CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 3, April 2012
132
AFRICA
with lipid-lowering medication. Patients were considered to be
smokers if they were current smokers or had stopped smoking
less than three years before enrollment. Positive family history
of CAD was considered in patients with a history of at least one
first-degree relative affected by CAD before the age of 55 and 65
years for male and female relatives, respectively.
The main outcome evaluated in this study was recurrence of
a cardiovascular event, defined as recurrence of unstable angina
(UA), MI, performing coronary angioplasty during the study
period, coronary artery bypass graft (CABG) surgery, or sudden
cardiac death (SCD). Data regarding these outcomes were
collected when patients were visited at pre-scheduled one-month
and one-year intervals from their enrollment, to be evaluated
for short- and long-term prognosis, respectively. When patients
did not attend their appointments, they or their families were
contacted by telephone. If the participant had died, the cause of
death was determined by taking a history from family members
and reviewing medical documents. Otherwise participants were
asked to attend an appointment for follow up.
The main independent variable evaluated in the study was
H pylori
infection, defined as seropositivity for anti-
H pylori
antibodies. Both IgG and IgA antibodies against
H pylori
were tested for each participant at enrollment, and a patient
was considered seropositive if positive for either IgA or IgG
antibodies.
All participants were enrolled on their first day of admission
for ACS, and all blood samples were collected the next morning
and tested the same day.
H pylori
IgG and IgA antibodies were
determined using commercially available ELISA assays (RADIM
kit, Milan, Italy). According to the kit’s reference values, levels
of more than 30 and less than 15 units? were considered positive
and negative, respectively. Levels between these values were
considered borderline. A total of 10 participants had borderline
levels for
H pylori
IgA, and were designated as having a
negative result. At enrollment and during each follow-up visit,
patients underwent a complete evaluation, including a physical
examination, electrocardiography, and review of all paraclinical
data and medical history pertaining to study outcomes and/or
H
pylori
eradication.
Statistical analysis
Data are presented as number (percent) or mean (
±
SD).
Possible association between categorical variables was assessed
using the
χ
2
test. Relative risk was calculated to estimate the
increase in risk of incidence of outcomes among
H pylori
seropositive patients, compared to seronegative participants.
Logistic regression models were used to identify significant
determinants of incidence of outcomes. All statistical tests used
are reported with two-tailed estimates of type I error (
p
-value);
p
<
0.05 was considered significant. All statistical analysis was
done using SPSS, version 13.0 (SPSS Inc, Chicago, USA).
Results
A total of 450 subjects were enrolled in the study, of whom
433 completed the study and were followed up for one year.
Seventeen patients were excluded either because of
H pylori
eradication during the study period or becoming unavailable for
follow up. Patients’ ages ranged between 29 and 85 years, with
a mean of 60.9 (
±
12.3) years. Of 433 patients, 245 (56.6
%
)
were male. Of all the participants, 204 patients (47.1
%
) were
seropositive for
H pylori
. Of the whole study population, 69
(15.9
%
) patients developed short-term outcomes, defined as
being diagnosed with UA or MI, or undergoing angioplasty,
CABG or SCD during the first month after enrollment; 194
(44.8
%)
participants developed long-term outcomes, defined as
occurrence of the same conditions mentioned above during the
one-year follow-up period.
Table 1 summarises the prevalence of the five evaluated
classic risk factors among the participants at the time of
enrollment. Hypertension had the highest prevalence, as 221
(51.0
%
) patients were hypertensive, and DM showed the lowest
prevalence, as only 109 (25.1
%
) subjects were diabetic.
Table 1 presents the number and percentage of patients with
each risk factor who ultimately developed short- and long-term
outcomes. When evaluated by
χ
2
test, the rate of short- and long-
term outcomes among patients with and without each risk factor
did not differ significantly (
p
>
0.05). In other words, none of
the classic risk factors including hypertension, DM, smoking,
hyperlipidaemia and family history of CAD were associated with
occurrence of cardiovascular events during the study period. In
addition, the classic risk factors did not show any association
with
H pylori
seropositivity, as similar numbers of patients with
or without each risk factor were positive for
H pylori
antibodies
(
p
>
0.05).
TABLE 1. PREVALENCE OF CLASSIC CAD RISK FACTORSAMONG PARTICIPANTS, INCIDENCE
OFADVERSE OUTCOMES IN PATIENTSWITHANDWITHOUT EACH RISK FACTOR,
AND DISTRIBUTION OF
H PYLORI
SEROPOSITIVITYACROSS RISK FACTORS
Total study
population
(n = 433) n (
%
)
Patients with short-term
outcomes
n (
%
)
Patients with long- term
outcomes
n (
%
)
Patients with positive antibody to
H pylori
n (
%
)
Hypertension
+
221 (51.0)
33 (14.9)
100 (45.2)
106 (47.9)
-
212 (49.0)
36 (16.9)
94 (44.3)
98 (46.2)
Diabetes
+
109 (25.2)
19 (17.4)
55 (50.5)
51 (46.7)
-
324 (74.8)
50 (15.4)
139 (42.9)
153 (47.2)
Smoking
+
137 (31.6)
20 (14.6)
55 (40.1)
63 (45.9)
-
296 (68.4)
49 (16.5)
139 (46.9)
141 (47.6)
Hyperlipidaemia
+
126 (29.1)
22 (17.4)
53 (42.1)
57 (45.2)
-
307 (70.9)
47 (15.3)
141 (45.9)
147 (47.9)
Family history of CAD
+
111 (25.6)
15 (13.5)
53 (47.7)
49 (44.1)
-
322 (74.4)
54 (16.7)
141 (43.8)
155 (48.1)
