Using Machine Learning for Early Prediction of Cardiogenic Shock in Patients With Acute Heart Failure
Ontology highlight
ABSTRACT:
SUBMITTER: Rahman F
PROVIDER: S-EPMC11307874 | biostudies-literature | 2022 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Cardiogenic shock (CS) is a severe condition with in-hospital mortality of up to 50%. Patients who develop CS may have previous cardiac history, but that may not always be the case, adding to the challenges in optimally identifying and managing these patients. Patients may present to a medical facility with CS or develop CS while in the emergency department (ED), in a general inpatient ward (WARD) or in the critical care unit (CC). While different clinical pathways for management exist once CS is recognized, there are challenges in identifying the patients in a timely manner, in all settings, in a timeframe that will allow proper management. We therefore developed and evaluated retrospectively a machine learning model based on the XGBoost (XGB) algorithm which runs automatically on patient data from the electronic health record (EHR). The algorithm was trained on 8 years of de-identified data (from 2010 to 2017) collected from a large regional healthcare system. The input variables include demographics, vital signs, laboratory values, some orders, and specific pre-existing diagnoses. The model was designed to make predictions 2 h prior to the need of first CS intervention (inotrope, vasopressor, or mechanical circulatory support). The algorithm achieves an overall area under curve (AUC) of 0.87 (0.81 in CC, 0.84 in ED, 0.97 in WARD), which is considered useful for clinical use. The algorithm can be refined based on specific elements defining patient subpopulations, for example presence of acute myocardial infarction (AMI) or congestive heart failure (CHF), further increasing its precision when a patient has these conditions. The top-contributing risk factors learned by the model are consistent with existing clinical findings. Our conclusion is that a useful machine learning model can be used to predict the development of CS. This manuscript describes the main steps of the development process and our results.
| S-EPMC9326048 | biostudies-literature
Project description:ObjectiveThe management of cardiogenic shock (CS) in the elderly remains a major clinical challenge. Existing clinical prediction models have not performed well in assessing the prognosis of elderly patients with CS. This study aims to build a predictive model, which could better predict the 30-day mortality of elderly patients with CS.MethodsWe extracted data from the Medical Information Mart for Intensive Care III version 1.4 (MIMIC-III) as the training set and the data of validation sets were collected from the Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University. Three models, including the cox regression model, the Least Absolute Shrinkage and Selection Operator (LASSO) regression model, and the CoxBoost model, were established using the training set. Through the comparison of area under the receiver operating characteristic (ROC) curve (AUC), C index, net reclassification improvement (NRI), integrated discrimination improvement (IDI), and median improvement in risk score, the best model was selected. Then for external validation, compared the best model with the simplified acute physiology score II (SAPSII) and the CardShock risk score.ResultsA total of 919 patients were included in the study, of which 804 patients were in the training set and 115 patients were in the verification set. Using the training set, we built three models: the cox regression model including 6 predictors, the LASSO regression model including 4 predictors, and the CoxBoost model including 16 predictors. Among them, the CoxBoost model had good discrimination [AUC: 0.730; C index: 0.6958 (0.6657, 0.7259)]. Compared with the CoxBoost model, the NRI, IDI, and median improvement in risk score of other models were all<0. In the validation set, the CoxBoost model was also well-discriminated [AUC: 0.770; C index: 0.7713 (0.6751, 0.8675)]. Compared with the CoxBoost model, the NRI, IDI, and median improvement in risk score of SAPS II and the CardShock risk score were all < 0. And we constructed a dynamic nomogram to visually display the model.ConclusionIn conclusion, this study showed that in predicting the 30-day mortality of elderly CS patients, the CoxBoost model was superior to the Cox regression model, LASSO regression model, SAPS II, and the CardShock risk score.
| S-EPMC9120613 | biostudies-literature
Project description:Heart failure is a devastating disease that has high mortality rates and a negative impact on quality of life. Heart failure patients often experience emergency readmission after an initial episode, often due to inadequate management. A timely diagnosis and treatment of underlying issues can significantly reduce the risk of emergency readmissions. The purpose of this project was to predict emergency readmissions of discharged heart failure patients using classical machine learning (ML) models based on Electronic Health Record (EHR) data. The dataset used for this study consisted of 166 clinical biomarkers from 2008 patient records. Three feature selection techniques were studied along with 13 classical ML models using five-fold cross-validation. A stacking ML model was trained using the predictions of the three best-performing models for final classification. The stacking ML model provided an accuracy, precision, recall, specificity, F1-score, and area under the curve (AUC) of 89.41%, 90.10%, 89.41%, 87.83%, 89.28%, and 0.881, respectively. This indicates the effectiveness of the proposed model in predicting emergency readmissions. The healthcare providers can intervene pro-actively to reduce emergency hospital readmission risk and improve patient outcomes and decrease healthcare costs using the proposed model.
| S-EPMC10252957 | biostudies-literature
Project description:AimsThe present analysis from the multicentre prospective Altshock-2 registry aims to better define clinical features, in-hospital course, and management of cardiogenic shock complicating acutely decompensated heart failure (ADHF-CS) as compared with that complicating acute myocardial infarction (AMI-CS).Methods and resultsAll patients with AMI-CS or ADHF-CS enrolled in the Altshock-2 registry between March 2020 and February 2022 were selected. The primary objective was the characterization of ADHF-CS patients as compared with AMI-CS. In-hospital length of stay and mortality were secondary endpoints. One-hundred-ninety of the 238 CS patients enrolled in the aforementioned period were considered for the present analysis: 101 AMI-CS (80% ST-elevated myocardial infarction and 20% non-ST-elevated myocardial infarction) and 89 ADHF-CS. As compared with AMI-CS, ADHF-CS patients were younger [63 (IQR 59-76) vs. 67 (IQR 54-73) years, P = 0.01], but presented with higher creatinine [1.6 (IQR 1.0-2.6) vs. 1.2 (IQR 1.0-1.4) mg/dL, P < 0.001], bilirubin [1.3 (IQR 0.9-2.3) vs. 0.6 (IQR 0.4-1.1) mg/dL, P = 0.01], and central venous pressure values [14 mmHg (IQR 8-12) vs. 10 mmHg (IQR 7-14),P = 0.01]. Norepinephrine was the most common catecholamine used in AMI-CS (79.3%), whereas epinephrine was used more commonly in ADHF-CS (65.5%); 75.8% vs. 46.6% received a temporary mechanical support in AMI-CS and ADHF-CS, respectively (P < 0.001). Length of hospital stay was longer in the latter [28 (IQR 13-48) vs. 17 (IQR 9-29) days, P = 0.001]. Heart replacement therapies were more frequently used in the ADHF-CS group (heart transplantation 13.5% vs. 0% and left ventricular assist device 11% vs. 2%, P < 0.01 and 0.01, respectively). In-hospital mortality was 41.1% (38.6% AMI-CS vs. 43.8% ADHF-CS, P = 0.5).ConclusionsADHF-CS is characterized by a higher prevalence of end-organ and biventricular dysfunction at presentation, a longer hospital length of stay, and higher need of heart replacement therapies when compared with AMI-CS. In-hospital mortality was similar between the two aetiologies. Our data warrant development of new management protocols focused on CS aetiology.
| S-EPMC10682868 | biostudies-literature
Project description:BackgroundAcute decompensated heart failure (ADHF) complicated by cardiogenic shock (CS) has unique pathophysiological background requiring specific patient stratification, management and therapeutic targets. Accordingly, the aim of this study was to derive a simple stratification tool to predict survival in patients with ADHF complicated by CS.Methods and resultsUsing logistic regression, univariable testing was performed to identify the variables potentially associated with 28-day mortality. We propose a new logistic model (ALC-Shock score) based on three easy parameters (age, serum creatinine and serum lactate at the ICU admission) as a powerful predictor of survival or successful bridge to heart replacement therapy at 28-day follow-up in this specific population. A multivariable analysis (logistic model) was performed to evaluate the association between selected variables and outcome (overall death at 28-day follow up). The score was then validated in a different cohort of 93 ADHF-CS patients and compared to a previous developed score (the Cardshock score).Overall, 28-day mortality was 34%. The ALC-shock score showed better discrimination (Area Under the Curve-AUC- 0.82; 95% CI 0.73-0.91) as compared to the Cardshock score (AUC 0.67; 95% CI 0.55-0.79) (p = 0.009) to predict 28-days overall mortality. In the validation cohort the AUC for the ALC-shock score was 0.66.ConclusionsA simple score including age, lactates and creatinine on admission could be considered to predict short-term mortality in CS-ADHF patients in order to drive towards a treatment intensification.
| S-EPMC8188054 | biostudies-literature
Project description:AimsMyocardial infarction and heart failure are major cardiovascular diseases that affect millions of people in the USA with morbidity and mortality being highest among patients who develop cardiogenic shock. Early recognition of cardiogenic shock allows prompt implementation of treatment measures. Our objective is to develop a new dynamic risk score, called CShock, to improve early detection of cardiogenic shock in the cardiac intensive care unit (ICU).Methods and resultsWe developed and externally validated a deep learning-based risk stratification tool, called CShock, for patients admitted into the cardiac ICU with acute decompensated heart failure and/or myocardial infarction to predict the onset of cardiogenic shock. We prepared a cardiac ICU dataset using the Medical Information Mart for Intensive Care-III database by annotating with physician-adjudicated outcomes. This dataset which consisted of 1500 patients with 204 having cardiogenic/mixed shock was then used to train CShock. The features used to train the model for CShock included patient demographics, cardiac ICU admission diagnoses, routinely measured laboratory values and vital signs, and relevant features manually extracted from echocardiogram and left heart catheterization reports. We externally validated the risk model on the New York University (NYU) Langone Health cardiac ICU database which was also annotated with physician-adjudicated outcomes. The external validation cohort consisted of 131 patients with 25 patients experiencing cardiogenic/mixed shock. CShock achieved an area under the receiver operator characteristic curve (AUROC) of 0.821 (95% CI 0.792-0.850). CShock was externally validated in the more contemporary NYU cohort and achieved an AUROC of 0.800 (95% CI 0.717-0.884), demonstrating its generalizability in other cardiac ICUs. Having an elevated heart rate is most predictive of cardiogenic shock development based on Shapley values. The other top 10 predictors are having an admission diagnosis of myocardial infarction with ST-segment elevation, having an admission diagnosis of acute decompensated heart failure, Braden Scale, Glasgow Coma Scale, blood urea nitrogen, systolic blood pressure, serum chloride, serum sodium, and arterial blood pH.ConclusionThe novel CShock score has the potential to provide automated detection and early warning for cardiogenic shock and improve the outcomes for millions of patients who suffer from myocardial infarction and heart failure.
| S-EPMC11214586 | biostudies-literature
Project description:Gene expression profiles were generated from 199 primary breast cancer patients. Samples 1-176 were used in another study, GEO Series GSE22820, and form the training data set in this study. Sample numbers 200-222 form a validation set. This data is used to model a machine learning classifier for Estrogen Receptor Status. RNA was isolated from 199 primary breast cancer patients. A machine learning classifier was built to predict ER status using only three gene features.
2013-01-01 | E-GEOD-29210 | biostudies-arrayexpress
Project description:BackgroundScarce data on factors related to discharge disposition in patients hospitalized for acute heart failure (AHF) were available, and we sought to develop a parsimonious and simple predictive model for non-home discharge via machine learning.MethodsThis observational cohort study using a Japanese national database included 128,068 patients admitted from home for AHF between April 2014 and March 2018. The candidate predictors for non-home discharge were patient demographics, comorbidities, and treatment performed within 2 days after hospital admission. We used 80% of the population to develop a model using all 26 candidate variables and using the variable selected by 1 standard-error rule of Lasso regression, which enhances interpretability, and 20% to validate the predictive ability.ResultsWe analyzed 128,068 patients, and 22,330 patients were not discharged to home; 7,879 underwent in-hospital death and 14,451 were transferred to other facilities. The machine-learning-based model consisted of 11 predictors, showing a discrimination ability comparable to that using all the 26 variables (c-statistic: 0.760 [95% confidence interval, 0.752-0.767] vs. 0.761 [95% confidence interval, 0.753-0.769]). The common 1SE-selected variables identified throughout all analyses were low scores in activities of daily living, advanced age, absence of hypertension, impaired consciousness, failure to initiate enteral alimentation within 2 days and low body weight.ConclusionsThe developed machine learning model using 11 predictors had a good predictive ability to identify patients at high risk for non-home discharge. Our findings would contribute to the effective care coordination in this era when HF is rapidly increasing in prevalence.
| S-EPMC10955024 | biostudies-literature
Project description:BackgroundThe beta-blocker (BB) initiation in acute heart failure (AHF) patients is still controversial. Some show the benefit of BB employment in decreasing the mortality outcome. This study aims to assess the safety and efficacy of in-hospital and long-term outcomes of BB initiation in AHF hospitalized patients. We searched multiple databases examining the outcome of AHF patients who had administered BB as the therapy initiation. Primary outcomes were all-cause mortality, composite endpoint after BB initiation when hospitalized, and post-discharge all-cause mortality. The secondary outcomes were adverse events after in-hospital BB initiation, including hypotension and symptomatic bradycardia after BB initiation when hospitalization and rehospitalization.ResultsEight cohort studies with 16,639 patients suffering from AHF and cardiogenic shock, with 9923 participants allocated to the early BB group and 6,713 patients in the control group. The follow-up durations ranged from 2 to 24 months. Early BB administration significantly reduced in-hospital composite endpoints (RR: 0.42; 95% CI (0.30-0.58); p < 0.001), in-hospital all-cause mortality (RR: 0.43; 95% CI (0.31-0.61); p < 0.001), discharge mortality (RR: 0.51; 95% CI (0.41-0.63); p < 0.001), and rehospitalization (RR: 0.57; 95% CI (0.44-0.74); p < 0.001). There were no discernible differences in in-hospital BB-related adverse events between the two groups (p = 0.13). Subgroup analyses conducted on AHF patients presenting with cardiogenic shock revealed no significant differences in in-hospital composite endpoint and in-hospital mortality, and similar results were shown in the naive BB population.ConclusionsThe BB initiation in AHF patients shows advantages in efficacy and safety outcome.
| S-EPMC11399533 | biostudies-literature
Project description:AimsHeart failure (HF) is an impending complication to myocardial infarction. We hypothesized that the degree of complement activation reflects severity of HF following acute myocardial infarction.Methods and resultsThe LEAF trial (LEvosimendan in Acute heart Failure following myocardial infarction) evaluating 61 patients developing HF within 48 h after percutaneous coronary intervention-treated ST-elevation myocardial infarction herein underwent a post hoc analysis. Blood samples were drawn from inclusion to Day 5 and at 42 day follow-up, and biomarkers were measured with enzyme immunoassays. Regional myocardial contractility was measured by echocardiography as wall motion score index (WMSI). The cardiogenic shock group (n = 9) was compared with the non-shock group (n = 52). Controls (n = 44) were age-matched and sex-matched healthy individuals. C4bc, C3bc, C3bBbP, and sC5b-9 were elevated in patients at inclusion compared with controls (P < 0.01). The shock group had higher levels compared with the non-shock group for all activation products except C3bBbP (P < 0.05). At Day 42, all products were higher in the shock group (P < 0.05). In the shock group, sC5b-9 correlated significantly with WMSI at baseline (r = 0.68; P = 0.045) and at Day 42 (r = 0.84; P = 0.036). Peak sC5b-9 level correlated strongly with WMSI at Day 42 (r = 0.98; P = 0.005). Circulating endothelial cell activation markers sICAM-1 and sVCAM-1 were higher in the shock group during the acute phase (P < 0.01), and their peak levels correlated with sC5b-9 peak level in the whole HF population (r = 0.32; P = 0.014 and r = 0.30; P = 0.022, respectively).ConclusionsComplement activation discriminated cardiogenic shock from non-shock in acute ST-elevation myocardial infarction complicated by HF and correlated with regional contractility and endothelial cell activation, suggesting a pathogenic role of complement in this condition.
| S-EPMC5933968 | biostudies-literature