21 / 84
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 29, No 2, March/April 2018
AFRICA
83
insufficient echocardiographic and electrocardiographic data
were excluded.
The local ethics committee approved the study. All participants
provided written, informed consent prior to participation in the
study.
Transthoracic echocardiographic examinations were
performed using a commercially available cardiac ultrasound
scanner (Acuson Sequoia 512 system with 2.5–4.0 MHz
transducer, Siemens Mountain View, California, USA) in the
left lateral position, according to the criteria of the American
Society of Echocardiography.
16
During echocardiography a
continuous one-lead ECG recording was done.
Left ventricular end-diastolic and end-systolic volumes were
determined in the apical view, and stroke volume and EF were
measured using the modified Simpson’s equation.
16
LV mass
(LVM) was calculated with the Devereux formula as:
LVM (g)
=
1.04 [(LVID + PWT + IVST)³ – LVID³] – 14
Where LVID
=
LV internal dimension; PWT
=
posterior wall
thickness; IVST
=
interventricular septum thickness. LVM was
indexed to body surface area (BSA) by dividing LVM by BSA.
Peak early diastolic (E) velocity, atrial contraction (A)
velocity and E-wave deceleration time (DT) were measured
from the transmitral pulsed-wave Doppler spectra, and the
E/A ratio was calculated. Pulsed-wave tissue Doppler imaging
(TDI) was performed in an apical four-chamber window with
a sample volume of 5 mm and the monitor sweep speed was
set at 100 mm/s to optimise the spectral display of myocardial
velocities. All Doppler spectral velocities were averaged over
three consecutive beats. The average pulsed-wave TDI-derived
early (E
′
) diastolic myocardial velocity was obtained from the
lateral and septal sides of the mitral annulus. Then the E/E
′
ratio
was calculated to provide an estimation of LV filling pressures.
17
The TDI-derived late-diastolic wave (A
′
) was obtained from the
mitral lateral annulus.
LA diameter was measured from the parasternal long axis
with M-mode echocardiography. LA volumes were traced and
calculated by means of the modified Simpson’s method from
apical four- and two-chamber views, according to the guidelines
of the American Society of Echocardiography and European
Association of Cardiovascular Imaging.
16
LA volumes were
measured as: (1) just before the mitral valve opening, at
end-systole (maximal LA volume or V
max
); (2) at the onset of the
P wave on electrocardiography (pre-atrial contraction volume or
V
olp
); and (3) at mitral valve closure, at end-diastole (minimal LA
volume or V
min
). From these, the following measurements were
calculated:
•
LA passive emptying volume (PEV)
=
V
max
– V
olp
•
LA passive emptying fraction (PEF)
=
PEV/V
max
×
100
•
LA active emptying volume (AEV)
=
V
olp
– V
min
•
LA active emptying fraction (AEF)
=
AEV/V
olp
×
100
•
LA total emptying volume (TEV)
=
V
max
– V
min
•
LA total emptying fraction (TEF)
=
TEV/V
max
×
100.
Left atrial volumes were indexed to BSA in all patients.
18
Statistical analysis
Statistical analyses were performed with the MedCalc Statistical
Software version 12.7.7 (MedCal Software bvbv, Ostend,
Belgium; 2013). All continuous variables are expressed as mean
±
standard deviation and median (minimum–maximum). All
categorical variables are defined as frequency and percentage.
All continuous variables were checked with the Kolmogorov–
Smirnov normality test to show their distributions. Continuous
variables with normal distributions were compared using the
unpaired Student’s
t
-test, while continuous variables with
abnormal distributions were compared using the Mann–Whitney
U
-test. For categorical variables, the chi-squared test was used.
Pearson or Spearman’s correlation analyses were used to
determine the associations between LA volume and function,
and various laboratory parameters and 2D echocardiographic
diastolic parameters. Multivariate evaluations were performed
using linear regression analysis. The confounders that were
found to have a statistically significant impact on the dependent
variable on univariate analysis were described as the independent
variables in a multivariate linear regression analysis model. The
p-
values less than 0.05 were considered significant.
Sample size justification: according to the article ‘Effects
of diabetes mellitus on left atrial volume and functions in
normotensive patients without symptomatic cardiovascular
disease’,
8
the V
max
value for DM2 patients was 40.9
±
11.9 ml,
and for the control group, 34.6
±
9.3 ml. The mean difference was
assumed as 6.3 ml; the standard deviation of the DM2 group was
11.9 ml and of the control group, 9.3 ml. With the assumption of
5% of type 1 error (a) and 80% power (1b), the sample size was
calculated at 46 patients for each group. With a 20% drop-out
rate, a minimum of 56 patients (112 in total) would have to be
enrolled in the study.
Results
The study population consisted of 112 subjects (52 male, mean
age 51.7
±
7.0 years). Patient characteristics, analysed according
to the two groups, are shown in Table 1. The groups were similar
regarding age and gender. In the DM2 group, 44 (78.6%) patients
were hypertensive and 33 (58.9%) were receiving insulin and oral
antidiabetic agents. Patients in the DM2 group were also taking
more medications, such as acetylsalicylic acid, angiotensin
converting enzyme inhibitors, beta-blockers and statins than the
control group.
Body mass index (BMI) and levels of triglycerides (TG), high-
sensitivity C-reactive protein (hsCRP), uric acid, fasting glucose
and HbA
1c
were significantly higher in the DM2 group compared
with the control group (
p
<
0.05). There were no significant
differences regarding total cholesterol and low- (LDL) and high-
density lipoprotein (HDL) cholesterol levels between the groups
(
p
> 0.05) (Table 1).
Table 2 reports the results of 2D echocardiographic
parameters reflecting diastolic function with preserved systolic
function. Twelve (21.4%) subjects in the control group and 29
(51.8%) patients in the DM2 group had some degree of diastolic
dysfunction. Mitral A wave, E/E
′
ratio and mitral A
′
wave were
significantly higher, and mitral E
′
wave was significantly lower in
the DM2 group compared with the controls (
p
<
0.05).
There were no significant differences between the groups
regarding EF, mitral E wave and E/A ratio (
p
> 0.05). LA
diameter, and indexed V
max
, V
olp
, V
min
, AEV and TEV were found
to be significantly higher in the DM2 group compared with the
controls (
p
<
0.05). PEF was significantly lower in the DM2
group compared with the controls (
p
<
0.05). Between the two