Immune checkpoint inhibitors have reshaped cancer treatment, but the response patterns they produce—pseudoprogression, dissociated responses, and hyperprogressive disease (HPD)—fit awkwardly into the size-based logic of RECIST 1.1. This narrative review, prepared with reference to the SANRA criteria, follows the evolution of response assessment from RECIST 1.1 through immune-adapted criteria such as iRECIST to functional and molecular imaging and, most recently, artificial intelligence (AI) models trained on longitudinal imaging. One distinction runs through the entire argument: current AI and radiomics models are risk-stratification tools, not diagnostic classifiers of HPD. Because HPD is defined as acceleration of tumor growth relative to a pretreatment trajectory, its diagnosis requires at least three imaging time points (pre-baseline, baseline, and first on-treatment assessment); no single-time-point model, however sophisticated, can reconstruct that trajectory. We appraise Image Biomarker Standardisation Initiative-compliant radiomics, convolutional neural networks, vision transformers, and early longitudinal modeling, and we compare early, joint, and late strategies for fusing imaging with clinical and molecular data. The clinical meaning of reported performance gains is weighed against standard assessment practice. Most evidence remains retrospective, and persistent obstacles include heterogeneous acquisition protocols, small cohorts, missing pre-baseline scans, endpoint misclassification, weak external validation, and limited biological interpretability. Prospective multicenter data collection, transparent model reporting, calibration and domain-shift testing, and biologically anchored validation are prerequisites before imaging-based AI biomarkers can enter routine immuno-oncology care.
Nursing internship serves as a pivotal link between nursing education and clinical practice, functioning as a key platform for cultivating applied nursing professionals. The university-hospital collaboration model integrates the resource advantages of both academic institutions and medical institutions, promoting the deep integration of theoretical knowledge and clinical skills, thereby significantly enhancing the practical abilities and professional competitiveness of nursing undergraduates. Although this model has achieved positive results in expanding practical opportunities, deepening resource sharing, and strengthening professional competence cultivation, it still faces prominent issues such as uneven mentor qualifications, inadequate pre-service training systems, incomplete internship management mechanisms, insufficient student rights protection, and weakened practical education functions. This paper proposes that standardized selection and training of mentors, the establishment of a comprehensive practical teaching system, enhanced protection of interns’ rights and safety, implementation of humanistic care and psychological support, and the creation of a university-hospital collaborative management mechanism should be adopted to standardize nursing education and precisely align with clinical needs. The optimized university-hospital collaboration model can effectively improve the quality of nursing talent cultivation, stabilize the nursing workforce, adapt to the demands of public health services, promote professionalization and high-quality development in nursing, and provide solid talent support for the implementation of the Healthy China strategy.
Congenital pyloric atresia (CPA) is a rare cause of neonatal gastric outlet obstruction. Although usually isolated, CPA may coexist with additional gastrointestinal atresias. Its association with distal duodenal atresia (DA) is particularly relevant because the segment between the two obstructions may become a closed duodenal loop into which biliary and pancreatic secretions continue to drain. We report the case of a male newborn delivered at 35 + 3 weeks of gestation after prenatal detection of severe polyhydramnios, recurrent bowel dilatation, abdominal effusion, and a cystic abdominal lesion. Postnatal radiography did not show the classic double-bubble sign. Exploratory laparotomy revealed CPA, marked dilatation of the second duodenal portion, distal DA, extensive adhesions, and jejunal perforations. Surgical management included adhesiolysis, reduction duodenoplasty, gastrointestinal continuity reconstruction, and jejunal perforations repair. The postoperative course was complicated by hemorrhagic–hypovolemic shock, coagulopathy, liver dysfunction, and the need for a second surgery due to increasing abdominal free fluid and suspected pneumoperitoneum, followed by progressive recovery. Follow-up demonstrated adequate gastric emptying and duodenal transit, satisfactory growth, and age-appropriate neurodevelopment. A focused narrative review on concomitant CPA and DA was also conducted. Concomitant CPA and distal DA should be considered when prenatal imaging shows a cystic abdominal lesion associated with polyhydramnios and ascites. Recognizing closed duodenal loop syndrome may improve prenatal counseling, atypical imaging interpretation, and surgical planning.
ObjectiveThis study aims to elucidate the clinicopathological characteristics, molecular profiles, and prognostic outcomes of congenital neuroblastoma in Chinese infants.MethodsA retrospective single-center analysis was conducted on 15 pathologically confirmed congenital neuroblastoma cases in China from 2017 to 2025. Clinical, imaging, pathological features and FISH for MYCN amplification and KMT2A rearrangement were evaluated.ResultsThe cohort included 15 patients with a male-to-female ratio of 8:7. Eight cases were detected prenatally and seven within 28 postnatal days. Most tumors (12/15) arose in the adrenal gland. Preoperative serum NSE elevation was more sensitive than VMA. All patients underwent complete resection; one received adjuvant chemotherapy and one relapsed. No MYCN amplification or KMT2A rearrangement was found.ConclusionIn this single-center Chinese cohort, congenital neuroblastoma showed favorable biological behavior. Complete resection achieved good outcomes. Combination of NSE and imaging examinations is promising for postoperative prognostic evaluation and recurrence surveillance in these pediatric patients.
BackgroundFracture-related infection (FRI) is a major complication after open reduction and internal fixation (ORIF) of tibial plateau fractures. This study evaluated whether a preoperative computed tomography (CT)-derived soft-tissue envelope index (STEI) was associated with 12-month FRI and whether it provided incremental value for preoperative risk stratification.MethodsThis retrospective single-center cohort study included 347 adults with unilateral closed tibial plateau fractures who underwent ORIF between January 2020 and December 2024. The STEI was defined as the mean ratio of soft-tissue area to osseous contour area at three axial levels located 5, 15, and 25 mm below the tibial plateau articular reference plane. FRI was adjudicated according to consensus confirmatory criteria. Every included patient had a dated, traceable, archived outcome record generated during clinical care after the 12-month anniversary of definitive ORIF. Non-FRI status required both the absence of any confirmatory criterion during the first postoperative year and explicit confirmation in a subsequent archived record. Logistic regression, receiver operating characteristic (ROC) curve analysis, calibration, decision curve analysis, and internal validation using 1,000 bootstrap resamples were performed.ResultsThe median follow-up duration was 12.8 months [interquartile range (IQR), 12.2–13.9], and 41 patients developed FRI (11.82%). The STEI was higher in the FRI group than in the non-FRI group (3.08 ± 0.52 vs. 2.51 ± 0.48; P < 0.001), and the interobserver intraclass correlation coefficient (ICC) was 0.946 [95% confidence interval (CI), 0.920–0.963]. The STEI remained independently associated with FRI [adjusted odds ratio (OR) per 0.10-unit increase, 1.21; 95% CI, 1.12–1.31; P < 0.001]. The combined preoperative model yielded an area under the ROC curve (AUC) of 0.865 (95% CI, 0.806–0.923), an optimism-corrected AUC of 0.847, and a calibration slope of 0.91, and provided greater net benefit across threshold probabilities of 5%−30%.ConclusionThe preoperative CT-derived STEI was independently associated with 12-month FRI after ORIF of closed tibial plateau fractures and added predictive information beyond clinical and fracture-related factors. Prospective multicenter external validation of both the index and the combined model is required.
BackgroundChronic obstructive pulmonary disease (COPD) is a heterogeneous disorder involving varying contributions from airway abnormalities and emphysematous destruction. However, structural airway alterations in patients with airflow limitation but minimal emphysema remain incompletely characterized. Quantitative CT (QCT)-based airway morphology analysis may capture additional aspects of airway geometry associated with airway-predominant structural patterns.ObjectivesThis study aimed to evaluate lobar airway inter-tapering abnormalities on quantitative CT in patients with non-emphysematous COPD (NE-COPD) and investigate whether incorporation of inter-tapering features improves discrimination within the studied cohort.MethodsThis retrospective cross-sectional study included 173 participants who underwent inspiratory thin-section chest CT and post-bronchodilator pulmonary function testing between January 2023 and March 2025. NE-COPD was defined as airflow limitation with minimal emphysema quantified by low-attenuation area percentage below −950 HU (LAA%-950 < 5%). Automated airway segmentation and centerline-based analysis were used to quantify fourth-generation lobar inter-tapering indices. A clinical model incorporating age, sex, and smoking exposure was constructed. An imaging-augmented model was subsequently developed using LASSO-selected lobar inter-tapering features together with global airway branching complexity. Model performance was evaluated using receiver operating characteristic analysis, bootstrap validation, calibration assessment, decision curve analysis, and SHAP interpretation.ResultsAmong the 173 participants, 87 had NE-COPD and 86 were Non-COPD controls. Compared with Non-COPD controls, patients with NE-COPD demonstrated altered fourth-generation lobar inter-tapering patterns in multiple lobes. The clinical model achieved an area under the curve (AUC) of 0.717. After incorporation of quantitative airway features, the imaging-augmented model demonstrated improved discrimination, with an AUC of 0.851. Bootstrap validation demonstrated stable model performance. SHAP analysis indicated that four lobar fourth-generation inter-tapering indices accounted for a substantial proportion of feature contribution within the fitted model. Decision curve analysis suggested higher theoretical net benefit of the imaging-augmented model within the evaluated threshold probability range.ConclusionQuantitative CT-derived lobar inter-tapering was associated with pulmonary function-defined non-emphysematous COPD. These findings suggest that lobar inter-tapering represents a candidate geometric CT descriptor for characterizing airway-predominant structural patterns. Further multicenter studies with external validation and integration of complementary quantitative CT descriptors are required to determine its broader clinical utility.
BackgroundEvidence is limited on the prospective prognostic value of longitudinal blood pressure (BP) and resting heart rate (RHR) dynamics after percutaneous coronary intervention (PCI).AimThis study aimed to identify joint BP-RHR trajectories and evaluate their impact on net adverse clinical events (NACE).MethodsIn a prospective post-PCI cohort, latent class trajectory modeling identified 6-month patterns of systolic BP, diastolic BP, and RHR, classifying patients into four joint trajectories. Inverse probability of treatment weighting (IPTW) was employed to rigorously balance baseline confounders. The primary endpoint was NACE. The incremental predictive value of the joint trajectories was assessed using the C-index, continuous net reclassification improvement (cNRI), and integrated discrimination improvement (IDI).ResultsAmong 736 patients (71 NACE events during follow-up), four distinct joint trajectory groups were identified: Double Stable (n = 329), BP-Fluctuating (n = 215), RHR-Fluctuating (n = 108), and Double Fluctuating (n = 84). After IPTW adjustment, compared with the Double Stable group, the risk of NACE was significantly elevated in the BP-Fluctuating group (HR: 2.21, 95% CI: 1.23–3.97), the RHR-Fluctuating group (HR: 1.85, 95% CI: 1.09–3.13), and peaked in the Double Fluctuating group (HR: 2.69, 95% CI: 1.18–6.14). Furthermore, incorporating the joint trajectory groups into a baseline clinical risk model significantly improved prognostic discrimination and reclassification, with significant increases in the C-index (p < 0.001), IDI improvement (p = 0.033), and continuous net reclassification improvement.ConclusionJoint BP and RHR trajectories effectively stratify post-PCI NACE risk. Patients with combined fluctuations face the highest risk, highlighting the need for dynamic hemodynamic monitoring over static single-point assessments.
Background and aimsRefractory esophageal variceal bleeding (REVB) in patients with cirrhosis is associated with high mortality. Self-expanding metal stents (SEMS) serve as a salvage therapeutic option. However, data regarding their prolonged use as a bridge to transjugular intrahepatic portosystemic shunt (TIPS) remain limited. We report a case in which REVB was managed with prolonged SEMS placement before TIPS.Case presentationA 72-year-old female with decompensated cirrhosis presented with acute hematemesis. After failure of endoscopic injection sclerotherapy and band ligation to control active bleeding from a ruptured esophageal varix, a fully covered SEMS was successfully deployed. The patient declined the scheduled stent removal and was subsequently followed conservatively. TIPS was performed 387 days after stent placement, followed by a successful stent retrieval. Clinical outcomes, including rebleeding episodes and procedure-related complications, were recorded throughout follow-up until July 26, 2026.ResultsImmediate hemostasis was achieved following SEMS deployment. During the 387-day indwelling period, the patient developed intermittent acid regurgitation, retrosternal discomfort, and endoscopic evidence of mucosal ulceration and granulation tissue at the stent site. The stent was successfully removed after TIPS, and follow-up endoscopy revealed complete resolution of the esophageal varices. The patient remained free of recurrent bleeding till the last follow-up. However, two episodes of hepatic encephalopathy occurred after TIPS.ConclusionThis case suggests that successful removal of a fully covered esophageal SEMS may remain technically feasible even after an exceptionally prolonged retention period. Nonetheless, this observation should not be interpreted as support for routine prolonged stent placement, which lies outside current guideline recommendations and may entail substantial risks.
AimThis study aims to establish a standardized training indicator system for cardiac Intensive Care Unit (CICU) specialized nurses based on competency theory, so as to provide practical guidance for improving their core competencies and guaranteeing high-quality care for critically ill patients.MethodsThe training indicator system was developed through literature review, semi-structured qualitative interviews and Delphi method. Literature review was conducted to extract competency dimensions and indicators covering professional knowledge, professional skills, professional competence and practical competence; semi-structured qualitative interviews with clinical and educational specialists were further carried out to develop an initial training indicator system with 4 primary, 19 secondary and 60 tertiary indicators. Followed by two rounds of Delphi consultation, 23 experts specializing in CICU Medicine, CICU nursing management, CICU clinical nursing, and nursing education were invited to participate. Indicators were retained when importance score >3.50 and coefficient of variation <0.25. The Analytic Hierarchy Process (AHP) was used to determine the weights of indicators at different levels.ResultsBoth two rounds of expert surveys achieved a 100% response rate, with expert authority coefficients of 0.884 and 0.894, respectively. The Kendall’s ranged from 0.241 to 0.247 (p < 0.001), and the coefficient of variation ranged from 0.070 to 0.131. The final version of the training indicator system includes 4 first-level indicators, 19 secondary-level indicators, and 63 tertiary-level indicators. All indicators obtained importance scores above 4.47 with reasonable weight distribution.ConclusionThis competency-oriented CICU specialized nurse training indicator system possesses favorable scientificity and operability, and can serve as a reference for standardized professional development.
BackgroundMalignant melanoma is an aggressive skin cancer with rising incidence. Accurate early staging and standardized treatment are crucial for prognosis. This study evaluates seven large language models (LLMs)-including GPT-5.2, Gemini-3.1, and medically enhanced models-in assisting non-metastatic melanoma management amid clinical complexities.MethodsEmploying a prospective, simulated expert-blinded design, 59 virtual cases across TNM stages, ages, and comorbidities were assessed. Multiple senior oncologists independently evaluated model outputs using a 6-point Likert scale for staging accuracy, treatment rationality, and protocol standardization.ResultsGPT-5.2 (5.56 ± 1.12) and Gemini-3.1 (5.25 ± 1.3) achieved the highest staging accuracy, while AntAngelMed performed worst (2.93 ± 1.67). Performance declined significantly in complex Stage III cases. GPT-5.2 and Gemini-3.1 also led in treatment rationality, showing stability, whereas model performances converged in early stages but diverged in advanced ones. Gemini-3.1 excelled in protocol standardization (5.17 ± 0.57), though some models posed risks like insufficient surgical margin recommendations.ConclusionLeading LLMs demonstrate potential for high-quality melanoma management but exhibit inconsistent performance influenced by architecture and case complexity, with reduced reliability in advanced stages. Future tools require risk-stratified guidelines and real-world validation to improve patient outcomes.
ObjectiveThis study aims to develop and validate machine learning (ML) models for predicting treatment failure in trauma patients using comprehensive clinical and laboratory variables, and to identify key prognostic features.MethodsA retrospective cohort of 318 trauma patients was included. We included 44 characteristics, and the primary outcome was treatment failure at hospital discharge, defined as in-hospital death, an unimproved or worsened discharge status, discharge against medical advice or withdrawal of active treatment because of critical illness, or a GOS score of 1–3 in patients with concomitant traumatic brain injury. The dataset was randomly divided into a training set (70%) and a test set (30%). Features were selected via least absolute shrinkage and selection operator (LASSO) regression in the training set. 5 ML models—Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), k-Nearest Neighbors (KNN), and Extreme Gradient Boosting (XGBoost)—were trained and evaluated. The optimal model was interpreted using SHapley Additive exPlanations (SHAP) analysis.ResultsLASSO regression selected 13 candidate predictors for model development in the training set. The RF model demonstrated the best performance in the test set, with an AUC of 0.967 (95% CI: 0.936–0.997), sensitivity of 1.000, specificity of 0.855, and F1-score of 0.776. SHAP analysis identified Glasgow Coma Scale (GCS) score as the most influential predictor, followed by Multiple Organ Dysfunction Syndrome (MODS), Acute PHysiology and Chronic Health Evaluation (APACHE II) score, Injury Severity Score (ISS), and creatine kinase (CK). Higher APACHE II and ISS scores were positively associated with treatment failure, while higher GCS, absence of MODS and CK levels correlated with reduced risk.ConclusionUtilizing multiple trauma severity scores, laboratory parameters, and machine learning algorithms, we developed a predictive model to identify trauma patients at risk of treatment failure within the first 24 h of admission. Among the algorithms evaluated, RF model demonstrated superior internal discriminative performance in our single-center cohort. SHAP interpretability analysis reveals the core prognostic value of GCS, MODS, APACHE II, ISS, and CK, providing a potential transparent auxiliary tool for clinical risk stratification. However, its generalization performance still needs multi center external validation.
ObjectiveReliable biomarkers for identifying patients at increased risk of osteomyelitis recurrence during the early postoperative period are lacking. The C-reactive protein-to-albumin ratio (CAR), which integrates systemic inflammatory burden and nutritional status, has shown diagnostic value in musculoskeletal infections, but its prognostic role after surgical treatment of osteomyelitis remains unclear. This study evaluated whether early postoperative CAR predicts osteomyelitis recurrence and compared its performance with conventional inflammatory biomarkers.Patients and methodsThis retrospective cohort study included 90 consecutive adults who underwent surgical treatment for osteomyelitis between January 2020 and December 2024. Serum inflammatory biomarkers were recorded before surgery and at the first postoperative outpatient visit, approximately one month after surgery. The primary outcome was osteomyelitis recurrence during a minimum follow-up of 12 months. Receiver operating characteristic (ROC) analysis was used to evaluate the predictive performance of inflammatory biomarkers, and multivariable logistic regression was used to evaluate factors independently associated with recurrence.ResultsDuring a mean follow-up of 19 months, recurrence occurred in 19 patients (21.1%). Patients with recurrence had significantly higher baseline and one-month CAR values than those without recurrence (both p < 0.001). Among the evaluated biomarkers, one-month CAR demonstrated the numerically highest AUC for recurrence (AUC = 0.862; optimal cutoff = 1.41), with a sensitivity of 73.7% and specificity of 87.3%. However, its AUC was not significantly different from that of one-month CRP (AUC = 0.831; DeLong p = 0.223). In multivariable analysis, one-month CAR remained independently associated with recurrence within the parsimonious multivariable model (OR 5.42, 95% CI 2.38–12.36; p < 0.001), together with the Charlson Comorbidity Index (OR 1.94, 95% CI 1.27–2.96; p = 0.002). Leukocyte-derived inflammatory indices showed substantially lower prognostic performance during postoperative follow-up.ConclusionEarly postoperative CAR was associated with osteomyelitis recurrence and showed good discriminative ability, with the numerically highest AUC among the evaluated inflammatory biomarkers. However, its discriminative performance was not statistically superior to that of one-month CRP. CAR may therefore represent a practical adjunctive biomarker for postoperative risk stratification. Prospective multicenter studies are warranted to validate these findings and determine whether CAR-guided surveillance improves patient outcomes.
BackgroundDespite advances in immunosuppressive therapies, lupus nephritis (LN) continues to be a major contributor to morbidity and mortality. The scarcity of reliable biomarkers for predicting renal outcomes further complicates clinical management. While Cordyceps is recognized for its immunomodulatory properties as a medicinal fungus, its therapeutic potential in LN remains underexplored. Moreover, it is unclear whether its molecular targets identified through screening constitute viable therapeutic interventions for this disease.MethodsNetwork pharmacology was integrated with LN-associated databases and transcriptomic analysis of GSE200306. Protein–protein interaction, Gene Ontology, and pathway-enrichment analyses were performed to identify candidate targets and biological processes. Molecular docking was used to evaluate predicted interactions between Cordyceps constituents and selected proteins. MAPK1 and CFB expression was further assessed using Nephroseq, an independent transcriptomic cohort (GSE112943), and immunohistochemistry in renal tissues from 30 patients with LN and 10 controls. A multivariable model incorporating age, sex, estimated glomerular filtration rate, MAPK1, and CFB was evaluated using receiver-operating-characteristic, calibration, and decision-curve analyses.ResultsSeven active constituents and 111 putative Cordyceps targets were identified. Intersection with 707 LN-associated genes yielded 26 candidate therapeutic targets enriched in apoptosis, inflammatory responses, complement activation, and NF-κB and PI3K–Akt signaling. Analysis of GSE200306 identified 89 differentially expressed genes, among which MAPK1 and CFB intersected with the Cordyceps–LN target network. Both genes were upregulated in LN kidneys, enriched in the tubulointerstitial compartment, inversely associated with renal function, and supported by immunohistochemistry. The combined model achieved an area under the curve of 0.954 and showed potential clinical utility. In GSE112943, MAPK1 was significantly upregulated, whereas CFB showed a concordant but nonsignificant increase. Docking suggested plausible interactions between peroxyergosterol and MAPK1/CFB but did not establish direct target engagement.ConclusionsCollectively, these findings identify MAPK1 as a reproducible candidate biomarker and CFB as a compartment-sensitive candidate biomarker in LN. Both molecules represent hypothesis-generating intervention nodes through which Cordyceps-derived constituents may act; direct target-engagement and functional studies are required before they can be considered actionable therapeutic targets.
Machine learning can be used to support data-driven modeling of supercritical CO₂ processing. The method of machine learning modeling is applied in this work for evaluation of small-molecule processing under supercritical conditions. As a necessary step, the solute solubility in the solvent is evaluated via different models. The purpose of this investigation is to develop models for the solubility of Clobetasol Propionate (CP) based on the two parameters of temperature and pressure as the model's features. Adaptive Boosting (AdaBoost) is utilized in this study on top of three core models: Decision Tree (DT), Lasso, and Gaussian Process Regression (GPR). The models are tuned through the firefly algorithm (FA) to determine their unknown coefficients. The final models are called FB-DT (FA-optimized Boosted DT), FB-GPR (FA-optimized Boosted GPR), and FB-LASSO (FA-optimized Boosted LASSO) and have R2 scores of 0.934, 0.977, and 0.813, respectively, for fitting the solubility data of CP. Accordingly, FB-GPR is the most accurate model, with an RMSE of 1.11 × 10⁻2. The developed models demonstrate great performance in estimating CP solubility in supercritical CO₂ under different temperature and pressure conditions.
ObjectiveTo explore the efficacy of healthcare-associated infection (HAI) prevention and control management that is based on Healthcare Failure Mode and Effect Analysis (HFMEA) in the prevention of puerperal infection after cesarean section converted from trial of vaginal labor.MethodsAn HFMEA team was set up. The study was conducted from July to December 2023 (pre-intervention) and January to June 2024 (post-intervention), with 115 and 118 puerperae in the pre- and post-intervention periods, respectively. By mapping out the management process of cesarean section converted from trial of vaginal labor, key failure modes related to infection prevention and control were identified. Subsequently, potential causes of these failure modes were analyzed, and improvement plans were formulated, implemented, followed by the evaluation of their efficacy.ResultsFollowing the optimization of the management process for cesarean section converted from trial of vaginal labor using HFMEA, the Risk Priority Number (RPN) of potential failure modes leading to puerperal infection after this procedure was significantly reduced, with a notable improvement rate. The hand hygiene compliance rate increased from 60.9% (67/110) to 80.4% (90/112), the accuracy rate of surgical hand disinfection rose from 66.7% (80/120) to 90.7% (136/150), and the incidence of puerperal infection after the procedure decreased from 7.8% (9/115) to 1.7% (2/118), with statistically significant differences (P = 0.031). In contrast, the surgical site infection rate decreased from 1.7% (2/115) to 0% (0/118), with no statistically significant difference (P = 0.155).ConclusionThe application of HFMEA, as part of a multifaceted quality improvement initiative, was associated with a refinement of the management process for cesarean section converted from trial of vaginal labor and a lower incidence of puerperal infection. However, due to the before-and-after study design, causality cannot be definitively established.
BackgroundPhoenixin-14 (PNX-14) is a bioactive peptide associated with mitochondrial function, oxidative stress responses, and cellular protection. Although experimental data suggest a possible role in renal injury, its serum levels in chronic kidney disease (CKD) have not been previously studied. Given the limitations of conventional renal biomarkers, PNX-14 may offer additional information regarding renal dysfunction in CKD.MethodsThis single-center, prospective, observational, cross-sectional study included 140 adult patients with Stage 3 CKD (n = 30), Stage 4 CKD (n = 30), Stage 5 CKD (n = 30), and end-stage kidney disease receiving maintenance hemodialysis (HD) (n = 50). Serum PNX-14 was measured using a commercial ELISA kit, before and after dialysis in HD patients. Associations between PNX-14 and clinical and laboratory parameters were assessed using Spearman correlation. Receiver operating characteristic (ROC) curve analysis evaluated the discriminative performance of PNX-14 across CKD stages.ResultsSerum PNX-14 levels increased progressively with advancing CKD stage, from a median of 17.88 (IQR 15.16–20.20) pg/mL in Stage 3–43.14 (IQR 23.78–95.71) pg/mL in the HD group (p < 0.001). Post hoc analysis showed that PNX-14 levels were significantly lower in Stage 3 than in Stage 5 and HD, in Stage 4 than in HD, and in Stage 5 than in HD. PNX-14 showed significant positive correlations with creatinine and parathyroid hormone (PTH) and a significant negative correlation with eGFR (all p < 0.001). In the pre-dialysis subgroup, PNX-14 remained significantly correlated with creatinine, eGFR, urea, and PTH. In HD patients, no significant difference was observed between pre-dialysis and post-dialysis PNX-14 levels (p = 0.152), and no significant correlation was found between PNX-14 and Kt/V. ROC analysis demonstrated good discriminative performance for distinguishing Stage 3 from advanced disease (Stage 5 + HD), with an AUC of 0.839 (95% CI: 0.755–0.913). The optimal cut-off was 19.59 pg/mL, with 87.5% sensitivity and 73.3% specificity.ConclusionsSerum PNX-14 levels increase with worsening CKD stage and are strongly associated with renal function parameters. Its limited change after HD and its discriminative performance suggest that PNX-14 may provide clinically relevant information regarding disease severity in CKD. Further prospective and longitudinal studies are needed to clarify its potential clinical utility.
Background and aimSplenic infarction after endoscopic cyanoacrylate injection for gastric varices (GV) is poorly characterized. We estimated its proportion among computed tomography (CT)-evaluable patients and explored clinical features and short-term outcomes.MethodsWe retrospectively screened 255 adult patients undergoing endoscopic glue injection for GV between January 2018 and February 2025. Eligible patients had CT within 30 days and no pre-existing infarction or alternative causes; imaging was clinically selected rather than protocol-mandated. Expanded 1:4 propensity score matching included clinical, laboratory, and index-session endoscopic features. Exploratory Firth regression addressed sparse events and complete separation.ResultsOf 183 CT-evaluable patients, 16 (8.7%) had radiologically confirmed splenic infarction. After matching, 16 cases and 64 controls were analyzed, although residual imbalance remained. All cases had ≥2 previous injection sessions versus 15.6% of matched controls, producing complete separation; the Firth estimate was therefore not considered a stable effect size. Exploratory models showed statistical signals for repeated injections, larger splenic diameter, large spontaneous portosystemic shunts, and greater index-session glue volume. These features may represent advanced portal hypertension and treatment complexity rather than independent risk factors. Most cases (93.8%) were managed conservatively. Thirty-day mortality and rebleeding did not differ, whereas hospital stay was longer (median, 14 vs. 10 days; p = 0.008).ConclusionsSplenic infarction was identified in 8.7% of this selected CT-evaluable cohort. The observed features should be regarded as exploratory clinical markers, not causal or actionable predictors. Imaging selection, residual imbalance, 16 events, complete separation, and wide uncertainty preclude incidence estimation, precise effect-size interpretation, or individual risk prediction.
BackgroundDental and oral health students experience substantial stress, yet the intervention evidence published over the past decade has not been comprehensively synthesized.MethodsWe searched MEDLINE/PubMed, Embase, Scopus, Web of Science Core Collection, CENTRAL, ERIC, and PsycINFO for reports published from January 2015 through June 2026. Randomized, non-randomized controlled, crossover, and uncontrolled pre–post studies were eligible if they evaluated an intervention and reported a validated perceived-stress, psychological-distress, burnout, or resilience outcome. Risk of bias was assessed with RoB 2, ROBINS-I, and design-specific criteria for uncontrolled studies. Synthesis followed SWiM and prioritized direction and magnitude, design, sample size, and precision rather than statistical significance alone; certainty was rated with GRADE.ResultsFourteen studies from 11 countries reported study sample sizes totaling 750 participants (median 47.5; range 5–120), including 102 participants from one mixed health-professions trial in which dentistry-specific outcomes were not separable; outcome-specific denominators were often smaller. Three were explicitly randomized controlled trials, one quasi-experimental study reported random group allocation, four used other controlled or crossover designs, and six were uncontrolled. Interventions comprised mindfulness or meditation (four studies), coaching/counseling/cognitive-behavioral or psychoeducational approaches (five), movement or sensory/relaxation approaches (three), and resilience training (two). Within-group improvements were common, but comparative findings were mixed: several apparent benefits arose from very small, self-selected, or high-attrition samples. One uncontrolled study (n = 69) directly measured burnout and found immediate reductions in personal and study-related domains, whereas colleague- and teacher-related domains did not change. Long-term follow was uncommon. Certainty was very low for perceived stress/psychological distress, burnout, and resilience.ConclusionsStudent-facing programs may improve selected short-term perceived-stress, psychological-distress, coping, or burnout outcomes, but durable effectiveness is unestablished. Confidential counseling/referral and brief, supported skills modules appear feasible as evaluated adjuncts; stand-alone digital or sensory interventions and claims of burnout prevention should remain experimental. These programs should not substitute for organization-level prevention.Systematic Review Registrationhttps://www.crd.york.ac.uk/PROSPERO/view/CRD420261449695.
