CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 5, June 2013
166
AFRICA
One hundred and thirty-one STEMI patients were the study
subjects. All patients were admitted to Wonju Christian Hospital.
They had received reperfusion therapy using primary PCI
within 12 hours of symptom onset and had blood sampling
with a planned schedule for BNP. We then compared clinical
variables including plasma BNP levels and echocardiographic
data between the remodelling (RG) and non-remodelling groups
(NRG). The study period was from April 2006 to March 2009.
Inclusion criteria were ischaemic chest pain lasting
≥
30 min;
electrocardiographic ST-segment elevation
>
0.1 mV in two or
more limb leads, or
>
2 mV in two or more precordial leads or
new-onset left bundle branch block; and elevation in the level of
CK-MB or troponin I to
≥
twice the normal range.
Exclusion criteria were previous myocardial infarction
(MI); severe valvular heart disease; cardiomyopathy; impaired
renal function (creatinine
>
1.5 mg/dl); inadequate quality of
echocardiographic images; cardiogenic shock (initial systolic
blood pressure
<
90 mmHg); advanced heart failure (Killip class
≥
III); or life-limiting non-cardiac disease.
Success of revascularisation using primary PCI was defined
as residual stenosis
<
50% and if coronary flow in the culprit
vessel after primary PCI resulted in thrombolysis in myocardial
infarction (TIMI) grade
≥
2. The primary PCI procedure and
type of stent used were at the discretion of the interventional
cardiologist.
Coronary angiographic analysis during primary PCI was
performed by interventional cardiologists in the Ilsan Paik
Hospital cardiovascular centre, who were blinded to clinical
and plasma BNP findings. All patients were prescribed aspirin
(300 mg p.o.), clopidogrel (600 mg p.o.), and heparin (70 IU/
kg p.o.) before the procedure. Each patient was maintained
on aspirin (100 mg) and clopidogrel (75 mg) for
≥
12 months
after revascularisation. Also, most patients received
β
-blockers,
calcium-channel blockers, angiotensin receptor blockers (ARBs)
or angiotensin-converting enzyme inhibitors (ACEIs), and statins
at the discretion of the attending physician.
Blood samples were taken for BNP measurement on hospital
admission (acute phase), at two to five days (early phase), three
to four weeks (late phase), and at six months (long term) after
symptom onset. All plasma samples were obtained in plastic
tubes containing potassium ethylene diamine tetra-acetic acid
(EDTA; Becton Dickinson, Franklin Lakes, NJ, USA) with
amounts that ranged from 3–5 ml. All samples were centrifuged,
and plasma was tested singly for BNP using the Biosite
Triage Assay, a point-of-care device that uses a fluorescence
immunoassay technique (Biosite, San Diego, CA, USA).
The total coefficient of variation at different levels of plasma
BNP was reported to be
<
7% using control samples provided by
the manufacturer. The sensitivity for BNP in these measurements
ranges from 5–5 000 pg/ml. Levels of CK-MB and troponin
I were evaluated after symptom onset. The peak release of
CK-MB and troponin I was determined every four hours after
hospital admission for three days.
Two-dimensional echocardiography was undertaken at
baseline and at the six-month follow up. Echocardiographic
examinations and data were obtained using a commercially
available imaging system (Vivid 7; GE Medical Systems,
Milwaukee, WI, USA). Echocardiograhic data were sent to the
echocardiography laboratory in Ilsan PaikHospital cardiovascular
centre and analysed by echocardiography physicians blinded to
the laboratory data.
Apical four- and two-chamber views as well as apical long-
axis views were obtained from all patients. To assess regional
wall motion abnormalities, the wall of the LV was divided into
16 segments, as recommended by the American Society of
Echocardiography.
9
For each segment, the WMSI was derived.
LV end-diastolic volume (LVEDV), LV end-systolic volume
(LVESV) and LV ejection fraction (LVEF) were calculated using
a modified version of Simpson’s method.
Assessment of diastolic function was carried out by measuring
the mitral inflow pattern with pulsed-wave Doppler [E/A ratio,
and deceleration time (DT) of the E wave], pulmonary venous
inflow, and tissue Doppler velocities of the mitral annulus. The
ratio of early diastolic mitral annulus velocity (E/E
′
) was used as
an indicator for LV filling pressures.
10
PMIR was defined as
>
20% increment in LVEDV estimated
at the six-month follow-up echocardiography compared with
baseline results using a modified version of Simpson’s method.
2
Intra- and inter-observer variability of LVEDV and LVESV was
<
5% in this study.
Statistical analyses
Data were analysed using the SPSS statistical package, version
15.0 (SPSS Incorporated, Chicago, IL, USA). Data are mean
±
SD for continuous variables and frequency with percentages for
categorical variables. Because mean BNP levels were uneven,
natural log transformation was used in the regression analyses to
satisfy modelling assumptions.
Continuous variables were compared using the paired Student’s
t
-test. Categorical variables were compared using chi-square
analyses. Differences in proportions were compared using
Pearson’s chi-square test. Repeated-measures analysis of variance
(ANOVA) was used to analyse inter- and intra-group differences
between the RG and NRG with regard to plasma BNP levels;
p
<
0.05 was considered significant. Univariate and multiple logistic
regression analyses were carried out to estimate independent
predictors of PMIR. Variable selection in multivariable modelling
was based on statistical significance from univariate analysis.
The optimal time of BNP sampling for the prediction of
PMIR was determined by a multivariate model. The BNP cut-off
value for prediction of PMIR was assessed by receiver operator
characteristic (ROC) curve analyses. The predictive value of
plasma BNP level for PMIR was evaluated using estimation of
the area under the curve (AUC) separately for each parameter.
Results
The clinical characteristics of the study population are shown
in Table 1. All patients treated with primary PCI received at
least one stent implantation. PMIR was detected in 42 patients.
The mean age was older in the RG (RG vs NRG; 63.1
±
11.9
vs 58.1
±
11.1 years,
p
=
0.02). The mean time from symptom
onset to reperfusion was later in the RG, but was not statistically
significant (RG vs NRG; 5.4
±
2.3 vs 4.8
±
2.2 h,
p
=
0.07).
There were significant differences in the percentage of New
York Heart Association class I between the two groups (RG vs
NRG 57.1 vs 78.7%,
p
=
0.03). Moreover, mean peak levels of
CK-MB (RG vs NRG; 246.8
±
88.1 vs 170.9
±
109.9 ng/ml,
p
<
0.01) and troponin I (RG vs NRG; 48.3
±
28.3 vs 33.7
±
25.1
ng/ml,
p
<
0.01) were significantly higher in the RG. At hospital
