Cox-LASSO Analysis for Hospital Mortality in Patients With Sepsis Received Continuous Renal Replacement Therapy: A MIMIC-III Database Study

Predictive models of sepsis-associated acute kidney injury based on machine learning: a scoping review

BACKGROUND

With the development of artificial intelligence, the application of machine learning to develop predictive models for sepsis-associated acute kidney injury has made potential breakthroughs in early identification, grading, diagnosis, and prognosis determination.

METHODS

Here, we conducted a systematic search of the PubMed, Cochrane Library, Embase (Ovid), Web of Science, and Scopus databases on April 28, 2023, and screened relevant literature. Then, we comprehensively extracted relevant data related to machine learning algorithms, predictors, and predicted objectives. We subsequently performed a critical evaluation of research quality, data aggregation, and analyses.

RESULTS

We screened 25 studies on predictive models for sepsis-associated acute kidney injury from a total of originally identified 2898 studies. The most commonly used machine learning algorithm is traditional logistic regression, followed by eXtreme gradient boosting. We categorized these predictive models into early identification models (60%), prognostic prediction models (32%), and subtype identification models (8%) according to their predictive purpose. The five most commonly used predictors were serum creatinine levels, lactate levels, age, blood urea nitrogen concentration, and diabetes mellitus. In addition, a single data source, insufficient assessment of clinical utility, lack of model bias assessment, and hyperparameter adjustment may be the main reasons for the low quality of the current research.

CONCLUSIONS

However, studies on the nondeath prognostic outcomes, the long-term clinical outcomes, and the subtype identification models are insufficient. Additionally, the poor quality of the research and the insufficient practicality of the model are problems that need to be addressed urgently.

Humoral and cellular immune response after severe acute respiratory syndrome coronavirus 2 messenger ribonucleic acid vaccination in heart transplant recipients: An observational study in France

INTRODUCTION

Heart transplant (HT) recipients have a high risk of developing severe COVID-19. Immunoglobulin G antibodies are considered to provide protective immunity and T-cell activity is thought to confer protection from severe disease. However, data on T-cell response to mRNA vaccination in a context of HT remains limited.

METHODS

In 96 HT patients, a IFN-γ release assay and an anti-Spike antibody test were used to evaluate the ability of SARS-CoV-2 mRNA vaccines to generate cellular and humoral immune response. Blood samples were collected few weeks to 7 months after vaccination. Multiple fractional polynomial and LASSO regression models were used to define predictors of T-cell response.

RESULTS

Three to five months after vaccination, three doses of vaccine induced a positive SARS-CoV-2 T-cell response in 47% of recipients and a positive humoral response in 83% of recipients, 11.1% of patients remained negative for both T and B cell responses. Three doses were necessary to reach high IgG response levels (>590 BAU/mL), which were obtained in a third of patients. Immunity was greatly amplified in the group who had three vaccine doses plus COVID-19 infection.

CONCLUSION

Our study revealed that T and B immunity decreases over time, leading us to suggest the interest of a booster vaccination at 5 months after the third dose. Moreover, a close follow-up of immune response following vaccination is needed to ensure ongoing immune protection. We also found that significant predictors of higher cellular response were infection and active smoking, regardless of immunosuppressive treatment with mycophenolate mofetil (MMF).