Background: Adults hospitalized with COVID-19, influenza A/B, or respiratory syncytial virus (RSV) frequently receive empirical antibiotics, but antibiotic prescribing, confirmed bacterial co-infection, antimicrobial resistance (AMR), and outcomes have not been jointly synthesized across these infections. Methods: We conducted a PRISMA 2020 systematic review and meta-analysis of 39 studies including 839,531 hospitalized adults. Random-effects models with Freeman–Tukey double-arcsine transformation pooled prevalence estimates; sensitivity and publication-bias analyses were performed where appropriate. Results: Pooled antibiotic use was 62.56% (95% CI, 53.75–70.97%) for COVID-19, 57.48% (25.76–86.09%) for influenza A/B, and 76.03% (67.62–83.53%) for RSV, with very high heterogeneity. Confirmed bacterial co-infection was lower: 5.31% (3.43–7.56%), 18.66% (9.98–29.30%), and 24.36% (18.53–30.70%), respectively. Prescribing-to-confirmed infection ratios ranged from 3.0 to 46.2. AMR evidence was restricted to COVID-19 studies and was dominated by carbapenem-resistant Gram-negative organisms, mainly in secondary, ICU-associated, or healthcare-associated infections. Confirmed bacterial complications were associated with ICU admission, longer hospitalization, and higher mortality. Conclusions: Antibiotic prescribing exceeded confirmed bacterial infection across all viral groups, but estimates require cautious interpretation due to heterogeneity, diagnostic uncertainty, observational evidence, and the absence of low-risk-of-bias studies. The evidence base was dominated by COVID-19 cohorts, while influenza A/B and RSV data, especially virus-specific AMR evidence, remain limited. COVID-19-specific AMR findings should not be generalized to influenza A/B or RSV. Virus-specific stewardship should prioritize rapid diagnostics, systematic sampling, reassessment, and de-escalation when bacterial infection is not confirmed.

Background/Objectives: Accurate staging of mediastinal lymph nodes is essential for therapeutic decisions and prognostic assessment in non-small cell lung cancer (NSCLC). This systematic review evaluates diagnostic performance, validation strategies, and clinical significance of artificial intelligence (AI)-based 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET/CT) radiomics models for mediastinal nodal staging in NSCLC. Methods: Systematic literature searching was conducted in PubMed, ScienceDirect, and Scopus according to the PRISMA 2020 guidelines. Eligible studies used radiomic or AI-based approaches for mediastinal lymph node (LN) evaluation in NSCLC, with histopathology as a reference standard. Extracted data included study design, cohort characteristics, imaging method, validation strategy, and diagnostic performance metrics. Methodological quality was assessed by the QUADAS-2 tool. Results: Thirteen studies were included which are mainly retrospective in their designs with cohort sizes varying between 87 and 3265 patients. Models evaluated on external or prospective validation cohorts generally showed lower performance compared with training or internal datasets. However, clinically significant discriminative ability has been preserved across heterogeneous populations. In studies that directly compared methods, composite models integrating radiomic features with clinical factors and conventional PET metrics, sometimes including deep learning-derived features, consistently outperformed radiomics-only models. Additionally, selected approaches addressing FDG-related false-positive uptake improved distinction between benign and metastatic mediastinal lymph nodes; this is reflected by reduced false-positive classifications plus higher specificity compared with conventional PET/CT interpretation. Conclusions: AI-based 18F-FDG PET/CT radiomics show a promising discriminative capacity for mediastinal nodal staging in NSCLC, especially when it is integrated with clinical and conventional imaging variables. Although the model performance remains clinically significant within independent validation cohorts, attenuation compared with training datasets is commonly observed. Methodological heterogeneity, predominantly retrospective study designs, and the scarcity of prospective multicenter validation currently limit routine clinical implementation.

Background/Objectives: CT-guided transthoracic core needle biopsy (CT-TTNB) is the standard technique for histological characterization of pulmonary lesions, yet it carries a 15–42% rate of pneumothorax and a 5–27% rate of alveolar hemorrhage. Accurate identification of modifiable and non-modifiable procedural determinants is essential for pre-procedural risk stratification, technique optimization, and post-procedural care. Methods: We conducted a single-centre retrospective cohort study of 240 consecutive CT-TTNB procedures performed between November 2023 and January 2025. The variables were extracted from medical records and PACS and entered into univariate and multivariate analysis models. Model performance was assessed by AUC-ROC, Nagelkerke R², and the Hosmer–Lemeshow test. Results: Pneumothorax occurred in 79 patients (32.9%), with chest tube drainage required in 14 (5.8%). Alveolar hemorrhage was identified in 49 patients (20.4%). Four independent predictors for pneumothorax were identified: needle–pleural angle (OR 0.837/1°; p < 0.001), pleura-to-lesion distance (OR 1.675/10 mm; p < 0.001), fissure traversal (OR 26.718; p = 0.003), and patient age (OR 1.076/year; p = 0.002); AUC-ROC 0.927. The alveolar hemorrhage model identified pleura-to-lesion distance (OR 1.768/10 mm; p < 0.001), RUL apical segment (OR 4.281; p < 0.001), and carcinomatous lymphangitis as protective factors (OR 0.359; p = 0.030); AUC-ROC 0.822. The needle–pleural angle was not independently associated with alveolar hemorrhage, confirming mechanistically distinct complication pathways. Conclusions: Needle–pleural angle is the dominant modifiable determinant of pneumothorax and pneumothorax-related complications. Optimizing the trajectory towards perpendicularity does not increase hemorrhage risk. These findings support the implementation of angle-based trajectory planning protocols and risk-stratified post-procedural monitoring.