Search

Introduction:There is paucity of evidence on long-term nationwide trends and seasonal variation in incidence of Intracerebral Hemorrhage (ICH) and Acute Ischemic Stroke (AIS).Methods:Using the Nationwide Inpatient Sample (2004 - 2019), we identified adults (≥ 18 years) discharged with primary diagnosis of ICH and AIS. We aggregated hospitalizations across quarters: Jan - Mar (Q1); Apr - Jun (Q2); Jul - Sep (Q3) and Oct - Dec (Q4). We used survey weights to calculate nationally representative estimates of quarter-specific incidence across age, sex, and race/ethnicity sub-cohorts and obtained time-specific denominators from the US Census Bureau. We calculated per quarter incidence rate (IR) across Q1-Q4 and Q2-Q3. We used multivariable Poisson regression model to report the adjusted incidence rate ratio (aIRR) and 95% confidence interval (CI) for the differences in quarterly incidence between Q1Q4 and Q2Q3.Results:Across 16 years, 985 959 ICH and 7 067 178 AIS hospitalizations were evaluated. The average quarterly crude IR of ICH in Q1Q4 of 6 per 100 000 was higher than the average rate of 5.4 per 100 000 observed in Q2Q3 (aIRR, CI: 1.10, 1.09 - 1.10). However, for AIS the average quarterly IR of ICH in Q1Q4 (41.2 per 100 000) was similar to the incidence rate in Q2Q3 (41.2 per 100 000) (aIRR, CI: 1.00, 1.00 - 1.00). Furthermore, we observed that average quarterly incidence is increasing overtime (aIRR, CI) for both ICH (1.01, 1.01 - 1.01) and AIS (1.03, 1.03 -1.03). The aIRR and CI for age, race, and sex are provided in the figure.Conclusion:Both stroke phenotypes demonstrate an increasing nationwide incidence. However, unlike AIS, ICH incidence is consistently and significantly higher in the months of October to March. Additional evaluation of factors such as holiday stress, seasonal variations in health behaviors (diet, physical activity, medication adherence) and physiological parameters (blood pressure, blood viscosity, fibrinogen, and cholesterol levels) is warranted.

Introduction:Bayesian analyses provide meaningful interpretations of clinical trial data in terms of probabilities of treatment effect. This extends the value of results beyond that obtained from traditional binary cut offs. Such interpretations provide actionable information for conditions with high mortality and no proven treatments, such as intracerebral hemorrhage (ICH).Methods:The minimally invasive surgery with thrombolysis in ICH evacuation (MISTIE 3) trial randomized 506 adults with spontaneous, supratentorial ICH of ≥ 30mL to receive either MISTIE intervention or standard medical care. Using reference priors (minimally informative, enthusiastic, skeptical) and a MISTIE 2-derived prior, we estimated probabilities that the effect of MISTIE intervention exceeds pre-specified thresholds of relative risks (RR) for achieving a 365-day Modified Rankin Scale (mRS) score of 0-3. We also computed the effects of MISTIE intervention on 180 and 30-day mRS and 365, 180 and 30-day mortality.Results:Using a minimally informative prior, the posterior probability of MISTIE intervention having any beneficial effect (RR > 1) on 365-day mRS was 70%. The probabilities of observing RR > 1.02 and > 1.10 were 63% and 34%, respectively. The probabilities of observing RR > 1, 1.02 and 1.10, with other priors are: 87%, 82% and 55% (enthusiastic); 68%, 60% and 30% (skeptical); and 73%, 66% and 38% (MISTIE 2-derived). The probabilities that MISTIE intervention has any beneficial effect on 180 and 30-day mRS, and 365, 180, and 30-day mortality are 65%,80%, 93%, 98% and 99%, respectively. RR, 95% credible intervals across various priors are shown in Figure.Conclusions:The probability that the MISTIE intervention has a favorable functional outcome at 365-days among ICH patients is 68 to 87%. Bayesian-derived probabilities of treatment effect may facilitate shared decision making in ICH management, above and beyond p-value-based frequentist interpretation of trial results.

Background:The current AHA guidelines recommend (class IIb) advanced cardiac imaging in embolic stroke of unknown source (ESUS). We aim to better characterize the role of cardiac CTA (cCTA) in the evaluation of stroke patients.Method:A retrospective review of stroke patients at a comprehensive stroke center, who had cCTA between 12/2016 and 11/2020 was conducted using institutional registries. Ischemic stroke patients with ESUS, cryptogenic (with ESUS as a competing etiology) and cardio-embolic etiologies were included. Only cardioembolic etiologies in which a suspicion of intracardiac thrombus with either a known diagnosis of atrial fibrillation or presence of LVAD were included. Cases with TIA, large vessel disease, small vessel disease, dissection or hemodynamic etiologies were excluded. TTE and cCTA results from all cases were reviewed for remarkable findings, most notably intracardiac thrombus, and analyzed if a change in anti-thrombotic therapy was directly attributable to cCTA results.Results:As per our criteria, 96 cases (Female 42.7%, Black 37.5%, Age mean: 63.7 years) were included, of whom 45 (46.9%) were ESUS, 21 (21.9%) were cryptogenic, and 30 (31.2%) were cardioembolic. cCTA revealed intracardiac thrombus in 5 (5.2%) cases, all with LA thrombus. Of these, 3 (60%) had escalation of anti-thrombotic therapy. Interestingly, in one patient who had both cMRI and cCTA imaging, the cMRI revealed an apical LV thrombus which was not detected on cCTA.Conclusion:In this retrospective analysis of 96 select stroke cases, escalation of anti-thrombotic therapy from anti-platelets to full dose anti-coagulation was indicated in 3 patients (3.1%) with addition of cCTA. Compared to a similar cohort of patients from the same institutional registry using cMRI as the advanced cardiac imaging modality, cCTA did not reveal any LV thrombi and even missed a case of LV thrombus seen on cMRI suggesting lower sensitivity for cCTA to detect LV thrombus. Further analysis of data is in process to determine the subset of stroke patients who would benefit the most from cCTA.

Introduction:Contemporary nationwide estimates of outcomes associated with craniotomy procedures (CP) among patients with primary intracerebral hemorrhage (ICH) are lacking.Methods:From the nationwide inpatient sample (2016 - 2019), we identified adults discharged with primary ICH diagnosis and those who received CP. We fit a series of multivariable logistic regression models to report the adjusted odds ratios (aOR) and 95% confidence interval (CI) for factors associated with receiving CP and the association of CP with in-hospital mortality, home discharge, and development of major complication (MCC): brain abscess, sepsis, deep vein thrombosis, urinary tract infection, and pneumonia. We also fit generalized linear model (γ family) to report the adjusted mean ratio (aMR) and CI for the association of craniotomy with length of stay (LOS) and cost of care.Results:Among 271 840 ICH hospitalizations (mean age [SD] 68.2 [14.9] years; 48% females), 15.3% received CP. Utilization of CP increased from 14.6% in 2016 to 15.8% in 2019 (aOR, CI: 1.10, 1.00 -1.21). Factors independently associated with lower odds of receiving CP (aOR, CI) are older age (0.97, 0.97 - 0.97); Hispanic ethnicity (0.87, 0.78 - 0.97) and Asian race (0.84, 0.73 - 0.96) (vs. Non-Hispanic White); congestive heart failure (0.86, 0.79 - 0.95); diabetes (0.79, 0.72 - 0.87) and renal failure (0.68, 0.62 - 0.74). Private (vs. Medicare) insurance (1.23, 1.13 - 1.34); and residence in large metropolitan (vs. non-metropolitan) counties (1.14, 1.03 - 1.25) is associated with higher odds of receiving a CP. CP is associated with significantly lower odds of in-hospital mortality (aOR, CI: 0.66, 0.61 - 0.72) and home discharge (0.51, 0.45 - 0.57). However, CP ICH patients had longer LOS (aMR, CI: 1.79. 1.75 - 1.84) and higher cost of care (1.82, 1.79 - 1.85). CP is not associated with MCC (aOR, CI: 0.99, 0.92 - 1.07).Conclusions:Modern-day craniotomy procedures likely improve ICH in-patient survival and are not associated with major complications. Large, pragmatic clinical trials for ICH patients are needed to evaluate effectiveness of craniotomy, particularly minimally invasive procedures, on longer term mortality and functional outcomes.

Background:Systolic blood pressure variability (SBPV) has been associated with poor outcomes among patients with intracerebral hemorrhage (ICH). Intravenous nicardipine (IVN) is frequently used for blood pressure management among hospitalized ICH patients, however SBPV in response to IVN has not been characterized.Methods:Data for primary ICH patients who received IVN were retrieved from a stroke-specific bioinformatics pipeline. SBPV was quantified as the coefficient of variation (CV) of cuff-measured systolic blood pressure, and CV was calculated over the time periods 0-24 hrs prior to (Pre24), 0-24 hrs after (Post24), and 24-48 hrs after (Post48) initial IVN administration. Group differences were assessed using Friedman’s Test and Bonferroni-corrected Wilcoxon Signed-rank tests. Contributions of SBPV at each timepoint to poor in-hospital mortality or discharge to hospice (poor outcome) were assessed in separate multivariable logistic regression models adjusted for patient characteristics. Adjusted Odds ratios (aOR) and 95% confidence intervals (CI) are reported.Results:Data included 370 ICH patients with a median [interquartile range] age of 65 [54 - 76]. Patients included were 46.0% female, 41.1% Non-Hispanic White, 25.1% Non-Hispanic Black, 23.8% Hispanic, 8.1% Asian, and 1.9% other. Patients had a median CV of 9.2 [6.1-13.6] in the Pre24 period, 11.6 [9.46 - 13.9] in Post24, and 9.5 [7.6 - 12.1] in Post48. SPBV showed significant differences across timepoints (p=0.000), with Post24 showing elevated SBPV (vs. Pre24 (p=0.000) or Post48 (p=0.000), Figure 1A, B). In multivariable modelling, patients in the highest quartile of SPBV during the Post24 period showed increased risk of poor outcome: aOR (CI) 2.91 (1.06-8.01), Figure 1C.Conclusions:SBPV increases over the first 24 hours of IVN administration and patients with higher SPBV during this time are at higher risk for in-hospital mortality. SPBV management protocols need to be evaluated.

Introduction:Contemporary nationwide estimates of outcomes associated with craniotomy procedures (CP) among patients with primary intracerebral hemorrhage (ICH) are lacking.Methods:From the nationwide inpatient sample (2016 - 2019), we identified adults discharged with primary ICH diagnosis and those who received CP. We fit a series of multivariable logistic regression models to report the adjusted odds ratios (aOR) and 95% confidence interval (CI) for factors associated with receiving CP and the association of CP with in-hospital mortality, home discharge, and development of major complication (MCC): brain abscess, sepsis, deep vein thrombosis, urinary tract infection, and pneumonia. We also fit generalized linear model (γ family) to report the adjusted mean ratio (aMR) and CI for the association of craniotomy with length of stay (LOS) and cost of care.Results:Among 271 840 ICH hospitalizations (mean age [SD] 68.2 [14.9] years; 48% females), 15.3% received CP. Utilization of CP increased from 14.6% in 2016 to 15.8% in 2019 (aOR, CI: 1.10, 1.00 -1.21). Factors independently associated with lower odds of receiving CP (aOR, CI) are older age (0.97, 0.97 - 0.97); Hispanic ethnicity (0.87, 0.78 - 0.97) and Asian race (0.84, 0.73 - 0.96) (vs. Non-Hispanic White); congestive heart failure (0.86, 0.79 - 0.95); diabetes (0.79, 0.72 - 0.87) and renal failure (0.68, 0.62 - 0.74). Private (vs. Medicare) insurance (1.23, 1.13 - 1.34); and residence in large metropolitan (vs. non-metropolitan) counties (1.14, 1.03 - 1.25) is associated with higher odds of receiving a CP. CP is associated with significantly lower odds of in-hospital mortality (aOR, CI: 0.66, 0.61 - 0.72) and home discharge (0.51, 0.45 - 0.57). However, CP ICH patients had longer LOS (aMR, CI: 1.79. 1.75 - 1.84) and higher cost of care (1.82, 1.79 - 1.85). CP is not associated with MCC (aOR, CI: 0.99, 0.92 - 1.07).Conclusions:Modern-day craniotomy procedures likely improve ICH in-patient survival and are not associated with major complications. Large, pragmatic clinical trials for ICH patients are needed to evaluate effectiveness of craniotomy, particularly minimally invasive procedures, on longer term mortality and functional outcomes.

Background:We evaluated the frequency and outcomes of Coronavirus Disease 2019 (COVID-19) associated stroke (ischemic and hemorrhagic) and myocardial infarction (MI) in a global research network.Methods:All adult (≥ 18 years) patients with ICD-10 diagnoses of COVID-19 were included and those with a stroke (ischemic stroke, intracerebral hemorrhage [ICH], transient ischemic attach [TIA] or subarachnoid hemorrhage [SAH]) or myocardial infarction (MI) ± 28 days from COVID-19 diagnosis were flagged. Individuals with a prior history of stroke / MI were excluded. COVID-19 associated stroke (CAS) and MI (CAM) frequencies were compared to prior AHA-reported population-based incidences. Severe COVID-19 disease (use of ventilator, intubation, or life support) and 90-day mortality were evaluated among no-stroke/MI and stroke/MI subgroups utilizing general linear models adjusted for age, sex, race/ethnicity, and Charlson Comorbidity Index scores. Adjusted odds ratios (aOR) and 95% confidence intervals (CI) are reported.Results:Between Jan 2020 and Jul 2021, we identified 637,820 COVID-19 patients of whom 4,559 (0.71%) had a CAS and 6,972 (1.09%) had a CAM. COVID-19 was associated with significantly higher stroke/MI rates compared to population based reported incidence (ORs, CI for CAS 1.99, 1.93 - 2.05 and CAM 3.09, 3.01 - 3.16). Severe COVID disease was observed among 1.07% of COVID-19 only patients, 11.25% of CAS patients (aOR, CI: 4.82, 4.35 - 5.34), and 14.81% of CAM patients (aOR, CI: 5.77, 5.34 - 5.22). 90-Day mortality was 1.92% for COVID-19 only patients, 14.35% for CAS patients (aOR, CI: 3.16, 2.87 - 3.46), and 20.52% for CAM patients (aOR, CI: 4.12, 3.85 - 4.41). Group specific demographic and outcome proportions are reported (Figure).Conclusion:The COVID-19 pandemic has tremendously exacerbated the burden of cerebrovascular and cardiovascular disease globally. Continued work is needed to understand drivers of poor outcomes among COVID-19 patients.

Background:Delirium in-hospital (DIH) is common among the critically ill. However, DIH incidence and outcomes are not well characterized among ischemic stroke (IS) patients, particularly those treated with intravenous tissue plasminogen activator (tPA) and / or mechanical thrombectomy (MT).Methods:Utilizing data from a healthcare system with standardized delirium screening protocols, DIH was determined by a positive 4AT / CAM-ICU screen or diagnosis codes. IS patients with tPA or MT were flagged and a subset with available 90-day modified Rankin Scale (mRS) were analyzed for shifts in mRS scores associated with DIH, via ordinal logistic regression models adjusted for age, stroke severity, tPA or MT, Charlson Comorbidity Index [CCI], prior stroke and sepsis / infections. Common odds ratios (OR) and 95% confidence intervals (CI) are reported.Results:Between May 2016 and June 2021, IS was the primary discharge diagnosis in 12,415 encounters (10,878 unique patients). DIH was documented in 41.6% of IS encounters, compared to 20.0% of non-IS encounters. Stroke-DIH patients (vs no-DIH Stroke) were older (median: 75 vs 65 years), more frequently female (53.3% vs 48.7%), with higher comorbidity burden (median CCI: 7 vs 5), longer hospital stays (median: 6 vs 3 days), higher in-hospital mortality (3.1% vs 0.5%), and fewer home discharges (36.2% vs 75.2%). Among a sub-cohort of 2,785 IS patients with 90-day mRS, fully adjusted model indicated lower mRS (OR, CI: 0.48, 0.41-0.57) for those with tPA or MT, and worse outcomes for DIH patients (OR, CI: 2.70, 2.26-3.23). Among 948 treated IS patients, DIH remained a significant risk for worse outcomes (OR, CI: 2.54, 1.89-3.43).Conclusion:Delirium was twice as common in IS patients and was a negative prognostic indicator of short and long-term outcomes among non-treated and treated IS patients. Active screening and management of DIH is critically important to improve stroke outcomes.

Background:Impact of socioeconomic disadvantage on outcomes among acute ischemic stroke (AIS) patients has not been well characterized.Methods:Clinical data on AIS patients were extracted from electronic medical records and 90-day modified Rankin Scale (mRS) scores were collected as a part of prospective stroke registry. Exact patient addresses were geocoded, and state-level Area Deprivation Index (ADI) ranks were categorized as low, medium, high. Patients with a 90-day mRS score ≥ 4 were categorized as severe disability or death (SDD). Logistic regression models (adjusted for treatment with intravenous tissue plasminogen activator or mechanical thrombectomy, age, sex, race/ethnicity, insurance, prior stroke, vascular risk factors) were fitted to compute odds ratios (OR) and 95% confidence intervals (CI) for total effect of high ADI on SDD. Structural equation modeling was used to assess mediation effects of stroke severity as measured by National Institutes of Health Stroke Scale (NIHSS).Results:Between May 2016 and Apr 2021, a total of 2,900 AIS patients (mean age: 68.5 years; 50.1% male; 28.4% non-Hispanic Black; 12.9% Hispanic) with complete outcomes data were included. In an adjusted model, high ADI was significantly associated with SDD (OR, CI: 1.14, 1.02-1.28). In the mediation analysis, patients in higher ADI neighborhoods had a 28% increased likelihood of having higher NIHSS (OR: 1.28, CI: 1.15-1.44). Likewise, higher NIHSS was associated with SDD (OR: 7.10, CI: 5.96-8.51). The effect of neighborhood disadvantage on SDD was fully mediated by NIHSS (average causal mediation effect of ADI on SDD: P=0.002), with 77% of the total effect pathway mediated through NIHSS. The proportions of 90-day mRS by ADI categories are reported (Figure).Conclusion:Neighborhood disadvantage leads to poor stroke outcomes mediated via stroke severity. Tracking social determinants of health may identify opportunities for reducing stroke related disability.

Introduction:There is limited data on real-world efficacy of left atrial appendage closure (LAAC) procedures compared to anticoagulants (AC) for stroke prevention among patients with atrial fibrillation (AF).Methods:We utilized a retrospective cohort of over 63 million patients from 51 healthcare organizations across 6 countries via a harmonized electronic medical record-based research data platform. Utilizing international classification of disease version 10 and current procedural terminology codes, adult (≥ 18 years) patients with AF (I48) were identified and grouped by treatment type (LAAC (33340, 02L73DK) vs AC (1015112, Z79.0, BL110)). Patients with a history of stroke prior to AC initiation or LAAC procedure were excluded. Subjects were followed for 5 years for incident ischemic stroke (I63), intracerebral hemorrhage (I61), and all-cause mortality. Treatment groups were propensity score matched by age, sex, race, ethnicity, and comorbidities. Risk ratio(RR) and 95% confidence intervals(CI) among unmatched and matched populations are reported.Results:Among a total of 1,980,130 AF patients; 1,374,013 were flagged for AC use and 8,004 were treated via LAAC. Treatment cohorts were propensity score matched by age, race, sex, hyperlipidemia, and hypertension resulting in an exact 1:1 matched cohort of 8,004 subjects, across all co-variates. In the matched population the mean (SD) age was 75.2 (8.05) years, 63.2% were male; 86.6% were white, 4.4% black, 0.89% Asian, and 3.5% Hispanic; with 64.4% hyperlipidemic and 79.2% hypertensive. Prior to matching, AC (vs. LAAC) was significantly associated with a higher 5-year risk of ischemic stroke, intracerebral hemorrhage, and all-cause mortality. The protective effect of LAAC (vs. AC) was maintained for 5-year risk of ischemic stroke (RR, CI: 0.68, 0.58 - 0.79) and all-cause mortality (RR, CI: 0.42, 0.39 - 0.45). However, the difference in 5-year risk for intracerebral hemorrhage for LAAC (vs. AC) was not statistically lower (RR, CI: 0.72, 0.50 - 1.05).Conclusion:Notwithstanding the possibility of residual confounding in our analyses, LAAC seems to be associated with a lower long-term risk of ischemic stroke and all-cause mortality as compared to AC treated AF patients in large real-world data.

Background:Delirium in-hospital (DIH) results in poor in-patient outcomes. However, it’s longer-term effects among intracerebral hemorrhage (ICH) patients are not well characterized. We sought to evaluate the association between DIH and 90-day readmission (RA) among primary ICH patients.Method:Clinical, imaging and outcomes data, from May 2016 to June 2021, were obtained from the Neurological Outcomes Registry for ICH (NEURO-RICH); an informatics pipeline across 7 comprehensive and primary stroke centers which implement protocolized delirium assessments via 4AT / CAM-ICU scales. Demographic (age, sex, race, ethnicity, marital status), Glasgow Coma Scale (GCS), systolic and diastolic blood pressure (S/DBP), sepsis, systemic inflammatory response syndrome, and comorbidity data were analyzed. Survival analysis for time-to-90-Day RA was performed with death modeled as a competing risk. Sub hazard ratios (SHR) and 95% confidence intervals (CI) are reported. Subgroup with imaging data (ICH score and cerebral small vessel disease) was analyzed.Results:Final analyses included 1,434 ICH patients (mean age: 66.0 years, 47.2% female, 24.8% non-Hispanic Black, 19.9% Hispanic, median GCS: 13.2, SBP: 163 mmHg). In the fully adjusted model, DIH was significantly associated with higher rates of 90-Day RA (SHR, CI: 2.24, 1.04 - 4.81) (Figure 1A). Anticoagulant therapy (1.36, 1.03 - 1.80), hypertension (1.95, 1.38 - 2.74), diabetes (1.13, 1.09 - 1.17), and Charlson Comorbidity Index (1.08, 1.04 - 1.13) were also independently associated with 90-Day RA (SHR, CI reported). In the imaging subgroup of 523 patients, DIH retained independent association with 90-Day RA (SHR: 3.94 [1.37 - 11.36]) (Figure 1B).Discussion:In addition to traditional predictors of poor outcomes among ICH patients, DIH demonstrated strong and independent association with 90-day RA. Screening, identification, and active management of DIH is critical to reduce long-term disease burden of ICH.

Background:The current AHA guidelines recommend (class IIb) advanced cardiac imaging in an embolic stroke of unknown source (ESUS). We aim to better characterize the role of cardiac MRI (cMRI) in the evaluation of stroke patients and determine patient characteristics that may increase the positive predictive value of cMRI.Method:A retrospective review of stroke patients at a comprehensive stroke center, who had cMRI between 12/2016 and 11/2020 was conducted using institutional registries. Ischemic stroke patients with ESUS, cryptogenic (with ESUS as a competing etiology), and cardioembolic etiologies were included. Cases with TIA, large vessel, small vessel, or hemodynamic etiologies were excluded. Cardioembolic etiology was defined as newly diagnosed Afib patients with TTE abnormalities including wall motion abnormalities and/or reduced LVEF <50%, stroke while compliant with anticoagulation for known Afib, or among whom the timing of restarting anticoagulation was in question. TTE and cMRI results from all cases were reviewed for remarkable findings, most notably intracardiac thrombus, and analyzed if a change in anti-thrombotic therapy was directly attributable to cMRI results.Results:As per our criteria, 250 cases (Female 44.4%, Black 29.6%, Age mean: 65.2 years) were included, of whom 146 (58.4%) were ESUS, 86 (34.4%) were cryptogenic, and 18 (7.2%) were cardioembolic. Sixteen (6.4%) revealed intracardiac thrombus, of which 10 had LV thrombus, 5 had LA thrombus, and 1 had RA thrombus along with questionable PFO on cMRI. Of these, 14 (87.5%) had an escalation of anti-thrombotic therapy. For the other 2 cases, despite cMRI obtained due to low LVEF and revealing LA thrombus, antithrombotic regimen was not changed. Eight (3.2%) additional patients were placed on anticoagulation secondary to incidental findings.Conclusion:In this large retrospective analysis of 250 select stroke cases, escalation of anti-thrombotic therapy from anti-platelets to full dose anti-coagulation was indicated in 14 patients (5.6%) with addition of cMRI. Further analysis of data is in process to determine the subset of stroke patients who would benefit the most from cMRI.