Purpose: This study evaluated the potential of wearable-derived longitudinal data for early detection of mild cognitive impairment (MCI). We examined whether variability in nocturnal sleep and heart rate (HR) patterns reflect autonomic nervous system instability associated with cognitive decline. Materials and Methods: Nocturnal data from 162 participants (111 cognitively normal; 51 MCI) were collected via smart rings over an average of 70 days. To mitigate signal dilution from long-term averaging (LTA), we derived variability-based features (including measures of dispersion, distributional properties, and time-series indices) and compared them to mean-based models. Logistic regression models were constructed through stepwise integration and validated with cross-validation. Key predictors were identified using SHapley Additive exPlanations (SHAP). Results: Mean-only models showed limited discrimination. In contrast, variability-focused models improved performance, with the model that integrated extended mean, distributional, and time-series features achieving an area under the curve of 0.861 and sensitivity of 0.820. LTA analyses revealed instability of mean effects (signed Cohen's d changed in magnitude/direction across 35-70-day windows), supporting a shift away from long-window means. SHAP identified HR_drop variability (short-term variability, mean absolute deviation-based distributional instability, time-bin fluctuation) as top contributors, while sleep-behavior rhythm (e.g., median/trimmed mean of daily sleep count) and circadian-timing features (e.g., mode of sleep midpoint, variability in the timing of the lowest HR) provided complementary information. Conclusion: Longitudinal wearable monitoring suggests that variability in sleep and HR patterns can capture subtle signs of cognitive decline more effectively than conventional, mean-based measures. These variability-based digital biomarkers may support early MCI screening and personalized intervention strategies.
Purpose: Cricothyroidotomy is a critical procedure for managing difficult airways, yet it is infrequently performed, leading to limited familiarity among medical professionals. This study explores the effectiveness of augmented reality (AR)-based training using HoloLens compared to traditional non-head mounted device (HMD) training in teaching cricothyroidotomy. This study aims to confirm the effect of technical practice using AR on participants' performance ability, performance confidence, learning selfefficacy, and practical satisfaction, and to present basic data for using AR in medical education. Materials and Methods: All participants received initial guidance via PowerPoint (PPT) on laptops. The AR group performed the procedure with three-dimensional models via HoloLens, while the non-HMD group used the initial PPT guidance. Effectiveness was assessed through a pre- and post-test design, including cricothyroidotomy estimated time, skill confidence, NASA-Task Load Index, System Usability Scale, and participant satisfaction. Results: The average pre-test procedure time for the AR group was 188 seconds, and 150 seconds for the non-HMD group. In the post-test, the AR group recorded 146 seconds and the non-HMD group 136 seconds, showing no statistically significant difference in procedure time (p>0.05). The HMD group showed a significant improvement in confidence levels (median increase from 1.0 to 5.0), and both groups demonstrated similar proficiency in cricothyroidotomy performance over time. Satisfaction rates were similar, but the AR group reported lower mental demand. Conclusion: Despite initial technical challenges, AR showed potential for enhancing learning and confidence without compromising performance, indicating the need for further research to optimize AR in medical training.
Purpose: Systemic cancer and ischemic stroke are major causes of mortality and share several risk factors, including older age, smoking, obesity, and inflammation. This study investigated the incidence of newly diagnosed cancers in stroke survivors and their outcomes. Materials and Methods: This retrospective observational study included patients with acute ischemic stroke enrolled between January 2010 and December 2019. We excluded patients with active or newly diagnosed cancer within 30 days of stroke onset. Primary outcomes were the incidence of new systemic cancer, recurrent ischemic stroke, and mortality. Results: A total of 5048 patients were included and followed for a median of 3.4 years (interquartile range 1.5-6.1 years). Among them, 242 (4.79%) were diagnosed with a new cancer, and 524 (10.4%) died during follow-up. At the time of cancer diagnosis, 48 patients (19.8%) had metastases. Factors associated with cancer occurrence included male sex [sub-distribution hazard ratio (SHR), 1.84; 95% confidence interval (CI), 1.27-2.67] and older age (SHR, 1.03; 95% CI, 1.01-1.04). During follow-up, 398 patients (7.9%) developed recurrent stroke, with a higher incidence in those with cancer than those without (14.9% vs. 7.5%, p<0.001). In multivariate Fine-Gray competing risk regression analyses, cancer was not significantly associated with recurrent stroke (SHR, 1.22; 95% CI, 0.77-1.91; p=0.395), but was significantly associated with increased mortality [hazard ratio, 2.57; 95% CI, 1.97-3.36; p<0.001]. Conclusion: Cancer frequently developed in stroke survivors. Patients who developed cancer had a higher risk of mortality. Efforts for early cancer detection are warranted.
PURPOSE:Patients with familial hypercholesterolemia (FH) are at high risk of coronary artery disease (CAD), and its prediction is of great clinical importance. This study aimed to identify predictors of CAD and construct an effective risk prediction model for Korean patients with FH. MATERIALS AND METHODS:Clinical and laboratory data of 245 patients were collected from the Korean FH registry. CAD was defined as ≥50% coronary stenosis on invasive or computed tomographic angiography. The cholesterol efflux capacity (CEC) was measured using J774A1 cells and radiolabeled cholesterol. Predictors of CAD were identified by multivariable logistic regression analysis, and the best-performing prediction model was determined based on its statistical indices. RESULTS:The participants' mean age was 49.1 years; 92 (37.6%) were male, and 41 (16.7%) had CAD. Age [odds ratio (OR) 1.05; p=0.02], hypertension (OR 7.28; p<0.0001), low-density lipoprotein-cholesterol (OR 1.60; p=0.02), and high-density lipoprotein-cholesterol (HDL-C) (OR 0.93; p=0.002) were identified as predictors of CAD in the multivariable analysis. In another analysis, age, hypertension, body mass index (OR 1.16; p=0.03), and apolipoprotein A1 (OR 0.97; p=0.002) were predictive of CAD. CEC did not have a significant predictive value. The best-performing CAD prediction model included six clinical and laboratory variables, with an area under the curve of 0.878. The effects of hypertension, smoking, and HDL-C were particularly influential. CONCLUSION:Traditional risk factors, including hypertension and HDL-C, were predictors of CAD in Korean patients with FH, whereas CEC was not. The resulting CAD prediction model demonstrated satisfactory performance in this population.
Purpose: The relationship between cognitive impairment and biological age (BA) is unclear. This study aimed to investigate the association between BA and cognitive impairment. Materials and Methods: A total of 2202 participants from the 2011-2012 and 2013-2014 National Health and Nutrition Examination Survey were included. Klemera-Doubal method age (KDM-Age), phenotypic age (PhenoAge), and their acceleration values were calculated based on laboratory parameters. Six machine learning models were constructed to compare performance metrics such as area under the receiver operating characteristic curve (AUC-ROC) and accuracy, and Shapley additive explanations (SHAP) was used to interpret feature contributions Results: After fully adjusting for confounders, each 1-year increase in PhenoAge was associated with a 6%, 8%, and 5% higher risk of cognitive impairment on the Consortium to Establish a Registry for Alzheimer's Disease (CERAD), Digit Symbol Substitution Test (DSST), and Animal Fluency Test (AFT) tests, respectively (all p<0.001); the highest quartile group exhibited a 5.98-17.48-fold in- crease in risk compared to the lowest. KDM-Age showed only a marginal association with DSST [odds ratio (OR)=1.02, p=0.049]. Among all models, the random forest (RF) performed best for CERAD prediction (AUC-ROC=0.997, sensitivity=99.5%), significantly outperforming the other algorithms (p<0.001). SHAP results demonstrated that sociodemographic factors had mean contribution values (0.014-0.135) significantly higher than those of BA indicators (<0.011), underscoring the dominant role of social determinants. Conclusion: PhenoAge is a biological aging indicator with suggestive value for identifying the risk of cognitive impairment in old- er adults. Combining the RF model with key sociodemographic information may help facilitate early risk screening.
Purpose: Lumbar fusion surgeries have increased substantially, making patient satisfaction an important indicator of surgical outcomes and quality of care. Conventional statistical approaches have limitations in predicting clinically significant improvement (CSI). This study aimed to develop a machine learning model to predict CSI after lumbar fusion surgery using only preoperative factors to support clinical decision-making. Materials and Methods: A total of 359 patients who underwent lumbar fusion surgery between January 2021 and December 2023 were included. Twenty-two preoperative variables, including demographic characteristics and comorbidities, were used for model development. A multi-label classification approach was applied to predict improvements in the 36-item short-form survey (SF-36) mental component summary (MCS) and physical component summary (PCS). CSI was defined as improvement when the average postoperative SF-36 score (PCS or MCS) across available follow-up time points exceeded the preoperative baseline value. Six machine learning algorithms were evaluated using 5-fold cross-validation. Model performance was assessed using seven evaluation metrics, with emphasis on the F1-score and the area under the receiver operating characteristic curve (AUROC). Model interpretability was examined using SHapley Additive exPlanations (SHAP) analysis and waterfall plots. Results: The Extra Trees model achieved the best performance, with an F1-score of 0.850 and an AUROC of 0.835. SHAP analysis identified hypertension, diabetes mellitus, body mass index, bone mineral density, and revision status as key predictors. Postoperative complications and revision surgery were also analyzed for their associations with comorbidities and outcomes. Conclusion: The machine learning model accurately predicts CSI after lumbar fusion surgery using preoperative factors and may assist clinicians and patients in preoperative decision-making.
Purpose: We aimed to develop deep learning models that can identify and separate the shapes of 14 bones in the foot and ankle using multi-view radiographs and predict their masks. Materials and Methods: We retrospectively collected 273 radiographs from 99 patients with anatomically normal feet, including anteroposterior (AP), oblique (OBL), and lateral (LAT) views, obtained between January 2020 and December 2021. In each view, 14 bones were segmented using AP and OBL radiographs and 6 bones using LAT radiographs. Ground truth masks were manually annotated by two radiology technologists and reviewed by an emergency medicine physician. Two deep learning models, a fully convolutional network (FCN) with ResNet-50 and DeepLabv3 with ResNet-50, were independently fine-tuned for semantic segmentation and evaluated using five-fold cross-validation. Results: In the AP and OBL views, both models attained mean intersection over union (mIoU) values ranging from 0.899 to 0.975 and from 0.875 to 0.978, respectively. In the LAT view, mIoU values varied from 0.926 to 0.976 for FCN-ResNet-50 and from 0.872 to 0.961 for DeepLabv3. FCN-ResNet-50 achieved slightly higher mIoU values than DeepLabv3, with statistically significant differences identified between the two models in the overall OBL view and across the 14 specific bones (p<0.05). Conclusion: The FCN-ResNet-50 and DeepLabv3 models could be effective in automatically segmenting foot and ankle bone structures using multi-view radiographs.
PURPOSE:Numerous studies have shown that the atherogenic index of plasma (AIP) is a strong predictor of the risk of cardiovascular diseases (CVDs) and an independent predictor of cardiovascular events (CEs) and related mortality. The predictive value of AIP for in-hospital mortality (IHM) in individuals with sepsis remains uncertain. Therefore, this study aimed to investigate the association between AIP and IHM among patients with sepsis using a large sample from the Medical Information Mart for Intensive Care (MIMIC) database. MATERIALS AND METHODS:Individuals with sepsis were identified from the MIMIC-IV database and categorized into four groups according to AIP quartiles. The IHM was the primary outcome. The association between AIP and IHM was evaluated using logistic regression and restricted cubic spline (RCS) models. RESULTS:A total of 2243 patients (59% male) were included this study. In univariable logistic regression, higher AIP was significantly associated with increased IHM (odds ratio: 1.18 [95% confidence interval: 1.05-1.33]; p of Wald test=0.005). The RCS model showed no evidence of a nonlinear association between AIP and IHM (p for nonlinearity>0.05). Sensitivity analysis demonstrated consistent findings regarding the magnitude and direction of the effects across different subgroups, indicating stable results. CONCLUSION:Elevated AIP was associated with higher IHM in individuals with sepsis.
PURPOSE:There are few national studies of waiting-list mortality for heart transplantation (HT) in Korea. Our multicenter study examined mortality on the HT waiting list and its associated risk factors. MATERIALS AND METHODS:We retrospectively analyzed 1101 consecutive patients who were wait-listed for HT between 2012 and 2017 in four centers. Time on the HT waiting list was defined as the time from initial wait-listing to delisting due to HT, death, or recovery. Subjects were censored at the time of transplantation or recovery. RESULTS:Of the whole cohort, 327 (29.7%) patients needed mechanical circulatory support (MCS) while on the waiting list, 314 (28.5%) were treated with extracorporeal membrane oxygenation (ECMO) as a bridge, and 13 (1.2%) were treated with a ventricular assist device (VAD) as a bridge. The waiting-list survival rate for ECMO-bridged patients was significantly lower than that for VAD-bridged or non-MCS-bridged patients. The waiting-list survival rate for patients with a status of 0 or 1 was significantly lower than for patients with a status of 2 or 3. Multivariate analysis revealed that the independent risk factors for waiting-list mortality were congenital heart disease and restrictive cardiomyopathy compared with dilated cardiomyopathy, status 0 compared with status 2 and 3, low hemoglobin, history of ventricular arrhythmia, high MELD-XI score, and ECMO bridging during the waiting period. CONCLUSION:This study showed that ECMO as a bridge to transplantation and status 0 were associated with significantly higher waiting-list mortality. Advanced end-organ damage at the time of listing was also found to be an independent risk factor for waiting-list mortality.
PURPOSE:In cardiovascular patients, precise risk stratification is crucial for minimizing exercise-related risks and optimizing treatment efficacy during cardiac rehabilitation. Existing classification criteria vary among organizations, necessitating a comprehensive assessment of the underlying conditions and medical tests. We developed machine-learning models to assist in standardizing risk classification for cardiopulmonary exercise training. MATERIALS AND METHODS:Using retrospective data from 1163 patients across three institutions, we developed an AI model to reproduce clinician-assigned risk stratification for exercise-related cardiovascular events. Data pre-processing involved excluding variables with numerous missing values and conducting multiple imputation using chained equations, resulting in 46 variables for analysis. Logistic regression, support vector machines, extreme gradient boosting (XGBoost), and random forests were employed, optimizing hyperparameters via a grid search. In feature selection, the least absolute shrinkage and selection operator (LASSO), with a range of regularization strengths, was applied to the L1 term to determine the optimal penalty value. RESULTS:Several machine-learning models were evaluated for their ability to replicate clinician-assigned cardiovascular risk-type classifications. XGBoost achieved the best performance with 76.72% accuracy and an area under the curve (AUC) of 88.76%. After applying LASSO-based feature selection, performance improved, with the best result (α=0.001) showing 79.31% accuracy and an AUC of 89.40%. Feature importance analysis identified ventilatory efficiency, maximal systolic blood pressure, and maximal oxygen consumption as key features for risk stratification based on aerobic exercise load test data, along with weight. CONCLUSION:A machine-learning model was established to categorize exercise-related risk types during cardiopulmonary rehabilitation, serving as a decision-support tool for the safe and effective training of cardiovascular patients.
PURPOSE:The incidence of invasive fungal infection (IFI) in pediatric liver transplant (LT) recipients in Korea has not been characterized; consequently, antifungal prophylaxis is not covered by the national health insurance. We aimed to determine the incidence and epidemiology of IFI. MATERIALS AND METHODS:We conducted a single-center, retrospective study of children (age ≤19 years) who underwent LT at Severance Children's Hospital, Republic of Korea (2012-2023). IFI was defined as fungal isolation from a sterile site (fluid/blood/tissue) with compatible clinical features. RESULTS:Of the 126 LT cases involving 115 children, 69.0% (n=87) were performed due to biliary atresia. Twenty-four IFI episodes in 20 recipients yielded a 90-day post-LT crude incidence of 19.0% and an incidence rate of 0.14 per 1000 patient-days. Invasive candidiasis predominated (95.4%) with Candida albicans (40.9%), C. parapsilosis (31.8%), and C. auris (9.1%). The most common clinical manifestations were peritonitis (54.2%) and fungemia (16.7%). Emergency LT [subdistribution hazard ratio (sHR), 3.97; 95% confidence interval (CI), 1.39-11.3; p=0.010] and reoperation or interventional procedures (sHR, 4.05; 95% CI, 1.53-10.7; p=0.005) were independently associated with IFI. One-year overall survival was significantly lower in the IFI group than in the non-IFI group (58.4% vs. 82.8%, p=0.037). CONCLUSION:IFIs impose a burden on pediatric LT recipients in Korea and are associated with poorer survival. These data support the consideration of universal antifungal prophylaxis. However, multicenter studies are warranted to validate these findings and define the prophylaxis regimen.
PURPOSE:Both South and North Korea are identified by the World Health Organization (WHO) as having the potential to eliminate malaria. However, the increase in malaria cases during the COVID-19 response has prompted an investigation into the reasons behind this unexpected rise. MATERIALS AND METHODS:Epidemiological, entomological, and meteorological data from national and international surveillance systems were analyzed. Using the WHO's analytical framework, which can be effectively applied during the elimination phase, we assessed drivers of malaria resurgence in North and South Korea. RESULTS:In North Korea, border closures caused a severe shortage of rapid diagnostic test kits, reducing malaria testing by over 90% by 2021. Microscopy use increased but remained insufficient, while confirmed cases rose from 1819 in 2020 to 3160 in 2023. In South Korea, public health resources were diverted to COVID-19, weakening malaria control activities. The response time for epidemiological investigations increased from 1.3 days in 2019 to 4.3 days in 2021. Relapse cases, which were zero in 2021, increased to eight in 2023, and confirmed cases rose from 274 in 2021 to 673 in 2023. There were no meaningful changes in the drivers of malaria resurgence, including importation of malaria parasites or human susceptibility. With no notable shifts in vector density, variations in temperature and precipitation did not explain the rise in either country, and there were no reports of emerging resistance to insecticides or antimalarial drugs. CONCLUSION:These COVID-19-related disruptions, including supply chain failures and the diversion of human resources, impaired malaria control in countries on the Korean Peninsula.
PURPOSE:Thiopurines remain an effective treatment for maintaining remission in inflammatory bowel disease (IBD), yet their use is often limited by thiopurine-induced leukopenia (TIL), a potentially life-threatening adverse effect. We aimed to investigate pharmacogenetic associations with TIL in patients with IBD. MATERIALS AND METHODS:We retrospectively analyzed 218 thiopurine-treated patients from two independent IBD cohorts with whole-exome sequencing. Pharmacogenetic subgroups were defined using Clinical Pharmacogenetics Implementation Consortium star allele-based molecular phenotypes or genotype-based classifications. Genetic associations with TIL, defined as white blood cell count ≤3000/µL, were analyzed using Andersen-Gill models adjusted for clinical covariates. RESULTS:NUDT15 intermediate metabolizers [hazard ratio (HR) 5.21, p<0.001], poor metabolizers (HR 6.42, p<0.001), and IL6 heterozygotes (HR 4.35, p<0.001) showed reproducible, independent associations with increased TIL risk. Incorporating IL6 into the traditional TPMT/NUDT15 model significantly improved sensitivity (p=0.0156) and negative predictive value (p=0.046). CONCLUSION:Our findings confirm the central role of NUDT15 in TIL susceptibility and identify IL6 as a novel, independent contributor in Korean patients with IBD. Incorporating IL6 into existing pre-emptive pharmacogenetic testing may support safer thiopurine therapy.
PURPOSE:Traumatic cervical spinal cord injury (TCSCI) frequently necessitates mechanical ventilation, and early tracheostomy has been shown to reduce complications in patients requiring prolonged ventilation. This study aimed to develop a machine learning model to predict the need for tracheostomy in TCSCI patients, utilizing early clinical data to improve patient outcomes. MATERIALS AND METHODS:This study was conducted using data from 2017 to 2024, obtained from the single institution database. A total of 267 TCSCI patients were included, of whom 49 underwent tracheostomy. Variables selected for the model included demographics, comorbidities, injury level, medical research council motor grading, Glasgow Coma Scale (GCS), and treatment details. This study also implemented SHapley Additive exPlanations analysis to interpret the predictive model and identify significant risk predictors contributing to the outcomes. RESULTS:The CatBoost model outperformed other models, achieving the highest performance metrics. The model that incorporated GCS and the American Spinal Injury Association (ASIA) impairment scale (AIS) yielded the highest area under the curve (AUC) score. Following variable selection, the CatBoost model, utilizing age, GCS, AIS score, surgery, and injury levels (C3/4, C4/5, and C2/3), achieved an AUC score of 0.8166 and an accuracy of 0.8652. CONCLUSION:Our machine learning model effectively predicted the need for tracheostomy in TCSCI patients. Important predictors were age, GCS, AIS score, cervical surgery, and injury level. This model may improve outcomes for patients requiring tracheostomy.
Purpose: Optimizing antibiotic use is essential for overcoming antibiotic resistance. In this study, we identified strategies for improving antibiotic use for urinary tract infections (UTIs). Materials and Methods: This retrospective study was conducted between July 2022 and June 2023 to evaluate the effect of quarterly qualitative assessments of antibiotic prescriptions for inpatients with UTIs. Appropriateness was evaluated based on antibiotic selection, dosage, administration route, and duration, and feedback was shared with medical staff to enhance prescription practices. Evaluations were performed at 3-month intervals, with the first quarter as baseline. Changes in appropriateness were analyzed using linear regression. Results: Overall, 1473 antibiotic prescriptions from 638 patients were analyzed. Third-generation cephalosporins were the most prescribed class. For lower UTIs, significant improvements were observed in treatment duration (40.8%, p=0.050), administration route (22.9%, p=0.039), and dosage (10.5%, p<0.001), thereby increasing the proportion of appropriate prescriptions from 28.6% to 68.0% (p=0.010). For upper UTIs, significant improvements were observed in dosage (6.7%, p=0.032) and duration (20.2%, p=0.032), with the proportion of appropriate prescriptions increasing from 55.9% to 79.0% (p=0.043). Overall, qualitative assessments and feedback improved prescribing appropriateness from 47.1% to 75.5% (p=0.013) without adverse effects on mortality or length of stay. Conclusion: Regular qualitative assessments of antibiotic prescriptions significantly improved prescriptions for UTIs without negative outcomes. These findings support the role of qualitative assessments in antibiotic stewardship; however, further studies are required to evaluate their long-term impact and broader applicability.
PURPOSE:Conflicting results have been published for the association between egg consumption and the risk of diabetes mellitus (DM). The present study was to assess the risk of incident DM in relation to egg consumption. MATERIALS AND METHODS:Study participants were 91005 non-diabetic adults who periodically received health check-ups. They were categorized into six groups by egg consumption (<1/week, ≥1 and <3/week, ≥3 and <7/week, ≥1 and <2/day, ≥2 and <3/day, ≥3/day) and followed for incident DM during median 6.9 years. The Cox proportional hazard model was used to longitudinally evaluate the hazard ratio (HR) and 95% confidence interval (CI) for incident DM according to the six groups of egg consumption [adjusted HR (95% CI)]. Subgroup analysis was conducted by sex and age (<45 years or ≥45 years). The risk of DM conferred by a daily increment of one egg consumption was also analyzed. RESULTS:In all participants, egg consumption ≥1/week was not significantly associated with an increased risk of DM, compared with the reference [<1/week: reference, ≥1 and <3/week: 1.01 (0.91-1.12), ≥3 and <7/week: 0.96 (0.87-1.06), ≥1 and <2/day: 1.06 (0.95-1.19), ≥2 and <3/day: 1.04 (0.88-1.22), ≥3/day: 1.29 (0.98-1.70)]. This pattern of relationship was similarly observed across all subgroups. A daily increment of one egg consumption showed only a marginal association with the risk of DM in men [1.07 (1.02-1.13)] and younger individuals [1.07 (1.02-1.13)]. CONCLUSION:Egg consumption appears to have little association with the risk of incident DM.
PURPOSE:Hepatopancreatoduodenectomy (HPD) is a potentially curative option for selected patients with advanced cholangiocarcinoma or gallbladder cancer, but it is associated with substantial morbidity. In this study, we evaluated short- and long-term outcomes after HPD and identified prognostic factors. MATERIALS AND METHODS:Patients who underwent HPD at the Seoul National University Hospital, South Korea from 2000 to 2023 were included. Prospectively collected data on patient and tumor characteristics, perioperative and survival outcomes were analyzed. Prognostic factors were assessed using Cox proportional hazards models, and risk factors for disease-free survival (DFS) <12 months were explored using logistic regression. RESULTS:The 30-day mortality rate was 0%, the 90-day mortality rate was 6%, and major complications (Clavien-Dindo grade ≥III) occurred in 58% of patients. The median overall survival (OS) was 21 months, and median DFS was 12 months. The 5-year survival rate was 16%. In multivariable analysis, angiolymphatic invasion was an independent prognostic factor for both OS and DFS. Adjuvant treatment was associated with an improved OS. For DFS <12 months, preoperative carbohydrate antigen (CA) 19-9 level and angiolymphatic invasion were independent risk factors. CONCLUSION:Despite high morbidity, HPD can provide meaningful long-term survival in carefully selected patients. Preoperative CA 19-9 and angiolymphatic invasion may aid risk stratification. Associations between adjuvant treatment and survival outcomes should be interpreted cautiously and warrant confirmation in larger, prospective multicenter studies.
PURPOSE:Reduced cardiorespiratory fitness (CRF) is associated with hospitalization risk in heart failure (HF). Traditional peak oxygen uptake (VO₂) scaling uses total body weight (TBW), potentially underestimating CRF due to adiposity. The prognostic value of fat-free mass (FFM)-adjusted peak VO₂ remains unclear, particularly in Asian populations. MATERIALS AND METHODS:A retrospective cohort study included HF patients who underwent cardiopulmonary exercise testing and bioelectrical impedance analysis. Two peak VO₂ cutoffs-14 mL/TBW kg/min and 19 mL/FFM kg/min-were applied to predict all-cause and HF-specific 1-year hospitalization. The prognostic performance was assessed using Cox proportional hazards models adjusted for the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) risk score. Likelihood ratio tests were conducted to evaluate the incremental value of adding the FFM-adjusted cutoff. RESULTS:A total of 83 patients (mean age, 60.0 years; 14 female; 37 obese) were analyzed. Both TBW- and FFM-adjusted cutoffs were significantly associated with increased risk of all-cause hospitalization [hazard ratio (HR)=3.86, 95% confidence interval (CI), 1.40-10.63 vs. HR=5.02, 95% CI, 1.98-12.72] and HF-specific hospitalization (HR=4.00, 95% CI, 1.00-16.05 vs. HR=7.26, 95% CI, 1.92-27.46). Adding the FFM-adjusted cutoff significantly improved model fit when added to a model with the TBW-adjusted cutoff (p<0.05). The difference in c-indices between the two cutoffs after bootstrapping was not statistically significant. CONCLUSION:The FFM-adjusted cutoff can complement the traditional TBW-adjusted cutoff by correcting the confounding bias of excessive adiposity or low muscle mass, providing incremental prognostic value for risk stratification in Asian patients with HF.
