Vol. 20, No. 2 Printed in Great Britain
International Journal of Epidemiology © International Epidemiological Association 1991
Edinburgh Artery Study: Prevalence of Asymptomatic and Symptomatic Peripheral Arterial Disease in the General Population Fowkes F G R (Wolfson Unit for Prevention of Peripheral Vascular Diseases, Department of Community Medicine, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG, UK), Housley E, Cawood EHH, Macintyre CCA, Ruckley CV and Prescott RJ. Edinburgh Artery Study: Prevalence of asymptomatic and symptomatic arterial artery disease in the general population. Internationa/Journal of Epidemiology 1991; 20: 384-392. Intermittent claudication has been studied in cardiovascular surveys but limited information is available on asymptomatic peripheral arterial disease. The purpose of this paper is to describe the prevalence of both asymptomatic and symptomatic disease and relation to ischaemic heart disease in the Edinburgh Artery Study. A cross-sectional survey was conducted on an age-stratified sample of men and women aged 55 to 74 years selected from age-sex registers in ten general practices in the city. Arterial disease was assessed in 1592 participants by means of the WHO questionnaire on intermittent claudication and measurement of the ankle brachial systolic pressure index (ABPI) and change in ankle systolic pressure during reactive hyperaemia. The prevalence of intermittent claudication was 4.5% (95% confidence interval (Cl): 3.5%-5.5%). Major asymptomatic disease causing a significant impairment of blood flow occurred in 8.0% (95% Cl: 6.6%-9.4%). Afurther 16.6% (95% Cl: 14.6%-18.5%) had criteria considered abnormal in clinical practice: 9.0% had ABPI 20%. Intermittent claudication was equally common in both sexes. The ABPI and reactive hyperaemia results suggested a slight preponderance of asymptomatic disease in males and were consistent with an increasing prevalence with age and lower social class. Mean ABPI was higher in normal men than women, and was lower in the left leg than the right suggesting a unilateral predisposition to disease. Subjects with major asymptomatic disease had more evidence of ischaemic heart disease than In the normal population (relative risk (RRJ 1.6; 95% Cl: 1.3-1.9). In conclusion, major asymptomatic atherosclerosis affecting the lower limbs occurs commonly in the general population and is associated with ischaemic heart disease such that further study is required of natural history and prospects for prevention of major vascular events.
questions we report on the first survey in the UK to describe the distribution in the general population of measures of both symptomatic and asymptomatic peripheral arterial disease and relationships with ischaemic heart disease. Our intention is to follow up this population during subsequent years and to study the aetiology and natural history of peripheral arterial disease. In assessing asymptomatic disease in the general population, angiography is not feasible but indirect measures of the haemodynamic effects of arterial narrowing can be used.2 In clinical practice an ankle brachial systolic pressure ratio or index (ABPI) of 2O-35% >35% No result Total
4
5 11 7 27
9 2 4 1 16
22 2 0 5 29
34 15 10 4 63
130 20 10 23 183
X).9 1061 103 18 82 1264
ABPI result missing on one subject with claudication and nine subjects without claudication
Age and Sex The prevalence of intermittent claudication increased with age (Figure 1). Men aged 65-69 years were an exception but this may be explained by small numbers in each age category. After adjusting for age, the odds ratio of intermittent claudication in women compared to men was 1.11 (95% CI: 0.70-1.79) suggesting that prevalence was similar in men and women. Mean ABPIs decreased with age in both sexes (Figure 1) and, after adjusting for age, were found to be higher in men than women, the difference being 0.053 (95% CI: 0.036-0.071). However, the ABPI correlated with height (males: r = 0.14, females: r = 0.11) and after adjusting for age and height, the sex difference was reduced to 0.020 (95% CI: 0.004-0.044). Despite a higher overall mean, men had relatively more ABPIs 0.9 and reactive hyperaemic drop of 0.9 and reactive hyperaemia result 50% narrowing of an artery. A further 16.6% of the population had an ABPI 20% reduction in systolic pressure during reactive hyperaemia. Although these cut-off points have been shown to be valid in identifying angiogram positive disease in patients in the clinical setting,23'6 validity in the general population is not clearly known. Thus, the classification of the extent of disease in an individual according to one of these measures is not possible and estimated predictive values in the general population will require further investigation using noninvasive 'gold standards' such as duplex scanning and nuclear magnetic resonance. However, the results within groups probably reflect the extent of disease particularly since the trends with age and other parameters are as expected and we found that mean ABPIs were related to the severity of disease detected by duplex scanning (Unpublished observations). Also, the distribution of the ABPI had a negative skew and the reactive hyperaemia test a positive skew, both in keeping with the impact of disease on these parameters. One-third of claudicants had an ABPI >0.9 and reactive hyperaemia pressure reduction
International Journal of Epidemiology © International Epidemiological Association 1991
Edinburgh Artery Study: Prevalence of Asymptomatic and Symptomatic Peripheral Arterial Disease in the General Population Fowkes F G R (Wolfson Unit for Prevention of Peripheral Vascular Diseases, Department of Community Medicine, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG, UK), Housley E, Cawood EHH, Macintyre CCA, Ruckley CV and Prescott RJ. Edinburgh Artery Study: Prevalence of asymptomatic and symptomatic arterial artery disease in the general population. Internationa/Journal of Epidemiology 1991; 20: 384-392. Intermittent claudication has been studied in cardiovascular surveys but limited information is available on asymptomatic peripheral arterial disease. The purpose of this paper is to describe the prevalence of both asymptomatic and symptomatic disease and relation to ischaemic heart disease in the Edinburgh Artery Study. A cross-sectional survey was conducted on an age-stratified sample of men and women aged 55 to 74 years selected from age-sex registers in ten general practices in the city. Arterial disease was assessed in 1592 participants by means of the WHO questionnaire on intermittent claudication and measurement of the ankle brachial systolic pressure index (ABPI) and change in ankle systolic pressure during reactive hyperaemia. The prevalence of intermittent claudication was 4.5% (95% confidence interval (Cl): 3.5%-5.5%). Major asymptomatic disease causing a significant impairment of blood flow occurred in 8.0% (95% Cl: 6.6%-9.4%). Afurther 16.6% (95% Cl: 14.6%-18.5%) had criteria considered abnormal in clinical practice: 9.0% had ABPI 20%. Intermittent claudication was equally common in both sexes. The ABPI and reactive hyperaemia results suggested a slight preponderance of asymptomatic disease in males and were consistent with an increasing prevalence with age and lower social class. Mean ABPI was higher in normal men than women, and was lower in the left leg than the right suggesting a unilateral predisposition to disease. Subjects with major asymptomatic disease had more evidence of ischaemic heart disease than In the normal population (relative risk (RRJ 1.6; 95% Cl: 1.3-1.9). In conclusion, major asymptomatic atherosclerosis affecting the lower limbs occurs commonly in the general population and is associated with ischaemic heart disease such that further study is required of natural history and prospects for prevention of major vascular events.
questions we report on the first survey in the UK to describe the distribution in the general population of measures of both symptomatic and asymptomatic peripheral arterial disease and relationships with ischaemic heart disease. Our intention is to follow up this population during subsequent years and to study the aetiology and natural history of peripheral arterial disease. In assessing asymptomatic disease in the general population, angiography is not feasible but indirect measures of the haemodynamic effects of arterial narrowing can be used.2 In clinical practice an ankle brachial systolic pressure ratio or index (ABPI) of 2O-35% >35% No result Total
4
5 11 7 27
9 2 4 1 16
22 2 0 5 29
34 15 10 4 63
130 20 10 23 183
X).9 1061 103 18 82 1264
ABPI result missing on one subject with claudication and nine subjects without claudication
Age and Sex The prevalence of intermittent claudication increased with age (Figure 1). Men aged 65-69 years were an exception but this may be explained by small numbers in each age category. After adjusting for age, the odds ratio of intermittent claudication in women compared to men was 1.11 (95% CI: 0.70-1.79) suggesting that prevalence was similar in men and women. Mean ABPIs decreased with age in both sexes (Figure 1) and, after adjusting for age, were found to be higher in men than women, the difference being 0.053 (95% CI: 0.036-0.071). However, the ABPI correlated with height (males: r = 0.14, females: r = 0.11) and after adjusting for age and height, the sex difference was reduced to 0.020 (95% CI: 0.004-0.044). Despite a higher overall mean, men had relatively more ABPIs 0.9 and reactive hyperaemic drop of 0.9 and reactive hyperaemia result 50% narrowing of an artery. A further 16.6% of the population had an ABPI 20% reduction in systolic pressure during reactive hyperaemia. Although these cut-off points have been shown to be valid in identifying angiogram positive disease in patients in the clinical setting,23'6 validity in the general population is not clearly known. Thus, the classification of the extent of disease in an individual according to one of these measures is not possible and estimated predictive values in the general population will require further investigation using noninvasive 'gold standards' such as duplex scanning and nuclear magnetic resonance. However, the results within groups probably reflect the extent of disease particularly since the trends with age and other parameters are as expected and we found that mean ABPIs were related to the severity of disease detected by duplex scanning (Unpublished observations). Also, the distribution of the ABPI had a negative skew and the reactive hyperaemia test a positive skew, both in keeping with the impact of disease on these parameters. One-third of claudicants had an ABPI >0.9 and reactive hyperaemia pressure reduction
