Cardiac Intensive Care Unit Patients Research Articles

Changes in Inhaled Nitric Oxide Use Across ICUs After Implementation of a Standard Pathway.

Inhaled nitric oxide (iNO) is a selective pulmonary vasodilator. It is expensive, frequently used, and not without risk. There is limited evidence supporting a standard approach to initiation and weaning. Our objective was to optimize the use of iNO in the cardiac ICU (CICU), PICU, and neonatal ICU (NICU) by establishing a standard approach to iNO utilization. A quality improvement study using a prospective cohort design with historical controls. Four hundred seven-bed free standing quaternary care academic children's hospital. All patients on iNO in the CICU, PICU, and NICU from January 1, 2017 to December 31, 2022. Unit-specific standard approaches to iNO initiation and weaning. Sixteen thousand eighty-seven patients were admitted to the CICU, PICU, and NICU with 9343 in the pre-iNO pathway era (January 1, 2017 to June 30, 2020) and 6744 in the postpathway era (July 1, 2020 to December 31, 2022). We found a decrease in the percentage of CICU patients initiated on iNO from 17.8% to 11.8% after implementation of the iNO utilization pathway. We did not observe a change in iNO utilization between the pre- and post-iNO pathway eras in either the PICU or NICU. Based on these data, we estimate 564 total days of iNO (-24%) were saved over 24 months in association with the standard pathway in the CICU, with associated cost savings. Implementation of a standard pathway for iNO use was associated with a statistically discernible reduction in total iNO usage in the CICU, but no change in iNO use in the NICU and PICU. These differential results likely occurred because of multiple contextual factors in each care setting.

Exposure to hyperoxemia and mortality in cardiac intensive care unit patients

Abstract Funding Acknowledgements None. Background Hyperoxemia (a high PaO2, i.e. >200-300 mmHg) has been associated with adverse outcomes in general critical care settings as well as in acute cardiovascular care conditions. The association between hyperoxemia and mortality has not been characterized in the heterogeneous population admitted to the cardiac intensive care unit (CICU). We aimed to examine the association between hyperoxemia on CICU admission with in-hospital mortality. Methods We retrospectively analyzed Mayo Clinic CICU patients from January 2007 to April 2018. PaO2 closest on admission was analyzed as a continuous variable (per 100 mmHg) and categorized as: 60-100 mmHg (51.9%), 101-150 mmHg (28.6%), 151-200 mmHg (10.6%), 201-300 mmHg (6.4%), and >300 mmHg (2.5%). Those without PaO2 on admission and those with hypoxemia were excluded. The primary outcome was all-cause in-hospital mortality evaluated using logistic regression, after adjusting for age, Charlson Comorbidity Index, M-CARS, and use of IMV and NIPPV. Results A total of 3,368 patients were included (median 70.3 years, 39.4% women). Admission diagnoses included ACS (37.0%, with 19.8% STEMI), HF (67.8%), shock (33.0%) and cardiac arrest (25.3%). Median admission PaO2 was 99 (IQR 78, 136) mmHg, with no difference in median PaO2 between hospital survivors and inpatient deaths (99 versus 98, p = 0.38). A J-shaped relationship was observed between admission PaO2 and in-hospital mortality (Figure A), with slightly higher mortality below 100 mmHg and incrementally higher mortality above this threshold (esp. at >200 mmHg). The incremental association between higher PaO2 and mortality (adjusted OR 1.17 per 100 mmHg higher, 95% CI 1.01-1.34, P = 0.03) was focused in the highest PaO2 group (i.e., >300 mmHg; Figure B). Conclusion Hyperoxia (esp. PaO2 >300 mmHg) on CICU admission is associated with increased in-hospital mortality, demonstrating a J-shaped association with an inflection point of lowest mortality around 100 mmHg. Further studies are warranted to assess if specific oxygen targets influence outcomes in critically ill cardiac patients.

Global longitudinal strain as a predictor of mortality in patients with cardiogenic shock

Abstract Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): Aarhus University Dagmar Marshalls Fond. Background The assessment of left ventricular (LV) function through transthoracic echocardiography (TTE) is essential for guiding diagnosis and treatment in cases of cardiogenic shock (CS). Myocardial global longitudinal strain (GLS) provides an objective and reproducible measure of systolic function, independently predicting outcomes in acute heart failure, regardless of left ventricular ejection fraction (LVEF). However, whether assessment of GLS can predict mortality in patients with CS remains unknown. Purpose In this study, we aimed to evaluate the prognostic significance of GLS in patients with CS. Methods Data from cardiac intensive care unit (CICU) patients admitted between 2007 and 2018 with CS who received vasoactive agents and/or mechanical circulatory support were analysed. GLS (as absolute values) and biplane LVEF were measured via TTE performed within 24 hours of CICU admission. Patients with CS were categorised based on the GLS quartile: mild (GLS >9.7%), moderate (7.0% < GLS <9.6%), severe (5.0% < GLS < 6.9%), and very severe (GLS ≤ 4.9%). CICU and in-hospital mortality were compared across GLS categories, and the relationship between GLS and mortality was assessed using multivariable logistic regression. Results The study included 626 patients with a median age of 68.6 years, a median LVEF of 31% (IQR 24 to 40%), and a median GLS of 7.0% (IQR 5.0 to 9.7%). In-hospital mortality was 29.1%. As GLS deteriorated, there was an increasing need for vasopressors (P<0.001). Patients with progressively lower GLS exhibited significantly higher CICU and in-hospital mortality across the GLS quartiles (Figure A; P<0.001). Patients in the severe (adjusted OR 2.67; p<0.001) and very severe category (adjusted OR 3.54; P<0.001) had significantly increased risk of in-hospital death compared with mild category. In multivariable analysis, every 1% decrease in GLS corresponded to a 14% higher risk of in-hospital mortality (P<0.001). Finally, GLS outperformed LVEF for discrimination of in-hospital mortality (P=0.002), and when both measurements were included in the same multivariable model, only GLS remained significant. When patients were divided according to whether they had GLS and/or LVEF below the median value, patients with low GLS had higher mortality irrespective of the LVEF (Figure B). Conclusion In patients with CS, GLS assessment surpasses LVEF in its prognostic value and improves risk stratification. Routine evaluation of GLS should be considered for patients with CS.Figure AFigure B

Association between the shock index on admission and in-hospital mortality in the cardiac intensive care unit.

An elevated shock index (SI) predicts worse outcomes in multiple clinical arenas. We aimed to determine whether the SI can aid in mortality risk stratification in unselected cardiac intensive care unit patients. We included admissions to the Mayo Clinic from 2007 to 2015 and stratified them based on admission SI. The primary outcome was in-hospital mortality, and predictors of in-hospital mortality were analyzed using multivariable logistic regression. We included 9,939 unique cardiac intensive care unit patients with available data for SI. Patients were grouped by SI as follows: < 0.6, 3,973 (40%); 0.6-0.99, 4,810 (48%); and ≥ 1.0, 1,156 (12%). After multivariable adjustment, both heart rate (adjusted OR 1.06 per 10 beats per minute higher; CI 1.02-1.10; p-value 0.005) and systolic blood pressure (adjusted OR 0.94 per 10 mmHg higher; CI 0.90-0.97; p-value < 0.001) remained associated with higher in-hospital mortality. As SI increased there was an incremental increase in in-hospital mortality (adjusted OR 1.07 per 0.1 beats per minute/mmHg higher, CI 1.04-1.10, p-Value < 0.001). A higher SI was associated with increased mortality across all examined admission diagnoses. The SI is a simple and universally available bedside marker that can be used at the time of admission to predict in-hospital mortality in cardiac intensive care unit patients.

Stress hyperglycemia ratio linked to worse prognosis in Cardiac Intensive Care Unit patients: A retrospective observational study

AimsThis study aimed to delineate correlation between stress hyperglycemia ratio (SHR) and clinical outcomes among patients in the cardiac intensive care unit (CICU). MethodsParticipants were categorized based on their SHR threshold values. Key outcomes were short-term mortality and major adverse cardiovascular events (MACEs) at 1-year follow-up. The association between SHR and outcomes was estimated using inverse probability of treatment weighting (IPTW) and Kaplan-Meier analyses. The C-statistic was used to gauge the predictive capability of SHR. ResultsThe study included 1,133 patients from the Medical Information Mart for Intensive Care IV and 412 from the Second Affiliated Hospital of Wenzhou Medical University. Kaplan-Meier curves revealed that individuals with elevated SHR exhibited higher 90-day mortality and MACEs. When considering SHR levels and diabetes status simultaneously, those with increased SHR but non-diabetes had the highest 90-day mortality and MACEs. SHR was associated with short-term mortality and MACEs (HRadjusted 1.63 95%CI 1.15–2.30; HRIPTW 1.47 95%CI 1.05–2.05). Upon integrating SHR into the foundational model, the C-statistic was 0.821, outperforming other hyperglycemia metrics. ConclusionSHR is a valuable indicator for predicting poor outcomes in CICU patients. Its utility spans potential risk stratification and offers insights for tailoring prognostic treatments to CICU patients.

Métodos de aprendizaje automático para el desarrollo de un modelo predictivo de delirio durante el ingreso en unidades de cuidados intensivos cardiacos

Introduction and objectivesDelirium, recognized as a crucial prognostic factor in the cardiac intensive care unit (CICU), has evolved in response to the changing demographics among critically ill cardiac patients. This study aimed to create a predictive model for delirium for patients in the CICU. MethodsThis study included consecutive patients admitted to the CICU of the Samsung Medical Center. To assess the candidate variables for the model: we applied the following machine learning methods: random forest, extreme gradient boosting, partial least squares, and Plmnet-elastic.net. After selecting relevant variables, we performed a logistic regression analysis to derive the model formula. Internal validation was conducted using 100-repeated hold-out validation. ResultsWe analyzed 2774 patients, 677 (24.4%) of whom developed delirium in the CICU. Machine learning-based models showed good predictive performance. Clinically significant and frequently important predictors were selected to construct a delirium prediction scoring model for CICU patients. The model included albumin level, international normalized ratio, blood urea nitrogen, white blood cell count, C-reactive protein level, age, heart rate, and mechanical ventilation. The model had an area under the receiver operating characteristics curve (AUROC) of 0.861 (95%CI, 0.843-0.879). Similar results were obtained in internal validation with 100-repeated cross-validation (AUROC, 0.854; 95%CI, 0.826-0.883). ConclusionsUsing variables frequently ranked as highly important in four machine learning methods, we created a novel delirium prediction model. This model could serve as a useful and simple tool for risk stratification for the occurrence of delirium at the patient's bedside in the CICU.

Family Perceptions of Virtual Family-Centered Rounds in a Quaternary Cardiac Intensive Care Unit.

Background: Family-centered rounds (FCR) reduce the risk of psychological comorbidities of family members and improve the quality of communication between providers and families. Materials and methods: We conducted a pilot quality improvement study analyzing family perceptions of virtual FCR. Family members of previously admitted cardiac ICU patients who participated in at least one session of virtual FCR between April 2020 and June 2021 at Massachusetts General Hospital were surveyed post-ICU discharge. Results: During the study, 82 family members enrolled and participated in virtual FCR with 29 completing the post-admission telephone survey. Many cardiac ICU patients were male (n = 53), and a majority were discharged home (43%) with the patient's wives being the most common respondents to the questionnaire (n = 18). Across all questions in the survey, more than 75% of the respondents perceived the highest level of care in trust, communication, relationship, and compassion with their provider. Participants perceived the highest level of care in trust (96%), explanation (88%), as well as care and understanding (89%). Conclusions: Family members of cardiac ICU patients positively rated the quality of communication and perceived a high level of trust and communication between their providers on the virtual format.

Patients Aged 90 Years and Above With Acute Coronary Syndrome in the Cardiac Intensive Care Unit: Management and Outcomes

Limited data exist regarding outcomes after coronary angiography (CAG) and percutaneous coronary intervention (PCI) in patients aged ≥90 years admitted to the cardiac intensive care unit (CICU) with acute coronary syndrome (ACS). We studied sequential CICU patients ≥90 years admitted with ACS from 2007 to 2018. Three therapeutic approaches were defined: (1) No CAG; (2) CAG without PCI (CAG/No PCI); and (3) CAG with PCI (CAG/PCI). In-hospital mortality was evaluated using multivariable logistic regression. All-cause 1-year mortality was evaluated using Kaplan-Meier and multivariable Cox proportional hazards analysis. The study included 239 patients with a median age of 92 (range 90 to 100) years (57% females; 45% ST-elevation myocardial infarction; 8% cardiac arrest; 16% shock). The No CAG group had higher Day 1 Sequential Organ Failure Assessment scores, more co-morbidities, worse kidney function, and fewer ST-elevation myocardial infarctions. In-hospital mortality was 20.8% overall and did not differ between the No CAG (n = 103; 21.4%), CAG/No PCI (n = 47; 21.3%), and CAG/PCI (n = 90; 20.0%) groups, before or after adjustment. Overall 1-year mortality was 52.5% and did not differ between groups before or after adjustment. Median survival was 6.9 months overall and 41.2% of hospital survivors died within 1 year of CICU admission. CICU patients aged ≥90 years with ACS have a substantial burden of illness with high in-hospital and 1-year mortality that was not lower in those who underwent CAG or PCI. These results suggest that careful patient selection for invasive coronary procedures is essential in this vulnerable population.

Machine learning methods for developing a predictive model of the incidence of delirium in cardiac intensive care units

Introduction and objectivesDelirium, recognized as a crucial prognostic factor in the cardiac intensive care unit (CICU), has evolved in response to the changing demographics among critically ill cardiac patients. This study aimed to create a predictive model for delirium for patients in the CICU. MethodsThis study included consecutive patients admitted to the CICU of the Samsung Medical Center. To assess the candidate variables for the model: we applied the following machine learning methods: random forest, extreme gradient boosting, partial least squares, and Plmnet-elastic.net. After selecting relevant variables, we performed a logistic regression analysis to derive the model formula. Internal validation was conducted using 100-repeated hold-out validation. ResultsWe analyzed 2774 patients, 677 (24.4%) of whom developed delirium in the CICU. Machine learning-based models showed good predictive performance. Clinically significant and frequently important predictors were selected to construct a delirium prediction scoring model for CICU patients. The model included albumin level, international normalized ratio, blood urea nitrogen, white blood cell count, C-reactive protein level, age, heart rate, and mechanical ventilation. The model had an area under the receiver operating characteristics curve (AUROC) of 0.861 (95%CI, 0.843-0.879). Similar results were obtained in internal validation with 100-repeated cross-validation (AUROC, 0.854; 95%CI, 0.826-0.883). ConclusionsUsing variables frequently ranked as highly important in four machine learning methods, we created a novel delirium prediction model. This model could serve as a useful and simple tool for risk stratification for the occurrence of delirium at the patient's bedside in the CICU.

Usefulness of the intracardiac electrocardiogram recorded using central venous catheters on P-wave magnification

BackgroundThe detection of atrial electrical activity is extremely valuable in recognizing complex cardiac arrhythmias. However, P-wave detection on a surface electrocardiogram (S-ECG) can sometimes be challenging. The intracardiac electrocardiogram (IC-ECG), recorded by a central venous catheter loaded with saline solution, has proven to be a safe and effective method for amplifying atrial electrical activity. We aim to compare the P-wave amplitude recorded in the S-ECG and the IC-ECG in different venous accesses, catheters, heart rhythms, and atrial dimensions. MethodsWe compared the P wave amplitude obtained by the IC-ECG and the S-ECG recordings from cardiac intensive care unit patients. ResultsIn 109 nonconsecutive patients, a total of 166 IC-ECG were collected. The median amplitude of the P wave was 0.1 (0.083–0.3) mV in the S-ECG and 0.4 (0.25–2.4) mV in the IC-ECG; p < 0.001. This difference remained significant regardless of the patient's heart rhythm, left atrial dimension, and catheter or vascular access used. ConclusionThe IC-ECG acquired using central venous catheters significantly increases atrial electrical activity signals. This technique might help identify complex cardiac arrhythmias.

Admission Total Leukocyte Count as a Predictor of Mortality in Cardiac Intensive Care Unit Patients

BackgroundInflammation is a sequela of cardiovascular critical illness and a risk factor for mortality. ObjectivesThis study aimed to evaluate the association between white blood cell count (WBC) and mortality in a broad population of patients admitted to the cardiac intensive care unit (CICU). MethodsThis retrospective cohort study included patients admitted to the Mayo Clinic CICU between 2007 and 2018. We analyzed WBC as a continuous variable and then categorized WBC as low (<4.0 × 103/mL), normal (≥4.0 to <11.0 × 103/mL), high (≥11.0 to <22.0 × 103/mL), or very high (≥22.0 × 103/mL). The association between WBC and in-hospital mortality was evaluated using multivariable logistic regression and random forest models. ResultsWe included 11,699 patients with a median age of 69.3 years (37.6% females). Median WBC was 9.6 (IQR: 7.4-12.7). Mortality was higher in the low (10.5%), high (12.0%), and very high (33.3%) WBC groups relative to the normal WBC group (5.3%). A rising WBC was incrementally associated with higher in-hospital mortality after adjustment (AICc adjusted OR: 1.03 [95% CI: 1.02-1.04] per 1 × 103 increase in WBC). After adjustment, only the high (AICc adjusted OR: 1.37 [95% CI: 1.15-1.64]) and very high (AICc adjusted OR: 1.99 [1.47-2.71]) WBC groups remained associated with increased risk of in-hospital mortality. ConclusionsLeukocytosis is associated with an increased mortality risk in a diverse cohort of CICU patients. This readily available marker of systemic inflammation may be useful for risk stratification within the increasingly complex CICU patient population.

Serial Assessment of Shock Severity in Cardiac Intensive Care Unit Patients.

One-time assessment of the Society for Cardiovascular Angiography and Interventions (SCAI) shock classification robustly predicts mortality in the cardiac intensive care unit (CICU). We sought to determine whether serial SCAI shock classification could improve risk stratification. Unique admissions to a single academic level 1 CICU from 2015 to 2018 were included in this retrospective cohort study. Electronic health record data were used to assign the SCAI shock stage during 4-hour blocks of the first 24 hours of CICU admission. Shock was defined as hypoperfusion (SCAI shock stage C, D, or E). In-hospital death was evaluated using logistic regression. Among 2918 unique CICU patients, 1537 (52.7%) met criteria for shock during ≥1 block, and 266 (9.1%) died in the hospital. The SCAI shock stage on admission was: A, 37.6%; B, 31.5%; C, 25.9%; D, 1.8%; and E, 3.3%. Patients who met SCAI criteria for shock on admission (first 4 hours) and those with worsening SCAI shock stage after admission were at higher risk for in-hospital death. Each higher admission (adjusted odds ratio, 1.36 [95% CI, 1.18-1.56]; area under the receiver operating characteristic curve, 0.70), maximum (adjusted odds ratio, 1.59 [95% CI, 1.37-1.85]; area under the receiver operating characteristic curve, 0.73) and mean (adjusted odds ratio, 2.42 [95% CI, 1.99-2.95]; area under the receiver operating characteristic curve, 0.78) SCAI shock stage was incrementally associated with a higher in-hospital mortality rate. Discrimination was highest for the mean SCAI shock stage (P<0.05). Each additional 4-hour block meeting SCAI criteria for shock predicted a higher mortality rate (adjusted odds ratio, 1.15 [95% CI, 1.07-1.24]). Dynamic assessment of shock using serial SCAI shock classification assignment can improve mortality risk stratification in CICU patients by quantifying the magnitude and duration of shock.

Noninvasive Hemodynamic Characterization of Shock and Preshock Using Echocardiography in Cardiac Intensive Care Unit Patients.

Shock and preshock are defined on the basis of the presence of hypotension, hypoperfusion, or both. We sought to determine the hemodynamic underpinnings of shock and preshock noninvasively using transthoracic echocardiography (TTE). We included Mayo Clinic cardiac intensive care unit patients from 2007 to 2015 with TTE within 1 day of admission. Hypotension and hypoperfusion at the time of cardiac intensive care unit admission were used to define 4 groups. TTE findings were evaluated across these groups, and in-hospital mortality was evaluated according to TTE findings in each group. We included 5375 patients with a median age of 69.2 years (36.8% women). The median left ventricular ejection fraction was 50%. Groups based on hypotension and hypoperfusion were assigned as follows: no hypotension or hypoperfusion, 59.7%; isolated hypotension, 15.3%; isolated hypoperfusion, 16.4%; and both hypotension and hypoperfusion, 8.7%. Most TTE variables of interest varied across these groups, with worse biventricular function, lower forward flow, and higher filling pressures as the degree of hemodynamic compromise increased. In-hospital mortality occurred in 8.2%, and inpatient deaths had more TTE parameter abnormalities. In-hospital mortality increased with the degree of hemodynamic compromise, and a marked gradient in in-hospital mortality was observed when the clinical classification of shock and preshock was combined with TTE findings reflecting worse biventricular function, lower forward flow, or higher filling pressures. Substantial differences in cardiac function are observed between cardiac intensive care unit patients with preshock and shock using TTE, and the combination of the clinical and TTE hemodynamic assessment provides robust mortality risk stratification.

The correlation between death anxiety, loneliness and hope levels in patients treated in the cardiac intensive care unit.

A sense of hope plays an important role in relieving stress and psychological distress of cardiology patients, as well as improving their physical well-being. The aim of this study is to investigate the correlation between death anxiety, loneliness and hope levels in patients receiving treatment in a cardiac intensive care unit (ICU). This is a prospective, descriptive and correlational study. The study was completed with 150 cardiac ICU patients in Istanbul, Turkey. The data were collected using a Patient Information Form, the Templer Death Anxiety Scale (TDAS), the Herth Hope Index (HHI) and the UCLA Loneliness Scale (UCLA-LS). The patients had a mean age of 63.56 ± 12.74 years. Most of the patients (82%) were treated in the ICU for heart failure. There was a statistically significant positive correlation between total scores of TDAS and UCLA-LS (r = .337; p < .001) and a statistically significant negative correlation between total scores of UCLA-LS and HHI (r = -.292; p < .001). Also, there was a statistically significant negative correlation between the scores of UCLA-LS and Positive Readiness and Expectancy Subscale (r = -.164; p = .044). The multiple linear regression indicated that the model was statistically significant (F = 7.177, p < .001). The variables of age and UCLA-LS among those included in the model were statistically significant predictors of the death anxiety scores of the patients (23.1%) (p < .05). The cardiology patients who received treatment in the ICU had a high level of death anxiety and moderate levels of loneliness and hope. The age and loneliness level were statistically significant predictors of death anxiety. It is recommended that individualized nursing care be planned and provided to conscious cardiology patients who are treated in the ICU, considering their age and loneliness levels and that nursing care be planned for individuals who are at risk of fear, anxiety, loneliness and hopelessness by periodically assessing their death anxiety, loneliness and hope levels.

Machine learning score focused only on echocardiographic data to predict in-hospital outcomes in ICCU patients. A study from the ADDICT ICCU cohort

Abstract Background While several studies have shown the strong prognostic value of several transthoracic echocardiography (TTE) parameters in patients admitted in intensive cardiac care unit (ICCU), the interest of a severity score integrating these parameters has not been established. Machine learning (ML) represents an innovative and accurate statistical method to embed TTE parameters into a scoring system. Purpose To investigate the feasibility and accuracy of ML integrating TTE parameters measured at admission in ICCU to predict in-hospital major adverse cardiovascular events (MACE) and to compare its performance with existing scores. Methods In April 2021, 1,499 patients admitted to 39 French ICCU were included in the ADDICT-ICCU Cohort (NCT05063097). A comprehensive TTE was performed within the first 24 hours of admission. Twenty TTE parameters were systematically collected. ML involved automated feature selection by Boruta to select variables and prevent model overfitting. Several ML models (LASSO, Random Forest and XGBoost) were then trained on 70% of patients and evaluated on the other 30% as internal validation. Their performances were compared against standard logistic regression model, using receiver operating characteristics (ROC) curves, precision-recall (PR) curves and area-under-the curves (AUC). The primary composite outcome was in-hospital MACEs defined by all-cause mortality, cardiogenic shock (requiring medical or mechanical haemodynamic support) and resuscitated cardiac arrest. The external validation cohort of the ML score was performed in another center on 300 patients. Results Of 1,499 consecutive patients screened (63±15 years, 70% male), MACEs occurred in 67 patients (4.5%) including: 27 (1.8%) deaths, 38 (2.53%) cardiogenic shocks, and 14 (0.9%) resuscitated cardiac arrests. To build the ML models, 5 TTE parameters were selected as being the most important in predicting MACEs: LVOT VTI, E/e’ ratio, sPAP, TAPSE, and LVEF. The XGBoost model showed the best performance compared with the other ML models (AUROC for XGBoost: 0.83, Random-forest: 0.81, LASSO regression 0.79 and to a regression model: 0.76). Using the XGBoost model as our final ML-score, SHAP values for the parameters were: 0.047 for LVOT VTI, 0.041 for E/e’ ratio, 0.028 for SPAP, 0.021 for TAPSE and 0.01 for LVEF. Our ML-score exhibited a higher AUC compared with existing scores (TIMI, GRACE, qSOFA, Acute-HF) (Figure 1), and an incremental prognostic value for predicting in-hospital MACEs over and above clinical or biological data (Chi-square 59.72 p&amp;lt;0.001; C-Index 0.8 [0.71;0.89], p=0.012) (Figure 2). Conclusion The ML score including only TTE parameters exhibited a higher prognostic value to predict in-hospital MACE compared with any other existing traditional methods in patients admitted in ICCU. This ML score could be implemented in clinical practice given its ease of use (only 5 TTE parameters to analyze) and its prognostic implication.Figure 1.PerformanceFigure 2.Prognostic value

Abstract 15559: Mortality Risk Stratification Using Artificial Intelligence Augmented Prediction of Serum Potassium in Cardiac Intensive Care Unit Patients

Introduction: Hyperkalemia has been associated with increased mortality in cardiac intensive care unit (CICU) patients. An artificial intelligence (AI) augmented electrocardiogram (ECG) can predict hyperkalemia, and other AI-ECG algorithms have predicted mortality in CICU patients. Hypothesis: We hypothesized that the AI-ECG hyperkalemia algorithm could stratify mortality risk beyond the serum potassium measurement. Methods: We included 11 234 unique Mayo Clinic CICU patients admitted from 2007 to 2018 who had both a 12 lead ECG and blood potassium (K) level obtained at admission with K&gt;=5 mEq/L defining hyperkalemia. ECGs underwent AI evaluation for probability of hyperkalemia (probability &gt;=0.5 defined as positive). Hospital mortality was analyzed using logistic regression and survival to one year was estimated using Kaplan-Meier and Cox analysis. Results: The mean age was 70 years, 37.8% were females, and 92.4% were white. Chronic kidney disease was present in 20%. The mean K value was 4.2 mEq/L, and 13% (n=1460) had hyperkalemia. AI-ECG predicted hyperkalemia in 34% (n=3820). In-hospital mortality was higher in patients with laboratory and/or AI-ECG hyperkalemia, and these patients had lower survival out to 1 year. Mortality was lowest in patients with normal K and negative AI-ECG and highest in patients with both laboratory-confirmed hyperkalemia and AI-ECG predicted hyperkalemia. Patients with isolated laboratory hyperkalemia (false-negative AI-ECG) had higher mortality than those with isolated AI-ECG hyperkalemia (false-positive AI-ECG). Conclusions: AI-ECG probability of hyperkalemia, even with a normal laboratory K, was associated with higher short-term and long-term mortality in CICU patients. The laboratory K value and AI-ECG probability of hyperkalemia are complementary risk factors. The AI algorithm can detect prognostically important ECG abnormalities which may reflect the effects of K levels on the myocardium.

Abstract 13964: Dynamic Assessment of Shock Severity in Cardiac Intensive Care Unit Patients

Introduction: A single assessment of the SCAI shock classification robustly predicts mortality, and a repeat assessment may improve prognostication. Hypothesis: We hypothesized that frequent serial assignment of the SCAI shock stage could further improve risk stratification. Methods: Unique AHA Level 1 cardiac intensive care unit (CICU) admissions at a single center from 2015 to 2018 were reviewed retrospectively. Time-stamped electronic health record data were used to assign the SCAI shock stage in each 4-hour block during the first 24 hours of CICU admission. Shock was defined as SCAI shock stage C, D, or E. In-hospital mortality was evaluated using logistic regression. Results: Among 2,918 CICU patients, 1,537 (52.7%) met criteria for shock during &gt;=1 block and 266 (9.1%) died in hospital. The SCAI shock stage on admission was: A, 37.6%; B, 31.5%; C, 25.9%; D, 1.8%; E, 3.3%. Patients with worsening SCAI shock stage after admission (first 4 hours) were at higher risk of mortality ( Figure A ), as were patients who met SCAI criteria for shock (particularly those with shock on admission, Figure B ). The pattern was consistent in patients with acute coronary syndromes, heart failure, or cardiac arrest (who had very high mortality). Each higher admission (aOR 1.36, 95% CI 1.18-1.56, AUC 0.70), maximum (aOR 1.59, 95% CI 1.37-1.85, AUC 0.73) and 24 hour mean (aOR 2.42, 95% CI 1.99-2.95, AUC 0.78) SCAI shock stage were incrementally associated with increasing in-hospital mortality. Discrimination was highest for the mean SCAI shock stage (p &lt;0.05). Each additional 4 hour block meeting SCAI criteria for shock was associated with higher mortality (aOR 1.15, 95% CI 1.07-1.24). Conclusions: Dynamic assessment of shock using serial SCAI shock classification assignment can improve mortality risk stratification in CICU patients. Cumulative time in shock was a strong mortality risk indicator, highlighting the area under the curve of shock severity as an important determinant of outcomes.

Association of Shock Index with Echocardiographic Parameters in Cardiac Intensive Care Unit

BackgroundA high shock index (SI), the ratio of heart rate (HR) to systolic blood pressure (SBP), has been associated with unfavorable outcomes. We sought to determine the hemodynamic underpinnings of an elevated SI using 2-D and doppler Transthoracic Echocardiography (TTE) in unselected cardiac intensive care unit (CICU) patients. MethodsWe included Mayo Clinic CICU admissions from 2007 to 2018 who were in sinus rhythm at the time of TTE. The SI was calculated using HR and SBP at the time of TTE. Patients were grouped according to SI: <0.7, 4012 (64%); 0.7–0.99, 1764 (28%); and ≥ 1.0, 513 (8%). Pearson's correlation coefficient was used to assess associations between continuous variables. ResultsWe included 6289 unique CICU patients, 58% of whom had acute coronary syndrome. The median age was 67.9 years old and 37.8% were females. The mean SI was 0.67 BPM/mmHg. As the SI increased, markers of left ventricular (LV) systolic function and forward flow decreased, including left ventricular ejection fraction (LVEF), fractional shortening, left ventricular outflow tract (LVOT) velocity time integral (VTI), stroke volume, LV stroke work index, and cardiac power output. Biventricular filling pressures increased, and markers of right ventricular function worsened with rising SI. Most TTE measurements reflecting LV function and forward flow were inversely correlated with SI, including LV stroke work index (r = −0.59) and LVOT VTI (r = −0.41), as were both systemic vascular resistance index (r = −0.43) and LVEF (r = −0.23). ConclusionCICU patients with elevated SI have worse biventricular function and systemic hemodynamics, particularly decreased stroke volume and related calculated TTE parameters. The SI is an easily available marker that can be used to identify CICU patients with unfavorable hemodynamics who may require further assessment.

Racial and Ethnic Variation in ECMO Utilization and Outcomes in Pediatric Cardiac ICU Patients

BackgroundPrevious studies have reported racial disparities in extracorporeal membrane oxygenation (ECMO) utilization in pediatric cardiac patients. ObjectivesThe objective of this study was to determine if there was racial/ethnic variation in ECMO utilization and, if so, whether mortality was mediated by differences in ECMO utilization. MethodsThis is a multicenter, retrospective cohort study of the Pediatric Cardiac Critical Care Consortium clinical registry. Analyses were stratified by hospitalization type (medical vs surgical). Logistic regression models were adjusted for confounders and evaluated the association between race/ethnicity with ECMO utilization and mortality. Secondary analyses explored interactions between race/ethnicity, insurance, and socioeconomic status with ECMO utilization and mortality. ResultsA total of 50,552 hospitalizations from 34 hospitals were studied. Across all hospitalizations, 2.9% (N = 1,467) included ECMO. In medical and surgical hospitalizations, Black race and Hispanic ethnicity were associated with severity of illness proxies. In medical hospitalizations, race/ethnicity was not associated with the odds of ECMO utilization. Hospitalizations of other race had higher odds of mortality (adjusted odds ratio [aOR]: 1.61; 95% CI: 1.22-2.12; P = 0.001). For surgical hospitalizations, Black (aOR: 1.24; 95% CI: 1.02-1.50; P = 0.03) and other race (aOR: 1.50; 95% CI: 1.17-1.93; P = 0.001) were associated with higher odds of ECMO utilization. Hospitalizations of Hispanic patients had higher odds of mortality (aOR: 1.31; 95% CI: 1.03-1.68; P = 0.03). No significant interactions were demonstrated between race/ethnicity and socioeconomic status indicators with ECMO utilization or mortality. ConclusionsBlack and other races were associated with increased ECMO utilization during surgical hospitalizations. There were racial/ethnic disparities in outcomes not explained by differences in ECMO utilization. Efforts to mitigate these important disparities should include other aspects of care.

Systematic Assessment of Shock Severity in Postoperative Cardiac Surgery Patients

BackgroundThe Society for Cardiovascular Angiography and Interventions (SCAI) shock classification has been shown to provide robust mortality risk stratification in a variety of cardiovascular patients. ObjectivesThis study sought to evaluate the SCAI shock classification in postoperative cardiac surgery intensive care unit (CSICU) patients. MethodsThis study retrospectively analyzed 26,792 postoperative CSICU admissions at a heart center between 2012 and 2022. Patients were classified into SCAI shock stages A to E using electronic health record data. Moreover, the impact of late deterioration (LD) as an additional risk modifier was investigated. ResultsThe proportions of patients in SCAI shock stages A to E were 24.4%, 18.8%, 8.4%, 35.5%, and 12.9%, and crude hospital mortality rates were 0.4%, 0.6%, 3.3%, 4.9%, and 30.2%, respectively. Similarly, the prevalence of postoperative complications and organ dysfunction increased across SCAI shock stages. After multivariable adjustment, each higher SCAI shock stage was associated with increased hospital mortality (adjusted OR: 1.26-16.59) compared with SCAI shock stage A, as was LD (adjusted OR: 8.2). The SCAI shock classification demonstrated a strong diagnostic performance for hospital mortality (area under the receiver operating characteristic: 0.84), which noticeably increased when LD was incorporated into the model (area under the receiver operating characteristic: 0.90). ConclusionsThe SCAI shock classification effectively risk-stratifies postoperative CSICU patients for mortality, postoperative complications, and organ dysfunction. Its application could, therefore, be extended to the field of cardiac surgery as a triage tool in postoperative care and as a selection criterion in research.