Danish Cardiovascular Screening Trial II

Background

Although CVDs have decreased, they are still among the most predominant cause of morbidity and mortality in the western world, incl. Denmark, where about 420,000 people have recognized CVD symptoms. Due to an aging population, the decline has not reduced CVD admissions and healthcare costs. In Denmark, the CVD related hospital admission costs are DKK 4.6 billion and the pharmaceutical cost DKK 2.4 billion. The Danish National Board of Health has reported that CVD carries the second largest socioeconomic difference in burden of disease. Unfortunately, population-based health checks and screening for risk factors has proven not efficient. Consequently, screening of asymptomatic CVD is discussed intensively. In the investigators first unique CVD screening RCT (2008-11), the Viborg Vascular (VIVA) trial, more than 50.000 65-74 year old men were randomised for population-based ultrasound screening for abdominal aortic aneurysm (AAA), peripheral arterial disease (PAD) and hypertension. In case of positive finding, preventive medical actions were initiated. A significant reduction in overall mortality by 7% after 5 years was observed. Using a non-contrast computered tomography (CT) scan, instead of the ultrasound based screening approach, has the opportunity to identify aneurysms in the entire aortic, coronary artery calcification (CAC) and arterial fibrillation, so individualized risk assessment and initiation of preventive actions on those with sign of early asymptomatic CVD is possible. A second trial - The Danish Cardiovascular Screening (DANCAVAS) trial initiated was initiated in 2014 randomizing 45.000 65-74 year old men with the potential of a huge beneficial effect on health, quality of life (QoL) and survival. However, screening is impaired by lower social class, and adherence to initiated prevention could be impacted as well. Consequently, we want to conduct a third RCT (Fighting Social Inequality in Cardiovascular Health (FISICH) to test a number of add-ons to screening that potentially balance the benefits across socio economic groups. The perspective is to establish a clear decision foundation for public health care policy incl. benefits, cost effectiveness and impact on social inequality of alternative variants of population screening for CVD.

Hypotheses

The primary hypothesis is that an extensive circulatory screening and intervention programme reduces social inequality in cardiovascular health and fulfills the WHO criteria for screening. However, this reduction can be even more pronounced, if factors reducing the social selection to attend screening and adherence to preventive actions initiated are identified.

Aims

The aims are to

1. Estimate benefits of using DANCAVAS screening methodology in men aged 60-64. This evaluation will be overall, and subgroup-analyzed concerning educational level, income level, and psychiatric disease.

Materials and methods 20.000 60-64 year old men are randomized to the control group, while another 5.000 are randomized to the screening and intervention program for CAC, aortic and iliac aneurysms, atrial fibrillation, PAD, hypertension, diabetes and hypercholesterolaemia. There is no exclusion criteria. The screening setup is similar to DANCAVAS:

1. A small questionnaire on life style, medical history, and the QoL a.o. will be enclosed with the invitation. Non-responders are re-invited once.
2. The participant will be informed at attendance to the screening visit, and their consent will be obtained together with the questionnaire, weight, height, and waist circumference.
3. The CT scan will cover the area from the mandibular bone distally to the proximal third of the femur. Calcium scores for the common carotids, coronary arteries, aorta, and common iliac, and femoral arteries will be calculated. The aorta are visualized, and the diameter is measured in ascending, arcus, descending and abdominal, and if possible in the iliac arteries. Further the heart rhythm during the CT scan is evaluated.
4. Bilateral blood pressure will be recorded three times after 5 minutes of supine rest, and concurrently the ankle blood pressure are measured.
5. The HbA1c and lipid parameters will be measured. Biobank blood samples are then taken, centrifuged, labelled, cooled, and stored at -80 degrees Celsius.

Follow-up visit after screening

If the CAC is above the median or if an aneurism of peripheral arterial disease are detected the participant is informed of the finding and its implications at a follow-up visit. At this visit, the patient will be recommended suitable prophylactic measures, including smoking cessation, walking/exercise, a low-fat diet. Additionally to start treatment with aspirin 75 mg/day and atorvastatin 40 mg/day. If an aneurism is large the patient is referred vascular surgical assessment for the repair. Otherwise, an annual check-up of the aneurism including a CT scan will be offered.

If no positive findings (CAC above the median, aneurysm or PAD) are detected, the participants will be informed of the findings by e-mail or ordinary post as preferred.

Independent of the above findings, the patients will be encouraged to see their GP for further assessment if potential undiagnosed hypertension (systolic blood pressure >160 mmHg), diabetes mellitus (HbA1c >48 mmol/mol), or significant isolated hypercholesterolemia (total-cholesterol >8.0 mmol/l) are observed, as possible continuous medical treatments will be better managed by the GPs.

The GPs will be informed by a letter of all negative and positive results and the initiated actions.

Enrolment started mid 2017 with sites in Silkeborg, Svendborg, Odense, Vejle and Nykøbing Sjælland.

Power calculations and Randomisations

A study with 4 controls per invited subject is planned based upon the mortality reduction in the VIVA study, a hazard ratio of 0.93 in the screening group is expected. If so, with a 0.05 significance level, and 90% power, we´ll need 5245 experimental subjects and 20980 controls. The randomisation procedure will be stratified by geographical area (municipalities), as sociodemographics, attendance, and disease prevalence are expected to differ across areas. Randomisation will be performed in SPSS by providing each individual a random number from 1-20. Those numbered +16 will be invited to participate in the screening program.

Baseline variables

Age, smoking, previous or current stroke, ischemic heart disease, PAD, chronic obstructive pulmonary disease, diabetes, hypertension, use of statins, use of antithrombotics, body mass index, systolic- and diastolic blood pressure, ankle brachial index, marital status, highest educational level, personal- and in house income, psychiatric morbidity defined as any diagnosis and/or use of medications for mental illness, and quality of life (QoL).

Baseline and outcome variables from national registries

The CPR number assigned to Danish citizens enables individual-level linkage to multiple nation-wide healthcare and administrative registries which have proved valid. Registry-based information on outpatient visits, hospitals admissions and procedures (The Danish National Patient Registry), relevant prescribed drugs dispensed (The Danish National Prescription Registry), socio economic status etc. (Registries at Statistics Denmark) and primary care service use (National Health Insurance Service Registry) will be obtained.

Outcomes

The primary outcome is

1. All-cause mortality (time to event or censoring).

Secondary outcomes are:

1. Positive findings by screening, in all and by specific condition,
2. Mortality due to CVD (time to death),
3. Initiation of first redeement of prophylactic pharmaceutical therapy in participants not in therapy at baseline - split on antithrombotic,- lipid lowering and antihypertensive agents,
4. Initiation of prophylactic surgery and avoidance of acute surgery (time to first procedure concerning AAA, PAD, and coronary revascularization,
5. Morbidity due to CVD (time to first hospital-based diagnosis and total number of inpatient days) as a composite measure,
6. Adverse events (time to first hospital-based diagnosis) split on diabetes, intracerebral haemorrhage, and cancer,
7. Safety (30d postoperative mortality) split on AAA, PAD and coronary revascularization,
8. QoL,
9. Health care costs and cost-effectiveness.

Statistical analysis

Baseline characteristics will be analysed using conventional summary statistics. Outcomes representing a time to event are analysed and compared between randomisation groups including subgroups defined by social status and mentally ill using Cox proportional hazards model. All outcomes representing density of events are analysed and compared between randomisation groups including corresponding subgroups using plain incidence rate ratios.

Health economic models

Multivariate regression analysis will be undertaken to analyse health care costs (one year before and after screening) across socioeconomic groups due to pharmacological over-/undertreatment. A cost-effectiveness analysis with 5 years of follow-up adopting a health care perspective will be performed to investigate cost-effectiveness of screening across socioeconomic groups (population heterogeneity). All cause mortality is the primary outcome and costs include screening costs and primary- and secondary care costs. Finally, a health economic decision model based on available evidence (literature review and meta analysis) and original data from the three trials will be developed to inform decision makers about the (distributional) consequences of implementing different variants of population screening in Denmark. The specification of the model will be based on a Markov state model allowing for the analysis of the total number of expected life years and the total health care costs for a lifetime perspective in different scenarios with different variants of screening. The modelling of social inequality (tagging who receives the benefit of screening) is novel in this literature and a unique opportunity given the Danish trials and register data.

Biobank

As in the DANCAVAS study, a biobank will be organised. 40 mL of blood form each of the participants are centrifuged, labelled, and stored at -80 degrees Celsius. The purpose of the biobank is to perform pathophysiological and translational studies, and to facilitate new knowledge of development and treatment of atherosclerosis. Analyses of (1) atherosclerosis and (2) inflammation in the vessel wall, (3) diseases in the myocardium, (4) calcium / phosphate metabolism and (4) proteomics are intended.

Remaining blood samples are planned to be stored for a ten-year period for future research, and subsequently data will be anonymous. This is to be approved by the Danish Data Protection Agency. Prior to future analyses approval by the Region Committee on Biomedical Research Ethics