Background/Objectives: Sepsis is a severe medical condition characterized by a dysregulated host response to infection, with potentially fatal outcomes, requiring early diagnosis and rapid intervention. The limitations of traditional sepsis identification methods, as well as the complexity of clinical data generated in intensive care, have driven increased interest in applying artificial intelligence in this field. The aim of this narrative review article is to analyze how artificial intelligence is being used in the diagnosis and prognosis of sepsis, to present the most relevant current models and algorithms, and to discuss the challenges and opportunities related to integrating these technologies into clinical practice. Methods: We conducted a structured literature search for this narrative review, covering studies published between 2016 and 2024 in databases such as PubMed/Medline, Scopus, Web of Science, IEEE Xplore, and Google Scholar. The review covered models based on machine learning (ML), deep neural networks (DNNs), Recurrent Neural Networks (RNNs), and clinical alert systems implemented in hospitals. The clinical data sources used, algorithms applied, system architectures, and performance outcomes are presented. Results: Numerous artificial intelligence models demonstrated superior performance compared to conventional clinical scores (qSOFA, SIRS), achieving AUC values above 0.90 in predicting sepsis and mortality. Systems such as Targeted Real-Time Early Warning System (TREWS) and InSight have been clinically validated and have significantly reduced the time to treatment initiation. However, challenges remain, such as a lack of model transparency, algorithmic bias, difficulties integrating into clinical workflows, and the absence of external validation in multicenter settings. Conclusions: Artificial intelligence has the potential to transform sepsis management through early diagnosis, risk stratification, and personalized treatment. A responsible, multidisciplinary approach is necessary, including rigorous clinical validation, enhanced interpretability, and training of healthcare personnel to effectively integrate these technologies into everyday practice.

Viral infections caused by exposure to viruses such as Epstein–Barr, cytomegalovirus, or Parvovirus B19 have always been considered predisposing environmental factors for the onset of autoimmune diseases. More recently, autoimmune mechanisms such as molecular mimicry, T-cell activation, transient immunosuppression and inflammation have also been observed in cases of SARS-CoV-2 infection. Several newly diagnosed autoimmune disorders have been reported post-COVID-19, such as COVID-19-associated multisystemic inflammatory syndrome in children (MIS-C), type 1 diabetes mellitus, systemic lupus erythematosus, or rheumatoid arthritis. In this article, we present a new case of paediatric systemic lupus erythematosus (SLE) with haematological (macrophage activation syndrome), renal (stage 2), cutaneous (urticarial vasculitis) and digestive involvement, onset three and a half months post-COVID-19. In the dynamics, de novo infection generated by Epstein–Barr exposure was associated. The diagnosis was confirmed based on EULAR/ACR 2019 criteria. The aim of the article is to present a possible correlation between SARS-CoV-2 and Epstein–Barr as extrinsic factors in triggering or activating paediatric systemic lupus erythematosus. Keywords: paediatric systemic lupus erythematosus; post-COVID-19; Epstein–Barr; SARS- CoV-2; case report; paediatric patient.

Pediatric patients with polyuria polydipsia syndrome (PPS) represent a diagnostic challenge for clinicians because of the technical difficulties in performing the gold standard water deprivation test (WDT). Copeptin, a stable biomarker representing the C-terminal portion of the polypeptide chain of the antidiuretic hormone, is a reliable diagnostic tool. To assess the diagnostic accuracy of baseline copeptin dosing, arginine/hypertonic saline copeptin stimulation tests, and WDT. This study aimed to establish the diagnostic utility of copeptin in pediatric patients by distinguishing between central diabetes insipidus, nephrogenic diabetes insipidus, and primary polydipsia. Comparative and non-comparative primary studies published between January 2018 and August 2024 focusing on children were searched and included in PubMed, Cochrane Library, Web of Science, ScienceDirect, Scopus, and Google Scholar. The QUADAS-2 tool was used to assess the risk of bias and applicability. Meta-analyses used fixed effects models because of low heterogeneity and the HSROC model. Eleven studies were included with an overall low bias and no significant applicability concerns. The mean pooled sensitivity = 0.98 (95% CI: 0.936–1.025), pooled specificity = 0.947 (95% CI: 0.920–0.973), and AUC = 0.972 (95% CI: 0.952–0.992), indicating excellent diagnostic accuracy. Stimulation methods for copeptin dosing represent an effective and less invasive diagnostic test for children with PPS, and future development of standard copeptin testing protocols is needed.

COVID-19-associated multisystem inflammatory syndrome in children (MIS-C) is a rare autoimmune disorder characterized by a range of polymorphic manifestations, similar to but distinct from other well-known inflammatory syndromes in children. We conducted a retrospective–descriptive study in which we summarized the clinical presentation of, biomarker variations in, and complications occurring in patients diagnosed with MIS-C, admitted to the Emergency Clinical Hospital for Children “Sf. Ioan”, Galati, between July 2020 and June 2024. A total of 36 children met the MIS-C classification criteria according to the WHO-approved case definitions. A total of 41.7% (n = 15) were male and 58.3% (n = 21) were female. The median age of the study group was 4 years (IQR: 1.75–9.25 years). Surgical involvement was suspected in 16.7% (n = 6) of the patients, while 52.8% (n = 19) required intensive care. Clinically, fever was the most common symptom present in 89% (n = 32) of the cases. Gastrointestinal disorders were also common, with 50% (n = 18) presenting with inappetence, 42% (n = 15) with vomiting, and 39% (n = 14) with abdominal pain from admission, which worsened over time. Paraclinically, all patients exhibited signs of inflammation, and 86.1% (n = 31) had hydroelectrolytic and acid–base imbalances. The median hospital stay was 10 days (IQR: 7–12 days), with a stagnant clinical course in most cases. The inflammatory mechanisms in MIS-C, which can affect the secretion of antidiuretic hormone (ADH), were correlated with hydroelectrolytic disturbances and may lead to severe complications. For this reason, it is imperative to evaluate hydroelectrolytic disorders in the context of MIS-C and use diagnostic and prognostic biomarkers to develop effective therapeutic management strategies, ultimately improving the quality of life of affected children.

The COVID-19 pandemic has introduced challenges in pediatric care, especially due to the emergence of Multisystem Inflammatory Syndrome in Children (MIS-C), a severe condition associated with SARS-CoV-2 infection. This study investigated the impact of hyponatremia and antidiuretic hormone (ADH) secretion corelated to clinical outcomes in these patients. We conducted a retrospective cohort study, including 118 pediatric patients, with a detailed sub-cohort analysis of 53 patients for ADH secretion markers. Hyponatremia, defined by age-specific sodium thresholds, was present in 47.22% of MIS-C cases and 28.04% of COVID-19 cases. Ordinal logistic regression analysis revealed that severe hyponatremia significantly increased the likelihood of more severe clinical outcomes (β = 3.514, p < 0.001). A significant correlation was found between hyponatremia and prolonged hospitalization. For ADH secretion, a predictive model using ridge regression was analysed, which demonstrated that serum sodium level, U/P ratio, and hospitalization duration are key predictors of SIADH. This model fit was assessed using the ROC curve with an AUC of 0.96, indicating reliable model performance. Our findings underscore the significant role of hyponatremia on the clinical severity and hospitalization outcome of COVID-19 and MIS-C in pediatric patients.

Differentiating central diabetes insipidus (CDI), nephrogenic diabetes insipidus (NDI), and primary polydipsia (PP) in pediatric patients with polyuria–polydipsia syndrome (PPS) remains a clinical challenge. The water deprivation test (WDT) is the traditional gold standard; however, it is time-consuming, burdensome, and prone to equivocal results. Stimulated copeptin, a surrogate marker of vasopressin, has emerged as a promising diagnostic alternative. We conducted a prospective, observational, cross-sectional study involving 27 pediatric patients (ages 2–17) presenting with PPS. Each patient underwent a WDT with desmopressin and hypertonic saline infusion (3% NaCl) for stimulated copeptin testing. Diagnostic accuracy was assessed using clinical diagnoses as a reference. The WDT showed high accuracy with an area under the curve (AUC) of 0.97, and there was an increased optimal threshold of ≥14% urine osmolality after desmopressin acetate (1-deamino-8-D-arginine vasopressin, DDAVP) administration (sensitivity 88.9%, specificity 100%). Stimulated copeptin at a threshold of <6.5 pmol/L demonstrated 100% sensitivity and specificity (AUC = 1.00) for CDI versus PP. Basal copeptin ≥21.4 pmol/L accurately identified all NDI cases. The agreement between the WDT and copeptin was low (κ = 0.06, McNemar p = 0.021), suggesting that copeptin has greater specificity, particularly for borderline or partial CDI. These results support the use of stimulated copeptin as a first-line diagnostic tool in pediatric PPS, offering improved objectivity, tolerability, and diagnostic clarity compared with the WDT. Basal copeptin also demonstrated excellent performance in rapid noninvasive NDI identification.