CARDIOVASCULAR JOURNAL OF AFRICA • Vol 22, No 3, May/June 2011
126
AFRICA
pattern, compared to the hypertensives with normal geometry.
Those with concentric hypertrophy had the highest Tei index and
were significantly different from those with concentric remod-
elling. However, there was no significant difference in the Tei
index between those with eccentric and concentric hypertrophy.
Univariate correlation coefficients of the echocardiographic
parameters with Tei index (myocardial performance index) are
shown in Table 3. Ejection fraction, fractional shortening, mitral
E/A ratio and left ventricular internal systolic dimensions were
significantly correlated with the Tei index. Multiple regression
analysis revealed that the Tei index was significantly related to
only ejection fraction (–0.232), fractional shortening (
r
=
–0.142)
and mitral E/A ratio (
r
=
–0.280) (Table 4).
Discussion
This study revealed some important findings. Firstly, the Tei
index was significantly different between hypertensive subjects
with normal and those with abnormal left ventricular geometry.
Secondly, concentric hypertrophy was the commonest abnormal
pattern of left ventricular geometric pattern in this study, followed
by concentric remodelling and then eccentric hypertrophy. Only
17.7% of the hypertensive subjects had normal left ventricular
geometry. Thirdly, the Tei index was highest among hyperten-
sive subjects with concentric hypertrophy, closely followed by
those with eccentric hypertrophy, with no significant difference
between the two groups. Fourthly, multiple regression analysis
suggests that there was no significant association between left
ventricular geometric pattern and Tei index.
The pattern of Tei index in this study therefore indicated
the pattern/severity of combined systolic and diastolic function
associated with each type of left ventricular geometry among
these hypertensive subjects and was not a function of the left
ventricular geometry itself. Earlier studies have found concentric
and eccentric hypertrophy to be associated with a higher degree
of cardiovascular events, and systolic and diastolic dysfunction.
4-6
The Tei index was also not associated with age, left ventricular
mass or body mass index in this study population.
Ejection fraction, fractional shortening and mitral E/A ratio
were the main echocardiographic parameters associated with
Tei index among hypertensive Nigerian subjects. This study
therefore demonstrates the potential clinical usefulness of the Tei
index in stratifying left ventricular systolic and diastolic dysfunc-
tion across all age strata and patient groups, notwithstanding the
body mass index and left ventricular mass.
Use of the Tei index for assessing cardiac performance has a
potential clinical advantage over the use of other classical echo-
cardiographic indices for estimating left ventricular diastolic and
systolic function. Ejection fraction and left ventricular volumes,
which are measures of systolic function, are prone to large errors
when the ellipsoid shape of the heart is altered. Age, rhythm,
preload and afterload changes affect the transmitral Doppler
signal, which is a measure of diastolic function. Therefore the
use of these conventional indices when left ventricular geometry
is altered may be further error-prone.
Left ventricular geometric pattern has been shown to be
related to diastolic and systolic function.
8-10
However, the Tei
index is not affected by these factors and is therefore more likely
to be a better estimate of systolic and diastolic function in a large
group of the population, irrespective of confounding factors, as
shown in various studies.
11,18,19
This study further corroborates
other studies that have shown that the Tei index is independent
of flow haemodynamics, preload, afterload, age and left ventricu-
lar geometry among a variety of subjects, including paediatric
subjects.
11,19-21
Several geometric variations are cardiac adaptations to pres-
TABLE 3. CORRELATION OF ECHOCARDIOGRAPHIC
PARAMETERSWITHTEI INDEX
Variable
Correlation
p
LVIDd (mm)
0.124
0.114
LVIDs (mm)
0.159
0.044*
EF
–0.209
0.008*
Mitral E/A ratio
–0.198
0.006*
FS
–0.187
0.018*
IVSd
–0.013
0.822
PWTd
0.018
0.818
RWT
–0.018
0.816
*Statistically significant. SBP: systolic blood pressure, DBP: diastolic blood
pressure, LVIDd: left ventricular internal dimension in diastole, LVIDs: left
ventricular internal dimension in systole, PWTd: posterior wall thickness in
diastole, ET: ejection time, IVCT: isovolumic contraction time, IVRT: isovolu-
mic relaxation time, BMI: body mass index, LVMI: left ventricular mass index,
RWT: relative wall thickness.
TABLE 4. CORRELATION COEFFICIENTS OF
LINEAR REGRESSIONANALYSIS BETWEEN
THE TEI INDEXAND OTHER PARAMETERS
Clinical parameter
Correlation coefficient
p
Age
0.101
0.218
SBP
0.101
0.220
DBP
0.049
0.551
BMI
0.145
0.079
PWTd
0.018
0.830
IVSd
–0.005
0.954
MEARAT
–0.280
0.045*
EF
–0.232
0.004*
FS
–0.142
0.032*
RWT
0.001
0.886
*Statistically significant. SBP: systolic blood pressure, DBP: diastolic blood
pressure, LVIDd: left ventricular internal dimension in diastole, LVIDs: left
ventricular internal dimension in systole, PWTd: posterior wall thickness in
diastole, ET: ejection time, IVCT: isovolumic contraction time, IVRT: isovolu-
mic relaxation time, BMI: body mass index, LVMI: left ventricular mass index,
RWT: relative wall thickness.
TABLE 2. CLINICALAND ECHOCARDIOGRAPHIC PARAMETERS
AMONGVARIOUS LEFTVENTRICULAR GEOMETRIC PATTERNS
IN HYPERTENSIVES
Variable
Normal
geometry
(29)
Concentric
remodelling
(43)
Concentric
hypertrophy
(68)
Eccentric
hypertrophy
(24)
Age (years)
54.4
±
12.0 55.1
±
12.2 58.3
±
12.1 56.1
±
11.0
IVSd (mm)
10.7
±
3.0
12.3
±
2.5
‡
15.7
±
7.9
‡†
13.7
±
6.1
‡
PWTd (mm) 10.4
±
2.3
11.5
±
1.4
‡
15.0
±
1.9
‡†
10.4
±
2.2
‡
ET (msec)
283.0
±
32.8 275.1
±
37.8 280.0
±
59.5 243.0
±
86.7
†
IVRT (msec) 97.1
±
25.6 97.8
±
25.9 100.7
±
31.9 101.9
±
30.1
IVCT (msec) 99.1
±
35.4 94.8
±
39.3 103.3
±
56.1
†
105.0
±
37.9
RWT
0.38
±
0.05 0.60
±
0.11
‡
0.70
±
0.69
‡†
0.36
±
0.07
†
Tei index
0.61
±
0.2
0.71
±
0.2
‡
0.83
±
1.0
‡†
0.80
±
0.2
‡
‡
p
<
0.05 vs normals;
†
p
<
0.05 vs concentric remodelling.
PWTd: posterior wall thickness in diastole, ET: ejection time, IVCT: isovolu-
mic contraction time, IVRT: isovolumic relaxation time, RWT: relative wall
thickness.
