National Institute For Health Research Research Articles

Cost-effectiveness of statin therapy in categories of patients in the UK

Abstract Background Cardiovascular disease (CVD) mortality has declined steadily over the last few decades across Europe and North America. Purpose To provide contemporary estimates of long-term effectiveness and cost-effectiveness of statin therapy in different categories of patients in UK. Methods The CTT-UKB micro-simulation model, developed using the Cholesterol Treatment Trialists' Collaboration data (CTT: 118,000 participants; 5 years follow-up), and calibrated in the UK Biobank cohort (UKB: 502,000 participants; 9 years follow-up). The model integrates parametric risk equations for incident myocardial infarction, stroke, coronary revascularization, diabetes, cancer and vascular and nonvascular death, and projects annually these endpoints and survival using patient characteristics at entry. UKB data and linked primary and hospital care data informed healthcare costs in the model (2020 UK£); 2021 UK NHS Drug Tariff informed statin costs (atorvastatin 40mg at £1.22 and 80mg at £1.68 per 28 tablets); and Health Survey for England data informed health-related quality of life in the model. Previous CTT meta-analysis, atorvastatin dose-response randomized trials, and further meta-analyses of statin trials and cohort studies informed effects of 40mg/80mg atorvastatin therapy daily on rates of incident myocardial infarction, stroke, coronary revascularization, vascular death, diabetes, myopathy and rhabdomyolysis. The model was used to project gains in quality-adjusted life years (QALYs) and additional cost per QALY with lifetime use of atorvastatin 40mg or 80mg daily in categories of UKB participants by sex, age at statin initiation (40–49; 50–59 and 60–70 years), and 10-year CVD risk (QRISK3 risk (%): <5; 5–10, 10–15, 15–20, ≥20). Further scenarios explored effects of 5-year delay of statin initiation in people under 45 years of age or stopping statin therapy at 80 years of age. Results Across men and women in categories by age and CVD risk, lifetime use of atorvastatin 40mg daily was associated with increases in survival by 0.44–1.69 years (0.28–1.02 QALYs), and atorvastatin 80mg daily with increases in survival of 0.45–1.87 years (0.32–1.13 QALYs; Figure 1) with gains larger among participants at higher CVD risk. Both atorvastatin 40mg and 80mg doses were in the range of cost-effective treatments with incremental cost per QALY gained with atorvastatin 40mg daily versus no statin therapy below £7200/QALY and with atorvastatin 80mg vs 40mg daily below £16000/QALY (Figure 2) across all patient categories studied. Compared to lifetime statin therapy, stopping therapy at 80 years of age substantially reduced benefits and was not cost-effective in any patient category studied. Similarly, compared to immediate initiation, 5-year delay of statin therapy in 40–45 years old patients was not a cost-effective. Conclusions In the UK, statin therapy remains highly cost-effective across men and women 40–70 years old, including those at 10-year CVD risk <5%. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): UK National Institute for Health Research (NIHR) Health Technology Assessment (HTA) Programme, UK National Institute for Health Research (NIHR) Health Technology Assessment (HTA) Programme, UK Medical Research Council (MRC), British Heart Foundation

Left ventricular dysfunction, heart failure, and mortality risk factors in de novo pacemaker recipients and those requiring pacemaker generator replacement

Abstract Background Pacing the right ventricle (RV) can lead to adverse remodelling of the left ventricle (LV), LV dysfunction, increased risk of heart failure hospitalisation (HFH) and mortality. Despite RV pacing avoidance programming becoming commonplace, pacemaker patients remain at risk of pacing-induced cardiomyopathy. New pacing strategies and cardiac resynchronisation therapy are available therapeutic options, but it is unclear which patients should be targeted, and when. Purpose This study examined the effects of RV pacing on LV function, the risk of heart failure requiring hospitalisation and all-cause mortality in patients during the first 12 months of pacemaker therapy, compared to those at pacemaker generator replacement (PGR). Methods Data were obtained from a cohort of patients who underwent de novo pacemaker implantation between 2014 and 2017, and patients requiring PGR between 2008 and 2011 at a single tertiary UK hospital, with follow-up from 12 months. Clinical, echocardiographic, and pacemaker variables, medication, and past medical history data were collected. Predictors of a combined endpoint of all-cause mortality or HFH were assessed using cox-regression analysis, with predictors of impaired LV function analysed using multivariable regression analysis. Results 514 newly implanted (NI) patients (mean age, 75 years; 66% male) were recruited, and 491 patients (mean age, 76 years; 56% male) requiring PGR. After a mean follow up of 887 days, 79 NI patients (16%) were deceased (n=27) or had been hospitalised for HF (n=52), whereas after a mean follow up of 668 days, 56 patients after PGR (12%) were deceased (n=34) or had been hospitalised (n=22) for HF. After 12 months of pacemaker therapy, 182 (35%) NI patients had a LV ejection fraction (LVEF) <50%, which had a higher incidence at PGR of 197 (40%). Age was the only significant predictor of mortality or HFH for both NI and PGR (hazard ratio (HR), 1.068; 95% confidence interval (CI), 1.039 to 1.098; p<0.001 vs 1.035; 95% CI, 1.007 to 1.064; p=0.014), respectively). Univariate analysis revealed baseline LVEF (2.439; 95 percent CI, 1.279 to 4.659; p=0.007), RV pacing burden, medication and blood chemistry were significant predictors of the combined outcome at PGR but not at NI (Table 2). Multivariable analysis of predictors of LV impairment (<50% LVEF) showed history of MI (odds ratio (OR), 0.47; 95% CI, 0.29 to 0.78; p=0.003), RV pacing burden (1.01; 95% CI, 1.01 to 1.02; p<0.001) and creatinine (1.06; 95% CI, 1.02 to 1.10; p=0.004) were independently associated in both cohorts. Conclusions Our data suggest that there remains a similar subgroup of people from initial implantation to PGR at risk of LV impairment that might benefit from medication optimisation or novel pacing strategies. Further research is required to better identifying these people to direct more complex therapies to those with the most to gain. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): The research is supported by a National Institute for Health Research (NIHR) clinician scientist fellowship (NIHRCS-2012-032). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. This research was supported by the NIHR Leeds Clinical Research Facility.

Female sex and systolic blood pressure are independently associated with coronary microvascular dysfunction in asymptomatic adults with type 2 diabetes

Abstract Introduction Coronary microvascular dysfunction is frequently reported in people with type 2 diabetes (T2D), is associated with reduced exercise capacity, and is a prognostic marker. Identifying modifiable risk factors associated with microvascular dysfunction may facilitate early intervention to improve outcomes in these patients. Purpose To identify independent determinants of myocardial perfusion reserve (MPR) in asymptomatic adults with T2D and no prevalent cardiovascular disease. Methods Prospective cross-sectional study. People with and without T2D and no signs, symptoms or evidence of cardiovascular disease underwent comprehensive phenotyping with echocardiography, coronary artery calcium scoring, and multiparametric cardiac MRI including adenosine stress and rest perfusion with automated pixel-wise myocardial blood flow (MBF) mapping. Participants with regional perfusion defects indicating obstructive coronary disease or silent myocardial infarct on late-gadolinium enhancement were excluded from analysis. Univariable and multivariable linear regression was performed to identify independent determinants of MPR. Results Two-hundred people with T2D (diabetes duration 11±8 years) were compared with 39 sex- and ethnicity-matched non-diabetic controls (Table 1). People with T2D had higher body mass index (BMI) and ambulatory 24-hour systolic blood pressure (SBP). There was evidence of concentric left ventricular (LV) remodelling (higher LV mass/volume), extracellular matrix expansion (higher ECV fraction), and both systolic and diastolic dysfunction (lower global longitudinal systolic strain and E/A ratio, respectively) in those with T2D. Resting MBF was similar between groups, but stress MBF tended to be lower in T2D compared to controls with significantly reduced MPR in T2Ds (2.87±0.86 vs 3.18±0.82, p=0.043). In univariable analysis, MPR correlated with sex, 24-hour SBP, and E/e' ratio. In a multivariable model adjusting for clinical (age, sex, smoking status, BMI, ambulatory SBP, diabetes duration, HbA1c, low-density lipoprotein, albuminuria) and imaging variables (E/e' ratio, LV mass/volume, global longitudinal strain, myocardial ECV, coronary calcium score) known to affect coronary perfusion, female sex (β=−0.227, p=0.013) and 24-hour SBP (β=−0.275, p=0.001) were the only variables independently associated with MPR. Conclusion Female sex is associated with coronary microvascular dysfunction in asymptomatic people with T2D but not LV mass or myocardial extracellular volume. Systolic BP is the only modifiable independent determinant of MPR and may be an early target for intervention to prevent heart failure development in these patients. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): National Institute for Health Research (NIHR) United Kingdom through a Research Professorship award (RP-2017-08-ST2-007).British Heart Foundation through a Clinical Research Training Fellowship award (FS/16/47/32190).

Unmasking early diastolic dysfunction in type 2 diabetes using the peak early-to-late diastolic strain rate ratio by cardiac MRI feature tracking

Abstract Introduction Type 2 diabetes (T2D) is associated with early alterations in left ventricular (LV) diastolic function, which precedes the development of heart failure. These may be quantified by cardiac MRI feature tracking, which permits a direct assessment of myocardial relaxation through calculation of diastolic strain rates. Previous studies have not fully exploited this technique by integrating early and late LV diastolic relaxation in asymptomatic adults with T2D. Purpose To explore whether the peak early-to-late diastolic strain rate ratio (PEDSR:PLDSR), derived from cardiac MRI feature tracking, could: 1) identify diastolic dysfunction in an asymptomatic cohort of adults with T2D, 2) compare this technique with conventional echocardiographic measures of diastolic function. Methods Prospective cross-sectional study. People with T2D (n=253, mean age 63±7 years, 62% male, diabetes duration 11±8 years) without signs, symptoms or history of cardiovascular disease and non-diabetic controls (n=40, mean age 61±8 years, 63% male) underwent comprehensive phenotyping including echocardiography and cardiac MRI. The LV PEDSR:PLDSR was derived from cardiac MRI balanced steady-state free precession cine images using feature tracking. Correlations between circumferential PEDSR:PLDSR and echocardiography-derived E:A and E/e' were explored. Results Compared to controls, people with T2D had evidence of concentric LV remodelling, diastolic dysfunction, and diffuse myocardial fibrosis as demonstrated by increased LV mass/volume, E/e' ratio, and extracellular volume fraction, respectively (Table 1). People with T2D had lower PEDSR (0.87±0.23 vs 0.95±0.24 s–1, p=0.043), higher PLDSR (0.80±0.22 vs 0.67±0.16 s–1, p<0.001), and lower PEDSR:PLDSR ratio (1.18±0.46 vs 1.53±0.61, p<0.001). PEDSR:PLDSR was moderately correlated with E:A (r=0.495, p<0.001) (Figure 1) but not E/e' (r=0.087, p=0.182). Conclusions Compared to people without diabetes, lower PEDSR:PLDSR in asymptomatic subjects with T2D highlights impaired myocardial relaxation in early diastole and greater dependence on atrial contraction for LV filling in this group. This novel MRI measure may be promising tool to assess global diastolic function in people undergoing cardiac MRI and especially useful in those with poor echocardiographic windows. The functional and prognostic significance of PEDSR:PLDSR warrants further assessment in longitudinal studies. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): National Institute for Health Research (NIHR) United Kingdom through a Research Professorship award (RP-2017-08-ST2-007).British Heart Foundation through a Clinical Research Training Fellowship (FS/16/47/32190).

Association of hepatic steatosis with subclinical cardiac dysfunction in asymptomatic people with type 2 diabetes

Abstract Introduction Non-alcoholic fatty liver disease is highly prevalent among people with type 2 diabetes (T2D) and is an emerging risk factor for heart failure with preserved ejection fraction (HFpEF). Whether excess liver adiposity is simply a marker of the coexisting adverse cardiometabolic risk profile or independently contributes to the development of HFpEF is unclear. Purpose To assess the association between liver fat fraction and subclinical cardiac dysfunction in adults with T2D. Methods Prospective cross-sectional study. Two-hundred and thirty-eight adults with T2D (mean age 63±7 years, 62% males, HbA1c 7.2±1.5%, diabetes duration 10±8 years) with no signs, symptoms, or evidence of cardiovascular disease and 40 age-, sex-, and ethnicity-matched non-diabetic controls (mean age 61±8 years, 63% males, HbA1c 5.2±1.2%) underwent comprehensive phenotyping with echocardiography and multiparametric cardiac MRI including adenosine stress and rest perfusion. Volumetric liver fat fraction (VLFF) was measured using a histologically validated, proprietary MRI technique blinded to all participant details. Inter-study reproducibility was assessed in participants (n=28) who underwent a repeat MRI within two weeks. Linear regression analysis was performed to assess any independent associations between VLFF and identified markers of subclinical cardiac dysfunction in subjects with T2D. Results People with T2D had evidence of concentric left ventricular (LV) remodelling (higher LV mass/volume), extracellular matrix expansion (higher ECV fraction), both systolic and diastolic dysfunction (lower global longitudinal systolic strain and E/A ratio, respectively), and coronary microvascular dysfunction (lower myocardial perfusion reserve) (Table 1). VLFF demonstrated excellent inter-study reproducibility with an intra-class correlation coefficient (ICC) of 0.988 (0.974–0.994). T2Ds had higher VLFF compared to controls [7.5 (3.8–13.7)% vs 2.9 (1.7–4.7)%, p<0.001]. In multivariable regression analysis adjusting for age, sex, ethnicity, body mass index, ambulatory systolic blood pressure, HbA1c, and low-density lipoprotein, VLFF (β=−0.161, p=0.027) was independently associated with E/A, but not other imaging measures of subclinical cardiac dysfunction. Conclusion Liver fat fraction is elevated in people with T2D and is independently associated with early LV diastolic dysfunction. These results add to the growing evidence that ectopic fat plays an important role in the pathogenesis of HFpEF and may be a potential target for intervention. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): National Institute for Health Research (NIHR) United Kingdom through a Research Professorship award (RP-2017-08-ST2-007).British Heart Foundation through a Clinical Research Training Fellowship (FS/16/47/32190).

Melanocortin 4 receptor agonism enhances sexual brain processing in women with hypoactive sexual desire disorder.

BACKGROUNDHypoactive sexual desire disorder (HSDD) is characterized by a persistent deficiency of sexual fantasies and desire for sexual activity, causing marked distress and interpersonal difficulty. It is the most prevalent female sexual health problem globally, affecting approximately 10% of women, but has limited treatment options. Melanocortin 4 receptor (MC4R) agonists have emerged as a promising therapy for women with HSDD, through unknown mechanisms. Studying the pathways involved is crucial for our understanding of normal and abnormal sexual behavior.METHODSUsing psychometric, functional neuroimaging, and hormonal analyses, we conducted a randomized, double-blinded, placebo-controlled, crossover clinical study to assess the effects of MC4R agonism compared with placebo on sexual brain processing in 31 premenopausal heterosexual women with HSDD.RESULTSMC4R agonism significantly increased sexual desire for up to 24 hours after administration compared with placebo. During functional neuroimaging, MC4R agonism enhanced cerebellar and supplementary motor area activity and deactivated the secondary somatosensory cortex, specifically in response to visual erotic stimuli, compared with placebo. In addition, MC4R agonism enhanced functional connectivity between the amygdala and the insula during visual erotic stimuli compared with placebo.CONCLUSIONThese data suggest that MC4R agonism enhanced sexual brain processing by reducing self-consciousness, increasing sexual imagery, and sensitizing women with HSDD to erotic stimuli. These findings provide mechanistic insight into the action of MC4R agonism in sexual behavior and are relevant to the ongoing development of HSDD therapies and MC4R agonist development more widely.TRIAL REGISTRATIONClinicalTrials.gov NCT04179734.FUNDINGThis is an investigator-sponsored study funded by AMAG Pharmaceuticals Inc., the Medical Research Council (MRC) (MR/T006242/1), and the National Institute for Health Research (NIHR) (CS-2018-18-ST2-002 and RP-2014-05-001).

Survival among people with heart failure and atrial fibrillation; a population cohort study

Abstract Background People with chronic heart failure (HF) have a poor prognosis, with survival rates at five year follow-up close to 50%.1 More than half of patients with HF will develop atrial fibrillation (AF). The presence of AF in people with HF has been associated with a poor prognosis, irrespective of left ventricular ejection fraction. 2,3 However, the majority of studies to date have analysed prognosis among secondary care cohorts or randomised trial participants, who may not be representative of patients with chronic HF in the community.2 Purpose To examine the association between survival in patients with HF and AF compared to either condition alone, among a large primary care cohort. Methods We extracted data from the Clinical Practice Research Datalink of primary care records from 1st January 2000 to 31st December 2018 and included all patients aged 45 years and over who were registered at an up-to-standard practice for a minimum of 12 months. Records were linked to Hospital Episode Statistics for secondary care data. The primary outcome was all-cause mortality. Exposure groups were defined as HF+AF, HF or AF, with exposure status treated as a time-varying covariate across follow-up. We used Cumulative Hazard plots to compare survival in people with HF and AF, compared to people with either condition alone or neither. We also report a Cox regression model for risk of all-cause mortality among people with HF and AF, adjusting for age, sex, ethnicity, smoking status and comorbid cardiovascular disease. Results There were 314,042 deaths during the study follow-up. The average age of participants was 58.0 years (SD 10.6) and 51.4% were women. At some point across follow-up, 94,990 people had HF alone, 147,815 had AF alone and 74,470 had both HF and AF. In an unadjusted Cox regression analysis, people with HF and AF were at the greatest risk of death (HR 17.94, 95% CI 17.75 to 18.13), followed by people with HF alone (HR 12.00, 95% CI 11.87 to 12.13), and AF alone (HR 6.14, 95% CI 6.08 to 6.21) compared to people with neither HF nor AF. In the fully adjusted analysis, the risk of death remained highest among people with HF and AF (HR 3.78, 95% CI 3.73 to 3.83), followed by people with HF alone (HR 3.06, 95% CI 3.02 to 3.10), then people with AF alone (HR 1.85, 95%, CI 1.82 to 1.87). In a cumulative hazard plot, the risk of death across follow-up was similar among people with HF and AF, compared to those with HF alone. Conclusion In our large community cohort, we found HF and AF was associated with a worse prognosis than either condition alone. Both HF and AF were also associated with a poor prognosis. These results support the findings of previous secondary care and trial studies regarding the importance of AF as a prognostic indicator among people with HF. Further research could aim to identify preventive strategies that might improve prognosis among this high-risk group of patients. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): The study was undertaken as part of NRJ's Doctoral Research Fellowship, supported by the Wellcome Trust (grant number 203921/Z/16/Z), with additional funding for this project from the National Institute for Health Research (NIHR) Collaboration for Applied Health Research (CLAHRC) Oxford at Oxford Health NHS Foundation Trust (P2-001).

Body mass index and survival in people with heart failure

Abstract Background People with obesity are twice as likely to develop heart failure (HF) compared to people with a healthy body mass index (BMI) [1]. However, among people with HF a higher BMI has been linked to a reduced risk of all-cause mortality, a concept known as the “obesity paradox” [2]. Purpose To examine the association between BMI and survival in patients with chronic HF among a large primary care cohort. Methods We extracted data from the Clinical Practice Research Datalink of primary care records from 1st January 2000 to 31st December 2017 and included 47,531 patients with an incident diagnosis of HF, who were aged 45 years and over and who had a recorded BMI. Patients were stratified into categories of baseline BMI as underweight (BMI <18.5 kg/m2), healthy weight (BMI 18.5 to 24.9 kg/m2), overweight (BMI 25.0 to 29.9 kg/m2) or obese, with obesity split into class I (30.0–34.9 kg/m2), class II (35.0–39.9 kg/m2) and class III (40 kg/m2 and over). The primary outcome was all-cause mortality. We used Kaplan-Meier curves and log rank tests to compare survival in people with HF, based on baseline BMI. We also report a Cox regression model for risk of all-cause mortality among people with HF comparing BMI categories. Results There were 25,013 deaths during the study follow-up. The average age of participants was 77.1 years (SD 10.6) and mean BMI was 27.9 (SD 6.1). In an age- and sex-adjusted analysis, people who were underweight were at increased risk of all-cause mortality compared to people with healthy weight (HR 1.52, 95% CI 1.41 to 1.64). People with overweight (HR 0.81, 95% CI 0.79 to 0.84), obesity class I (HR 0.79, 95% CI 0.76 to 0.82) and obesity class II (HR 0.78, 95% CI 0.74 to 0.82) were at decreased risk of all-cause mortality. People with obesity class III had no difference in risk of death compared to people with healthy weight (HR 0.95, 95% CI 0.88 to 1.02). In a Kaplan-Meier analysis, there was an inverse relationship between body weight and risk of death, even within the first year of follow-up. Conclusion In our large community cohort of people with HF, we found an inverse relationship between BMI and survival. Underweight people with HF have the poorest prognosis and should be identified as high-risk. Conversely, people with HF who are overweight or obese (class I and II) are at lower risk of death confirming the obesity paradox in a real-world primary care population. These findings suggest a more cautious approach to weight management in overweight and obese patients may be needed for people with HF in primary care. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): The SurviveHF study was funded by the National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care (CLAHRC) Oxford at Oxford Health NHS Foundation Trust and the Wellcome Institutional Strategic Fund. The funders did not have any role in the design of the study, analysis and interpretation of the data, or writing of the results for publication.

Impact of cardiovascular events on primary and hospital care costs: findings from UK Biobank study

Abstract Background Need for primary and secondary healthcare increases following cardiovascular disease (CVD) events but there is no data on comparative increases in costs. Purpose To estimate annual primary care and hospital inpatient costs associated with key CVD and other adverse events using the UK Biobank (UKB) individual participant data. Methods UKB participants with linked primary care data (192,983 participants) or hospital inpatient episodes data (all 501,807 participants) contributed data to this study. The three categories of primary care services (patient consultations, diagnostic and monitoring tests, prescription medications), and hospital episodes were costed (2020 UK£) using the NHS England reference costs. Annual primary care costs and, separately, annual hospital inpatient costs were modelled as functions of participant characteristics at entry (socio-demographic, clinical, prior diseases) and time-updated first occurrences of myocardial infarction, stroke, coronary revascularization, incident cancer, incident diabetes, vascular death and non-vascular death during follow-up (p-value <0.01 in stepwise covariate selection). One-part generalized linear regression model (GLM) with Poisson distribution and identity link function was used for primary care costs, and two-part model was used for inpatient costs (part 1: logistic regression models probability of incurring costs; part 2: GLM with Poisson distribution and identity link function models costs conditional on incurring any). Separate models were fitted among participants with and without previous CVD at entry into UKB. Results Most adverse events were associated with excess primary care and hospital inpatient costs. Compared to people without previous CVD, people with previous CVD had on average larger excess primary care and hospital inpatient costs in years with myocardial infarction, stroke and vascular death; but similar excess costs in years with other events. Among both people without and with previous CVD, the excess annual primary care costs were less than 7% of the excess annual hospital inpatient costs for vascular events (Table). However, following diabetes diagnosis the excess annual primary care costs were higher than the excess annual hospital inpatient costs (Table). Conclusions These excess primary and hospital care costs associated with CVD events could inform assessments of interventions and policies to reduce CVD risks in UK. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): UK National Institute for Health Research (NIHR) Health Technology Assessment (HTA) Programme, UK Medical Research Council (MRC), British Heart Foundation

Cardiac magnetic resonance imaging predictors of ventricular arrhythmia in mid-cavity obstructive hypertrophic cardiomyopathy

Abstract Background/Introduction Left ventricular (LV) mid-cavity obstruction (LVMCO) in hypertrophic cardiomyopathy (HCM) is an uncommon phenotypic feature predisposed to the formation of myocardial fibrosis and apical aneurysms (LVAA). These features may be independently proarrhythmic, and LVAA is considered a class 2a indication for implantable cardioverter defibrillator (ICD) in current US, but not European guidelines for the primary prevention of sudden cardiac death (SCD). Cardiac magnetic resonance (CMR) imaging is the preferred modality for detecting these and other phenotypic features critical to SCD risk assessment. Purpose To assess the ability of CMR imaging parameters to predict occurrence of non-sustained ventricular tachycardia (NSVT) in HCM patients with Doppler-derived evidence of LVMCO. Methods Multi-modality imaging records were retrospectively assessed to identify HCM patients with Doppler-LVMCO and CMR scans. CMR images were assessed by an investigator blinded to clinical status. Late gadolinium enhancement (LGE) was quantified using the full-width, half-maximum technique. CMR imaging parameters were assessed for predictive ability using Cox proportional hazards during univariate and multivariate analyses, accounting for time to event (NSVT or censorship of follow-up). Results The study cohort included 58 patients (57±11 years, 74% male) with a median follow-up of 6.2 (IQR 4.3) years. Mean mid-cavity gradient was 33±23 mmHg. NSVT was detected in 27/58 (47%) patients, was 4 beats or longer in 23/27 (85%) and was monomorphic in 21/27 (77%). On univariate analysis, predictors of NSVT during follow-up include LV mass index (HR 1.02, 95% CI 1.00–1.04, p=0.03), LGE in grams (HR 1.04, 95% CI 1.01–1.06, p=0.005), and LVAA (HR 2.57, 95% CI 1.14–5.79, p=0.023). After multivariate adjustment (Table 2), none were significantly associated. Conclusions In LVMCO, magnitude of LV hypertrophy, extent of LGE and the presence of an apical aneurysm may not be independent predictors of ventricular arrhythmias. SCD algorithms based on qualitative assessments of these features may overestimate risk. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): National Institute of Health Research (NIHR)

Impaired myocardial calcium handling in people with type 2 diabetes: an in vivo manganese-enhanced magnetic resonance imaging study

Abstract Background There is a high prevalence of subclinical cardiac dysfunction in people with type 2 diabetes (T2D) which is associated with subsequent development of heart failure. Dysregulated myocardial calcium handling has been demonstrated in animal models of T2D and may be a key mechanism driving the development of heart failure. Manganese-enhanced cardiac magnetic resonance imaging (MEMRI) provides a unique method to assess in vivo myocardial calcium handling. Purpose To determine whether myocardial calcium handling is perturbed in people with T2D with no history of cardiovascular disease. We hypothesised that myocardial manganese uptake would be reduced in people with T2D compared with healthy volunteers. Methods Cross-sectional case-control study, adults with (n=20) and without (n=9) T2D underwent both gadolinium-enhanced MRI and MEMRI. Standard gadolinium-enhanced MRI was used to assess cardiac structure, function and tissue characteristics. MEMRI scans were performed within two weeks of the initial scan. Native T1 maps were obtained in the mid-short axis slice position using a Modified Look-Locker Inversion recovery sequence. An intravenous infusion of manganese dipyridoxyl diphosphate (5 μmol/kg (0.1 mL/kg) at 1 mL/min) was administered and T1 maps at the same location were repetitively acquired every 2.5 min for 30 min. Regions of interest were drawn in the inferoseptal segment and blood pool for all T1 maps from 0 to 30 min by a single observer. The primary outcome was the rate of manganese uptake which was assessed by Patlak modelling as a measure of myocardial calcium handling. Manganese uptake constants were compared using analysis of co-variance, with age, sex and body mass index as co-variates. Results Subjects with T2D were older (62±7 vs. 57±5 years, p=0.046) but body mass index (29.0±4.5 vs. 26.2±3.4 kg/m2, p=0.106), systolic (135±16 vs. 134±17 mmHg, p=0.809) and diastolic (81±10 vs. 83±9 mmHg, p=0.736) blood pressures were similar. Compared to control subjects, participants with T2D had normal systolic function but more concentric left ventricular remodelling (mass/volume ratio 0.90±0.14 vs. 0.71±0.06 g/mL, p<0.001) and reduced peak early diastolic strain rate (0.64±0.17 vs. 0.91±0.26 s–1, p=0.002). Myocardial manganese uptake was substantially reduced in people with T2D compared with controls (6.51±1.46 vs. 8.45±2.52 ml/100 g of tissue/min, p=0.003) (Figure 1). Conclusions For the first time, we have demonstrated in vivo that despite no history of cardiovascular disease and normal systolic function, patients with T2D have marked impairment of myocardial calcium handling. This has potential major implications for the pathogenesis, diagnosis and treatment of diabetic cardiomyopathy. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): British Heart Foundation and National Institute for Health Research

Digoxin improves systolic cardiac function in patients with AF and HFpEF: the RATE-AF randomised trial

Abstract Background The RAte control Therapy Evaluation in permanent AF trial (RATE-AF; NCT02391337) was the first head-to-head controlled trial of beta-blockers versus digoxin in patients with permanent atrial fibrillation (AF) and symptoms of heart failure. Patients randomised to digoxin had similar physical-related quality of life and heart rate, with significantly improved functional class, reduced N-terminal pro-brain natriuretic peptide (NT-proBNP) and substantially less adverse events. The impact of rate control therapy on measures of cardiac function is not currently understood. Purpose To compare the effect of digoxin versus beta-blockers on systolic and diastolic cardiac function according to heart failure sub-type. Methods Blinded echocardiograms assessing systolic and diastolic function were performed at baseline and 12 month follow-up, using a robust method to account for rhythm irregularity (average of three index-beats acquired in appropriate cardiac cycles). Outcomes were the change in left-ventricular ejection fraction (LVEF), systolic tissue Doppler velocity (s'), stroke volume, global longitudinal strain (GLS), diastolic tissue Doppler (e'), mitral E wave deceleration time, E/e', pulmonary vein diastolic deceleration time, isovolumic relaxation time and left atrial ejection fraction. Analyses were stratified by baseline LVEF (≥50%, 40–50% and <40%). Results 160 patients were randomised, of which 145 patients survived to 12-month follow-up with median age 75 years (IQR 69–82) and 44% women. Median baseline heart rate was 96 beats/min (IQR 86–112), blood pressure 135/85 mmHg (IQR 124/77–146/91), NTproBNP 1049 pg/mL (744–1463) and mean NYHA class 2.4 (SD 0.6). In 119 patients with LVEF ≥50% at baseline, diastolic and systolic parameters improved over time with digoxin therapy. There was a significantly greater improvement in systolic function in 63 patients on digoxin compared to 67 with beta-blockers; Figure 1. Patients randomised to digoxin had a higher LVEF at follow-up (adjusted mean difference [AMD] 2.3%, 95% CI 0.3–4.2; p=0.021), higher s' (1.1cm/s, 1.0–1.2; p=0.003) and higher stroke volume (6.5mL, 0.4–12.6; p=0.037) compared to beta-blockers, without any difference in diastolic parameters (Figure 2). In 16 patients with LVEF 40–50% at baseline, s' significantly increased with digoxin compared to beta-blockers (AMD 1.5 cm/s, 1.2–1.7; p=0.001), with no difference for other systolic or diastolic parameters. 10 patients with LVEF <40% at baseline showed no difference between digoxin and beta-blockers for any echocardiographic measures. Conclusion Patients randomised to digoxin with permanent AF, heart failure symptoms and preserved LVEF have significantly greater improvement in multiple parameters of systolic function compared to conventional treatment with beta-blockers. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): National Institute of Health Research

Transaortic flow rate to predict short and long term outcomes in individuals with asymptomatic aortic stenosis

Abstract Background Echocardiographic derived transaortic flow rate (TFR), defined as stroke volume over left ventricular ejection time, has been shown to be associated with increased mortality in asymptomatic mild to severe aortic stenosis (AS) and superior to stroke volume index (SVi) in individuals with symptomatic discordant AS undergoing aortic valve replacement. However, TFR has not been explored alongside SVi in asymptomatic moderate to severe AS, who are a group of interest in risk stratifying for early intervention. Moreover, there is no data where TFR is indexed to body surface area (TFRi). Purpose We explored the prognostic value of TFR, TFRi and SVi in a homogenous cohort of asymptomatic patients with moderate to severe AS. Methods Subjects with asymptomatic moderate to severe AS were prospectively recruited to the Prognostic Importance of Microvascular Dysfunction in asymptomatic patients with AS (PRIMID) study, a multi-centre observational study in the UK conducted between April 2012 and November 2014. All subjects underwent extensive phenotyping with transthoracic echocardiography, bicycle exercise testing and cardiovascular magnetic resonance (CMR) imaging, with blinded core-lab analysis. Patients were followed up in person for a minimum of 12 months, and through health records thereafter. The composite outcome of interest was: cardiovascular mortality, AVR for symptoms and major adverse cardiovascular events (hospitalisation with heart failure, myocardial infarction, syncope and arrhythmia) at one-year and at five years. A cox proportional hazards model was used to calculate a hazard ratio (HR) and 95% confidence intervals (95% CI). Known co-variables associated with the composite outcome were added into the multivariable model. Results Overall, 173 individuals were included with a mean age of 66.3—-±13.3 years and 76.4% were male. Most individuals had severe AS (71.1%, n=123). There were 47 (64.4%) primary outcome events at one-year and 110 (63.6%) events at five-years. Age, sex, N-terminal pro brain natriuretic peptide (NT-pro-BNP), peak aortic velocity (AV Vmax), a positive exercise tolerance test (ETT), myocardial perfusion reserve and right ventricular ejection fraction measured on cardiac magnetic resonance were included in the multivariable model in addition to TFR or TFRi or SVi. Decreasing TFR and TFRi remained independently associated with one-year and five-year composite outcome (Figure 1). However, SVi was only associated with the composite outcome at five-years. AV Vmax (HR: 4.36, 95% CI: 2.59, 7.34, p<0.01) and a positive ETT (HR: 1.87, 95% CI: 1.03, 3.37, p=0.04) were independently associated with the primary outcome at one-year. Conclusion Both TFR and TFRi have a potential role in risk stratifying asymptomatic patients with AS and identifying those for earlier intervention, and may be superior to SVi. However, further prospectively designed studies are needed before this becomes part of the routine clinical practice. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Independent research from a Post-Doctoral Fellowship supported by the National Institute for Health Research (NIHR-PDF 2011-04-51 Geral P McCann).

A service evaluation of patch technology compared with traditional holter monitoring in a stroke pathway

Abstract Background Atrial fibrillation (AF) is a known risk factor for embolic stroke. Establishing a diagnosis of AF will guide clinical decision making and help with downstream secondary stroke prevention. Patch technology is increasingly used for long-term cardiac monitoring for arrhythmia detection, as healthcare organisations look for new innovative ways to streamline service delivery. We were selected as a host site to evaluate the service and patient benefits of Zio XT in AF detection for cryptogenic stroke, as part a Artificial Intelligence Award. Purpose To evaluate the service and patient benefits of patch technology versus traditional Holter in a Stroke / TIA pathway. Methods The Zio XT data were collected between March and December 2021. A total of 778 patients were included. We only selected patients who were referred for a long-term ambulatory monitoring following a recent diagnosis of stroke or transient ischaemic attack (TIA) of an unknown cause (cryptogenic). The stroke pathway operates across three sites and all were included. We undertook a retrospective analysis of our previous pathway using Holter monitoring and sampled 100 stroke / TIA patients who underwent 24–72 hours of monitoring between January to June 2019. In order to obtain the requisite data a variety of sources were scrutinised: referral letters, test reports, patient medical records, internal clinical systems and third-provider platforms. Results The comparison of the ECG technologies showed the Zio XT used less staff time. We found a 50% reduction in-clinic monitor fitting when compared with Holter. Zio XT were all analysed by the provider staff resulting in significant savings in service resource time for analysis. The Zio XT had a faster test turnaround time (referral to report available), 19 days (inclusive of 14 days wear time) for Zio compared with 60 days for Holters. The rapid Zio turnaround led to a 27.5% reduction in repeat clinic appointments, due to results being unavailable for the consultation (table 1). Zio XT also had fewer repeat testing due to inconclusive results, 1% for Zio versus 16% for Holter. AF yield did not vary significantly between technologies, 3% yield for both Holter and Zio, which could be due to the young mean age of the stroke/ TIA the patient cohort: 63.8 years for Holter versus 62.9 years for Zio XT. Conclusion Introducing new ECG technologies such as Zio XT could increase outpatient and diagnostic capacity due to patches being more readily available than Holters, reduced reliance on clinical staff for fitting and analysis as well as fewer repeat clinic appointments. Overall, access to rapid long term cardiac monitoring has benefits over traditional Holter monitoring for TIA/stroke patients and service delivery. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): AI award from the Accelerated Access Collaborative (ACC) in partnership with NHSX and National Institute for Health Research

Epicardial adipose tissue volume and density is associated with cardiac dysfunction in asymptomatic people with type 2 diabetes

Abstract Background Type 2 diabetes (T2D) is associated with several perturbations of cardiac structure and function, which are precursors to the development of heart failure. Excess accumulation of epicardial adipose tissue (EAT) may contribute to cardiac dysfunction in individuals with T2D. Lipid-laden adipocytes have a lower computed tomography (CT) attenuation and can be readily identified using cardiac CT. Using a multimodality cardiac imaging approach, we aimed to assess the association of total and low attenuation EAT volume with early markers of cardiac dysfunction in people with T2D. Methods Prospective case-control study, in which participants with and without T2D and no known cardiovascular disease, underwent comprehensive cardiovascular phenotyping including multiparametric cardiac magnetic resonance imaging (MRI), echocardiography and non-contrast cardiac CT. EAT volume was measured from CT scans using a deep learning method and volumes indexed to body surface area. Total EAT was defined according to CT adipose tissue attenuation range of −30 to −190 Hounsfield Units (HU) and low attenuation EAT as −90 to −190 HU. Left ventricular (LV) volumes, function and strain measurements were derived from cardiac MRI images and diastolic function also assessed using echocardiography. Markers of early cardiac dysfunction in those with T2D were assessed for associations with EAT in T2D participants using multivariable linear regression analyses. Results Two hundred and fifty-four participants were included: demographic, anthropometric and imaging variables are displayed in Table 1. Subjects with T2D had increased LV concentric remodelling (higher LV mass/volume ratio), diastolic dysfunction (lower circumferential peak early diastolic strain rate (PEDSR) and average E/e') and reduced systolic function (global longitudinal strain, GLS) compared with controls. Total and low attenuation indexed EAT volumes were 1.6-fold and 2-fold higher, respectively, in participants with T2D compared to controls (Figure 1). After adjustment for age, gender, ethnicity, insulin resistance, systolic blood pressure and waist/hip ratio, total and low attenuation indexed EAT volume were independently associated with LV mass/volume ratio (total indexed EAT volume: β=0.002, p=0.02, low attenuation indexed EAT volume: β=0.004, p=0.01) and LV GLS (total indexed EAT volume: β=−0.02, p<0.01, low attenuation indexed EAT volume: β=−0.04, p=0.02) in subjects with T2D, but not indices of diastolic dysfunction. Conclusion Total and low attenuation EAT volumes are higher in individuals with T2D, and excess EAT accumulation is independently associated with early markers of cardiac dysfunction. Further studies into the underlying mechanisms of this interaction may facilitate the development of interventions targeted at EAT, which could mitigate against the development of heart failure in people with T2D. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): United Kingdom National Institute for Health Research

Direct oral anticoagulants compared to vitamin K antagonists for the treatment of left ventricular thrombi

Abstract Background Left ventricular thrombus (LVT) complicates around one in six cases of acute and chronic left ventricular systolic dysfunction and is associated with an increased risk of stroke, major systemic embolism and death, believed to be ameliorated by anticoagulation. Off-label use of direct oral anticoagulants (DOACs) for LVT has steadily increased, largely based on favourable outcomes in atrial fibrillation and venous thromboembolism, but the safety and efficacy of DOACs versus vitamin K antagonists (VKA) for LVT remains uncertain. Purpose The main aim of our study was to compare treatment of LVT with VKA to DOAC, focusing on all-cause mortality, stroke, major systemic emboli and major bleeding. Methods We conducted a retrospective observational longitudinal study of patients presenting to two large quaternary centres between 2011 and 2021 with a diagnosis of LVT. Patients were eligible if they had a documented LVT and received anticoagulation with either VKA or DOAC. Baseline data, thrombus characteristics, treatment type and duration, follow up imaging and clinical events were recorded using electronic health care records. Outcome measures included thrombus resolution, stroke and systemic embolism (SSE), major bleeding and mortality. Results A total of 955 patients were identified, of whom 901 received treatment with either a VKA (567 pts, 62.9%) or a DOAC (334 pts, 37.1%) and were included in the analysis. Underlying aetiologies included acute myocardial infarction (AMI) (38.3%), chronic ischaemic cardiomyopathy (38.0%) and non-ischaemic cardiomyopathy (23.7%). Rivaroxaban (43.4%) was the most frequently prescribed DOAC followed by apixaban (35.9%), and the remaining on edoxaban (20.7%). AMI related LVT was more commonly treated with DOAC (53.0%) and chronic ischaemic cardiomyopathy with VKA (72.9%). There was a lower baseline LVEF in the VKA cohort (29.5±13.2 vs 33.1±14.2, p<0.0001). Other demographic features were comparable. Median follow up was 2.5 years (IQR: 1–3.5). There were no differences in follow up duration between the two treatments (p=0.17). Greater rates of thrombus resolution were seen in the DOAC group compared to VKA (1 year: 78.4% vs 51.4%, p<0.0001), with higher rates of persistent thrombus over the follow-up period seen in the VKA group (25.1% vs 12.9%, p<0.0001). Rates of stroke and systemic embolization were similar between the groups (VKA 9.3% vs 9.6% DOAC, p=0.93). Higher rates of bleeding (BARC >3, 8.1% VKA, 3.6% DOAC, p=0.031) (Figure 1A) and mortality (VKA 18.5%, DOAC 10.2%, p=0.001) (Figure 1B) were seen in the VKA group over the follow-up period. Conclusions In a large multi-centre registry of LVT of mixed aetiology, anticoagulation with DOAC was associated with earlier and greater rates of thrombus resolution and consequential reduced adverse events (major bleeding and mortality) during follow up. A funding application to support a multi-centre randomised control trial is underway. Funding Acknowledgement Type of funding sources: Other. Main funding source(s): This work was supported by the British Heart Foundation (Fellowship FS/CRTF/21/24190 to HM) and the National Institute for Health Research (Biomedical Research Centre Award to Guy's and St Thomas' NHS FT and King's College London).

Inter-field strength agreement of cardiovascular magnetic resonance cine-derived strain and strain rate measures: a randomised study

Abstract Background Left ventricular (LV) strain and strain rate measurements can be derived from routinely acquired cardiovascular magnetic resonance (CMR) cine images by feature tracking techniques. However, the inter-field strength agreement of strain measurements derived from these techniques is not known. We hypothesised that there would be excellent inter-field strength agreement (between 1.5 and 3 Tesla [T]) for the measurement of global strain and strain rate derived from cine imaging. Methods Prospective, randomised cross-over observational study. Healthy volunteers each underwent CMR scans at 1.5T and 3T within 30 minutes on the same day in a randomised order. Retrospectively ECG gated, short and long-axis balanced steady state free precession cine images were obtained using standardised acquisition parameters at both field strengths. Two software packages were used to derive LV global longitudinal, circumferential and long and short axis radial systolic strain, peak systolic, early diastolic and late diastolic strain rates. All strain values are expressed as positive numbers. Results Twenty-two subjects (mean age 36±12 years; 45% male) were studied. No differences in heart rate and blood pressure measurements during scanning were observed between field strengths. The abstract figure shows an example of strain analysis and Bland-Altman plots for global longitudinal and circumferential strain. Minimal bias was seen in all strain and strain rate measurements between field strengths using the first software package. Strain and strain rate values derived from long axis images (longitudinal and long axis radial) showed poor to fair agreement (intraclass correlation co-efficient (ICC) range 0.39–0.71), whereas measures derived from short axis images (circumferential and short axis radial) showed good to excellent agreement between field strengths (ICC range 0.78–0.91). Similar results were observed with the second software package, though the differences in agreement between long and short axis derived measures were less pronounced. Conclusion Longitudinal strain and strain rate measures derived from CMR feature tracking have poor inter-field strength agreement between 1.5T and 3T. By contrast, agreement of circumferential and short axis radial strain and strain rate measurements at 1.5T and 3T is good. These results need to be considered when assessing strain at different field strengths. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): United Kingdom National Institute for Health Research

Temporal trends and patterns in atrial fibrillation incidence: a population-based study of 3.4 million individuals

Abstract Background Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia worldwide; confers an increased risk of adverse outcomes and once diagnosed most patients will require lifelong-treatment. Population-based studies of AF incidence are needed to inform health-service planning. However, few reports refer to a standard population and estimates of incidence in high income countries vary by over 12-fold. Purpose We aimed to assess trends in the crude and standardised atrial fibrillation incidence by sex, age, socioeconomic status and region in a large general population cohort from England. We also investigated the comorbidity profile of patients over almost two decades. Methods We used linked primary and secondary electronic health records of 3.4 million individuals from the Clinical Practice Research Datalink (CPRD). The CPRD database contains anonymised patient data from approximately 7% of the UK population and is broadly representative in terms of age, sex, and ethnicity. Eligible patients aged 16 years and older contributed data between Jan 1 1998 and Dec 31 2017. We defined incident AF diagnosis as the first record of AF in primary care or hospital admission records from any diagnostic position. For incidence calculations, we excluded all individuals who had a diagnosis of AF before the study start date, or within the first 12 months of registration with their general practice. For patients with incident AF, we extracted baseline characteristics, comorbidities, socioeconomic status and geographic region. We calculated standardised rates by applying direct age and sex standardisation to the 2013 European Standard Population. We inferred crude rates by applying year-specific, age-specific and sex-specific incidence to UK census mid-year population estimates. Results From 1998 to 2017 AF incidence (standardised by age and sex) increased by 30% (from 247 to 322 per 100,000 person-years; adjusted incidence ratio [IRR] 1.30, 95% CI 1.27–1.33) (Figure 1). Absolute number of incident AF increased by 72% (from 117,880 in 1998 to 202,333 in 2017), due to an increasing number of older people. Comorbidity burden at diagnosis of AF increased (2.58 [SD 1.83] vs 3.74 [2.29] conditions; adjusted difference 1.26, 95% CI 1.14–1.39). Age-standardised incidence was higher in men than women (IRR 1.49; 95% CI 1.46–1.52), and men were younger at diagnosis (adjusted difference 5.53 years; 95% CI 5.36–5.69). Socioeconomically deprived individuals had more comorbidities and were more likely to develop AF than the most affluent individuals (IRR 1.20; 95% CI 1.15–1.24). Over time, the age of AF diagnosis declined disproportionately in the most deprived individuals (Figure 2). Conclusion In England AF incidence has increased, and the socioeconomic gradient in age at diagnosis and comorbidity burden widened. This changing burden of AF requires policy-based interventions to prevent associated morbidity and mortality. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): British Heart Foundation, National Institute of Health Research

Benefit accrual with cardiovascular disease prevention and effects of discontinuation: a modelling study

Abstract Background Statin therapy reduces rates of heart attacks and strokes and improves survival in people at increased cardiovascular disease (CVD) risk. However, there is some uncertainty when to start and how long to persist with statin therapy so as to optimise benefits. Purpose To project the accrual of benefit with statin therapy in population groups by age at therapy initiation using a newly developed micro-simulation model. Methods Participants without previous CVD (N=44,412) and with previous CVD (N=13,061) at entry were randomly selected from the UK Biobank cohort, ensuring sufficient representation in respective categories by age, LDL cholesterol, diabetes and 10-year CVD risk categories (QRISK3 score, for those without previous CVD only). The CTT-UKB model, a CVD micro-simulation model [1], was used to predict subsequent survival and quality-adjusted life years (QALYs) of the participants using their characteristics at entry. Treatment with atorvastatin 40mg daily was used as an example to illustrate the effect of the therapy compared to no such therapy. Scenarios include: (1) lifelong preventive therapy, (2) preventive therapy stopped at 80 years of age, and (3) delayed initiation of preventive therapy by 5 years in participants under 45 years of age. Results Statin treatment benefits, measured in QALYs gained, accrue over lifetime. The majority of benefits accrue later in life. Men accumulate larger benefits and earlier than women (Figure 1A). The pattern of benefits accrual is similar for participants with and without previous CVD (data not shown). The higher the participants' CVD risk, the larger and earlier the benefits, with younger participants accruing larger benefits (Figure 1B). Compared with lifelong prevention, stopping treatment at 80 years of age leads to large reductions in overall benefits, especially in women and those at lower CVD risk. For example, compared to lifelong therapy, people without previous CVD who initiate therapy in their 50s, would lose 47% of QALYs benefit (if men), 66% (if women), 73% (if with CVD risk <5%), and 35% (if with CVD risk ≥20%), respectively, if they stop treatment when they reach 80 years of age. Five-year delay of statin therapy initiation in people under 45 years of age reduces their benefits by about 4% on average, though the loss is somewhat larger in people at higher CVD risk (Figure 2). Conclusion Benefits from lifelong cardiovascular prevention accrue over peoples' lifespan with large share of benefits accruing at older age. Stopping treatment earlier substantially reduces benefits. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): UK NationalInstitute for Health Research (NIHR) Health Technology Assessment (HTA) Programme, UK Medical Research Council (MRC), and British Heart Foundation

Measurable residual disease in childhood B-cell acute lymphoblastic leukemia.

Measurable residual disease in childhood B-cell acute lymphoblastic leukemia.