Vascular screening programme: Have we gone far enough?

An outline proposal delivered by the Department of Health (‘Putting prevention first – vascular checks: risk assessment and management’ DoH, 2008a) calls for a system of vascular checks to be carried out in primary care. This document advocates a ‘predict and prevent’ approach to identify vulnerability to vascular diseases such as heart disease, stroke, diabetes and kidney disease. 

A second programme of vascular screening to identify men aged 65 with abdominal aortic aneurysms at high risk of rupture is predicted to save more than 1600 lives each year. At present £3 million has been set aside for pilot projects initially offering screening to 32 000 men in England with an overall target of 270 000 men a year and will be rolled out over the next 10 years.

Vascular checks: Risk assessment and management
The aim is to prevent up to 9500 heart attacks and strokes every year and save 2000 lives by screening all people aged 40 to 74 years in 2009–10 for vascular risk (DoH, 2008a) (Figure 1). In addition, at least 4000 people a year may be prevented from developing diabetes and a further 25000 cases of diabetes or kidney disease may be detected earlier (DoH, 2008a). The programme, set to cost around £250m per year, aims to shift the emphasis to primary prevention of vascular disease (coronary heart disease [CHD], stroke, diabetes and kidney disease) and reduce high-cost secondary care treatment-dependent complications. This primary prevention programme will involve GPs, practice nurses, pharmacists and other community practitioners. The maximum impact is aimed within disadvantaged communities, in particular, ethnic groups that are disproportionately affected by vascular disease. 

The handbook for vascular risk assessment, risk reduction and risk management, outlines the current state of knowledge regarding vascular assessment (National Screening Committee, 2008). CVD is defined as: CHD, stroke, transient ischaemic attacks, type 2 diabetes, peripheral arterial disease and chronic kidney disease. This suggests that information regarding risk reduction should be widely available to all, but there is a high-risk subset of the population who require more intervention to change lifestyle factors and consideration of pharmacological therapy (see Figure 1). 

This document advocates a four-stage approach:

1. Risk surveillance 
2. Risk communication 
3. Risk assessment 
4. Risk management.

Risk surveillance
Opportunities for surveillance have been widened to include self-assessment through a public health campaign to screening by independent and extended practitioners. High-risk patients will also be identified from practice registers with an emphasis on modifiable risk factors such as: smoking; physical inactivity and sedentary lifestyle; hypertension; hyperlipidaemia and obesity. Population-based screening is likely to commence from the age of 40 years and be repeated every 5 years, to give the largest overall health benefit with a favourable incremental cost per QALY (quality adjusted life year) compared with more intense screening or at a higher age group. 

The handbook (National Screening Committee, 2008) outlines a framework for screening for diabetes. The controversies regarding implementation based on best practice were debated by Goyder et al (2008). They concluded that mass screening for diabetes was unlikely to be cost effective but this needed to be considered as part of a risk assessment. 

Screening tools for vascular risk
One of the ambitions of the primary prevention programme is to reduce health inequalities and therefore an assessment tool sympathetic to this variable is required. A modified version of the FINDRISC (Finnish diabetes risk score) self-assessment tool has been suggested as a screening tool but has not been validated in a UK or BME setting (Lindstrom and Tuomilehto, 2003; Rathmann et al, 2005). The ASSIGN score (Assessing cardiovascular risk using SIGN guidelines to ASSIGN preventative treatment [http://www.assign-score.com]) has been developed to include social deprivation (using family history or postcode) as a risk factor (Scottish Intercollegiate Guidelines Network, 2007; Woodward et al, 2006). A further screening tool, QRISK, appears to be better calibrated to a UK population (Hippisley-Cox et al, 2007a, Hippisley-Cox et al, 2007b). Pilots are underway to determine which tool will identify the populations at highest risk needing intervention and ensuring integration of this information into current information systems and holistic care. 

Risk communication
Once the risk has been assessed, patients may need additional blood tests (cholesterol, in some cases glucose and tests for kidney disease, urine and blood). Screening may occur at different opportunities and therefore communication is important to ensure an overall risk reduction plan is developed for that individual patient. The success of this risk reduction will be largely dependent upon involving the patient in the long-term plan to reduce cardiovascular risk and improving the patient’s understanding of the disease processes. 

Patients at low-risk may need general advice on how to maintain good health, while those at moderate risk may need weight management including exercise and smoking cessation. Those at high-risk might require pharmaceutical management and possibly referral to secondary care in addition to these interventions. People need to understand that they are on a continuum rather than the traditional ‘at-risk or not at-risk’ and the aim is risk reduction from their own individual baseline (see Table 1). 

Capacity planning
A typical GP practice with 5600 patients on its list would offer vascular checks to all patients aged 40–74 years every five years: an estimated 330 additional vascular checks per year. Of those who attend, an estimated 65% would be eligible to one or more of the lifestyle interventions, and an estimated 20% would require statins or antihypertensives in the first round of tests (National Screening Committee, 2008). Practice and specialist nurses and other community practitioners will play a significant role in carrying out screening and it must be a key government priority to ensure there are enough nurses with the right skills in place (Tweddell, 2008). 

Abdominal aneurysm screening
Nearly 5000 patients a year die as a result of a ruptured abdominal aortic aneurysm in England and Wales (Office for National Statistics, 2006). Seventy-five per cent of patients die before reaching hospital, and of the survivors, only half make it to the operating theatre (Brown and Powell, 1999). The operative mortality for ruptured aneurysm remains at about 40% at 30 days, compared with a mortality of 5–6% for open elective surgery and 1.6% for endovascular repair (EVAR Trial Participants, 2007). Risk factors include smoking, hypertension, positive family history and elevated LDL cholesterol. There are no data to suggest that aggressive risk modification reverses aneurysm growth but screened patients have better opportunities for secondary prevention. Ultrasonography is a simple and reliable test for aneurysm screening. 

Analysis of cost effectiveness suggested that an initial aneurysm screen cost £20.39 compared to £6909 for an open elective repair and £11176 as an emergency. After 4 years the screened group cost £98.42 per patient compared to £35.03 per patient in the control group. The initial cost per life year gained by the screening programme after four years was £28389. After 10 years the cost per life year gained fell to £8000 or around £10000 per QALY gained (Ashton et al, 2002; Ashton et al, 2007). A Cochrane review showed a highly significant reduction in aneurysm related mortality (odds ratio 0.60; 95% CI 0.47 to 0.78) although the 5% all cause mortality reduction was non significant (Cosford and Leng, 2007). 

Capacity planning
The proposal document suggests that a single screening unit will cover a population of around 800000, and scan up to 7000 men per year. The target population will be identified from GP registers (National Screening Committee, 2007). 

The screening procedure is divided into four stages. 

1. Basic screen to detect any abnormality.
2. Screening assessment (based on aortic measurements) of the abdominal aortic aneurysm to determine immediate referral to a vascular surgeon or regular surveillance. 
3. Clinical assessment following referral to a suitable vascular unit. 
4. Surgical treatment of the screen-detected aneurysm as appropriate.

These screening clinics will be held within primary care to allow maximum access as uptake of invitations from screening studies range from 63% to 80%. Although the size of the aneurysm will be communicated to the patient by the screening team, there is no doubt that this will also increase the GP’s or nurse specialist’s workload. These newly diagnosed people will need information regarding risk as well as secondary prevention. In addition there are data suggesting an increase in anxiety – patients often liken the risk of an aneurysm to a ticking time-bomb.

Peripheral arterial disease
Although the definition for CVD includes CHD, stroke, transient ischaemic attacks, type 2 diabetes, peripheral arterial disease and chronic kidney disease there are no guidelines for the assessment of risk for people with peripheral arterial disease.

Screening of high-risk groups has been advocated because most patients are asymptomatic. People with peripheral arterial disease remain an unrecognized group with a disease that can be detected by a simple, non-invasive, and accurate test (ankle brachial pressure index [ABPI]) and in whom secondary preventive treatment can prevent vascular events and death. 

The prevalence of symptomatic peripheral arterial disease rises with age: 3–6% in men 60 years of age, 10.6% in men 65–69 years of age and 23.3% in men 75–79 years of age (Van Ganse et al, 1972). The prevalence of asymptomatic disease is higher, data suggest 10%–25% of men and women over 55 years of age have an ABPI less than 0.9 suggesting that for every patient with peripheral arterial disease there are probably another three with asymptomatic disease (Dormandy et al, 1999; see Table 2 for ABPI interpretations). There is an inverse relationship between the numerical value of ABPI and the risk of cardiovascular morbidity and mortality (Steg et al, 2007) and patients with critical ischaemia have a poor survival (Criqui et al, 1992). The relative five year mortality of a patient with severe peripheral arterial disease is 44% which is equivalent to that of a Dukes B carcinoma of the colon (McKenna et al, 1991).

Despite this, there are no national screening programmes as it is suggested that there are no data to show that the overall benefit of treating patients with peripheral arterial disease outweighs the adverse consequences and costs associated with screening and follow-up. 

Peripheral arterial disease, coronary artery disease and stroke are all manifestations of atherosclerosis and therefore, it is not surprising that the three conditions commonly occur together. A sixth of the population in the REACH registry (reduction of atherothrombosis for continued health) had evidence of polyvascular disease. Each individual disease location carried an increased overall cardiovascular risk but presence in all three vascular beds increased the risk exponentially. The incidence of a further cardiovascular event within 1 year in this high-risk group was 26% (Bhatt et al, 2006). Patients enrolled in this observational study with evidence of risk factors but not manifest disease had a 1 in 20 risk of a cardiovascular event within the first year, a diagnosis of peripheral arterial disease doubled this risk. 

It is essential, just based on this data that patients with peripheral arterial disease are optimally treated and that patients with multiple sites of atherosclerosis are not missed.  

Screening tools for peripheral arterial disease
A diagnosis of peripheral arterial disease may be made on clinical history but this is not sensitive for asymptomatic disease. The Edinburgh Claudication Questionnaire (Modified Rose/WHO) suggests that the pain associated with peripheral arterial disease is defined by five characteristics (Table 2). It has been shown to be highly specific (99%) and sensitive (91%) in detecting those with symptomatic peripheral arterial disease compared with physicians’ assessments of symptoms and has been adapted for self-administration (Rose et al, 1977; Leng and Fowkes, 1992).

Ankle brachial pressure index
It has been suggested that a resting ABPI of 0.9 is up to 95% sensitive in detecting angiogram-positive disease, and almost 100% specific in identifying apparently healthy individuals. A resting ABPI of less than 0.9 is the most common definition of peripheral arterial disease (Table 3) and total mortality is very significantly increased in people with an ABPI of 0.8 to 0.9, with very little difference with an ABPI of 0.9 to 1 or over 1.0 (Vogt et al, 1993). This technique can reproduce results and is non-invasive.

Diabetes and peripheral arterial disease: A high risk population
Up to 70% of non-traumatic amputations are performed in people with diabetes, and a person with diabetes who smokes runs an approximately 30% risk of amputation within 5 years (Dormandy et al, 1999).

Diabetes is more prevalent in BME populations and the government is investing in screening to reduce long-term risk in people with diabetes. An opportunity will be lost if peripheral arterial disease is not detected in this vulnerable group. This diagnosis not only identifies people with diabetes as the highest-risk group but also as people at high risk of diabetic foot ulceration and amputation. 

Carotid screening
Stroke is the third biggest cause of death in the UK and each year more than 110000 people in England will suffer from a stroke which costs the NHS over £2.8 billion. The National Service Framework has improved overall stroke services and a National Stroke Strategy has been developed. This aims to dramatically reduce strokes which are also the single largest cause of adult disability. A campaign of reduction in health inequalities and promoting an understanding of stroke and transient ischaemic attacks to both the medical community and general public may allow this to be achieved. 

The aim is secondary prevention and risk reduction. Carotid screening is not discussed within the handbook (National Screening Committee, 2008). Screening of the general population is not cost-effective and not recommended as the prevalence of asymptomatic carotid artery stenosis ranges between 2% and 18% among the general population (Quershi et al, 2007). Asymptomatic carotid stenosis warrants surgical treatment – if patients have a surgical risk of less than 3% and a life expectancy of at least 5 years, then carotid endarterectomy may be considered for stenoses exceeding 70%. Screening may need to be considered for asymptomatic people aged 65 years of age or older with at least three cardiovascular risk factors; especially diabetes, peripheral vascular disease, a history of smoking, or previous stroke. 

Conclusions
‘Putting prevention first – vascular checks: risk assessment and management’ (DOH, 2008a) is a laudable attempt to improve the health of the nation. The aim is to detect and stratify patients into low-, moderate- and high-risk groups. The high-risk groups will need tighter targets for risk factor control and lifestyle changes. 

An opportunity may be lost for identifying further high-risk groups who may also benefit from advisory, pharmaceutical and clinical interventions especially in patients with peripheral arterial disease.