Using machine learning to predict carotid artery symptoms from CT angiography: A radiomics and deep learning approach.

Purpose: To assess radiomics and deep learning (DL) methods in identifying symptomatic Carotid Artery Disease (CAD) from carotid CT angiography (CTA) images. We further compare the performance of these novel methods to the conventional calcium score. Methods: Carotid CT angiography (CTA) images from symptomatic patients (ischaemic stroke/transient ischaemic attack within the last 3 months) and asymptomatic patients were analysed. Carotid arteries were classified into culprit, non-culprit and asymptomatic. The calcium score was assessed using the Agatston method. 93 radiomic features were extracted from regions-of-interest drawn on 14 consecutive CTA slices. For DL, convolutional neural networks (CNNs) with and without transfer learning were trained directly on CTA slices. Predictive performance was assessed over 5-fold cross validated AUC scores. SHAP and GRAD-CAM algorithms were used for explainability. Results: 132 carotid arteries were analysed (41 culprit, 41 non-culprit, and 50 asymptomatic). For asymptomatic vs symptomatic arteries, radiomics attained a mean AUC of 0.96(± 0.02), followed by DL 0.86(± 0.06) and then calcium 0.79(± 0.08). For culprit vs non-culprit arteries, radiomics achieved a mean AUC of 0.75(± 0.09), followed by DL 0.67(± 0.10) and then calcium 0.60(± 0.02). For multi-class classification, the mean AUCs were 0.95(± 0.07), 0.79(± 0.05), and 0.71(± 0.07) for radiomics, DL and calcium, respectively. Explainability revealed consistent patterns in the most important radiomic features. Conclusions: Our study highlights the potential of novel image analysis techniques in extracting quantitative information beyond calcification in the identification of CAD. Though further work is required, the transition of these novel techniques into clinical practice may eventually facilitate better stroke risk stratification.

Biologic therapy for inflammatory bowel disease: Real-world comparative effectiveness and impact of drug sequencing in 13,222 patients within the UK IBD BioResource.

BACKGROUND AND AIMS: To compare effectiveness of different biologic therapies and sequences in patients with Inflammatory Bowel Disease (IBD) using real-world data from a large cohort with long exposure. METHODS: Demographic, disease, treatment and outcome data were retrieved for patients in the UK IBD BioResource. Effectiveness of treatment was based on persistence free of discontinuation or failure, analysed by Kaplan-Meier survival analysis with inverse probability of treatment weighting to adjust for differences between groups. RESULTS: 13,222 evaluable patients received at least one biologic. In ulcerative colitis (UC) first line vedolizumab (VDZ) demonstrated superior effectiveness over five years compared to anti-TNF agents (p=0.006). VDZ was superior to both infliximab (IFX) and adalimumab (ADA) after ADA and IFX failure respectively (p<0.001 and p<0.001). Anti-TNF therapy showed similar effectiveness when used first-line, or after failure of VDZ. In Crohn's disease (CD) we found significant differences between first line treatments over ten years (p=0.045), with superior effectiveness of IFX compared to ADA in perianal CD. Non-anti-TNF biologics were superior to a second anti-TNF after first line anti-TNF failure in CD (p=0.035). Patients with UC or CD experiencing TNF-failure due to delayed loss of response or intolerance had superior outcomes when switching to a non-anti-TNF biologic, rather than a second anti-TNF. CONCLUSIONS: We provide real-world evidence to guide biologic selection and sequencing in a range of common scenarios. Our findings challenge current guidelines regarding drug selection after loss of response to first anti-TNF.

Circulating testosterone levels and health outcomes in chronic obstructive pulmonary disease: results from ECLIPSE and ERICA.

The relationship of circulating testosterone levels with health outcomes in people with chronic obstructive pulmonary disease (COPD) is unknown. AIM: To determine whether serum testosterone levels predict hospitalised acute exacerbations of COPD (H-AECOPD), cardiovascular disease outcome, and mortality in people with COPD. METHODS: Separate analyses were carried out on two observational, multicentre COPD cohorts, Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points (ECLIPSE) and Evaluation of the Role of Inflammation in Chronic Airways Disease (ERICA), both of which had serum testosterone measured using a validated liquid chromatography assay at the same laboratory. Data from 1296 male participants in ECLIPSE and 386 male, 239 female participants in ERICA were analysed. All analyses were sex-specific. Multivariate logistic regression was used to determine associations with H-AECOPD during follow-up (3 years ECLIPSE, 4.5 years ERICA), a composite endpoint of cardiovascular hospitalisation and cardiovascular death, and all-cause mortality. RESULTS: Mean (SD) testosterone levels were consistent across cohorts; 459 (197) and 455 (200) ng/dL for males in ECLIPSE and ERICA, respectively, and in ERICA females: 28 (56) ng/dL. Testosterone was not associated with H-AECOPD (ECLIPSE: OR: 0.76, p=0.329, ERICA males: OR (95% CI): 1.06 (0.73 to 1.56), p=0.779, ERICA females: OR: 0.77 (0.52 to 1.12), p=0.178) or cardiovascular hospitalisation and death. Testosterone was associated with all-cause mortality in Global Initiative for Obstructive Lung Disease (GOLD) stage 2 male patients only, in ECLIPSE (OR: 0.25, p=0.007) and ERICA (OR: (95% CI): 0.56 (0.32 to 0.95), p=0.030). CONCLUSIONS: Testosterone levels do not relate to H-AECOPD or cardiovascular outcome in COPD, but are associated with all-cause mortality in GOLD stage 2 COPD male patients, although the clinical significance of this finding is uncertain.

Primary hypertension, anti-hypertensive medications and the risk of severe COVID-19 in UK Biobank.

Hypertension appears to be one of the commonest comorbidities in COVID-19 patients, although whether hypertensive individuals have a higher risk of severe COVID-19 compared with non-hypertensives is unclear. It is also unclear whether the absolute level of systolic blood pressure, or the type of anti-hypertensive medication is related to this risk. Analyses were conducted using data from the UK Biobank and linked health records. Logistic regression models were fitted to assess the impact of hypertension, systolic blood pressure (SBP) and medications on the risk of severe COVID-19. 16,134 individuals tested positive for severe acute respiratory syndrome-coronavirus, 22% (n = 3,584) developed severe COVID-19 and 40% (n = 6,517) were hypertensive. Hypertension was associated with 22% higher odds of severe COVID-19 (Odds ratio (OR) 1.22; 95% confidence interval (CI) 1.12, 1.33), compared with normotension after adjusting for confounding variables. In those taking anti-hypertensive medications, elevated SBP showed a dose-response relationship with severe COVID-19 (150-159mmHg versus 120-129mmHg (OR 1.91; 95% CI 1.44, 2.53), >180+mmHg versus 120-129mmHg (OR 1.93; 95% CI 1.06, 3.51)). SBP <120mmHg was associated with greater odds of severe COVID-19 (OR 1.40; 95% CI 1.11, 1.78). Angiotensin-converting enzyme inhibitors or angiotensin-II receptor blockers were not associated with altered risk of severe COVID-19. Hypertension is an important risk factor for COVID-19. A better understanding of the underlying mechanisms is warranted in case of more severe strains or other viruses in the future.

MRI Feature Tracking Strain in Pulmonary Hypertension: Utility of Combined Left Atrial Volumetric and Deformation Assessment in Distinguishing Post- From Pre-capillary Physiology.

Aims: Pulmonary hypertension (PH) is dichotomized into pre- and post-capillary physiology by invasive catheterization. Imaging, particularly strain assessment, may aid in classification and be helpful with ambiguous hemodynamics. We sought to define cardiac MRI (CMR) feature tracking biatrial peak reservoir and biventricular peak systolic strain in pre- and post-capillary PH and examine the performance of peak left atrial strain in distinguishing the 2 groups compared to TTE. Methods and Results: Retrospective cross-sectional study from 1 Jan 2015 to 31 Dec 2020; 48 patients (22 pre- and 26 post-capillary) were included with contemporaneous TTE, CMR and catheterization. Mean pulmonary artery pressures were higher in the pre-capillary cohort (55 ± 14 vs. 42 ± 9 mmHg; p < 0.001) as was pulmonary vascular resistance (median 11.7 vs. 3.7 WU; p < 0.001). Post-capillary patients had significantly larger left atria (60 ± 22 vs. 25 ± 9 ml/m2; p < 0.001). There was no difference in right atrial volumes between groups (60 ± 21 vs. 61 ± 29 ml/m2; p = 0.694), however peak RA strain was lower in post-capillary PH patients (8.9 ± 5.5 vs. 18.8 ± 7.0%; p < 0.001). In the post-capillary group, there was commensurately severe peak strain impairment in both atria (LA strain 9.0 ± 5.8%, RA strain 8.9 ± 5.5%). CMR LAVi and peak LA strain had a multivariate AUC of 0.98 (95% CI 0.89-1.00; p < 0.001) for post-capillary PH diagnosis which was superior to TTE. Conclusion: CMR volumetric and deformation assessment of the left atrium can highly accurately distinguish post- from pre-capillary PH.

Physiological effects and subjective tolerability of prone positioning in COVID-19 and healthy hypoxic challenge.

BACKGROUND: Prone positioning has a beneficial role in coronavirus disease 2019 (COVID-19) patients receiving ventilation but lacks evidence in awake non-ventilated patients, with most studies being retrospective, lacking control populations and information on subjective tolerability. METHODS: We conducted a prospective, single-centre study of prone positioning in awake non-ventilated patients with COVID-19 and non-COVID-19 pneumonia. The primary outcome was change in peripheral oxygenation in prone versus supine position. Secondary outcomes assessed effects on end-tidal CO2, respiratory rate, heart rate and subjective symptoms. We also recruited healthy volunteers to undergo proning during hypoxic challenge. RESULTS: 238 hospitalised patients with pneumonia were screened; 55 were eligible with 25 COVID-19 patients and three non-COVID-19 patients agreeing to undergo proning - the latter insufficient for further analysis. 10 healthy control volunteers underwent hypoxic challenge. Patients with COVID-19 had a median age of 64 years (interquartile range 53-75). Proning led to an increase in oxygen saturation measured by pulse oximetry (SpO2) compared to supine position (difference +1.62%; p=0.003) and occurred within 10 min of proning. There were no effects on end-tidal CO2, respiratory rate or heart rate. There was an increase in subjective discomfort (p=0.003), with no difference in breathlessness. Among healthy controls undergoing hypoxic challenge, proning did not lead to a change in SpO2 or subjective symptom scores. CONCLUSION: Identification of suitable patients with COVID-19 requiring oxygen supplementation from general ward environments for awake proning is challenging. Prone positioning leads to a small increase in SpO2 within 10 min of proning though is associated with increased discomfort.

Role of the IL-33/ST2 axis in cardiovascular disease: A systematic review and meta-analysis.

Interleukin (IL)-33 and its unique receptor, ST2, play a pivotal role in the immune response to infection and stress. However, there have been conflicting reports of the role of IL-33 in cardiovascular disease (CVD) and the potential of this axis in differentiating CVD patients and controls and with CVD disease severity, remains unclear. AIMS: 1) To quantify differences in circulating IL-33 and/or sST2 levels between CVD patients versus controls. 2) Determine association of these biomarkers with mortality in CVD and community cohorts. METHODS AND RESULTS: Using Pubmed/MEDLINE, Web of Science, Prospero and Cochrane databases, systematic review of studies published on IL-33 and/or sST2 levels in patients with CVD (heart failure, acute coronary syndrome, atrial fibrillation, stroke, coronary artery disease and hypertension) vs controls, and in cohorts of each CVD subtype was performed. Pooled standardised mean difference (SMD) of biomarker levels between CVD-cases versus controls and hazard ratios (HRs) for risk of mortality during follow-up in CVD patients, were assessed by random effects meta-analyses. Heterogeneity was evaluated with random-effects meta-regressions. From 1071 studies screened, 77 were meta-analysed. IL-33 levels were lower in HF and CAD patients vs controls, however levels were higher in stroke patients compared controls [Meta-SMD 1.455, 95% CI 0.372-2.537; p = 0.008, I2 = 97.645]. Soluble ST2 had a stronger association with risk of all-cause mortality in ACS (Meta-multivariate HR 2.207, 95% CI 1.160-4.198; p = 0.016, I2 = 95.661) than risk of all-cause mortality in HF (Meta-multivariate HR 1.425, 95% CI 1.268-1.601; p<0.0001, I2 = 92.276). There were insufficient data to examine the association of IL-33 with clinical outcomes in CVD. CONCLUSIONS: IL-33 and sST2 levels differ between CVD patients and controls. Higher levels of sST2 are associated with increased mortality in individuals with CVD. Further study of IL-33/ST2 in cardiovascular studies is essential to progress diagnostic and therapeutic advances related to IL-33/ST2 signalling.

Investigating the Lowest Threshold of Vascular Benefits from LDL Cholesterol Lowering with a PCSK9 mAb Inhibitor (Alirocumab) in Patients with Stable Cardiovascular Disease (INTENSITY-HIGH): protocol and study rationale for a randomised, open label, parallel group, mechanistic study.

INTRODUCTION: Elevated low-density lipoprotein cholesterol (LDL-C) is a strong independent risk predictor of cardiovascular (CV) events, while interventions to reduce it remain the only evidence-based approach to reduce CV morbidity and mortality. Secondary prevention statin trials in combination with ezetimibe and/or proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors showed that there is no 'J shaped curve' in LDL-C levels with regard to CV outcomes. The lowest threshold beyond which reduction of LDL-C confers no further CV benefits has not been identified.The INTENSITY-HIGH study seeks to explore physiological mechanisms mediating CV benefits of LDL-C lowering by PCSK9 inhibition in patients with established cardiovascular disease (CVD). The study examines the changes in measures of endothelial function and vascular inflammation imaging following intervention with PCSK9 and against standard of care. METHODS AND ANALYSIS: This is a single-centre, randomised, open label, parallel group, mechanistic physiological study. It will include approximately 60 subjects with established CVD, with LDL-C of <4.1 mmol/L on high-intensity statins. All eligible participants will undergo 18-fluorodeoxyglucose positron emission tomography/CT (FDG-PET/CT) scanning of the aorta and carotid arteries, as well as baseline endothelial function assessment. Subsequently, they will be randomised on a 1:1 basis to either alirocumab 150 mg or ezetimibe 10 mg/day. Repeat FDG-PET/CT scan and vascular assessments will be undertaken after 8 weeks of treatment. Any changes in these parameters will be correlated with changes in lipid levels and systemic inflammation biomarkers. ETHICS AND DISSEMINATION: The study received a favourable opinion from the Wales Research Ethics Committee 4, was registered on clinicaltrials.gov and conformed to International Conference for Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use Good Clinical Practice. The results of this study will be reported through peer-reviewed journals and conference presentations. TRIAL REGISTRATION NUMBER: NCT03355027.

Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events.

Radiomics, quantitative feature extraction from radiological images, can improve disease diagnosis and prognostication. However, radiomic features are susceptible to image acquisition and segmentation variability. Ideally, only features robust to these variations would be incorporated into predictive models, for good generalisability. We extracted 93 radiomic features from carotid artery computed tomography angiograms of 41 patients with cerebrovascular events. We tested feature robustness to region-of-interest perturbations, image pre-processing settings and quantisation methods using both single- and multi-slice approaches. We assessed the ability of the most robust features to identify culprit and non-culprit arteries using several machine learning algorithms and report the average area under the curve (AUC) from five-fold cross validation. Multi-slice features were superior to single for producing robust radiomic features (67 vs. 61). The optimal image quantisation method used bin widths of 25 or 30. Incorporating our top 10 non-redundant robust radiomics features into ElasticNet achieved an AUC of 0.73 and accuracy of 69% (compared to carotid calcification alone [AUC: 0.44, accuracy: 46%]). Our results provide key information for introducing carotid CT radiomics into clinical practice. If validated prospectively, our robust carotid radiomic set could improve stroke prediction and target therapies to those at highest risk.

Abnormal Pulmonary Venous Filling: An Adjunct Feature in the Computed Tomography Pulmonary Angiogram Assessment of Chronic Thromboembolic Pulmonary Hypertension.

Background Hypodense filling defects within the pulmonary veins on computed tomography described as pulmonary vein sign (PVS) have been noted in acute pulmonary embolism and shown to be associated with poor prognosis. We evaluated venous flow abnormalities in chronic thromboembolic pulmonary hypertension (CTEPH) to determine its usefulness in the computed tomography assessment of CTEPH. Methods and Results Blinded retrospective computed tomography analysis of 50 proximal CTEPH cases and 3 control groups-50 acute pulmonary embolism, 50 nonthromboembolic cohort, and 50 pulmonary arterial hypertension. Venous flow reduction was assessed by the following: (1) presence of a filling defect of at least 2 cm in a pulmonary vein draining into the left atrium, and (2) left atrium attenuation (>160 Hounsfield units). PVS was most prevalent in CTEPH. Compared with all controls, sensitivity and specificity of PVS for CTEPH is 78.0% and 85.3% (95% CI, 64.0-88.5 and 78.6-90.6, respectively) versus 34.0% and 70.7% (95% CI, 21.2-48.8 and 62.7-77.8) in acute pulmonary embolism, 8.0% and 62% (95% CI, 2.2-19.2 and 53.7-69.8) in nonthromboembolic and 2.0% and 60% (95% CI, 0.1-10.7 and 51.7-67.9) in pulmonary arterial hypertension. In CTEPH, lobar and segmental arterial occlusive disease was most commonly associated with corresponding absent venous flow. PVS detection was highly reproducible (Kappa=0.96, 95% CI, 0.90-1.01, P<0.001). Conclusions PVS is easy to detect with higher sensitivity and specificity in CTEPH compared with acute pulmonary embolism and is not a feature of pulmonary arterial hypertension. Asymmetric enhancement of pulmonary veins may serve as an additional parameter in the computed tomography assessment of CTEPH and can be used to differentiate CTEPH from pulmonary arterial hypertension.

Investigating the lowest threshold of vascular benefits from LDL cholesterol lowering with a PCSK9 mAb inhibitor (alirocumab) in healthy volunteers - a mechanistic physiological study (INTENSITY-LOW): protocol and study rationale.

Objective: Whether reducing low density lipoprotein cholesterol (LDL-C) is associated with cardiovascular benefits in low risk normocholesterolaemic subjects is unknown. The INTENSITY LOW [Investigating the lowest threshold of vascular benefits from LDL-cholesterol lowering with a PCSK9 mAb inhibitor (alirocumab) in healthy volunteers] study aims to assess whether lowering LDL-C by alirocumab monotherapy can improve endothelial-dependent vascular function compared with placebo (primary objective) in low-risk normocholesterolaemic healthy individuals. Changes in endothelial-dependent or endothelial-independent vascular function, arterial stiffness and biomarkers of systemic inflammation by alirocumab, atorvastatin or their combination are secondary objectives. Study design and methods: This is a single-center, randomized, two-period, single-blind, placebo-controlled clinical trial. The study was registered on clinicaltrials.gov (N03273972). It will include 30 healthy low-risk subjects with LDL-C < 4.1 mmol/l. After passing the screening visit (Visit 1), eligible participants will be randomized 1:1 to either subcutaneous alirocumab 150 mg or placebo. These will be administered as single doses in 2 visits 14 days apart (Visits 2 and 3). Atorvastatin 20 mg once nightly will be prescribed for 14 days at Visit 3 in both groups through to Visit 4. At baseline (Visit 2) and during all post-dose visits (Visits 3-4), endothelial function will be assessed using venous occlusion plethysmography. Specifically, changes in forearm blood flow responses to intra-arterial infusions of acetylcholine, sodium nitroprusside and L-NG-monomethyl-arginine acetate will be assessed as surrogates of endothelial-dependent and -independent vasodilatation. Additionally, arterial stiffness and carotid intima-media thickness will be evaluated at the same timepoints. The above-mentioned changes will be correlated with changes in lipid and systemic inflammation biomarkers.

A zero coronary artery calcium score in patients with stable chest pain is associated with a good prognosis, despite risk of non-calcified plaques.

Objectives: To estimate the prevalence of non-calcified coronary artery disease (CAD) in patients with suspected stable angina and a zero coronary artery calcification (CAC) score, and to assess the prognostic significance of a zero CAC in these symptomatic patients. Methods: In this prospective cohort study, consecutive patients with stable chest pain underwent CAC scoring ± CT coronary angiography (CTCA) as part of routine clinical care at a single tertiary centre over 7 years. Major adverse cardiac event (MACE) was defined as cardiac death, non-fatal myocardial infarction and/or non-elective revascularisation. Results: A total of 915 of 1753 (52.2%) patients (mean age 56.8 ± 12.0 years; 46.2% male) had a zero CAC score. Of the 751 (82.1%) patients with a zero CAC in whom CTCA was performed, 674 (89.7%) had normal coronary arteries, 63 (8.4%) had non-calcified CAD with < 50% stenosis and 14 (1.9%) had ≥ 50% stenosis in at least one coronary artery. The negative predictive value of a zero CAC for excluding a ≥ 50% CTCA stenosis was 98.1%. Over a median follow-up period of 2.2 years (range 1.0-7.0 years), the absolute annualised rates of MACE were as follows: zero CAC 1.9 per 1000 person-years and non-zero CAC 7.4 per 1000 person-years (HR 3.8, p = 0.009). However, after adjusting for age, gender and cardiovascular risk factors using a multivariable Cox proportional hazards model, there was no statistically significant difference in the risk of MACE between the two patient cohorts (p = 0.19). After adjusting for age, gender and cardiovascular risk factors, the HR for all-cause mortality among the zero CAC cohort vers non-zero CAC was 2.1 (p = 0.27). Conclusion: A zero CAC score in patients undergoing CT scanning for suspected stable angina has a high negative predictive value for the exclusion of obstructive CAD and is associated with a good medium-term prognosis.

Randomised controlled trial of high versus ad libitum water intake in patients with autosomal dominant polycystic kidney disease: rationale and design of the DRINK feasibility trial.

INTRODUCTION: Vasopressin stimulates cyst growth in autosomal dominant polycystic kidney disease (ADPKD) leading to enlarged kidneys, hypertension and renal failure. Vasopressin receptor blockade slows disease progression. Physiological suppression of vasopressin secretion through high water (HW) intake could achieve a similar effect, necessitating a definitive large-scale trial of HW intake in ADPKD. The objective of the DRINK trial is to answer the key design and feasibility questions required to deliver a successful definitive water intake trial. METHODS AND ANALYSIS: We describe the design of a single-centre, open-label, prospective, randomised controlled trial. The "Determining feasibility of R andomisation to high vs. ad libitum water In take in Polycystic K idney Disease" (DRINK) trial aims to enrol 50 patients with ADPKD, over the age of 16 years with an estimated glomerular filtration rate (eGFR) ≥20 mL/min/1.73 m2. Participants will be randomised 1:1 to HW intake based on an individualised water intake prescription, or to ad libitum (AW) water intake. The HW group will aim for a dilute urine (urine osmolality ≤270 mOsm/kg) as a surrogate marker of vasopressin suppression, and those in the AW group will target more concentrated urine. Participants will have an 8-week treatment period, and will be seen at weeks 0, 2, 4 and 8, undergoing assessments of fluid status, renal function and serum and urine osmolalities. They will receive dietary advice, and self-monitor urine specific gravity and fluid intake. The trial employs smartphone technology to permit home monitoring and remote direct data capture. The primary feasibility end points are recruitment rate and separation between arms in measured urinary osmolality. Key secondary assessments include acceptability, adherence, health-related quality of life, acute effects of HW intake on measured (51Cr-EDTA) and eGFR and ADPKD-related pain. ETHICS AND DISSEMINATION: Ethical approval was awarded by the East of England Essex Research Ethics Committee (16/EE/0026). The results of DRINK will be submitted to peer-reviewed journals, and presented to patients via the PKD Charity. TRIAL REGISTRATION NUMBER: NCT02933268 and ISCRTN16794957.