Volume 32, Issue 5 , Pages 494-499, November 2006
The Low Incidence of Surgery for Non-Cardiac Vascular Disease in UK Asians may be Explained by a Low Prevalence of Disease
Article Outline
Aims
Firstly, to compare rates of surgery for non-cardiac vascular disease in Caucasians and Asians and secondarily to assess the prevalence of peripheral arterial disease (PAD) and abdominal aortic aneurysm (AAA) in the male UK Asian population.
Methods
Analysis of a prospective database followed by an epidemiological survey of 100 unselected Pakistani males, in which demographic and anthropometric data were collected alongside aortic ultrasonography and measurement of ankle: brachial pressure index (ABPI).
Results
Although 14.1% of our catchment area is Asian, after correction for age, they only accounted for 64/2268 (2.8%) of procedures for PAD and AAA. Specifically, Asians were 10 times less likely to undergo AAA repair and 3 times less likely to undergo procedures for lower limb peripheral bypass, amputation and endovascular intervention. In the epidemiological study, 26 subjects had a significant history of ischaemic heart disease, 21 were diabetic, 32 had hypertension and 60 were current or ex-smokers. Median aortic diameter [IQR] was 17.6
mm [16.3–19.1mm] and no subject had an AAA. In 200 limbs, median ABPI [IQR] was 1.12 [1.04–1.21]. Only 2 patients had an ABPI <0.9.
Conclusion
Despite a high prevalence of cardiovascular risk factors and ischaemic heart disease, the prevalence of PAD and AAA is much lower than would have been expected in an age- and sex-matched Caucasian population. These data suggest that the reduced incidence of surgery for PAD and AAA in UK Asians is due to a low prevalence of disease.
Keywords: Ethnicity, Asians, Caucasians, Peripheral arterial disease, Abdominal aortic aneurysm, Surgery
Introduction
Non-cardiac, often referred to as peripheral, vascular disease is a common cause of morbidity and mortality among Caucasians living in developed countries. Specifically, lower limb peripheral arterial disease, intermittent claudication and abdominal aortic aneurysm (AAA) are present in approximately 20%, 5% and 5% respectively of middle-aged men,1, 2 ruptured AAA accounts for at least 2% of all male deaths over 65 years2 and carotid artery disease is responsible for up to half of all ischaemic strokes.3 By contrast, almost nothing is known of the epidemiology of non-cardiac vascular disease among non-Caucasians living in these countries.4 This is an important gap in our knowledge, as an understanding of the epidemiology and natural history of disease within different populations is a pre-requisite for the adequate provision of health care. With regard to the UK, Asians make up 4.4% of the population of England and Wales and approximately 20% of the catchment area of our own institution (2001 UK national census data). The primary aim of this study was to compare the rates of surgery for non-cardiac vascular disease in Caucasians and Asians within our own population. Secondly, in order to attempt to explain the differences found, we conducted a pilot study to investigate the logistical difficulties and feasibility of establishing the prevalence of lower limb peripheral arterial disease and AAA in our local Asian population.
Methods
Incidence of surgery for non-cardiac vascular disease
Details of all non-cardiac vascular operations and endovascular procedures performed in this institution between January 1997 and April 2002 were obtained from a prospectively collected database and categorised as follows: AAA repair; peripheral vascular surgical procedures (overwhelmingly lower limb bypass, the great majority for critical limb ischaemia), carotid endarterectomy; major lower limb amputations (above, through and below knee), minor lower limb amputations (digit and forefoot) and endovascular interventions (digital subtraction arteriography and percutaneous transluminal balloon angioplasty). Patients were categorised as Asian or non-Asian by name analysis, as previously described.5 Population data for the catchment area of this Trust were obtained from the Office of National Statistics (2001 UK National Census data). The national census ethnic groupings of Pakistani, Indian, Bangladeshi and other Asian were considered together as ‘Asian’, and the remaining population (overwhelmingly Caucasian) were considered ‘non-Asian’. The expected number of procedures for Asians was calculated by indirect standardisation using non-Asian age-specific operative rates and the Asian age structure. Standardised operation ratios, with Poisson confidence intervals, for Asians were calculated from the expected and observed values.
Prevalence of non-cardiac vascular disease in the local Asian population
After local ethical committee approval, all men over the age of 55 years attending a local mosque were invited to take part in the study. Written informed consent was taken with the aid of an interpreter. Consenting subjects underwent a structured history and clinical and ultrasound examination at the mosque.
Clinical history: A positive history of cardiac vascular disease was made if the subject had a history of treated angina, myocardial infarction, percutaneous transluminal coronary angioplasty, or coronary artery bypass grafting. Cerebrovascular disease was deemed present if the subject had a history of previous cerebrovascular accident, transient ischaemic attack or carotid endarterectomy. Diabetes was defined as the current use of insulin, oral hypoglycaemics, a history of diabetes controlled by diet alone or the finding of a random blood glucose level >11.1mmol/l. Hypertension was defined as the current use of anti-hypertensive medication or the finding of a resting blood pressure >160/95mmHg. Details of current medications and previous vascular investigations and intervention were also recorded.
Anthropometry: Height was measured without shoes to the nearest 0.5cm and weight was measured without shoes and with the subject lightly clad to the nearest 0.1kg. Body mass index was calculated by dividing the weight (kg) by height2 (m2). Waist circumference was measured to the nearest 1cm with the patient standing, at the mid point between the lower costal margin and the iliac crest after gentle expiration. Waist: height ratio was used as an indirect measure of intra-abdominal obesity in South Asians as previously described.6 Blood pressure was measured in both arms after resting for 10 minutes in a seated position (Omron M5-I, Omron Healthcare UK, Henfield, UK).
Serum analysis: A random, non-fasting venous blood was collected into clot activator serum tubes for later analysis in an accredited laboratory for serum glucose, total cholesterol, HDL cholesterol and triglycerides.
Ankle: brachial pressure index (ABPI): ABPI was calculated by measuring the highest ankle pressure (anterior tibial, posterior tibial, perforating peroneal) in each leg (Mini Dopplex®, Huntleigh Diagnostics, Cardiff, UK) and dividing it by the highest brachial systolic pressure. An ABPI of <0.9 at rest was taken to indicate the presence of peripheral arterial disease.
Aortic diameter: The maximum diameter of the infra-renal aorta was measured in its longitudinal plane by means of ultrasound (SonoSite 180 Plus, SonoSite Inc, Bothwell, Washington, USA). As previously described, AAA was defined by a diameter of ≥30mm.7
Results
Incidence of surgery for non-cardiac vascular disease
This institution serves a population of approximately half a million people of whom 21.4% (110,216) are recorded by the 2001 Census as being Asian. The Asian population is significantly younger than the non-Asian population with Asians accounting for 8.5% of those over 50 years of age.
However, over the 5-year study period, only 64 (2.8%) of 2268 non-cardiac vascular procedures were conducted on Asians (Table 1). Even when adjusted for age, Asians were 10 times less likely to undergo AAA repair and 3 times less likely to undergo carotid endarterectomy and procedures for lower limb peripheral arterial disease, including major lower limb amputation and endovascular interventions. No Asian patient attended this hospital with a ruptured AAA. Asians undergoing carotid endarterectomy were significantly younger than non-Asians (mean age [s.d.], 57 [6.2] vs. 67.9 [9.7], p=0.02, Mann–Whitney), but there were no demographic differences with regard to the other procedure categories.
Table 1. Ethnic breakdown of vascular procedures performed between January 1997 and April 2002 and standardised operation ratios for Asians. Expected numbers of procedures for Asians have been calculated by indirect standardisation using non-Asian age-specific operative rates and the Asian age structure. Standardised operation ratios, with Poisson confidence intervals, for Asians are shown, calculated from the expected and observed values
| Operation | Median Age (IQR) | Procedures performed | Expected number of procedures in Asians | Standardised Operation Ratio (95% CI) | P value | ||
|---|---|---|---|---|---|---|---|
| Total | Non-Asian | Asian (%) | |||||
| AAA | 73 (67–78) | 248 | 246 | 2 (0.8) | 20.1 | 10.0 (1.2–36) | <0.001 |
| Peripheral Bypass | 70 (62–78) | 406 | 395 | 11 (2.7) | 37.3 | 29.5 (14.7–52.7) | <0.001 |
| Carotid Endarterectomy | 68 (62–75) | 140 | 136 | 4 (2.9) | 13.1 | 30.5 (8.3–78.2) | 0.003 |
| Major Amputation | 74 (66–79) | 228 | 222 | 6 (2.6) | 19.8 | 30.4 (11.1–66.1) | <0.001 |
| Minor Amputation | 67 (58–77) | 168 | 159 | 9 (5.4) | 15.7 | 57.5 (26.3–109.1) | 0.05 |
| Endovascular Intervention | 69 (61–76) | 1078 | 1046 | 32 (3.0) | 100.4 | 31.9 (21.8–45.0) | <0.001 |
Prevalence of non-cardiac vascular disease in the local Asian population
The first 100 men over the age of 55 years accepting the invitation to take part were studied. It is estimated that there are 5000 Muslim men in the mosque catchment area and that approximately 1000 (20%) of these attend the mosque each week with >70% of these men being over the age of 55 years. The study cohort therefore represents approximately 15% of the men, over the age of 55 years, attending this particular mosque.
The median [IQR] age of the subjects was 67 [62.3–72.8] years. All had been born on the Indian Subcontinent originating from the Kashmiri region of Pakistan (76), other Pakistani regions (24) or Bangladesh (2). The median [IQR] length of UK residence was 41 [38–43] years. There was a high prevalence of diabetes (22%), hypertension (42%) and cardiac vascular disease (26%) (Table 2). Of the latter, only 81% were on an anti-platelet agent and 54% were on a statin. No subject had a history of AAA repair or intervention for lower limb peripheral vascular disease. However, a single subject had undergone carotid endarterectomy 5 years previously. The median body mass index (BMI) was 26.3; 44% were overweight (BMI 25–30), 18% were obese (BMI>30) and 97 subjects had a waist: height ratio >0.50. The total cholesterol: HDL ratio was ≥5 in 50% of subjects (Table 3).
Table 2. Demographic details, co-morbidity, medication and measured clinical parameters of subjects undergoing screening (n=100). Figures are expressed as median [interquartile range] or numbers of subjects where appropriate
| Variable | n |
|---|---|
| Age (years) | 67 [62.3–72.8] |
| Time in UK (years) | 41 [38–43] |
| Cardiac vascular disease | 26 |
| Angina | 17 |
| Myocardial infarction | 11 |
| Percutaneous transluminal coronary angioplasty | 11 |
| Coronary artery bypass graft | 9 |
| Cerebral vascular disease | 5 |
| Transient ischaemic attack | 3 |
| Cerebrovascular accident | 3 |
| Carotid endarterectomy | 1 |
| Diabetes Mellitus | 22 |
| Insulin | 2 |
| Orohypoglycaemics | 11 |
| Diet | 9 |
| Reported hypertension or BP>160/95mmHg at screening | 42 |
| Systolic blood pressure | 142 [129–155] |
| Diastolic blood pressure | 85 [77–92] |
| Smoking | |
| Current smoker | 20 |
| Ex smoker | 40 |
| Never smoker | 40 |
| Pack years of smokers/ex smokers | 10 [4.5–20] |
| Medication | |
| Antiplatelet agent | 27 |
| Statin | 17 |
| Antihypertensive | 25 |
| ACE inhibitor | 7 |
| Warfarin | 1 |
| BMI (kg/m2) | 26.3 [23.7–29.2] |
| BMI≥25 | 62 |
| BMI≥30 | 18 |
| Waist: height ratio | 0.6 [0.57–0.64] |
| <0.50 | 3 |
| 0.50–0.59 | 49 |
| 0.60–0.69 | 42 |
| >0.70 | 6 |
| Aortic diameter (mm) | |
| 12–14.9 | 10 |
| 15–17.9 | 45 |
| 18–20.9 | 33 |
| 21–23.9 | 9 |
| 24–26.9 | 3 |
| Ankle: Brachial Pressure Index (n=200) | |
| 0.50–0.69 | 1 |
| 0.70–0.89 | 1 |
| 0.90–1.09 | 85 |
| 1.10–1.29 | 91 |
| 1.30–1.49 | 22 |
Table 3. Serum lipid and glucose levels of subjects undergoing screening (n=100). Figures are expressed as median [interquartile range] or numbers of subjects where appropriate
| Variable | n |
|---|---|
| Total Cholesterol (mmol/l) | 5.1 [4.3–5.8] |
| ≥6.5mmol/l | 14 |
| HDL Cholesterol (mmol/l) | 1.0 [0.9–1.2] |
| ≤0.9mmol/l | 28 |
| Total cholesterol: HDL cholesterol | 5.0 [3.9–6.1] |
| ≥5 | 50 |
| ≥6.5 | 15 |
| LDL Cholesterol (mmol/l) | 3.4 [2.7–4.0] |
| Triglycerides (mmol/l) | 2.2 [1.6–3.3] |
| Glucose (mmol/l) | 5.9 [5.2–7.6] |
It was possible to visualise the infra-renal aorta in all 100 subjects. No AAAs were detected and median [IQR] aortic diameter was 18 [16–19] mm (range 13–26mm). The median [IQR] ABPI was 1.12 [1.04–1.21] (range 0.58–1.43) and only 2 (of 200 limbs) had an ABPI <0.9. One of these subjects was asymptomatic and the other had disabling claudication symptoms that had previously been labelled as sciatica. After further investigation he underwent a successful iliac angioplasty with complete resolution of symptoms. The ABPI was ≥1.3, possibly indicating non-incompressible vessels, in 14 subjects (22 limbs). However, the prevalence of diabetes in this sub-group (3 of 14, 21%) was the same as in the group as whole.
Although the sample size is relatively small, the differences between the ‘observed’ prevalence of 2% for PAD and 0% for AAA and the ‘expected’ prevalence (from Caucasian data) of 17% for PAD and 5% for AAA are so large that the findings are meaningful both statistically and epidemiologically. This can be quantified using the hypergeometric distribution, which in the case of PAD estimates that the probability of only finding a 2% prevalence in Asians given a 17% prevalence in the Caucasian population to be <0.0000001. In the case of AAA it estimates that the probability of finding a 0% prevalence in Asians given a 5% prevalence in the Caucasian population to be 0.005.
Discussion
Incidence of surgery for non-cardiac vascular disease
The present study has shown that, even after adjusting for age, the incidence of surgery and endovascular intervention for non-cardiac vascular disease in our institution is very much lower among the Asian than amongst the non-Asian (overwhelmingly Caucasian) population. These age-standardised differences are highly statistically significant for all six of the procedure categories. Specifically, Asians were 10 times less likely to undergo AAA repair and 3 times less likely to undergo procedures for lower limb peripheral arterial disease (including endovascular intervention), major lower limb amputation, and carotid endarterectomy. By contrast, Asians were only half as likely to undergo minor amputations. Amongst Asians undergoing amputation there was a higher ratio of minor to major amputations when compared to non-Asians (Asians 1.5:1, non-Asians 0.7:1). Although present data do not allow a definite conclusion to be drawn, this possibly reflects the high prevalence of diabetes in the Asian population and the resulting requirement for minor foot amputations, usually as a result of sepsis secondary to neuropathy. The small numbers of Asians undergoing non-cardiac vascular operations and the relatively short time scale (5 years) of the study make it impossible to assess any time trends in presentation however it will be interesting to see if these observations change for 2nd or 3rd generation immigrants.
The key question is whether this very low incidence of non-cardiac vascular surgery reflects unmet need due to cultural differences or language barriers or whether they are due to a low prevalence of disease. In an attempt to answer this question we conducted a pilot study to investigate the logistical difficulties and feasibility of establishing the prevalence of lower limb peripheral vascular disease and AAA in our local Asian population.
Prevalence of non-cardiac vascular disease in the local Asian population
What is most striking about the data from the pilot study is the contrast between the high prevalence of cardiac disease and risk factors such as smoking, diabetes, unfavourable lipid profile, hypertension and abdominal obesity, and the very low prevalence of non-cardiac vascular disease; specifically, lower limb peripheral arterial disease and aortic dilation. How do these data compare with the findings of others? Unfortunately, although the epidemiology of cardiac vascular disease in the UK Asian population has received considerable attention,8, 9, 10 to our knowledge no other study has specifically examined the prevalence of non-cardiac vascular disease in the Asian community. Only a few studies have looked at the incidence of non-cardiac vascular surgery amongst Asians. One study, from Bradford investigated the incidence of surgery for AAA and reported no operations on Asians (14% of their catchment population) over a 7-year period.11 A case controlled study from London estimated that diabetic South Asians have about a quarter of the risk of lower limb amputation when compared to diabetic Europeans and although it was postulated that this was due to a reduced prevalence of peripheral arterial disease, ankle: brachial pressure indices were not measured.12
By contrast, the epidemiology of non-cardiac vascular disease is well established in the UK Caucasian population. Specifically, the Edinburgh Artery Study found a prevalence of lower limb peripheral arterial disease (ABPI<0.9) to be 17%,1 a figure almost 10 times higher than that seen in our survey of Asian men of a similar age. Furthermore, it is now well established that the prevalence of AAA (aortic diameter>3.0cm) among middle-aged Caucasian men living in the UK is in the region of 5%.13, 14 By contrast, none of the middle-aged Asian men in the current study had an aortic diameter over 26mm and over half were less than 20mm. Taken in conjunction with the current and the previous Bradford information on the very low incidence of surgery for AAA, these novel ultrasound data strongly suggest that age-for-age Asian men appear very much less likely to develop aortic dilatation in response to vascular risk factors than Caucasians.
Limitations of the survey
As the current study is survey of a self-selecting group of Asian men, interesting and novel though the data are, it is important that they are not over-interpreted and that the investigators address possible sources of bias. It is possible that men with vascular disease might have selected themselves out and that this might account for the low prevalence of non-cardiac vascular disease. However, the high prevalence of vascular risk factors and of cardiac vascular disease in the study group suggests that this is not the case. Indeed, the authors' impression is quite the opposite; men who had already suffered from cardiovascular disease and/or considered themselves at greatest risk appeared to be the most enthusiastic to attend. It has also been suggested that patients with symptomatic lower limb peripheral arterial disease would not have been able to attend because of exercise-limiting claudication. While this is possible, though perhaps unlikely given the importance of mosque attendance within the Muslim community, the prevalence of asymptomatic peripheral arterial disease detected on the basis of reduced ABPI would not have been affected. Similarly, while it is possible that Asian men who were already aware that they had an AAA may have decided not to attend, this would not have affected the ultrasound-based data on aortic diameter. Furthermore, as discussed above, concerns that the prevalence of AAA has been under-estimated are mitigated by hospital-based data indicating that elective or emergency AAA repairs are much less common among Asians. The present study only looked at men and while we are not aware of any data on UK Asian women, it seems likely that their prevalence of non-cardiac vascular disease will be lower rather than higher. Finally, it is well-known that different Asian groups, like different Caucasian Europeans, have different vascular risk factor profiles related to both life-style and, possibly, genetic differences.8, 9 Although one of the strengths of the present study is that the subject group is relatively homogenous being predominantly (>75%) Kashmiri-born Pakistani Muslims, obviously, present data cannot be extrapolated to other racial, ethnic and religious Asian groupings. Nor may they be relevant to Asian men born and raised in the UK and who are now approaching middle age.15
The need for and feasibility of future epidemiological studies
Notwithstanding the issues of possible bias and other limitations discussed above, the present study has provided novel and thought-provoking data regarding the prevalence of non-cardiac vascular disease in Asians; especially when they are interpreted in the context of the hospital-based information from the same catchment area on the incidence of surgery. However, a properly powered, cross-sectional, epidemiological survey of randomly selected, male and female, cohorts of different Asian groupings would be necessary to elucidate further these complex issues. The cultural, communication and logistical problems that the investigators have encountered and had to overcome in during this pilot study indicate that completing such a study, while possible, would be a major challenge and require very significant funding.
Having said that, in the future, relative changes in birth rates mean that so-called ‘ethnic minorities’ are likely to account for an increasing proportion of the populations of many developed countries.16 Understanding the epidemiology of any condition is essential for the planning and delivery of clinically and cost-effective health care for affected individuals. Currently the screening, diagnosis, investigation, medical and surgical treatment of non-cardiac vascular disease is based almost exclusively upon data gathered from Caucasians. Such care may be inappropriate, even dangerous, in other racial and ethnic groups. It is hoped that a better understanding of the prevalence of these common and morbid conditions among UK Asians will improve the healthcare needs of this large population.
Acknowledgements
We thank Ellen Drew and Emma Burke, vascular research nurses, for their help with data collection and we thank Arif Choudhary, general secretary of the Noor-ul-Uloom mosque, Small Heath, Birmingham for allowing us access to the premises to study the subjects. We also thank Tim Marshall, University of Birmingham for his help with the statistical analysis.
Simon Hobbs is supported by a British Heart Foundation Junior Research Fellowship (FS/03/026/15467) and the Royal College of Surgeons (England) ‘Lea Thomas’ Research Fellowship.
References
- . Edinburgh Artery Study: prevalence of asymptomatic and symptomatic peripheral arterial disease in the general population. International Journal of Epidemiology. 1991;20:384–392
- A 21-year experience of abdominal aortic aneurysm operations in Edinburgh. British Journal of Surgery. 1998;85:645–647
- Stroke care: reducing the burden of disease. The Stroke Association, 1998.
- . Ethnicity and peripheral arterial disease. European Journal of Vascular and Endovascular Surgery. 2003;25:505–512
- . What's in a name? Accuracy of using surnames and forenames in ascribing ethnic identity in English populations. Journal of Epidemiology and Community Health. 1986;40:364–368
- A comparison of proxy measures of abdominal obesity in Chinese, European and South Asian adults. Diabetic Medicine. 1999;16:853–860
- . The incidence of small abdominal aortic aneurysms and the change in infrarenal aortic diameter: implications for screening. European Journal of Vascular and Endovascular Surgery. 2001;21:165–170
- . What is the risk of coronary heart disease in South Asians? A review of UK research. Journal of Public Health Medicine. 2000;22:375–385
- . Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in South London. Heart. 1997;78:555–563
- . Differences in biological risk factors for cardiovascular disease between three ethnic groups in the Whitehall II study. Atherosclerosis. 1999;142:279–286
- . Epidemiology of abdominal aortic aneurysms in the Asian community. British Journal of Surgery. 2001;88:382–384
- . Risk of diabetes-related amputation in South Asians vs. Europeans in the UK. Diabetes Medicine. 2002;19:99–104
- . Epidemiology and potential for prevention of abdominal aortic aneurysm. British Journal of Surgery. 1998;85:155–162
- . Abdominal aortic aneurysm in 4237 screened patients: prevalence, development and management over 6 years. British Journal of Surgery. 1991;78:1122–1125
- Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California. Incidence of myocardial infarction and death from coronary heart disease. American Journal of Cardiology. 1977;39:239–243
- . London: Office for National Statistics; 1999;
Presented to The Vascular Society AGM, Harrogate, UK, November 2004.None of the authors have any financial and personal relationships with other people or organisations that could inappropriately influence (bias) this work.
PII: S1078-5884(06)00163-8
doi:10.1016/j.ejvs.2006.03.010
© 2006 Elsevier Ltd. All rights reserved.
Volume 32, Issue 5 , Pages 494-499, November 2006
