0-4.4%) compared to 10.5% (95% CI, 10.3-10.6%, p<0.001) after DVT. After adjustment for demographics and comorbidities, CVT was associated with a significantly lower hazard of PE when compared to DVT (hazard ratio, 0.31; 95% CI, 0.26-0.38).

Conclusion: In a large, heterogeneous population, we found that the risk of PE after CVT was significantly lower than that of PE after DVT. Among patients with CVT, the greatest risk for PE was apparent during the index hospitalization.

Symptomatic Carotid Occlusion Is Frequently Associated With Microembolization

BACKGROUND AND PURPOSE

Symptomatic carotid artery disease is associated with significant morbidity and mortality. The pathophysiologic mechanisms of cerebral ischemia among patients with carotid occlusion remain underexplored.

METHODS

We conducted a prospective observational cohort study of patients hospitalized within 7 days of ischemic stroke or transient ischemic attack because of ≥50% carotid artery stenosis or occlusion. Transcranial Doppler emboli detection was performed in the middle cerebral artery ipsilateral to the symptomatic carotid. We describe the prevalence of microembolic signals (MES), characterize infarct topography, and report clinical outcomes at 90 days.

RESULTS

Forty-seven patients, 19 with carotid occlusion and 28 with carotid stenosis, had complete transcranial Doppler recordings and were included in the final analysis. MES were present in 38%. There was no difference in MES between those with carotid occlusion (7/19, 37%) compared with stenosis (11/28, 39%; P=0.87). In patients with radiographic evidence of infarction (n=39), 38% had a watershed pattern of infarction, 41% had a nonwatershed pattern, and 21% had a combination. MES were present in 40% of patients with a watershed pattern of infarction. Recurrent cerebral ischemia occurred in 9 patients (19%; 6 with transient ischemic attack, 3 with ischemic stroke). There was no difference in the rate of recurrence in those with compared to those without MES.

CONCLUSIONS

Cerebral embolization plays an important role in the pathophysiology of ischemia in both carotid occlusion and stenosis, even among patients with watershed infarcts. The role of aggressive antithrombotic and antiplatelet therapy for symptomatic carotid occlusions may warrant further investigation given our findings.

E-Mail Is an Effective Tool for Rapid Feedback in Acute Stroke

OBJECTIVE

To determine whether e-mail is a useful mechanism to provide prompt, case-specific data feedback and improve door-to-needle (DTN) time for acute ischemic stroke treated with intravenous tissue plasminogen activator (IV-tPA) in the emergency department (ED) at a high-volume academic stroke center.

METHODS

We instituted a quality improvement project at Columbia University Medical Center where clinical details are shared via e-mail with the entire treatment team after every case of IV-tPA administration in the ED. Door-to-needle and component times were compared between the prefeedback (January 2013 to March 2015) and postfeedback intervention (April 2015 to June 2016) periods.

RESULTS

A total of 273 cases were included in this analysis, 102 (37%) in the postintervention period. Median door-to-stroke code activation (2 vs 0 minutes, P < .01), door-to-CT Scan (21 vs 18 minutes, P < .01), and DTN (54 vs 49 minutes, P = .17) times were shorter in the postintervention period, although the latter did not reach statistical significance. The proportion of cases with the fastest DTN (≤45 minutes) was higher in the postintervention period (29.2% vs 42.2%, P = .03).

CONCLUSION

E-mail is a simple and effective tool to provide rapid feedback and promote interdisciplinary communication to improve acute stroke care in the ED.

Cardiac magnetic resonance imaging: a new tool to identify cardioaortic sources in ischaemic stroke

Stroke of undetermined aetiology or 'cryptogenic' stroke accounts for 30-40% of ischaemic strokes despite extensive diagnostic evaluation. The role and yield of cardiac imaging is controversial. Cardiac MRI (CMR) has been used for cardiac disorders, but its use in cryptogenic stroke is not well established. We reviewed the literature (randomised trials, exploratory comparative studies and case series) on the use of CMR in the diagnostic evaluation of patients with ischaemic stroke. The literature on the use of CMR in the diagnostic evaluation of ischaemic stroke is sparse. However, studies have demonstrated a potential role for CMR in the diagnostic evaluation of patients with cryptogenic stroke to identify potential aetiologies such as cardiac thrombi, cardiac tumours, aortic arch disease and other rare cardiac anomalies. CMR can also provide data on certain functional and structural parameters of the left atrium and the left atrial appendage which have been shown to be associated with ischaemic stroke risk. CMR is a non-invasive modality that can help identify potential mechanisms in cryptogenic stroke and patients who may be targeted for enrolment into clinical trials comparing anticoagulation to antiplatelet therapy in secondary stroke prevention. Prospective studies are needed to compare the value of CMR as compared to transthoracic and transesophageal echocardiography in the diagnostic evaluation of cryptogenic stroke.

International Classification of Diseases, Ninth Revision (ICD-9) Diagnosis Codes Can Identify Cerebral Venous Thrombosis in Hospitalized Adults

BACKGROUND

Cerebral venous thrombosis (CVT) is a relatively rare and understudied disease. We sought to determine the accuracy of International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes to identify CVT.

METHODS

Retrospective chart review using the electronic medical record (EMR) to identify all patients discharged with CVT following admission or emergency department visit from May 1, 2010 to May 1, 2015 at our center.

RESULTS

We identified 111 patients with an ICD-9 discharge diagnosis code of 325.0 (cerebral sinovenous thrombosis, excluding nonpyogenic cases and cases associated with pregnancy and the puerperium), 437.6 (CVT of nonpyogenic origin), or 671.5 (CVT complicating pregnancy, childbirth, or the puerperium) in any position. Of these 111 patients, 84 (75.7%) had confirmed CVT after EMR review. Searching outpatient and radiology records, we found an additional 24 patients with CVT who were not identified via query of ICD-9 discharge diagnosis codes. The ICD-9 codes 325.0, 437.6, or 671.5 in any position had a combined sensitivity of 77.8% and specificity of 92.7%; in the primary position, they had a sensitivity of 28.7% and specificity of 98.3%.

CONCLUSION

The ICD-9 codes 325.0, 437.6, and 671.5 can be used to identify CVT with acceptable sensitivity and specificity.

Abstract 123: Door To Needle 60, Creative Solutions in Academia

Introduction: National guidelines endorse that eligible acute ischemic stroke (AIS) patients should be treated with intravenous tissue plasminogen activator (IV tPA) within 60 minutes of arrival to an emergency department (ED). We participated in the American Heart Association’s Target: Stroke program which successfully reduced door to needle (DTN) times through 10 best practices, but academic hospitals face a unique challenge as junior residents evaluate and manage AIS patients. We hypothesized that a “stroke boot camp” could improve resident efficiency during stroke codes and shorten DTN times through faster stroke code to tPA times.

Methods: A neurology resident educational protocol was developed and implemented in April 2013 using a Socratic case-based discussion to emphasize focused history and exam, medication history, and tPA exclusion criteria. We distributed cards with IV tPA risks/benefits and a checklist for tPA exclusion criteria. We compared pre-intervention (January 2010-April 2013) to post-intervention (April 2013-April 2014) patient demographics, comorbidities, resident level, relevant times, and outcomes using appropriate tests.

Results: We analyzed 122 consecutive AIS patients treated with IV tPA in our ED during the study period. Pre and post intervention groups did not differ by demographics except gender (p = 0.005). There were no difference in comorbidities, baseline NIHSS, or resident post graduate year (PGY). After the intervention, stroke-code-to-tPA was significantly reduced (75 min vs. 45 min; p < 0.001), whereas door-to-stroke-code (7 min vs. 6 min, p = 0.56) and door-to-CT (18 min vs. 19 min, p = 0.44) did not change. The proportion of patients treated within 60 minutes increased (16.4% vs. 51.4%, P < 0.001) and median DTN time decreased (81 min vs. 60 min P < 0.001) significantly after the intervention. Time reductions were consistent across PGY levels without increased adverse outcomes.

Conclusion: Reduction in stroke code-to-tPA times after implementation of a “stroke boot camp” led to a significant reduction in DTN time. Focused neurology resident acute stroke education should be implemented at academic institutions to improve rapid IV tPA administration