Dissecting contributions of individual systemic inflammatory response syndrome criteria from a prospective algorithm to the prediction and diagnosis of sepsis in a polytrauma cohort

A nomogram for predicting short-term mortality in ICU patients with coexisting chronic obstructive pulmonary disease and congestive heart failure

OBJECTIVE

This study aimed to develop and validate a nomogram for predicting 28-day and 90-day mortality in intensive care unit (ICU) patients who have chronic obstructive pulmonary disease (COPD) coexisting with congestive heart failure (CHF).

METHODS

An extensive analysis was conducted on clinical data from the Medical Information Mart for Intensive Care IV database, covering patients over 18 years old with both COPD and CHF, who were were first-time ICU admissions between 2008 and 2019. The least absolute shrinkage and selection operator (LASSO) regression method was employed to screen clinical features, with the final model being optimized using backward stepwise regression guided by the Akaike Information Criterion (AIC) to construct the nomogram. The predictive model's discrimination and clinical applicability were evaluated via receiver operating characteristic (ROC) curves, calibration curves, the C-index, and decision curve analysi s (DCA).

RESULTS

This analysis was comprised of a total of 1948 patients. Patients were separated into developing and validation cohorts in a 7:3 ratio, with similar baseline characteristics between the two groups. The ICU mortality rates for the developing and verification cohorts were 20.8 % and 19.5 % at 28 days, respectively, and 29.4 % and 28.3 % at 90 days, respectively. The clinical characteristics retained by the backward stepwise regression include age, weight, systolic blood pressure (SBP), respiratory rate (RR), oxygen saturation (SpO2), red blood cell distribution width (RDW), lactate, partial thrombosis time (PTT), race, marital status, type 2 diabetes mellitus (T2DM), malignant cancer, acute kidney failure (AKF), pneumonia, immunosuppressive drugs, antiplatelet agents, vasoactive agents, acute physiology score III (APS III), Oxford acute severity of illness score (OASIS), and Charlson comorbidity index (CCI). We developed two separate models by assigning weighted scores to each independent risk factor: nomogram A excludes CCI but includes age, T2DM, and malignant cancer, while nomogram B includes only CCI, without age, T2DM, and malignant cancer. Based on the results of the AUC and C-index, this study selected nomogram A, which demonstrated better predictive performance, for subsequent validation. The calibration curve, C-index, and DCA results indicate that nomogram A has good accuracy in predicting short-term mortality and demonstrates better discriminative ability than commonly used clinical scoring systems, making it more suitable for clinical application.

CONCLUSION

The nomogram developed in this study offers an effective assessment of short-term mortality risk for ICU patients with COPD and CHF, proving to be a superior tool for predicting their short-term prognosis.

Development and validation of a machine learning-based model to assess probability of systemic inflammatory response syndrome in patients with severe multiple traumas

BACKGROUND

Systemic inflammatory response syndrome (SIRS) is a predictor of serious infectious complications, organ failure, and death in patients with severe polytrauma and is one of the reasons for delaying early total surgical treatment. To determine the risk of SIRS within 24 h after hospitalization, we developed six machine learning models.

MATERIALS AND METHODS

Using retrospective data about the patient, the nature of the injury, the results of general and standard biochemical blood tests, and coagulation tests, six models were developed: decision tree, random forest, logistic regression, support vector and gradient boosting classifiers, logistic regressor, and neural network. The effectiveness of the models was assessed through internal and external validation.

RESULTS

Among the 439 selected patients with severe polytrauma in 230 (52.4%), SIRS was diagnosed within the first 24 h of hospitalization. The SIRS group was more strongly associated with class II bleeding (39.5% vs. 60.5%; OR 1.81 [95% CI: 1.23-2.65]; P = 0.0023), long-term vasopressor use (68.4% vs. 31.6%; OR 5.51 [95% CI: 2.37-5.23]; P < 0.0001), risk of acute coagulopathy (67.8% vs. 32.2%; OR 2.4 [95% CI: 1.55-3.77]; P < 0.0001), and greater risk of pneumonia (59.5% vs. 40.5%; OR 1.74 [95% CI: 1.19-2.54]; P = 0.0042), longer ICU length of stay (5 ± 6.3 vs. 2.7 ± 4.3 days; P < 0.0001) and mortality rate (64.5% vs. 35.5%; OR 10.87 [95% CI: 6.3-19.89]; P = 0.0391). Of all the models, the random forest classifier showed the best predictive ability in the internal (AUROC 0.89; 95% CI: 0.83-0.96) and external validation (AUROC 0.83; 95% CI: 0.75-0.91) datasets.

CONCLUSIONS

The developed model made it possible to accurately predict the risk of developing SIRS in the early period after injury, allowing clinical specialists to predict patient management tactics and calculate medication and staffing needs for the patient.

LEVEL OF EVIDENCE

Level 3.

TRIAL REGISTRATION

The study was retrospectively registered in the ClinicalTrials.gov database of the National Library of Medicine (NCT06323096).

Identifying biomarkers deciphering sepsis from trauma-induced sterile inflammation and trauma-induced sepsis

Objective

The purpose of this study was to identify a panel of biomarkers for distinguishing early stage sepsis patients from non-infected trauma patients.

Background

Accurate differentiation between trauma-induced sterile inflammation and real infective sepsis poses a complex life-threatening medical challenge because of their common symptoms albeit diverging clinical implications, namely different therapies. The timely and accurate identification of sepsis in trauma patients is therefore vital to ensure prompt and tailored medical interventions (provision of adequate antimicrobial agents and if possible eradication of infective foci) that can ultimately lead to improved therapeutic management and patient outcome. The adequate withholding of antimicrobials in trauma patients without sepsis is also important in aspects of both patient and environmental perspective.

Methods

In this proof-of-concept study, we employed advanced technologies, including Matrix-Assisted Laser Desorption/Ionization (MALDI) and multiplex antibody arrays (MAA) to identify a panel of biomarkers distinguishing actual sepsis from trauma-induced sterile inflammation.

Results

By comparing patient groups (controls, infected and non-infected trauma and septic shock patients under mechanical ventilation) at different time points, we uncovered distinct protein patterns associated with early trauma-induced sterile inflammation on the one hand and sepsis on the other hand. SYT13 and IL1F10 emerged as potential early sepsis biomarkers, while reduced levels of A2M were indicative of both trauma-induced inflammation and sepsis conditions. Additionally, higher levels of TREM1 were associated at a later stage in trauma patients. Furthermore, enrichment analyses revealed differences in the inflammatory response between trauma-induced inflammation and sepsis, with proteins related to complement and coagulation cascades being elevated whereas proteins relevant to focal adhesion were diminished in sepsis.

Conclusions

Our findings, therefore, suggest that a combination of biomarkers is needed for the development of novel diagnostic approaches deciphering trauma-induced sterile inflammation from actual infective sepsis.