Prediction model of in-hospital mortality in elderly patients with acute heart failure based on retrospective study

Admission Blood Glucose Associated with In-Hospital Mortality in Critically III Non-Diabetic Patients with Heart Failure: A Retrospective Study

Background

Heart failure (HF) is a primary public health issue associated with a high mortality rate. However, effective treatments still need to be developed. The optimal level of glycemic control in non-diabetic critically ill patients suffering from HF is uncertain. Therefore, this study examined the relationship between initial glucose levels and in-hospital mortality in critically ill non-diabetic patients with HF.

Methods

A total of 1159 critically ill patients with HF were selected from the Medical Information Mart for Intensive Care-III (MIMIC-III) data resource and included in this study. The association between initial glucose levels and hospital mortality in seriously ill non-diabetic patients with HF was analyzed using smooth curve fittings and multivariable Cox regression. Stratified analyses were performed for age, gender, hypertension, atrial fibrillation, CHD with no MI (coronary heart disease with no myocardial infarction), renal failure, chronic obstructive pulmonary disease (COPD), estimated glomerular filtration rate (eGFR), and blood glucose concentrations.

Results

The hospital mortality was identified as 14.9%. A multivariate Cox regression model, along with smooth curve fitting data, showed that the initial blood glucose demonstrated a U-shape relationship with hospitalized deaths in non-diabetic critically ill patients with HF. The turning point on the left side of the inflection point was HR 0.69, 95% CI 0.47-1.02, p = 0.068, and on the right side, HR 1.24, 95% CI 1.07-1.43, p = 0.003. Significant interactions existed for blood glucose concentrations (7-11 mmol/L) (p-value for interaction: 0.009). No other significant interactions were detected.

Conclusions

This study demonstrated a U-shape correlation between initial blood glucose and hospital mortality in critically ill non-diabetic patients with HF. The optimal level of initial blood glucose for non-diabetic critically ill patients with HF was around 7 mmol/L.

Interpretable machine learning for 28-day all-cause in-hospital mortality prediction in critically ill patients with heart failure combined with hypertension: A retrospective cohort study based on medical information mart for intensive care database-IV and eICU databases

Background

Heart failure (HF) combined with hypertension is an extremely important cause of in-hospital mortality, especially for the intensive care unit (ICU) patients. However, under intense working pressure, the medical staff are easily overwhelmed by the large number of clinical signals generated in the ICU, which may lead to treatment delay, sub-optimal care, or even wrong clinical decisions. Individual risk stratification is an essential strategy for managing ICU patients with HF combined with hypertension. Artificial intelligence, especially machine learning (ML), can develop superior models to predict the prognosis of these patients. This study aimed to develop a machine learning method to predict the 28-day mortality for ICU patients with HF combined with hypertension.

Methods

We enrolled all critically ill patients with HF combined with hypertension in the Medical Information Mart for IntensiveCare Database-IV (MIMIC-IV, v.1.4) and the eICU Collaborative Research Database (eICU-CRD) from 2008 to 2019. Subsequently, MIMIC-IV was divided into training cohort and testing cohort in an 8:2 ratio, and eICU-CRD was designated as the external validation cohort. The least absolute shrinkage and selection operator (LASSO) Cox regression with internal tenfold cross-validation was used for data dimension reduction and identifying the most valuable predictive features for 28-day mortality. Based on its accuracy and area under the curve (AUC), the best model in the validation cohort was selected. In addition, we utilized the Shapley Additive Explanations (SHAP) method to highlight the importance of model features, analyze the impact of individual features on model output, and visualize an individual’s Shapley values.

Results

A total of 3,458 and 6582 patients with HF combined with hypertension in MIMIC-IV and eICU-CRD were included. The patients, including 1,756 males, had a median (Q1, Q3) age of 75 (65, 84) years. After selection, 22 out of a total of 58 clinical parameters were extracted to develop the machine-learning models. Among four constructed models, the Neural Networks (NN) model performed the best predictive performance with an AUC of 0.764 and 0.674 in the test cohort and external validation cohort, respectively. In addition, a simplified model including seven variables was built based on NN, which also had good predictive performance (AUC: 0.741). Feature importance analysis showed that age, mechanical ventilation (MECHVENT), chloride, bun, anion gap, paraplegia, rdw (RDW), hyperlipidemia, peripheral capillary oxygen saturation (SpO₂), respiratory rate, cerebrovascular disease, heart rate, white blood cell (WBC), international normalized ratio (INR), mean corpuscular hemoglobin concentration (MCHC), glucose, AIDS, mean corpuscular volume (MCV), N-terminal pro-brain natriuretic peptide (Npro. BNP), calcium, renal replacement therapy (RRT), and partial thromboplastin time (PTT) were the top 22 features of the NN model with the greatest impact. Finally, after hyperparameter optimization, SHAP plots were employed to make the NN-based model interpretable with an analytical description of how the constructed model visualizes the prediction of death.

Conclusion

We developed a predictive model to predict the 28-day mortality for ICU patients with HF combined with hypertension, which proved superior to the traditional logistic regression analysis. The SHAP method enables machine learning models to be more interpretable, thereby helping clinicians to better understand the reasoning behind the outcome and assess in-hospital outcomes for critically ill patients.

Development and validation of a clinical predictive model for 1-year prognosis in coronary heart disease patients combine with acute heart failure

Background

The risk factors for acute heart failure (AHF) vary, reducing the accuracy and convenience of AHF prediction. The most common causes of AHF are coronary heart disease (CHD). A short-term clinical predictive model is needed to predict the outcome of AHF, which can help guide early therapeutic intervention. This study aimed to develop a clinical predictive model for 1-year prognosis in CHD patients combined with AHF.

Materials and methods

A retrospective analysis was performed on data of 692 patients CHD combined with AHF admitted between January 2020 and December 2020 at a single center. After systemic treatment, patients were discharged and followed up for 1-year for major adverse cardiovascular events (MACE). The clinical characteristics of all patients were collected. Patients were randomly divided into the training (n = 484) and validation cohort (n = 208). Step-wise regression using the Akaike information criterion was performed to select predictors associated with 1-year MACE prognosis. A clinical predictive model was constructed based on the selected predictors. The predictive performance and discriminative ability of the predictive model were determined using the area under the curve, calibration curve, and clinical usefulness.

Results

On step-wise regression analysis of the training cohort, predictors for MACE of CHD patients combined with AHF were diabetes, NYHA ≥ 3, HF history, Hcy, Lp-PLA2, and NT-proBNP, which were incorporated into the predictive model. The AUC of the predictive model was 0.847 [95% confidence interval (CI): 0.811–0.882] in the training cohort and 0.839 (95% CI: 0.780–0.893) in the validation cohort. The calibration curve indicated good agreement between prediction by nomogram and actual observation. Decision curve analysis showed that the nomogram was clinically useful.

Conclusion

The proposed clinical prediction model we have established is effective, which can accurately predict the occurrence of early MACE in CHD patients combined with AHF.

Acute heart failure in elderly patients: a review of invasive and non-invasive management

Acute heart failure (AHF) is a major cause of unplanned hospitalisations in the elderly and is associated with high mortality. Its prevalence has grown in the last years due to population aging and longer life expectancy of chronic heart failure patients. Although international societies have provided guidelines for the management of AHF in the general population, scientific evidence for geriatric patients is often lacking, as these are underrepresented in clinical trials. Elderly have a different risk profile with more comorbidities, disability, and frailty, leading to increased morbidity, longer recovery time, higher readmission rates, and higher mortality. Furthermore, therapeutic options are often limited, due to unfeasibility of invasive strategies, mechanical circulatory support and cardiac transplantation. Thus, the in-hospital management of AHF should be tailored to each patient's clinical situation, cardiopulmonary condition and geriatric assessment. Palliative care should be considered in some cases, in order to avoid unnecessary diagnostics and/or treatments. After discharge, a strict follow-up through outpatient clinic or telemedicine is can improve quality of life and reduce rehospitalisation rates. The aim of this review is to offer an insight on current literature and provide a clinically oriented, patient-tailored approach regarding assessment, treatment and follow-up of elderly patients admitted for AHF.

Prediction model of in-hospital mortality in intensive care unit patients with heart failure: machine learning-based, retrospective analysis of the MIMIC-III database

OBJECTIVE

The predictors of in-hospital mortality for intensive care units (ICUs)-admitted heart failure (HF) patients remain poorly characterised. We aimed to develop and validate a prediction model for all-cause in-hospital mortality among ICU-admitted HF patients.

DESIGN

A retrospective cohort study.

SETTING AND PARTICIPANTS

Data were extracted from the Medical Information Mart for Intensive Care (MIMIC-III) database. Data on 1177 heart failure patients were analysed.

METHODS

Patients meeting the inclusion criteria were identified from the MIMIC-III database and randomly divided into derivation (n=825, 70%) and a validation (n=352, 30%) group. Independent risk factors for in-hospital mortality were screened using the extreme gradient boosting (XGBoost) and the least absolute shrinkage and selection operator (LASSO) regression models in the derivation sample. Multivariate logistic regression analysis was used to build prediction models in derivation group, and then validated in validation cohort. Discrimination, calibration and clinical usefulness of the predicting model were assessed using the C-index, calibration plot and decision curve analysis. After pairwise comparison, the best performing model was chosen to build a nomogram according to the regression coefficients.

RESULTS

Among the 1177 admissions, in-hospital mortality was 13.52%. In both groups, the XGBoost, LASSO regression and Get With the Guidelines-Heart Failure (GWTG-HF) risk score models showed acceptable discrimination. The XGBoost and LASSO regression models also showed good calibration. In pairwise comparison, the prediction effectiveness was higher with the XGBoost and LASSO regression models than with the GWTG-HF risk score model (p<0.05). The XGBoost model was chosen as our final model for its more concise and wider net benefit threshold probability range and was presented as the nomogram.

CONCLUSIONS

Our nomogram enabled good prediction of in-hospital mortality in ICU-admitted HF patients, which may help clinical decision-making for such patients.

The Additional Prognostic Value of Ghrelin for Mortality and Readmission in Elderly Patients with Acute Heart Failure

Purpose

To evaluate the prognostic value of ghrelin, a growth hormone-releasing peptide, for mortality and readmission in elderly patients with acute heart failure (AHF).

Patients and Methods

We measured plasma ghrelin and pro B-type natriuretic peptide (NT-proBNP) levels upon emergency admission in 241 prospectively recruited elderly AHF patients (61.0% men). The outcomes were all-cause mortality and/or readmission due to heart failure (HF). Multivariate Cox proportional hazards regression analyses were used to evaluate the prognostic value of ghrelin. Discrimination, calibration, and reclassification indices were compared between models, with or without ghrelin.

Results

During 1.2 years of follow-up, we observed 90 events (57 deaths and 33 readmissions due to HF). Plasma ghrelin levels were significantly elevated in elderly AHF patients, when compared to healthy control subjects (P < 0.001). Patients with events had significantly higher baseline ghrelin levels, when compared to those without (P < 0.001). Ghrelin levels were positively correlated with NT-proBNP levels and HF severity, whereas they were negatively correlated with nutritional status (all P < 0.05). Log transformed ghrelin levels were independently associated with AHF events (hazard ratio = 2.64, 95% confidence interval = 1.11–6.25, P = 0.028). The incorporation of ghrelin into the reference model, or reference with the NT-proBNP model, both improved C-statistics (from 0.742–0.780 and 0.836–0.857; P = 0.074 and 0.044, respectively), resulting in an improvement in net reclassification index (14.42% and 10.45%, P = 0.020 and 0.025, respectively), and integrated discrimination index (5.64% and 3.60%, both P < 0.001). Patients who displayed the above NT-proBNP and ghrelin median levels had a markedly higher risk of AHF adverse events (P < 0.001).

Conclusion

Plasma ghrelin is an independent predictor of adverse events in elderly AHF patients. Ghrelin may provide additional value to clinical parameters or NT-proBNP for prognostic risk stratification in AHF.

Admission Hyperglycemia as a Predictor of Mortality in Acute Heart Failure: Comparison between the Diabetics and Non-Diabetics

Background: To investigate the impact of admission hyperglycemia (HGL) on in-hospital death (IHD) and 1-year mortality in acute heart failure (AHF) patients with or without diabetes mellitus (DM). Methods: Among 5625 AHF patients enrolled in a nationwide registry, 5541 patients were divided into four groups based on the presence of admission HGL and diabetes mellitus (DM). Admission HGL was defined as admission glucose level > 200 mg/dL. IHD and 1-year mortality were compared. Results: IHD developed in 269 patients (4.9%), and 1-year death developed in 1220 patients (22.2%). DM was a significant predictor of 1-year death (24.8% in DM vs. 20.5% in non-DM, p < 0.001), but not for IHD. Interestingly, admission HGL was a significant predictor of both IHD (7.6% vs. 4.2%, p < 0.001) and 1-year death (26.2% vs. 21.3%, p = 0.001). Admission HGL was a significant predictor of IHD in both DM and non-DM group, whereas admission HGL was a significant predictor of 1-year death only in non-DM (27.8% vs. 19.9%, p = 0.003), but not in DM group. In multivariate analysis, admission HGL was an independent predictor of 1-year mortality in non-DM patients (HR 1.32, 95% CI 1.03–1.69, p = 0.030). Conclusion: Admission HGL was a significant predictor of IHD and 1-year death in patients with AHF, whereas DM was only a predictor of 1-year death. Admission HGL was an independent predictor of 1-year mortality in non-DM patients with AHF, but not in DM patients. Careful monitoring and intensive medical therapy should be considered in AHF patients with admission HGL, regardless of DM.

Natriuretic peptide family as diagnostic/prognostic biomarker and treatment modality in management of adult and geriatric patients with heart failure: remaining issues and challenges

B-type natriuretic peptide (BNP) and N-terminal proBNP (NT-proBNP), the key members of natriuretic peptide family have been recommended as the gold standard biomarkers for the diagnosis and prognosis of heart failure (HF) according to the current clinical guidelines. However, recent studies have revealed many previously unrecognized features about the natriuretic peptide family, including more accurate utilization of BNP and NT-proBNP in diagnosing HF. The pathophysiological mechanisms behind natriuretic peptide release, breakdown, and clearance are very complex and the diverse nature of circulating natriuretic peptides and fragments makes analytical detection particularly challenging. In addition, a new class of drug therapy, which works via natriuretic peptide family, has also been considered promising for cardiology application. Under this context, our present mini-review aims at providing a critical analysis on these new progresses on BNP and NT-proBNP with a special emphasis on their use in geriatric cardiology settings. We have focused on several remaining issues and challenges regarding the clinical utilization of BNP and NT-proBNP, which include: (1) Different prevalence and diagnostic/prognostic values of BNP isoforms; (2) methodological issues on detection of BNP; (3) glycosylation of proBNP and its effect on biomarker testing; (4) specificity and comparability of BNP/NT-proBNP resulted from different testing platforms; (5) new development of natriuretic peptides as HF treatment modality; (6) BNP paradox in HF; and (7) special considerations of using BNP/NT-proBNP in elderly HF patients. These practical discussions on BNP/NT-proBNP may be instrumental for the healthcare providers in critically interpreting laboratory results and effective management of the HF patients.