Objective: To evaluate the adherence of large language model (LLM)-based healthcare research to the Minimum Reporting Items for Clear Evaluation of Accuracy Reports of Large Language Models in Healthcare (MI-CLEAR-LLM) checklist, a framework designed to enhance the transparency and reproducibility of studies on the accuracy of LLMs for medical applications. Materials and Methods: A systematic PubMed search was conducted to identify articles on LLM performance published in high-ranking clinical medicine journals (the top 10% in each of the 59 specialties according to the 2023 Journal Impact Factor) from November 30, 2022, through June 25, 2024. Data on the six MI-CLEAR-LLM checklist items: 1) identification and specification of the LLM used, 2) stochasticity handling, 3) prompt wording and syntax, 4) prompt structuring, 5) prompt testing and optimization, and 6) independence of the test data—were independently extracted by two reviewers, and adherence was calculated for each item. Results: Of 159 studies, 100% (159/159) reported the name of the LLM, 96.9% (154/159) reported the version, and 91.8% (146/159) reported the manufacturer. However, only 54.1% (86/159) reported the training data cutoff date, 6.3% (10/159) documented access to web-based information, and 50.9% (81/159) provided the date of the query attempts. Clear documentation regarding stochasticity management was provided in 15.1% (24/159) of the studies. Regarding prompt details, 49.1% (78/159) provided exact prompt wording and syntax but only 34.0% (54/159) documented prompt-structuring practices. While 46.5% (74/159) of the studies detailed prompt testing, only 15.7% (25/159) explained the rationale for specific word choices. Test data independence was reported for only 13.2% (21/159) of the studies, and 56.6% (43/76) provided URLs for internet-sourced test data. Conclusion Although basic LLM identification details were relatively well reported, other key aspects, including stochasticity, prompts, and test data, were frequently underreported. Enhancing adherence to the MI-CLEAR-LLM checklist will allow LLM research to achieve greater transparency and will foster more credible and reliable future studies.

Background: Despite the fact that an increasing number of studies have focused on developing therapies for acute lung injury, managing acute respiratory distress syndrome (ARDS) remains a challenge in intensive care medicine.Whether the pathology of animal models with acute lung injury in prior studies differed from clinical symptoms of ARDS, resulting in questionable management for human ARDS. To evaluate precisely the therapeutic effect of trans‑ planted stem cells or medications on acute lung injury, we developed an animal model of severe ARDS with lower lung function, capable of keeping the experimental animals survive with consistent reproducibility. Establishing this animal model could help develop the treatment of ARDS with higher efficiency. Results: In this approach, we intratracheally delivered bleomycin (BLM, 5 mg/rat) into rats’ left trachea via a needle connected with polyethylene tube, and simultaneously rotated the rats to the left side by 60 degrees. Within sevendays after the injury, we found that arterial blood oxygen saturation ­(SpO2 ) significantly decreased to 83.7%, partial pressure of arterial oxygen ­(PaO2 ) markedly reduced to 65.3 mmHg, partial pressure of arterial carbon dioxide ­(PaCO2 )amplified to 49.2 mmHg, and the respiratory rate increased over time. Morphologically, the surface of the left lung appeared uneven on Day 1, the alveoli of the left lung disappeared on Day 2, and the left lung shrank on Day 7. A his‑ tological examination revealed that considerable cell infiltration began on Day 1 and lasted until Day 7, with a larger area of cell infiltration. Serum levels of IL-5, IL-6, IFN-γ, MCP-1, MIP-2, G-CSF, and TNF-α substantially rose on Day 7. Conclusions This modified approach for BLM-induced lung injury provided a severe, stable, and one-sided (left-lobe) ARDS animal model with consistent reproducibility. The physiological symptoms observed in this severe ARDS animal model are entirely consistent with the characteristics of clinical ARDS. The establishment of this ARDS animal model could help develop treatment for ARDS.

This report provides a summary of the 2024 survey results on the external quality assessment (EQA) scheme for the general transfusion medicine test and the general transfusion antibody test programs in Korea. Proficiency testing materials were prepared at the Asan Medical Center for bi-annual distribution to participating laboratories. The accuracy rates and number of participating laboratories for the bi-annual EQAs were: ABO typing, 99.6%–99.9% (n=944, n=945); RhD typing, 99.9%–100.0% (n=929, n=930);crossmatching, 95.0%–99.2% (n=825, n=825); unexpected antibody scre ening, 99.5%–100.0% (n=363, n=367); direct antiglobulin test (DAT) using a polyspecific reagent, 99.3%–100.0% (n=296, n=299); DAT using an antiimmunoglobulin G monospecific reagent, 100.0% (n=74, n=72); and DAT using an anti-C3d monospecific reagent, 98.6%–100.0% (n=72, n=71). The 2024 EQA scheme for the transfusion medicine program has improved and maintained the standards of the participating laboratories.

Purpose: Postoperative ileus (POI) typically occurs after abdominal surgery but can also affect patients undergoing spinal surgery via the lateral retroperitoneal approach, such as oblique lumbar interbody fusion (OLIF). Therefore, this study aimed to investigate the incidence and risk factors associated with POI in OLIF. Methods: This retrospective study examined a cohort of 465 patients who underwent OLIF from 2015 to 2023. Patient demographics, comorbidities, pre- and postoperative laboratory test results, and perioperative status were assessed. General condition of patients was assessed using the modified frailty index-11 (mFI-11), prognostic nutrition index, and geriatric nutrition risk index. In OLIF, the size and location of the psoas muscle involved in retraction and its relationship with the vertebral body were also investigated. Results: POI occurred in 19 patients (4%). Lower mFI-11 was linked to a higher risk of POI. While psoas muscle size had no significant effect on the risk of POI, the anterior location of the psoas muscle relative to the vertebral body was associated with a higher occurrence of POI. Multivariate logistic regression analysis of POI identified mFI-11 as the most significant risk factor (p = 0.003). Conclusion This study demonstrated that frailty and nutritional status can influence the occurrence of POI after OLIF. Additionally, bowel manipulation associated with the location of psoas muscle and vertebral body was identified as a risk factor. Proper assessment and improvement in patient frailty and nutritional status before surgery can help predict and prevent the occurrence of postoperative POI.

Objective: Whether elevation in plasma levels of amyloid and tau protein biomarkers are better indicators of cognitive decline than higher baseline levels in patients with amnestic mild cognitive impairment (aMCI) and Alzheimer’s disease (AD) remains understudied. Methods: We included 67 participants with twice testing for AD-related plasma biomarkers via immunomagnetic reduction (IMR) assays (amyloid beta [Aβ]1-40, Aβ1-42, total tau [t-Tau], phosphorylated tau [p-Tau] 181, and alpha-synuclein [α-Syn]) and the Mini-Mental State Examination (MMSE) over a 1-year interval. We examined the correlation between biomarker levels (baseline vs. longitudinal change) and annual changes in the MMSE scores. Receiver operating characteristic curve analysis was conducted to compare the biomarkers. Results: After adjustment, faster cognitive decline was correlated with lower baseline levels of t-Tau (β=0.332, p=0.030) and p-Tau 181 (β=0.369, p=0.015) and rapid elevation of t-Tau (β=-0.330, p=0.030) and p-Tau 181 levels (β=-0.431, p=0.004). However, the levels (baseline and longitudinal changes) of Aβ1-40, Aβ1-42, and α-Syn were not correlated with cognitive decline. aMCI converters had lower baseline levels of p-Tau 181 (p=0.002) but larger annual changes (p=0.001) than aMCI non-converters. The change in p-Tau 181 levels showed better discriminatory capacity than the change in t-Tau levels in terms of identifying AD conversion in patients with aMCI, with an area under curve of 86.7% versus 72.2%. Conclusion We found changes in p-Tau 181 levels may be a suitable biomarker for identifying AD conversion.

Objective: The treatment rate for alcohol use disorder (AUD) in Korea is significantly lower than its prevalence rate. Because untreated AUD can have harmful consequences, it is important to identify the factors that contribute to individuals with AUD seeking mental health services. Methods: We collected nationally representative data from the National Mental Health Survey of Korea 2021 and analyzed responses from 643 individuals with AUD, of which 76.8% were male. Factors related to mental health service utilization among individuals with AUD were classified into three categories: sociodemographic (such as sex, age, marital status, education, and monthly household income), clinical (including symptom severity, psychiatric comorbidity, suicidality, and physical illness), and psychological characteristics (like perceived stigma, loneliness and social isolation, and resilience). We used multiple logistic regression analyses to examine each characteristic separately and combined in a single model to determine the most significant factors. Results: The three logistic regression models revealed that sex, psychiatric comorbidity, physical illness, and perceived stigma are significantly linked to the utilization of mental health services among individuals with AUD. Results from the comprehensive model indicated that only physical illness and perceived stigma have significant associations with mental health service utilization. Conclusion These findings can assist in developing targeted interventions for individuals with AUD.

PURPOSE: The aim of this study was to evaluate the clinical performance and survival rate of implant-supported zirconia-lithium disilicate (Zr-LiSi) bilayered ceramic prostheses over 3 years. MATERIALS AND METHODS: This study included 71 patients, including 34 with implant-supported metal-ceramic prostheses (control group) and 37 with implant-supported Zr-LiSi bilayered ceramic prostheses (test group). The implant survival rate and incidence of prosthetic and biological complications (veneer fractures, dislodgement of screw-access hole filling material, screw loosening, peri-implant mucositis and peri-implantitis, and marginal bone loss) were investigated. The survival rate was analyzed using Kaplan-Meier survival curves, and the identity between two groups was confirmed by the log-rank test. RESULTS: Both groups showed a 100% survival rate, whereas the prosthetic survival rates were 77% and 73% for the metal-ceramic and Zr-LiSi groups, respectively. Biological complications did not appear in the metal-ceramic group, and 16.2% of peri-implant mucositis occurred in the Zr-LiSi group, which was significant (P < .05). Prosthetic complications occurred in 5.8% of the metal-ceramic group with veneer fractures and did not occur in the Zr-LiSi bilayered ceramic group. CONCLUSION This study revealed that posterior Zr-LiSi bilayered ceramic implant prostheses showed high survival rates and similar survival rates to metal-ceramic implant prostheses; however, additional consideration should be given to avoid overcontouring. Zr-LiSi bilayered ceramic implant prostheses may be an option for posterior implant-supported prosthetic treatment.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely prescribed for their anti-inflammatory and analgesic effects; however, their prolonged use significantly contributes to peptic ulcer disease (PUD) and its complications, such as bleeding and perforation. The pathogenesis primarily involves cyclooxygenase (COX) enzyme inhibition and direct mucosal injury, leading to impaired gastrointestinal defense mechanisms. Multiple risk factors, including advanced age, a history of ulcers, and the concurrent use of anticoagulants or corticosteroids, significantly increase the risk of ulcers and related complications. Global epidemiological studies demonstrate considerable geographical variation in prevalence rates. Despite higher NSAID usage, high-income countries exhibit relatively lower rates, primarily due to well-established preventive strategies. Prevention should be based on careful risk stratification that accounts for both gastrointestinal and cardiovascular factors. Proton pump inhibitors have demonstrated superior efficacy in both prevention and treatment, while selective COX-2 inhibitors offer an alternative strategy, though they require careful cardiovascular risk assessment. The synergistic interaction between NSAID use and Helicobacter pylori infection necessitates testing and eradication, particularly in high-risk patients. NSAID discontinuation remains the primary therapeutic strategy when feasible, with studies showing significantly improved healing rates compared with continued use. Recent advances include the emergence of potassium-competitive acid blockers, which provide rapid and sustained acid suppression, offering promising alternatives for both prevention and treatment. Continued research aimed at optimizing preventive strategies and developing novel therapeutic approaches remains essential for improving clinical outcomes in NSAID-induced PUD.

Background/Aims: Atrial fibrillation (AF) significantly contributes to global morbidity and mortality. Paroxysmal atrial fibrillation (PAF) is particularly common among patients with cryptogenic strokes or transient ischemic attacks and has a silent nature. This study aims to develop reliable artificial intelligence (AI) algorithms to detect early signs of AF in patients with normal sinus rhythm (NSR) using a 12-lead electrocardiogram (ECG). Methods: Between 2013 and 2020, 552,372 ECG traces from 318,321 patients were collected and split into training (n = 331,422), validation (n = 110,475), and test sets (n = 110,475). Deep neural networks were then trained to predict AF onset within one month of NSR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). An explainable AI technique was employed to identify the inference evidence underlying the predictions of deep learning models. Results: The AUROC for early diagnosis of PAF was 0.905 ± 0.007. The findings reveal that the vicinity of the T wave, including the ST segment and S-peak, significantly influences the ability of the trained neural network to diagnose PAF. Additionally, comparing the summarized ECG in NSR with those in PAF revealed that nonspecific ST-T abnormalities and inverted T waves were associated with PAF. Conclusions Deep learning can predict AF onset from NSR while detecting key features that influence decisions. This suggests that identifying undetected AF may serve as a predictive tool for PAF screening, offering valuable insights into cardiac dysfunction and stroke risk.

Background/Aims: This study evaluated the clinical utility of serum periostin measured at diagnosis in reflecting activity at diagnosis and predicting all-cause mortality during follow-up in patients with antineutrophil cytoplasmic antibody-associated vasculitis (AAV). Methods: This study included 76 patients with AAV whose serum periostin was measured from sera collected and stored at diagnosis. The correlation of either serum periostin or the Birmingham Vasculitis Activity Score (BVAS) with other variables was evaluated. Cumulative survival rates were compared using Kaplan–Meier survival analysis. The variables at diagnosis were compared between deceased and surviving patients. Hazard ratios were obtained by Cox proportional hazard analysis. Results: The median age of the 76 patients was 64.0 years and 60.5% were female. The median BVAS and serum periostin were 5.0 and 10.9 ng/mL, respectively. Five of the 76 patients (6.6%) died. Serum periostin was independently correlated with cross-sectional BVAS, the Vasculitis Damage Index (VDI), white blood cell count, and serum albumin. Patients with serum periostin ≥ 15.9 ng/mL at diagnosis had a significantly lower cumulative survival rate than those without. In addition to high VDI, dyslipidaemia frequency, and C-reactive protein, deceased patients showed higher serum periostin than surviving patients. In multivariable Cox analysis, however, only dyslipidaemia rather than serum periostin was identified as an independent predictor of all-cause mortality. Conclusions This study is the first to demonstrate that serum periostin at diagnosis could independently reflect cross-sectional BVAS and further partially contribute to all-cause mortality prediction in patients with AAV.