Dipak Kotecha

Introduction: Blood pressure treatment response is variable in individual patients, and the choice of medical therapy is often dependent on clinician experience. Treatment choices could be personalised by patient empowerment and metabolomic profiles, and augmented by machine learning, but robust evaluation is lacking on how these can be combined to enhance clinical effectiveness. The HYPERMARKER trial will evaluate how an individualised choice of medication class can address the avoidable global health and economic burdens of hypertension. Methods and analysis: The HYPERMARKER trial is a proof-of-concept, pragmatic, multicentre, adaptive, open-label strategy trial embedded into routine clinical practice with stratified individual patient randomisation. The trial was co-designed with a patient and public involvement team. The intervention is a digital portal that supports shared decision-making on hypertension therapy class using clinical features plus metabolomic profiles determined with liquid chromatography-mass spectrometry. Eligible patients are aged≥18 years with a systolic blood pressure≥140 mm Hg and a clinical indication for antihypertensive therapy. 400 participants will be randomised to the usual standard of care for treatment selection or to the intervention group. Remote follow-up will occur through a patient smartphone application and linked blood pressure monitor to assess the primary outcome of change in home systolic blood pressure during a 4-week period after medication changes. Secondary outcomes will include patient-reported adverse effects and quality of life, treatment withdrawal, healthcare utilisation and a health economic analysis. In the second phase of the trial, all participants will receive an updated version of the intervention regardless of the original randomised group. Ethics and dissemination: Ethical approval will be obtained for all sites. Approval in England: North West – Greater Manchester West Research Ethics Committee (REC) (25/NW/0296). Trial results will be disseminated via peer-reviewed publications and plain-language patient summaries. Trial registration number: NCT07294794 and ISRCTN29385951 .

U-Net style networks are commonly utilized in unsupervised image registration to predict dense displacement fields in the full-resolution spatial domain. For high-resolution volumetric image data, this process is, however, resource-intensive and time-consuming. To address this challenge, we propose Fourier-Net+, an image-domain deformable registration framework that operates on real-valued images for broad modality compatibility. Fourier-Net+ uses deterministic Fourier-domain band-limiting for efficient down- and up-sampling and employs a parameter-free, model-driven decoder to learn a band-limited, low-dimensional representation of the displacement field; all learnable network layers are real-valued. In addition, to enhance the registration performance and encourage diffeomorphism, we propose the cascaded and diffeomorphic versions of Fourier-Net+. We evaluate the proposed methods on five datasets, including two brain MRI, one 3-D cardiac MRI (3-D-CMR), one abdominal CT-MR dataset, and one noisy ultrasound cardiac dataset, comparing them against various state-of-the-art approaches. Our Fourier-Net+ and its variants achieve comparable results with these approaches while exhibiting faster training and inference speeds with a lower memory footprint and fewer multiply add operations (mult-adds). For example, on the 3-D-CMR dataset, our Diff-Fourier-Net+ significantly outperforms strong baselines such as TransMorph, TransMatch, and SACB-Net in Dice and HD, as well as in clinical metrics including end-diastolic (ED) volume and ejection fraction estimation, while using substantially less memory and computational cost. This efficiency enables large-scale 3-D registration training on low-VRAM GPUs.

Aims: Co-existence of atrial cardiomyopathy (AtCM) and heart failure with reduced ejection fraction (HFrEF) has prognostic implications. We aimed to quantify the association of a health records data biomarker, FIND-AF, with outcomes in patients with HFrEF and AtCM without atrial fibrillation (AF). Methods: Patients with HFrEF and AtCM without AF at baseline were identified from the ECG-HF registry (2010–6) and linked to the territory-wide Clinical Data Analysis and Reporting System. AtCM was operationalized using a pragmatic multi-domain framework based on the 2025 clinical consensus statement of the Heart Failure Association of the ESC, using electrocardiographic and echocardiographic parameters. Patients were stratified into high and low risk by sex-specific FIND-AF thresholds. The primary outcome was a composite of incident AF, ischaemic stroke, and all-cause mortality. All patients were followed until 31 December 2019. Results: We included 164 patients (mean age 62.3 ± 13.0 years, 73.8% men). The primary outcome occurred in 23/32 (71.9%) high-risk vs 67/132 (50.8%) low-risk patients (OR 2.48, 95% CI 1.07–5.76; P = .035). In the model including both FIND-AF risk group and AtCM severity, high FIND-AF risk remained associated with the composite outcome (adjusted OR 2.70, 95% CI 1.14–6.40; P = .024). Compared with CHA2DS2-VASc, FIND-AF showed superior discrimination for the composite outcome (AUC .735 vs .641; DeLong P = .007) and ischaemic stroke (AUC .700 vs .575; DeLong P = .008). Conclusions: High FIND-AF risk is associated with adverse prognosis in a cohort of patients with HFrEF and AtCM but without baseline AF. These exploratory data require external validation and a prospective study.

Aims: Ethnicity can influence patient outcomes and treatment efficacy, but knowledge is limited on how multimorbidity interacts with clinical events, for example when heart failure (HF) and atrial fibrillation (AF) combine. Methods and Results: 16 713 patients were included from 12 randomized placebo-controlled trials in HF (11 vs beta-blockers and 1 vs spironolactone), of which 13 568 patients (81.2%) were in sinus rhythm and 3145 (18.8%) had comorbid AF at baseline. Non-white ethnicity was recorded in 1899 (11%), with these patients being younger than those of white ethnicity (median age 58 vs 67 years), higher rates of diabetes and hypertension, and lower left-ventricular ejection fraction (median 25% vs 30%). During median follow-up of 1.4 years (interquartile range 0.8–2.3), the primary outcome of all-cause mortality occurred in 394 (21%) non-white patients and 2142 (15%) white patients, with confounder-adjusted hazard ratio (HR) 1.36, 95% CI 1.20–1.54; P < .001. The impact of ethnicity on death was greater in patients with coexisting HF and AF (non-white vs white HR 2.05; 95% CI 1.55–2.70; P < .001) than in those with HF in sinus rhythm (HR 1.24; 95% CI 1.08–1.41; P = .002). The interaction P-value was .003, and confirmed using propensity-score matching to account for baseline differences (P = .009). Similar disparities with ethnicity were seen for the secondary outcomes of cardiovascular and HF-related death, and cardiovascular and HF-related hospitalization. Conclusion: Non-white patients with HF and reduced ejection fraction suffer from substantially higher rates of death than white patients, with comorbid atrial fibrillation leading to significant worsening of this ethnicity-related disparity.

Background and Objective: Blood pressure treatment response is variable in individual patients, and the choice of medical therapy is often dependent on clinician experience. Treatment choices can be personalised by patient empowerment, metabolomic profiles and augmented by machine learning, but robust evaluation is lacking on how these can be combined to enhance clinical effectiveness. The HYPERMARKER trial will evaluate how an individualised choice of medication class can address the avoidable global health and economic burdens of hypertension. Design and Intervention: The HYPERMARKER trial is a proof-of-concept, pragmatic, multicentre, adaptive, open-label strategy trial embedded into routine clinical practice with stratified individual patient randomisation. The trial was co-designed with a patient and public involvement team. The intervention is a digital portal that supports shared decision making on hypertension therapy class using clinical features plus metabolomic profiles determined with liquid chromatography-mass spectrometry. Participants and Outcomes: Eligible patients are aged ≥18yrs with a systolic blood pressure ≥140mmHg and a clinical indication for antihypertensive therapy. 400 participants will be randomised to usual standard of care for treatment selection, or to the intervention group. Remote follow-up will occur through a patient smartphone application and linked blood pressure monitor to assess the primary outcome of change in home systolic blood pressure during a four-week period after medication changes. Secondary outcomes will include patient-reported adverse effects and quality of life, treatment withdrawal, healthcare utilisation and a health economic analysis. In the second phase of the trial, all participants will receive an updated version of the intervention, regardless of original randomised group. Ethics and Dissemination: Ethical approval will be obtained for all sites. Approval in England: North West – Greater Manchester West Research Ethics Committee (REC) (25/NW/0296). Trial results will be disseminated via peer-reviewed publications and plain language patient summaries. Trial registration: Clinicaltrials.gov : NCT07294794 ; ISRCTN: ISRCTN29385951 . STRENGTH AND LIMITATIONS OF THIS STUDY: Hypertension is a major cause of preventable morbidity and mortality, and this study aims to reduce those burdens through machine learning-based integration of metabolomics with clinical factors to enable better personalisation of blood pressure lowering therapy. The HYPERMARKER trial was co-created with patient and public representatives, using digital technology with remote monitoring to facilitate a high level of shared care and patient empowerment. HYPERMARKER is a pragmatic proof-of-concept randomised trial designed to evaluate the potential for future pharmacometabolomic strategies to aid clinical decision-making.

Background and Aims: Atrial fibrillation (AF) disease burden is increasing. Pharmacotherapy of cardio-renal-metabolic diseases may prevent incident AF. This meta-analysis estimates the effect of different pharmacotherapies on risk of incident AF across cardio-renal-metabolic diseases. Methods: The Medline, Embase, and Cochrane Central databases were searched to 7 October 2025 for randomized clinical trials (RCTs) comparing the effect of a non-antiarrhythmic cardio-renal-metabolic medication with control or another agent for incident AF. Random-effects meta-analysis using the Mantel–Haenszel method, with between-study variance estimated using the DerSimonian–Laird method, was performed to synthesize risk ratios (RR) with 95% confidence intervals (CI). Results: Two hundred and forty-nine RCTs involving 745 041 patients were included, of which 207 identified AF through adverse event reports, 161 were placebo-controlled, and 15 had AF as a pre-specified endpoint. In placebo-controlled trials, significant differences in incident AF were observed with treatment of heart failure with reduced ejection fraction with angiotensin-converting enzyme inhibitors and angiotensin receptor blockers (RR 0.69, 95% CI 0.60–0.80), mineralocorticoid receptor antagonists (RR 0.62, 95% CI 0.43–0.90), and sodium-glucose co-transporter 2 (SGLT2) inhibitors (RR 0.62, 95% CI 0.44–0.87); treatment of chronic kidney disease with SGLT2 inhibitors (RR 0.53, 95% CI 0.33–0.85); and treatment of obesity with glucagon-like peptide-1 receptor agonists (RR 0.79, 95% CI 0.63–0.99). However, the number of AF events per trial was low and none were adequately powered for incident AF. Conclusions: Prospective RCTs with AF as a pre-specified outcome should be integrated into the design of future trials of cardio-renal-metabolic medications to determine whether they reduce incident AF.

Aims: Anaemia is common in patients with chronic heart failure (CHF), worsening functional status and prognosis. The role of erythropoiesis-stimulating agents (ESAs) to reduce hospitalization remains unclear. Methods and results: A pre-registered systematic search was performed of randomized controlled trials in adults with CHF and anaemia, comparing any ESA at any dose versus placebo or no treatment. Fourteen trials were included with 3128 participants, with moderate quality of evidence for most outcomes and variable risk of bias. Meta-analysis during 5604 person-years of follow-up demonstrated no significant difference in first CHF hospitalization comparing ESAs with control: Peto odds ratio (OR) 0.93, 95% CI 0.78–1.10, P = .37; trial heterogeneity I2 = 36%. ESAs significantly reduced total CHF hospitalization with 622 events/2698 person-years, compared with 761/2914 person-years for control: incidence rate ratio 0.81, 95% CI 0.73 to 0.90, P < .001; no heterogeneity between trials, I2 = 0%. There were no significant differences between treatment groups for all-cause mortality (OR 1.01, 95% CI 0.86–1.18, P = .88; I2 = 35%) or incident adverse events. Patients randomized to ESAs increased their haemoglobin level [mean difference (MD) 1.6 g/dL compared with control, 95% CI 1.6–1.7, P < .001] and exercise tolerance (MD 69 metres, 95% CI 17–122, P = .009), with lower NYHA class on follow-up (MD −0.73 class, 95% CI −1.11 to −0.36, P < .001). Conclusion: ESA treatment in patients with CHF and mild anaemia can reduce recurrent CHF hospitalization and improve functional capacity, without any impact on adverse events. Including ESAs in CHF management could be considered in patients with anaemia at high risk of recurrent hospital admissions.

Background: The presence and role of endogenous digoxin-like cardiotonic steroids (CTS) in humans is controversial. This study utilises a novel pipeline to quantify CTS and examines their interaction with digoxin within a randomised trial. Methods: The RAte control Therapy Evaluation in permanent Atrial Fibrillation (RATE-AF) trial randomised patients with permanent AF and symptoms of heart failure to low-dose digoxin or beta-blocker therapy; clinicaltrials.gov NCT02391337. Circulating CTS were detected and quantified using a new ultra-high-performance liquid chromatography tandem mass spectrometry (LC–MS/MS) pipeline. Results: All 160 participants of the RATE-AF trial were included, with mean age 76 years (SD 8) and 46% women. Endogenous CTS detected and quantified in baseline samples included digoxigenin and digitoxigenin, plus low or unquantifiable levels of ouabain, telocinobufagin, cinobufagin, marinobufagenin, bufalin, cinobufotalin, dihydroouabain, and ouabagenin. Compared to beta-blockers, patients randomised to digoxin had better functional outcomes at 12 months for heart failure (− 0.57 New York Heart Association class, 95% CI − 0.82 to − 0.32; p < 0.001) and atrial fibrillation (odds ratio 2.24 for a two-class improvement in modified European Heart Rhythm Association class, 95% CI 1.43–3.84; p < 0.001), with lower NT-pro-B-type natriuretic peptide (geometric mean ratio 0.78, 95% CI 0.61 to 0.99; p = 0.006). No interactions were observed for any baseline CTS with each outcome. Digoxin was associated with fewer adverse events (odds ratio 0.16, 95% CI 0.07–0.34; p < 0.001), again without any interaction from circulating CTS. Digoxin levels by LC–MS/MS were strongly correlated with measurement by a clinical immunoassay ( r = 0.87; p < 0.001), and treatment with digoxin did not affect CTS concentrations at 6-month follow-up. Conclusions: A range of CTS are detected in the circulation of patients with atrial fibrillation and heart failure. Within this randomised trial but limited by low circulating levels, CTS do not appear to interact with the ability of digoxin to improve wellbeing compared to conventional first-line treatment with beta-blockers. Graphical