Study Participants:
Baseline characteristics and LV end-diastolic diameter:
Of 82 patients with reperfused chronic myocardial infarction, there are 69 male patient (84.1%) and the mean age was 58.16(range from 34 to 74 years). The 50 healthy controls included 36 male(72%) with an average age of 59.74±8.99 years(range from 51 to 69 years). The booster function of left ventricle of the patients was lower than the controls.(LVEF, 49.00±14.81 vs 64.25±8.55). Instead, the LVEDV, LVESV, SV, myocardial mass in the phase of both diastolic and systolic are higher in the patients.(LVEDV, 189.44±89.99 vs 130.38±31.22; LVESV, 94.12±76.90 vs 47.14±18.18; SV, 96.68±37.23 vs 83.24±20.22; Myo Mass(diastolic), 133.44±52.51 vs 91.87±25.31, Myo Mass(systolic), 137.05±65.59 vs 80.63±29.64, (P<0.05 for all)
There was no evident difference between the patients with reperfused chronic myocardial infarction and healthy control from the aspects of gender, age, height, weight, BMI, CO and CI(p>0.05 for all). The baseline characteristics of the patients with reperfused chronic myocardial infarction are recorded in Table 1 and Figure 2
Table 1 Baseline characteristics and LV end-diastolic diameter(LVEDD)
|
Reperfused chronic myocardial infarction
|
Healthy control
|
P=value
|
|
n=82
|
n=50
|
|
Gender, male, n(%)
|
69(84.1)
|
36(72)
|
0.12
|
Age(years)
|
58.16±9.58
|
59.74±8.99
|
0.348
|
Height(cm)
|
170.07±6.36
|
170.76±7.32
|
0.584
|
Weight(kg)
|
69.59±12.98
|
70.24±5.14
|
0.685
|
BMI(kg/cm2)
|
23.97±3.40
|
24.28±1.18
|
0.446
|
LVEF(%)
|
49.00±14.81
|
64.25±8.55
|
<0.001
|
LVEDV(cm3)
|
189.44±89.99
|
130.38±31.22
|
<0.001
|
LVESV(cm3)
|
94.12±76.90
|
47.14±18.18
|
<0.001
|
SV(cm3)
|
96.68±37.23
|
83.24±20.22
|
<0.001
|
CO
|
6.47±2.75
|
6.03±1.41
|
0.238
|
CI
|
3.52±1.49
|
3.43±0.90
|
0.688
|
Myo Mass(Diastolic)
|
133.44±52.51
|
91.87±25.31
|
<0.001
|
Myo Mass(systolic)
|
137.05±65.59
|
80.63±29.64
|
<0.001
|
Heart rate
|
67.50±12.77
|
73.08±9.11
|
<0.001`
|
Culprit coronary artery, n(%)
|
82
|
|
|
LAD, n (%)
|
41(50%)
|
|
|
LCx, n (%)
|
31(38%)
|
|
|
RCA, n (%)
|
10(12%)
|
|
|
LM, n (%)
|
0
|
|
|
Heart failure index
|
Value/Sum
|
|
|
NT-pro BNP
|
776.6/51
|
|
|
BNP
|
404.9/34
|
|
|
Data are expressed as mean±SD, or as n(%),
LA geometry and functional parameters between the reperfused chronic myocardial infarction and healthy control:
LA geometry parameters of LAVmax(cm3), LAVpac(cm3), LAVmin(cm3)(P<0.01 for all) of reperfused chronic myocardial infarction are higher than healthy control. LAVmax(cm3), LAVpac(cm3), LAVmin(cm3) increased from 59.91±20.73 to 87.92±51.77, from 67.80±40.41 to 41.08±17.39, from 45.93±33.75 to 24.92±13.72, respectively. On the contrary, it decreased a lot in the LATEF(%)(49.94±14.68 vs 60.32±11.85, P<0.01), LAAEF(%)(35.11±14.17 vs 41.65±13.39, P=0.02), LAPEF(%)(23.30±11.55 vs 32.11±12.13, P<0.01), MAPSE inferior(mm)(10.76±4.98 vs 15.55±2.73, P<0.01), MAPSE anterior(mm)(9.99±7.30 vs 12.36±2.70, P=0.01)(Table 2).The evaluation are more precise in the LATEF(%), LAAEF(%), LAPEF(%), MAPSE inferior(mm), MAPSE anterior(mm) with AUC of 0.675, 0.593, 0.686, 0.834, 0.757, respectively.(Fig. 3).
Table 2 LA geometry and functional parameters between the reperfused chronic myocardial infarction and healthy control
Parameters
|
Reperfused chronic myocardial infarction
|
Healthy control
|
P=value
|
n=82
|
n=50
|
|
LA function parameters
|
LAVmax(cm3)
|
87.92±51.77
|
59.91±20.73
|
<0.001
|
LAVpac(cm3)
|
67.80±40.41
|
41.08±17.39
|
<0.001
|
LAVmin(cm3)
|
45.93±33.75
|
24.92±13.72
|
<0.001
|
LATEF(%)
|
49.94±14.68
|
60.32±11.85
|
<0.001
|
LAAEF(%)
|
35.11±14.17
|
41.65±13.39
|
0.02
|
LAPEF(%)
|
23.30±11.55
|
32.11±12.13
|
<0.001
|
Other parameters
|
MAPSE inferior(mm)
|
10.76±4.98
|
15.55±2.73
|
<0.001
|
MAPSE anterior(mm)
|
9.99±7.30
|
12.36±2.70
|
0.01
|
Data are expressed as mean±SD, or as n(%),
LA strain parameters between the reperfused chronic myocardial infarction and healthy control:
Compared with the healthy controls, the patients of reperfused chronic myocardial infarction had lower strain parameters including total longitudinal strain(29.75±15.82 vs 40.38±11.39, P<0.01), passive longitudinal strain(16.24±10.55 vs 24.53±8.46, P<0.01), booster longitudinal strain(13.11±6.87 vs 15.85±5.39, P=0.01), total radial strain(16.47±6.38 vs 20.65±3.32, P<0.01), passive radial strain(7.20±3.54 vs 9.49±2.52, P<0.01) and booster radial strain(8.96±3.94 vs 11.16±3.17, P<0.01). GLSRs(1.50±0.88 vs 1.87±0.67, P=0.02), GLSRe(-1.46±0.92 vs -1.97±0.77, P<0.01), GRSRs(-0.81±0.75 vs -1.60±0.35, P=0.01), GRSRe(0.49±0.55 vs 0.76±0.26, P<0.01) and GRSRa(1.02±0.69 vs 1.31±0.38, P<0.01) differed a lot between the patients and healthy control. There is no obvious difference between the reperfused chronic myocardial infarction and healthy controls in GLSRa(P=0.28), peak radial displacement(P=0.05) and peak longitudinal displacement(P=0.58)(Table 3). In addition, passive longitudinal strain showed the greatest distinctive ability with the largest area under the ROC curve(AUC) of 0.762. Following passive longitudinal strain, the distinctive ability of total longitudinal strain, total radial strain and passive radial strain decease a little with the area under the ROC curve(AUC) of 0.729, 0.719 and 0.723(Fig.3). Although decreasing a lot among the patients with reperfused chronic myocardial infarction, booster longitudinal strain, booster radial strain, GLSRs, GLSRe, GLSRa, GRSRs, GRSRe, GRSRa had a little weaker differentiating ability with the AUC of 0.613, 0.650, 0.665, 0.695, 0.584, 0.635, 0.692, 0.620, respectively.(Fig. 4).
The correlations between LATEF(%) and geometry variables
In regards of the correlations between LATEF(%) and imaging or structural variables, LATEF(%) showed moderate but outstanding correlation with total longitudinal strain(r=0.574, P<0.001), booster longitudinal strain(r=0.577, P<0.001), GLSRs(r=0.540, P<0.001), GLSRa(r=0.567, P<0.001) and LAVmin(r=-0.674, P<0.001)(Fig. 6). However, Variety of radial strain and strain rate parameters mildly correlated to LATEF(%) with r value of 0.525, 0.363, 0.477, -0.461, 0.340, 0.493, respectively.(P<0.001 for all). Except for LAVmin with r value of -0.674, LA function is also mildly related to LATEF(%) with the r value of -0.404, -0.544, respectively.(P<0.001 for all). The r value of passive longitudinal strain, GLSRe, peak radial displacement, peak longitudinal displacement, MAPSE inferior and MAPSE anterior are 0.503, -0.531, 0.428, 0.491, 0.332, 0.378, respectively.(P<0.001 for all) (Fig. 5)
The correlations between LVEF and geometry variables
As far as the relation between LVEF and these different strain and strain rate parameters, booster longitudinal strain, total radial strain, booster radial strain, GLSRa, GRSRa are significantly but mildly in relation to LVEF with r value of -0.231, -0.221, -0.287, 0.286, -0.321, respectively. (P<0.05 for all). LVEF was negatively in relation to the LATEF(%)(r=-0.255, P=0.021) and LAPEF(%)(r=-0.306, P=0.005). Else, BMI(r=0.229, P=0.008), LVESV(r=-0.636,P<0.001), LVEDV(r=-0.749, P<0.001), Myo Mass(systolic)(r=-0.507, P<0.001), Myo Mass(diastolic)(r=-0.523, P<0.001) correlated to LVEF significantly.(Fig. 6)
Univariable and multivariable regression with LA CMR-derived parameters as a dichotomous variable
This univariable regression analysis shows that passive radial strain(HR,0.612;P=0.047), booster radial strain(HR,1.226;P=0.032), GLSRa(HR,19.144;P=0.017), MAPSE(inferior)(HR,1.411;P<0.001), MAPSE(anterior)(HR,1.208,P=0.017) are significantly associated with primary results among the whole variables(Table 4). Based on univariable regression analysis, the consequent multivariable regression analysis reveals that among covariates including passive radial strain, booster radial strain, GLSRa(longitudinal), MAPSE(inferior),MAPSE(anterior), passive(HR,0.797;P=0.005) and booster radial strain(HR,0.794;P=0.013) and MAPSE(inferior)(HR,0.789;P<0.001) are only three parameters significantly in relation to primary results(Table 4). The C-statistics of this model is 0.838, reflecting strong reliability, with likelihood ratio test P<0.001.
Table 4 Univariable and multivariable regression with LA CMR-derived parameters as a dichotomous variable
|
Univariable Analysis
|
Multivariable Analysis
|
Variables
|
Unadjusted Hazard Ratio
|
P value
|
Unadjusted Hazard Ratio
|
P value
|
LA strain parameters
|
Total longitudinal strain
|
1.104(0.832,1.465)
|
0.491
|
|
|
Passive longitudinal strain
|
1.108(0.780,1.575)
|
0.567
|
|
|
Booster longitudinal strain
|
0.968(0.868,1.080)
|
0.564
|
|
|
Total radial strain
|
1.438(1.001,2.065)
|
0.050
|
|
|
Passive radial strain
|
0.612(0.378,0.993)
|
0.047*
|
0.797(0.679,0.934)
|
0.005*
|
Booster radial strain
|
1.226(1.018,1.476)
|
0.032*
|
0.794(0.662,0.953)
|
0.013*
|
LA strain rate
|
GLSRs
|
1.770(0.366,8.555)
|
0.477
|
|
|
GLSRe
|
6.173(0.925,41.184)
|
0.060
|
|
|
GLSRa
|
19.144(1.684,217.593)
|
0.017*
|
0.496(0.215,1.144)
|
0.100
|
GRSRs
|
1.610(0.139,18.577)
|
0.703
|
|
|
GRSRe
|
15.837(0.799,313.900)
|
0.070
|
|
|
GRSRa
|
2.361(0.231,24.122)
|
0.469
|
|
|
LA function parameters
|
LAVmax(cm3)
|
0.886(0.703,1.117)
|
0.306
|
|
|
LAVpac(cm3)
|
1.083(0.712,1.648)
|
0.709
|
|
|
LAVmin(cm3)
|
1.013(0.772,1.330)
|
0.925
|
|
|
LATEF(%)
|
0.721(0.448,1.161)
|
0.178
|
|
|
LAAEF(%)
|
1.236(0.921,1.659)
|
0.158
|
|
|
LAPEF(%)
|
1.315(0.889,1.944)
|
0.170
|
|
|
Other parameters
|
Peak displacement radial (mm)
|
0.847(0.447,1.604)
|
0.610
|
|
|
Peak displacement long. (mm)
|
2.105(0.958,4.627)
|
0.064
|
|
|
MAPSE inferior(mm)
|
1.411(1.192,1.669)
|
P<0.001*
|
0.789(0.699,0.892)
|
P<0.001*
|
MAPSE anterior(mm)
|
1.208(1.034,1.411)
|
0.017*
|
1.000(0.920,1.088)
|
0.991
|
*Statistically significant difference.