Citations

Hypertension is among the most important risk factors for cardiovascular disease, chronic kidney disease, and dementia. The artificial intelligence (AI) field is advancing quickly, and there has been little discussion on how AI could be leveraged for improving the diagnosis and management of hypertension. AI technologies, including machine learning tools, could alter the way we diagnose and manage hypertension, with potential impacts for improving individual and population health. The development of successful AI tools in public health and health care systems requires diverse types of expertise with collaborative relationships between clinicians, engineers, and data scientists. Unbiased data sources, management, and analyses remain a foundational challenge. From a diagnostic standpoint, machine learning tools may improve the measurement of blood pressure and be useful in the prediction of incident hypertension. To advance the management of hypertension, machine learning tools may be useful to find personalized treatments for patients using analytics to predict response to antihypertension medications and the risk for hypertension-related complications. However, there are real-world implementation challenges to using AI tools in hypertension. Herein, we summarize key findings from a diverse group of stakeholders who participated in a workshop held by the National Heart, Lung, and Blood Institute in March 2023. Workshop participants presented information on communication gaps between clinical medicine, data science, and engineering in health care; novel approaches to estimating BP, hypertension risk, and BP control; and real-world implementation challenges and issues.

Abstract Summary Copy number variation (CNV) and alteration (CNA) analysis is a crucial component in many genomic studies and its applications span from basic research to clinic diagnostics and personalized medicine. CNVpytor is a tool featuring a read depth-based caller and combined read depth and B-allele frequency (BAF) based 2D caller to find CNVs and CNAs. The tool stores processed intermediate data and CNV/CNA calls in a compact HDF5 file—pytor file. Here, we describe a new track in igv.js that utilizes pytor and whole genome variant files as input for on-the-fly read depth and BAF visualization, CNV/CNA calling and analysis. Embedding into HTML pages and Jupiter Notebooks enables convenient remote data access and visualization simplifying interpretation and analysis of omics data. Availability and implementation The CNVpytor track is integrated with igv.js and available at https://github.com/igvteam/igv.js. The documentation is available at https://github.com/igvteam/igv.js/wiki/cnvpytor. Usage can be tested in the IGV-Web app at https://igv.org/app and also on https://github.com/abyzovlab/CNVpytor.

Background Multiple randomized trials could not establish the noninferiority of endovascular thrombectomy (EVT) alone without preceding intravenous thrombolysis (IVT) or superiority of IVT followed by EVT in anterior circulation large‐vessel occlusion stroke. The role of prior IVT in posterior circulation large‐vessel occlusion remains controversial. Methods In this multicenter, retrospective study, patients with stroke with large‐vessel occlusion in the posterior circulation who received EVT alone or with IVT were selected from the stroke thrombectomy and aneurysm registry between 2013 and 2022. Effects of IVT followed by thrombectomy on favorable functional outcome (defined as modified Rankin scale≤3 at 90 days) and safety were investigated using multivariable logistic and linear regression models. Results Of the 588 included patients, 67% (n = 394) were treated with EVT alone and 29% (n = 170) with EVT after IVT, and 4% (n = 24) have missing values on this variable. Controlling for multiple confounding factors, IVT was not associated with a higher likelihood of favorable functional outcome at 90 days (odds ratio, 1.04 [95% CI, 0.52–2.09; P = 0.901). Thrombectomy alone did not show any safety advantages compared with those receiving IVT. Conclusion Similar functional outcomes and complication rates were seen in patients with posterior circulation large‐vessel occlusion treated with EVT alone versus EVT after IVT. Further prospective studies are required to determine the utility of IVT in posterior circulation stroke, especially in patients being directly admitted to thrombectomy centers.

AbstractAimsRecent advances in artificial intelligence, particularly with large language models like GPT‐4Vision (GPT‐4V)—a derivative feature of ChatGPT—have expanded the potential for medical image interpretation. This study evaluates the accuracy of GPT‐4V in image classification tasks of histopathological images and compares its performance with a traditional convolutional neural network (CNN).MethodsWe utilised 1520 images, including haematoxylin and eosin staining and tau immunohistochemistry, from patients with various neurodegenerative diseases, such as Alzheimer's disease (AD), progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD). We assessed GPT‐4V's performance using multi‐step prompts to determine how textual context influences image interpretation. We also employed few‐shot learning to enhance improvements in GPT‐4V's diagnostic performance in classifying three specific tau lesions—astrocytic plaques, neuritic plaques and tufted astrocytes—and compared the outcomes with the CNN model YOLOv8.ResultsGPT‐4V accurately recognised staining techniques and tissue origin but struggled with specific lesion identification. The interpretation of images was notably influenced by the provided textual context, which sometimes led to diagnostic inaccuracies. For instance, when presented with images of the motor cortex, the diagnosis shifted inappropriately from AD to CBD or PSP. However, few‐shot learning markedly improved GPT‐4V's diagnostic capabilities, enhancing accuracy from 40% in zero‐shot learning to 90% with 20‐shot learning, matching the performance of YOLOv8, which required 100‐shot learning to achieve the same accuracy.ConclusionsAlthough GPT‐4V faces challenges in independently interpreting histopathological images, few‐shot learning significantly improves its performance. This approach is especially promising for neuropathology, where acquiring extensive labelled datasets is often challenging.

Summary Background & Aims The development of acute kidney injury (AKI) in the setting of alcohol‐associated hepatitis (AH) portends a poor prognosis. Whether the presence of AH itself drives worse outcomes in patients with cirrhosis and AKI is unknown. Methods Retrospective cohort study of 11 hospital networks of consecutive adult patients admitted in 2019 with cirrhosis and AKI. AKI phenotypes, clinical course, and outcomes were compared between AH and non‐AH groups. Results A total of 2062 patients were included, of which 303 (15%) had AH, as defined by National Institute on Alcohol Abuse and Alcoholism (NIAAA) criteria. Patients with AH, compared to those without, were younger and had higher Model for End‐stage Liver Disease‐Sodium (MELD‐Na) scores on admission. AKI phenotypes significantly differed between groups ( p < 0.001) with acute tubular necrosis occurring more frequently in patients with AH. Patients with AH reached more severe peak AKI stage, required more renal replacement therapy, and had higher 90‐day cumulative incidence of death (45% [95% CI: 39%–51%] vs. 38% [95% CI: 35%–40%], p = 0.026). Using no AH as reference, the unadjusted sHR for 90‐day mortality was higher for AH (sHR: 1.24 [95% CI: 1.03–1.50], p = 0.024), but was not significant when adjusting for MELD‐Na, age and sex. However, in patients with hepatorenal syndrome, AH was an independent predictor of 90‐day mortality (sHR: 1.82 [95% CI: 1.16–2.86], p = 0.009). Conclusions Hospitalised patients with cirrhosis and AKI presenting with AH had higher 90‐day mortality than those without AH, but this may have been driven by higher MELD‐Na rather than AH itself. However, in patients with hepatorenal syndrome, AH was an independent predictor of mortality.

AbstractPurposeThe purpose of this guideline is to provide evidence‐based guidance to clinicians of all specialties on the evaluation, management, and treatment of idiopathic overactive bladder (OAB). The guideline informs the reader on valid diagnostic processes and provides an approach to selecting treatment options for patients with OAB through the shared decision‐making process, which will maximize symptom control and quality of life, while minimizing adverse events and burden of disease.MethodsAn electronic search employing OVID was used to systematically search the MEDLINE and EMBASE databases, as well as the Cochrane Library, for systematic reviews and primary studies evaluating diagnosis and treatment of OAB from January 2013 to November 2023. Criteria for inclusion and exclusion of studies were based on the Key Questions and the populations, interventions, comparators, outcomes, timing, types of studies and settings (PICOTS) of interest. Following the study selection process, 159 studies were included and were used to inform evidence‐based recommendation statements.ResultsThis guideline produced 33 statements that cover the evaluation and diagnosis of the patient with symptoms suggestive of OAB; the treatment options for patients with OAB, including Noninvasive therapies, pharmacotherapy, minimally invasive therapies, invasive therapies, and indwelling catheters; and the management of patients with BPH and OAB.ConclusionOnce the diagnosis of OAB is made, the clinician and the patient with OAB have a variety of treatment options to choose from and should, through shared decision‐making, formulate a personalized treatment approach taking into account evidence‐based recommendations as well as patient values and preferences.

Abstract The ZRSR2 gene is a member of the spliceosome gene family which are frequently mutated in myeloid neoplasms. ZRSR2 mutations (ZRSR2m) occur in less than 5% of MDS, CMML, and AML. Our study included 164 ZRSR2m patients (98.8% males) and 98 ZRSR2wt MDS control cohort. In the ZRSR2m group, there were 53 MDS (32%), 39 CCUS (24%), 33 MPN (20%), 23 MDS/MPN overlap (14%), 15 AML (9%), and 1 MPAL (0.6%). Most MDS patients were the low blast subtype (n=37, 70%). Twenty-seven MDS patients (51%), and 23 CCUS patients (59%) had absolute monocyte count ≥0.5 x109/L and 18 of 23 MDS/MPN overlap were CMML (78%). Mutations in ZRSR2 were spread across the entire gene. The median number of co-mutations was 2, with TET2 (51%) and ASXL1 (32%) being the most common. U2AF1 and SRSF2, previously described as mutually exclusive with ZRSR2, were found in 10.4% of patients. Median overall survival (OS) was 51 months, and significantly varied among MNs (p=0.004). ZRSR2m MDS patients had better mOS than the MDS control cohort with ZRSR2wt (35 vs 22 months, p=0.013). ZRSR2m patients with isolated TET2 co-mutation and higher hemoglobin showed improved survival, while patients with RUNX1m, higher WBC count showed worse OS.

Objective This study aimed to determine the prevalence and heteroplasmy level(s) of MT-RNR1 variants m.1555A > G and m.1494C > T, which are associated with aminoglycoside-induced hearing loss, in a general perinatal population. This study also aimed to characterize the association of these variants and their heteroplasmy levels with hearing loss outcomes with and without aminoglycoside exposure. Study Design Droplet digital polymerase chain reaction was performed on 479 maternal DNA samples from a general perinatal biobank at our institution to detect the presence and heteroplasmy levels of MT-RNR1 variants m.1555A > G and m.1494C > T. Testing of paired neonatal specimen(s) was planned for positive maternal tests. A retrospective chart review was performed to characterize the population, identify aminoglycoside exposures, and determine hearing outcomes. Results All maternal samples tested negative for MT-RNR1 variants m.1555A > G and m.1494C > T. Maternal and neonatal subjects had high rates of aminoglycoside exposure (15.9 and 13.9%, respectively). No subjects with sensorineural or mixed hearing loss had documented aminoglycoside exposure. Conclusion This study demonstrated that a larger sample size is needed to establish the prevalence of these variants as no subjects tested positive. Determination of variant prevalence in the neonatal population, association of variant heteroplasmy levels with hearing outcomes, and reliability of maternal testing as a surrogate for neonatal testing are important next steps toward universal prenatal or newborn screening. Key Points

Abstract: Paratesticular mesothelioma (malignant mesothelioma arising from the tunica vaginalis of the testis) represents a small proportion of mesothelial neoplasms, and cutaneous involvement by paratesticular mesothelioma is very rare. Cutaneous involvement can manifest as scrotal subcutaneous nodules from regional spread, distant metastasis, or direct extension through surgical scars. Mesothelioma has 3 histopathologic classifications that include epithelioid, biphasic, and sarcomatoid, which is rarely seen in paratesticular mesothelioma. Given the rarity of this condition, cutaneous mesothelioma may be misdiagnosed as histologic mimics, such as metastatic adenocarcinoma or adnexal neoplasms; thus, appropriate immunohistochemical workup and clinical correlation are required to make an accurate diagnosis. In this case, a 75-year-old man with a history of paratesticular mesothelioma, status postorchiectomy, presented with right-sided scrotal swelling, erythema, and subcutaneous nodules. These nodules were identified as local recurrence with cutaneous involvement by paratesticular mesothelioma on histopathologic examination. This case highlights the clinical and histopathologic features of this diagnosis and underscores the importance of dermatopathologists being aware of this condition to ensure accurate diagnosis.