Characterising atherothrombosis in Hong Kong: results of the Hong Kong data from a global atherothrombosis epidemiological survey.

Abstract

To describe the characteristics of patients in Hong Kong with or at risk of atherothrombosis, to determine the proportion of symptomatic patients with more than one vascular bed affected, and to assess the relationship between ankle brachial index and disease severity.
Local participation in an international prevalence study.
Five centres in Hong Kong.
A total of 210 subjects were recruited (105 women and 105 men). Patients were divided into the symptomatic group (with current or previous atherothrombotic symptoms, n=101) and at-risk group (with no current or previous symptoms, but aged over 55 years with at least two specified risk factors, n=109).
Patient characteristics were described, including the number of arterial beds affected, ankle brachial index, presence of risk factors, and medications taken.
Of the symptomatic patients, 30% had more than one arterial bed involved. A total of 55.4% of the symptomatic group and 18.4% of the at-risk group had abnormal ankle brachial index values. Lower ankle brachial indices were associated with a greater number of affected arterial beds. Diabetes mellitus and hypertension were the most prevalent risk factors in the at-risk group. Symptomatic patients were commonly treated with antihypertensive and antiplatelet agents, whereas at-risk patients were mostly treated with antihypertensive and antidiabetic agents. Only 20% of at-risk patients were taking antiplatelet agents.
Ankle brachial index is a useful tool for predicting those at risk of atherothrombosis. This simple measurement can be used as part of the screening process in the general practice. The role of antiplatelet agents in primary prevention of atherothrombotic events in at-risk patients deserves further attention.

Full text

Page 1

36 Hong Kong Med J Vol 11 No 1 February 2005
ORIGINAL ARTICLE
Characterising atherothrombosis in
Hong Kong: results of the Hong Kong
data from a global atherothrombosis
epidemiological survey
?? !"#$%&'()*+,-./0123456
??
Key words:
Arteriosclerosis;
Cardiovascular diseases;
Epidemiologic factors;
Hong Kong
?? !
?? !"
?? !"#
?? !"#$
??
Hong Kong Med J 2005;11:36-41
○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○○
United Christian Hospital, 130 Hip Wo
Street, Kwun Tong, Hong Kong:
Diabetes Ambulatory Care Centre,
Department of Medicine and Geriatrics
MW Tsang, MB, BS, FRCP
YW Kam, MB, ChB, FHKAM (Medicine)
Department of Surgery
KH Kwong, MB, ChB, FHKAM (Surgery)
Flat B, 30/F, Villa Veneto, 3 Kotewall Road,
Hong Kong
CL Chu, MB, BS
Room 1815, East Point Centre, 555
Hennessy Road, Causeway Bay, Hong Kong
CK Lam, MD, FHKAM (Medicine)
Room 2409-10, Hang Lung Centre, 2-20
Paterson Street, Causeway Bay, Hong Kong
SY Ngan, MB, ChB, FRACGP
M/F, 475 Nathan Road, Yaumatei, Kowloon,
Hong Kong
YK Yu, MB, BS, FRCS (Edin)
This paper was supported by funding from
Sanofi-Synthelabo HK Ltd.
Correspondence to: Dr MW Tsang
(e-mail: mwtsang@ha.org.hk)
MW Tsang ??
CL Chu ??
YW Kam ??
KH Kwong ??
CK Lam ??
SY Ngan ??
YK Yu ??
Objectives. To describe the characteristics of patients in Hong Kong with
or at risk of atherothrombosis, to determine the proportion of sympto-
matic patients with more than one vascular bed affected, and to assess the
relationship between ankle brachial index and disease severity.
Design. Local participation in an international prevalence study.
Setting. Five centres in Hong Kong.
Participants. A total of 210 subjects were recruited (105 women and 105
men). Patients were divided into the symptomatic group (with current or
previous atherothrombotic symptoms, n=101) and at-risk group (with no
current or previous symptoms, but aged over 55 years with at least two
specified risk factors, n=109).
Main outcome measures. Patient characteristics were described, includ-
ing the number of arterial beds affected, ankle brachial index, presence of
risk factors, and medications taken.
Results. Of the symptomatic patients, 30% had more than one arterial
bed involved. A total of 55.4% of the symptomatic group and 18.4% of
the at-risk group had abnormal ankle brachial index values. Lower ankle
brachial indices were associated with a greater number of affected
arterial beds. Diabetes mellitus and hypertension were the most prevalent
risk factors in the at-risk group. Symptomatic patients were commonly
treated with antihypertensive and antiplatelet agents, whereas at-risk
patients were mostly treated with antihypertensive and antidiabetic agents.
Only 20% of at-risk patients were taking antiplatelet agents.
Conclusions. Ankle brachial index is a useful tool for predicting those at
risk of atherothrombosis. This simple measurement can be used as part of
the screening process in the general practice. The role of antiplatelet agents
in primary prevention of atherothrombotic events in at-risk patients
deserves further attention.
?? ?? !"#$#%&"#'()*+,-./01234567
?? !"#$%&'()*+,'-./0123456789:;<
?? !
?? ?? !"#$% &'()*
?? ?? !"#$%&'
?? !?ONM?? !"#$%&'()*+,-.!"/0$12)
?? !"#$?? !"#$%&'()*+,-./01234
?NMN?? ?? !"#$% ?? !"#$%&RR=?? !"#$%&

Page 2

Hong Kong Med J Vol 11 No 1 February 2005 37
Characterising atherothrombosis in Hong Kong
Introduction
Cardiovascular disease, ischaemic heart disease, and
stroke are the leading causes of death worldwide,
accounting for 52% of all deaths in 2000.1 In Hong
Kong, heart disease and cerebrovascular disease
are the second and third main causes of death, repre-
senting 26.8% of all mortalities in the same year.2 It is
now recognised that atherothrombosis is the common
factor in all of these conditions. In the past, coronary
heart disease, cerebrovascular disease, and peripheral
artery disease (PAD) were regarded as discrete local-
ised entities, but they are actually clinical manifesta-
tions of the atherothrombosis. Atherothrombosis is
an extensive vascular disease affecting the coronary,
cerebral, and peripheral arterial beds to varying
degrees of severity. The atherothrombotic process
is progressive, in which the ruptured or eroded
atherosclerotic plaque activates platelets, causing a
thrombus formation and resultant vascular injury.
Factors in the different arterial beds including blood
flow patterns, vessel diameter, and arterial wall struc-
ture all affect the precise nature of the atherothrombotic
process.3 The cascade of events leads to various clini-
cal manifestations, such as intermittent claudication,
transient ischaemic attacks, stroke, or myocardial
infarction (MI).
Individual patients can often have atherothrombosis
in more than one arterial bed—in the Clopidogrel
Versus Aspirin in Patients at Risk of Ischaemic Events
(CAPRIE) trial, at least two arterial beds were in-
volved in 26% of patients.4 Hence, an instance of
atherothrombosis would indicate that this pathologi-
cal process is already manifest in other arterial
territories, regardless of whether symptoms are
evident or not. Indeed, patients who survive their first
atherothrombotic episode have a greatly increased risk
of recurrent attacks, whether in the same or separate
arterial bed. It has been reported that between 16%
and 24% of patients with MI will have a stroke or
transient ischaemic attack within 10 years, whereas
25% to 45% of patients who have had a stroke will
develop coronary artery disease in the same period.5
The prevalence of atherothrombosis is increasing
worldwide and co-morbidities, such as hypertension,
obesity, and diabetes, exacerbate its development
resulting in the increased risk of stroke or MI.6 The
well-documented increasing burden of obesity and
diabetes worldwide means that effective diagnosis and
management of atherothrombosis will become ever
more important in future years.
One simple method of predicting atherothrombotic
risk is the ankle brachial index (ABI). The ratio of
highest ankle pressure divided by the highest brachial
systolic pressure is typically between 0.9 and 1.1 in
healthy individuals, whereas values below 0.9 indicate
atherosclerosis with associated thrombotic risk.7 Leng
et al8 found that in patients with an ABI of ≤0.9, the
relative risk of non-fatal MI, stroke, and cardiovascu-
lar death in the 5 years post-baseline was 1.38, 1.98,
and 1.85, respectively, compared with subjects whose
ABI was above 0.9. This simple, accessible measure-
ment can be useful to warn a patient of the relative
risk of atherothrombosis and to indicate the priority
for treatment.
A Global ATHerothrombosis Assessment
(AGATHA) was an international cross-sectional
survey with the objective of better describing the
characteristics of patients with or at risk of
atherothrombosis.9 The study described the prevalence
with specific risk factors, such as diabetes or
hypertension, the number of symptomatic
atherothrombotic patients with more than one
vascular bed involved, and the relationship of ABI
values to disease severity. This paper discusses the
results of a subanalysis of the Hong Kong data gener-
ated in this study.
Methods
The AGATHA is an international prevalence survey
of 8950 patients from 471 sites in 24 countries. Nine
countries from Asia participated, which represented
?? !"#$?%&' NMV???
?? !"#$?? !"#$%&'()*+ ,-./01,.2?34#56789:;
?? ?? !"#$%&'()*+,-./01PMB?? !"#$%&'()*+,-.)*/0-*
?? !RRKQB?NUKQB?? !"#$%&'()*+,"-.#/?01234567829:;<=>)
?? !"#?$%&'()*+,-./01+-23456#?78&9:;)*+,-.1+<=?/
?? !"#$OMB ?? !"#$%&'(
?? ?? !"#$%&'()#*+,-./0123"456789:;<53=>?@ABC6DEF
?? !"#$%&'()*+(,-./0123456789:;<=%>?@ABC9DE"

Page 3

Tsang et al
38 Hong Kong Med J Vol 11 No 1 February 2005
Figure legends
optional carotid echo-Doppler, which was performed
in only 5.7% of patients in Hong Kong).
Fig 1.
Fig 2.
25% of the worldwide recruitment. General
practitioners, angiologists, cardiologists, neurologists,
diabetologists, internists, and vascular surgeons acted
as referral agents. In Hong Kong, 210 patients from
five sites were enrolled consecutively from April to
October 2002. Patients participated in one study visit,
with no subsequent follow-up.
Patient selection criteria
Patients were categorised into one of two groups if
they met either of the following criteria:
(1) The symptomatic group, who had previous or
current atherothrombotic symptoms, was identi-
fied as one of the following: prior ischaemic
stroke, transient ischaemic attack, prior MI,
history of stable or unstable angina, established
PAD, or prior vascular intervention; or
(2) The at-risk group who had no history of a prior
event or current symptoms, but was aged 55 years
or over with at least two of the following risk
factors:
(i) Diabetes mellitus (type 1 or type 2)
(ii) Dyslipidaemia or hyperlipidaemia, defined as
recent (≤3 months) with laboratory findings of:
• total cholesterol greater than or equal to
6.21 mmol/L (≥240 mg/dL), or
• low-density lipoprotein cholesterol greater than
or equal to 4.14 mmol/L (≥160 mg/dL), or
•high-density lipoprotein cholesterol smaller
than or equal to 0.9 mmol/L (≤35 mg/dL), or
• triglycerides greater than or equal to
2.26 mmol/L (≥200 mg/dL), or
• currently using a lipid-lowering agent for
dyslipidaemia
(iii) Hypertension, defined as:
•systolic blood pressure ≥140 mm Hg, or
•diastolic blood pressure ≥90 mm Hg, or
• currently on antihypertensive treatment
(iv) Obesity, defined as body mass index (BMI)
≥25 kg/m2
(v) Current (or former) smoker with consumption of
≥10 pack-years (1 pack-year is equal to smoking
1 pack of 20 cigarettes per day for 1 year, or 2
packs per day for half a year, etc)
Subjects were excluded from the study if they
had cerebral disease of non-atherothrombotic origin
or neurological signs and symptoms due to non-
ischaemic causes (eg neoplasm). Written informed
consent was obtained from every participant in accord-
ance with the Good Clinical Practice’s requirements.
Data analysis
At each centre, patients were seen by the investigator
who recorded their sex, age, race, any history of
previous vascular events, any current cardiovascular
symptoms, presence of the specified risk factors, and
any current medications. Subjects were evaluated by
physical examination (weight, height, BMI, heart
rate, and blood pressure) and by clinical tests
(electrocardiography, ABI measurement, and the
Ankle brachial index was measured using the
ELITE 100R Doppler (Nicolet Vascular, Wisconsin,
United States) with one 5-MHz vascular probe—
the identical model was used at all AGATHA
sites. Training of using this equipment was provided
to all participating sites before patient enrolment. The
brachial systolic pressure was taken in both arms by
use of the Doppler probe in the antecubital fossa.
Systolic blood pressure was then recorded at the
left and right posterior tibial arteries and dorsalis
pedis arteries. The ABI was calculated for each side
by taking the highest ankle pressure of each measured
side, divided by the absolute highest brachial pressure
(from either side). The lower of the two resultant
ABI values was used for risk classification. In this
study, ABI values ≥0.90 were considered normal,
0.71-0.89 mild obstruction, 0.31-0.70 moderate
obstruction, and 0.00-0.30 severe obstruction. A
descriptive analysis was performed on patient charac-
teristics and extent of atherothrombosis. Continuous
parameters were summarised using mean, median, and
95% confidence interval, whereas categorical param-
eters were summarised using counts and percentages.
Results
Subject characteristics are shown in Table 1. All
participants were Chinese with an equal male to
female ratio for the total number of subjects. The
median age was 69.6 years, the youngest enrolled
patient was aged 36.9 years, and the oldest was aged
91.4 years. The median weight was 59 kg, and the
median BMI was 24.5. This index was associated with
‘observed risk’ for type 2 diabetes and cardiovascular
disease in the Asian population.10 The two groups were
very similar in terms of age, weight, height, BMI, heart
rate, and blood pressure. Abnormal electrocardiograms
were more common in the symptomatic group (60.4%
compared with 28.4% in the at-risk group); however,
this measurement was not taken in 19% of subjects so
that these data were incomplete.
Results from the symptomatic group showed that
29.7% patients had atherothrombosis in two or more

Page 4

Hong Kong Med J Vol 11 No 1 February 2005 39
Characterising atherothrombosis in Hong Kong
arterial beds, more than 54% had prior history or
current symptoms of coronary artery disease, whereas
43.6% had symptoms of PAD, and 39.6% had a
cerebrovascular event. By definition, all the at-risk
patients were over 55 years of age and had at least two
risk factors; however, the majority of subjects (58.7%)
had three or more of the specified risk factors. The
overall breakdown of the results was 41.3% with two
risk factors, 41.3% with three, 16.5% with four, and
0.9% with five. Of these results, diabetes mellitus and
hypertension were the most prevalent risk factors in
the at-risk population (Fig 1).
The pattern of ABI results varied between the
symptomatic and the at-risk subjects (Fig 2). Among
the at-risk patients, 81.6% had normal ABI ratios,
and percentage of patients declined steeply when the
severity categories increased. In contrast, there was a
more even distribution of the symptomatic patients
across the abnormal ABI categories compared with the
at-risk group (Fig 2). The total percentage of sympto-
matic patients with an abnormal ABI value was 55.4%,
which indicated the further progression of the condi-
tion compared with the at-risk group with 18.4% who
had abnormal ABI values.
The percentage of symptomatic subjects with
abnormal ABI values greatly increased with the number
of arterial beds affected (Fig 3). All patients had
abnormal ABI values when three arterial beds were
affected. Abnormal ABI values were not restricted to
patients with peripheral disease; 89% of patients with
PAD had an abnormal ABI, whereas 30% of patients
without PAD also registered abnormal results. This
again indicates that ABI measurement can reflect
not only peripheral disease but also systemic
atherothrombotic disease.
An analysis of the medications showed that, as
expected, the symptomatic group were prescribed more
c i t s i r e t ca r ah
e f / e l am(xe
) s r aey (eg
k ( t hg i eW
c ( t hg i eH
mydoB
e t a rt r aeH
c i l o t syS
c i l o t sa i D
mr onbA
a r be l knA
* Data are expressed as median (95% confidence interval)
†Not performed in 19% of patients
C
S
A
) 901= n (p
0 . 6
8-
01-
1-0
-4 . 71 (
-74 (0
01 (0
06 ( 58
4 . 82
4 . 81
uorg
5 / 0 . 44
5 (1 . 86
95
651
9 . 42
7
61
k s i r - tA ) 101= n (puor
6 . 3
9-
41-
1-0
-3 . 41 (
-35 (2
01 (0
0 5 (0
4 . 06
4 . 55
gc i t am
4 / 4 . 65
3 (3 . 17
95
751
2 . 42
7
51
8
o tpmyS) 012=
0 . 0
-
41-
-0
-3 . 41 (
-74 (2
01 (0
0 5 (48
8 . 34
2 . 63
n ( l a t oT
5 / 0 . 05
3 (6 . 96
95
751
5 . 42
7
61
]%[ ) e l am
) g
m
sa
)
m
*
/ g
) n i
e r uss
e r usse r p
o i d r aco r t ce l e
dn i l a i hc
k (
m
xed
/ s t ae
e r pdo
doo l b
l a
n is
b (
2)
) g
) gH
%
Hm
mm
m
%(
m
(
(o l b
)(a r g
9 . 0<)xe
) 5 . 7
) 9
) 18
) 1 . 53
) 90
) 022
) 0 11
5 . 5
73 (
41 (
1
-5
-
) 4 . 1
) 7
) 97
) 6 . 75
) 56
) 012
) 50
9 . 6
2 3 (
21 (
1
-
1
0
-
) 4 . 19
) 7
) 181
) 6 . 75
) 56
) 022-
) 011
9 . 6
23 (
21 (
1
0
-
Table 1. Characteristics of Hong Kong patients with atherothrombosis from A Global ATHerothrombosis
Assessment survey
*
*
*
†
*
*
*
HypertensionDiabetes
mellitus
DyslipidaemiaSmoking
history
Obesity
Risk factor
0
20
40
60
80
100
No. of patients (%)
Fig 1. Prevalence of risk factors in the at-risk group
A patient may have two or more risk factors
NormalMild
obstruction
Moderate
obstruction
Severe
obstruction
ABI category
0
20
40
60
80
100
At-risk group
Symptomatic group
No. of patients (%)
Fig 2. Distribution of ankle brachial index (ABI) in
at-risk and symptomatic patients
(i) ABI values: ≥0.90=normal; 0.71-0.89=mild obstruction;
0.31-0.70=moderate obstruction; 0.00-0.30=severe
obstruction. (ii) Mild obstruction, moderate obstruction,
and severe obstruction come into abnormal ABI category

Page 5

Tsang et al
40 Hong Kong Med J Vol 11 No 1 February 2005
medications (Table 2) with most taking antihyperten-
sive and antiplatelet agents, whereas half of them
were on vasodilators, lipid-lowering or antidiabetic
agents. In contrast, the at-risk group were commonly
prescribed antihypertensive and antidiabetic agents.
Only 20.2% of the at-risk group were on an antiplatelet
therapy (acetylsalicylic acid, clopidogrel, ticlopidine
or dipyridamole) compared with 85.1% of the symp-
tomatic group.
Discussion
Atherothrombotic disease and its sequelae represent
a major health burden worldwide. This burden is
predicted to increase globally in the coming decades.
The continuing ‘westernisation’ of diet and lifestyle
in Hong Kong and China is leading to increased
susceptibility to atherosclerosis.11 Prevalent risk
factors, such as diabetes, smoking, hypertension, and
hypercholesterolaemia are commonly found in Hong
Kong patients with arterial disease, as demonstrated
in this and other studies.12 Prompt and appropriate
management of patients will be increasingly impor-
tant in limiting the future burden of this disease.
Patients with atherothrombosis often have more
than one vascular bed affected. In the current study,
29.7% of Hong Kong patients had atherothrombosis
in two or more arterial beds, which was similar to the
result of 26% that was reported in the CAPRIE trial.4
Hence, a diagnosis of PAD should indicate a serious,
potentially fatal, and an underlying systemic problem
that requires treatment.
Effective treatments for atherothrombosis exist, but
there are apparent discrepancies in their application
in clinical practice. Patients who have experienced
coronary or cerebral events are more likely to receive
antiplatelet therapy together with intensive lipid and
hypertension treatments compared with those with
PAD.13 Awareness of atherothrombosis associated with
systemic disease does not, as yet, appear to have
changed treatment patterns—an affected peripheral
arterial bed seems to be regarded as a less important
indication for treatment than affected cerebral or
coronary beds. However, PAD is still a powerful
predictor of future MI, stroke, and increased mortality.
The 10-year mortality rate of PAD patients with
cardiovascular or coronary heart disease is approxi-
mately 6 times higher than in healthy controls, and for
a subgroup of severely symptomatic PAD patients, the
risk is increased to 15-fold.14
Screening at-risk patients for atherothrombosis is
vital to prevent potentially devastating strokes or heart
attacks. The application of ABI measurement as a
screening tool has many benefits. Ankle brachial
index is an inexpensive, non-invasive, and reliable
tool that enables general practitioners to identify
patients at risk of atherothrombosis. In this study, 30%
of non-PAD patients had an abnormal ABI, and a
recent German study of PAD in primary care settings
found that approximately every fifth unselected
(ie randomly chosen) patient (age-adjusted prevalence,
19.8%) had an ABI of <0.9, indicating generalised
atherothrombosis.15 Compared with angiography,
sensitivity and specificity of the ABI is about 90% and
98%, respectively, for a stenosis of 50% or more in
leg arteries.15 Relying on the appearance of symptoms
to make the diagnosis, for example, claudication, is
not recommended, because classic claudication is
found in fewer than 11% of PAD patients.13 Despite
the fact that Doppler ultrasound was used in our study,
it is re-assuring to learn from a recent publication by
Ostergren et al16 that ABI obtained by manual palpa-
tion of the arterial pulse is clinically valid. Moreover,
Normal ABIAbormal ABI
ABI category
1 arterial bed
2 arterial beds
3 arterial beds
0
20
40
60
80
100
No. of patients (%)
Fig 3. Distribution of ankle brachial index (ABI) in
symptomatic patients
The index is inversely associated with the number of
arterial beds affected
* Patients may take more than one medication
Table 2. Percentage of patients taking prescribed
medication*
no i t ac i deMks i r - tA
org
=n
%(
1 . 88
, pu
01
)
9
c i t am
, pu
101
)%
1 . 68
o tp
org
=n
(
myS, l a t oT
2=n
%(
01
)
ev i sne t r epyh i t nA
ega
aV
t e l e t a l p i t nA
o l -d i p i L
a i d i t nA
s t n
r o t a l i doss t ne
s t ne
ga
s t neg
g
g
g
a
a
a
s t ne n i r ew
c i t eb
8 3 . 7
2 . 02
6 . 73
9 . 98
5 . 15
1 . 58
5 . 15
5 . 45
1 . 78
6 . 82
4 . 15
3 . 44
9 . 27