Risk factors of sepsis among patients with qSOFA<2 in the emergency department

Interpretable machine learning for predicting sepsis risk in emergency triage patients

The study aimed to develop and validate a sepsis prediction model using structured electronic medical records (sEMR) and machine learning (ML) methods in emergency triage. The goal was to enhance early sepsis screening by integrating comprehensive triage information beyond vital signs. This retrospective cohort study utilized data from the MIMIC-IV database. Two models were developed: Model 1 based on vital signs alone, and Model 2 incorporating vital signs, demographic characteristics, medical history, and chief complaints. Eight ML algorithms were employed, and model performance was evaluated using metrics such as AUC, F1 Score, and calibration curves. SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) methods were used to enhance model interpretability. The study included 189,617 patients, with 5.95% diagnosed with sepsis. Model 2 consistently outperformed Model 1 across most algorithms. In Model 2, Gradient Boosting achieved the highest AUC of 0.83, followed by Extra Tree, Random Forest, and Support Vector Machine (all 0.82). The SHAP method provided more comprehensible explanations for the Gradient Boosting algorithm. Modeling with comprehensive triage information using sEMR and ML methods was more effective in predicting sepsis at triage compared to using vital signs alone. Interpretable ML enhanced model transparency and provided sepsis prediction probabilities, offering a feasible approach for early sepsis screening and aiding healthcare professionals in making informed decisions during the triage process.

A clinical prediction model for safe early discharge of patients with an infection at the emergency department

BACKGROUND

Every hospital admission is associated with healthcare costs and a risk of adverse events. The need to identify patients who do not require hospitalization has emerged with the profound increase in hospitalization rates due to infectious diseases during the last decades, especially during the COVID-19 pandemic. This study aimed to identify predictors of safe early discharge (SED) in patients presenting to the emergency department (ED) with a suspected infection meeting the Systemic Inflammatory Response Syndrome (SIRS) criteria.

METHODS

We conducted a prospective cohort study on adult non-trauma patients with a suspected infection and at least two SIRS criteria. We defined SED as hospital discharge within 24 h (e.g. direct ED discharge or rapid ward discharge) without disease-related readmission to our hospital or death during the first seven days. A prediction model for SED was developed using multivariate logistic regression analysis and tested with k-fold cross-validation.

RESULTS

We included 1381 patients, of whom 1027 (74.4 %) were hospitalized for longer than 24 h or re-admitted within seven days and 354 (25.6 %) met SED criteria. Parameters associated with SED were relatively young age, absence of comorbidities, living independently, yellow or green triage urgency, lack of ambulance transport or general practitioner referral, normal clinical impression scores, and risk scores (i.e., qSOFA, PIRO, MEDS, NEWS, and SIRS), normal vital sign measurements and absence of kidney and respiratory failure. The model performance metrics showed an area under the curve of 0.824. The validation showed a minimal drop in performance and indicated a good fit.

CONCLUSION

We developed and validated a model to identify patients with an infection at the ED who can be safely discharged early.

Prediction of sepsis among patients with major trauma using artificial intelligence: a multicenter validated cohort study

BACKGROUND

Sepsis remains a significant challenge in patients with major trauma in the ICU. Early detection and treatment are crucial for improving outcomes and reducing mortality rates. Nonetheless, clinical tools for predicting sepsis among patients with major trauma are limited. This study aimed to develop and validate an artificial intelligence (AI) platform for predicting the risk of sepsis among patients with major trauma.

PATIENTS AND METHODS

This study involved 961 patients, with a prospective analysis of data from 244 patients with major trauma at our hospital and a retrospective analysis of data from 717 patients extracted from a database in the United States. The patients from our hospital constituted the model development cohort, and the patients from the database constituted the external validation cohort. The patients in the model development cohort were randomly divided into a training cohort and an internal validation cohort at a ratio of 8:2. The machine-learning algorithms used to train models included logistic regression, decision tree, extreme gradient boosting machine (eXGBM), neural network (NN), random forest, and light gradient boosting machine (LightGBM).

RESULTS

The incidence of sepsis for the model development cohort was 43.44%. Twelve predictors, including gender, abdominal trauma, open trauma, red blood cell count, heart rate, respiratory rate, injury severity score, sequential organ failure assessment score, Glasgow coma scale, smoking, total protein concentrations, and hematocrit, were used as features in the final model. Internal validation showed that the NN model had the highest area under the curve (AUC) of 0.932 (95% CI: 0.917-0.948), followed by the LightGBM and eXGBM models with AUCs of 0.913 (95% CI: 0.883-0.930) and 0.912 (95% CI: 0.880-0.935), respectively. In the external validation cohort, the eXGBM model (AUC: 0.891, 95% CI: 0.866-0.914) had the highest AUC value, followed by the LightGBM model (AUC: 0.886, 95% CI: 0.860-0.906), and the AUC value of the NN model was only 0.787 (95% CI: 0.751-0.829). Considering the predictive performance for both the internal and external validation cohorts, the LightGBM model had the highest score of 82, followed by the eXGBM (81) and NN (76) models. Thus, the LightGBM has emerged as the optimal model, and it was deployed online as an AI application.

CONCLUSIONS

This study develops and validates an AI application to effectively assess the susceptibility of patients with major trauma to sepsis. The AI application equips healthcare professionals with a valuable tool to promptly identify individuals at high risk of developing sepsis. This will facilitate clinical decision-making and enable early intervention.

Prediction model of death risk in patients with sepsis and screening of biomarkers for prognosis of patients with myocardial injury

This study aimed to create a robust prediction model for sepsis patient mortality and identify key biomarkers in those with myocardial injury. A retrospective analysis of 261 sepsis inpatients was conducted, with 44 deaths and 217 recoveries. Key factors were assessed via univariate and multivariate analyses, revealing myocardial injury, shock, and pulmonary infection as independent mortality risk factors. Using LASSO regression, a reliable prediction model was developed and internally validated. Additionally, procalcitonin (PCT) emerged as a sensitive biomarker for myocardial injury prediction in sepsis patients. In summary, this study highlights myocardial injury, shock, and pulmonary infection as independent risk factors for sepsis-related deaths. The LASSO-based prediction model effectively forecasts the prognosis of septic patients with myocardial injury, with PCT showing promise as a predictive biomarker.

Interpretable machine-learning model for real-time, clustered risk factor analysis of sepsis and septic death in critical care

BACKGROUND AND OBJECTIVE

Interpretable and real-time prediction of sepsis and risk factor analysis could enable timely treatment by clinicians and improve patient outcomes. To develop an interpretable machine-learning model for the prediction and risk factor analysis of sepsis and septic death.

METHODS

This is a retrospective observational cohort study based on the Medical Information Mart for Intensive Care (MIMIC-IV) dataset; 69,619 patients from the database were screened. The two outcomes include patients diagnosed with sepsis and the death of septic patients. Clinical variables from ICU admission to outcomes were analyzed: demographic data, vital signs, Glasgow Coma Scale scores, laboratory test results, and results for arterial blood gasses (ABGs). Model performance was compared using the area under the receiver operating characteristic curve (AUROC). Model interpretations were based on the Shapley additive explanations (SHAP), and the clustered analysis was based on the combination of K-means and dimensionality reduction algorithms of t-SNE and PCA.

RESULTS

For the analysis of sepsis and septic death, 47,185 and 2480 patients were enrolled, respectively. The XGBoost model achieved a predictive value of area under the curve (AUC): 0.745 [0.731-0.759] for sepsis prediction and 0.8 [0.77, 0.828] for septic death prediction. The real-time prediction model was trained to predict by day and visualize the individual or combined risk factor effects on the outcomes based on SHAP values. Clustered analysis separated the two phenotypes with distinct risk factors among patients with septic death.

CONCLUSION

The proposed real-time, clustered prediction model for sepsis and septic death exhibited superior performance in predicting the outcomes and visualizing the risk factors in a real-time and interpretable manner to distinguish and mitigate patient risks, thus promising immense potential in effective clinical decision making and comprehensive understanding of complex diseases such as sepsis.

Role of the Renin Angiotensin Aldosterone System in the Pathogenesis of Sepsis-Induced Acute Kidney Injury: A Systematic Review

BACKGROUND

Sepsis is a life-threatening condition responsible for up to 20% of all global deaths. Kidneys are among the most common organs implicated, yet the pathogenesis of sepsis-induced acute kidney injury (S-AKI) is not completely understood, resulting in the treatment being nonspecific and responsive. In situations of stress, the renin angiotensin aldosterone system (RAAS) may play a role. This systematic review focuses on analyzing the impact of the RAAS on the development of S-AKI and discussing the use of RAAS antagonists as an emerging therapeutic option to minimize complications of sepsis.

METHODS

Studies were identified using electronic databases (Medline via PubMed, Google Scholar) published within the past decade, comprised from 2014 to 2023. The search strategy was conducted using the following keywords: sepsis, S-AKI, RAAS, Angiotensin II, and RAAS inhibitors. Studies on human and animal subjects were included if relevant to the keywords.

RESULTS

Our search identified 22 eligible references pertaining to the inclusion criteria. Treatment of sepsis with RAAS inhibitor medications is observed to decrease rates of S-AKI, reduce the severity of S-AKI, and offer an improved prognosis for septic patients.

CONCLUSION

The use of RAAS antagonists as a treatment after the onset of sepsis has promising findings, with evidence of decreased renal tissue damage and rates of S-AKI and improved survival outcomes.

REGISTRATION

INPLASY202360098.

The prognostic utility of prehospital qSOFA in addition to emergency department qSOFA for sepsis in patients with suspected infection: A retrospective cohort study

BACKGROUND

The quick sequential organ failure assessment (qSOFA) was widely used to estimate the risks of sepsis in patients with suspected infection in the prehospital and emergency department (ED) settings. Due to the insufficient sensitivity of qSOFA on arrival at the ED (ED qSOFA), the Surviving Sepsis Campaign 2021 recommended against using qSOFA as a single screening tool for sepsis. However, it remains unclear whether the combined use of prehospital and ED qSOFA improves its sensitivity for identifying patients at a higher risk of sepsis at the ED.

METHODS

We retrospectively analyzed the data from the ED of a tertiary medical center in Japan from April 2018 through March 2021. Among all adult patients (aged ≥18 years) transported by ambulance to the ED with suspected infection, we identified patients who were subsequently diagnosed with sepsis based on the Sepsis-3 criteria. We compared the predictive abilities of prehospital qSOFA, ED qSOFA, and the sum of prehospital and ED qSOFA (combined qSOFA) for sepsis in patients with suspected infection at the ED.

RESULTS

Among 2,407 patients with suspected infection transported to the ED by ambulance, 369 (15%) patients were subsequently diagnosed with sepsis, and 217 (9%) died during hospitalization. The sensitivity of prehospital qSOFA ≥2 and ED qSOFA ≥2 were comparable (c-statistics for sepsis [95%CI], 0.57 [0.52-0.62] vs. 0.55 [0.50-0.60]). However, combined qSOFA (cutoff, ≥3 [max 6]) was more sensitive than ED qSOFA (cutoff, ≥2) for identifying sepsis (0.67 [95%CI, 0.62-0.72] vs. 0.55 [95%CI, 0.50-0.60]). Using combined qSOFA, we identified 44 (12%) out of 369 patients who were subsequently diagnosed with sepsis, which would have been missed using ED qSOFA alone.

CONCLUSIONS

Using both prehospital and ED qSOFA could improve the screening ability of sepsis among patients with suspected infection at the ED.

Association between comorbid mental illness and preceding emergency department visits in unplanned admissions

Aim

To investigate the association between comorbid mental illness and preceding emergency department (ED) visits in patients with unplanned admission.

Methods

This is a retrospective observational study using data from the EDs of three large tertiary medical facilities in Japan. We included adult patients who were admitted to these hospitals via the ED from 2017 to 2020. To investigate whether patients with mental illness were more likely to have preceding ED visits within 30 days prior to unplanned admissions compared with those without, we used univariate and multivariable logistic regression models. In the multivariable model, we adjusted for age category, gender, facility, year, and ambulance use.

Results

Out of 15,429 total admissions, 766 (5.0%) cases had documented comorbid mental illness and 14,663 (95.0%) did not. The prevalence of preceding ED visits among patients with mental illness was significantly higher than in those without (17.1% vs 8.8%; unadjusted odds ratio 2.15, 95% confidence interval [CI] 1.76-2.61; P < 0.001). This association was more prominent in the multivariable regression model (adjusted odds ratio 2.40, 95% CI 1.97-2.94; P < 0.001).

Conclusions

The presence of mental illness was significantly associated with a higher prevalence of preceding ED visits within 30 days prior to the unplanned admission. The result suggests that physicians should be more cautious in discharging patients with mental illness from the EDs and in providing care after ED discharge.

Characteristics of patients with hip fractures and comorbid fall-related injuries in the emergency department

Aim

Hip fracture is one of the most common fall‐related injuries in the elderly population. Although falls may cause multiple types of injuries, no study has investigated the details of fall‐related injuries accompanied by hip fractures. This study aimed to characterize the features of such injuries.

Methods

This is a cross‐sectional study using data from four tertiary emergency departments in Japan. We identified patients diagnosed with hip fracture including femoral neck fracture, trochanter fracture, or subtrochanteric fracture from May 12, 2014 to July 12, 2021. Among patients with hip fracture, we included those with fall‐related hip fracture. We excluded patients ages <40 years old and whose fall was high energy onset, defined as fall from more than three steps or 1 m.

Results

Among 326 emergency departments patients diagnosed with fall‐related hip fracture, 288 patients were eligible for the analysis. Seventeen patients (6%) had injuries in addition to hip fractures. The most frequent injury was upper limb injury (e.g., distal radial fracture; n = 5, 30%), followed by head injury (e.g., subdural hematoma; n = 4, 24%), chest injury (e.g., pneumothorax; n = 2, 12%), and trunk injury (vertebral compression fracture; n = 2, 12%). There were no significantly different clinical characteristics between patients with hip injuries and those without.

Conclusion

A total of 6% of patients diagnosed with hip fracture had other fall‐related injuries. The most frequent were upper limb injury and head injury. Our findings underscore the importance of whole‐body assessment in patients with fall‐related hip fracture in the emergency department.

External validation of the POP score for predicting obstetric and gynecological diseases in the emergency department

BACKGROUND

The POP score was developed as an easy screening tool for predicting obstetrics and gynecological (OBGYN) diseases in the emergency department (ED), and consists of three predictors, each representing one point: past history of OBGYN diseases, no fever or digestive symptoms, and peritoneal irritation signs). However, its external validity has not yet been evaluated. We aimed to perform the external validation of the POP score.

METHODS

This is a multi-center, retrospective cohort study using ED data of three tertiary care hospitals in Japan between Jan 2017 and October 2020. Young adult women aged 16-49 years with abdominal pain were included in the analysis. The probability of OBGYN diseases was calculated using a logistic regression model of the POP score. Predictions were compared with observations to evaluate the calibration of the model. Further, the diagnostic ability (sensitivity, specificity, and likelihood ratio) of the POP score was evaluated.

RESULTS

Of 66,599 ED visits, 1026 young adult women (median age [interquartile range]: 31 [23-41] years) were included for the analysis. The c-statistic was 0.645 [95% confidence interval (CI): 0.603-0.687]. The predicted probabilities of OBGYN diseases was generally well-calibrated to the observations. When the cut-off was set between 2 and 3 points for the ruling in of OBGYN diseases, the positive likelihood ratio was 9.72 [95% CI: 3.33-28.4]. When the cut-off was set between 0 and 1 points for ruling out of OBGYN diseases, negative likelihood ratio was 0.181 [95% CI: 0.059-0.558].

CONCLUSIONS

Using ED data of three tertiary care hospitals, we externally validated the POP score for prediction of OBGYN diseases in the ED. The POP score likely has clinical value for screening OBGYN diseases in young adult women with abdominal pain in the ED.