Predicted Risk Of Mortality Research Articles

Early inflammation as a footprint of increased mortality risk in infants living with HIV from three African countries

In this work our aim was to identify early biomarkers in plasma samples associated with mortality in children with perinatal HIV treated early in life, to potentially inform early intervention targeting this vulnerable group. 20/215 children (9.3%) with perinatal HIV, enrolled within 3 months of age died prematurely within the first year of the study, despite early ART initiation. Using a propensity score, we selected 40 alive study participants having similar clinical and virological records compared to the deceased group. 13 HIV unexposed (HU) healthy children were additionally used as controls. Baseline plasma samples were analyzed using a targeted proteomic approach, and to assess pathogen-associated and damage-associated molecular patterns (PAMPs, DAMPs) levels. Data from deceased participants were compared to both control groups, with multivariate logistic regression models used to evaluate the association between mortality and plasma proteins. We developed a machine learning model to predict mortality risk, finding that IL-6 and CXCL11 not only were higher in deceased children than Matched-children with HIV (p < 0.001 and p = 0.0034) but also predictive of mortality (accuracy of 77%); levels of PAMPs were higher in deceased children (p = 0.0016). Thus, measuring early inflammatory biomarkers, particularly IL-6, could help mortality risk prediction and potentially guide targeted interventions.

Associations between diabesity and mortality over 7 years: a Maltese population-level case study

Abstract Introduction Diabesity presents a growing global epidemic. Demographic and socioeconomic associations have been documented, yet, the literature on mortality risk associated with diabesity is scarce. The aim of this study was to assess whether diabetes and/or diabesity increas the risk of all-cause mortality across a period of 7 years at a population level. Methods A nation-wide health examination survey in Malta was conducted between 2014 and 2016. This collected sociodemographic information, bio-anthropometric measurements (waist circumference, body mass index [BMI], body adiposity index [BAI]) and blood samples. Mortality information was acquired in 2021 by cross-referencing with the national mortality register. Statistical analysis was conducted using chi-square tests and Cox proportional hazard models. Results A mortality rate of 1.67% was observed (n = 3947), with diabetes and obesity being responsible for 6.06% and 3.03% of deaths, respectively. Examining the deceased individuals alongside their baseline characteristics (2014-2015) revealed that 25.76% (CI95%: 16.75 - 37.44) had diabetes, and 30.30% (CI95%: 20.55 - 42.22) were obese. Diabetes was associated with significant mortality risk (HR: 2.65 CI95%: 1.20 - 5.85; p = 0.02), while diabesity showed borderline significance (p = 0.06). After adjustment for potential confounding factors the mortality risk linked to diabetes remained significant and BAI was the only anthropometric measurement positively associated with mortality. Conclusions These findings emphasise the urgency for comprehensive public health policies targeting the concurrent prevention and management of diabetes and obesity at a population level. Our results also indicate that incorporating body adiposity index (BAI) into risk assessment protocols may provide valuable insights into mortality risk prediction and inform tailored interventions. Key messages • Diabetes and a high BAI score were significantly associated with mortality in the Maltese population, while diabesity had moderate significance, emphasising the urgency of tackling this epidemic. • These findings call for broad health policies to address diabesity at a population level and suggest integrating BAI into risk assessment for better mortality prediction and tailored interventions.

Diastolic shock index: a novel prognostic parameter unveiling insights into vasodilatory cardiogenic shock

Abstract Introduction Systemic vasodilation is an increasingly acknowledged feature in cardiogenic shock (CS), defining a subset of patients with low systemic vascular resistances and higher vasopressor requirements. While a diastolic shock index (DSI) above 2.0 predicts vasoplegia and mortality in septic shock, its cut-off and prognostic significance in CS remain unclear. Our study aimed to characterize vasoplegia in CS using the DSI and assess its impact on prognosis. Methods Single-center retrospective cohort study on Cardiac Intensive Care Unit (CICU) admissions for CS from January 2017 to October 2023. Data included patient variables, CS features, vasoactive drugs, mechanical circulatory support, and 30-day mortality. DSI (heart rate (HR) to invasive diastolic blood pressure (DBP) ratio) impact on 30-day all-cause mortality was assessed with univariate Cox regression. Optimal DSI cut-off for adverse outcomes was determined through ROC curve analysis, and survival based on this cut-off was analyzed using Kaplan-Meier curves. Results 185 patients (mean age 59 ± 12 years, 63% male) were included. Acute myocardial infarction-related CS (47%) and heart failure-related CS (29%) were predominant. According to SCAI classification, 86% were class C, 22% class D, and 4% class E, with 26% experiencing previous cardiac arrest. Median Vasoactive-Inotropic Score (VIS) was 48 (IQR 21-85), and 25% used mechanical circulatory support. At admission, median HR was 105 bpm (IQR 90-125), DBP was 60 mmHg (IQR 55-65), and DSI was 1.71 (IQR 1.31-2.18). Thirty-day all-cause mortality was 36% (n=67), with median time to death of 4 days (IQR 1-10). DSI at 30 days predicted mortality (OR 1.42, 95%CI 1.06-1.91, p=0.020). DSI showed fair discriminative power for 30-day mortality (C-Statistic 0.60, 95% CI 0.51-0.68, p=0.031) with a cut-off of 2.2. Patients with DSI &amp;gt;2.2 exhibited heightened likelihood of prior cardiac arrest (38% vs. 20%, p=0.032), and higher serum lactate (5.5 vs. 3.9 mmol/L, p=0.008), and serum creatinine (2.64 vs. 1.88 mg/dL, p=0.003) at admission. Despite higher VIS at 48 hours (120±136 vs. 65±93.9, p=0.04), MCS need did not significantly differ. This group had superior 30-day mortality risk (log-rank p=0.011). Conclusion In our cohort, the admission Diastolic Shock Index (DSI) emerged as a promising identifier for CS vasoplegic patients and was a predictor of a higher risk for mortality. A DSI above 2.2 identified a subset of patients with previous cardiac arrest and higher lactate levels, enlightening possible risk factors associated with vasoplegia. Further studies are warranted to establish the robustness of this novel prognostic parameter in CS patients and its therapeutic implications.

Interpretable machine learning model for predicting long-term cardiovascular mortality risk in chronic kidney disease patients using SHAP explainers

Abstract Background Patients with chronic kidney disease (CKD) exhibited a pronounced burden of cardiac mortality. However, early prediction of cardiovascular mortality risk in CKD patients remains an unmet medical need. Purpose This study aimed to evaluate the performance of machine learning (ML) algorithms and identify the most influential factors in predicting long-term cardiac mortality among CKD patients. Methods Six ML models were developed, with the most effective one chosen for predicting and categorising patients into high-risk and low-risk groups based on the maximal Youden’s index. Differences in survival rates were assessed using the log-rank test on Kaplan-Meier curves. Cox regression analysis was employed to investigate the relationship between ML-predicted risk scores and mortality. Furthermore, the SHapley Additive exPlanations (SHAP) method was implemented to provide personalised explanations for model decisions. Results The auto-encoder (AE) model achieved the highest AUC of 0.94, with a sensitivity of 80.95% and specificity of 96.34%. Compared to individuals in low-risk group, those in high-risk group exhibited a significantly elevated risk of cardiovascular mortality (adjusted hazard ratio [HR]: 48.24; P &amp;lt;0.001). According to the SHAP model, age, C-reactive protein, blood urea nitrogen and hypertension were identified as the four most influential factors in the AE model. Conclusions Our study successfully developed an AE machine learning model for predicting 10-year cardiovascular mortality among CKD patients. The combination of ML model and SHAP aids doctors in understanding key features in the model, improving their insight into decision-making related to disease severity assessment, thus facilitating early preventive strategies in clinical practice.

Development and validation of a prognostic score integrating remote heart failure symptoms and clinical variables for mortality risk prediction after myocardial infarction. The PragueMi score

Abstract Background While heart failure (HF) symptoms are associated with adverse prognosis after myocardial infarction (MI), they are not routinely used for patients’ stratification. Purpose The primary objective of this study was to develop and validate a score to predict mortality risk after MI, combining remotely recorded HF symptoms and clinical risk factors, and to compare it against the guideline-recommended GRACE score. Methods A cohort study design using prospectively collected data from consecutive patients hospitalized for MI at a large tertiary heart centre between June 2017 and September 2022 was used. Results Data from 1,135 patients (aged 64±12 years, 26.7% women), were split into derivation (70%) and validation cohort (30%). Components of the 23-item Kansas City Cardiomyopathy Questionnaire (KCCQ) questionnaire and clinical variables were used as possible predictors. The best model included the following variables – age, heart failure history, admission creatinine and heart rate, ejection fraction at hospital discharge, and HF symptoms 1 month after discharge including walking impairment, leg swelling, and change in HF symptoms. Based on these variables, the PragueMi score was developed. In the validation cohort, the PragueMi score showed superior discrimination to the GRACE score for 6 months (AUC 90.1, 95% CI 81.8-98.4 vs. 77.4, 95% CI 62.2-92.5, p=0.04) and 1-year risk prediction (AUC 89.7, 95% CI 83.5-96.0 vs. 76.2, 95% CI 64.7-87.7, p=0.004). Conclusion The PragueMi score combining heart failure symptoms and clinical variables performs better than the currently recommended GRACE score.

Development and validation of a predicting model for risk of mortality in heart failure patients with mildly reduced ejection fraction post discharge

Abstract Background Prognostic prediction for heart failure with mildly reduced ejection fraction (HFmrEF) patients remains challenging. We aimed to create and validate a mortality risk prediction model for HFmrEF patients at 6-month, 1-year, and 3-year post discharge. Methods Clinical data of 1691 HFmrEF patients registered in the heart failure registry from 2015 to 2020 at the Heart Center of our hospital were analyzed. Patients were assigned into a training (1183 patients) and validation cohort (508 patients) at 7:3 ratio. Predictive variables for mortality at various period post discharge were identified by the Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis. The predicting model was then established based on these variables. Model performance was evaluated with ROC curve and Decision Curve Analysis (DCA). Results Eight predictors were identified including age, NT-proBNP, hemoglobin levels, use of beta-blockers, ACEI/ARB, invasive ventilation, PCI procedures, and pulmonary artery systolic pressure values. The model achieved a C-index of 0.748 (training set) and 0.755 (validation set). Area Under the Curve (AUC) for training set at 6 months, 1 year, and 3 years were 0.813, 0.784, and 0.774, and AUC for validation set were 0.775, 0.744, and 0.770, respectively. The DCA analysis confirmed the favorable net benefit of the established nomogram model. Conclusion This predicting model might be used for post-discharge risk stratification and individual decision-making of post-discharge management in HFmrEF patients.

Machine learning for prediction of all-cause mortality in acute coronary syndrome

Abstract Background Acute coronary syndrome (ACS) remains one of the leading causes of death globally. Accurate and reliable mortality risk prediction of ACS patients is essential for developing targeted treatment strategies and improve prognostication. Traditional models for risk stratification such as the Global Registry of Acute Coronary Events (GRACE) and the Thrombolysis In Myocardial Infarction (TIMI) risk scores offer moderate discriminative value, and do not incorporate contemporary predictors of ACS prognosis. Machine learning (ML) models have emerged as an alternate method that may offer improved risk assessment. This study aims to compare machine learning models with traditional risk scores for predicting all-cause mortality in patients with ACS. Methods PubMed, EMBASE, Web of Science and Cochrane databases were searched until 1st November 2023 for studies comparing ML models with traditional statistical methods for event prediction of ACS patients. The primary outcome was comparative discrimination measured by C-statistics with 95% confidence intervals between ML models and traditional methods in estimating the risk of all-cause mortality. Results Ten studies were included (239627 patients). The summary C-statistic of all ML models across all endpoints was 0.89 (95% CI, 0.86-0.92), compared to traditional methods 0.82 (95% CI, 0.79-0.85). The difference in C-statistic between all ML models and traditional methods was 0.07 (p&amp;lt;0.05). Three models undertook external validation, and calibration was inconsistently reported. Conclusion ML models demonstrated superior discrimination of all-cause mortality for ACS patients compared to traditional risk scores. Despite outperforming well-established prognostic tools such as the GRACE and TIMI scores, current clinical applications of ML approaches remain uncertain, particularly in view of the methodological limitations of existing models and the need for greater validation.

The value of Age-adjusted Charlson and Elixhauser-Van Walraven comorbidity index in predicting prognosis for patients undergoing heart valve surgery

ObjectiveThis study aims to compare the efficacy of the Age-adjusted Charlson Comorbidity Index (ACCI) and the Elixhauser-Van Walraven Comorbidity Index (ECI-VW) in predicting mortality risk among patients undergoing heart valve surgery.MethodsClinical data were extracted from the INSPIRE Database using R language. The Receiver Operating Characteristic (ROC) Curve was employed to assess the predictive accuracy of ACCI and ECI-VW for in-hospital all-cause mortality and post-surgical all-cause mortality at 7 and 28 days. Subgroup analysis was conducted to validate the application efficacy, and the optimal cutoff value was identified.ResultsThe study included 996 patients, with 931 survivors and 65 cases of in-hospital all-cause mortality. The area under the curve (AUC) for ACCI in predicting in-hospital all-cause mortality was 0.658 (95% CI: 0.584, 0.732), while the AUC for ECI-vw in predicting the same outcome was 0.663 (95% CI: 0.584, 0.741). For predicting all-cause mortality within 7 days post-surgery, the AUC of ACCI was 0.680 (95% CI: 0.04, 0.56), and for ECI-vw, it was 0.532 (95% CI: 0.353, 0.712). Regarding the prediction of all-cause mortality within 28 days after surgery, the AUC for ACCI was 0.724 (95% CI: 0.622, 0.827), and for ECI-vw, it was 0.653 (95% CI: 0.538, 0.69). Patients were categorized into two groups based on the ACCI cutoff value of 3.5, including Group 1 (ACCI < 3.5 points, 823 cases) and Group 2 (ACCI > 3.5 points, 173 cases). The overall survival rate for these two patient groups was calculated using the Kaplan-Meier method, revealing that the 28-day postoperative survival rate for patients in Group 1 was significantly higher than that for patients in Group 2 (P < 0.0001).ConclusionsACCI demonstrates significant predictive value for in-hospital all-cause mortality within 28 days following cardiac valve disease surgery. Patients presenting with an ACCI greater than 3.5 exhibit an increased risk of mortality within 28 days post-surgery compared to those with an ACCI less than 3.5. This finding suggests that the ACCI can serve as a preliminary tool for assessing the prognosis of patients undergoing this type of surgical intervention.

Prediction of in-hospital mortality risk for patients with acute ST-elevation myocardial infarction after primary PCI based on predictors selected by GRACE score and two feature selection methods

IntroductionAccurate in-hospital mortality prediction following percutaneous coronary intervention (PCI) is crucial for clinical decision-making. Machine Learning (ML) and Data Mining methods have shown promise in improving medical prognosis accuracy.MethodsWe analyzed a dataset of 4,677 patients from the Regional Vascular Center of Primorsky Regional Clinical Hospital No. 1 in Vladivostok, collected between 2015 and 2021. We utilized Extreme Gradient Boosting, Histogram Gradient Boosting, Light Gradient Boosting, and Stochastic Gradient Boosting for mortality risk prediction after primary PCI in patients with acute ST-elevation myocardial infarction. Model selection was performed using Monte Carlo Cross-validation. Feature selection was enhanced through Recursive Feature Elimination (RFE) and Shapley Additive Explanations (SHAP). We further developed hybrid models using Augmented Grey Wolf Optimizer (AGWO), Bald Eagle Search Optimization (BES), Golden Jackal Optimizer (GJO), and Puma Optimizer (PO), integrating features selected by these methods with the traditional GRACE score.ResultsThe hybrid models demonstrated superior prediction accuracy. In scenario (1), utilizing GRACE scale features, the Light Gradient Boosting Machine (LGBM) and Extreme Gradient Boosting (XGB) models optimized with BES achieved Recall values of 0.944 and 0.954, respectively. In scenarios (2) and (3), employing SHAP and RFE-selected features, the LGB models attained Recall values of 0.963 and 0.977, while the XGB models achieved 0.978 and 0.99.DiscussionThe study indicates that ML models, particularly the XGB optimized with BES, can outperform the conventional GRACE score in predicting in-hospital mortality. The hybrid models' enhanced accuracy presents a significant step forward in risk assessment for patients post-PCI, offering a potential alternative to existing clinical tools. These findings underscore the potential of ML in optimizing patient care and outcomes in cardiovascular medicine.

Trends in Transcatheter Aortic Valve Replacement Outcomes

Although transcatheter aortic valve replacement (TAVR) outcomes in the US have improved substantially since 2011, it is unknown whether these trends have continued since 2019. To examine changes in risk-adjusted TAVR outcomes from 2019 to 2022 and to examine any noteworthy trends over time. This cohort study examined data from patients with severe aortic stenosis treated with TAVR at 786 US hospitals between January 1, 2019, and March 31, 2022, included in the Society of Thoracic Surgeons (STS)/American College of Cardiology (ACC) Transcatheter Valve Therapies (TVT) Registry. Patients who underwent TAVR. The primary outcome was 30-day mortality, and the secondary outcomes were in-hospital mortality and 30-day composite adverse events. To understand factors explaining these trends, a series of logistic regression models was constructed for each outcome, with time as the primary explanatory variable. After adjusting for changing patent characteristics and procedural factors, a series of exploratory analyses was performed to examine the extent to which these findings could be explained by several plausible hypotheses. This study's analytic cohort included a total of 210 495 patients. Median (IQR) patient age was 79 (73-85) years, and 91 313 patients (43.4%) were female. Median (IQR) STS predicted risk of mortality (PROM) was 3.3% (2.0%-5.3%). There were no significant changes in unadjusted 30-day mortality from quarter 1 of 2019 (2.4%) to the end of quarter 1 of 2022 (2.2%) (P for trend = .10), with an unadjusted odds ratio (OR) for time of 0.98 per year (95% CI, 0.94-1.01). After adjusting for patient characteristics, the OR increased to 1.05 per year (95% CI, 1.02-1.08), which increased further after adjusting for procedural characteristics to 1.09 per year (95% CI, 1.05-1.13). In exploratory analyses, there were no meaningful changes in the adjusted odds of death after excluding sites that entered the STS/ACC TVT Registry in 2019 or later (OR, 1.09; 95% CI, 1.05-1.13), low-volume sites (OR, 1.09; 95% CI, 1.06-1.13), low-risk patients (OR, 1.11; 95% CI, 1.07-1.15), patients with a bicuspid aortic valve (OR, 1.09; 95% CI, 1.05-1.13), in-hospital deaths (OR, 1.08; 95% CI, 1.03-1.14), or patients who experienced a major vascular complication (OR, 1.09; 95% CI, 1.05-1.12). In this observational cohort study performing a national analysis of outcomes after TAVR, it was found that risk-adjusted 30-day mortality increased modestly from January 2019 to March 2022. However, no site-level, patient-related, or process-related factors were identified that could explain these findings. Although the absolute increase in risk-adjusted mortality during the study period was relatively small, these findings warrant continued surveillance.

Prediction of 90-day mortality risk after unplanned emergency department visits of advanced stage cancer patients

PurposeCancer represents the leading cause of mortality in high-income countries. In the last years, the rate of emergency department (ED) visits by cancer patients has increased 5.5-fold. These ED visits impose a significant economic burden and may indicate the progression of the oncologic disease. The goal of this retrospective study was to identify patient-derived risk factors, especially focusing on serum albumin and body mass index (BMI) for 90-day mortality following unplanned ED visits by cancer patients.MethodsA retrospective chart review of all patients with an ICD-10 diagnosis for cancer undergoing palliative treatment presenting at the ED between 2016 and 2018 at the General Hospital of Vienna was performed. Laboratory values, emergency severity index (ESI), and BMI were collected at the ED presentation. 90-day mortality (90MM) was calculated from the ED presentation.ResultsA total of 448 cancer patients were included. Lung cancer (19.2%) and pancreaticobiliary cancer (15.6%) were the most frequent diagnoses. The main reasons for ED visits were pain (20.5%) and fever (17.4%). Sixty-nine percent of patients had to be admitted and 17.5% of patients died during hospitalization. 90MM was highest for patients with low albumin (< 35 g/L vs. > 35 g/L: 60.4% vs. 31.4%; p < .0001). When incorporating albumin levels and BMI, patients with both values below the cutoff had the highest risk for death (HR 4.01, 95% CI 2.30–7.02).ConclusionCancer patients face a high risk for hospitalization when presenting at the ED. The 90MM rate is highest in patients with low BMI and albumin levels. This highlights an especially vulnerable cohort of cancer patients for whom supportive care and palliative care have to be optimized.

Association between Birth Weight of the Babies and Mode of Delivery: A Cross-Sectional Study

Background: Birth weight has significant short and long-term effects on health and survival and is a useful predictor of prenatal and postnatal development. Birth weight is likely to play a key role for the infant’s survival because it is one of the strongest predictors of infant mortality risk. The aim of the study was to find out the association between birth weight and mode of delivery. Methods: This was a cross-sectional study conducted at Combined Military Hospital, Ghatail from 15 March 2019 to 30 June 2019. Secondary data were used, which were collected from hospital register during the period of 01 January 2018 to 31 December 2018. The study population was all the newborn babies delivered in CMH, Ghatail during that period and mothers of the newborn babies were respondents of the study. Results: Out of 526 respondents majority (64.83%) were in the age group of 23-32 years with mean age 25.14(range: 18-40) years. Of the newborn babies around half (48.3%) were male. The mean birth weight was 2.92 kg with minimum weight 1.2 kg and maximum 4.2 kg. The data showed that the prevalence of Low Birth Weight (LBW) 13.1%. Majority (84.6%) of the mothers didn’t have any comorbidity, few (15.4%) suffered from comorbidity. 53.8% babies were born by caesarean section (CS) and 46.2% were delivered by normal vaginal delivery. Among the low birth weight babies 65.2% were born by CS. Statistically significant association was found between birth weight of the babies and mode of delivery (p&lt;0.05). Conclusion: Caesarean section cannot be routinely recommended, unless there are other obstetric indications. Bangladesh Armed Forces Med J Vol 57 No (1) June 2024, pp 8-13

Brighter nights and darker days predict higher mortality risk: A prospective analysis of personal light exposure in >88,000 individuals

Light enhances or disrupts circadian rhythms, depending on the timing of exposure. Circadian disruption contributes to poor health outcomes that increase mortality risk. Whether personal light exposure predicts mortality risk has not been established. We therefore investigated whether personal day and night light, and light patterns that disrupt circadian rhythms, predicted mortality risk. UK Biobank participants (N = 88,905, 62.4 ± 7.8 y, 57% female) wore light sensors for 1 wk. Day and night light exposures were defined by factor analysis of 24-h light profiles. A computational model of the human circadian pacemaker was applied to model circadian amplitude and phase from light data. Cause-specific mortality was recorded in 3,750 participants across a mean (±SD) follow-up period of 8.0 ± 1.0 y. Individuals with brighter day light had incrementally lower all-cause mortality risk (adjusted-HR ranges: 0.84 to 0.90 [50 to 70th light exposure percentiles], 0.74 to 0.84 [70 to 90th], and 0.66 to 0.83 [90 to 100th]), and those with brighter night light had incrementally higher all-cause mortality risk (aHR ranges: 1.15 to 1.18 [70 to 90th], and 1.21 to 1.34 [90 to 100th]), compared to individuals in darker environments (0 to 50th percentiles). Individuals with lower circadian amplitude (aHR range: 0.90 to 0.96 per SD), earlier circadian phase (aHR range: 1.16 to 1.30), or later circadian phase (aHR range: 1.13 to 1.20) had higher all-cause mortality risks. Day light, night light, and circadian amplitude predicted cardiometabolic mortality, with larger hazard ratios than for mortality by other causes. Findings were robust to adjustment for age, sex, ethnicity, photoperiod, and sociodemographic and lifestyle factors. Minimizing night light, maximizing day light, and keeping regular light-dark patterns that enhance circadian rhythms may promote cardiometabolic health and longevity.

Association of weight-adjusted-waist index with all-cause and cardiovascular mortality in individuals with diabetes or prediabetes: a cohort study from NHANES 2005–2018

Weight-adjusted waist index (WWI) is a new marker of central obesity. This study explored the association of WWI with all-cause and cardiovascular disease (CVD) mortality in individuals with diabetes or prediabetes. 6551 participants with diabetes or prediabetes from the National Health and Nutrition Examination Survey (NHANES) records between 2005 and 2018 were included. The association of WWI with all-cause and CVD mortality was assessed using Kaplan-Meier survival analysis, Cox proportional hazards model (Cox regression), and restricted cubic spline (RCS). The predictive value of WWI for mortality was analyzed using time-dependent receiver operating characteristic curves (ROC). There were 1083 all-cause deaths and 360 CVD deaths. Multivariable-adjusted Cox regression analyses showed WWI was positively correlated with the risk of all-cause and CVD mortality in subjects with diabetes or prediabetes. Multivariate-adjusted RCS analyses showed a linear and positive correlation of WWI with all-cause mortality risk, and a nonlinear relationship with CVD mortality, with a threshold of 12.35. The area under the curve (AUC) for 3, 5, and 10-years survival for all-cause mortality was 0.795, 0.792, and 0.812, respectively, and for CVD mortality was 0.815, 0.833, and 0.831, respectively. WWI is a valuable predictor of all-cause mortality risk in patients with diabetes and prediabetes, and a valuable predictor of CVD mortality risk when patients with diabetes and prediabetes are considered as a whole.

Machine learning-based prediction of mortality in acute myocardial infarction with cardiogenic shock.

In the ICU, patients with acute myocardial infarction and cardiogenic shock (AMI-CS) often face high mortality rates, making timely and precise mortality risk prediction crucial for clinical decision-making. Despite existing models, machine learning algorithms hold the potential for improved predictive accuracy. In this study, a predictive model was developed using the MIMIC-IV database, with external validation performed on the eICU-CRD database. We included ICU patients diagnosed with AMI-CS. Feature selection was conducted using the Boruta algorithm, followed by the construction and comparison of four machine learning models: Logistic Regression (LR), eXtreme Gradient Boosting (XGBoost), Adaptive Boosting (AdaBoost), and Gaussian Naive Bayes (GNB). Model performance was evaluated based on metrics such as AUC (Area Under the Curve), accuracy, sensitivity, specificity, and so on. The SHAP method was employed to visualize and interpret the importance of model features. Finally, we constructed an online prediction model and conducted external validation in the eICU-CRD database. In this study, a total of 570 and 391 patients with AMI-CS were included from the MIMIC-IV and eICU-CRD databases, respectively. Among all machine learning algorithms evaluated, LR exhibited the best performance with a validation set AUC of 0.841(XGBoost: 0.835, AdaBoost: 0.839, GNB: 0.826). The model incorporated five variables: prothrombin time, blood urea nitrogen, age, beta-blockers and Angiotensin-Converting Enzyme Inhibitors or Angiotensin II Receptor Blockers. SHAP plots are employed to visualize the importance of model features and to interpret the results. An online prediction tool was developed, externally validated with the eICU-CRD database, achieving an AUC of 0.755. Employing the LR algorithm, we developed a predictive model for assessing the mortality risk among AMI-CS patients in the ICU setting. Through model predictions, this facilitates early detection of high-risk individuals, ensures judicious allocation of healthcare resources.

Utility of ISARIC 4C Mortality Score, Vaccination History, and Anti-S Antibody Titre in Predicting Risk of Severe COVID-19.

The ISARIC 4C Mortality score was developed to predict mortality risk among patients with COVID-19. Its performance among vaccinated individuals is understudied. This is a retrospective study of all patients with SARS-CoV-2 infection admitted to the National Centre for Infectious Diseases, Singapore, from January-2020 to December-2021. Demographic, clinical, and laboratory data were extracted, and multiple logistic regression (MLR) models were developed to predict the relationship between ISARIC score, vaccination status, anti-S antibody titre, and severe COVID-19. A total of 6377 patients were identified, of which 5329 met the study eligibility criteria. The median age of the patients was 47 years (IQR 35-71), 1264 (23.7%) were female, and 1239 (25.7%) were vaccinated. Severe disease occurred in 499 (9.4%) patients, including 133 (2.5%) deaths. After stratification, 3.0% of patients with low (0-4), 17.8% of patients with moderate (5-9), and 36.2% of patients with high (≥10) ISARIC scores developed severe COVID-19. Vaccination was associated with a reduced risk of progression to severe COVID-19 in the MLR model: aOR 0.88 (95% CI: 0.86-0.90), and the risk of severe COVID-19 decreased inversely to anti-S antibody titres. The anti-S antibody titre should be further investigated as an adjunct to the ISARIC score to triage COVID-19 patients for hospital admission and antiviral therapy.

Longitudinal frailty assessment in the prediction of survival among patients with advanced chronic kidney disease: a prospective observational single-centre cohort study

ObjectivesThis study aims to describe the prevalence, characteristics and longitudinal changes in frailty among outpatient chronic kidney disease (CKD) and haemodialysis (HD) populations and their impact on survival.DesignProspective observational cohort...

DynProfiler: a Python package for comprehensive analysis and interpretation of signaling dynamics leveraged by deep learning techniques.

Signaling dynamics encode important features and regulatory mechanisms of biological systems, and recent studies have reported the use of simulated signaling dynamics with mechanistic modeling as biomarkers for human diseases. Given the success of deep learning techniques, it is expected that they can extract informative patterns from simulation results more effectively than traditional approaches involving manual feature selection, which can be used for subsequent analyses, such as patient stratification and survival prediction. Here, we propose DynProfiler, which utilizes the entire signaling dynamics, including intermediate variables, as input and leverages deep learning techniques to extract informative features without requiring any labels. Furthermore, DynProfiler incorporates a modern explainable AI solution to provide quantitative time-dependent importance scores for each dynamics. Using simulated dynamics of patients with breast cancer as an example, we demonstrate DynProfiler's ability to extract high-quality features that can predict mortality risk and identify important dynamics, highlighting upregulated phosphorylated GSK3β as a biomarker for poor prognosis. Overall, this tool can be useful for clinical application, as well as for elucidating biological system dynamics. The DynProfiler Python library is available in GitHub at https://github.com/okadalabipr/DynProfiler.

Social Determinants of Health and Risk-Adjusted Sepsis Mortality in the Nationwide Veterans Affairs Healthcare System.

Traditional risk prediction and risk adjustment models have focused on clinical characteristics, but accounting for social determinants of health (SDOH) and complex health conditions could improve understanding of sepsis outcomes and our ability to predict outcomes, treat patients, and assess quality of care. To evaluate the impact of SDOH and health scales in sepsis mortality risk prediction and hospital performance assessment. Observational cohort study. One hundred twenty-nine hospitals in the nationwide Veterans Affairs (VA) Healthcare System between 2017 and 2021. Veterans admitted through emergency departments with community-acquired sepsis. Individual- and community-level SDOH (race, housing instability, marital status, Area Deprivation Index [ADI], and rural residence) and two health scales (the Care Assessment Need [CAN] score and Claims-Based Frailty Index [CFI]). The primary outcome was 90-day mortality from emergency department arrival; secondary outcomes included 30-day mortality and in-hospital mortality. Among 144,889 patients admitted to the hospital with community-acquired sepsis, 139,080 were men (96.0%), median (IQR) age was 71 (64-77) years, and median (IQR) ADI was 60 (38-81). Multivariable regression models had good calibration and discrimination across models that adjusted for different sets of variables (e.g., AUROC, 0.782; Brier score, 1.33; and standardized mortality rate, 1.00). Risk-adjusted hospital performance was similar across all models. Among 129 VA hospitals, three hospitals shifted from the lowest or highest quintile of performance when comparing models that excluded SDOH to models that adjusted for all variables. Models that adjusted for ADI reported odds ratios (CI) of 1.00 (1.00-1.00), indicating that ADI does not significantly predict sepsis mortality in this cohort of patients. In patients with community-acquired sepsis, adjusting for community SDOH variables such as ADI did not improve 90-day sepsis mortality predictions in mortality models and did not substantively alter hospital performance within the VA Healthcare System. Understanding the role of SDOH in risk prediction and risk adjustment models is vital because it could prevent hospitals from being negatively evaluated for treating less advantaged patients. However, we found that in VA hospitals, the potential impact of SDOH on 90-day sepsis mortality was minimal.

Performance of SARC-F and SARC-CalF scores to predict risk of mortality in hemodialysis patients: a cohort study

This study aimed to evaluate the association between the questionnaires SARC-F and SARC-CalF with risk of mortality in patients undergoing hemodialysis (HD). A cohort study, with patients on HD age ≥ 18 years, both sex, between June 2019 and April 2023. Body composition (anthropometry and bioelectrical impedance), muscle functional (handgrip strength and gait speed), screening of sarcopenia using the SARC-F and SARC-CalF, nutritional status and laboratory data were assessed. Follow-up for mortality up to 47 months. The sample consisted of 243 participants and the prevalence of risk of sarcopenia using SARC-F and SARC-CalF were 30% and 45%, respectively; 65 died for all reasons and three patients were censored due to transplantation. Multivariate analysis identified SARC-CalF as predictor of mortality in HD patients [hazard ratio 1.96; 95% CI (1.01–3.79); p = 0.04]. The survival analysis showed that there was a significant difference in the survival curves among the groups stratified by SARC-F and SARC-CalF for log-rank test. A higher specificity was found for SARC-CalF than SARC-F (80% vs. 77%) in receiver operating characteristic (ROC) curve. Both questionnaires were associated with anthropometric, parameters of body composition, physical measurements, and SARC-CalF was predictor of risk for mortality in HD patients.