William Woodward’s research while affiliated with University of Oxford and other places

Background People born preterm (<37 weeks' gestation) have a potentially adverse cardiac phenotype that progresses with blood pressure elevation. Objectives The authors investigated whether preterm-born and term-born adults exhibit similar cardiac structural and functional remodeling following a 16-week aerobic exercise intervention. Methods We conducted a randomized controlled trial in 203 adults (aged 18-35 years) with elevated blood pressure or stage 1 hypertension. Participants were randomized 1:1 to a 16-week aerobic exercise intervention or to a control group. In a prespecified cardiovascular magnetic resonance imaging (CMR) substudy, CMR was performed at 3.0-Tesla to assess left and right ventricular (LV and RV) structure and function before and after intervention. Results A total of 100 participants completed CMR scans at baseline and after the 16-week intervention, with n = 47 in the exercise intervention group (n = 26 term-born; n = 21 preterm-born) and n = 53 controls (n = 32 term-born; n = 21 preterm-born). In term-born participants, LV mass to end-diastolic volume ratio decreased (−3.43; 95% CI: −6.29 to −0.56; interaction P = 0.027) and RV stroke volume index increased (5.53 mL/m²; 95% CI: 2.60, 8.47; interaction P = 0.076) for those in the exercise intervention group vs controls. No significant effects were observed for cardiac structural indices in preterm-born participants. In preterm-born participants, LV basal- and mid-ventricular circumferential strain increased (−1.33; 95% CI: −2.07 to −0.60; interaction P = 0.057 and −1.54; 95% CI: −2.46 to −0.63; interaction P = 0.046, respectively) and RV global longitudinal strain increased (1.99%; 95% CI: −3.12 to −0.87; interaction P = 0.053) in the exercise intervention group vs controls. No significant effects were observed for myocardial deformation parameters in term-born participants. Conclusions Aerobic exercise training induces improved myocardial function but not cardiac structure in preterm-born adults.

Aims To assess the real world impact of updated clinical guidelines and literature on the management of patients undergoing stress echocardiography for the assessment of inducible ischaemia across a national health service. Methods and Results A total of 13,819 patients from 32 UK hospitals, referred for stress echocardiography between 2015-2023, were analysed across two phases: phase 1 (2015-2020) and phase 2 (2020-2023). Follow-up data for one year was available for 4,920 participants through NHS Digital. Patients in phase 2 were younger, and presented with a higher cardiovascular risk profile, although sex distribution remained similar across phases. There was an observed reduction in invasive angiography referrals within one year following a positive stress echocardiogram (p<0.01), which appeared to be attributed to changes in management of patients with moderate ischaemia (3-4 segments; p<0.01). For those who did receive invasive assessment, there were no changes in intervention rate (p=0.27), regardless of ischaemic burden. This trend was most evident in centres performing a higher volume of stress echocardiograms. Conclusion Coronary disease management pathways have changed within the UK and fewer patients with moderate ischaemia are undergoing invasive coronary angiography. However, coronary intervention rates are unchanged, suggesting stress echocardiography is being used to improve patient selection for invasive procedures, while minimising unnecessary referrals. Future work will assess if this reduction in angiography referrals is maintained long term, and if there are any effects on patient outcomes.

Background/Introduction Recent guidelines and randomised controlled trials have proposed less frequent use of invasive strategies for patients with stable chest pain, and demonstrated the non-inferiority of an initial medical management strategy. However, the impact of these updates on clinical practice in the United Kingdom has not been previously reported. Purpose We aimed to identify if recent guidelines and literature have changed the management of patients being referred for stress echocardiography to assess inducible ischaemia in a real-world, multi-centre national healthcare system, and determine if this has significantly impacted patient outcomes. Methods Patients were recruited to the EVAREST/BSE-NSTEP prospective cohort study between 2015 and 2023 across the UK as part of two phases (phase 1: Mar 2015-Sep 2020; phase 2: Oct 2020-Sep 2023). Patient demographics, and stress echocardiogram procedural details were collected by study team members at each participating hospital, and 1 year medical outcome data was collected via the Data Access Request Service from NHS England. Patient management decisions and outcomes including referral for invasive coronary angiography, mortality, and cardiovascular events were compared between recruitment phases using Kaplan-Meier analysis and Cox proportional hazard ratios. Results 5,082 participants were included in this analysis of patient management and outcomes (2,613 in phase 1, and 2,469 in phase 2), with a consistent age, sex, and BMI between recruitment cohorts. There was a higher rate of smoking and hypercholesterolaemia in phase 2 participants, but a decrease in hypertension and peripheral vascular disease (Table 1). There was a decrease in referral rate for invasive angiography for participants with a positive stress echocardiogram in phase 2 (p<0.05; Fig. 1B). Overall, participants in phase 2 had a higher rate of mortality (p<0.01) than those in phase 1, but there was no difference in cardiac event rate between phases (Fig. 1C-F). These results remained consistent after adjusting for covariates, with a reduced hazard ratio for invasive angiography in phase 2 participants who had a positive stress echocardiogram (HR 0.78, 95% CI 0.66-0.92, p<0.01). Additionally, phase 2 participants had an increased hazard ratio for mortality (HR 1.93, 95% CI 1.05-3.55, p<0.05), but no significant increase in risk of cardiac event (HR 0.69, 95% CI 0.43-1.12, p=0.13). Conclusions Since 2020, participants are being less frequently referred for invasive coronary angiography after stress echocardiography possibly in favour of an initial medical management strategy. However, this reduction in referral to coronary angiography does not appear to have a negative impact on patient health as measured by cardiac event rate.

Background/Introduction EVAREST is a large-scale prospective, multicenter, observational study evaluating the use and accuracy of stress echocardiography (SE) in patients recruited from 31 NHS hospitals. Purpose Sub-study analysis to assess whether SE practice and outcome varies with age of patients. Methods Patients were recruited sequentially on attendance at the SE clinic between March 2015 and March 2020. Differences in SE characteristics and outcomes were compared between two age groups separated by the median age of the study population into Younger and Older patients. Results We studied 7846 pts with a median age for all pts of 66 yrs. The mean age in the Younger Group was 55 (range 20-65) years and in the Older Group was 74 (range 66-94) years. An abnormal SE response occurred more commonly in the Older Group (19% vs 13%, p<0.001). Key demographic and risk factors that differed between groups were a higher proportion of females in the Older Group (2185 male, 1769 female) compared to the Younger group (2255 male, 1626 female), p=0.011, higher mean BMI in the Younger Group (29.61±5.8 vs 28.1±5.2 kg/m², p<0.001) and lower resting pulse pressure in the Older Group at 56.3±15.9 (95%CI 55.8-56.8) compared to 65.9±16.6 (95%CI 65.3-66.4), p<0.001. Exercise test was performed more commonly in the Younger Group compared to the Older Group (39% vs 23%, p<0.001). For those undergoing dobutamine stress, atropine was used more commonly in the Younger Group (36% vs 34%, p<0.001). However, dobutamine dosing was similar between groups, with the highest dose (40 µg/kg/min) achieved in 13% of the Younger Group and 14% of the Older Group, p=0.05. Ultrasound contrast was used less frequently in the Older Group (73% vs 70%, p=0.03) compared to the Younger group, (p<0.001). Complication rates were similar between groups with the Older Group having a lower peak SBP (150.7±29.3 (95%CI 149.7-151.7) vs 154.9±30.3 mm Hg (95%CI 153.8-155.9), p<0.001) and similar rates of an exaggerated BP response (defined as a peak SBP ≥220 mm Hg) evident in both groups (20% of the Younger Group compared to 21% of the Older Group, p=0.55). A low BP response (defined as peak systolic blood pressure <120 mm Hg) occurred in 8% of the Younger Group and 11% of the Older Group (p=0.001), while a severe hypotensive response (peak SBP<82 mm Hg) was very rare and occurred 0.31% of the Younger Group and 0.56% of the Older Group (p=0.09). Conclusions Stress echocardiography is being performed in patients with a wide age range from 20 to 94 years of age within the UK. Older patients undergoing stress echocardiography are more likely to be female and have a relatively lower cardiovascular risk profile. Delivery of stress echocardiography differs significantly with age, with lower contrast use and more frequent use of dobutamine. However, there is no evidence of higher rates of complications in older patients that might limit use of stress echocardiography in older populations.

Background/Introduction Stress echocardiography is widely used to assess patients with chest pain. The clinical value of a positive or negative test result to inform on likely longer-term outcomes when applied in real-world practice across a national healthcare system has not been previously reported. Purpose We report patient outcome up to five years after stress echocardiogram for patients undergoing a clinically-indicated stress echocardiogram for ischaemic heart disease in a real-world, multi-centre national healthcare system and determine the event-free ‘warranty’ period following a normal stress echocardiogram. Methods Patients participating in the EVAREST/BSE-NSTEP (Echocardiography: Value and Accuracy at Rest and Stress/ British Society of Echocardiography National Review of Stress Echo Practice) prospective cohort study were recruited from 32 UK NHS hospitals between 2018 and 2022. Medical outcome data up to 2023 was available from NHS England. Stress echocardiography results were related to outcomes, including death, procedures, hospital admissions and relevant cardiovascular diagnoses, using Kaplan-Meier analysis and Cox proportional hazard ratios. Warranty period following a negative stress echocardiogram was defined as time the cumulative event rate of myocardial infarction or cardiac-related death remained below 5%. Results 5503 patients were included in the analysis with a median follow-up of 829 days (IQR 224 – 1434). A positive stress echocardiogram was associated with a greater risk of myocardial infarction (HR 2.71, 95% CI 1.73 – 4.24, P<0.001), revascularisation (HR 10.52, 95% CI 8.26 – 13.42, P<0.001) and reaching a composite endpoint of cardiac-related mortality and myocardial infarction (HR 2.04, 95% CI 1.41 – 2.93, P<0.001). Hazard ratios for all events increased with ischaemic burden identified at stress echocardiography (Figure 1). A negative stress echocardiogram identified an event-free ‘warranty period’ of at least five years in patients with no prior history of coronary artery disease, and four years for those with disease (Figure 2). Conclusions In real-world practice, the degree of ischaemia recorded by clinicians at stress echocardiography correctly categorises risk of future events over the next five years. Reporting a stress echocardiogram as negative correctly identifies patients with no greater than a background risk of cardiovascular events over a similar time period. Figure 1 Figure 2

Introduction The British Society of Echocardiography National Review of Stress Echo Practice (BSE N-STEP) is a large, prospective, multicentre study investigating stress echocardiography (SE) practice within the UK National Health Service. We used this dataset to identify, categorise and compare the characteristics of the workforce groups delivering SE within this real-world study cohort. Method Recruitment occurred between September 2020 and June 2023 across 34 NHS trusts. All patients provided informed consent and underwent SE following the local policy of each trust. Baseline characteristics of participants and details of SE procedure were recorded including stress modality and test outcome data. In addition, workforce involved in delivery of each test was self-reported including a description of staff responsible for test delivery, reporting and overall test supervision. Based on this, studies were categorised as either doctor-led (DL) which included both consultants and doctors in training, or cardiac physiologist/nurse-led (CNL). Frequency data was reported for each workforce group with Chi-square analysis used to identify differences between categorical variables. Results Data from 8506 patients were included in this analysis. SE supervision was identified as DL in 4839 (56.9%) and CNL in 3636 (42.7%) of which 79% was supervised by a cardiac physiologist. 28 hospitals reported both DL and CNL services. CNL services tended to be within higher volume centres, so that 4 trusts (two regional hospitals and two tertiary) contributed 71% of the total SE activity. Patients in CNL services were of similar age compared to DL service but were more likely to be male (p=0.008) and had a higher risk factor profile including hypertension (54.9% vs 50.1%), hypercholesterolaemia (50.3% vs 39.6%), and diabetes (24.9% vs 18.8%) (p=<0.001). The proportion of patients seen for ischaemic heart disease tests was similar between DL and CNL clinics (89.1% vs 89.7%, p=ns) but CNL services performed more viability (0.8% vs 1.2%, p= 0.04) and pre-op studies (2.6% vs 3.4%, p= 0.03). DL services more commonly performed dobutamine stress studies (63.0% vs 56.3%, p=<0.001) whilst the CNL group performed more exercise SE (42.8% vs 36.4%, p=<0.001). Positivity rates for SE were similar across DL and CNL groups (17.1% vs 17.7%) as were inconclusive/ abandoned tests (3.8% vs 3.6%, p=ns). However, there was a lower rate of reported complications in the CNL services (2.2% vs 5.3% (p= <0.001). Discussion: Supervision of SE testing has traditionally been led by cardiologists with nursing and physiologist staff acting in supporting roles. This study demonstrates that the UK SE workforce is evolving, with non-medical-led services contributing significantly to the volume of SE activity although clear differences in the test modality remain between DL and CNL services at present.

Aims Stress echocardiography is widely used to assess patients with chest pain. The clinical value of a positive or negative test result to inform on likely longer-term outcomes when applied in real world practice across a healthcare system has not been previously reported. Methods and Results 5503 patients recruited across 32 UK NHS hospitals between 2018 and 2022, participating in the EVAREST/BSE-NSTEP prospective cohort study, with data on medical outcomes up to 2023 available from NHS England were included in analysis. Stress echocardiography results were related to outcomes, including death, procedures, hospital admissions and relevant cardiovascular diagnoses, based on Kaplan-Meier analysis and Cox proportional hazard ratios. Median follow-up was 829 days (IQR 224–1434). A positive stress echocardiogram was associated with a greater risk of myocardial infarction (HR 2.71, 95% CI 1.73–4.24, P<0.001), and a composite endpoint of cardiac-related mortality and myocardial infarction (HR 2.03, 95% CI 1.41–2.93, P<0.001). Hazard ratios increased with ischaemic burden. A negative stress echocardiogram identified an event-free ‘warranty period’ of at least five years in patients with no prior history of coronary artery disease, and four years for those with disease. Conclusions In real-world practice, the degree of myocardial ischaemia recorded by clinicians at stress echocardiography correctly categorises risk of future events over the next five years. Reporting a stress echocardiogram as negative correctly identifies patients with no greater than a background risk of cardiovascular events over a similar time period.

Background People born preterm (<37 weeks’ gestation) have a potentially adverse cardiac phenotype that progresses with blood pressure elevation. We sought to determine whether preterm-born adults with elevated and stage 1 hypertension exhibit similar cardiac structural and functional remodeling following a 16-week aerobic exercise intervention as their term-born peers. Methods We conducted an open, parallel, two-arm superiority randomized controlled (1:1) trial in n=203 adults aged 18-35 years old with elevated or stage 1 hypertension. Participants were randomized to a 16-week aerobic exercise intervention targeting three, 60-minute supervised sessions per week at 60-80% peak heart rate (exercise intervention group) or sign-posted to educational materials on hypertension and recommended lifestyle behaviors (control group). In a pre-specified cardiovascular magnetic resonance imaging (CMR) sub-study, CMR was performed on a 3.0-Tesla Siemens scanner to assess left ventricular (LV) and right ventricular (RV) structure and function before and after the 16-week intervention period. Group differences in outcome variables after the intervention were examined using analysis of covariance (ANCOVA) adjusting for baseline outcome variables, sex, and age. Interactions between birth category and intervention effect were investigated by including an interaction term in the intervention ANCOVA analyses. Results One hundred participants completed CMR scans both at baseline and after the 16-week intervention, with n=47 in the exercise intervention group (n=26 term-born; n=21 preterm-born) and n=53 controls (n=32 term-born; n=21 preterm-born). There was a significantly different response to the exercise intervention between preterm- and term-born participants. In term-born participants, LV mass to end-diastolic volume ratio decreased (adjusted mean difference: -3.43, 95% CI: -6.29,-0.56, interaction term p=0.027) and RV stroke volume index increased (adjusted mean difference: 5.53mL/m ² , 95% CI: 2.60,8.47, interaction term p=0.076) for those in the exercise intervention group versus controls. No significant effects were observed for cardiac structural indices in preterm-born participants. In preterm-born participants, LV basal- and mid-ventricular circumferential strain increased (adjusted mean difference: -1.33, 95% CI: -2.07,-0.60, interaction term p=0.057 and adjusted mean difference: -1.54, 95% CI: -2.46,-0.63, interaction term p=0.046, respectively) and RV global longitudinal strain increased (adjusted mean difference=-1.99%, 95% CI=-3.12,-0.87, interaction p=0.053) for those in the exercise intervention group versus controls. No significant effects were observed for myocardial deformation parameters in term-born participants. Conclusions Aerobic exercise training induces improved myocardial function but not cardiac structure in preterm-born adults.

Background People born preterm (<37 weeks’ gestation) have a potentially adverse cardiac phenotype that progresses with blood pressure elevation and may explain their increased risk of early heart failure and cardiovascular-related mortality. It remains unknown whether lifestyle or pharmacological interventions lead to beneficial left ventricle (LV) and right ventricle (RV) structural and/or functional changes in preterm-born adults. Purpose To determine whether preterm-born adults with elevated and stage 1 hypertension exhibit similar cardiac remodelling following a 16-week aerobic exercise intervention as their term-born peers with comparable blood pressures. Methods We conducted an open, parallel, two-arm superiority randomised controlled (1:1) trial in n=203 adults aged 18-35 years old with elevated or stage 1 hypertension. Participants were randomised to a 16-week aerobic exercise intervention targeting three, 60-minute supervised sessions per week at 60-80% peak heart rate (exercise intervention) or sign-posted to educational materials (controls). In a pre-specified cardiovascular magnetic resonance imaging (CMR) sub-study in n=100 individuals, CMR was performed before and after the 16-week intervention. Group differences in outcome variables after the intervention were examined using ANCOVA adjusting for baseline outcome variables, sex, and age. Interactions between birth category and intervention effect were investigated by including an interaction term in the intervention ANCOVA analyses. Results One hundred participants completed CMR scans at baseline and after the 16-week intervention, with n=47 in the exercise intervention group (n=26 term-born; n=21 preterm-born) and n=53 controls (n=32 term-born; n=21 preterm-born). Cardiac structural measures in preterm-born adults did not change following the intervention. However, term-born adults in the exercise intervention group decreased LV mass to end-diastolic volume ratio (adjusted mean difference: -0.03, 95% CI: -0.06,-0.01, interaction term p=0.027) and increased RV stroke volume (adjusted mean difference: 5.53mL/m2, 95% CI: 2.60,8.47, interaction term p=0.076) versus controls. Preterm-born adults in the exercise intervention group increased LV basal- and mid-ventricular circumferential strain versus controls (adjusted mean difference: -1.33%, 95% CI: -2.07,-0.60, interaction term p=0.057 and adjusted mean difference: -1.54%, 95% CI: -2.46,-0.63, interaction term p=0.046, respectively), which was not observed in term-born adults. Similarly, preterm-born adults improved RV longitudinal strain versus controls (adjusted mean difference: -0.51%, 95% CI: -1.49,0.46, interaction term p=0.053), which was also not observed in term-born adults. Conclusions Aerobic exercise training improves myocardial function but not cardiac structure in preterm-born adults. Modified exercise protocols or combination with pharmacological intervention may lead to additional cardiac remodelling benefits.