Neonatal infections pose a significant threat to the health of newborns. Associated morbidity and mortality risks underscore the urgency of prompt diagnosis and treatment with appropriate empiric antibiotics. Delay in treatment can be fatal; thus, early detection improves outcomes. However, diagnosing early is a challenge as signs and symptoms of neonatal infection are non-specific and overlap with non-infectious conditions. Machine learning (ML) offers promise in early detection, utilizing various data sources and methodologies. However, ML models require rigorous validation and consideration of various challenges, including false alarms and user acceptance requiring careful integration and ongoing evaluation for successful implementation.

Diagnostic errors in health care pose significant risks to patient safety and are disturbingly common. In the emergency department (ED), the chaotic and high-pressure environment increases the likelihood of these errors, as emergency clinicians must make rapid decisions with limited information, often under cognitive overload. Artificial intelligence (AI) offers promising solutions to improve diagnostic errors in three key areas: information gathering, clinical decision support (CDS), and feedback through quality improvement. AI can streamline the information-gathering process by automating data retrieval, reducing cognitive load, and providing clinicians with essential patient details quickly. AI-driven CDS systems enhance diagnostic decision making by offering real-time insights, reducing cognitive biases, and prioritizing differential diagnoses. Furthermore, AI-powered feedback loops can facilitate continuous learning and refinement of diagnostic processes by providing targeted education and outcome feedback to clinicians. By integrating AI into these areas, the potential for reducing diagnostic errors and improving patient safety in the ED is substantial. However, successfully implementing AI in the ED is challenging and complex. Developing, validating, and implementing AI as a safe, human-centered ED tool requires thoughtful design and meticulous attention to ethical and practical considerations. Clinicians and patients must be integrated as key stakeholders across these processes. Ultimately, AI should be seen as a tool that assists clinicians by supporting better, faster decisions and thus enhances patient outcomes.

Sepsis is a significant health burden on a global scale. Timely identification and treatment of sepsis can greatly improve patient outcomes, including survival rates. However, time-consuming laboratory results are often needed for screening sepsis. We aimed to develop a quick sepsis screening tool (qSepsis) based on patients' non-laboratory clinical data at the emergency department (ED) using machine learning (ML), and compare its performance with established clinical scores.

This retrospective study included patients admitted to the ED of Zhongnan Hospital of Wuhan University (Wuhan, China) from 1/1/2015 to 5/31/2022. Patients who were under 18 years of age, had cardiopulmonary arrest upon arrival at the ED, or had missing and abnormal medical record data were excluded. The qSepsis was derived by three ML algorithms, including logistic regression (LR), random forest (RF), and extreme gradient boosting (XGB). To benchmark the existing clinical tools for assessing the risk of sepsis in the ED, qSepsis was compared with the Systemic Inflammatory Response Syndrome (SIRS), the Quick Sepsis-Related Organ Failure Assessment (qSOFA), and the Modified Early Warning Score (MEWS). The external validation was performed with the Medical Information Mart for Intensive Care IV ED database (United States), and adopted the same inclusion and exclusion criteria. The predictive power of qSepsis and other clinical scores was measured using the area under the receiver operating characteristic curve (AUROC). The primary outcome of the study was the diagnosis of sepsis in the ED based on the Sepsis 3.0 criteria, which served as the basis for developing the qSepsis tool.

A total of 414,864 patients were finally included in the cohort (median ([IQR]) patient age, 43 (29, 60) years; 202,730 (48.87%) females, 212,134 (51.13%) males), and 200,089 in the external testing cohort (median (SD) patient age, 57 (39, 71) years; 107,427 (53.69%) females, 92,663 (46.31%) males). For internal testing, LR outperformed RF and XGB with an AUROC of 0.862 (95% CI, 0.855-0.869). In external testing, the AUROC decreased to 0.766 (95% CI, 0.758-0.774) for LR, 0.725 (95% CI, 0.717-0.733) for RF, and 0.735 (95% CI, 0.728-0.742) for XGB. In addition, the AUROC for the qSOFA, MEWS, and SIRS scores in external validation cohort were 0.579 (95% CI, 0.563-0.596), 0.600 (95% CI, 0.578-0.622), and 0.704 (95% CI, 0.683-0.725), respectively. The area under the precision-recall curve (AUPRC) for the qSepsis model was 0.213 (95% CI: 0.204-0.222). The AUPRC values for the other scores were as follows: SIRS, 0.071 (95% CI: 0.013-0.099); qSOFA, 0.096 (95% CI: 0.003-0.186); and MEWS, 0.083 (95% CI: 0.063-0.111).

This retrospective study demonstrated that qSepsis had better predictive performance in terms of AUROC and area under the precision-recall curve (AUPRC) compared to existing assessment scores. It has the potential to be used in pre-hospital settings with limited access to laboratory tests and in the ED for quick screening of patients with sepsis. However, due to its low positive predictive value (PPV), the false alarms may increase in actual clinical practice.

Transformation of Scientific and Technological Achievements Fund Project of Zhongnan Hospital of Wuhan University.

The influence of disease-causing pathogen on acute kidney injury (AKI) in septic patients is poorly understood.

We examined the association of microbial pathogen with AKI among patients with community-onset sepsis.

This was a retrospective cohort study. Patient data were acquired from the nationwide multicenter PINC AI Healthcare Database (2016-2020). Participants included adult patients with Centers for Disease Control and Prevention-defined community-onset sepsis.

The primary exposure was pathogen type identified by culture growth. Microbial cultures from any site were included. The primary endpoint was development of AKI within 7 days of admission using the Kidney Disease: Improving Global Outcomes serum creatinine criteria. We used multilevel logistic regression to examine the association between pathogen type and AKI. Escherichia coli-positive cultures were used as the reference category.

We included 119,733 patients with community-onset sepsis. The median age was 67 years, 33.3% were mechanically ventilated, 36.1% received vasopressors, and hospital mortality was 13.1%. Forty-two thousand twenty-seven patients (35.1%) developed stage 1 AKI, 22,979 (19.2%) developed stage 2 AKI, and 25,073 (20.9%) developed stage 3 AKI. Relative to patients with E. coli infection (odds ratio [OR], 1.0), Proteus species (OR, 1.26; 95% CI, 1.06-1.50), and Streptococcus species (OR, 1.24; 95% CI, 1.10-1.41) were associated with increased odds of AKI. Meanwhile, Pseudomonas aeruginosa (OR, 0.56; 95% CI, 0.49-0.64) and Serratia species (OR, 0.70; 95% CI, 0.52-0.94) were associated with decreased odds of AKI.

The causative pathogen in patients with sepsis may influence the development of AKI. Further mechanistic and clinical research is needed to confirm these findings and to explore how different pathogens may affect AKI risk in critically ill patients.

Aims/Background Sepsis is a life-threatening condition resulting from dysregulated immune responses to infection, leading to organ dysfunction. High-density lipoprotein (HDL) and red cell distribution width (RDW) have shown significant correlations with sepsis severity, yet the combined prognostic value of HDL and RDW in evaluating sepsis severity and outcomes remains unclear. This study examines the relationship between HDL and RDW levels and sepsis severity, as well as evaluates the combined utility of these markers in predicting disease severity and patient outcomes. Methods This retrospective study included 103 patients diagnosed with sepsis. Clinical data, including HDL and RDW levels, were collected for analysis. Patients were divided into shock and non-shock groups based on the presence of septic shock and into survival and death groups based on 30-day in-hospital mortality. Multivariate logistic regression was used to identify factors influencing sepsis severity and prognosis, while the predictive value of HDL in combination with RDW was evaluated using receiver operating characteristic (ROC) curve analysis. Results Multivariate analysis identified sequential organ failure assessment (SOFA) score (OR = 6.566), interleukin-6 (IL-6) (OR = 2.568), HDL (OR = 0.864), and RDW (OR = 4.052) as independent predictors of sepsis severity (p < 0.05 for all). ROC analysis demonstrated that HDL combined with RDW yielded the highest diagnostic accuracy for sepsis severity, with an area under curve (AUC) of 0.962, sensitivity of 97.56%, and specificity of 91.94%. Additionally, SOFA score (OR = 2.354), interleukin-6 (IL-6) (OR = 1.446), HDL (OR = 0.870), and RDW (OR = 3.502) were independent prognostic indicators (p < 0.05 for all). ROC analysis for prognosis showed that HDL combined with RDW had the highest predictive efficacy for the prognosis of sepsis, with an AUC of 0.922, sensitivity of 79.31%, and specificity of 93.24%. Conclusion The combination of HDL and RDW is a robust indicator for the evaluation of sepsis severity and is a valuable prognostic tool for assessing 30-day mortality risk in sepsis patients.

Electrolyte disturbances are prevalent complications in critically ill patients with sepsis, significantly impacting patient prognosis. However, the specific association between serum potassium levels and mortality risk in this population remains poorly understood. This study aimed to investigate the association between serum potassium levels during hospitalization and the risk of 28-day and 90-day mortality in critically ill patients with sepsis.

Data were obtained from the Medical Information Mart for Intensive Care (MIMIC-IV) database, and patients with severe sepsis requiring ICU admission were stratified into quartiles based on their mean serum potassium levels. Outcomes assessed included 28-day and 90-day mortality. A multivariate Cox proportional hazards model was used to investigate the association between serum potassium levels and mortality, with restricted cubic splines to identify potential nonlinear correlations. A dichotomous Cox proportional hazards model was applied to analyze the association further, and Kaplan-Meier analysis assessed the mortality risk across different potassium ranges.

A total of 25,203 patients were included, with 28-day and 90-day mortality rates of 27.84% and 40.48%, respectively. Multivariate analysis showed a significant association between serum potassium levels and mortality. Restricted cubic splines identified an inflection point at 4.4 mmol/L, with potassium levels above this threshold associated with higher mortality (28-day mortality: HR 2.96, 95% CI = 2.43-3.60; 90-day mortality: HR 2.19, 95% CI = 1.81-2.64). Kaplan-Meier analysis confirmed a significantly higher risk of death for patients with serum potassium levels above 4.4 mmol/L compared to those within the 3.5-4.4 mmol/L range (P<0.001).

In critically ill patients with sepsis, serum potassium levels exceeding 4.4 mmol/L are associated with an increased risk of death. Maintaining the average serum potassium level within the range of 3.5-4.4 mmol/L appears to be safe and may contribute to better outcomes in this patient population.

Background: Missing data in electronic health records (EHRs) presents significant challenges in medical studies. Many methods have been proposed, but uncertainty exists regarding the current state of missing data addressing methods applied for EHR and which strategy performs better within specific contexts. Methods: All studies referencing EHR and missing data methods published from their inception until 2024 March 30 were searched via the MEDLINE, EMBASE, and Digital Bibliography and Library Project databases. The characteristics of the included studies were extracted. We also compared the performance of various methods under different missingness scenarios. Results: After screening, 46 studies published between 2010 and 2024 were included. Three missingness mechanisms were simulated when evaluating the missing data methods: missing completely at random (29/46), missing at random (20/46), and missing not at random (21/46). Multiple imputation by chained equations (MICE) was the most popular statistical method, whereas generative adversarial network-based methods and the k nearest neighbor (KNN) classification were the common deep-learning-based or traditional machine-learning-based methods, respectively. Among the 26 articles comparing the performance among medical statistical and machine learning approaches, traditional machine learning or deep learning methods generally outperformed statistical methods. Med.KNN and context-aware time-series imputation performed better for longitudinal datasets, whereas probabilistic principal component analysis and MICE-based methods were optimal for cross-sectional datasets. Conclusions: Machine learning methods show significant promise for addressing missing data in EHRs. However, no single approach provides a universally generalizable solution. Standardized benchmarking analyses are essential to evaluate these methods across different missingness scenarios.

How to cite this article: Pande R, Pandey M. The Sepsis Score Dilemma: Balancing Precision and Utility. Indian J Crit Care Med 2024;28(10):906-907.

Sepsis impacts 1.7 million Americans annually. It is a life-threatening disruption of organ function because of the body's host response to infection. Sepsis remains a condition frequently encountered in emergency departments (ED) with an estimated 850,000 annual visits affected by sepsis each year in the United States. The pillars of managing sepsis remain timely identification, initiation of antimicrobials while aiming for source control and resuscitation with a goal of restoring tissue perfusion. The focus herein is current evidence and best practice recommendations for state-of-the-art sepsis care that begins in the ED.

This comprehensive review offers a detailed exposition of contemporary strategies in sepsis management, encompassing predictors, diagnostic tools, and therapeutic advances. The analysis elucidates the dynamic nature of sepsis, emphasizing the crucial role of early detection and intervention. The multifaceted strategies advocate for a holistic and personalized approach to sepsis care from traditional clinical methodologies to cutting-edge technologies. The implications for clinical practice underscore clinicians' need to adapt to evolving definitions, integrate advanced diagnostic tools, and embrace precision medicine. Integrating artificial intelligence and telemedicine necessitates a commitment to training and optimization. Judicious antibiotic use and recognition of global health disparities emphasize the importance of a collaborative, global effort in sepsis care. Looking ahead, recommendations for future research underscore priorities such as longitudinal studies on biomarkers, precision medicine trials, implementation science in technology, global health interventions, and innovative antibiotic stewardship strategies. These research priorities aim to contribute to transformative advancements in sepsis management, ultimately enhancing patient outcomes and reducing the global impact of this critical syndrome.

Sepsis is a global public health burden. The sequential organ failure assessment (SOFA) is the most commonly used scoring system for diagnosing sepsis and assessing severity. Due to the widespread use of endotracheal intubation and sedative medications in sepsis, the accuracy of the Glasgow Coma Score (GCS) is the lowest in SOFA. We designed this multicenter, cross-sectional study to investigate the predictive efficiency of SOFA with or without GCS on ICU mortality in patients with sepsis.

First, 3048 patients with sepsis admitted to Peking Union Medical College Hospital (PUMCH) were enrolled in this survey. The data were collected from June 8, 2013 to October 12, 2022. Second, 18,108 patients with sepsis in the eICU database were enrolled. Third, 2397 septic patients with respiratory system ≥ 3 points in SOFA in the eICU database were included. We investigated the predictive efficiency of SOFA with or without GCS on ICU mortality in patients with sepsis in various ICUs of PUMCH, and then we validated the results in the eICU database.

In data of ICUs in PUMCH, the predictive efficiency of SOFA without GCS (AUROC [95% CI], 24 h, 0.724 [0.688, 0.760], 48 h, 0.734 [0.699, 0.769], 72 h, 0.748 [0.713, 0.783], 168 h, 0.781 [0.747, 0.815]) was higher than that of SOFA with GCS (AUROC [95% CI], 24 h, 0.708 [0.672, 0.744], 48 h, 0.721 [0.685, 0.757], 72 h, 0.735 [0.700, 0.757], 168 h, 0.770 [0.736, 0.804]) on ICU mortality in patients with sepsis, and the difference was statistically significant (P value, 24 h, 0.001, 48 h, 0.003, 72 h, 0.004, 168 h, 0.005). In septic patients with respiratory system ≥ 3 points in SOFA in the eICU database, although the difference was not statistically significant (P value, 24 h, 0.148, 48 h, 0.178, 72 h, 0.132, 168 h, 0.790), SOFA without GCS (AUROC [95% CI], 24 h, 0.601 [0.576, 0.626], 48 h, 0.625 [0.601, 0.649], 72 h, 0.639 [0.615, 0.663], 168 h, 0.653 [0.629, 0.677]) had a higher predictive efficiency on ICU mortality than SOFA with GCS (AUROC [95% CI], 24 h, 0.591 [0.566, 0.616], 48 h, 0.616 [0.592, 0.640], 72 h, 0.628 [0.604, 0.652], 168 h, 0.651 [0.627, 0.675]).

In severe sepsis, it is realistic and feasible to discontinue the routine GCS for SOFA in patients with a respiratory system ≥ 3 points, and even better predict ICU mortality.

Perhaps nowhere else in the healthcare system than in the intensive care unit environment are the challenges to create useful models with direct time-critical clinical applications more relevant and the obstacles to achieving those goals more massive. Machine learning-based artificial intelligence (AI) techniques to define states and predict future events are commonplace activities of modern life. However, their penetration into acute care medicine has been slow, stuttering and uneven. Major obstacles to widespread effective application of AI approaches to the real-time care of the critically ill patient exist and need to be addressed.

Clinical decision support systems (CDSSs) in acute and critical care environments support clinicians, not replace them at the bedside. As will be discussed in this review, the reasons are many and include the immaturity of AI-based systems to have situational awareness, the fundamental bias in many large databases that do not reflect the target population of patient being treated making fairness an important issue to address and technical barriers to the timely access to valid data and its display in a fashion useful for clinical workflow. The inherent "black-box" nature of many predictive algorithms and CDSS makes trustworthiness and acceptance by the medical community difficult. Logistically, collating and curating in real-time multidimensional data streams of various sources needed to inform the algorithms and ultimately display relevant clinical decisions support format that adapt to individual patient responses and signatures represent the efferent limb of these systems and is often ignored during initial validation efforts. Similarly, legal and commercial barriers to the access to many existing clinical databases limit studies to address fairness and generalizability of predictive models and management tools.

AI-based CDSS are evolving and are here to stay. It is our obligation to be good shepherds of their use and further development.

Sepsis is one of the most serious hospital conditions associated with high mortality. Sepsis is the result of a dysregulated immune response to infection that can lead to multiple organ dysfunction and death. Due to the wide variability in the causes of sepsis, clinical presentation, and the recovery trajectories, identifying sepsis sub-phenotypes is crucial to advance our understanding of sepsis characterization, to choose targeted treatments and optimal timing of interventions, and to improve prognostication. Prior studies have described different sub-phenotypes of sepsis using organ-specific characteristics. These studies applied clustering algorithms to electronic health records (EHRs) to identify disease sub-phenotypes. However, prior approaches did not capture temporal information and made uncertain assumptions about the relationships among the sub-phenotypes for clustering procedures.

We developed a time-aware soft clustering algorithm guided by clinical variables to identify sepsis sub-phenotypes using data available in the EHR.

We identified six novel sepsis hybrid sub-phenotypes and evaluated them for medical plausibility. In addition, we built an early-warning sepsis prediction model using logistic regression.

Our results suggest that these novel sepsis hybrid sub-phenotypes are promising to provide more accurate information on sepsis-related organ dysfunction and sepsis recovery trajectories which can be important to inform management decisions and sepsis prognosis.

Most sepsis patients could potentially experience advantageous outcomes from targeted medical intervention, such as fluid resuscitation, antibiotic administration, respiratory support, and nursing care, promptly upon arrival at the emergency department (ED). Several scoring systems have been devised to predict hospital outcomes in sepsis patients, including the Sequential Organ Failure Assessment (SOFA) score. In contrast to prior research, our study introduces the novel approach of utilizing the National Early Warning Score 2 (NEWS2) as a means of assessing treatment efficacy and disease progression during an ED stay for sepsis.

To evaluate the sepsis prognosis and effectiveness of treatment administered during ED admission in reducing overall hospital mortality rates resulting from sepsis, as measured by the NEWS2.

The present investigation was conducted at a medical center from 1997 to 2020. The NEWS2 was calculated for patients with sepsis who were admitted to the ED in a consecutive manner. The computation was based on the initial and final parameters that were obtained during their stay in the ED. The alteration in the NEWS2 from the initial to the final measurements was utilized to evaluate the benefit of ED management to the hospital outcome of sepsis. Univariate and multivariate Cox regression analyses were performed, encompassing all clinically significant variables, to evaluate the adjusted hazard ratio (HR) for total hospital mortality in sepsis patients with reduced severity, measured by NEWS2 score difference, with a 95% confidence interval (adjusted HR with 95% CI). The study employed Kaplan-Meier analysis with a Log-rank test to assess variations in overall hospital mortality rates between two groups: the "improvement (reduced NEWS2)" and "non-improvement (no change or increased NEWS2)" groups.

The present investigation recruited a cohort of 11,011 individuals who experienced the first occurrence of sepsis as the primary diagnosis while hospitalized. The mean age of the improvement and non-improvement groups were 69.57 (± 16.19) and 68.82 (± 16.63) years, respectively. The mean SOFA score of the improvement and non-improvement groups were of no remarkable difference, 9.7 (± 3.39) and 9.8 (± 3.38) years, respectively. The total hospital mortality for sepsis was 42.92% (4,727/11,011). Following treatment by the prevailing guidelines at that time, a total of 5,598 out of 11,011 patients (50.88%) demonstrated improvement in the NEWS2, while the remaining 5,403 patients (49.12%) did not. The improvement group had a total hospital mortality rate of 38.51%, while the non-improvement group had a higher rate of 47.58%. The non-improvement group exhibited a lower prevalence of comorbidities such as congestive heart failure, cerebral vascular disease, and renal disease. The non-improvement group exhibited a lower Charlson comorbidity index score [4.73 (± 3.34)] compared to the improvement group [4.82 (± 3.38)] The group that underwent improvement exhibited a comparatively lower incidence of septic shock development in contrast to the non-improvement group (51.13% versus 54.34%, P < 0.001). The improvement group saw a total of 2,150 patients, which represents 38.41% of the overall sample size of 5,598, transition from the higher-risk to the medium-risk category. A total of 2,741 individuals, representing 48.96% of the sample size of 5,598 patients, exhibited a reduction in severity score only without risk category alteration. Out of the 5,403 patients (the non-improvement group) included in the study, 78.57% (4,245) demonstrated no alteration in the NEWS2. Conversely, 21.43% (1,158) of patients exhibited an escalation in severity score. The Cox regression analysis demonstrated that the implementation of interventions aimed at reducing the NEWS2 during a patient's stay in the ED had a significant positive impact on the outcome, as evidenced by the adjusted HRs of 0.889 (95% CI = 0.808, 0.978) and 0.891 (95% CI = 0.810, 0.981), respectively. The results obtained from the Kaplan-Meier analysis indicated that the survival rate of the improvement group was significantly higher than that of the non-improvement group (P < 0.001) in the hospitalization period.

The present study demonstrated that 50.88% of sepsis patients obtained improvement in ED, ascertained by means of the NEWS2 scoring system. The practical dynamics of NEWS2 could be utilized to depict such intricacies clearly. The findings also literally supported the importance of ED management in the comprehensive course of sepsis treatment in reducing the total hospital mortality rate.

Sepsis is a multi-organ dysfunction characterized by an unregulated host response to infection. It is associated with high morbidity, rapid disease progression, and high mortality. Current therapies mainly focus on symptomatic treatment, such as blood volume supplementation and antibiotic use, but their effectiveness is limited. Th17/Treg balance, based on its inflammatory property, plays a crucial role in determining the direction of the inflammatory response and the regression of organ damage in sepsis patients. This review provides a summary of the changes in T-helper (Th) 17 cell and regulatory T (Treg) cell differentiation and function during sepsis, the heterogeneity of Th17/Treg balance in the inflammatory response, and the relationship between Th17/Treg balance and organ damage. Th17/Treg balance exerts significant control over the bloom and wanes in host inflammatory response throughout sepsis.