medRxiv : serveur de preprint pour les sciences médicales

Background Relationships between low estimated glomerular filtration rate (eGFR) and albuminuria with poorer cognitive performance in patients with ischemic stroke are less clear. Our aim was to retrospectively ascertain the associations between these renal measures and cognitive performance in patients with ischemic stroke. Methods Retrospective analysis was performed on 608 patients with acute ischemic stroke. Urine albumin-creatinine ratio (UACR) and eGFR were obtained from inpatient medical records. Global cognitive function via mini-mental state exam (MMSE) and Montreal Cognitive Assessment (MoCA) was determined one month after hospital discharge. The relationship between renal measures and cognitive performance was assessed using univariate and multiple linear regression analyses. Results Patients had an average age of 66.6{+/-}4.1 years, and 48% were females. Average eGFR and UACR were 88.4{+/-}12.9 ml/min/1.73m2 and 83.6{+/-}314.2 mg/g, respectively. The number of patients with an eGFR of >=90, 60 to 89, and <60 ml/min/1.73 m2 were (371, 61%), (207, 34%) and (30, 5%) respectively. The proportions with a UACR <30 mg/g, 30-300mg/g and >300 mg/g were 56%, 39% and 5%. Multivariate adjusted models showed that eGFR was independently associated with MMSE ({beta}= -0.4; 95% CI= -0.5,-0.4; p <0.001) and MoCA ({beta} = -0.6; 95% CI= -0.7,-0.5; p <0.001). However, the correlations between UACR and MMSE and MoCA were statistically non-significant. Conclusion In patients with ischemic stroke, reduced eGFR but not albuminuria was associated with lower cognitive performance. These results show that the eGFR decline could be an effective indicator of cognitive impairment after a stroke. Therefore, regular monitoring and early detection of mild renal dysfunction in patients with acute ischemic stroke might be needed.

Quantification of the myelin content of the white matter is important for studying demyelination in neurodegenerative diseases such as Multiple Sclerosis (MS), particularly for longitudinal monitoring. A novel noninvasive MRI method, called Microstructure-Informed Myelin Mapping (MIMM), is developed to quantify the myelin volume fraction (MVF) by utilizing a multi gradient echo sequence (mGRE) and a detailed biophysical model of tissue microstructure. Myelin is modeled as anisotropic negative susceptibility source based on the Hollow Cylindrical Fiber Model (HCFM), and iron as isotropic positive susceptibility source in the extracellular region. Voxels with a range of biophysical parameters are simulated to create a dictionary of MR echo time magnitude signals and total susceptibility values. MRI signals measured using a mGRE sequence are then matched voxel-by-voxel to the created dictionary to obtain the spatial distributions of myelin and iron. Three different MIMM versions are presented to deal with the fiber orientation dependent susceptibility effects of the myelin sheaths: a basic variation, which assumes fiber orientation is an unknown to fit, two orientation informed variations, which assume the fiber orientation distribution is available either from a separate diffusion tensor imaging (DTI) acquisition or from a DTI atlas based fiber orientation map. While all showed a significant linear correlation with the reference method based on T2-relaxometry (p < 0.0001), DTI orientation informed and atlas orientation informed variations reduced overestimation at white matter tracts compared to the basic variation. Finally, the implications and usefulness of attaining an additional iron susceptibility distribution map are discussed.

Introduction: Thyroid cancer is a common endocrine malignancy with a rising incidence. However, to improve patient outcomes, it is essential to understand the molecular mechanisms driving its progression and metastasis, and the metabolomics can unveil alterations in metabolic pathways that contribute to thyroid cancer. Objective: To identify the metabolic pathways altered in thyroid cancer progression and metastasis. Methods: Multiple bioinformatics tools were employed in the research. Gene expression data was obtained from the Gene Expression Omnibus and The Cancer Genome Atlas. Functional assessment of the expressed genes in thyroid cancer was performed using gene set enrichment analysis. The Kyoto Encyclopedia of Genes and Genomes database was utilized to identify the metabolic pathway involved in thyroid cancer progression and metastasis. A computational algorithm was developed to estimate the activity levels of the identified metabolic pathways and construct a signaling pathway. Results: The altered metabolic pathways in thyroid cancer progression and metastasis were identified based on the following algorithm: activation of growth factor signaling, activation of multiple signaling pathways, regulation by transcription factors, dysregulation of downstream signaling cascades, changes in cellular metabolism, tumor progression, invasion and metastasis, and feedback regulation. Conclusion: By applying a comprehensive algorithm, we were able to uncover key molecular events driving the aggressive behavior of thyroid cancer. These findings provide insights into the underlying mechanisms of thyroid cancer progression and metastasis. Keywords: Thyroid cancer, metabolic pathways, metastasis.

Introduction Sepsis is a major cause of morbidity and mortality worldwide. In the updated, 2016 Sepsis-3 criteria, sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, where organ dysfunction can be represented by an increase in the Sequential Organ Failure Assessment (SOFA) score of 2 points or more. We sought to apply the Sepsis-3 criteria the characterise the septic cohort in the Amsterdam University Medical Centres database (AmsterdamUMCdb). Methods We examined adult intensive care unit (ICU) admissions in the Amsterdam UMCdb, which contains de-identified data for patients admitted to a mixed surgical-medical ICU at a tertiary academic medical centre in the Netherlands. We operationalised the Sepsis-3 criteria, defining organ dysfunction as an increase in the SOFA score of 2 points or more, while infection was defined as a new course of antibiotics or an escalation in antibiotic therapy, with at least one antibiotic given intravenously. Patients with sepsis were determined to be in septic shock if they additionally required the use of vasopressors and had a lactate level >2 mmol/L. Results We identified 18,221 ICU admissions from 16,408 patients in our cohort. There were 6,371 unique sepsis episodes, of which 30.1% met the criteria for septic shock. A total of 4,958/6,371 sepsis (77.8%) episodes occurred on ICU admission. Forty-eight percent of emergency medical admissions and 37.0% of emergency surgical admissions were for sepsis. Overall, there was a 12.5% ICU mortality rate; patients with septic shock had a higher ICU mortality rate (38.5%) than those without shock (11.3%). Conclusions We successfully operationalised the Sepsis-3 criteria to the Amsterdam UMCdb, allowing the characterization and comparison of sepsis epidemiology across different centres.

Background: Despite reporting very few mpox cases early in 2023, Mainland China observed a surge with over 500 cases during the summer. Amid ambiguous prevention strategies and stigma surrounding mpox, the epidemic escalated silently. This study aims to quantify the epidemic scale and assess the transmission dynamics by estimating the effective reproduction number (R_e) during its early phase. Methods: Publicly available data sources were aggregated to retrieve daily mpox incidences in Mainland China. The R_e value was estimated employing an exponential growth model. This estimate was compared with R_e values from 16 other national outbreaks in 2022 selected based on a cumulative incidence exceeding 700 symptomatic cases by the end of that year. A meta-analytic approach was adopted to compute the pooled mean of R_e for selected outbreaks. Results: The estimated mean R_e for Mainland China was 1.76 (95% credible interval: 1.51-2.06), suggesting a case doubling time of approximately two weeks. R_e estimates from selected outbreaks of 2022 ranged from 1.17 for Portugal to 2.88 for Columbia. The computed pooled mean stood at 1.66 (1.40-1.92), aligning closely with the R_e for Mainland China. Limitations: The study does not distinguish between indigenous and imported infections, which might impact the R_e estimates. Furthermore, the analysis is limited by reporting delays and underascertainment, especially during the early phase of the epidemic.

From 2003 to 25 February 2023, the avian influenza (H5N1) virus was confirmed in 59 human infections, including 39 deaths (~66% case-fatality rate) reported in 13 of 25 provinces in Cambodia. We aimed to assess current knowledge, attitude, and practice toward changes in poultry handling behaviors, poultry consumption, and poultry mortality reporting among rural villagers in areas affected by Avian influenza (H5N1) in Cambodia. A cross-section survey was conducted in August 2023. There were 208 participants residing in Prey Veng province who were invited to be interviewed face-to-face. Descriptive statistics were performed using STATA V17. The participants' average age was 55 years old (SD=13.3 years), 78.4% were female, 59% had completed primary school, 56.7% were farmers, 68.3% raised chickens in their backyards, and 10.2% raised ducks, 23% of participants cooked sick or dead birds for their families, 32% knowing information about avian influenza virus was a lower proportion from healthcare providers, 10.6% from village health support groups were, and 2% from village animal health workers were only, 49% have been reported poultry illness and deaths to local authorities. The avian influenza epidemic in Cambodia is a genuine threat to animals and a possible concern to humans. To prevent and control this, we strongly advise everyone who works with poultry or wild game birds always to be prepared to follow appropriate hygiene standards and to cook poultry meat properly.

Background The impact of migration on HIV risk among non-migrating household members is poorly understood. We measured HIV incidence among non-migrants living in households with and without migrants in Uganda. Methods We used four survey rounds of data collected from July 2011-May 2018 from non-migrant participants aged 15-49 years in the Rakai Community Cohort Study, an open, population-based cohort. Non-migrants were individuals with no evidence of migration between surveys or at the prior survey. The primary exposure, household migration, was assessed using census data and defined as 1 or more household members migrating in or out of the house from another community between surveys (~18 months). Incident HIV cases tested positive following a negative result at the preceding visit. Incidence rate ratios (IRR) with 95% confidence intervals were estimated using Poisson regression with generalized estimating equations and robust standard errors. Analyses were stratified by gender, migration into or out of the household, and the relationship between non-migrants and migrants (i.e., any household migration, spouse, child). Findings Overall, 11,318 non-migrants (5,674 women) were followed for 37,320 person-years. 28% (6,059/21,370) of non-migrant person-visits had recent migration into or out of the household, and 240 HIV incident cases were identified in non-migrating household members. Overall, non-migrants in migrant households were not at greater risk of acquiring HIV. However, HIV incidence among men was significantly higher when the spouse had recently migrated in (adjIRR:2.12;95%CI:1.05-4.27) or out (adjIRR:4.01;95%CI:2.16-7.44) compared to men with no spousal migration. Women with in- and out-migrant spouses also had higher HIV incidence, but results were not statistically significant. Interpretation HIV incidence is higher among non-migrating persons with migrant spouses, especially men. Targeted HIV testing and prevention interventions such as pre-exposure prophylaxis could be considered for those with migrant spouses. Funding National Institutes of Health, US Centers for Disease Control and Prevention

Background: Clinical guidelines recommend incorporating non-cognitive markers like mild behavioral impairment (MBI) and sleep disturbance (SD) into dementia screening to improve detection. Objective: We investigated the longitudinal associations between MBI, SD, and incident dementia. Methods: Participant data were from the National Alzheimer's Coordinating Center in the United States. MBI was derived from the Neuropsychiatric Inventory Questionnaire (NPI-Q) using a published algorithm. SD was determined using the NPI-Q nighttime behaviors item. Cox proportional hazard regressions with time-dependant variables for MBI, SD, and cognitive diagnosis were used to model associations between baseline 1) MBI and incident SD (n=11277); 2) SD and incident MBI (n=10535); 3) MBI with concurrent SD and incident dementia (n=13544); and 4) MBI without concurrent SD and incident dementia (n=11921). Models were adjusted for first-visit age, sex, education, cognitive diagnosis, race, and for multiple comparisons using the Benjamini-Hochberg method. Results: The rate of developing SD was 3.1-fold higher in older adults with MBI at baseline compared to those without MBI (95%CI: 2.8-3.3). The rate of developing MBI was 1.5-fold higher in older adults with baseline SD than those without SD (95%CI: 1.3-1.8). The rate of developing dementia was 2.2-fold greater in older adults with both MBI and SD, as opposed to SD alone (95%CI:1.9-2.6). Conclusions: There is a bidirectional relationship between MBI and SD. Older adults with SD develop dementia at higher rates when co-occurring with MBI. Future studies should explore the mechanisms underlying these relationships, and dementia screening may be improved by assessing for both MBI and SD.

Maternal smoking has been linked to adverse health outcomes in newborns but the extent to which it impacts newborn health has not been quantified through an aggregated cord blood DNA methylation (DNAm) score. Here we propose the first cord blood DNAm score to capture the epigenetic signature of maternal smoking. We first examined the association between individual CpGs and cigarette smoking during pregnancy in two White European birth cohorts (n = 763), and smoking exposure in the White European and the South Asian birth cohorts (n = 887). Several previously reported genes for maternal smoking were supported, with the strongest and most consistent signal from the GFI1 gene (ps < 5x10-5). Leveraging previously reported CpGs for maternal smoking, we constructed a cord blood epigenetic score of maternal smoking that was internally validated in one of the European-origin cohorts. This score was then tested for association with smoking status, secondary smoking exposure during pregnancy, and health outcomes in offspring measured after birth. The epigenetic maternal smoking score was strongly associated with smoking status during pregnancy (p=1.04x10-23) and self-reported smoking exposure (p=6.56x10-8) in White Europeans, but not with self-reported exposure (p>0.05) in South Asians. The same score was consistently associated with smaller birth size (p =1.19x10-6) and lower birth weight (p=8.79x10-7) in the combined South Asian and White European cohorts. This cord blood epigenetic score can help identify babies exposed to maternal smoking and assess its long-term impact on growth. Notably, even minimal smoking exposure in South Asian mothers, without active smoking, still showed a DNAm response to small body size and low weight in newborns.

Musculoskeletal conditions have been ranked as the leading cause of disability worldwide. Low back pain is the single largest contributor in 160 countries including Nepal. Nurses working in hospitals in Nepal are overworked and at risk of developing musculoskeletal disorders. The study aimed to determine the 12-month periods and point prevalence of work-related musculoskeletal disorders (WMSDs); the predictors, and the perceived risk factors among nurses. A self-administered standard Nordic questionnaire was distributed among all eligible 118 nurses working at two hospitals. Bivariate and multivariate logistic regression analyses were applied to identify the predictors of WMSDs. Study findings revealed that nearly half (47.46%) of nurses had WMSDs most significantly affecting the lower back (18.93%). Nurses exceeding 30 years of age had almost seven times higher odds of having WMSDs compared to their counterparts aged less than 30 adjusting the effects of BMI, department, years of clinical experience, and type of hospital (OR= 6.92; CI =1.67-28.58). There was decreased odds of experiencing WMSDs by 0.33 times for nurses working in a private hospital than in a government hospital adjusting the effect of age, BMI, department, and years of clinical experience (OR=0.33; 95% CI=0.12-0.91). Similarly, nurses working in critical units had 5.24 times higher odds of having musculoskeletal disorders than in general units (OR=5.25; 95% CI=1.79-15.38). Nurses working for more than five years had 7.53 times higher odds of having WMSDs than those with less than and equal to five years of work experience (OR=7.53; CI 2.78-20.32). WMSDs are common among nurses primarily affecting the lower back and the odds of having MSDs is high with increasing age, BMI, and work experience, and for nurses who worked in critical care units and public hospital. Prompt preventive measures should be adopted to avoid risk factors of MSDs in work settings.

Development of deep learning models to assess the degree of cognitive impairment on magnetic resonance imaging (MRI) scans has high translational significance. Performance of such models is often affected by potential variabilities stemming from independent protocols for data generation, imaging equipment, radiology artifacts, and demographic distributional shifts. Domain generalization (DG) frameworks have the potential to overcome these issues by learning signal from one or more source domains that can be transferable to unseen target domains. We developed an approach that leverages model interpretability as a means to improve generalizability of classification models across multiple cohorts. Using MRI scans and clinical diagnosis obtained from four independent cohorts (Alzheimer's Disease Neuroimaging Initiative (ADNI, n = 1, 821), the Framingham Heart Study (FHS, n = 304), the Australian Imaging Biomarkers and Lifestyle Study of Ageing (AIBL, n = 661), and the National Alzheimer's Coordinating Center (NACC,n = 4, 647)), we trained a deep neural network that used model-identified regions of disease relevance to inform model training. We trained a classifier to distinguish persons with normal cognition (NC) from those with mild cognitive impairment (MCI) and Alzheimer's disease (AD) by aligning class-wise attention with a unified visual saliency prior computed offline per class over all training data. Our proposed method competes with state-of-the-art methods with improved correlation with postmortem histology, thus grounding our findings with gold standard evidence and paving a way towards validating DG frameworks.

Amid a potential menthol ban, electronic cigarette (e-cigarette) companies are incorporating synthetic cooling agents like WS-3 and WS-23 to replicate menthol/mint sensations. This study examines public views on synthetic cooling agents in e-cigarettes via Twitter data. From May 2021 to March 2023, we used Twitter Streaming API, to collect tweets related to synthetic cooling agents with keywords such as WS-23, 'ice,' and 'frozen.' Two deep learning roBERTa models, classified attitudes expressed in tweets about synthetic cooling agents and identified e-cigarette users. The BERTopic deep-learning model identified major topics of positive and negative tweets. Two proportion Z-tests were used to compare the proportions of positive and negative attitudes between e-cigarette users (vapers) and non-e-cigarette-users (non-vapers). Of 6,940,065 e-cigarettes related tweets, 5,788 non-commercial tweets related to synthetic cooling agents. The longitudinal trend analysis showed a clear upwards trend in discussions. Vapers posted most of the tweets (73.05%, 4,228/5,788). Nearly half (47.87%, 2,771/5,788) held a positive attitude toward synthetic cooling agents, which is significantly higher than those with a negative attitude (19.92%,1,153/5,788) with a P-value < 0.0001. The likelihood of Vapers expressing positive attitudes (60.17%, 2,544/4,228) was significantly higher (P < 0.0001) than that from non-vapers (14.55%, 227/1,560). Conversely, negative attitudes from non-vapers (30%, 468/1,560) was significantly (P < 0.0001) higher than vapers (16.2%, 685/4,228). Prevalent topics from positive tweets included "enjoyment of specific vape flavors," "preference for lush ice vapes," and "liking of minty/icy feelings." Major topics from negative tweets included "disliking certain vape flavors" and "dislike of others vaping around them." On Twitter, vapers are more likely to have a positive attitude toward synthetic cooling agents than non-vapers. Our study provides important insights into how the public perceives synthetic cooling agents in e-cigarettes, insights that are crucial for shaping future FDA regulations aimed at safeguarding public health.

Background People infected with COVID-19 may continue to experience symptoms for several weeks or even months after acute infection, a condition known as long COVID. Cognitive problems such as memory loss are among the most commonly reported symptoms of long COVID. However, a comprehensive evaluation on the risks of cognitive decline following COVID infection among different sociodemographic groups has not been undertaken at the national level in the United States. Methods We conducted a secondary analysis on the datasets from U.S. Census Bureau Household Pulse Survey, encompassing the data collected from June 1, 2022 to December 19, 2022. Based on a cohort of 385,370 individuals aged 18 or older, we employed logistic regression analyses to examine the association between self-reported cognitive deficits and different sociodemographic factors among individuals with long COVID conditions. Results Among individuals aged 18 or older, 44.7% percent of survey respondents report having been diagnosed with COVID in the past, and 29.0% of those with previous COVID infection experienced long COVID symptoms lasting for more than 3 months. We have demonstrated that individuals with long COVID had significantly higher risk of experiencing cognitive deficits compared to those with no history of COVID infection. Furthermore, females, young adults, people with multiple races, or low levels of education attainment are at high risk of cognitive deficits if they experience long COVID. At the state level, the prevalence of cognitive deficits among long COVID patients varied across different US states, with the highest prevalence in West Virginia and Kentucky, and the lowest prevalence in Connecticut and Maryland. The variation could be due to differences in racial composition and education level among long COVID patients in the four states. Conclusions The risks of cognitive deficits among adults with post-COVID conditions are substantial. Various sociodemographic groups can have different risks of developing cognitive deficits after experiencing long COVID. Findings of this large-scale study can help identify sociodemographic groups at higher risk of cognitive deficits, and facilitate medical interventions and guide resource allocation to target populations at risk and to prioritize areas with a high rate of cognitive decline.

Background Real-time surveillance of emerging infectious diseases necessitates a dynamically evolving, computable case definition, which frequently incorporates symptom-related criteria. For symptom detection, both population health monitoring platforms and research initiatives primarily depend on structured data extracted from electronic health records. Objective To validate and test an artificial intelligence (AI) based Natural Language Processing (NLP) pipeline for detecting COVID-19 symptoms from physician notes. Methods Subjects in this retrospective cohort study are patients 21 years old and younger, who presented to a pediatric emergency department (ED) at a large academic children's hospital between March 1, 2020 and May 31, 2022. ED notes for all patients were processed with an NLP pipeline tuned to detect the mention of 11 COVID-19 symptoms based on CDC criteria. For a gold standard, 3 subject matter experts labeled 226 ED notes and had strong agreement (F1=98.6; PPV=97.2; Recall=100.0). F1, PPV, and recall were used to compare the performance of both NLP and ICD-10 to the gold standard chart review. As a formative use case, variations in symptom patterns were measured across SARS-Cov2 variant eras. Results There were 85,678 ED encounters during the study period, 4.0% with patients with COVID-19. NLP was more accurate at identifying encounters with patients that had any of the COVID-19 symptoms (F1=79.6) than ICD-10 codes (F1=45.1%). NLP accuracy was higher for positive symptoms (recall=93%) than ICD-10 (recall=30%). However, ICD-10 accuracy was higher for negative symptoms (specificity=99.4%) than NLP (specificity=91.7%). Congestion or runny nose showed the highest accuracy difference: NLP F1=82.8%, ICD-10 F1=4.2%. Prevalence of NLP symptoms among patients with COVID-19 differed across variant eras. And patients with COVID-19 were more likely to have each symptom than patients without this disease. Effect sizes (odds ratios) varied across pandemic eras. Conclusions This study establishes the value of AI based NLP as a highly effective tool for real-time COVID-19 symptom detection in pediatric patients, outperforming traditional ICD-10 methods. It also reveals the evolving nature of symptom prevalence across different virus variants, underscoring the need for dynamic, technology-driven approaches in infectious disease surveillance.

Background The colliding epidemic of infectious and non-communicable diseases in South Africa could potentially increase the prevalence of kidney disease. This study determines the prevalence of kidney damage and known risk factors in a rural community of the Eastern Cape province, South Africa. Methods This observational cross-sectional study was conducted in the outpatient department of the Mbekweni Community Health Centre in the Eastern Cape between May and July 2022. Relevant data on demography, medical history, anthropometry and blood pressure were obtained. The glomerular filtration rate was estimated using the Chronic Kidney Disease Epidemiology Collaboration Creatinine (CKD-EPICreatinine) equation and the re-expressed four-variable Modification of Diet in Renal Disease (MDRD) equation, without any adjustment for black ethnicity. Significant kidney damage was defined as low eGFR (<60mL/min per 1.73m2) and/or the presence of proteinuria. We used the logistic regression model analysis to identify the independent risk factors for significant kidney damage. Results The mean ({+/-} standard deviation) age of the 389 participants was 52.3 ({+/-} 17.5) years. The prevalence of significant kidney damage was 17.2% (n=67), as estimated by the CKD-EPICreatinine, with slight difference from the MDRD equation (17.7%; n=69), while the prevalence of proteinuria was 7.2%. Risk factors for significant kidney damage were older age (OR=0.94, 95% CI 0.91 - 0.96, p<0.001) and the presence of proteinuria (OR=0.98, 95% CI 0.97 - 0.99, p 0.002). Proteinuria was strongly associated with hypertension (OR=4.46, 95% CI 1.33 - 14.92, p<0.015) and elevated serum creatinine (OR=1.01, 95% CI 1 - 1.02, p=0.004). Conclusions This study found a high prevalence of kidney damage (17.2%) and proteinuria (7.2%) in this rural community, largely attributed to advanced age and hypertension, respectively. Early detection of proteinuria and decreased renal function could lead to prompt preventative measures and management to delay the progression to end-stage kidney failure and mortality.

Background: Electrocardiographic left ventricular hypertrophy (LVH) could predict adverse renal outcome in patients with hypertension. This study aimed to investigate the association between electrocardiographic LVH and future decline of renal function in the general population. Methods: This retrospective cohort study included individuals who received population-based health checkups from 2010 to 2019 in Japan. Electrocardiographic LVH was defined according to the Minnesota code. Renal function decline was defined as a decrease of 25% in the estimated glomerular filtration rate from baseline to <60 mL/min/1.73 m2. Multivariate adjusted Cox regression analysis was employed to evaluate the association between electrocardiographic LVH at baseline and renal function decline. Results: Of the 19,825 study participants, 1,263 exhibited electrocardiographic LVH at the baseline visit. The mean follow-up period was 3.4 {+/-} 1.9 years. The incidence rates of renal function decline were 0.30 and 0.78 per 100 person-years in the non-LVH group and LVH groups, respectively. Electrocardiographic LVH was associated with the risk for renal function decline in both unadjusted analysis (hazard ratio 2.50, 95% confidence interval 1.73-3.60, P < 0.001) and adjusted analysis (hazard ratio 1.69, 95% confidence interval 1.14-2.50, P = 0.009). This association was comparable across subgroups stratified by age, sex, body mass index, diagnosed hypertension, systolic blood pressure, hemoglobin A1c, and urinary protein. Conclusions: In the general population, electrocardiographic LVH was associated with future renal function decline. To detect high-risk individuals for renal function decline, electrocardiographic LVH may be useful in the setting of health checkups in the general population.

Rituximab, the first monoclonal antibody approved for the treatment of lymphoma, eventually became one of the most popular and versatile drugs ever in terms of clinical application and revenue. Since its patent expiration, and consequently, the loss of exclusivity of the original biologic, its repurposing as an off-label drug has increased dramatically, propelled by the development and commercialization of its many biosimilars. Currently, rituximab is prescribed worldwide to treat a vast range of autoimmune diseases mediated by B cells. Here, we present a comprehensive overview of rituximab repurposing in 116 autoimmune diseases across 17 medical specialties, sourced from over 1,530 publications. Our work highlights the extent of its off-label use and clinical benefits, underlining the success of rituximab repurposing for both common and orphan immune-related diseases. We discuss the scientific mechanism associated with its clinical efficacy and provide additional indications for which rituximab could be investigated. Our study presents rituximab as a flagship example of drug repurposing owing to its central role in targeting CD20+ B cells in over 100 autoimmune diseases.

Background: Proprotein convertase subtilisin/kexin type 9 inhibitors (PCSK9I) have potential benefits against cardiovascular disease. The comparative risks of new-onset major adverse cardiovascular events (MACE) between PCSK9I and ezetimibe remain unknown. Objective: This real-world study compared the risks of MACE upon exposure to PCSK9I and ezetimibe. Methods: This was a retrospective population-based cohort study of patients with dyslipidaemia on either PCSK9I or ezetimibe between 1st January 2015 and 30th October 2022 using a territory-wide database from Hong Kong. The primary outcome was new-onset MACE. The secondary outcomes were myocardial infarction, heart failure, stroke/transient ischaemic attack, and all-cause mortality. Propensity score matching (1:3 ratio) using the nearest neighbour search was performed. Multivariable Cox regression was used to identify significant associations. Results: This cohort included 42450 dyslipidaemia patients (median age: 65.0 years old [SD: 11.1]; 64.54 % males). The PCSK9I and ezetimibe groups consisted of 1477 and 40973 patients, respectively. After matching, 67 and 235 patients suffered from MACE in the PCSK9I and ezetimibe groups, respectively, over a total of 14514.5 person-years. PCSK9I was associated with lower risks of MACE (Hazard ratio [HR]: 0.59; 95% Confidence Interval [CI]: 0.37-0.92) compared to ezetimibe use after adjusting for demographics, past comorbidities, other medications, and time-weighted means of lipid and glucose tests. Besides, while both alirocumab and evolocumab were associated with lower risks of MACE, evolocumab was associated with significantly lower risks of myocardial infarction, heart failure, and stroke/transient ischaemic attack. The results remained consistent in the competing risk and sensitivity analyses. Conclusions: PCSK9I use amongst dyslipidaemia patients was associated with lower risks of new-onset MACE and outcomes compared to ezetimibe after adjustments. Evolocumab might perform better than Alirocumab in reducing the risks of cardiovascular diseases.

Objective: To estimate the prevalence of patients presenting in pain to an inner-city emergency department (ED), describing this population, their treatment, and the effect of the COVID-19 pandemic. Materials and Methods: We applied a clinical text deep learning model to the free text nursing assessments to identify the prevalence of pain on arrival to the ED. Using interrupted time series analysis, we examined the prevalence over three years. We describe this population pre- and post-pandemic in terms of their demographics, arrival patterns and treatment. Results: 55.16% (95%CI 54.95% - 55.36%) of all patients presenting to this ED had pain on arrival. There were significant differences in demographics, arrival and departure patterns between those patients with and without pain. The COVID-19 pandemic initially precipitated a decrease followed by a sharp, sustained rise in the prevalence of pain on arrival, altering the population arriving in pain and their treatment. Discussion The application of a clinical text deep learning model has successfully identified the prevalence of pain on arrival. The description of this population and their treatment forms the basis of intervention to improve care for patients presenting with pain. The combination of the clinical text deep learning model and interrupted time series analysis has identified the effects of the COVID-19 pandemic on pain care in the ED. Conclusion A clinical text deep learning model has led to identifying the prevalence of pain on arrival and was able to identify the effect a major pandemic had on pain care in this ED.

Background Patients and their families often explore the information available online for information about health status. A dialogue-based artificial intelligence (AI) language model (ChatGPT) has been developed for complex question and answer. We sought to assess whether AI model had knowledge of cardiovascular disease (CVD) racial disparities, including disparities associated with CVD risk factors and associated diseases. Methods To assess ChatGPT's responses to topic in cardiovascular disease disparities, we created six questions of twenty sets, with each set consisting of three questions with changes to text prompt based on differences in patient demographics. Each question was asked three times to assess variability in responses. Results A total of 180 responses were tabulated from ChatGPT's responses to 60 questions asked in triplicate to assess. Despite some variation in wording, all responses to the same prompt were consistent across different sessions. ChatGPT's responses to 63.4% of the questions (38 out of 60 questions) were appropriate, 33.3% (20 out of 60 questions) were inappropriate and 3.3% unreliable (2 out of 60 questions). Of the 180 prompt entries into ChatGPT, 141 (78.3%) were correct, 28 (15.5%) were hedging or indeterminate responses that could not be binarized into a correct or incorrect response, and 11 (6.1%) were incorrect. There were consistent themes in incorrect or hedging responses by ChatGPT, with 91% and 79% of incorrect and hedging responses related to a cardiovascular disease disparity that affects a minority or underserved racial group. Conclusion Our study showed that an online chat-based AI model has a broad knowledge of CVD racial disparities, however persistent gaps in knowledge about minority groups. Given that these models might be used by the general public, caution should be advised in taking responses at face value.