Yonsei Medical Journal最新文献

Purpose: This study investigates the risk of developing dementia in breast cancer patients diagnosed with depression within 2 years of their cancer diagnosis, comparing those taking antidepressants alone versus those taking both antidepressants and benzodiazepines.

Materials and methods: Utilizing data from the Korean National Health Insurance claims database collected over 14 years, we included a cohort of 197917 breast cancer patients. Among them, 19170 were diagnosed with depression within 2 years of their cancer diagnosis. After matching for comorbidities and age, we included 1376 patients in each group for 1:1 matching, and 1326 patients taking only antidepressants and 5304 patients taking both antidepressants and benzodiazepines for 1:4 matching.

Results: In the initial 1:1 matched analysis, no significant difference in dementia incidence was observed between the group taking only antidepressants and the group taking both antidepressants and benzodiazepines. However, further detailed analysis revealed that patients taking higher doses or using benzodiazepines for longer durations had an increased risk of dementia. In the 1:4 matched analysis, the group taking both medications exhibited a significantly higher incidence of dementia compared to those taking only antidepressants (hazard ratio, 1.807; 95% confidence interval, 1.263-2.583; p=0.0012).

Conclusion: This study underscores the importance of cautious benzodiazepine use in breast cancer patients diagnosed with depression, given its potential to significantly increase dementia risk.

Purpose: The morphological and functional characteristics and clinical significance of the left atrial appendage (LAA) are well established in patients with non-valvular atrial fibrillation (AF). However, data on the LAA characteristics in patients with mitral valve (MV) disease are limited. This study aimed to identify the LAA characteristics in AF patients with severe MV disease.

Materials and methods: A total of 506 AF patients who underwent cardiac computed tomography (CT) as preoperative evaluations for MV surgery were retrospectively analyzed. The prevalences of different LAA morphologies (cactus, cauliflower, windsock, chicken wing), LAA ostium diameter, LAA volume, and LAA flow stasis or thrombus were assessed. The LAA variables were compared according to the predominant MV dysfunction.

Results: The most common LAA morphology was cactus (n=211, 41.7%), followed by cauliflower (n=143, 28.3%), windsock (n=90, 17.8%), and chicken wing (n=60, 11.9%). The average LAA ostium maximal diameter and LAA volume were 35.3±8.0 mm and 22.1±15.1 mL, respectively. LAA stasis was found in 215 patients (42.5%) and LAA thrombus in 93 patients (18.4%). Patients with mitral stenosis predominance showed significantly smaller LAA volume compared to those with mitral regurgitation predominance (17.8±11.7 mL vs. 26.9±16.8 mL, p<0.001). However, LAA flow stasis [190 (71.7%) vs. 25 (10.4%), p<0.001] and thrombus [89 (33.6%) vs. 4 (1.7%), p<0.001] were remarkably prevalent in these patients.

Conclusion: Due to advanced LAA remodeling in AF patients with severe MV disease, the morphologic distribution of LAA types differs from that established in patients without MV disease.

Purpose: This study investigated the risk factors for heterotopic ossification (HO) in patients with traumatic brain injury (TBI).

Materials and methods: This was a retrospective study using the Korean National Health Insurance Service (KNHIS) database and included as many relevant factors as possible. Data were collected from the KNHIS cohort, a nationwide cohort covering the entire Korean population. Patients diagnosed with TBI from 2004 to 2018 were included. TBI was defined as individuals who 1) had been hospitalized, 2) were diagnosed with TBI under ICD-10 code S06, and 3) underwent brain imaging within 1 week before or after diagnosis. Among 637315 adult patients, 1909 (0.30%) developed HO. This study aimed to clarify the relationship between HO and various factors, including demographic and medical history, medication history, complications, and accompanying injuries in TBI patients.

Results: Among TBI patients, HO was more common in female and peaked in patients in their 50s, with a lower incidence in their 70s. Preobesity and obesity were significant risk factors, while smokers had a reduced risk. HO was more common in patients with rheumatic diseases and medical comorbidities, and those who had undergone tracheostomies. The use of antiseptics was associated with an increased risk, whereas anticonvulsants, antithrombotics, steroids, and non-steroidal anti-inflammatory drugs were associated with lower risks. HO was more common in registered patients with disability due to brain lesion.

Conclusion: In conclusion, middle-age range, female sex, obesity, comorbidities, injury severity, systemic inflammation, and bony metabolism-affecting medications appear to increase the risk of HO in patients with TBI.

Purpose: In 2021, 23.9% of Korean workers were self-employed, with the majority working in small enterprises. As the unemployment rate increased after the COVID-19 outbreak, the number of small enterprises increased annually. Although interest in the health of small enterprises is increasing, insufficient research has been conducted. This study compared the well-being levels of self-employed and salaried workers in small enterprises and identified the factors affecting them.

Materials and methods: Data were obtained from the 6th Korean Working Conditions Survey (KWCS). Adults aged ≥40 years with Korean nationality working in small enterprises (less than 50 employees) in the service/sales private sector were included. Of the 50538 respondents, 2511 (942 male, 1569 female) were included. The general characteristics of the participants were identified. The well-being levels were analyzed using subjective health items and the WHO-5 well-being index, included in the 6th KWCS. Statistical analysis was performed using logistic regression analysis.

Results: People without chronic illnesses and salaried workers reported higher levels of well-being. Additionally, people who were younger, more educated, and had higher monthly incomes reported higher levels of well-being. Salaried workers showed higher levels of well-being compared to self-employed workers when stratified by the number of employees, working hours per week, and chronic illness.

Conclusion: Self-employed workers had significantly lower levels of well-being compared to salaried workers in small enterprises. Therefore, future health policies should include measures to improve the well-being or mental health of self-employed workers. Given past trends, involuntary self-employment tends to rise during recessions, making policy interventions crucial during such periods.

Purpose: This study aimed to examine 1) the relationship between domain-specific physical activity (PA) and the prevalence of chronic kidney disease (CKD), as well as 2) the association between meeting PA and resistance exercise (RE) guideline and CKD prevalence in individuals with diabetes.

Materials and methods: The study analyzed data from the 2019-2021 Korea National Health and Nutrition Examination Survey, a cross-sectional study that included 22559 participants. From this group, 2381 adults with diabetes were selected. CKD was defined as an estimated glomerular filtration rate <60 mL/min/1.73m², a urinary albumin-to-creatinine ratio ≥30 mg/g, or a physician's diagnosis. Logistic regression models were used to assess the association between compliance with the WHO's PA guidelines and CKD prevalence, with further stratification according to known CKD risk factors.

Results: Individuals with diabetes who met the PA guidelines through leisure physical activity (LPA) and RE were significantly inversely associated with the odds of CKD [odds ratio (OR): 0.55, 95% confidence interval (CI) 0.34-0.89]. This inverse association was pronounced in individuals with lower body mass index (OR: 0.31, 95% CI 0.15-0.65). However, individuals who met the recommended amount of PA through work-related physical activity and RE guideline were not significantly associated with the odds of CKD (OR: 1.46, 95% CI 0.44-4.82).

Conclusion: Meeting PA and RE guidelines are associated with reduced prevalence of CKD in individuals with diabetes. These findings underscore the potential benefits of LPA and RE in the prevention of CKD in individuals with diabetes.

Purpose: Insulin resistance (IR) is a condition closely associated with cardiovascular risk factors and metabolic dysfunction-associated steatotic liver disease (MASLD) is emerging as a significant IR-related complication. We aimed to develop a predictive model for IR in youths and implicate this model for MASLD.

Materials and methods: A total of 1588 youths from the population-based data were included in the training set. For the test sets, 121 participants were included for IR and 50 for MASLD from real-world clinic data. Logistic regression analysis, random forest, extreme gradient boosting (XGBoost), light gradient boosting machine (GBM), and deep neural network (DNN) were used to develop the models. A nomogram scoring system was constructed based on a model used to predict the probability of IR and MASLD.

Results: After stepwise selection, age, body mass index (BMI) standard deviation score (SDS), waist circumference (WC), systolic blood pressure, HbA1c, high-density lipoprotein cholesterol, triglyceride, and alanine aminotransferase levels were included in the model. A nomogram scoring system was constructed based on a multivariable logistic regression model. The areas under the curves (AUCs) of the models for IR prediction in external validation were 0.75 (logistic regression), 0.78 (random forest), 0.72 (XGBoost), 0.71 (light GBM), and 0.71 (DNN). For MASLD prediction, the AUCs were 0.93 (logistic regression), 0.95 (random forest), 0.90 (XGBoost), 0.91 (light GBM), and 0.85 (DNN). BMI SDS and WC SDS were the most important contributors to IR prediction in all models.

Conclusion: The Pediatric Insulin Resistance Assessment Score is a novel scoring system for predicting IR and MASLD in youths.

Purpose: To introduce heatmap-in-heatmap (HIH)-based model for automated diagnosis of foot deformities using weight-bearing foot radiographs, aiming to address the labor-intensive and variable nature of manual diagnosis.

Materials and methods: From January 2004 to September 2022, a dual-center retrospective study was conducted. In the first center, 1561 anterior-posterior (AP) and 1536 lateral images from 806 patients were used for model training, while 374 AP and 373 lateral images from 196 patients were allocated to the validation set. For external validation at the second center, 527 AP and 529 lateral images from 270 patients were allocated. Five deformities were diagnosed using four and three angles between the predicted landmarks in the AP and lateral images, respectively. The results were compared with those of the baseline model (FlatNet).

Results: The HIH model demonstrated robust performance in diagnosing multiple foot deformities. On the test set, it outperformed FlatNet with higher accuracy (FlatNet vs. HIH: 78.9% vs. 85.1%), sensitivity (78.9% vs. 84.1%), specificity (79.0% vs. 85.9%), positive predictive value (77.3% vs. 84.4%), and negative predictive value (80.5% vs. 85.7%). Additionally, HIH exhibited significantly lower absolute pixel and angle errors, lower normalized mean errors, higher successful detection rate, faster training and inference speeds, and fewer parameters.

Conclusion: The HIH model showed robust performance in diagnosing multiple foot deformities with high efficacy in internal and external validation. Our approach is expected to be effective for various tasks using landmarks in medical imaging.

Purpose: A high index of suspicion is crucial for early anastomotic leak (AL) diagnosis, enabling timely intervention and conservative management. Although recent studies have highlighted the potential of drain amylase (dAmy) as a predictive marker for AL, its application in colorectal surgery remains underexplored.

Materials and methods: A retrospective study was conducted among patients who underwent colorectal resection between March 2020 and November 2023. A total of 299 patients with at least one dAmy and serum amylase (sAmy) measurement between postoperative days 1 to 5 were included, after excluding patients with fecal diversion.

Results: Multiple logistic regression identified the drain/serum amylase ratio (d/s Amy) as an independent predictor of AL (odds ratio 1.032, 95% confidence interval 1.009-1.056; p=0.007). The receiver operating characteristic curve demonstrated significant diagnostic ability for AL [area under the curve (AUC)=0.691], with a cut-off value of 2.54, a sensitivity of 48.4%, and a specificity of 94.2%. Patients with d/s Amy ≥2.54 had a significantly higher incidence of AL with a faster diagnosis compared to conventional methods (3.5 days vs. 5 days, p=0.006). In patients who underwent ileal anastomosis, the d/s Amy ratio had an AUC of 0.936, with a sensitivity of 87.5% and a specificity of 96.6%.

Conclusion: The postoperative d/s Amy ratio is valuable for early AL detection in patients undergoing colorectal surgery, particularly in those with ileal anastomosis. This simple and noninvasive test can aid in timely diagnosis, offering earlier intervention compared with conventional methods.

Purpose: Children undergoing mechanical ventilation (MV) in the pediatric intensive care unit (PICU) require effective sedation to reduce anxiety and discomfort. Dexmedetomidine, an α2-receptor agonist, presents as a viable sedative alternative. However, its clinical outcomes for critically ill, mechanically ventilated children remain to be fully established. We performed a systematic review and meta-analysis of randomized controlled trials (RCTs) to assess the clinical outcomes and adverse effects of dexmedetomidine in such patients.

Materials and methods: A systematic search was conducted up to April 2024. RCTs that compare dexmedetomidine with other sedatives in mechanically ventilated children were included. This analysis focused on both the clinical and safety outcomes through meta-analysis.

Results: Included in the analysis were eight trials, involving a total of 387 mechanically ventilated children. Compared to other sedatives, dexmedetomidine reduced the duration of MV [mean difference -3.54 hours; 95% confidence interval (CI), -6.49 to -0.59], particularly in post-operative patients and when compared to fentanyl. However, dexmedetomidine did not significantly impact the length of ICU stay, duration of sedation, or the necessity for additional sedatives. Dexmedetomidine was associated with a significantly increased risk of bradycardia [odds ratio (OR) 6.14; 95% CI, 2.20 to 17.12] and hypotension (OR 8.14; 95% CI, 1.37 to 48.31) compared to other sedatives.

Conclusion: Although dexmedetomidine notably diminished the duration of MV, the potential for adverse effects necessitates further investigation. Large RCTs are needed to validate our findings and refine sedation management in mechanically ventilated children in PICU.

Purpose: Black-blood (BB) magnetic resonance images (MRI) offer superior image contrast for the detection and segmentation of brain metastases (BMs). This study investigated the efficacy and accuracy of deep learning (DL) architectures and post-processing for BMs detection and segmentation with BB images.

Materials and methods: The BB images of 50 patients were collect to train (40) and test (10) the DL model. To ensure consistency, we implemented piecewise linear histogram matching for intensity normalization and resampling. Modified U-Net, including combination with generative adversarial network (GAN), was applied to enhance the segmentation performance. The U-Net-based networks generated bounding boxes indicating regions of interest, which were then processed in a post-processing using the Segment Anything Model (SAM). We quantitatively assessed the three U-Net-based models and their post-processed counterparts in terms of lesion-wise sensitivity (LWS), patient-wise dice similarity coefficient (DSC), and average false-positive rate (FPR).

Results: The modified U-Net with GAN yielded a patient-wise DSC of 0.853 and a LWS of 89.19%, which outperformed the standard U-Net (patient-wise DSC of 0.815) and modified U-Net only (patient-wise DSC of 0.846). Combining GAN architecture with modified U-Net also reduced the FPR, less than 1 on average. Post-processing with SAM further did not affect LWS and FPR, but effectively enhanced the patient-wise DSC by 2%-3% for the U-Net-based models.

Conclusion: The modifications to standard U-Net notably improves the detection and segmentation of BMs in BB images, and applying SAM as post-processing can further enhance the precision of segmentation results.