Identify Sepsis Patients Research Articles

Transfer learning-enabled outcome prediction for guiding CRRT treatment of the pediatric patients with sepsis

Continuous renal replacement therapy (CRRT) is a life-saving procedure for sepsis but the benefit of CRRT varies and prediction of clinical outcomes is valuable in efficient treatment planning. This study aimed to use machine learning (ML) models trained using MIMIC III data for identifying sepsis patients who would benefit from CRRT. We first selected patients with sepsis and CRRT in the ICU setting and their gender, and an array of routine lab results were included as features to train machine learning models using 30-day mortality as the primary outcome. A total of 4161 patients were included for analysis, among whom there were 1342 deaths within 30 days. Without data augmentation, extreme gradient boosting (XGBoost) showed an accuracy of 64.2% with AUC-ROC of 0.61. Data augmentation using a conditional generative adversarial neural network (c-GAN) resulted in a significantly improved accuracy (82%) and ROC-AUC (0.78%). To enable prediction on pediatric patients, we adopted transfer learning approaches, where the weights of all but the last hidden layer were fixed, followed by fine-tuning of the weights of the last hidden layer using pediatric data of 200 patients as the inputs. A significant improvement was observed using the transfer learning approach (AUCROC = 0.76) compared to direct training on the pediatric cohort (AUCROC = 0.62). Through this transfer-learning-facilitated patient outcome prediction, our study showed that ML can aid in clinical decision-making by predicting patient responses to CRRT for managing pediatric sepsis.

Improving Early Sepsis & Severe Sepsis Identification and Management on Internal Medicine service at the Detroit John D. Dingell VA Medical Center

Background: Sepsis is a life-threatening organ dysfunction due to infection, with an incidence of nearly 50 million people each year, causing about 11 million in mortality worldwide. Adhering to Sepsis protocols is a quality metric monitored by both the Centers for Medicare & Medicaid Services (CMS) via the SEP 1 measure and the VA Healthcare System through SEP 90, which consists of an all-or-nothing score comprised of individual metrics. These metrics include specific actions such as early management bundle utilization, lactate monitoring, timely fluid resuscitation, and vasopressor administration, all within defined timeframes. Based on FY 2023 Data, the John D. Dingell VAMC performed nationally in the last quintile with > 72% fallouts. Given that the vast majority of our Sepsis patients are treated and managed by the inpatient Internal Medicine resident service, we initiated a Quality Improvement project to educate them and thereby decrease our number of fallouts. We aim to improve Sepsis, severe Sepsis, & Septic shock Identification and Management by 50% over the course of 8 months on the internal medicine service. Methods: Internal medicine residents are central to the identification and management of early Sepsis and Severe Sepsis at the Detroit VA. Given the monthly turnover of resident teams, we focused on our model for improvement on interventions at the start of each month with targeted educational sessions to define Sepsis, severe Sepsis, and septic shock (and their management). We have also implemented targeted EMR training on utilizing a recently developed “Sepsis Bundle order set” for patients who have been identified as Sepsis. Sepsis Menu data usage was collected based on health factor tags on patients with Sepsis as their ICD-10 code on discharge from the internal medicine service. Results: Internal Medicine Resident educational sessions and EMR training commenced on July 1st, 2024. Sepsis Bundle was implemented in the EMR late 2022 due to rising fallouts of Sepsis and severe sepsis cases. However, its usage remained minimal, as depicted in Figure One, when compared to the number of discharged patients. Following the educational sessions and EMR training, both the identification and management of Sepsis have increased. Prior to intervention, from October 2022 to June 2023, a mean sepsis bundle usage rate of 10% was observed for identified sepsis patients, whereas post-intervention, from July 2023 to March 2024, the mean has increased to 30%. Conclusion/Discussion: Based on the early data, we can confidently note that our monthly education sessions, paired with our process measure (Sepsis bundle use), have been effective. Ongoing opportunities for improvement are improved coordination and handoff between ER and IM/MICU. We continue to see the need for ongoing resident education on identification and appropriate documentation of Sepsis severity along with obtaining labs within a certain time frame.

A novel clinical prediction model for in-hospital mortality in sepsis patients complicated by ARDS: A MIMIC IV database and external validation study

BackgroundSepsis complicated by ARDS significantly increases morbidity and mortality, underscoring the need for robust predictive models to enhance patient management. MethodsWe collected data on 6390 patients with ARDS-complicated sepsis from the MIMIC IV database. Following rigorous data cleaning, including outlier management, handling missing values, and transforming variables, we conducted univariate analysis and logistic multivariate regression. We employed the LASSO machine learning algorithm to identify risk factors closely associated with patient outcomes. These factors were then used to develop a new clinical prediction model. The model underwent preliminary assessment and internal validation, and its performance was further tested through external validation using data from 225 patients at a major tertiary hospital in China. This validation assessed the model's discrimination, calibration, and net clinical benefits. ResultsThe model, illustrated by a concise nomogram, demonstrated significant discrimination with an area under the curve (AUC) of 0.711 in the internal validation set and 0.771 in the external validation set, outperforming conventional severity scores such as the SOFA and SAPS II. It also showed good calibration and net clinical benefits. ConclusionsOur model serves as a valuable tool for identifying sepsis patients with ARDS at high risk of in-hospital mortality. This could enable the implementation of personalized treatment strategies, potentially improving patient outcomes.

Clinical predictive value of the CRP-albumin-lymphocyte index for prognosis of critically ill patients with sepsis in intensive care unit: a retrospective single-center observational study.

Sepsis is a complex syndrome characterized by physiological, pathological, and biochemical abnormalities caused by infection. Its development is influenced by factors such as inflammation, nutrition, and immune status. Therefore, we combined C-reactive protein (CRP), albumin, and lymphocyte, which could reflect above status, to be the CRP-albumin-lymphocyte (CALLY) index, and investigated its association with clinical prognosis of critically ill patients with sepsis. This retrospective observational study enrolled critically ill patients with sepsis who had an initial CRP, albumin, and lymphocyte data on the first day of ICU admission. All data were obtained from the Affiliated Hospital of Jiangsu University. The patients were divided into quartiles (Q1-Q4) based on their CALLY index. The outcomes included 30-/60-day mortality and acute kidney injury (AKI) occurrence. The association between the CALLY index and these clinical outcomes in critically ill patients with sepsis was evaluated using Cox proportional hazards and logistic regression analysis. A total of 1,123 patients (63.0% male) were included in the study. The 30-day and 60-day mortality rates were found to be 28.1 and 33.4%, respectively, while the incidence of AKI was 45.6%. Kaplan-Meier analysis revealed a significant association between higher CALLY index and lower risk of 30-day and 60-day mortality (log-rank p < 0.001). Multivariate Cox proportional hazards analysis indicated that the CALLY index was independently associated with 30-day mortality [HR (95%CI): 0.965 (0.935-0.997); p = 0.030] and 60-day mortality [HR (95%CI): 0.969 (0.941-0.997); p = 0.032]. Additionally, the multivariate logistic regression model showed that the CALLY index served as an independent risk predictor for AKI occurrence [OR (95%CI): 0.982 (0.962-0.998); p = 0.033]. The findings of this study indicated a significant association between the CALLY index and both 30-day and 60-day mortality, as well as the occurrence of AKI, in critically ill patients with sepsis. These findings suggested that the CALLY index may be a valuable tool in identifying sepsis patients who were at high risk for unfavorable outcomes.

SIMPLIFIED IMMUNE-DYSREGULATION INDEX: A NOVEL MARKER PREDICTS 28-DAY MORTALITY OF INTENSIVE CARE PATIENTS WITH SEPSIS.

Background: Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. There is currently no simple immune-imbalance-driven indicator for patients with sepsis. Methods: This study was conducted in Peking Union Medical College Hospital. Patients with sepsis were identified according to Sepsis 3.0 after reviewing patient data from May 2018 through October 2022. Least absolute shrinkage and selection operator logistic regression was used for features selection. Receiver operating characteristic curves for 28-day mortality were used to compare the predictive performance of level of interleukin 6 (IL-6) and lymphocyte count (LY#) with that of the combined ratio, namely, the IL-6/LY# ratio. A Cox hazard model was also used to test the predictive performance of IL-6/LY# versus several other measurements. The dynamic trend of IL-6/LY# based on day 1 IL-6/LY# level was analyzed. Results: The mortality rate was 24.5% (220/898) in the study cohort. The LY#, IL-6 level, blood platelet count, Sequential Organ Failure Assessment score, Acute Physiology and Chronic Health Evaluation II score, heart rate, age and Fi o2 level were identified as key factors for predicting 28-day mortality. IL-6/LY# was identified as a core indicator according to Least absolute shrinkage and selection operator logistic regression analysis. IL-6/LY# was significantly higher in nonsurvivors than in survivors (348 [154.6-1371.7] vs. 42.3 [15.4-117.1]). IL-6/LY# yielded a higher area under the curve (0.852 [95% CI = 0.820-0.879]) than the level of IL-6 (0.776 [95% CI = 0.738-0.809]) and LY# (0.719 [95% CI = 0.677-0.755]) separately. Survival analysis of mortality risk versus the IL-6/LY# ratio suggested that IL-6/LY# was significantly more predictive of patient risk than the Sequential Organ Failure Assessment score or the other factors ( P = 1.5 × 10 -33 ). In trend analysis, as the trend of D1-D3-D7 IL-6/LY# decreases, the morality rate is lower than increase or fluctuate group (42.1% vs. 58.3%, 37.9% vs. 43.8%, 37.5% vs. 38.5% in high, moderate, and low D1 IL-6/LY# group separately). Conclusion: IL-6/LY# examined on first day in intensive care unit can be used as an immune-imbalance alert to identify sepsis patients with higher risk of 28-day mortality. Decreasing trend of IL-6/LY# suggests a lower 28-day mortality rate of sepsis patients.

The value of lymphocyte-to-C-reactive protein ratio for predicting clinical outcomes in patients with sepsis in intensive care unit: a retrospective single-center study.

The lymphocyte-to-C-reactive protein ratio (LCR) was a novel biomarker of inflammation that had been implicated in various diseases. Nevertheless, the role of LCR in the context of sepsis patients admitted to the Intensive Care Unit (ICU) had not been thoroughly elucidated. This study aimed to determine the significance of the LCR in predicting the prognosis of sepsis patients within ICU. A sample of sepsis patients requiring ICU care was selected from the Affiliated Hospital of Jiangsu University. These patients were then segmented into four quartiles based on their LCR levels. The primary endpoint of the study was 30-day mortality and the secondary endpoint was the occurrence of Acute Kidney Injury (AKI). Survival analysis, via the Kaplan-Meier method and log-rank test, was conducted to assess survival rates. Cox proportional hazards regression and logistic regression models were employed to investigate the association between LCR and clinical outcomes. Additional subgroup analyses were conducted to evaluate the influence of other confounding factors on the relationship between LCR and patient outcomes. A total of 1,123 patients were enrolled in this study, with a median age of 75 (65-84) years, and 707 (63.0%) of them were male. The 30-day mortality rate was 28.1%, while the incidence of AKI was 45.6%. A progressive decrease in LCR levels was found to be associated with an increased cumulative incidence of 30-day mortality (log-rank P < 0.001). Multivariable Cox proportional hazards analyses demonstrated that LCR was an independent predictor of 30-day mortality [per 1-unit increase in LCR: HR (95%CI): 0.370 (0.142-0.963); P = 0.042]. Additionally, multivariable logistic regression analysis revealed a significant association between LCR and AKI occurrence [per 1-unit increase in LCR: OR (95%CI): 0.541 (0.307-0.953); P = 0.034]. Furthermore, subgroup analysis indicated a stronger correlation for patients aged over 65 years compared to those aged 65 or younger (p for interaction <0.05) in predicting 30-day mortality or AKI occurrence based on LCR. A reduction in LCR was notably linked to 30-day mortality and the occurrence of AKI in sepsis patients. These findings suggested that LCR could potentially serve as a valuable tool in identifying sepsis patients at a heightened risk of adverse outcomes.

A MACHINE LEARNING MODEL DERIVED FROM ANALYSIS OF TIME-COURSE GENE-EXPRESSION DATASETS REVEALS TEMPORALLY STABLE GENE MARKERS PREDICTIVE OF SEPSIS MORTALITY.

Sepsis is associated with significant mortality and morbidity among critically ill patients admitted to intensive care units and represents a major health challenge globally. Given the significant clinical and biological heterogeneity among patients and the dynamic nature of the host immune response, identifying those at high risk of poor outcomes remains a critical challenge. Here, we performed secondary analysis of publicly available time-series gene-expression datasets from peripheral blood of patients admitted to the intensive care unit to elucidate temporally stable gene-expression markers between sepsis survivors and nonsurvivors. Using a limited set of genes that were determined to be temporally stable, we derived a dynamical model using a Support Vector Machine classifier to accurately predict the mortality of sepsis patients. Our model had robust performance in a test dataset, where patients' transcriptome was sampled at alternate time points, with an area under the curve of 0.89 (95% CI, 0.82-0.96) upon 5-fold cross-validation. We also identified 7 potential biomarkers of sepsis mortality (STAT5A, CX3CR1, LCP1, SNRPG, RPS27L, LSM5, SHCBP1) that require future validation. Pending prospective testing, our model may be used to identify sepsis patients with high risk of mortality accounting for the dynamic nature of the disease and with potential therapeutic implications.

Discharge Against Medical Advice After Hospitalization for Sepsis: Predictors, 30-Day Readmissions, and Outcomes

BackgroundSepsis is a leading cause of death worldwide. However, little has been known concerning the status of discharge against medical advice (DAMA) in sepsis patients. ObjectiveTo identify factors associated with DAMA, evaluate the association of DAMA with 30-day unplanned readmission and readmitted outcomes after sepsis hospitalization. MethodsUsing the National Readmission Database, we identified sepsis patients who discharged routinely or DAMA in 2017. Multivariable models were used to identify factors related to DAMA, evaluate the association between DAMA and readmission, and elucidate the relationship between DAMA and outcomes in patients readmitted within 30 days. ResultsAmong 1,012,650 sepsis cases, patients with DAMA accounted for 3.88% (n = 39,308). The unplanned 30-day readmission rates in patients who discharged home and DAMA were 13.08% and 27.21%, respectively. Predictors of DAMA in sepsis included Medicaid, diabetes, smoking, drug abuse, alcohol abuse, and psychoses. DAMA was statistically significantly associated with 30-day (odds ratio [OR] 2.18, 95% confidence interval [CI] 2.09–2.28), 60-day (OR 1.98, 95% CI 1.90–2.06), and 90-day (OR 1.88, 95% CI 1.81–1.96) readmission. DAMA is also associated with higher mortality in patients readmitted within 30 days (OR 1.38, 95% CI 1.17–1.63), whereas there were no statistically significant differences in length of stay and costs between patients who discharged home or DAMA. ConclusionsDAMA occurs in nearly 3.88% of sepsis patients and is linked to higher readmission and mortality. Those at high risk of DAMA should be early identified to motivate intervention to avoid premature discharges and associated adverse outcomes.

Early predicting 30-day mortality in sepsis in MIMIC-III by an artificial neural networks model

ObjectiveEarly identifying sepsis patients who had higher risk of poor prognosis was extremely important. The aim of this study was to develop an artificial neural networks (ANN) model for early predicting clinical outcomes in sepsis.MethodsThis study was a retrospective design. Sepsis patients from the Medical Information Mart for Intensive Care-III (MIMIC-III) database were enrolled. A predictive model for predicting 30-day morality in sepsis was performed based on the ANN approach.ResultsA total of 2874 patients with sepsis were included and 30-day mortality was 29.8%. The study population was categorized into the training set (n = 1698) and validation set (n = 1176) based on the ratio of 6:4. 11 variables which showed significant differences between survivor group and nonsurvivor group in training set were selected for constructing the ANN model. In training set, the predictive performance based on the area under the receiver-operating characteristic curve (AUC) were 0.873 for ANN model, 0.720 for logistic regression, 0.629 for APACHEII score and 0.619 for SOFA score. In validation set, the AUCs of ANN, logistic regression, APAHCEII score, and SOFA score were 0.811, 0.752, 0.607, and 0.628, respectively.ConclusionAn ANN model for predicting 30-day mortality in sepsis was performed. Our predictive model can be beneficial for early detection of patients with higher risk of poor prognosis.

Novel biomarker panel for the diagnosis and prognosis assessment of sepsis based on machine learning.

Background: The authors investigated a panel of novel biomarkers for diagnosis and prognosis assessment of sepsis using machine learning(ML) methods. Methods: Hematological parameters, liver function indices and inflammatory marker levels of 332 subjects were retrospectively analyzed. Results: The authors constructed sepsis diagnosis models and identified the random forest (RF) model to be the most optimal. Compared with PCT (procalcitonin) and CRP (C-reactive protein), the RF model identified sepsis patients at an earlier stage. The sepsis group had a mortality rate of 36.3%, and the RF model had greater predictive ability for the 30-day mortality risk of sepsis patients. Conclusion: The RF model facilitated the identification of sepsis patients and showed greater accuracy in predicting the 30-day mortality risk of sepsis patients.

Prospective, multi-site study of patient outcomes after implementation of the TREWS machine learning-based early warning system for sepsis.

Early recognition and treatment of sepsis are linked to improved patient outcomes. Machine learning-based early warning systems may reduce the time to recognition, but few systems have undergone clinical evaluation. In this prospective, multi-site cohort study, we examined the association between patient outcomes and provider interaction with a deployed sepsis alert system called the Targeted Real-time Early Warning System (TREWS). During the study, 590,736 patients were monitored by TREWS across five hospitals. We focused our analysis on 6,877 patients with sepsis who were identified by the alert before initiation of antibiotic therapy. Adjusting for patient presentation and severity, patients in this group whose alert was confirmed by a provider within 3 h of the alert had a reduced in-hospital mortality rate (3.3%, confidence interval (CI) 1.7, 5.1%, adjusted absolute reduction, and 18.7%, CI 9.4, 27.0%, adjusted relative reduction), organ failure and length of stay compared with patients whose alert was not confirmed by a provider within 3 h. Improvements in mortality rate (4.5%, CI 0.8, 8.3%, adjusted absolute reduction) and organ failure were larger among those patients who were additionally flagged as high risk. Our findings indicate that early warning systems have the potential to identify sepsis patients early and improve patient outcomes and that sepsis patients who would benefit the most from early treatment can be identified and prioritized at the time of the alert.

Variations among Electronic Health Record and Physiologic Streaming Vital Signs for Use in Predictive Algorithms in Pediatric Severe Sepsis

AbstractObjective This study sought to describe the similarities and differences among physiologic streaming vital signs (PSVSs) and electronic health record (EHR)-documented vital signs (EVSs) in pediatric sepsis.Methods In this retrospective cohort study, we identified sepsis patients admitted to the pediatric intensive care unit. We compared PSVS and EVS measures of heart rate (HR), respiratory rate, oxyhemoglobin saturation, and blood pressure (BP) across domains of completeness, concordance, plausibility, and currency.Results We report 1,095 epochs comprising vital sign data from 541 unique patients. While counts of PSVS measurements per epoch were substantially higher, increased missingness was observed compared with EVS. Concordance was highest among HR and lowest among BP measurements, with bias present in all measures. Percent of time above or below defined plausibility cutoffs significantly differed by measure. All EVS measures demonstrated a mean delay from time recorded at the patient to EHR entry.Conclusion We measured differences between vital sign sources across all data domains. Bias direction differed by measure, possibly related to bedside monitor measurement artifact. Plausibility differences may reflect the more granular nature of PSVS which can be critical in illness detection. Delays in EVS measure currency may impact real-time decision support systems. Technical limitations increased missingness in PSVS measures and reflect the importance of systems monitoring for data continuity. Both PSVS and EVS have advantages and disadvantages that must be weighed when making use of vital signs in decision support systems or as covariates in retrospective analyses.

Early warning model for death of sepsis via length insensitive temporal convolutional network.

Sepsis is a life-threatening systemic syndrome characterized by various biological, biochemical, and physiological abnormalities. Due to its high mortality, identifying sepsis patients with high risk of in-hospital death early and accurately will help doctors make optimal clinical decisions and reduce the mortality of sepsis patients. In this paper, we propose a length insensitive TCN-based model to predict sepsis patient's death risk in the future k hours, which is the first work for sepsis death risk early warning model only based on vital signs time series to our best knowledge. Furthermore, we design residual connections between temporal residual blocks to improve the prediction performance and stability especially on short input sequences. We validate and evaluate our model on two freely-available datasets, i.e., MIMIC-IV and eICU, from which 16,520 and 29,620 patients are selected respectively. The experiment results show that our model outperforms LSTM and other machine learning methods, as it has the highest sensitivity and Youden index in almost all cases. Meanwhile, the Youden index of the TCN-based model only slightly decreases by 0.0233 and 0.0307 when the time range of the input sequence changes from 24 to 4 h for k equal to 6 and 12, respectively.

8. Sepsis-Associated Acute Kidney Injury and Acute Kidney Disease: A 15-Year Cohort Study of 4,226 Adult Sepsis Inpatient Survivors at a Tertiary Medical Center in Taiwan

BackgroundSepsis is the most common cause of acute kidney injury (AKI) and about one-third of patients with sepsis-associated AKI (SA-AKI) develop acute kidney diseases (SA-AKD) and may progress to unfavorable outcomes. We aimed to study the characteristics and outcomes associated with SA-AKI and SA-AKD.MethodsThis cohort study included adult inpatients with first-time sepsis who were admitted during 2003-2017, had qualifying serum creatinine (SCr) measurements at baseline (-365 to -3 days), -2 to +7 days, and +8 to +90 days of sepsis index day, and survived the first 90 days (Figure 1). Sepsis was identified using an electronic medical records-based Sepsis-3 criteria. We classified sepsis inpatients into SA-AKI(-), SA-AKD(-), SA-relapsed-AKD, and SA-nonrecovery-AKD (Figure 2). ESRD and mortality were ascertained by linking to the Catastrophic Illness records and to National Death Registry, respectively. Multivariable Cox proportional hazard model was used to evaluate the risk of mortality and end-stage renal disease (ESRD) associated with SA-AKI/AKD subtypes.Figure 1. Flowchart of the selection process of adult sepsis survivors (N = 4226 patients). Figure 2. Definitions of sepsis associated-acute kidney injury (SA-AKI) and sepsis associated-acute kidney disease (SA-AKD). ResultsOf 4,226 eligible sepsis inpatient survivors, 47.1% developed SA-AKI and 10.1% progressed to SA-AKD (5.4% relapsed and 4.7% nonrecovery). Patient with AKI and non-recovered AKD had the worst baseline renal function (SCr, 1.3 mg/dL) (Table 1). The multivariable analyses revealed that SA-relapsed AKD was significantly associated with increased risk of all-cause mortality for 1-year (aHR 1.67; 95% CI 1.25, 2.24), 3-year (aHR 1.38; 95% CI 1.11, 1.71), and overall (aHR 1.35; 95% CI 1.12, 1.61), compared with SA-AKI(-). SA-relapsed AKD and SA-nonrecovery AKD were both significantly associated with 1-year, 3-year, and overall ESRD, with the risk of about 4-fold or higher than SA-AKI(-) (Table 2).Table 1. Baseline characteristics and outcomes among adult sepsis survivors, by different SA-AKI/AKD subtypes. Table 2. Risk of all-cause mortality and end stage renal disease (ESRD) among adult sepsis survivors. ConclusionSepsis survivors who initially had AKI and developed relapsed or nonrecovery AKD tended to have worse outcomes of all-cause and ESRD, compared with those without AKI. Unexpectedly, patients with non-recovered AKD did not have a higher mortality risk, possibly because we have selected those who survived the first 90 days of sepsis. We will develop two-stage prediction models to identify sepsis patients at risk of developing AKI and SA-AKI patients at risk of developing different types of AKD. Disclosures All Authors: No reported disclosures

Multi-Health System Comparison of Diagnostic Testing in Hospitalized Sepsis Patients

Abstract Introduction/Objective Comparisons of diagnostic testing across multiple health systems over multiple years are uncommon. Such comparisons would quantify variations in test use between health systems. Methods/Case Report Using the Vizient Clinical Database, we compared clinical laboratory and imaging utilization for hospitalized adult sepsis patients (N = 69,035) over three years (2017-2019) at 19 large academic-affiliated health systems across the United States. We used Medicare Severity Diagnosis Related Groups (MSDRG), employed by the US Centers for Medicare and Medicaid Services (CMS), and identified sepsis patients (MSDRG triplet 870/871/872). We stratified hospitalized sepsis patients by severity of illness (SOI) into high severity (MSDRG 870), moderate severity (MSDRG 871), and low severity (MSDRG 872) groups. SOI further categorizes patients within a diagnostic group, quantifying the extent of comorbid conditions and complications. We measured hospital length of stay (LOS), number of laboratory tests (CPT codes 80000-89999), and number of imaging studies (CPT codes 70000-79999). We divided the number of laboratory tests and radiology studies by mean hospital LOS (in days) to calculate laboratory tests and imaging studies per hospital day. Results (if a Case Study enter NA) Between health systems, lowest and highest values for laboratory and imaging utilization ranged from 50.1 to 141.3 tests per hospitalization and 1.4 to 7.2 studies per hospitalization, respectively. Differences in laboratory tests between health systems persisted after adjusting for SOI with low to high laboratory tests per hospitalization ranging from 127.8 to 405.7, 51.0 to 144.7, and 31.9 to 78.2, for high, moderate, and low SOI groups. Utilization ranges were smaller for imaging studies. After adjusting for hospital LOS, laboratory testing differences between health systems were more pronounced and with low and high laboratory test per day utilization of 6.5 to 24.3, 6.1 to 18.5, and 6.0 to 17.1, for high, moderate, and low SOI groups. Differences in radiology studies were not as pronounced after adjusting for LOS. Conclusion There is considerable variation among health systems in laboratory and radiology resource utilization for hospitalized sepsis patients. This variation persists, especially for laboratory testing, after adjusting for SOI and LOS.

Machine Learning Model to Identify Sepsis Patients in the Emergency Department: Algorithm Development and Validation.

Accurate stratification of sepsis can effectively guide the triage of patient care and shared decision making in the emergency department (ED). However, previous research on sepsis identification models focused mainly on ICU patients, and discrepancies in model performance between the development and external validation datasets are rarely evaluated. The aim of our study was to develop and externally validate a machine learning model to stratify sepsis patients in the ED. We retrospectively collected clinical data from two geographically separate institutes that provided a different level of care at different time periods. The Sepsis-3 criteria were used as the reference standard in both datasets for identifying true sepsis cases. An eXtreme Gradient Boosting (XGBoost) algorithm was developed to stratify sepsis patients and the performance of the model was compared with traditional clinical sepsis tools; quick Sequential Organ Failure Assessment (qSOFA) and Systemic Inflammatory Response Syndrome (SIRS). There were 8296 patients (1752 (21%) being septic) in the development and 1744 patients (506 (29%) being septic) in the external validation datasets. The mortality of septic patients in the development and validation datasets was 13.5% and 17%, respectively. In the internal validation, XGBoost achieved an area under the receiver operating characteristic curve (AUROC) of 0.86, exceeding SIRS (0.68) and qSOFA (0.56). The performance of XGBoost deteriorated in the external validation (the AUROC of XGBoost, SIRS and qSOFA was 0.75, 0.57 and 0.66, respectively). Heterogeneity in patient characteristics, such as sepsis prevalence, severity, age, comorbidity and infection focus, could reduce model performance. Our model showed good discriminative capabilities for the identification of sepsis patients and outperformed the existing sepsis identification tools. Implementation of the ML model in the ED can facilitate timely sepsis identification and treatment. However, dataset discrepancies should be carefully evaluated before implementing the ML approach in clinical practice. This finding reinforces the necessity for future studies to perform external validation to ensure the generalisability of any developed ML approaches.

Prediction of 90-Day Mortality among Sepsis Patients Based on a Nomogram Integrating Diverse Clinical Indices.

Background Sepsis is prevalent among intensive care units and is a frequent cause of death. Several studies have identified individual risk factors or potential predictors of sepsis-associated mortality, without defining an integrated predictive model. The present work was aimed at defining a nomogram for reliably predicting mortality. Methods We carried out a retrospective, single-center study based on 231 patients with sepsis who were admitted to our intensive care unit between May 2018 and October 2020. Patients were randomly split into training and validation cohorts. In the training cohort, multivariate logistic regression and a stepwise algorithm were performed to identify risk factors, which were then integrated into a predictive nomogram. Nomogram performance was assessed against the training and validation cohorts based on the area under receiver operating characteristic curves (AUC), calibration plots, and decision curve analysis. Results Among the 161 patients in the training cohort and 70 patients in the validation cohort, 90-day mortality was 31.6%. Older age and higher values for the international normalized ratio, lactate level, and thrombomodulin level were associated with greater risk of 90-day mortality. The nomogram showed an AUC of 0.810 (95% CI 0.739 to 0.881) in the training cohort and 0.813 (95% CI 0.708 to 0.917) in the validation cohort. The nomogram also performed well based on the calibration curve and decision curve analysis. Conclusion This nomogram may help identify sepsis patients at elevated risk of 90-day mortality, which may help clinicians allocate resources appropriately to improve patient outcomes.

MiR-486-5p Serves as a Diagnostic Biomarker for Sepsis and Its Predictive Value for Clinical Outcomes.

BackgroundAs a molecular detection method, miRNA can quickly diagnose and prevent diseases, intervene in disease as early as possible, and reduce mortality. This study was to investigate the potential clinical diagnostic and predictive significance of miR-486-5p in sepsis and its correlation with inflammation and disease severity.MethodsThe serum miR-486-5p in 108 sepsis, 60 pneumonia-infected, and 101 healthy controls were detected by RT-qPCR. Spearman coefficient detects the correlation between serum miRNA and disease severity indicators (APACHE II, SOFA scores), and inflammation indicators (CRP, PCT), respectively. The diagnostic significance of miR-486-5p in sepsis was analyzed by the ROC curve. Kaplan–Meier estimator and Cox regression hazards analysis of the predictive significance of serum miR-486-5p in 28-day survival from sepsis.ResultsSerum miR-486-5p was increased in sepsis patients compared with healthy control and pneumonia-infected patients (P < 0.001). And increased serum miR-486-5p was positively associated with disease severity (SOFA score and APACHE II score) and inflammation (CRP and PCT). Serum miR-486-5p can not only identify sepsis patients from healthy controls (AUC = 0.914) but also significantly distinguish sepsis patients from pneumonia-infected patients (AUC = 0.814), showing good potential as a diagnostic biomarker for sepsis. In addition, serum miR-486-5p was an independent predictor of 28-day survival (log-rank P = 0.012), and patients with high levels of miR-486-5p had a poorer overall 28-day survival (HR = 3.057, 95% CI = 1.385–17.817, P = 0.014).ConclusionmiR-486-5p is a potential diagnostic biomarker for sepsis, and its high level is significantly correlated with the disease severity and inflammation. In addition, miR-486-5p were predictive risk factors for 28-day survival in sepsis patients.

What's New in Shock, July 2021?

What's New in Shock, July 2021?

Vasoactive-Inotropic Score as an Early Predictor of Mortality in Adult Patients with Sepsis

Vasoactive and inotropic medications are essential for sepsis management; however, the association between the maximum Vasoactive-Inotropic score (VISmax) and clinical outcomes is unknown in adult patients with sepsis. We investigated the VISmax as a predictor for mortality among such patients in the emergency department (ED) and compared its prognostic value with that of the sequential organ failure assessment (SOFA) score. This single-center retrospective study included 910 patients diagnosed with sepsis between January 2016 and March 2020. We calculated the VISmax using the highest doses of vasopressors and inotropes administered during the first 6 h on ED admission and categorized it as 0–5, 6–15, 16–30, 31–45, and >45 points. The primary outcome was 30-day mortality. VISmax for 30-day mortality was significantly higher in non-survivors than in survivors. The mortality rates in the five VISmax groups were 17.2%, 20.8%, 33.3%, 54.6%, and 70.0%, respectively. The optimal cut-off value of VISmax to predict 30-day mortality was 31. VISmax had better prognostic value than the cardiovascular component of the SOFA score and initial lactate levels. VISmax was comparable to the APACHE II score in predicting 30-day mortality. Multivariable analysis showed that VISmax 16–30, 31–45, and >45 were independent risk factors for 30-day mortality. VISmax in ED could help clinicians to identify sepsis patients with poor prognosis.