INTRODUCTION:Nosocomial infections (NIs) in cirrhosis are associated with high mortality but could be preventable. Logistic regression (LR) models have failed to identify high-risk patients. We aimed to develop machine learning (ML) models to predict NI. METHODS:The CLEARED consortium consists of prospectively enrolled cirrhosis inpatients from >120 centers. Using day-of-admission clinical data, 3 ML approaches (random forest [RF], extreme gradient boosting, and neural networks [NNs]) were used to predict NI. Data were split 80:20 for training and testing stratified by the outcome. Models were compared using area under the receiver operating characteristic curve (AUC). RESULTS:In total, 8,263 patients (55.90 ± 13.34 years; 64.1% men) from 127 centers in 37 countries were included. NI developed in 869 (10.5%), a median of 6 (4-11) days of postadmission. Major NIs were respiratory (29.6%) and urinary tract infection (15.7%), and spontaneous bacterial peritonitis (13.5%). NIs occurred more frequently in patients from low/low-middle income countries and those with severe liver disease, alcohol etiology, and admission infections. NIs were associated with inpatient mortality (31.9% vs 8.1%, P < 0.001) and liver transplantation (4.8% vs 1.9%, P < 0.001). Although the RF model (AUC 0.69) showed good calibration (Brier score 0.09), outperforming extreme gradient boosting, neural network, and LR models (AUC 0.66 for all; LR comparison P = 0.043), no model achieved AUC ≥0.80 for clinical utility. At 10% predicted probability threshold, the RF model demonstrated only 75.4% sensitivity, 52.9% specificity, and 15.9% positive predictive value (PPV). DISCUSSION:NIs cannot be accurately predicted from day-of-admission data using ML models, even in a large, prospective, global cirrhosis cohort. Every hospitalized patient with cirrhosis should receive protocolized infection control measures.
Introduction and Objectives Delayed diagnostic paracentesis has been associated with worse outcomes in patients with decompensated cirrhosis and suspected spontaneous bacterial peritonitis (SBP). Data from Latin American populations remain limited.To evaluate the association between early paracentesis (24 hours from hospital admission) and in-hospital mortality in patients with decompensated cirrhosis and suspected SBP. Materials and Methods Retrospective observational study including hospitalized adults with decompensated cirrhosis and SBP. Patients were classified according to the timing of paracentesis: early (24 hours) or delayed (>24 hours). Clinical, laboratory, and survival outcomes were compared between groups. Cox regression analysis was performed to evaluate the association between early paracentesis and in-hospital mortality. Results A total of 102 patients were included. Mean age was 54.7±13.6 years, and 63.7% were men. Early paracentesis was performed in 33.3% of patients, whereas 66.7% underwent delayed paracentesis. In-hospital mortality was significantly lower in the early paracentesis group compared with the delayed group (38.2% vs 61.8%, p=0.021). Six-month survival was also higher in patients undergoing early paracentesis (58.8% vs 38.8%, p=0.045). Patients with delayed paracentesis had significantly higher bilirubin levels (14.02 vs 4.52 mg/dL, p=0.007), AST (104.5 vs 59.5 U/L, p=0.001), and ALT levels (52.5 vs 31.5 U/L, p=0.004). On Cox regression analysis, early paracentesis was independently associated with lower in-hospital mortality (HR 0.383, 95% CI 0.164–0.894, p=0.026). Conclusions Early paracentesis within the first 24 hours of admission was independently associated with lower in-hospital mortality in patients with decompensated cirrhosis and suspected SBP. Timely diagnostic paracentesis may represent an important quality-of-care measure.
Hyponatremia remains the most common electrolyte disorder in clinical practice, with a wide range of causes ranging from endocrine dysregulation to iatrogenic causes. Recent developments in point-of-care ultrasonography, vasopressin physiology, and osmoregulation have transformed diagnostics by emphasizing accuracy above purely clinical judgment. Although traditional management based on fluid restriction and hypertonic solutions remain the cornerstone of treatment, a careful and cautious monitoring is required to avoid osmotic demyelination and overcorrection. The Furst equation, a validated predictor of response to fluid restriction, should be integrated into routine practice, together with objective tools such as point-of-care ultrasonography, to improve the accuracy of volume assessment and guide more precise therapeutic decisions. Likewise, different therapeutic options have been studied for non-responders to fluid restriction, including urea, desmopressin, and selective vasopressin receptor antagonists (vaptans), showing promising results. Ongoing developments in artificial intelligence and laboratory-based decision support tools are expected to improve personalized management and predictive accuracy, potentially reconciling the disparity between evidence and clinical practice.
Bullous pemphigoid (BP) is the most common autoimmune subepidermal blistering disease and predominantly affects older adults. In recent years, dipeptidyl peptidase-4 inhibitors (DPP-4i), widely prescribed for type 2 diabetes mellitus, have emerged as important pharmacologic triggers of BP. DPP-4 inhibitors as a class have been associated with BP, with the strongest evidence reported for vildagliptin, although cases involving linagliptin have also been documented. Linagliptin has also been increasingly recognized as a potential trigger of drug-associated disease. We report the case of an 81-year-old woman with long-standing type 2 diabetes mellitus treated with metformin and linagliptin who developed a generalized pruritic vesiculobullous eruption. Dermatologic examination demonstrated multiple tense bullae, erosions, crusted plaques, and post-inflammatory hyperpigmented lesions involving the trunk, extremities, and intertriginous regions. Histopathologic examination revealed a subepidermal blister with prominent eosinophilic infiltration. Direct immunofluorescence demonstrated linear C3 and IgG deposition along the basement membrane zone, while indirect immunofluorescence localized immunoreactants to the roof of the split, confirming the diagnosis of BP. The temporal association with linagliptin exposure and the absence of alternative triggers supported the diagnosis of DPP-4 inhibitor-associated BP. Linagliptin was discontinued, and treatment with prednisone was initiated, resulting in progressive improvement and complete cessation of new blister formation at follow-up. This case highlights the importance of recognizing medication-induced BP in older diabetic patients and reviews current evidence regarding the epidemiology, pathogenesis, clinical presentation, diagnosis, and management of DPP-4 inhibitor-associated BP.
Purpose: To develop and validate a machine learning–based prognostic model using real-world data from a tertiary care center in Mexico to assess the impact of diagnostic delay and identify clinical, histopathological, and biochemical predictors of disease progression and mortality in testicular cancer, and to determine whether a Random Forest model based on clinical stage and metastatic status can accurately predict these adverse outcomes in a low- and middle-income setting. Methods: We retrospectively analyzed 223 testicular cancer cases, defining diagnostic delay as symptoms > 6 months. Predictors of progression and mortality were identified by multivariate analysis, and a Random Forest model based on clinical stage and metastasis status was trained (80:20 split) to predict adverse outcomes and assess feature importance using Python. Results: The Random Forest model achieved 83.3% accuracy for predicting progression and mortality, with clinical stage (79.1%) and metastatic status (20.9%) as the main contributors. Among 223 patients (mean age 27.8 years), non-seminomatous tumors predominated (53.4%), and 24.7% were poor-risk by IGCCCG. Diagnostic delay > 6 months occurred in 25.6% and was strongly associated with mortality (OR 12.98, p < 0.001). High AFP, elevated LDH, and non-seminomatous histology were additional predictors of adverse outcomes. Conclusions: A diagnostic delay > 6 months significantly increased mortality risk. A machine learning model using only clinical stage and metastasis achieved 83.3% accuracy, highlighting the potential of simple, low-cost tools for early risk stratification and improved clinical decision-making in resource-limited settings.
