The role of platelet to mean platelet volume ratio in the identification of adult-onset still's disease from sepsis

Identification of discriminatory factors and construction of nomograms for differentiating AOSD and sepsis

OBJECTIVE

This study aimed to develop nomogram prediction models to differentiate between adult-onset Still's disease (AOSD) and sepsis.

METHODS

We retrospectively collected laboratory test data from 107 hospitalized patients with AOSD and sepsis at the Affiliated Hospital of Xuzhou Medical University. Multivariate binary logistic regression was used to develop nomogram models using arthralgia, WBC, APTT, creatinine, PLT, and ferritin as independent factors. The performance of the model was evaluated by the bootstrap consistency index and calibration curve.

RESULTS

Model 1 had an AUC of 0.98 (95% CI, 0.96-1.00), specificity of 0.98, and sensitivity of 0.94. Model 2 had an AUC of 0.96 (95% CI, 0.93-1.00), specificity of 0.92, and sensitivity of 0.94. The fivefold cross-validation yielded an accuracy (ACC) of 0.92 and a kappa coefficient of 0.83 for Model 1, while for Model 2, the ACC was 0.87 and the kappa coefficient was 0.74.

CONCLUSION

The nomogram models developed in this study are useful tools for differentiating between AOSD and sepsis. Key Points • The differential diagnosis between AOSD and sepsis has always been a challenge • Delayed treatment of AOSD may lead to serious complications • We developed two nomogram models to distinguish AOSD from sepsis, which were not previously reported • Our models can be used to guide clinical practice with good discrimination.

Establishment of a differential diagnosis method and an online prediction platform for AOSD and sepsis based on gradient boosting decision trees algorithm

OBJECTIVE

The differential diagnosis between adult-onset Still's disease (AOSD) and sepsis has always been a challenge. In this study, a machine learning model for differential diagnosis of AOSD and sepsis was developed and an online platform was developed to facilitate the clinical application of the model.

METHODS

All data were collected from 42 AOSD patients and 50 sepsis patients admitted to Affiliated Hospital of Xuzhou Medical University from December 2018 to December 2021. In addition, 5 AOSD patients and 10 sepsis patients diagnosed in our hospital after March 2022 were collected for external validation. All models were built using the scikit-learn library (version 1.0.2) in Python (version 3.9.7), and feature selection was performed using the SHAP (Shapley Additive exPlanation) package developed in Python.

RESULTS

The results showed that the gradient boosting decision tree(GBDT) optimization model based on arthralgia, ferritin × lymphocyte count, white blood cell count, ferritin × platelet count, and α1-acid glycoprotein/creatine kinase could well identify AOSD and sepsis. The training set interaction test (AUC: 0.9916, ACC: 0.9457, Sens: 0.9556, Spec: 0.9578) and the external validation also achieved satisfactory results (AUC: 0.9800, ACC: 0.9333, Sens: 0.8000, Spec: 1.000). We named this discrimination method AIADSS (AI-assisted discrimination of Still's disease and Sepsis) and created an online service platform for practical operation, the website is http://cppdd.cn/STILL1/ .

CONCLUSION

We created a method for the identification of AOSD and sepsis based on machine learning. This method can provide a reference for clinicians to formulate the next diagnosis and treatment plan.

Platelet‑related parameters as potential biomarkers for the prognosis of sepsis

Early diagnosis and accurate prognosis are key for reducing the fatality rate and medical expenses associated with sepsis. Platelets are involved in the delayed tissue injury that occurs during sepsis. Therefore, the aim of the present study was to investigate the usefulness of platelets and associated parameters as prognostic markers of sepsis. The present study collected patient samples based on The Third International Consensus Definitions for Sepsis and Septic Shock criteria. Platelet-associated parameters were detected by flow cytometry and their correlation with clinical scores and prognoses was analyzed. Considering the association between endothelial cells and platelet activation, levels of plasma tumor necrosis factor-like weak inducer of apoptosis (TWEAK) and angiopoietin-2 (Ang-2) were analyzed by ELISA. The results showed significant differences in platelet P-selectin expression and phosphatidylserine exposure, mitochondrial membrane potential (Mmp)-Index values and plasma levels of TWEAK and Ang-2 between patients and healthy controls (P<0.05). Except for P-selectin and TWEAK levels, all parameters were correlated with clinical scores (acute physiology and chronic health evaluation II and sequential/sepsis-related organ failure assessment). Additionally, platelet Mmp-Index between admission and the end of therapy was only different in non-survivors (P<0.001) and platelet phosphatidylserine exposure was significantly lower in survivors (P=0.006). Therefore, of the parameters tested, the dynamic monitoring of phosphatidylserine exposure, platelet Mmp-Index values and plasma Ang-2 levels had the most potential for the assessment of disease severity and clinical outcomes.