Klaas, James P.

Background Computed tomography (CT) angiography collateral score (CTA-CS) is an important clinical outcome predictor following mechanical thrombectomy for ischemic stroke with large vessel occlusion (LVO). The present multireader study aimed to evaluate the performance of e-CTA software for automated assistance in CTA-CS scoring. Materials and Methods Brain CTA images of 56 patients with anterior LVO were retrospectively processed. Twelve readers of various clinical training, including junior neuroradiologists, senior neuroradiologists, and neurologists graded collateral flow using visual CTA-CS scale in two sessions separated by a washout period. Reference standard was the consensus of three expert readers. Duration of reading time, inter-rater reliability, and statistical comparison of readers’ performance metrics were analyzed between the e-CTA assisted and unassisted sessions. Results e-CTA assistance resulted in significant increase in mean accuracy (58.6% to 67.5%, p = 0.003), mean F1 score (0.574 to 0.676, p = 0.002), mean precision (58.8% to 68%, p = 0.007), and mean recall (58.7% to 69.9%, p = 0.002), especially with slight filling deficit (CTA-CS 2 and 3). Mean reading time was reduced across all readers (103.4 to 59.7 s, p = 0.001), and inter-rater agreement in CTA-CS assessment was increased (Krippendorff's alpha 0.366 to 0.676). Optimized occlusion laterality detection was also noted with mean accuracy (92.9% to 96.8%, p = 0.009). Conclusion Automated assistance for CTA-CS using e-CTA software provided helpful decision support for readers in terms of improving scoring accuracy and reading efficiency for physicians with a range of experience and training backgrounds and leading to significant improvements in inter-rater agreement.

Background: Seizures are a known complication of ischemic stroke. Some studies have reported a higher risk of post-stroke seizure with cardioembolic strokes, whereas others have not. The risk of post-stroke seizure for different stroke mechanisms is unclear. Objective: To compare different stroke etiologies and the risk of post-stroke seizure in a population-based study. Methods: Utilizing the Rochester Epidemiology Project medical records linkage system, we ascertained all first ischemic strokes occurring among Rochester, MN residents from 1990-1994, and followed the patients in the comprehensive medical record through March 2014 for occurrence of seizure. TOAST stroke subtype classification was determined by a consensus of cerebrovascular neurologists. New post-stroke seizure was identified by physician diagnosis and/or EEG report. Patients with a seizure before the index stroke were not included. Pre-morbid stroke risk factors were recorded. Associations between seizure after index ischemic stroke as outcome and other variables of interest were assessed using univariate logistic regression. Results: A total of 490 patients met inclusion criteria, with median follow-up of 6.5 years. Out of 490 patients, 35 had new onset seizures (7.1%). Median age was significantly lower at index stroke for patients with seizures (73 v. 77 years, p=0.017). The most common TOAST stroke subtype was cryptogenic (46.8%) followed by cardioembolic (24.9%). There was no significant difference in TOAST stroke subtypes between patients with post-stroke seizure and those without (p=0.35). There was also no difference between the two groups when comparing combined cardioembolic and large-vessel to small-vessel (p=0.28) or cryptogenic to small-vessel (p=0.9) mechanisms. There was also no difference in pre-morbid stroke risk factors between the two groups, including hypertension (p=0.22), total cholesterol (p=0.95), type 2 diabetes mellitus (p=0.13), tobacco use (p=0.27), atrial fibrillation (p=0.9), or previous myocardial infarction (p=0.67). Conclusion: This study suggests that the long-term risk of post-stroke seizures does not differ by ischemic stroke etiology.

Background: Coronary artery disease (CAD) is a significant cause of morbidity and mortality in ischemic stroke (IS) patients. IS and CAD often coexist and share common risk factors. Absolute risk and predictors of myocardial infarction (MI) in IS and by IS etiology subtype is not well known. We determined the risk of MI and its predictors in patients with incident IS in a population based cohort. Methods: Using the Rochester Epidemiology Project (REP) medical records linkage system we identified all residents with incident IS in a 5-year period between January 1, 1990, and December 31, 1994. Comprehensive medical records were reviewed and variables of interest were retrieved and analyzed. Univariate Cox proportional hazard regression models were constructed to assess hazard ratios and 95% CI for predictors of MI occurrence after IS. Results: We identified 490 persons with IS, mean age at IS was 76.6 (SD 12.9) years, mean follow up to death or final follow-up was 6.5 (SD 6.3, range 0.0-23.8) years. There were 126 recurrent strokes and 93 MIs during followup. The 1-year actuarial risk of MI was 3.6% (95% CI, 2.2% - 5.9%), and 5-year actuarial risk of MI was 11.4% (95% CI, 8.4% - 15.3%). Univariate predictors of MI included (HR; 95% CI): age (1.02;1.0-1.04), pre-IS history of MI (1.37;1.13-1.65), CHF (1.85;1.06-3.23), CABG (2.11;1.19-3.75), left ventricular dyskinesia (2.14;1.32-3.4), peripheral vascular disease (1.32;1.18-1.49), diabetes mellitus (1.54;1.21-1.96), elevated creatinine (1.93;1.44-2.59), LDL level (1.01;1.00-1.01). Conclusion: MI is common after first IS. Several factors predicted an increased risk of MI after stroke, and should be taken into account in determining CAD screening strategies at the time of IS. Recurrent stroke was more common than MI during long-term followup.

Introduction: Studies have suggested a high incidence of dementia after stroke; however data regarding risk factors are conflicting and few studies are population-based. The aims of this population-based study are to determine the frequency of diagnosis and predictors of clinical dementia after ischemic stroke. Methods: The Rochester Epidemiology Project medical records linkage system was used to ascertain all first ischemic strokes (IS) occurring among Rochester, MN residents between 1990 -1994. Comprehensive medical records were reviewed. Patient and stroke characteristics, medical comorbidities, and the development of clinical dementia any time after IS were recorded. Dementia was defined as any reflection in the chart by a clinician of memory loss resulting in functional impairment. Patients with a pre-stroke diagnosis of dementia were excluded. Associations between new dementia after the index IS and other variables of interest were analyzed using univariate logistic regression. Results: First IS was detected in 490 people, 82 of whom had a prior history of dementia, and 119 (29%) of whom developed dementia during a mean follow-up of 6.5 years (range 0-23.8). The mean age at index stroke was 76.6 years (SD 12.9). Those who developed dementia were a mean 4 years older at the time of IS (p = 0.003). The odds of developing dementia were 1.7 higher in those who had at least one recurrent stroke. There was no effect of sex, stroke location (cortical, subcortical, brainstem), stroke mechanism, or premorbid stroke risk factors (coronary artery disease, atrial fibrillation/flutter, hypertension, diabetes mellitus, left-sided valvular disease) on the occurrence of dementia. Conclusions: Dementia is commonly diagnosed following ischemic stroke. While stroke risk factors did not impact the risk of diagnosis of dementia, recurrent stroke increased this risk. Stroke risk factors may indirectly increase the risk of dementia by increasing the risk of recurrent IS. This suggests that aggressive secondary stroke prevention measures may significantly decrease the risk of being diagnosed with dementia following IS.

Background: Current guidelines recommend that initial management of patients with an intracerebral hemorrhage (ICH) occur in a neuroscience intensive care unit (ICU). However, some patients can be safely managed without ICU care. We evaluated factors that may predict the need for ICU level of care. Methods: We reviewed the medical records of all adult patients with supratentorial ICH presenting within 72 hours of symptom onset seen from 2009-2012. Patients admitted for palliative care or ICH due to a known structural lesion were excluded. Criteria defining need for ICU level of care were: need for intubation/mechanical ventilation, hyperosmolar therapy, neurosurgical procedure, prolonged ICU admission, or significant clinical deterioration. A univariate and multivariable logistic regression analysis was performed on patient demographics, medical comorbidities, time to presentation, vital signs, GCS/FOUR score, medications, laboratory studies, and imaging to identify independent predictors of the need for ICU admission. Results: 229 patients met inclusion criteria. Based on our triage criteria, 128 patients (56%) required ICU level of care. Multivariable analysis identified five factors that best predicted the need for ICU admission: younger age, shorter time to presentation, lower FOUR score, larger hemorrhage volume, and use of anticoagulants at presentation (table). Intraventricular hemorrhage and GCS score were not significant predictors on the multivariable analysis. Conclusion: Age, time to presentation, FOUR score, hemorrhage volume, and use of anticoagulants predict the need for ICU care after ICH. These variables will be used to develop a formula to aid clinicians in appropriate initial triage of patients with ICH.

Introduction: Digital subtraction angiography (DSA) is currently the gold standard for evaluation of moyamoya disease (MMD) and is the basis for several grading systems. Non-invasive modalities such as MRA are also used in assessment of MMD. Accurate implementation of these staging systems for use with MRA would imply a less-invasive method of MMD monitoring. Methods: This was a single-center retrospective study that compared MRA to reference DSA studies in determining MMD severity using Berlin and modified Suzuki (mSG) grading systems A single reviewer assigned modified Suzuki and Berlin grades to all included hemispheres on both MRA and DSA. The mean difference between MRA and DSA grades for each DSA reference grade was determined utilizing both grading systems. The distribution of scores was also compared between MRA and DSA using both grading systems. Results: A total of 166 cerebral hemispheres in 90 patients were included: The mean age at MMD diagnosis was 34.8 years (standard deviation [SD] = 14.2), and fifty-three patients (63.9%) were male. Mean MRA grades were significantly lower than DSA reference grades 3 and 4 utilizing the mSG system (p = <0.0001 in both cases); A similar pattern was observed for Berlin grades 2 and 3 (p = <0.0001, Table 1 ). The distribution of modified Suzuki grades on MRA was skewed toward relatively lower grades as compared to DSA (p = <0.0001): This pattern was also true for Berlin grades as well (p = <0.0001, Table 2 ). Conclusions: DSA is essential in the evaluation of MMD, as MRA is shown to underestimate disease stage, with a greater degree of underestimation in more severe disease. We suggest DSA be performed in order to aid in appropriate treatment planning.

Background Headache in patients with moyamoya disease is an under-addressed topic in the medical literature. Delay in the diagnosis of moyamoya disease or inappropriate treatment of headache could lead to devastating cerebrovascular outcome. With the evolving understanding of moyamoya disease, migraine pathophysiology, and various migraine-specific medications that have become available, it is crucial to provide an updated overview on this topic. Methods We searched PubMed for keywords including moyamoya disease, moyamoya syndrome, headache in moyamoya, surgical revascularization, surgical bypass, migraine and moyamoya, and calcitonin gene-related peptide (CGRP). We summarized the literature and provide a comprehensive review of the headache presentation, possible mechanisms, the impact of various surgical revascularizations on headache in patients with moyamoya disease, and the medical management of headache incorporating novel migraine-specific treatments. Results and conclusion: The most common headache phenotype is migraine; tension-type headache, hemiplegic migraine, and cluster headache have also been reported. Most patients experience improvement of headache after surgical revascularization, though some patients report worsening, or new-onset headache after surgery. Given the complexity of moyamoya disease, careful consideration of different types of medical therapy for headache is necessary to improve the quality of life while not increasing the risk of adverse cerebrovascular events. More prospective studies are warranted to better understand and manage headache in patients with moyamoya disease.