Tracheostomy in Acute Ischemic Stroke: Declining National Utilization, Independent Predictors, and In-Hospital Outcomes Among 854,660 Hospitalizations

Tracheostomy in Acute Ischemic Stroke: Declining National Utilization, Independent Predictors, and In-Hospital Outcomes Among 854,660 Hospitalizations | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Tracheostomy in Acute Ischemic Stroke: Declining National Utilization, Independent Predictors, and In-Hospital Outcomes Among 854,660 Hospitalizations Mian Urfy, Mariam Tariq Mir This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9546300/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background Tracheostomy in acute ischemic stroke (AIS) is a high-stakes decision yet contemporary national data on utilization, predictors, and outcomes are limited. We analyzed 854,660 AIS hospitalizations (NIS, 2016–2023) to characterize tracheostomy use, identify predictors, examine timing effects, and explore mechanisms underlying a national decline. Methods Tracheostomy was identified by ICD-10-PCS codes and classified as early (≤ 7 days) or late (> 7 days). Multivariable logistic regression identified predictors among all AIS patients (Model A) and among mechanically ventilated (MV) patients (Model B). Outcomes included length of stay, in-hospital mortality, PEG placement, and discharge disposition. Temporal trends were assessed by Cochran-Armitage test and logistic regression; mechanistic analyses tested five pre-specified hypotheses including a year-thrombectomy interaction model. Results Tracheostomy occurred in 5,373 hospitalizations (0.63% of AIS; 12.1% of MV patients). Median procedure day was 11 (IQR 8–16); 22.0% were early and 78.0% late. In-hospital mortality was 9.84% (early) and 8.23% (late), versus 39.7% in ventilated patients without tracheostomy. Mean LOS was significantly shorter with early versus late tracheostomy (23.8 vs 34.4 days; adjusted LOS ratio 0.662; p < 0.001). Prolonged MV (OR 62.3), hemorrhagic transformation (OR 1.35), and decompressive craniectomy (OR 2.55) were the dominant independent predictors. Tracheostomy rates declined from 12.9% to 9.44% of MV patients (Z = − 6.54; p < 0.0001), with the steepest drop in 2023. Mechanistic analyses showed thrombectomy utilization rose concurrently (15.3% to 22.2% among MV patients; Z = + 6.39; p < 0.0001), prolonged MV declined (Z = − 6.77; p < 0.0001), and a significant year-thrombectomy interaction (p = 0.027) suggested faster decline among thrombectomy-treated patients. PEG placement occurred in 78.3% of tracheostomy patients. Conclusions Declining tracheostomy rates in AIS likely reflect a convergence of factors: expanding thrombectomy utilization, shorter ventilator dependence, and evolving ICU practice — rather than a shift in the tracheostomy decision itself. Racial disparities persist after adjustment. These findings provide the largest contemporary national framework for tracheostomy decision-making and goals-of-care counseling in AIS. acute ischemic stroke tracheostomy mechanical ventilation neurocritical care tracheostomy timing outcomes National Inpatient Sample Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Mechanical ventilation (MV) complicates approximately 5% of acute ischemic stroke (AIS) hospitalizations and is strongly associated with increased morbidity, resource utilization, and mortality. 1 , 2 Among patients who require MV, a critical and consequential subset will ultimately require tracheostomy—either to facilitate ventilator weaning, provide long-term airway protection in the setting of impaired swallowing or consciousness, or manage complications such as aspiration pneumonia. Tracheostomy is among the most resource-intensive procedures performed in neurocritical care and is associated with prolonged hospitalizations, high costs, and uncertain long-term outcomes in the stroke population. 3 , 4 The optimal timing of tracheostomy after stroke has been the subject of considerable debate and investigation. Observational studies using national inpatient databases established that early tracheostomy was associated with reduced ventilator-associated pneumonia, shorter length of stay, and lower hospital costs compared with late placement. 5 , 6 These findings generated clinical enthusiasm for earlier tracheostomy practices in severe stroke. However, the landmark SETPOINT2 randomized clinical trial, which enrolled 382 patients with severe ischemic or hemorrhagic stroke at 26 neurocritical care centers, found that early tracheostomy (≤ 5 days) did not improve functional outcome at 6 months compared with a standard approach (tracheostomy from day 10 if needed). 7 A subsequent meta-analysis of over 17,000 critically ill stroke patients confirmed that tracheostomy timing was not associated with mortality, neurological outcomes, or ICU or hospital length of stay. 8 These findings have reframed the clinical question: while early tracheostomy may not improve functional recovery, its effects on short-term in-hospital outcomes such as LOS, complications, and resource utilization remain clinically important and practice-relevant. Contemporary nationally representative data characterizing tracheostomy in the AIS population are limited. Prior administrative database studies used data from 2007–2017 and focused either on stroke-specific tracheostomy populations, 4 on thrombectomy-specific cohorts, 6 or on all-stroke populations combining ischemic and hemorrhagic subtypes. 3 The period from 2016 onward represents a transformed era in AIS care, marked by widespread adoption of mechanical thrombectomy, expansion of thrombectomy-capable centers, and evolving neurocritical care practices. Studies examining tracheostomy specifically in AIS across this modern era, including the years during and after the COVID-19 pandemic, are lacking. Furthermore, several important and under characterized aspects of tracheostomy in AIS warrant dedicated investigation. The role of hemorrhagic transformation (HT) as a driver of tracheostomy has not been systematically examined at national scale. The mechanistic pathway from prolonged MV to tracheostomy has not been quantified in large populations. Racial disparities in tracheostomy use after AIS have received little attention despite being well-documented in other neurocritical care populations. And the relationship between tracheostomy and co-procedures such as gastrostomy (PEG) placement — a marker of anticipated prolonged disability — has not been examined in this context. Notably, a 2021 national survey found no consensus on optimal tracheostomy timing in neurocritical patients, with clinicians reporting inconsistent practice driven by local culture rather than evidence. 9 A recent multicenter study in neurocritical patients confirmed that early tracheostomy is associated with shorter LOS and lower total hospital costs in stroke patients, yet also found a 40% lower likelihood of discharge to rehabilitation in the late-tracheostomy group — underscoring that the consequences of timing extend beyond mortality. 10 The 2026 Neurocritical Care Society neuro-prognostication guidelines for critically ill AIS patients now explicitly emphasize tracheostomy decannulation prediction and goals-of-care counseling for large hemispheric infarction as clinical priorities, 11 creating an urgent need for contemporary population-level data to anchor these recommendations. Using a large, nationally representative cohort of 854,660 AIS hospitalizations from 2016–2023, we aimed to: (1) characterize the prevalence and temporal trends of tracheostomy in contemporary AIS care; (2) identify predictors of tracheostomy among all AIS patients and among those specifically requiring MV, including the novel contribution of decompressive craniectomy; (3) examine the association between tracheostomy timing (early vs. late) and in-hospital outcomes including LOS, mortality, PEG placement, and discharge destination; (4) compare outcomes across four clinically defined groups — no MV, MV without tracheostomy, early tracheostomy, and late tracheostomy; and (5) characterize racial and procedural subgroup differences in tracheostomy patterns. The cohort selection process is depicted in Fig. 1. Methods Data Source This retrospective observational study used data from the National Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP), for years 2016 through 2023. The NIS is the largest publicly available all-payer inpatient database in the United States, representing approximately 20% of all U.S. hospitalizations through a stratified sampling design. Discharge weights enable generation of nationally representative estimates. All analyses followed HCUP methodological guidance. The NIS contains fully de-identified data; institutional review board approval and informed consent were not required. 12 Study Population Hospitalizations with a principal diagnosis of acute ischemic stroke (AIS) were identified using ICD-10-CM codes I63.x. Patients aged 18 to 90 years were included. Exclusions were applied for missing age, sex, discharge disposition, or invalid sampling weights. The final analytic cohort comprised 854,660 unweighted AIS hospitalizations, representing approximately 4.27 million weighted discharges. The stepwise cohort selection process, exclusion criteria, and analytic group derivation are illustrated in Fig. 1. Mechanical ventilation (MV) was identified using ICD-10-PCS procedure codes 5A1935Z ( 96 hours). Tracheostomy and Timing Classification Tracheostomy was identified using ICD-10-PCS procedure codes for open (0B110F4, 0B110ZZ), percutaneous (0B113F4, 0B113ZZ), and percutaneous endoscopic (0B114F4, 0B114ZZ) approaches. Procedure day was extracted by matching tracheostomy procedure codes to corresponding PRDAY slots (PRDAY1 through PRDAY10). Tracheostomy timing was classified as early (≤ 7 days from admission) or late (> 7 days), consistent with recent administrative database studies in the stroke literature. 6 , 13 Patients with tracheostomy but missing procedure day data were classified separately. Gastrostomy tube placement (PEG) was identified using codes 0DH63UZ, 0DH64UZ, and 0DH60UZ. Variables Demographic variables included age, sex, and race/ethnicity. Comorbidities were derived using ICD-10-CM codes consistent with Elixhauser and Charlson classification systems: congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), atrial fibrillation (AFIB), chronic kidney disease (CKD), liver disease, coagulopathy, cancer, obesity, diabetes, and hypertension. Hemorrhagic transformation (HT) was identified by secondary ICD-10-CM codes I61.x or I62.x in the AIS cohort. Treatment exposures included mechanical thrombectomy (03CG3ZZ, 03CG4ZZ, 03CK3ZZ, 03CK4ZZ, 03CL3ZZ, 03CL4ZZ) and intravenous thrombolysis (3E033 series). In-hospital mortality was defined by discharge disposition code 20. Prolonged MV was defined as MV duration exceeding 24 hours (codes 5A1945Z or 5A1955Z). 14 Statistical Analysis Descriptive statistics compared tracheostomy and non-tracheostomy groups among all AIS patients (Table 1 ) and among mechanically ventilated patients (Table 2 ). Chi-square tests were used for categorical variables and t-tests for continuous variables. Two multivariable logistic regression models were developed: Model A identified predictors of tracheostomy among all AIS patients; Model B identified predictors among MV patients only, additionally including prolonged MV as a predictor. Results are reported as adjusted odds ratios (OR) with 95% confidence intervals. Table 1 Baseline demographic, clinical, and treatment characteristics stratified by tracheostomy status among all AIS patients (n = 854,660). Values are number (%) or mean (SD). Characteristic No Trach (n = 849,287) Trach (n = 5,373) p Age, mean (SD) 69.9 (14.0) 63.3 (13.6) *** Sex — Female 461,644 (49.4%) 2,251 (41.9%) *** Race — White 486,480 (57.3%) 2,174 (40.5%) *** Race — Black 124,842 (14.7%) 1,481 (27.6%) *** Race — Hispanic 60,986 (7.2%) 488 (9.1%) *** Race — Asian/PI 22,347 (2.6%) 199 (3.7%) *** Weekend Admission 237,754 (25.4%) 3,824 (26.3%) * Comorbidities Congestive Heart Failure 185,376 (19.8%) 4,336 (29.8%) *** COPD 135,809 (16.0%) 974 (18.1%) *** Atrial Fibrillation 213,347 (25.1%) 1,623 (30.2%) NS Chronic Kidney Disease 192,065 (20.5%) 3,127 (21.5%) ** Liver Disease 16,932 (2.0%) 299 (5.6%) *** Coagulopathy 50,201 (5.9%) 845 (15.7%) *** Cancer 42,121 (5.0%) 197 (3.7%) *** Obesity 139,638 (12.1%) 2,315 (15.9%) ** Diabetes 369,228 (39.5%) 5,175 (35.5%) *** Hypertension 723,897 (85.2%) 3,937 (73.3%) *** Hemorrhagic Transformation 43,206 (5.1%) 1,447 (26.9%) *** Treatments Mechanical Thrombectomy 32,004 (3.8%) 887 (16.5%) *** IV tPA 81,327 (9.6%) 647 (12.0%) *** MV (received) 37,339 (4.4%) 5,147 (95.8%) *** Prolonged MV > 24h 27,251 (3.2%) 5,119 (95.3%) *** PEG Tube 22,572 (2.7%) 4,103 (76.4%) *** LOS, mean (SD) 5.09 (6.79) 32.29 (26.38) *** In-hospital Mortality 33,971 (4.0%) 484 (9.0%) *** *** p < 0.001; ** p < 0.01; * p < 0.05; NS = not significant. AIS = acute ischemic stroke; COPD = chronic obstructive pulmonary disease; CKD = chronic kidney disease; MV = mechanical ventilation; PEG = percutaneous endoscopic gastrostomy; LOS = length of stay. Table 2 Characteristics stratified by tracheostomy status among mechanically ventilated AIS patients only (n = 42,486). Selected variables shown. Characteristic No Trach (n = 37,339) Trach (n = 5,147) p Age, mean (SD) 68.5 (14.3) 63.3 (13.6) *** Sex — Female 18,387 (49.2%) 2,151 (41.8%) *** Race — White 19,683 (52.7%) 2,083 (40.5%) *** Race — Black 5,896 (15.8%) 1,432 (27.8%) *** Congestive Heart Failure 10,471 (28.0%) 1,418 (27.6%) *** Atrial Fibrillation 13,889 (37.2%) 1,568 (30.5%) *** Coagulopathy 4,881 (13.1%) 811 (15.8%) *** Liver Disease 1,765 (4.7%) 285 (5.5%) NS Hemorrhagic Transformation 7,576 (20.3%) 1,400 (27.2%) *** Mechanical Thrombectomy 7,052 (18.9%) 861 (16.7%) *** Prolonged MV > 24h 27,251 (73.0%) 5,119 (99.5%) *** PEG Tube 3,962 (10.6%) 4,029 (78.3%) *** LOS, mean (SD) 10.8 (13.4) 32.1 (25.8) *** In-hospital Mortality 14,822 (39.7%) 463 (9.0%) *** *** p < 0.001; NS = not significant. Table restricted to selected clinically relevant variables. Full comparison available in supplementary material. Outcomes were compared across four clinically defined groups: no MV, MV without tracheostomy, early tracheostomy (≤ 7 days), and late tracheostomy (> 7 days). Between-group outcome comparisons used chi-square tests for proportions and t-tests for continuous variables. Adjusted mortality models compared early versus late tracheostomy, controlling for age, sex, and key comorbidities. Decompressive craniectomy was identified using ICD-10-PCS codes for skull excision (0NB series) and release of cerebral meninges (00N series), consistent with prior administrative database methodology. This variable was incorporated as an additional predictor in a pre-specified sensitivity model (Model B-extended) to quantify its independent contribution to tracheostomy risk among MV patients. Discharge disposition was categorized as: home or home health, skilled nursing facility (SNF) or long-term acute care hospital (LTACH), short-term hospital, or died. In-hospital complications including hospital-acquired pneumonia, sepsis, and acute kidney injury (AKI) were identified by secondary ICD-10-CM diagnosis codes. Temporal trends in tracheostomy rates were assessed using logistic regression with calendar year as a continuous predictor among MV patients; a Cochran-Armitage trend test was performed as a confirmatory analysis. To investigate potential mechanisms underlying the observed temporal decline in tracheostomy utilization, we performed a pre-specified mechanistic analysis testing five hypotheses: (1) rising mechanical thrombectomy utilization; (2) falling overall MV rates; (3) falling prolonged MV rates among MV patients; (4) case-mix severity shifts; and (5) COVID-19 confounding. Cochran-Armitage trend tests were applied to annual proportions for each hypothesis. A logistic interaction model (TRACH ~ YEAR × THROMBECTOMY + covariates) tested whether tracheostomy rates declined differentially by thrombectomy status. All analyses were performed in R version 4.5.2. Results Cohort and Tracheostomy Prevalence The final analytic cohort included 854,660 AIS hospitalizations, of which 42,486 (5.0%) required MV. Tracheostomy was performed in 5,373 hospitalizations (0.63% of all AIS; 12.1% of MV patients). Among non-MV patients, tracheostomy was exceedingly rare (n = 226, 0.03%), confirming that tracheostomy in AIS is almost exclusively a phenomenon of the ventilated population. Among the 5,373 tracheostomy cases, 5,147 (95.8%) had concurrent MV documentation. Patient Characteristics Tracheostomy patients were substantially younger than non-tracheostomy AIS patients (mean 63.3 ± 13.6 vs 69.9 ± 14.0 years; p < 0.001; SMD 0.482) and more frequently male (58.1% vs 50.6%; p < 0.001) (Table 1 ). Pronounced racial differences were observed: Black patients comprised 27.6% of tracheostomy cases compared with 14.7% of non-tracheostomy cases (p < 0.001), and White patients were underrepresented in the tracheostomy group (40.5% vs 57.3%; p < 0.001; SMD 0.358). Among MV patients, coagulopathy was present in 15.8% of tracheostomy versus 13.1% of non-tracheostomy patients, liver disease in 5.5% versus 4.7%, and CHF in 27.6% versus 28.0% (Table 2 ). Hemorrhagic transformation was identified in 27.2% of tracheostomy patients compared with 20.3% of non-tracheostomy MV patients (SMD 0.163; p 24h) compared with 73.0% of ventilated patients without tracheostomy (p < 0.001; Table 2 ). Tracheostomy Timing Among 5,217 tracheostomy patients with documented procedure day data, the median procedure day was 11 (IQR 8–16). Early tracheostomy (≤ 7 days) was performed in 1,146 patients (22.0%) and late tracheostomy (> 7 days) in 4,071 patients (78.0%). Procedure day data were missing in 156 patients (2.9%). The predominance of late tracheostomy is consistent with prevailing practice patterns in which tracheostomy is considered after failed weaning attempts typically beginning around day 7–10 after intubation. Tracheostomy Rate by MV Duration Tracheostomy rates increased markedly with MV duration. Among patients with short-duration MV ( 96h), 29.1%. This gradient confirms that tracheostomy in AIS is largely a function of ventilatory dependence rather than decision-making at the time of initial intubation. Predictors of Tracheostomy In Model A (all AIS patients; Table 3 ), hemorrhagic transformation demonstrated the strongest independent association with tracheostomy (adjusted OR 5.19; 95% CI 4.86–5.54; p < 0.001), followed by mechanical thrombectomy (OR 3.04; 95% CI 2.81–3.28), coagulopathy (OR 2.18; 95% CI 2.02–2.36), liver disease (OR 2.05; 95% CI 1.81–2.32), and CHF (OR 1.58; 95% CI 1.48–1.68). Age was strongly and inversely associated (OR 0.972 per year; 95% CI 0.970–0.974; p < 0.001). Cancer (OR 0.665; 95% CI 0.573–0.768) and White race (OR 0.557; 95% CI 0.482–0.648 vs Asian/PI reference) were independently associated with lower odds of tracheostomy. IV tPA was not independently associated with tracheostomy in Model A (OR 1.01; 95% CI 0.93–1.09; NS), and obesity was likewise not significant (OR 1.03; 95% CI 0.96–1.10; NS). Table 3 Multivariable logistic regression predictors of tracheostomy: Model A (all AIS patients) and Model B (mechanically ventilated patients only). Selected significant predictors shown. Variable Model A OR 95% CI p Model B OR † 95% CI p Age (per year) 0.972 0.970–0.974 < .001 0.981 0.979–0.983 < .001 Female (vs Male) 0.873 0.844–0.903 < .001 0.939 0.904–0.976 .001 Race — Black (vs Asian/PI) 1.28 1.10–1.49 < .001 1.13 0.96–1.35 NS Race — White (vs Asian/PI) 0.557 0.482–0.648 < .001 0.571 0.484–0.677 < .001 Comorbidities Hemorrhagic transformation 5.19 4.86–5.54 < .001 1.35 1.26–1.44 < .001 Coagulopathy 2.18 2.02–2.36 < .001 1.10 1.01–1.19 .028 Liver disease 2.05 1.81–2.32 < .001 1.00 0.88–1.15 NS CHF 1.58 1.48–1.68 < .001 0.95 0.88–1.02 NS COPD 1.22 1.14–1.31 < .001 1.05 0.97–1.14 NS Atrial Fibrillation 1.38 1.29–1.47 < .001 0.932 0.869–1.00 NS CKD 0.907 0.843–0.976 .010 0.864 0.796–0.936 < .001 Obesity 1.03 0.96–1.10 NS 1.16 1.07–1.25 < .001 Diabetes 1.08 1.02–1.14 .009 1.11 1.04–1.18 < .001 Treatments Mechanical thrombectomy 3.04 2.81–3.28 < .001 0.955 0.88–1.04 NS IV tPA 1.01 0.93–1.09 NS 0.863 0.786–0.945 24h — — — 62.3 43.9–92.6 < .001 Model A = all AIS patients (n = 854,660). Model B = mechanically ventilated AIS patients only (n = 42,486); additionally includes prolonged MV as predictor. Reference category for race = Asian/Pacific Islander. NS = not significant. OR = adjusted odds ratio; CI = confidence interval. Note the directional reversal of comorbidity ORs between Models A and B — see Discussion. In Model B (MV patients only; Table 3 ), the predictor structure changed substantially. Prolonged MV (> 24h) was overwhelmingly the dominant predictor (OR 62.3; 95% CI 43.9–92.6; p < 0.001), reflecting the near-deterministic relationship between extended ventilatory dependence and the tracheostomy decision. Among clinical variables, hemorrhagic transformation remained independently associated (OR 1.35; 95% CI 1.26–1.44; p < 0.001). Notably, traditional comorbidities including CHF (OR 0.95; NS), liver disease (OR 1.00; NS), and CKD (OR 0.864) no longer showed independent positive associations in the MV-only model—reflecting that among ventilated patients, those with severe systemic comorbidity die before reaching the tracheostomy decision threshold rather than surviving to require it. Age remained independently inversely associated (OR 0.981 per year; p < 0.001). In the pre-specified sensitivity analysis incorporating decompressive craniectomy (Model B-extended), this procedure was independently and strongly associated with tracheostomy (OR 2.55; 95% CI 2.31–2.80; p < 0.001), confirming at national scale that the decompressive craniectomy pathway in AIS is tightly coupled with subsequent tracheostomy placement — an observation previously documented only in single-center large vessel occlusion cohorts. 15 Outcomes by Tracheostomy Timing Group Outcomes differed substantially across the four clinical groups (Table 4 ; Figs. 3 and 4). In-hospital mortality was lowest in the no-MV group (2.32%) and highest in the MV-no-tracheostomy group (39.7%), reflecting the severity of acute illness in those who die before the tracheostomy threshold is reached. Mortality was 9.84% in the early tracheostomy group and 8.23% in the late tracheostomy group (unadjusted p = 0.007). After adjustment for age, sex, and comorbidities, early tracheostomy was associated with significantly higher mortality compared with late tracheostomy (adjusted OR 1.36; 95% CI 1.08–1.70; p = 0.007) — consistent with confounding by indication, in which patients selected for early procedure are systematically sicker. Table 4 In-hospital outcomes by ventilation and tracheostomy status group. Outcome No MV (n = 812,174) MV, No Trach (n = 37,339) Early Trach ≤7d (n = 1,146) Late Trach >7d (n = 4,071) LOS, mean (SD), days 4.8 (6.2) 10.8 (13.4) 23.8 (22.9) 34.4 (26.2) In-hospital mortality 2.32% 39.7% 9.84% 8.23% PEG placement 2.3% 10.6% 77.2% 78.5% Discharge home 54.4% 13.4% 4.5% 3.9% Discharge to SNF/LTACH 39.4% 41.7% 80.7% 84.2% All between-group differences p < 0.001 except early vs. late tracheostomy mortality (unadjusted p = 0.007; adjusted OR 1.36, 95% CI 1.08–1.70, p = 0.007 — indicating higher adjusted mortality with early tracheostomy, consistent with confounding by indication) and PEG co-placement between early and late tracheostomy (p = 0.38). LOS difference between early and late tracheostomy: mean 10.6 days (95% CI 9.3–12.4; p < 0.001); adjusted LOS ratio 0.662 (95% CI 0.638–0.687; p 7 days. SNF = skilled nursing facility; LTACH = long-term acute care hospital. MV = mechanical ventilation; PEG = percutaneous endoscopic gastrostomy. Discharge home includes home health care. Decompressive craniectomy was performed in 16.3% of tracheostomy and 4.4% of non-tracheostomy MV patients (p < 0.001). Length of stay differed significantly across groups. Early tracheostomy patients had a mean LOS of 23.8 ± 22.9 days, compared with 34.4 ± 26.2 days for late tracheostomy (mean difference 10.6 days; 95% CI 9.3–12.4 days; p < 0.001). PEG tube co-placement was present in 77.2% of early and 78.5% of late tracheostomy patients, compared with 10.6% in ventilated patients without tracheostomy and 2.3% in the no-MV group, indicating that the majority of tracheostomy patients are expected to require long-term enteral nutrition. Discharge destination paralleled these findings: 85.5% of early tracheostomy and 87.7% of late tracheostomy patients were discharged to institutional settings (skilled nursing facility, long-term acute care hospital, or inpatient rehabilitation), compared with 46.5% of mechanically ventilated patients without tracheostomy. Only 4.5% and 3.9% of early and late tracheostomy patients, respectively, were discharged home or to home health care, reinforcing the interpretation that tracheostomy in AIS signals anticipated long-term disability. Temporal Trends Tracheostomy rates among MV patients declined significantly over the study period, from 12.9% in 2016 to 9.44% in 2023 (Cochran-Armitage Z = − 6.54, p < 0.0001; logistic trend OR per year 0.958; 95% CI 0.946–0.970; Fig. 2). The steepest single-year decline occurred from 2022 to 2023 (− 2.78 percentage points). Median procedure day was stable at Day 11 in 2016–2020 and shifted to Day 12 in 2021–2023. Mechanistic analyses identified convergent structural explanations for the declining tracheostomy rate. The overall MV rate rose significantly over the study period (4.44% in 2016 to 5.04% in 2023; Cochran-Armitage Z = + 11.4, p 24 hours) declined significantly from 78.5% in 2016 to 73.9% in 2023 (Z = − 6.77, p 96 hours of ventilation reached a dataset low of 34.6% in 2023 — indicating that a growing proportion of AIS patients are being extubated before reaching the tracheostomy decision threshold. Mechanical thrombectomy utilization rose substantially over the study period. Among MV patients, the thrombectomy rate increased from 15.3% in 2016 to 22.2% in 2023 (Cochran-Armitage Z = + 6.39, p < 0.0001); across all AIS patients, thrombectomy utilization expanded from 2.86% to 4.77% (Z = + 20.9, p < 0.0001). A significant year-thrombectomy interaction (OR 0.960 per year, 95% CI 0.926–0.995; p = 0.027) confirmed that tracheostomy rates declined faster in thrombectomy-treated patients than in non-thrombectomy patients (7.28% vs 10.1% in 2023), consistent with thrombectomy-mediated recanalization enabling earlier neurological recovery and earlier extubation success. COVID-19 contributed a transient reduction in tracheostomy rates during 2020–2021, with partial rebound in 2022; the 2023 decline therefore represents a new post-COVID structural change. In-hospital mortality among MV patients who did not receive tracheostomy fell from 43.4% in 2016 to 36.6% in 2023, further supporting a shift toward a population with higher rates of successful ventilator liberation. Thrombectomy Subgroup Among MV patients, tracheostomy rates differed by thrombectomy status (10.9% in thrombectomy vs 12.4% in non-thrombectomy patients; p < 0.001). Thrombectomy patients who did receive tracheostomy had shorter LOS (30.7 vs 32.4 days) and markedly lower mortality (7.44% vs 8.87%), likely reflecting better baseline neurologic reserve and purposeful patient selection for tracheostomy among those with perceived recovery potential. Discussion In this nationally representative analysis of 854,660 AIS hospitalizations from 2016 to 2023, we demonstrate that tracheostomy occurs in 0.63% of all AIS patients and 12.1% of those requiring mechanical ventilation. These figures are consistent with prior national estimates spanning earlier eras 3 , 4 and provide the first national mechanistic analysis of declining tracheostomy rates in contemporary AIS care. Our findings extend prior work in several important dimensions: we demonstrate that the declining tracheostomy trend is driven primarily by upstream changes in ventilatory trajectory — specifically, rising thrombectomy utilization enabling earlier extubation and a structural decline in prolonged MV duration — rather than a shift in the tracheostomy decision itself; we characterize a mechanistic hierarchy of tracheostomy predictors; we document marked racial disparities in tracheostomy use; and we confirm that early tracheostomy is associated with shorter LOS but not survival benefit after adjustment, consistent with the SETPOINT2 trial. 7 , 8 The Dominant Role of Prolonged MV and Hemorrhagic Transformation The most striking finding of our regression analysis is the near-complete dominance of prolonged MV (OR 62.3) as a predictor of tracheostomy in the ventilated population. This finding, while mathematically expected given the high collinearity between prolonged ventilatory dependence and the clinical decision to place a tracheostomy, provides important quantitative context: once a patient with AIS has required more than 24 hours of MV, the probability of ultimately requiring tracheostomy is more than 62-fold higher than among those with short-duration MV. This suggests that the tracheostomy decision in AIS is less about comorbidity profiling and more about the trajectory of ventilatory dependence, consistent with prior single-center observations in large vessel occlusion patients. 15 Hemorrhagic transformation emerged as the second strongest predictor in the MV-only model (OR 1.35) and the strongest overall predictor across all AIS patients (OR 5.19). This finding has not been previously highlighted in the tracheostomy literature at national scale. HT likely contributes to tracheostomy need through multiple mechanisms: secondary neurologic deterioration impairing consciousness and airway protective reflexes, hemorrhagic expansion requiring surgical intervention, and increased complications including seizures and elevated intracranial pressure. The observation that HT is present in 27.2% of tracheostomy patients — more than a fivefold higher rate than non-tracheostomy AIS patients (5.1%) — suggests that HT should be incorporated into future tracheostomy prediction models and discussed in goals-of-care contexts when HT is identified. The Mortality Paradox: Selection Bias in the MV-No-Tracheostomy Group The stark mortality difference between MV-no-tracheostomy patients (39.7%) and those who received tracheostomy (9.84% early, 8.23% late) is counterintuitive but well-recognized as a survivorship phenomenon. Patients who require MV but die before the tracheostomy decision threshold — typically days 7–10 of hospitalization — populate the MV-no-tracheostomy group, while tracheostomy patients represent survivors to that decision point. This is not a causal protective effect of tracheostomy but rather a reflection of the natural history of acute respiratory failure in severe stroke, where early in-hospital mortality concentrates in those too critically ill for elective airway procedures. Similar observations have been made in prior stroke-specific analyses 5 , 16 and in the general critical care literature on tracheostomy. Stroke-associated pneumonia complicates a substantial proportion of ventilated AIS patients and independently worsens mortality and length of stay, 17 likely contributing to the higher early hazard observed in this group. Tracheostomy Timing and the SETPOINT2 Context Our finding that early tracheostomy (≤ 7 days) was associated with higher adjusted in-hospital mortality but with significantly shorter LOS (23.8 vs 34.4 days, p < 0.001) is broadly consistent with the SETPOINT2 trial's conclusion that early tracheostomy does not improve functional outcome at 6 months. 7 The Premraj et al. meta-analysis of over 17,000 stroke patients similarly found no association between tracheostomy timing and mortality, ICU LOS, or neurological outcome. 8 A more recent meta-analysis of 19 RCTs and 3,586 critically ill patients similarly reported only a modest mortality benefit with early tracheostomy alongside a clear reduction in ICU length of stay, 18 paralleling the LOS pattern observed here. However, the substantial LOS difference in our data — approximately 10 days shorter for early tracheostomy — represents a clinically meaningful finding for resource utilization and hospital capacity planning, consistent with prior NIS-based observations by Villwock et al. 5 and Albert et al. 3 The decline in tracheostomy rates is most parsimoniously explained by a convergence of structural forces rather than any single cause. The dominant mechanism appears to be the dramatic expansion of mechanical thrombectomy — which rose from 15.3% to 22.2% of MV patients over the study period — enabling recanalization of large vessel occlusions, faster neurological recovery, and earlier extubation success. This is directly supported by the significant year-thrombectomy interaction (p = 0.027) and by the parallel decline in prolonged MV duration, both of which predate SETPOINT2 publication and indicate a structural shift in the ventilatory trajectory of AIS patients. COVID-19–era aerosol precaution protocols contributed a transient suppression of tracheostomy rates in 2020–2021, with partial rebound in 2022. The additional sharp 2023 decline — the first full calendar year after SETPOINT2 publication — is temporally consistent with a practice-level shift in tracheostomy thresholds informed by that trial’s negative findings, though causality cannot be established from administrative data. The most defensible interpretation is that the 2023 decline reflects a convergence: thrombectomy-driven earlier liberation from MV, improving ICU care practices, post-COVID normalization, and possibly growing clinician comfort with deferring or declining tracheostomy in borderline cases following SETPOINT2. Racial Disparities in Tracheostomy Use Black patients were disproportionately represented among tracheostomy recipients (27.8% of tracheostomy vs 15.8% of non-tracheostomy MV patients), a disparity that was attenuated after adjustment for demographics and comorbidities (adjusted OR 1.13 vs Asian/PI reference in Model B; p = 0.15) but remained significant in the overall AIS model (OR 1.28; 95% CI 1.10–1.49; p = 0.001). Hispanic patients were under-represented after adjustment (adjusted OR 0.739 vs Asian/PI; p = 0.002), while White patients were significantly under-represented (40.5% vs 57.3%; adjusted OR 0.557 in Model A; OR 0.571 in Model B). This multi-racial pattern of disparity likely reflects a convergence of mechanisms operating at biological, structural, and interpersonal levels. Biologically, Black patients with AIS carry a disproportionate burden of comorbidities independently associated with tracheostomy — including coagulopathy, hemorrhagic transformation, and liver disease — and present at younger ages with more severe neurological impairment. Structurally, Black and Hispanic patients have historically had lower rates of thrombectomy access, higher residual deficits after stroke, and greater post-acute care needs, all of which increase the probability of prolonged MV and subsequent tracheostomy. At the interpersonal level, documented differences in goals-of-care communication quality, surrogate decision-making dynamics, and implicit bias in prognostication may alter the threshold at which tracheostomy is offered or pursued across racial groups. Prior work has demonstrated that minority race is independently associated with non-beneficial gastrostomy tube placement following stroke, 19 and that Black patients with intracerebral hemorrhage are significantly less likely to undergo withdrawal of life-sustaining therapy than White patients despite comparable in-hospital outcomes. 20 These findings are consistent with broader patterns of racial inequity in neurocritical care — including documented differences in supplemental oxygen delivery attributable to pulse oximetry bias in non-White patients 21 — and represent an important and incompletely understood driver of healthcare resource utilization after severe stroke. Prospective studies incorporating neurological severity, surrogate preferences, and communication quality are needed to disentangle these mechanisms. PEG Co-Placement as a Marker of Anticipated Disability PEG tube co-placement in 77.2–78.5% of tracheostomy patients — compared with 10.6% in ventilated patients without tracheostomy — highlights that tracheostomy in AIS almost universally signals anticipated long-term disability with impaired oral intake. The DECAST study demonstrated that decannulation was achieved in only 59% of stroke patients surviving 12 months after tracheostomy, and was associated with better functional outcomes compared with permanent cannulation. 16 Our discharge destination data extend this observation further: 80.7–84.2% of tracheostomy patients were discharged to SNF or LTACH, and fewer than 5% reached home, compared with 13.4% home discharge among ventilated patients without tracheostomy. A recent multicenter neurocritical care study found that early tracheostomy was associated with a 40% lower likelihood of discharge to rehabilitation versus late tracheostomy, suggesting timing itself may shape post-acute trajectories beyond its effect on LOS. 10 Taken together, the tracheostomy–PEG–LTACH triad represents a distinct, high-intensity care trajectory carrying profound implications for patients, families, and health systems. Goals-of-care discussions should ideally occur prospectively — at the point when prolonged MV becomes apparent — rather than reactively after procedures are already placed, because both the tracheostomy and PEG decisions signal a clinical trajectory that is rarely reversed. Thrombectomy Subgroup Implications The lower tracheostomy rate among thrombectomy MV patients (10.9% vs 12.4%) and their markedly better outcomes when tracheostomy is performed (mortality 7.44% vs 8.87%) likely reflect purposeful patient selection: thrombectomy patients undergo a reversibility-guided triage in which tracheostomy is reserved for those with anticipated neurological recovery. This is clinically important context for interpreting the association between thrombectomy and MV observed in our companion paper on mechanical ventilation predictors — thrombectomy-related intubation encompasses both procedural airway management and genuine respiratory failure, and the two groups have substantially different prognostic trajectories. Limitations This study has limitations inherent to administrative data. The NIS does not capture neurologic severity scores such as the NIH Stroke Scale, infarct location or volume, or the clinical indications and context surrounding individual tracheostomy decisions. Without stroke severity data, residual confounding in both predictor and outcome models cannot be excluded; in particular, the adjusted LOS comparison between early and late tracheostomy groups should be interpreted with appropriate caution, as patients selected for early tracheostomy may differ systematically from those receiving late tracheostomy in ways not captured by available covariates. Procedure day data were missing in 2.9% of tracheostomy patients, limiting precision in timing analyses. Causal inference is precluded in this observational design, and the survivorship bias inherent in comparing tracheostomy versus non-tracheostomy ventilated patients is explicitly addressed in the Discussion. The NIS does not capture long-term functional outcomes, tracheostomy decannulation rates, or post-discharge survival — outcomes that the 2026 Neurocritical Care neuro-prognostication guidelines identify as priority endpoints for this population. 11 Hospital region was not available in the analytic dataset and was therefore excluded from multivariable models; residual geographic confounding cannot be excluded. ICD-10-CM coding for hospital-acquired complications such as pneumonia and sepsis may reflect documentation variation across institutions rather than true incidence differences. Importantly, the NIS does not include the SETscore or equivalent variables — consciousness level, swallowing function, and predicted MV duration at admission — that would enable direct comparison with SETPOINT2 eligibility criteria. The racial disparity findings, while robust after adjustment, do not permit causal attribution to specific mechanisms such as implicit bias, goals-of-care communication differences, or structural access inequities; disentangling these requires prospective study designs with qualitative components. Despite these limitations, the NIS provides a uniquely powered platform for characterizing the national tracheostomy landscape in AIS during and immediately surrounding the SETPOINT2 era, with sample sizes and temporal range unavailable in any prior AIS-specific study. Conclusions Tracheostomy occurs in approximately 12.1% of mechanically ventilated AIS patients and is characterized by prolonged hospitalization, near-universal gastrostomy co-placement, predominantly institutional discharge, and significantly declining national rates — from 12.9% to 9.44% of MV patients over 2016–2023 (Cochran-Armitage Z = − 6.54; p < 0.0001). Prolonged ventilation and hemorrhagic transformation are the dominant independent predictors among ventilated patients; decompressive craniectomy carries an adjusted OR of 2.55 for tracheostomy and should be incorporated into future prediction models. Early tracheostomy is associated with significantly shorter hospitalization but not survival benefit after adjustment, consistent with the SETPOINT2 trial and subsequent meta-analyses. The median procedure day drifted toward later placement during 2020–2022, most likely driven by COVID-19–era caution rather than SETPOINT2 dissemination, as that trial was published in May 2022 — our study extends through 2023, capturing the first full year post-publication. Racial disparities persist after risk adjustment: Black patients had 28% higher adjusted odds of tracheostomy than Asian/PI patients in the overall AIS model (Model A, p = 0.001), though this difference was attenuated and not statistically significant when restricted to MV patients (Model B OR 1.13; p = 0.15); addressing these disparities requires prospective investigation of goals-of-care communication quality, surrogate decision-making dynamics, and structural access inequities. The tracheostomy–PEG–LTACH care pathway represents a high-intensity clinical trajectory with profound implications for patients and families that warrants prospective goals-of-care integration. These findings provide the largest and most contemporary national framework for tracheostomy in AIS, directly informing neurocritical care decision-making, resource planning, and the design of future prospective studies on timing optimization and equity. Declarations Disclosures: The authors report no conflicts of interest. Data Source: National Inpatient Sample, HCUP, AHRQ (2016–2023). De-identified data; IRB waiver applied. 1. Compliance with Instructions for Authors The authors confirm that this manuscript complies with all instructions outlined in the Neurocritical Care Instructions for Authors, including formatting, word count, reference style, table and figure preparation, and submission requirements for an Original Work (Clinical Investigation). 2. Author Contributions Mian Urfy, MD, MSc: Conceptualization; study design; data acquisition; statistical analysis; interpretation of results; drafting of the manuscript; critical revision for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work. Mariam Tariq Mir, MD: Literature review; interpretation of results; critical revision of the manuscript for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work. Both authors meet all four ICMJE authorship criteria. 3. Authorship Requirements and Final Approval All listed authors meet the authorship criteria established by the International Committee of Medical Journal Editors (ICMJE): (1) substantial contributions to the conception or design of the work, or the acquisition, analysis, or interpretation of data; (2) drafting the work or revising it critically for important intellectual content; (3) final approval of the version to be published; and (4) agreement to be accountable for all aspects of the work. The final manuscript has been read and approved by all authors. 4. Originality and Exclusive Submission This manuscript has not been published, in whole or in part, in any other journal, and is not currently under consideration for publication elsewhere. No portion of the work has appeared in any prior publication, including preprint servers, abstracts, or conference proceedings, beyond what is permitted by the journal’s policies. 5. Ethical Approval and Informed Consent This is a retrospective observational analysis of de-identified administrative data from the Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample (NIS), 2016–2023. Per the HCUP Data Use Agreement and 45 CFR 46.102(e), this work does not constitute human subjects research because it involves only secondary analysis of fully de-identified, publicly available data. Accordingly, Institutional Review Board (IRB) review and informed consent were not required. The authors followed all HCUP methodological and reporting requirements, including cell-size suppression rules where applicable. 6. Conflicts of Interest The authors declare that they have no conflicts of interest, financial or otherwise, relevant to the subject of this manuscript. Neither author has any financial relationships, consulting arrangements, equity interests, patents, or other competing interests that could be perceived to influence the conduct or reporting of this work. ICMJE Disclosure forms have been completed by each author and are available upon request. 7. Reporting Checklist This study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for cohort studies. A completed STROBE checklist is included as a supplemental file with this submission. 8. Funding This work received no specific grant or financial support from any funding agency in the public, commercial, or not-for-profit sectors. No external funding was used in the design, conduct, analysis, or reporting of this study. This statement is consistent with the Funding Information entered in the Editorial Manager submission portal. Corresponding Author Mian Urfy, MD, MSc Associate Professor of Neurology Chicago Medical School, Rosalind Franklin University Advocate Lutheran General Hospital, Neurocritical Care Unit 1775 Dempster Street, Park Ridge, IL 60068 [email protected] References Lahiri S, Mayer SA, Fink ME, et al. Mechanical ventilation for acute stroke: a multi-state population-based study. Neurocrit Care. 2015;23(1):28–32. 10.1007/s12028-014-0082-9 . Mayer SA, Copeland D, Bernardini GL, et al. Cost and outcome of mechanical ventilation for life-threatening stroke. Stroke. 2000;31(10):2346–53. 10.1161/01.str.31.10.2346 . Albert GP, McHugh DC, Hwang DY, et al. National cost estimates of invasive mechanical ventilation and tracheostomy in acute stroke, 2008–2017. Stroke. 2023;54(10):2602–12. 10.1161/STROKEAHA.123.043176 . Walcott BP, Kamel H, Castro B, et al. Tracheostomy after severe ischemic stroke: a population-based study. J Stroke Cerebrovasc Dis. 2014;23(5):1024–9. 10.1016/j.jstrokecerebrovasdis.2013.08.019 . Villwock JA, Villwock MR, Deshaies EM. Tracheostomy timing affects stroke recovery. J Stroke Cerebrovasc Dis. 2014;23(5):1069–72. 10.1016/j.jstrokecerebrovasdis.2013.09.008 . Shah S, Spirollari E, Ng C, et al. Early tracheostomy in patients undergoing mechanical thrombectomy for acute ischemic stroke. J Crit Care. 2023;78:154357. 10.1016/j.jcrc.2023.154357 . Bösel J, Niesen WD, Salih F, et al. Effect of early vs standard approach to tracheostomy on functional outcome at 6 months among patients with severe stroke receiving mechanical ventilation: the SETPOINT2 randomized clinical trial. JAMA. 2022;327(19):1899–909. 10.1001/jama.2022.4798 . Premraj L, Camarda C, White N, et al. Tracheostomy timing and outcome in critically ill patients with stroke: a meta-analysis and meta-regression. Crit Care. 2023;27(1):132. 10.1186/s13054-023-04417-6 . Zhao CW, Hwang DY. US practitioner attitudes toward tracheostomy timing, benefits, risks, and techniques for severe stroke patients: a national survey and National Inpatient Sample analysis. Neurocrit Care. 2021;34(2):669–73. 10.1007/s12028-020-01127-7 . Amano MA, Bahouth MN, Kassalow B, et al. Timing of tracheostomy and association with ventilator liberation, length of stay, and discharge outcomes in neurocritical patients. Tracheostomy. 2025;2(1):56–73. 10.62905/001c.133992 . Mainali S, Fontaine GV, Rajajee V, et al. Guidelines for neuroprognostication in critically ill adults with acute ischemic stroke. Neurocrit Care. 2026. 10.1007/s12028-026-02486-3 . Agency for Healthcare Research and Quality. HCUP National Inpatient Sample (NIS). Healthcare Cost and Utilization Project (HCUP). Rockville, MD. 2024. https://hcup-us.ahrq.gov/nisoverview.jsp Loggini A, Qureshi AI, Saleh Velez FG, et al. Early tracheostomy in patients with nontraumatic intracerebral hemorrhage is associated with lower in-hospital complications and reduced resource use without increased mortality. Neurocrit Care. 2026. 10.1007/s12028-025-02436-5 . Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–9. 10.1097/01.mlr.0000182534.19832.83 . Maier IL, Schramm K, Bähr M, et al. Predictive factors for the need of tracheostomy in patients with large vessel occlusion stroke being treated with mechanical thrombectomy. Front Neurol. 2021;12:728624. 10.3389/fneur.2021.728624 . Schneider H, Hertel F, Kuhn M, et al. Decannulation and functional outcome after tracheostomy in patients with severe stroke (DECAST): a prospective observational study. Neurocrit Care. 2017;27(1):26–34. 10.1007/s12028-017-0390-y . Hannawi Y, Hannawi B, Rao CPV, Suarez JI, Bershad EM. Stroke-associated pneumonia: major advances and obstacles. Cerebrovasc Dis. 2013;35(5):430–43. 10.1159/000350199 . Maugeri R, Iacopino DG, Graziano F, et al. Timing of tracheostomy in ICU patients: a systematic review and meta-analysis of randomized controlled trials. Life (Basel). 2024;14(9):1165. 10.3390/life14091165 . Faigle R, Carrese JA, Cooper LA, Urrutia VC, Gottesman RF. Minority race and male sex as risk factors for non-beneficial gastrostomy tube placements after stroke. PLoS ONE. 2018;13(2):e0191293. 10.1371/journal.pone.0191293 . Ormseth CH, Falcone GJ, Jasak SD, et al. Minority patients are less likely to undergo withdrawal of care after spontaneous intracerebral hemorrhage. Neurocrit Care. 2018;29(3):419–25. 10.1007/s12028-018-0554-4 . Gottlieb ER, Ziegler J, Morley K, Rush B, Celi LA. Assessment of racial and ethnic differences in oxygen supplementation among patients in the intensive care unit. JAMA Intern Med. 2022;182(8):849–58. 10.1001/jamainternmed.2022.2587 . Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 03 May, 2026 Reviewers invited by journal 03 May, 2026 Editor invited by journal 03 May, 2026 Editor assigned by journal 03 May, 2026 First submitted to journal 27 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9546300","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":633756519,"identity":"5dc77d80-ab32-41ae-b080-1580c1cc4a11","order_by":0,"name":"Mian Urfy","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYDACZgjFz8DeAGMTqUWygecAsVoYYFokEojUotvO/OwBY9s9CYObjx9+LmCwyZd3IKDF7DCbuQFjW7GEwe00Y+kZDGmWGw8Q1MJgJsHYllBncDvBQJqH4bCBYQNBLezfQFqADjv++TeRWnjMIFpu8JiBbZEnoAOkpUwi4VyChOSZnDJrHoM0AwOCWs4f3ybxoSxBgu/48c23eSpsDOQJOQwMEoBY4QCIBbTC4AAxWkAAbjhxtoyCUTAKRsFIAgCwSTmdeJVRYQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0009-0003-1531-0140","institution":"Rosalind Franklin University of Medicine and Science","correspondingAuthor":true,"prefix":"","firstName":"Mian","middleName":"","lastName":"Urfy","suffix":""},{"id":633756520,"identity":"f6520a3c-24a4-4412-8780-bc5248159244","order_by":1,"name":"Mariam Tariq Mir","email":"","orcid":"","institution":"Advocate South Suburban Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mariam","middleName":"Tariq","lastName":"Mir","suffix":""}],"badges":[],"createdAt":"2026-04-27 22:17:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9546300/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9546300/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109100000,"identity":"f585ca52-c4c8-4406-bf47-bceccafefd7b","added_by":"auto","created_at":"2026-05-12 14:19:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":486084,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"Figure1preview.png","url":"https://assets-eu.researchsquare.com/files/rs-9546300/v1/4f96af8156b55fe65102b52a.png"},{"id":109099999,"identity":"90831086-c4ed-4472-8d57-3ebbebccaa8e","added_by":"auto","created_at":"2026-05-12 14:19:40","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":520141,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"Figure2preview.png","url":"https://assets-eu.researchsquare.com/files/rs-9546300/v1/0c784aa5f145cbe296a29c6c.png"},{"id":109099949,"identity":"3832cd4d-0eae-468f-9934-2f93115b98c6","added_by":"auto","created_at":"2026-05-12 14:19:29","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":287405,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"Figure3preview.png","url":"https://assets-eu.researchsquare.com/files/rs-9546300/v1/94b7899153b6d733788f6600.png"},{"id":109099730,"identity":"a8169c33-f16f-400e-ba49-02cb6fca3bc2","added_by":"auto","created_at":"2026-05-12 14:18:07","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":408004,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"Figure4preview2.png","url":"https://assets-eu.researchsquare.com/files/rs-9546300/v1/415b908f82075d59b1899526.png"},{"id":109099948,"identity":"90d2877f-824d-4aeb-890d-ae0d59c96553","added_by":"auto","created_at":"2026-05-12 14:19:29","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":571787,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"Figure5preview.png","url":"https://assets-eu.researchsquare.com/files/rs-9546300/v1/a072b0c247b96bc3bb5e4ba6.png"},{"id":109204770,"identity":"a7add5eb-9cf9-47dc-bf9e-36af553604b3","added_by":"auto","created_at":"2026-05-13 15:02:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2091250,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9546300/v1/2f28a52b-0b0b-4327-8c4d-eb4b4092a5bf.pdf"}],"financialInterests":"","formattedTitle":"Tracheostomy in Acute Ischemic Stroke: Declining National Utilization, Independent Predictors, and In-Hospital Outcomes Among 854,660 Hospitalizations","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMechanical ventilation (MV) complicates approximately 5% of acute ischemic stroke (AIS) hospitalizations and is strongly associated with increased morbidity, resource utilization, and mortality.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e Among patients who require MV, a critical and consequential subset will ultimately require tracheostomy\u0026mdash;either to facilitate ventilator weaning, provide long-term airway protection in the setting of impaired swallowing or consciousness, or manage complications such as aspiration pneumonia. Tracheostomy is among the most resource-intensive procedures performed in neurocritical care and is associated with prolonged hospitalizations, high costs, and uncertain long-term outcomes in the stroke population.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThe optimal timing of tracheostomy after stroke has been the subject of considerable debate and investigation. Observational studies using national inpatient databases established that early tracheostomy was associated with reduced ventilator-associated pneumonia, shorter length of stay, and lower hospital costs compared with late placement.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e These findings generated clinical enthusiasm for earlier tracheostomy practices in severe stroke. However, the landmark SETPOINT2 randomized clinical trial, which enrolled 382 patients with severe ischemic or hemorrhagic stroke at 26 neurocritical care centers, found that early tracheostomy (\u0026le;\u0026thinsp;5 days) did not improve functional outcome at 6 months compared with a standard approach (tracheostomy from day 10 if needed).\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e A subsequent meta-analysis of over 17,000 critically ill stroke patients confirmed that tracheostomy timing was not associated with mortality, neurological outcomes, or ICU or hospital length of stay.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e These findings have reframed the clinical question: while early tracheostomy may not improve functional recovery, its effects on short-term in-hospital outcomes such as LOS, complications, and resource utilization remain clinically important and practice-relevant.\u003c/p\u003e \u003cp\u003eContemporary nationally representative data characterizing tracheostomy in the AIS population are limited. Prior administrative database studies used data from 2007\u0026ndash;2017 and focused either on stroke-specific tracheostomy populations,\u003csup\u003e4\u003c/sup\u003e on thrombectomy-specific cohorts,\u003csup\u003e6\u003c/sup\u003e or on all-stroke populations combining ischemic and hemorrhagic subtypes.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e The period from 2016 onward represents a transformed era in AIS care, marked by widespread adoption of mechanical thrombectomy, expansion of thrombectomy-capable centers, and evolving neurocritical care practices. Studies examining tracheostomy specifically in AIS across this modern era, including the years during and after the COVID-19 pandemic, are lacking.\u003c/p\u003e \u003cp\u003eFurthermore, several important and under characterized aspects of tracheostomy in AIS warrant dedicated investigation. The role of hemorrhagic transformation (HT) as a driver of tracheostomy has not been systematically examined at national scale. The mechanistic pathway from prolonged MV to tracheostomy has not been quantified in large populations. Racial disparities in tracheostomy use after AIS have received little attention despite being well-documented in other neurocritical care populations. And the relationship between tracheostomy and co-procedures such as gastrostomy (PEG) placement \u0026mdash; a marker of anticipated prolonged disability \u0026mdash; has not been examined in this context. Notably, a 2021 national survey found no consensus on optimal tracheostomy timing in neurocritical patients, with clinicians reporting inconsistent practice driven by local culture rather than evidence.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e A recent multicenter study in neurocritical patients confirmed that early tracheostomy is associated with shorter LOS and lower total hospital costs in stroke patients, yet also found a 40% lower likelihood of discharge to rehabilitation in the late-tracheostomy group \u0026mdash; underscoring that the consequences of timing extend beyond mortality.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e The 2026 Neurocritical Care Society neuro-prognostication guidelines for critically ill AIS patients now explicitly emphasize tracheostomy decannulation prediction and goals-of-care counseling for large hemispheric infarction as clinical priorities,\u003csup\u003e11\u003c/sup\u003e creating an urgent need for contemporary population-level data to anchor these recommendations.\u003c/p\u003e \u003cp\u003eUsing a large, nationally representative cohort of 854,660 AIS hospitalizations from 2016\u0026ndash;2023, we aimed to: (1) characterize the prevalence and temporal trends of tracheostomy in contemporary AIS care; (2) identify predictors of tracheostomy among all AIS patients and among those specifically requiring MV, including the novel contribution of decompressive craniectomy; (3) examine the association between tracheostomy timing (early vs. late) and in-hospital outcomes including LOS, mortality, PEG placement, and discharge destination; (4) compare outcomes across four clinically defined groups \u0026mdash; no MV, MV without tracheostomy, early tracheostomy, and late tracheostomy; and (5) characterize racial and procedural subgroup differences in tracheostomy patterns. The cohort selection process is depicted in Fig.\u0026nbsp;1.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData Source\u003c/h2\u003e \u003cp\u003eThis retrospective observational study used data from the National Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP), for years 2016 through 2023. The NIS is the largest publicly available all-payer inpatient database in the United States, representing approximately 20% of all U.S. hospitalizations through a stratified sampling design. Discharge weights enable generation of nationally representative estimates. All analyses followed HCUP methodological guidance. The NIS contains fully de-identified data; institutional review board approval and informed consent were not required.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStudy Population\u003c/h3\u003e\n\u003cp\u003eHospitalizations with a principal diagnosis of acute ischemic stroke (AIS) were identified using ICD-10-CM codes I63.x. Patients aged 18 to 90 years were included. Exclusions were applied for missing age, sex, discharge disposition, or invalid sampling weights. The final analytic cohort comprised 854,660 unweighted AIS hospitalizations, representing approximately 4.27\u0026nbsp;million weighted discharges. The stepwise cohort selection process, exclusion criteria, and analytic group derivation are illustrated in Fig.\u0026nbsp;1. Mechanical ventilation (MV) was identified using ICD-10-PCS procedure codes 5A1935Z (\u0026lt;\u0026thinsp;24 hours), 5A1945Z (24\u0026ndash;96 hours), and 5A1955Z (\u0026gt;\u0026thinsp;96 hours).\u003c/p\u003e\n\u003ch3\u003eTracheostomy and Timing Classification\u003c/h3\u003e\n\u003cp\u003eTracheostomy was identified using ICD-10-PCS procedure codes for open (0B110F4, 0B110ZZ), percutaneous (0B113F4, 0B113ZZ), and percutaneous endoscopic (0B114F4, 0B114ZZ) approaches. Procedure day was extracted by matching tracheostomy procedure codes to corresponding PRDAY slots (PRDAY1 through PRDAY10). Tracheostomy timing was classified as early (\u0026le;\u0026thinsp;7 days from admission) or late (\u0026gt;\u0026thinsp;7 days), consistent with recent administrative database studies in the stroke literature.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e Patients with tracheostomy but missing procedure day data were classified separately. Gastrostomy tube placement (PEG) was identified using codes 0DH63UZ, 0DH64UZ, and 0DH60UZ.\u003c/p\u003e\n\u003ch3\u003eVariables\u003c/h3\u003e\n\u003cp\u003eDemographic variables included age, sex, and race/ethnicity. Comorbidities were derived using ICD-10-CM codes consistent with Elixhauser and Charlson classification systems: congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), atrial fibrillation (AFIB), chronic kidney disease (CKD), liver disease, coagulopathy, cancer, obesity, diabetes, and hypertension. Hemorrhagic transformation (HT) was identified by secondary ICD-10-CM codes I61.x or I62.x in the AIS cohort. Treatment exposures included mechanical thrombectomy (03CG3ZZ, 03CG4ZZ, 03CK3ZZ, 03CK4ZZ, 03CL3ZZ, 03CL4ZZ) and intravenous thrombolysis (3E033 series). In-hospital mortality was defined by discharge disposition code 20. Prolonged MV was defined as MV duration exceeding 24 hours (codes 5A1945Z or 5A1955Z).\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eDescriptive statistics compared tracheostomy and non-tracheostomy groups among all AIS patients (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) and among mechanically ventilated patients (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Chi-square tests were used for categorical variables and t-tests for continuous variables. Two multivariable logistic regression models were developed: Model A identified predictors of tracheostomy among all AIS patients; Model B identified predictors among MV patients only, additionally including prolonged MV as a predictor. Results are reported as adjusted odds ratios (OR) with 95% confidence intervals.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eBaseline demographic, clinical, and treatment characteristics stratified by tracheostomy status among all AIS patients (n\u0026thinsp;=\u0026thinsp;854,660). Values are number (%) or mean (SD).\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCharacteristic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo Trach (n\u0026thinsp;=\u0026thinsp;849,287)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTrach (n\u0026thinsp;=\u0026thinsp;5,373)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, mean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e69.9 (14.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e63.3 (13.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex \u0026mdash; Female\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e461,644 (49.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2,251 (41.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace \u0026mdash; White\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e486,480 (57.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2,174 (40.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace \u0026mdash; Black\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e124,842 (14.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1,481 (27.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace \u0026mdash; Hispanic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e60,986 (7.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e488 (9.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace \u0026mdash; Asian/PI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e22,347 (2.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e199 (3.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeekend Admission\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e237,754 (25.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,824 (26.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComorbidities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCongestive Heart Failure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e185,376 (19.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4,336 (29.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOPD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e135,809 (16.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e974 (18.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAtrial Fibrillation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e213,347 (25.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1,623 (30.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChronic Kidney Disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e192,065 (20.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,127 (21.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLiver Disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16,932 (2.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e299 (5.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoagulopathy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e50,201 (5.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e845 (15.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCancer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e42,121 (5.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e197 (3.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObesity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e139,638 (12.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2,315 (15.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e369,228 (39.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5,175 (35.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e723,897 (85.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,937 (73.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHemorrhagic Transformation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e43,206 (5.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1,447 (26.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTreatments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMechanical Thrombectomy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e32,004 (3.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e887 (16.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV tPA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e81,327 (9.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e647 (12.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMV (received)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e37,339 (4.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5,147 (95.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProlonged MV \u0026gt;\u0026thinsp;24h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e27,251 (3.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5,119 (95.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePEG Tube\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e22,572 (2.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4,103 (76.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLOS, mean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.09 (6.79)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e32.29 (26.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIn-hospital Mortality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e33,971 (4.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e484 (9.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003e*** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01; * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; NS\u0026thinsp;=\u0026thinsp;not significant. AIS\u0026thinsp;=\u0026thinsp;acute ischemic stroke; COPD\u0026thinsp;=\u0026thinsp;chronic obstructive pulmonary disease; CKD\u0026thinsp;=\u0026thinsp;chronic kidney disease; MV\u0026thinsp;=\u0026thinsp;mechanical ventilation; PEG\u0026thinsp;=\u0026thinsp;percutaneous endoscopic gastrostomy; LOS\u0026thinsp;=\u0026thinsp;length of stay.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003e\u003cem\u003eCharacteristics stratified by tracheostomy status among mechanically ventilated AIS patients only (n\u0026thinsp;=\u0026thinsp;42,486). Selected variables shown.\u003c/em\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCharacteristic\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNo Trach (n\u0026thinsp;=\u0026thinsp;37,339)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTrach (n\u0026thinsp;=\u0026thinsp;5,147)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ep\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge, mean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e68.5 (14.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e63.3 (13.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSex \u0026mdash; Female\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e18,387 (49.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2,151 (41.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRace \u0026mdash; White\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,683 (52.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2,083 (40.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRace \u0026mdash; Black\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e5,896 (15.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1,432 (27.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCongestive Heart Failure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10,471 (28.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1,418 (27.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAtrial Fibrillation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e13,889 (37.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1,568 (30.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCoagulopathy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4,881 (13.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e811 (15.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLiver Disease\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1,765 (4.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e285 (5.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNS\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHemorrhagic Transformation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7,576 (20.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1,400 (27.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMechanical Thrombectomy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7,052 (18.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e861 (16.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eProlonged MV \u0026gt;\u0026thinsp;24h\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e27,251 (73.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e5,119 (99.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePEG Tube\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e3,962 (10.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4,029 (78.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLOS, mean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10.8 (13.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e32.1 (25.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIn-hospital Mortality\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e14,822 (39.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e463 (9.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003e*** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; NS\u0026thinsp;=\u0026thinsp;not significant. Table restricted to selected clinically relevant variables. Full comparison available in supplementary material.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eOutcomes were compared across four clinically defined groups: no MV, MV without tracheostomy, early tracheostomy (\u0026le;\u0026thinsp;7 days), and late tracheostomy (\u0026gt;\u0026thinsp;7 days). Between-group outcome comparisons used chi-square tests for proportions and t-tests for continuous variables. Adjusted mortality models compared early versus late tracheostomy, controlling for age, sex, and key comorbidities. Decompressive craniectomy was identified using ICD-10-PCS codes for skull excision (0NB series) and release of cerebral meninges (00N series), consistent with prior administrative database methodology. This variable was incorporated as an additional predictor in a pre-specified sensitivity model (Model B-extended) to quantify its independent contribution to tracheostomy risk among MV patients. Discharge disposition was categorized as: home or home health, skilled nursing facility (SNF) or long-term acute care hospital (LTACH), short-term hospital, or died. In-hospital complications including hospital-acquired pneumonia, sepsis, and acute kidney injury (AKI) were identified by secondary ICD-10-CM diagnosis codes. Temporal trends in tracheostomy rates were assessed using logistic regression with calendar year as a continuous predictor among MV patients; a Cochran-Armitage trend test was performed as a confirmatory analysis. To investigate potential mechanisms underlying the observed temporal decline in tracheostomy utilization, we performed a pre-specified mechanistic analysis testing five hypotheses: (1) rising mechanical thrombectomy utilization; (2) falling overall MV rates; (3) falling prolonged MV rates among MV patients; (4) case-mix severity shifts; and (5) COVID-19 confounding. Cochran-Armitage trend tests were applied to annual proportions for each hypothesis. A logistic interaction model (TRACH\u0026thinsp;~\u0026thinsp;YEAR \u0026times; THROMBECTOMY\u0026thinsp;+\u0026thinsp;covariates) tested whether tracheostomy rates declined differentially by thrombectomy status. All analyses were performed in R version 4.5.2.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eCohort and Tracheostomy Prevalence\u003c/h2\u003e \u003cp\u003eThe final analytic cohort included 854,660 AIS hospitalizations, of which 42,486 (5.0%) required MV. Tracheostomy was performed in 5,373 hospitalizations (0.63% of all AIS; 12.1% of MV patients). Among non-MV patients, tracheostomy was exceedingly rare (n\u0026thinsp;=\u0026thinsp;226, 0.03%), confirming that tracheostomy in AIS is almost exclusively a phenomenon of the ventilated population. Among the 5,373 tracheostomy cases, 5,147 (95.8%) had concurrent MV documentation.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePatient Characteristics\u003c/h3\u003e\n\u003cp\u003eTracheostomy patients were substantially younger than non-tracheostomy AIS patients (mean 63.3\u0026thinsp;\u0026plusmn;\u0026thinsp;13.6 vs 69.9\u0026thinsp;\u0026plusmn;\u0026thinsp;14.0 years; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; SMD 0.482) and more frequently male (58.1% vs 50.6%; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Pronounced racial differences were observed: Black patients comprised 27.6% of tracheostomy cases compared with 14.7% of non-tracheostomy cases (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and White patients were underrepresented in the tracheostomy group (40.5% vs 57.3%; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; SMD 0.358). Among MV patients, coagulopathy was present in 15.8% of tracheostomy versus 13.1% of non-tracheostomy patients, liver disease in 5.5% versus 4.7%, and CHF in 27.6% versus 28.0% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Hemorrhagic transformation was identified in 27.2% of tracheostomy patients compared with 20.3% of non-tracheostomy MV patients (SMD 0.163; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Among ventilated patients, 99.5% of tracheostomy recipients had prolonged MV (\u0026gt;\u0026thinsp;24h) compared with 73.0% of ventilated patients without tracheostomy (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eTracheostomy Timing\u003c/h2\u003e \u003cp\u003eAmong 5,217 tracheostomy patients with documented procedure day data, the median procedure day was 11 (IQR 8\u0026ndash;16). Early tracheostomy (\u0026le;\u0026thinsp;7 days) was performed in 1,146 patients (22.0%) and late tracheostomy (\u0026gt;\u0026thinsp;7 days) in 4,071 patients (78.0%). Procedure day data were missing in 156 patients (2.9%). The predominance of late tracheostomy is consistent with prevailing practice patterns in which tracheostomy is considered after failed weaning attempts typically beginning around day 7\u0026ndash;10 after intubation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eTracheostomy Rate by MV Duration\u003c/h2\u003e \u003cp\u003eTracheostomy rates increased markedly with MV duration. Among patients with short-duration MV (\u0026lt;\u0026thinsp;24h), tracheostomy was performed in 0.33%; in the 24\u0026ndash;96h group, 1.36%; and in patients with prolonged MV (\u0026gt;\u0026thinsp;96h), 29.1%. This gradient confirms that tracheostomy in AIS is largely a function of ventilatory dependence rather than decision-making at the time of initial intubation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePredictors of Tracheostomy\u003c/h2\u003e \u003cp\u003eIn Model A (all AIS patients; Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), hemorrhagic transformation demonstrated the strongest independent association with tracheostomy (adjusted OR 5.19; 95% CI 4.86\u0026ndash;5.54; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), followed by mechanical thrombectomy (OR 3.04; 95% CI 2.81\u0026ndash;3.28), coagulopathy (OR 2.18; 95% CI 2.02\u0026ndash;2.36), liver disease (OR 2.05; 95% CI 1.81\u0026ndash;2.32), and CHF (OR 1.58; 95% CI 1.48\u0026ndash;1.68). Age was strongly and inversely associated (OR 0.972 per year; 95% CI 0.970\u0026ndash;0.974; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Cancer (OR 0.665; 95% CI 0.573\u0026ndash;0.768) and White race (OR 0.557; 95% CI 0.482\u0026ndash;0.648 vs Asian/PI reference) were independently associated with lower odds of tracheostomy. IV tPA was not independently associated with tracheostomy in Model A (OR 1.01; 95% CI 0.93\u0026ndash;1.09; NS), and obesity was likewise not significant (OR 1.03; 95% CI 0.96\u0026ndash;1.10; NS).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eMultivariable logistic regression predictors of tracheostomy: Model A (all AIS patients) and Model B (mechanically ventilated patients only). Selected significant predictors shown.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModel A OR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eModel B OR \u0026dagger;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (per year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.972\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.970\u0026ndash;0.974\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.981\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.979\u0026ndash;0.983\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale (vs Male)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.844\u0026ndash;0.903\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.939\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.904\u0026ndash;0.976\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace \u0026mdash; Black (vs Asian/PI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.10\u0026ndash;1.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.96\u0026ndash;1.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace \u0026mdash; White (vs Asian/PI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.557\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.482\u0026ndash;0.648\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.571\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.484\u0026ndash;0.677\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComorbidities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHemorrhagic transformation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.86\u0026ndash;5.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.26\u0026ndash;1.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoagulopathy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.02\u0026ndash;2.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.01\u0026ndash;1.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e.028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLiver disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.81\u0026ndash;2.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.88\u0026ndash;1.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCHF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.48\u0026ndash;1.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.88\u0026ndash;1.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOPD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.14\u0026ndash;1.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.97\u0026ndash;1.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAtrial Fibrillation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.29\u0026ndash;1.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.932\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.869\u0026ndash;1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCKD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.907\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.843\u0026ndash;0.976\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.864\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.796\u0026ndash;0.936\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObesity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.96\u0026ndash;1.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.07\u0026ndash;1.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.02\u0026ndash;1.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.04\u0026ndash;1.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTreatments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMechanical thrombectomy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.81\u0026ndash;3.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.955\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.88\u0026ndash;1.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV tPA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.93\u0026ndash;1.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.786\u0026ndash;0.945\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProlonged MV \u0026gt;\u0026thinsp;24h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e62.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e43.9\u0026ndash;92.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eModel A\u0026thinsp;=\u0026thinsp;all AIS patients (n\u0026thinsp;=\u0026thinsp;854,660). Model B\u0026thinsp;=\u0026thinsp;mechanically ventilated AIS patients only (n\u0026thinsp;=\u0026thinsp;42,486); additionally includes prolonged MV as predictor. Reference category for race\u0026thinsp;=\u0026thinsp;Asian/Pacific Islander. NS\u0026thinsp;=\u0026thinsp;not significant. OR\u0026thinsp;=\u0026thinsp;adjusted odds ratio; CI\u0026thinsp;=\u0026thinsp;confidence interval. Note the directional reversal of comorbidity ORs between Models A and B \u0026mdash; see Discussion.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn Model B (MV patients only; Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), the predictor structure changed substantially. Prolonged MV (\u0026gt;\u0026thinsp;24h) was overwhelmingly the dominant predictor (OR 62.3; 95% CI 43.9\u0026ndash;92.6; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), reflecting the near-deterministic relationship between extended ventilatory dependence and the tracheostomy decision. Among clinical variables, hemorrhagic transformation remained independently associated (OR 1.35; 95% CI 1.26\u0026ndash;1.44; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Notably, traditional comorbidities including CHF (OR 0.95; NS), liver disease (OR 1.00; NS), and CKD (OR 0.864) no longer showed independent positive associations in the MV-only model\u0026mdash;reflecting that among ventilated patients, those with severe systemic comorbidity die before reaching the tracheostomy decision threshold rather than surviving to require it. Age remained independently inversely associated (OR 0.981 per year; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In the pre-specified sensitivity analysis incorporating decompressive craniectomy (Model B-extended), this procedure was independently and strongly associated with tracheostomy (OR 2.55; 95% CI 2.31\u0026ndash;2.80; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), confirming at national scale that the decompressive craniectomy pathway in AIS is tightly coupled with subsequent tracheostomy placement \u0026mdash; an observation previously documented only in single-center large vessel occlusion cohorts.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eOutcomes by Tracheostomy Timing Group\u003c/h2\u003e \u003cp\u003eOutcomes differed substantially across the four clinical groups (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Figs.\u0026nbsp;3 and 4). In-hospital mortality was lowest in the no-MV group (2.32%) and highest in the MV-no-tracheostomy group (39.7%), reflecting the severity of acute illness in those who die before the tracheostomy threshold is reached. Mortality was 9.84% in the early tracheostomy group and 8.23% in the late tracheostomy group (unadjusted p\u0026thinsp;=\u0026thinsp;0.007). After adjustment for age, sex, and comorbidities, early tracheostomy was associated with significantly higher mortality compared with late tracheostomy (adjusted OR 1.36; 95% CI 1.08\u0026ndash;1.70; p\u0026thinsp;=\u0026thinsp;0.007) \u0026mdash; consistent with confounding by indication, in which patients selected for early procedure are systematically sicker.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eIn-hospital outcomes by ventilation and tracheostomy status group.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOutcome\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo MV (n\u0026thinsp;=\u0026thinsp;812,174)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMV, No Trach (n\u0026thinsp;=\u0026thinsp;37,339)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEarly Trach \u0026le;7d (n\u0026thinsp;=\u0026thinsp;1,146)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLate Trach \u0026gt;7d (n\u0026thinsp;=\u0026thinsp;4,071)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLOS, mean (SD), days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.8 (6.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10.8 (13.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.8 (22.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e34.4 (26.2)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIn-hospital mortality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.32%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e39.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e8.23%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePEG placement\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e77.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.5%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDischarge home\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e54.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e13.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.9%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDischarge to SNF/LTACH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e39.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e41.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e80.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e84.2%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eAll between-group differences p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 except early vs. late tracheostomy mortality (unadjusted p\u0026thinsp;=\u0026thinsp;0.007; adjusted OR 1.36, 95% CI 1.08\u0026ndash;1.70, p\u0026thinsp;=\u0026thinsp;0.007 \u0026mdash; indicating higher adjusted mortality with early tracheostomy, consistent with confounding by indication) and PEG co-placement between early and late tracheostomy (p\u0026thinsp;=\u0026thinsp;0.38). LOS difference between early and late tracheostomy: mean 10.6 days (95% CI 9.3\u0026ndash;12.4; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001); adjusted LOS ratio 0.662 (95% CI 0.638\u0026ndash;0.687; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Early\u0026thinsp;=\u0026thinsp;\u0026le;\u0026thinsp;7 days from admission; Late\u0026thinsp;=\u0026thinsp;\u0026gt;\u0026thinsp;7 days. SNF\u0026thinsp;=\u0026thinsp;skilled nursing facility; LTACH\u0026thinsp;=\u0026thinsp;long-term acute care hospital. MV\u0026thinsp;=\u0026thinsp;mechanical ventilation; PEG\u0026thinsp;=\u0026thinsp;percutaneous endoscopic gastrostomy. Discharge home includes home health care. Decompressive craniectomy was performed in 16.3% of tracheostomy and 4.4% of non-tracheostomy MV patients (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eLength of stay differed significantly across groups. Early tracheostomy patients had a mean LOS of 23.8\u0026thinsp;\u0026plusmn;\u0026thinsp;22.9 days, compared with 34.4\u0026thinsp;\u0026plusmn;\u0026thinsp;26.2 days for late tracheostomy (mean difference 10.6 days; 95% CI 9.3\u0026ndash;12.4 days; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). PEG tube co-placement was present in 77.2% of early and 78.5% of late tracheostomy patients, compared with 10.6% in ventilated patients without tracheostomy and 2.3% in the no-MV group, indicating that the majority of tracheostomy patients are expected to require long-term enteral nutrition. Discharge destination paralleled these findings: 85.5% of early tracheostomy and 87.7% of late tracheostomy patients were discharged to institutional settings (skilled nursing facility, long-term acute care hospital, or inpatient rehabilitation), compared with 46.5% of mechanically ventilated patients without tracheostomy. Only 4.5% and 3.9% of early and late tracheostomy patients, respectively, were discharged home or to home health care, reinforcing the interpretation that tracheostomy in AIS signals anticipated long-term disability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eTemporal Trends\u003c/h2\u003e \u003cp\u003eTracheostomy rates among MV patients declined significantly over the study period, from 12.9% in 2016 to 9.44% in 2023 (Cochran-Armitage Z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;6.54, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001; logistic trend OR per year 0.958; 95% CI 0.946\u0026ndash;0.970; Fig.\u0026nbsp;2). The steepest single-year decline occurred from 2022 to 2023 (\u0026minus;\u0026thinsp;2.78 percentage points). Median procedure day was stable at Day 11 in 2016\u0026ndash;2020 and shifted to Day 12 in 2021\u0026ndash;2023. Mechanistic analyses identified convergent structural explanations for the declining tracheostomy rate. The overall MV rate rose significantly over the study period (4.44% in 2016 to 5.04% in 2023; Cochran-Armitage Z\u0026thinsp;=\u0026thinsp;+\u0026thinsp;11.4, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), excluding fewer severely ill patients reaching the ICU as an explanation. However, among already-ventilated patients, the proportion requiring prolonged MV (\u0026gt;\u0026thinsp;24 hours) declined significantly from 78.5% in 2016 to 73.9% in 2023 (Z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;6.77, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), and the proportion requiring\u0026thinsp;\u0026gt;\u0026thinsp;96 hours of ventilation reached a dataset low of 34.6% in 2023 \u0026mdash; indicating that a growing proportion of AIS patients are being extubated before reaching the tracheostomy decision threshold. Mechanical thrombectomy utilization rose substantially over the study period. Among MV patients, the thrombectomy rate increased from 15.3% in 2016 to 22.2% in 2023 (Cochran-Armitage Z\u0026thinsp;=\u0026thinsp;+\u0026thinsp;6.39, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001); across all AIS patients, thrombectomy utilization expanded from 2.86% to 4.77% (Z\u0026thinsp;=\u0026thinsp;+\u0026thinsp;20.9, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). A significant year-thrombectomy interaction (OR 0.960 per year, 95% CI 0.926\u0026ndash;0.995; p\u0026thinsp;=\u0026thinsp;0.027) confirmed that tracheostomy rates declined faster in thrombectomy-treated patients than in non-thrombectomy patients (7.28% vs 10.1% in 2023), consistent with thrombectomy-mediated recanalization enabling earlier neurological recovery and earlier extubation success. COVID-19 contributed a transient reduction in tracheostomy rates during 2020\u0026ndash;2021, with partial rebound in 2022; the 2023 decline therefore represents a new post-COVID structural change. In-hospital mortality among MV patients who did not receive tracheostomy fell from 43.4% in 2016 to 36.6% in 2023, further supporting a shift toward a population with higher rates of successful ventilator liberation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eThrombectomy Subgroup\u003c/h2\u003e \u003cp\u003eAmong MV patients, tracheostomy rates differed by thrombectomy status (10.9% in thrombectomy vs 12.4% in non-thrombectomy patients; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Thrombectomy patients who did receive tracheostomy had shorter LOS (30.7 vs 32.4 days) and markedly lower mortality (7.44% vs 8.87%), likely reflecting better baseline neurologic reserve and purposeful patient selection for tracheostomy among those with perceived recovery potential.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this nationally representative analysis of 854,660 AIS hospitalizations from 2016 to 2023, we demonstrate that tracheostomy occurs in 0.63% of all AIS patients and 12.1% of those requiring mechanical ventilation. These figures are consistent with prior national estimates spanning earlier eras\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e and provide the first national mechanistic analysis of declining tracheostomy rates in contemporary AIS care. Our findings extend prior work in several important dimensions: we demonstrate that the declining tracheostomy trend is driven primarily by upstream changes in ventilatory trajectory \u0026mdash; specifically, rising thrombectomy utilization enabling earlier extubation and a structural decline in prolonged MV duration \u0026mdash; rather than a shift in the tracheostomy decision itself; we characterize a mechanistic hierarchy of tracheostomy predictors; we document marked racial disparities in tracheostomy use; and we confirm that early tracheostomy is associated with shorter LOS but not survival benefit after adjustment, consistent with the SETPOINT2 trial.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eThe Dominant Role of Prolonged MV and Hemorrhagic Transformation\u003c/h2\u003e \u003cp\u003eThe most striking finding of our regression analysis is the near-complete dominance of prolonged MV (OR 62.3) as a predictor of tracheostomy in the ventilated population. This finding, while mathematically expected given the high collinearity between prolonged ventilatory dependence and the clinical decision to place a tracheostomy, provides important quantitative context: once a patient with AIS has required more than 24 hours of MV, the probability of ultimately requiring tracheostomy is more than 62-fold higher than among those with short-duration MV. This suggests that the tracheostomy decision in AIS is less about comorbidity profiling and more about the trajectory of ventilatory dependence, consistent with prior single-center observations in large vessel occlusion patients.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eHemorrhagic transformation emerged as the second strongest predictor in the MV-only model (OR 1.35) and the strongest overall predictor across all AIS patients (OR 5.19). This finding has not been previously highlighted in the tracheostomy literature at national scale. HT likely contributes to tracheostomy need through multiple mechanisms: secondary neurologic deterioration impairing consciousness and airway protective reflexes, hemorrhagic expansion requiring surgical intervention, and increased complications including seizures and elevated intracranial pressure. The observation that HT is present in 27.2% of tracheostomy patients \u0026mdash; more than a fivefold higher rate than non-tracheostomy AIS patients (5.1%) \u0026mdash; suggests that HT should be incorporated into future tracheostomy prediction models and discussed in goals-of-care contexts when HT is identified.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eThe Mortality Paradox: Selection Bias in the MV-No-Tracheostomy Group\u003c/h2\u003e \u003cp\u003eThe stark mortality difference between MV-no-tracheostomy patients (39.7%) and those who received tracheostomy (9.84% early, 8.23% late) is counterintuitive but well-recognized as a survivorship phenomenon. Patients who require MV but die before the tracheostomy decision threshold \u0026mdash; typically days 7\u0026ndash;10 of hospitalization \u0026mdash; populate the MV-no-tracheostomy group, while tracheostomy patients represent survivors to that decision point. This is not a causal protective effect of tracheostomy but rather a reflection of the natural history of acute respiratory failure in severe stroke, where early in-hospital mortality concentrates in those too critically ill for elective airway procedures. Similar observations have been made in prior stroke-specific analyses\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e and in the general critical care literature on tracheostomy. Stroke-associated pneumonia complicates a substantial proportion of ventilated AIS patients and independently worsens mortality and length of stay,\u003csup\u003e17\u003c/sup\u003e likely contributing to the higher early hazard observed in this group.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eTracheostomy Timing and the SETPOINT2 Context\u003c/h2\u003e \u003cp\u003eOur finding that early tracheostomy (\u0026le;\u0026thinsp;7 days) was associated with higher adjusted in-hospital mortality but with significantly shorter LOS (23.8 vs 34.4 days, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) is broadly consistent with the SETPOINT2 trial's conclusion that early tracheostomy does not improve functional outcome at 6 months.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e The Premraj et al. meta-analysis of over 17,000 stroke patients similarly found no association between tracheostomy timing and mortality, ICU LOS, or neurological outcome.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e A more recent meta-analysis of 19 RCTs and 3,586 critically ill patients similarly reported only a modest mortality benefit with early tracheostomy alongside a clear reduction in ICU length of stay,\u003csup\u003e18\u003c/sup\u003e paralleling the LOS pattern observed here. However, the substantial LOS difference in our data \u0026mdash; approximately 10 days shorter for early tracheostomy \u0026mdash; represents a clinically meaningful finding for resource utilization and hospital capacity planning, consistent with prior NIS-based observations by Villwock et al.\u003csup\u003e5\u003c/sup\u003e and Albert et al.\u003csup\u003e3\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThe decline in tracheostomy rates is most parsimoniously explained by a convergence of structural forces rather than any single cause. The dominant mechanism appears to be the dramatic expansion of mechanical thrombectomy \u0026mdash; which rose from 15.3% to 22.2% of MV patients over the study period \u0026mdash; enabling recanalization of large vessel occlusions, faster neurological recovery, and earlier extubation success. This is directly supported by the significant year-thrombectomy interaction (p\u0026thinsp;=\u0026thinsp;0.027) and by the parallel decline in prolonged MV duration, both of which predate SETPOINT2 publication and indicate a structural shift in the ventilatory trajectory of AIS patients. COVID-19\u0026ndash;era aerosol precaution protocols contributed a transient suppression of tracheostomy rates in 2020\u0026ndash;2021, with partial rebound in 2022. The additional sharp 2023 decline \u0026mdash; the first full calendar year after SETPOINT2 publication \u0026mdash; is temporally consistent with a practice-level shift in tracheostomy thresholds informed by that trial\u0026rsquo;s negative findings, though causality cannot be established from administrative data. The most defensible interpretation is that the 2023 decline reflects a convergence: thrombectomy-driven earlier liberation from MV, improving ICU care practices, post-COVID normalization, and possibly growing clinician comfort with deferring or declining tracheostomy in borderline cases following SETPOINT2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eRacial Disparities in Tracheostomy Use\u003c/h2\u003e \u003cp\u003eBlack patients were disproportionately represented among tracheostomy recipients (27.8% of tracheostomy vs 15.8% of non-tracheostomy MV patients), a disparity that was attenuated after adjustment for demographics and comorbidities (adjusted OR 1.13 vs Asian/PI reference in Model B; p\u0026thinsp;=\u0026thinsp;0.15) but remained significant in the overall AIS model (OR 1.28; 95% CI 1.10\u0026ndash;1.49; p\u0026thinsp;=\u0026thinsp;0.001). Hispanic patients were under-represented after adjustment (adjusted OR 0.739 vs Asian/PI; p\u0026thinsp;=\u0026thinsp;0.002), while White patients were significantly under-represented (40.5% vs 57.3%; adjusted OR 0.557 in Model A; OR 0.571 in Model B). This multi-racial pattern of disparity likely reflects a convergence of mechanisms operating at biological, structural, and interpersonal levels. Biologically, Black patients with AIS carry a disproportionate burden of comorbidities independently associated with tracheostomy \u0026mdash; including coagulopathy, hemorrhagic transformation, and liver disease \u0026mdash; and present at younger ages with more severe neurological impairment. Structurally, Black and Hispanic patients have historically had lower rates of thrombectomy access, higher residual deficits after stroke, and greater post-acute care needs, all of which increase the probability of prolonged MV and subsequent tracheostomy. At the interpersonal level, documented differences in goals-of-care communication quality, surrogate decision-making dynamics, and implicit bias in prognostication may alter the threshold at which tracheostomy is offered or pursued across racial groups. Prior work has demonstrated that minority race is independently associated with non-beneficial gastrostomy tube placement following stroke,\u003csup\u003e19\u003c/sup\u003e and that Black patients with intracerebral hemorrhage are significantly less likely to undergo withdrawal of life-sustaining therapy than White patients despite comparable in-hospital outcomes.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e These findings are consistent with broader patterns of racial inequity in neurocritical care \u0026mdash; including documented differences in supplemental oxygen delivery attributable to pulse oximetry bias in non-White patients\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e \u0026mdash; and represent an important and incompletely understood driver of healthcare resource utilization after severe stroke. Prospective studies incorporating neurological severity, surrogate preferences, and communication quality are needed to disentangle these mechanisms.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003ePEG Co-Placement as a Marker of Anticipated Disability\u003c/h2\u003e \u003cp\u003ePEG tube co-placement in 77.2\u0026ndash;78.5% of tracheostomy patients \u0026mdash; compared with 10.6% in ventilated patients without tracheostomy \u0026mdash; highlights that tracheostomy in AIS almost universally signals anticipated long-term disability with impaired oral intake. The DECAST study demonstrated that decannulation was achieved in only 59% of stroke patients surviving 12 months after tracheostomy, and was associated with better functional outcomes compared with permanent cannulation.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e Our discharge destination data extend this observation further: 80.7\u0026ndash;84.2% of tracheostomy patients were discharged to SNF or LTACH, and fewer than 5% reached home, compared with 13.4% home discharge among ventilated patients without tracheostomy. A recent multicenter neurocritical care study found that early tracheostomy was associated with a 40% lower likelihood of discharge to rehabilitation versus late tracheostomy, suggesting timing itself may shape post-acute trajectories beyond its effect on LOS.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Taken together, the tracheostomy\u0026ndash;PEG\u0026ndash;LTACH triad represents a distinct, high-intensity care trajectory carrying profound implications for patients, families, and health systems. Goals-of-care discussions should ideally occur prospectively \u0026mdash; at the point when prolonged MV becomes apparent \u0026mdash; rather than reactively after procedures are already placed, because both the tracheostomy and PEG decisions signal a clinical trajectory that is rarely reversed.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eThrombectomy Subgroup Implications\u003c/h2\u003e \u003cp\u003eThe lower tracheostomy rate among thrombectomy MV patients (10.9% vs 12.4%) and their markedly better outcomes when tracheostomy is performed (mortality 7.44% vs 8.87%) likely reflect purposeful patient selection: thrombectomy patients undergo a reversibility-guided triage in which tracheostomy is reserved for those with anticipated neurological recovery. This is clinically important context for interpreting the association between thrombectomy and MV observed in our companion paper on mechanical ventilation predictors \u0026mdash; thrombectomy-related intubation encompasses both procedural airway management and genuine respiratory failure, and the two groups have substantially different prognostic trajectories.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThis study has limitations inherent to administrative data. The NIS does not capture neurologic severity scores such as the NIH Stroke Scale, infarct location or volume, or the clinical indications and context surrounding individual tracheostomy decisions. Without stroke severity data, residual confounding in both predictor and outcome models cannot be excluded; in particular, the adjusted LOS comparison between early and late tracheostomy groups should be interpreted with appropriate caution, as patients selected for early tracheostomy may differ systematically from those receiving late tracheostomy in ways not captured by available covariates. Procedure day data were missing in 2.9% of tracheostomy patients, limiting precision in timing analyses. Causal inference is precluded in this observational design, and the survivorship bias inherent in comparing tracheostomy versus non-tracheostomy ventilated patients is explicitly addressed in the Discussion. The NIS does not capture long-term functional outcomes, tracheostomy decannulation rates, or post-discharge survival \u0026mdash; outcomes that the 2026 Neurocritical Care neuro-prognostication guidelines identify as priority endpoints for this population.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e Hospital region was not available in the analytic dataset and was therefore excluded from multivariable models; residual geographic confounding cannot be excluded. ICD-10-CM coding for hospital-acquired complications such as pneumonia and sepsis may reflect documentation variation across institutions rather than true incidence differences. Importantly, the NIS does not include the SETscore or equivalent variables \u0026mdash; consciousness level, swallowing function, and predicted MV duration at admission \u0026mdash; that would enable direct comparison with SETPOINT2 eligibility criteria. The racial disparity findings, while robust after adjustment, do not permit causal attribution to specific mechanisms such as implicit bias, goals-of-care communication differences, or structural access inequities; disentangling these requires prospective study designs with qualitative components. Despite these limitations, the NIS provides a uniquely powered platform for characterizing the national tracheostomy landscape in AIS during and immediately surrounding the SETPOINT2 era, with sample sizes and temporal range unavailable in any prior AIS-specific study.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eTracheostomy occurs in approximately 12.1% of mechanically ventilated AIS patients and is characterized by prolonged hospitalization, near-universal gastrostomy co-placement, predominantly institutional discharge, and significantly declining national rates \u0026mdash; from 12.9% to 9.44% of MV patients over 2016\u0026ndash;2023 (Cochran-Armitage Z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;6.54; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Prolonged ventilation and hemorrhagic transformation are the dominant independent predictors among ventilated patients; decompressive craniectomy carries an adjusted OR of 2.55 for tracheostomy and should be incorporated into future prediction models. Early tracheostomy is associated with significantly shorter hospitalization but not survival benefit after adjustment, consistent with the SETPOINT2 trial and subsequent meta-analyses. The median procedure day drifted toward later placement during 2020\u0026ndash;2022, most likely driven by COVID-19\u0026ndash;era caution rather than SETPOINT2 dissemination, as that trial was published in May 2022 \u0026mdash; our study extends through 2023, capturing the first full year post-publication. Racial disparities persist after risk adjustment: Black patients had 28% higher adjusted odds of tracheostomy than Asian/PI patients in the overall AIS model (Model A, p\u0026thinsp;=\u0026thinsp;0.001), though this difference was attenuated and not statistically significant when restricted to MV patients (Model B OR 1.13; p\u0026thinsp;=\u0026thinsp;0.15); addressing these disparities requires prospective investigation of goals-of-care communication quality, surrogate decision-making dynamics, and structural access inequities. The tracheostomy\u0026ndash;PEG\u0026ndash;LTACH care pathway represents a high-intensity clinical trajectory with profound implications for patients and families that warrants prospective goals-of-care integration. These findings provide the largest and most contemporary national framework for tracheostomy in AIS, directly informing neurocritical care decision-making, resource planning, and the design of future prospective studies on timing optimization and equity.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eDisclosures:\u0026nbsp;\u003c/strong\u003eThe authors report no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Source:\u0026nbsp;\u003c/strong\u003eNational Inpatient Sample, HCUP, AHRQ (2016\u0026ndash;2023). De-identified data; IRB waiver applied.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e1. Compliance with Instructions for Authors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors confirm that this manuscript complies with all instructions outlined in the Neurocritical Care Instructions for Authors, including formatting, word count, reference style, table and figure preparation, and submission requirements for an Original Work (Clinical Investigation).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2. Author Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMian Urfy, MD, MSc:\u0026nbsp;\u003c/strong\u003eConceptualization; study design; data acquisition; statistical analysis; interpretation of results; drafting of the manuscript; critical revision for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMariam Tariq Mir, MD:\u0026nbsp;\u003c/strong\u003eLiterature review; interpretation of results; critical revision of the manuscript for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work.\u003c/p\u003e\n\u003cp\u003eBoth authors meet all four ICMJE authorship criteria.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3. Authorship Requirements and Final Approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll listed authors meet the authorship criteria established by the International Committee of Medical Journal Editors (ICMJE): (1) substantial contributions to the conception or design of the work, or the acquisition, analysis, or interpretation of data; (2) drafting the work or revising it critically for important intellectual content; (3) final approval of the version to be published; and (4) agreement to be accountable for all aspects of the work. The final manuscript has been read and approved by all authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4. Originality and Exclusive Submission\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis manuscript has not been published, in whole or in part, in any other journal, and is not currently under consideration for publication elsewhere. No portion of the work has appeared in any prior publication, including preprint servers, abstracts, or conference proceedings, beyond what is permitted by the journal\u0026rsquo;s policies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5. Ethical Approval and Informed Consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis is a retrospective observational analysis of de-identified administrative data from the Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample (NIS), 2016\u0026ndash;2023. Per the HCUP Data Use Agreement and 45 CFR 46.102(e), this work does not constitute human subjects research because it involves only secondary analysis of fully de-identified, publicly available data. Accordingly, Institutional Review Board (IRB) review and informed consent were not required. The authors followed all HCUP methodological and reporting requirements, including cell-size suppression rules where applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e6. Conflicts of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no conflicts of interest, financial or otherwise, relevant to the subject of this manuscript. Neither author has any financial relationships, consulting arrangements, equity interests, patents, or other competing interests that could be perceived to influence the conduct or reporting of this work. ICMJE Disclosure forms have been completed by each author and are available upon request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e7. Reporting Checklist\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for cohort studies. A completed STROBE checklist is included as a supplemental file with this submission.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e8. Funding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work received no specific grant or financial support from any funding agency in the public, commercial, or not-for-profit sectors. No external funding was used in the design, conduct, analysis, or reporting of this study. This statement is consistent with the Funding Information entered in the Editorial Manager submission portal.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorresponding Author\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMian Urfy, MD, MSc\u003c/p\u003e\n\u003cp\u003eAssociate Professor of Neurology\u003c/p\u003e\n\u003cp\u003eChicago Medical School, Rosalind Franklin University\u003c/p\u003e\n\u003cp\u003eAdvocate Lutheran General Hospital, Neurocritical Care Unit\u003c/p\u003e\n\u003cp\u003e1775 Dempster Street, Park Ridge, IL 60068\u003c/p\u003e\n\u003cp\[email protected]\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLahiri S, Mayer SA, Fink ME, et al. Mechanical ventilation for acute stroke: a multi-state population-based study. Neurocrit Care. 2015;23(1):28\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12028-014-0082-9\u003c/span\u003e\u003cspan address=\"10.1007/s12028-014-0082-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMayer SA, Copeland D, Bernardini GL, et al. Cost and outcome of mechanical ventilation for life-threatening stroke. Stroke. 2000;31(10):2346\u0026ndash;53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/01.str.31.10.2346\u003c/span\u003e\u003cspan address=\"10.1161/01.str.31.10.2346\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlbert GP, McHugh DC, Hwang DY, et al. National cost estimates of invasive mechanical ventilation and tracheostomy in acute stroke, 2008\u0026ndash;2017. Stroke. 2023;54(10):2602\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/STROKEAHA.123.043176\u003c/span\u003e\u003cspan address=\"10.1161/STROKEAHA.123.043176\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalcott BP, Kamel H, Castro B, et al. Tracheostomy after severe ischemic stroke: a population-based study. J Stroke Cerebrovasc Dis. 2014;23(5):1024\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jstrokecerebrovasdis.2013.08.019\u003c/span\u003e\u003cspan address=\"10.1016/j.jstrokecerebrovasdis.2013.08.019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillwock JA, Villwock MR, Deshaies EM. Tracheostomy timing affects stroke recovery. J Stroke Cerebrovasc Dis. 2014;23(5):1069\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jstrokecerebrovasdis.2013.09.008\u003c/span\u003e\u003cspan address=\"10.1016/j.jstrokecerebrovasdis.2013.09.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShah S, Spirollari E, Ng C, et al. Early tracheostomy in patients undergoing mechanical thrombectomy for acute ischemic stroke. J Crit Care. 2023;78:154357. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jcrc.2023.154357\u003c/span\u003e\u003cspan address=\"10.1016/j.jcrc.2023.154357\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026ouml;sel J, Niesen WD, Salih F, et al. Effect of early vs standard approach to tracheostomy on functional outcome at 6 months among patients with severe stroke receiving mechanical ventilation: the SETPOINT2 randomized clinical trial. JAMA. 2022;327(19):1899\u0026ndash;909. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2022.4798\u003c/span\u003e\u003cspan address=\"10.1001/jama.2022.4798\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePremraj L, Camarda C, White N, et al. Tracheostomy timing and outcome in critically ill patients with stroke: a meta-analysis and meta-regression. Crit Care. 2023;27(1):132. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13054-023-04417-6\u003c/span\u003e\u003cspan address=\"10.1186/s13054-023-04417-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao CW, Hwang DY. US practitioner attitudes toward tracheostomy timing, benefits, risks, and techniques for severe stroke patients: a national survey and National Inpatient Sample analysis. Neurocrit Care. 2021;34(2):669\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12028-020-01127-7\u003c/span\u003e\u003cspan address=\"10.1007/s12028-020-01127-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmano MA, Bahouth MN, Kassalow B, et al. Timing of tracheostomy and association with ventilator liberation, length of stay, and discharge outcomes in neurocritical patients. Tracheostomy. 2025;2(1):56\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.62905/001c.133992\u003c/span\u003e\u003cspan address=\"10.62905/001c.133992\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMainali S, Fontaine GV, Rajajee V, et al. Guidelines for neuroprognostication in critically ill adults with acute ischemic stroke. Neurocrit Care. 2026. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12028-026-02486-3\u003c/span\u003e\u003cspan address=\"10.1007/s12028-026-02486-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgency for Healthcare Research and Quality. HCUP National Inpatient Sample (NIS). Healthcare Cost and Utilization Project (HCUP). Rockville, MD. 2024. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://hcup-us.ahrq.gov/nisoverview.jsp\u003c/span\u003e\u003cspan address=\"https://hcup-us.ahrq.gov/nisoverview.jsp\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoggini A, Qureshi AI, Saleh Velez FG, et al. Early tracheostomy in patients with nontraumatic intracerebral hemorrhage is associated with lower in-hospital complications and reduced resource use without increased mortality. Neurocrit Care. 2026. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12028-025-02436-5\u003c/span\u003e\u003cspan address=\"10.1007/s12028-025-02436-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/01.mlr.0000182534.19832.83\u003c/span\u003e\u003cspan address=\"10.1097/01.mlr.0000182534.19832.83\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaier IL, Schramm K, B\u0026auml;hr M, et al. Predictive factors for the need of tracheostomy in patients with large vessel occlusion stroke being treated with mechanical thrombectomy. Front Neurol. 2021;12:728624. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fneur.2021.728624\u003c/span\u003e\u003cspan address=\"10.3389/fneur.2021.728624\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider H, Hertel F, Kuhn M, et al. Decannulation and functional outcome after tracheostomy in patients with severe stroke (DECAST): a prospective observational study. Neurocrit Care. 2017;27(1):26\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12028-017-0390-y\u003c/span\u003e\u003cspan address=\"10.1007/s12028-017-0390-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHannawi Y, Hannawi B, Rao CPV, Suarez JI, Bershad EM. Stroke-associated pneumonia: major advances and obstacles. Cerebrovasc Dis. 2013;35(5):430\u0026ndash;43. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1159/000350199\u003c/span\u003e\u003cspan address=\"10.1159/000350199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaugeri R, Iacopino DG, Graziano F, et al. Timing of tracheostomy in ICU patients: a systematic review and meta-analysis of randomized controlled trials. Life (Basel). 2024;14(9):1165. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/life14091165\u003c/span\u003e\u003cspan address=\"10.3390/life14091165\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaigle R, Carrese JA, Cooper LA, Urrutia VC, Gottesman RF. Minority race and male sex as risk factors for non-beneficial gastrostomy tube placements after stroke. PLoS ONE. 2018;13(2):e0191293. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0191293\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0191293\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOrmseth CH, Falcone GJ, Jasak SD, et al. Minority patients are less likely to undergo withdrawal of care after spontaneous intracerebral hemorrhage. Neurocrit Care. 2018;29(3):419\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12028-018-0554-4\u003c/span\u003e\u003cspan address=\"10.1007/s12028-018-0554-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGottlieb ER, Ziegler J, Morley K, Rush B, Celi LA. Assessment of racial and ethnic differences in oxygen supplementation among patients in the intensive care unit. JAMA Intern Med. 2022;182(8):849\u0026ndash;58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamainternmed.2022.2587\u003c/span\u003e\u003cspan address=\"10.1001/jamainternmed.2022.2587\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"neurocritical-care","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"neca","sideBox":"Learn more about [Neurocritical Care](http://link.springer.com/journal/12028)","snPcode":"12028","submissionUrl":"https://www.editorialmanager.com/neca/default2.aspx","title":"Neurocritical Care","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"acute ischemic stroke, tracheostomy, mechanical ventilation, neurocritical care, tracheostomy timing, outcomes, National Inpatient Sample","lastPublishedDoi":"10.21203/rs.3.rs-9546300/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9546300/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eTracheostomy in acute ischemic stroke (AIS) is a high-stakes decision yet contemporary national data on utilization, predictors, and outcomes are limited. We analyzed 854,660 AIS hospitalizations (NIS, 2016\u0026ndash;2023) to characterize tracheostomy use, identify predictors, examine timing effects, and explore mechanisms underlying a national decline.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eTracheostomy was identified by ICD-10-PCS codes and classified as early (\u0026le;\u0026thinsp;7 days) or late (\u0026gt;\u0026thinsp;7 days). Multivariable logistic regression identified predictors among all AIS patients (Model A) and among mechanically ventilated (MV) patients (Model B). Outcomes included length of stay, in-hospital mortality, PEG placement, and discharge disposition. Temporal trends were assessed by Cochran-Armitage test and logistic regression; mechanistic analyses tested five pre-specified hypotheses including a year-thrombectomy interaction model.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eTracheostomy occurred in 5,373 hospitalizations (0.63% of AIS; 12.1% of MV patients). Median procedure day was 11 (IQR 8\u0026ndash;16); 22.0% were early and 78.0% late. In-hospital mortality was 9.84% (early) and 8.23% (late), versus 39.7% in ventilated patients without tracheostomy. Mean LOS was significantly shorter with early versus late tracheostomy (23.8 vs 34.4 days; adjusted LOS ratio 0.662; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Prolonged MV (OR 62.3), hemorrhagic transformation (OR 1.35), and decompressive craniectomy (OR 2.55) were the dominant independent predictors. Tracheostomy rates declined from 12.9% to 9.44% of MV patients (Z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;6.54; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), with the steepest drop in 2023. Mechanistic analyses showed thrombectomy utilization rose concurrently (15.3% to 22.2% among MV patients; Z\u0026thinsp;=\u0026thinsp;+\u0026thinsp;6.39; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), prolonged MV declined (Z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;6.77; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), and a significant year-thrombectomy interaction (p\u0026thinsp;=\u0026thinsp;0.027) suggested faster decline among thrombectomy-treated patients. PEG placement occurred in 78.3% of tracheostomy patients.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eDeclining tracheostomy rates in AIS likely reflect a convergence of factors: expanding thrombectomy utilization, shorter ventilator dependence, and evolving ICU practice \u0026mdash; rather than a shift in the tracheostomy decision itself. Racial disparities persist after adjustment. These findings provide the largest contemporary national framework for tracheostomy decision-making and goals-of-care counseling in AIS.\u003c/p\u003e","manuscriptTitle":"Tracheostomy in Acute Ischemic Stroke: Declining National Utilization, Independent Predictors, and In-Hospital Outcomes Among 854,660 Hospitalizations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-12 14:16:20","doi":"10.21203/rs.3.rs-9546300/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-05-04T00:36:46+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-04T00:10:54+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Neurocritical Care","date":"2026-05-03T13:56:24+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-03T13:37:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Neurocritical Care","date":"2026-04-27T18:16:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"neurocritical-care","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"neca","sideBox":"Learn more about [Neurocritical Care](http://link.springer.com/journal/12028)","snPcode":"12028","submissionUrl":"https://www.editorialmanager.com/neca/default2.aspx","title":"Neurocritical Care","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"6ba00653-48e4-4ff6-8d8e-38b2f320e8f3","owner":[],"postedDate":"May 12th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"","date":"2026-05-04T00:36:46+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-04T00:10:54+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Neurocritical Care","date":"2026-05-03T13:56:24+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-03T13:37:17+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-12T14:16:21+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-12 14:16:20","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9546300","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9546300","identity":"rs-9546300","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}