Predictors Of In-hospital Mortality Research Articles

Clinical usefulness of NT-proBNP as a prognostic factor for septic shock patients presenting to the emergency department

Plasma N-terminal prohormone of brain natriuretic peptide (NT-proBNP) level is primarily used as a biomarker for left ventricular (LV) dysfunction. It is influenced by various conditions, such as myocardial strain and situations affecting the clearance of NT-proBNP, including sepsis and shock. In this study, we investigated the appropriateness of NT-proBNP as a prognostic factor for septic shock. Patients with septic shock who visited the emergency department of the Ewha Womans’ University Mokdong Hospital between January 1, 2018, and December 31, 2020, were classified into the survival group (those who survived in the hospital and were discharged) and the death group (those who died in the hospital). The effectiveness of NT-proBNP, lactate, and blood urea nitrogen as predictive factors of in-hospital mortality was evaluated using the area under the receiver operating characteristic (AUROC) curve. The AUROC curve was 0.678 and 0.648 for lactate and NT-proBNP, respectively, with lactate showing the highest value. However, there was no significant difference between lactate and NT-proBNP levels in the comparison of their AUROC curve (p = 0.6278). NT-proBNP could be a useful predictor of in-hospital mortality in patients with septic shock who present to the emergency department.

Proposal and Validation of a Clinically Relevant Modification of the Japanese Association for Acute Medicine Disseminated Intravascular Coagulation Diagnostic Criteria for Sepsis.

Japanese Association for Acute Medicine (JAAM) disseminated intravascular coagulation (DIC) criteria were launched nearly 20 years ago. Following the revised conceptual definition of sepsis and subsequent omission of systemic inflammatory response syndrome (SIRS) score from the latest sepsis diagnostic criteria, we omitted the SIRS score and proposed a modified version of JAAM DIC criteria, the JAAM-2 DIC criteria. To validate and compare performance between new JAAM-2 DIC criteria and conventional JAAM DIC criteria for sepsis. We used three datasets containing adult sepsis patients from a multicenter nationwide Japanese cohort study (J-septic DIC, FORECAST, and SPICE-ICU registries). JAAM-2 DIC criteria omitted the SIRS score and set the cutoff value at ≥3 points. Receiver operating characteristic (ROC) analyses were performed between the two DIC criteria to evaluate prognostic value. Associations between in-hospital mortality and anticoagulant therapy according to DIC status were analyzed using propensity score weighting to compare significance of the criteria in determining introduction of anticoagulants against sepsis. Final study cohorts of the datasets included 2,154, 1,065, and 608 sepsis patients, respectively. ROC analysis revealed that curves for both JAAM and JAAM-2 DIC criteria as predictors of in-hospital mortality were almost consistent. Survival curves for the anticoagulant and control groups in the propensity score-weighted prediction model diagnosed using the two criteria were also almost entirely consistent. JAAM-2 DIC criteria were equivalent to JAAM DIC criteria regarding prognostic and diagnostic values for initiating anticoagulation. The newly proposed JAAM-2 DIC criteria could be potentially alternative criteria for sepsis management.

Development and Validation of an Explainable Deep Learning Model to Predict In-Hospital Mortality for Patients With Acute Myocardial Infarction: Algorithm Development and Validation Study.

Acute myocardial infarction (AMI) is one of the most severe cardiovascular diseases and is associated with a high risk of in-hospital mortality. However, the current deep learning models for in-hospital mortality prediction lack interpretability. This study aims to establish an explainable deep learning model to provide individualized in-hospital mortality prediction and risk factor assessment for patients with AMI. In this retrospective multicenter study, we used data for consecutive patients hospitalized with AMI from the Chongqing University Central Hospital between July 2016 and December 2022 and the Electronic Intensive Care Unit Collaborative Research Database. These patients were randomly divided into training (7668/10,955, 70%) and internal test (3287/10,955, 30%) data sets. In addition, data of patients with AMI from the Medical Information Mart for Intensive Care database were used for external validation. Deep learning models were used to predict in-hospital mortality in patients with AMI, and they were compared with linear and tree-based models. The Shapley Additive Explanations method was used to explain the model with the highest area under the receiver operating characteristic curve in both the internal test and external validation data sets to quantify and visualize the features that drive predictions. A total of 10,955 patients with AMI who were admitted to Chongqing University Central Hospital or included in the Electronic Intensive Care Unit Collaborative Research Database were randomly divided into a training data set of 7668 (70%) patients and an internal test data set of 3287 (30%) patients. A total of 9355 patients from the Medical Information Mart for Intensive Care database were included for independent external validation. In-hospital mortality occurred in 8.74% (670/7668), 8.73% (287/3287), and 9.12% (853/9355) of the patients in the training, internal test, and external validation cohorts, respectively. The Self-Attention and Intersample Attention Transformer model performed best in both the internal test data set and the external validation data set among the 9 prediction models, with the highest area under the receiver operating characteristic curve of 0.86 (95% CI 0.84-0.88) and 0.85 (95% CI 0.84-0.87), respectively. Older age, high heart rate, and low body temperature were the 3 most important predictors of increased mortality, according to the explanations of the Self-Attention and Intersample Attention Transformer model. The explainable deep learning model that we developed could provide estimates of mortality and visual contribution of the features to the prediction for a patient with AMI. The explanations suggested that older age, unstable vital signs, and metabolic disorders may increase the risk of mortality in patients with AMI.

Predicting in-hospital mortality in infective endocarditis: insights from complete blood count parameters and NT-proBNP

Abstract Funding Acknowledgements None. Introduction Infective endocarditis (IE) is a rare condition with an increasing incidence rate and complexity of cases over the last few years, accompanied by a high mortality rate ranging from 18% to 25%. Recent studies have shown the utility of new outcome predictors based on complete blood count (CBC), such as red blood cell distribution width (RDW) and neutrophils/lymphocytes ratio (NLR). Drawing retrospectively from an 8-year experience, we provide a perspective from a tertiary center on predictors of in-hospital mortality based on basic laboratory test results in patients with valve-related IE. Purpose The aim of the study was to assess the clinical accuracy of CBC-based parameters as in-hospital mortality predictors. Methods From a tertiary center database (serving a stable population of 2.5 million citizens), we selected all patients hospitalized due to valve-related infective endocarditis between 2015 and 2022. We analyzed clinical data (including sex, age, IE localization, and etiological factors, number of cases, management and mortality), laboratory test results at admission (including CBC parameters, C-reactive protein - CRP, procalcitonin - PCT, N-terminal prohormone of brain natriuretic peptide - NT-proBNP, estimated glomerular filtration rate - eGFR, Creatinine - Crea, urea - UR). For three NT-proBNP, NLR, and RDW we provided a further analysis to compare their clinical accuracy. Univariable logistic regression was used to assess the impact of specific factors on mortality. To determine the optimal cutoff values for predicting mortality, we analyzed the area under the curve (AUC) from receiver operating characteristic (ROC) curves and used Youden's index. P-values < 0.05 were considered statistically significant. Results A total of 194 consecutive patients with confirmed IE were included in the study, the majority of whom were men (75.3%) with an overall median age of 62 years. The overall mortality rate was 27.8%. The most common localization was left-sided IE (92.3%). In 39.7% of cases, the etiological factor was unidentified, while Staphylococcus spp. were the most common bacteria among the identified cases. We identified significant impacts on mortality in most of the analyzed laboratory tests (RDW, platelet count, hemoglobin, NLR, PCT, NT-proBNP, eGFR, Crea, UR). Results showed that for every 1000pg/ml of NT-proBNP the risk of in-hospital death increases by 6%. ROC analysis was performed for the three parameters (RDW, NLR, and NT-proBNP), and each parameter showed an acceptable area under the curve - AUC (RDW: 0.69; NLR: 0.62; NT-proBNP: 0.78). Although we found the highest AUC for NT-proBNP (p<0.001). Conclusions Although CBC parameters are easy to assess and useful in predicting in-hospital death in IE, the accuracy of NT-proBNP is higher. The management of IE cases requires a complex approach, and NT-proBNP, which also reflects the clinical state of patients, still remains a more useful tool.In-hospital deaths ROC analysis

Long-term prognostic significance of stress hyperglycemia in patients with ST-elevation myocardial infarction

Abstract Funding Acknowledgements None. Background/aim Stress hyperglycemia is defined as hyperglycemia that occurs in critically ill patients without diabetes mellitus (DM). It is associated with adverse in-hospital outcomes in patients with ST elevation myocardial infarction (STEMI). The aim of the present study is to evaluate short- and long-term prognostic impact of stress hyperglycemia in STEMI patients. Method We analysed 2505 consecutive STEMI patients treated with successful primary PCI (pPCI) without previous DM and without cardiogenic shock at admission. Successful pPCI was defined as post-procedural flow TIMI 3. Stress hyperglycemia was defined as plasma glucose level above 7.8mmol/L at any time during hospitalization. The follow-up period was 6 years. Results Stress hyperglycemia at admission was registered in 685 (27.3%) patients. Compared with patients without stress hyperglycemia, patients with stress hyperglycemia were older and presented more often with heart failure (Killip class II and III); they were more likely to have hypertension, chronic kidney disease, three vessel coronary disease (at initial angiogram) and lower predischarge left ventricular ejection fraction (EF). In-hospital, 30-day and 6-year mortality were significantly higher in patients with stress hyperglycemia vs patients without stress hyperglycemia: 7.1% vs 1.5%, 7.5% vs 1.7% and 12.5% vs 4.6% (p<0.001), respectively as shown in Figure 1. In Cox regression analysis stress hyperglycemia was an independent predictor for in-hospital, 30-days and 6-year mortality- HR 2.19 (95% 1.49-3.94), p=0.018, HR 1,58(95%CI 1.15-2.85), p=0.020 and HR 1.52 (95%CI 1.32-2.17), p=0.030, respectively. Conclusion In STEMI patients without diabetes mellitus treated with successful primary PCI stress hyperglycemia is an independent predictor for short- and long-term mortality (up to 6-years). These findings emphasize the importance of close monitoring of blood glucose level in STEMI patients regardless of diabetic status.

Prevalence and Prognostic Impact of Stroke in a National Cohort of Infective Endocarditis.

Stroke is a common complication of infective endocarditis (IE). Our aim was to describe the prevalence and prognostic impact of stroke in a national prospective cohort of IE. Consecutive inclusion at 46 Spanish hospitals between 2008 and 2021. Out of 5667 IE cases, 1125 had acute stroke (19.8%): 811 ischemic strokes (618 cardioembolic strokes, 193 cardioembolic strokes with hemorrhagic transformation, 4 transient ischemic attacks, 3 lacunar infarctions), 125 intracranial hemorrhages, and 29 other neurological complications (cerebral abscesses, encephalitis, meningitis, seizures). Compared to patients without stroke, those with stroke had a similar mean age (69 years) but were more frequently female (68.2% vs. 63.7%, p=0.04) and had a higher incidence of intracardiac complications (35% vs 30%, p=0.01), surgical indication (69.9% vs 65.9%, p=0.001), in-hospital mortality (40.9% vs. 22.0%, p<0.001), and one-year mortality (46.2% vs 27.9%, p<0.001). The following variables were independently associated with stroke: mitral location (odds ratio [OR] 1.5, 95% confidence interval [CI] 1.34-1.8, p<0.001), vascular phenomenon (OR 2.9, 95% CI 2.4-3.6, p=0.0001), acute renal failure (OR 1.2, 95% CI 1.0-1.4, p=0.021), septic shock (OR 1.3, CI 1.1-1.6, p=0.007), sepsis (OR 1.3, CI 1.1-1.6, p=0.005), surgery indicated but not performed (OR 1.4, 95% CI 1.2-1.7, p<0.001), community-acquired IE (OR 1.2, 95% CI 1-1.4, p=0.017), and peripheral embolization (OR 1.6, CI 1.4-1.9, p <0.001). Stroke was an independent predictor of in-hospital (OR 2.1, 95% CI: 1.78-2.51, p<0.001) and one-year mortality (hazard ratio 1.9, 95% CI 1.6-2.5). One fifth of patients with IE have concomitant stroke. Stroke is associated with mortality.

Continuous patient state attention model for addressing irregularity in electronic health records

BackgroundIrregular time series (ITS) are common in healthcare as patient data is recorded in an electronic health record (EHR) system as per clinical guidelines/requirements but not for research and depends on a patient’s health status. Due to irregularity, it is challenging to develop machine learning techniques to uncover vast intelligence hidden in EHR big data, without losing performance on downstream patient outcome prediction tasks.MethodsIn this paper, we propose Perceiver, a cross-attention-based transformer variant that is computationally efficient and can handle long sequences of time series in healthcare. We further develop continuous patient state attention models, using Perceiver and transformer to deal with ITS in EHR. The continuous patient state models utilise neural ordinary differential equations to learn patient health dynamics, i.e., patient health trajectory from observed irregular time steps, which enables them to sample patient state at any time.ResultsThe proposed models’ performance on in-hospital mortality prediction task on PhysioNet-2012 challenge and MIMIC-III datasets is examined. Perceiver model either outperforms or performs at par with baselines, and reduces computations by about nine times when compared to the transformer model, with no significant loss of performance. Experiments to examine irregularity in healthcare reveal that continuous patient state models outperform baselines. Moreover, the predictive uncertainty of the model is used to refer extremely uncertain cases to clinicians, which enhances the model’s performance. Code is publicly available and verified at https://codeocean.com/capsule/4587224.ConclusionsPerceiver presents a computationally efficient potential alternative for processing long sequences of time series in healthcare, and the continuous patient state attention models outperform the traditional and advanced techniques to handle irregularity in the time series. Moreover, the predictive uncertainty of the model helps in the development of transparent and trustworthy systems, which can be utilised as per the availability of clinicians.

Point-of-Care Serum Proenkephalin as an Early Predictor of Mortality in Patients Presenting to the Emergency Department with Septic Shock

Background: The aim of the present study is to investigate the prognostic utility of point-of-care (POC)-measured proenkephalin (PENK), a novel biomarker, in terms of predicting in-hospital mortality in patients presenting to the emergency department (ED) with septic shock. Methods: Bedside PENK was measured in consecutive patients presenting to the ED with septic shock according to the Sepsis-3 clinical criteria. The association of PENK with inflammatory and routine biomarkers, and its role as a predictor of in-hospital mortality, was examined. Results: Sixty-one patients with septic shock [53% females, median age 83 years (IQR 71–88)] were evaluated. Median (IQR) values of creatinine, plasma lactate, soluble urokinase plasminogen activator receptor (SuPAR), procalcitonin and PENK were 1.7 (1.0–2.9) mg/dL, 3.6 (2.1–6.8) mmol/L, 13.1 (10.0–21.4) ng/mL, 2.06 (0.84–3.49) ng/mL, and 205 (129–425) pmol/L, respectively. LogPENK significantly correlated with LogLactate (rho = 0.369, p = 0.004), LogCreatinine (rho = 0.537, p &lt; 0.001), LogProcalcitonin (rho = 0.557, p &lt; 0.001), and LogSuPAR (rho = 0.327, p = 0.011). During hospitalization, 39/61 (64%) patients died. In a multivariable logistic regression model, logPENK was an independent predictor of in-hospital mortality (OR 11.9, 95% CI: 1.7–84.6, p = 0.013). Conclusion: POC PENK levels measured upon presentation to the ED strongly correlated with metabolic, renal and inflammatory biomarkers, and may serve as a predictor of in-hospital mortality in patients with septic shock.

Predictors of infective endocarditis associated in-hospital mortality in Ayder Comprehensive Specialized Hospital, Tigray, North Ethiopia: Microbiological,clinical features, and management profiles.

Infective endocarditis (IE) is a continuously evolving disease with a high mortality rate despite different advances in treatment. In Ethiopia, there is a paucity of data regarding IE. Therefore, this study is aimed at assessing IE-related in-hospital mortality and characterization of IE patients based on their microbiological, clinical features, and management profiles in the Ayder Comprehensive Specified Hospital (ACSH). We conducted a hospital-based prospective follow-up study with all consecutive sampling techniques for suspected infective endocarditis patients admitted to ACSH from January 2020 to February 2022. Echocardiography was performed, and three sets of blood samples for blood culture were taken as per the standard protocol. We also performed isolation of microbial etiologies and antimicrobial susceptibility tests. The data was analyzed using STATA version 16. Stepwise logistic regression was run to identify predictors of in-hospital mortality. Effects were measured through the odds ratio at the 5% level of significance. Seventy-four cases of suspected infective endocarditis were investigated; of these, 54 episodes fulfilled modified Duke's criteria. Rheumatic heart disease (RHD) (85.2%) was the most common underlying heart disease. Murmur (94.4%), fever (68.5%), and pallor (57.4%) were the most common clinical findings. Vegetation was present in 96.3% of episodes. Blood culture was positive only in 7 (13%) episodes. Complications occurred in 41 (75.9%) cases, with congestive heart failure being the most common. All patients were managed medically, with no surgical intervention. The in-hospital mortality was 14 (25.9%). IE-related in-hospital mortality was significantly associated with surgery recommendation and myalgia clinical symptoms. IE occurred relatively in a younger population, with RHD as the most common underlying heart disease. There was a high rate of culture-negative endocarditis, and the majority of patients were treated empirically. Mortality was high. The establishment of cardiac surgery and strengthening microbiology services should be given top priority.

Time to maximum amplitude of thromboelastography can predict mortality in patients with severe COVID-19: a retrospective observational study

ObjectiveTo predict mortality in severe patients with COVID-19 at admission to the intensive care unit (ICU) using thromboelastography (TEG).MethodsThis retrospective, two-center, observational study involved 87 patients with PCR-and chest CT-confirmed severe COVID-19 who were admitted to at Wuhan Huoshenshan Hospital and the 908th Hospital of Chinese PLA Logistic Support Force between February 2020 and February 2023. Clinic demographics, laboratory results, and outcomes were compared between those who survived and those who died during hospitalization.ResultsThromboelastography showed that of the 87 patients, 14 were in a hypercoagulable state, 25 were in a hypocoagulable state, and 48 were normal, based on the time to maximum amplitude (TMA). Patients who died showed significantly lower α angle, but significantly longer R-time, K-time and TMA than patients who survived. Random forest selection showed that K-time, TMA, prothrombin time (PT), international normalized ratio (INR), D-dimer, C-reactive protein (CRP), aspartate aminotransferase (AST), and total bilirubin (Tbil) were significant predictors. Multivariate logistic regression identified that TMA and CRP were independently associated with mortality. TMA had a greater predictive power than CRP levels based on time-dependent AUCs. Patients with TMA ≥ 26.4 min were at significantly higher risk of mortality (hazard ratio 3.99, 95% Confidence Interval, 1.92–8.27, p &amp;lt; 0.01).ConclusionTMA ≥26.4 min at admission to ICU may be an independent predictor of in-hospital mortality for patients with severe COVID-19.

Triglyceride-glucose index as a potential predictor for in-hospital mortality in critically ill patients with intracerebral hemorrhage: a multicenter, case-control study.

The correlation between the triglyceride-glucose index (TyG) and the prognosis of ischemic stroke has been well established. This study aims to assess the influence of the TyG index on the clinical outcomes of critically ill individuals suffering from intracerebral hemorrhage (ICH). Patients diagnosed with ICH were retrospectively retrieved from the Medical Information Mart for Intensive Care (MIMIC-IV) and the eICU Collaborative Research Database (eICU-CRD). Various statistical methods, including restricted cubic spline (RCS) regression, multivariable logistic regression, subgroup analysis, and sensitivity analysis, were employed to examine the relationship between the TyG index and the primary outcomes of ICH. A total of 791 patients from MIMIC-IV and 1,113 ones from eICU-CRD were analyzed. In MIMIC-IV, the in-hospital and ICU mortality rates were 14% and 10%, respectively, while in eICU-CRD, they were 16% and 8%. Results of the RCS regression revealed a consistent linear relationship between the TyG index and the risk of in-hospital and ICU mortality across the entire study population of both databases. Logistic regression analysis revealed a significant positive association between the TyG index and the likelihood of in-hospital and ICU death among ICH patients in both databases. Subgroup and sensitivity analysis further revealed an interaction between patients' age and the TyG index in relation to in-hospital and ICU mortality among ICH patients. Notably, for patients over 60years old, the association between the TyG index and the risk of in-hospital and ICU mortality was more pronounced compared to the overall study population in both MIMIC-IV and eICU-CRD databases, suggesting a synergistic effect between old age (over 60years) and the TyG index on the in-hospital and ICU mortality of patients with ICH. This study established a positive correlation between the TyG index and the risk of in-hospital and ICU mortality in patients over 60years who diagnosed with ICH, suggesting that the TyG index holds promise as an indicator for risk stratification in this patient population.

Prognostic significance of preoperative osteosarcopenia on patient' outcomes after emergency surgery for gastrointestinal perforation.

Sarcopenia is a prognostic predictor in emergency surgery. However, there are no reports on the relationship between osteopenia and in-hospital mortality. This study clarified the effect of preoperative osteosarcopenia on patients with gastrointestinal perforation after emergency surgery. We included 216 patients with gastrointestinal perforations who underwent emergency surgery between January 2013 and December 2022. Osteopenia was evaluated by measuring the pixel density in the mid-vertebral core of the 11th thoracic vertebra. Sarcopenia was evaluated by measuring the area of the psoas muscle at the level of the third lumbar vertebra. Osteosarcopenia is defined as the combination of osteopenia and sarcopenia. Osteosarcomas were identified in 42 patients. Among patients with osteosarcopenia, older and female patients and those with an American Society of Anesthesiologists Physical Status of ≥ 3 were significantly more common, and the body mass index, hemoglobin value, and albumin level were significantly lower in these patients than in patients without osteosarcopenia. Furthermore, the osteosarcopenia group presented with more postoperative complications than patients without osteosarcopenia (P < 0.01). In the multivariate analysis, age ≥ 74years old (P = 0.04) and osteosarcopenia (P = 0.04) were independent and significant predictors of in-hospital mortality. Preoperative osteosarcopenia is a risk factor of in-hospital mortality in patients with gastrointestinal perforation after emergency surgery.

A COVID-19 specific multiparametric and ECG-based score for the prediction of in-hospital mortality: ELCOVID score.

We aimed to develop and validate a COVID-19 specific scoring system, also including some ECG features, to predict all-cause in-hospital mortality at admission. Patients were retrieved from the ELCOVID study (ClinicalTrials.gov identifier: NCT04367129), a prospective, multicenter Italian study enrolling COVID-19 patients between May to September 2020. For the model validation, we randomly selected two-thirds of participants to create a derivation dataset and we used the remaining one-third of participants as the validation set. Over the study period, 1014 hospitalized COVID-19 patients (mean age 74years, 61% males) met the inclusion criteria and were included in this analysis. During a median follow-up of 12 (IQR 7-22) days, 359 (35%) patients died. Age (HR 2.25 [95%CI 1.72-2.94], p < 0.001), delirium (HR 2.03 [2.14-3.61], p = 0.012), platelets (HR 0.91 [0.83-0.98], p = 0.018), D-dimer level (HR 1.18 [1.01-1.31], p = 0.002), signs of right ventricular strain (RVS) (HR 1.47 [1.02-2.13], p = 0.039) and ECG signs of previous myocardial necrosis (HR 2.28 [1.23-4.21], p = 0.009) were independently associated to in-hospital all-cause mortality. The derived risk-scoring system, namely EL COVID score, showed a moderate discriminatory capacity and good calibration. A cut-off score of ≥ 4 had a sensitivity of 78.4% and 65.2% specificity in predicting all-cause in-hospital mortality. ELCOVID score represents a valid, reliable, sensitive, and inexpensive scoring system that can be used for the prognostication of COVID-19 patients at admission and may allow the earlier identification of patients having a higher mortality risk who may be benefit from more aggressive treatments and closer monitoring.

Predicting in-hospital mortality among patients admitted with a diagnosis of heart failure: a machine learning approach.

Existing risk prediction models for hospitalized heart failure patients are limited. We identified patients hospitalized with a diagnosis of heart failure between 7 May 2013 and 26 April 2022 from a large academic, quaternary care medical centre (training cohort). Demographics, medical comorbidities, vitals, and labs were collected and were used to construct random forest machine learning models to predict in-hospital mortality. Models were compared with logistic regression, and to commonly used heart failure risk scores. The models were subsequently validated in patients hospitalized with a diagnosis of heart failure from a second academic, community medical centre (validation cohort). The entire cohort comprised 21802 patients, of which 14539 were in the training cohort and 7263 were in the validation cohort. The median age (25th-75th percentile) was 70 (58-82) for the entire cohort, 43.2% were female, and 6.7% experienced inpatient mortality. In the overall cohort, 7621 (35.0%) patients had heart failure with reduced ejection fraction (EF≤40%), 1271 (5.8%) had heart failure with mildly reduced EF (EF 41-49%), and 12910 (59.2%) had heart failure with preserved EF (EF≥50%). Random forest models in the validation cohort demonstrated a c-statistic (95% confidence interval) of 0.96 (0.95-0.97), sensitivity (SN) of 87.3%, and specificity (SP) of 90.6% for the prediction of in-hospital mortality. Models for those with HFrEF demonstrated a c-statistic of 0.96 (0.94-0.98), SN 88.2%, and SP 91.0%, and those for patients with HFpEF showed a c-statistic of 0.95 (0.93-0.97), SN 87.4%, and SP 89.5% for predicting in-hospital mortality. The random forest model significantly outperformed logistic regression (c-statistic 0.87, SN 75.9%, and SP 86.9%), and current existing risk scores including the Acute Decompensated Heart Failure National Registry risk score (c-statistic of 0.70, SN 69%, and SP 62%), and the Get With the Guidelines-Heart Failure risk score (c-statistic 0.69, SN 67%, and SP 63%); P<0.001 for comparison. Machine learning models built from commonly recorded patient information can accurately predict in-hospital mortality among patients hospitalized with a diagnosis of heart failure.

Predicting in-hospital mortality using Elixhauser comorbidity in patients underwent single and multiple coronary artery bypass surgery

BackgroundCoronary Artery Bypass Grafting (CABG) is a high-risk surgery. Cardiovascular diseases are strongly associated with comorbidities. This study aimed to assess the prediction of in-hospital mortality by comorbidities in patients who underwent CABG. MethodsThe National Inpatient Sample database was used to extract patients who received 1, 2, 3, and 4+ CABG between Q4 2015 and 2020. Best-fit model by logistic regressions was used to predict in-hospital mortality by Elixhauser Comorbidity Index (ECI). Moreover, age was adjusted in ECI prediction. ResultsThere were 190,524, 83,725, 48,147, and 13,540 patients who underwent 1, 2, 3, and 4+ CABG, respectively. In-hospital mortality was best predicted by ECI in 3 CABG (c-statistic = 0.63, 95 % CI = 0.62–0.65), followed by 4+ CABG (c-statistic = 0.63, 95 % CI = 0.60–0.66), 1 CABG (c-statistic = 0.62, 95 % CI = 0.61–0.63), and 2 CABG (c-statistic = 0.62, 95 % CI = 0.61–0.63). After adjusting for age, ECI adequately predicted in-hospital mortality in 4+ CABG (c-statistic = 0.72, 95 % CI = 0.69–0.75) and 3 CABG (c-statistic = 0.69, 95 % CI = 0.68–0.71). Predictive powers for age-adjusted ECI were comparable in 1 CABG (c-statistic=0.67, 95 % CI = 0.66–0.68) and 2 CABG (c-statistic = 0.67, 95 % CI = 0.65–0.68). ConclusionsECI was a moderate (c-statistic 0.6–0.7) predictor of in-hospital mortality in all CABG. Age-adjusted ECI could effectively predict in-hospital mortality, especially in patients who underwent 3 and 4+ CABG.

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.

Comparison of In-Hospital Mortality and Neurosurgical Intervention Between Intracranial Gunshot Wounds Arising From Long Guns and Handguns: A Propensity Score Matched Study

BACKGROUND AND OBJECTIVES: Patients with intracranial gunshot wounds (IC-GSWs) often present with severe neurological injuries requiring prompt neurological evaluation. Neurosurgical intervention is reserved for those with reasonable chances of survival. Handguns and long guns, such as shotguns and rifles, have differing mechanisms of injury which may influence surgical candidacy and outcomes. This study aims to compare rates and types of neurosurgical intervention and inpatient outcomes in patients with IC-GSWs handguns and long guns. METHODS: The National Trauma Data Bank was retrospectively queried for patients with IC-GSWs from 2017 to 2019. Patients with long gun IC-GSWs were propensity score matched with those with handgun IC-GSWs based on patient demographics, comorbidities, insurance status, injury extent and severity, and hospital trauma level. Group differences were compared using Student's t-tests and Pearson's χ2 tests, and multivariable logistic regression was used to identify predictors of in-hospital mortality. RESULTS: Overall, patients in the long gun group were more likely to undergo neurosurgical intervention (21% vs 17%, P = .02). Following propensity score matching, the long gun group had lower rates of in-hospital mortality (35% vs 43%, P &lt; .01), lower rates of cardiac arrest (5% vs 8%, P = .02), and lower rates of reoperation (0% vs 2%, P = .02) than the handgun group. In multivariable regression, independent predictors of survival included long gun IC-GSWs (odds ratio [OR] 0.65, CI 0.52-0.83), neurosurgical foreign body removal (OR 0.44, CI 0.33-0.58), intracranial debridement (OR 0.47, 0.33-0.67), and craniectomy (OR 0.46, CI 0.34-0.63). CONCLUSION: Patients with IC-GSWs present to the hospital with severe neurological injury. Neurosurgical intervention was independently associated with decreased mortality. After matching, patients with long gun IC-GSWs experienced lower in-hospital mortality rates compared with those from handguns. This study suggests that patients suffering from long gun IC-GSW may respond particularly well to neurosurgical intervention and firearm type should be considered when determining neurosurgical candidacy.

The relationship between red cell distribution width, serum calcium ratio, and in-hospital mortality among patients with acute respiratory failure: A retrospective cohort study of the MIMIC-IV database.

To investigate the impact of RDW/CA (the ratio of red cell distribution width to calcium) on in-hospital mortality in patients with acute respiratory failure (ARF). This retrospective cohort study analyzed the data of 6981 ARF patients from the Medical Information Mart for Intensive Care (MIMIC-IV) database 2.0. Critically ill participants between 2008 and 2019 at the Beth Israel Deaconess Medical Center in Boston. The primary outcome of interest was in-hospital mortality. A Cox proportional hazards regression model was used to determine whether the RDW/CA ratio independently correlated with in-hospital mortality. The Kaplan-Meier method was used to plot the survival curves of the RDW/CA. Subgroup analyses were performed to measure the mortality across various subgroups. After adjusting for potential covariates, we found that a higher RDW/CA was associated with an increased risk of in-hospital mortality (HR = 1.17, 95% CI: 1.01-1.35, P = .0365) in ARF patients. A nonlinear relationship was observed between RDW/CA and in-hospital mortality, with an inflection point of 1.97. When RDW/CA ≥ 1.97 was positively correlated with in-hospital mortality in patients with ARF (HR = 1.554, 95% CI: 1.183-2.042, P = .0015). The Kaplan-Meier curve indicated the higher survival rates for RDW/CA < 1.97 and the lower for RDW/CA ≥ 1.97 after adjustment for age, gender, body mass index, and ethnicity. RDW/CA is an independent predictor of in-hospital mortality in patients with ARF. Furthermore, a nonlinear relationship was observed between RDW/CA and in-hospital mortality in patients with ARF.

Time trajectories and within-subject correlations of matrix metalloproteinases 3, 8, 9, 10, 12, and 13 serum levels and their ability to predict mortality in polytraumatized patients: a pilot study

BackgroundManaging polytrauma victims poses a significant challenge to clinicians since applying the same therapy to patients with similar injury patterns may result in different outcomes. Using serum biomarkers hopefully allows for treating each multiple injured in the best possible individual way. Since matrix metalloproteinases (MMPs) play pivotal roles in various physiological processes, they might be a reliable tool in polytrauma care.MethodsWe evaluated 24 blunt polytrauma survivors and 12 fatalities (mean age, 44.2 years, mean ISS, 45) who were directly admitted to our Level I trauma center and stayed at the intensive care unit for at least one night. We determined their MMP3, MMP8, MMP9, MMP10, MMP12, and MMP13 serum levels at admission (day 0) and on days 1, 3, 5, 7, and 10.ResultsMedian MMP8, MMP9, and MMP12 levels immediately rose after the polytrauma occurred; however, they significantly decreased from admission to day 1 and significantly increased from day 1 to day 10, showing similar time trajectories and (very) strong correlations between each two of the three enzyme levels assessed at the same measurement point. For a two-day lag, autocorrelations were significant for MMP8 (− 0.512) and MMP9 (− 0.302) and for cross-correlations between MMP8 and MMP9 (− 0.439), MMP8 and MMP12 (− 0.416), and MMP9 and MMP12 (− 0.307). Moreover, median MMP3, MMP10, and MMP13 levels significantly increased from admission to day 3 and significantly decreased from day 3 to day 10, showing similar time trajectories and an (almost) strong association between every 2 levels until day 7. Significant cross-correlations were detected between MMP3 and MMP10 (0.414) and MMP13 and MMP10 (0.362). Finally, the MMP10 day 0 level was identified as a predictor for in-hospital mortality. Any increase of the MMP10 level by 200 pg/mL decreased the odds of dying by 28.5%.ConclusionsThe time trajectories of the highly varying individual MMP levels elucidate the involvement of these enzymes in the endogenous defense response following polytrauma. Similar time courses of MMP levels might indicate similar injury causes, whereas lead–lag effects reveal causative relations between several enzyme pairs. Finally, MMP10 abundantly released into circulation after polytrauma might have a protective effect against dying.

Cutoff of the reverse shock index multiplied by the Glasgow coma scale for predicting in-hospital mortality in adult patients with trauma: a retrospective cohort study

BackgroundEarly identification of patients at risk of potential death and timely transfer to appropriate healthcare facilities are critical for reducing the number of preventable trauma deaths. This study aimed to establish a cutoff value to predict in-hospital mortality using the reverse shock index multiplied by the Glasgow Coma Scale (rSIG).MethodsThis multicenter retrospective cohort study used data from 23 emergency departments in South Korea between January 2011 and December 2020. The outcome variable was the in-hospital mortality. The relationship between rSIG and in-hospital mortality was plotted using the shape-restricted regression spline method. To set a cutoff for rSIG, we found the point on the curve where mortality started to increase and the point where the slope of the mortality curve changed the most. We also calculated the cutoff value for rSIG using Youden's index.ResultsA total of 318,506 adult patients with trauma were included. The shape-restricted regression spline curve showed that in-hospital mortality began to increase when the rSIG value was less than 18.86, and the slope of the graph increased the most at 12.57. The cutoff of 16.5, calculated using Youden's index, was closest to the target under-triage and over-triage rates, as suggested by the American College of Surgeons, when applied to patients with an rSIG of 20 or less. In addition, in patients with traumatic brain injury, when the rSIG value was over 25, in-hospital mortality tended to increase as the rSIG value increased.ConclusionsWe propose an rSIG cutoff value of 16.5 as a predictor of in-hospital mortality in adult patients with trauma. However, in patients with traumatic brain injury, a high rSIG is also associated with in-hospital mortality. Appropriate cutoffs should be established for this group in the future.