3.1 Study population and baseline characteristics
In total, 112 consecutive participants were enrolled in this study. Table 1 shows the baseline characteristics of the two DAPT groups. There were no significant differences in mean age, BMI and proportion of males between the two groups. The patients receiving aspirin and clopidogrel had higher LVEF, more history of hypertension and less history of diabetes. The patients treated with aspirin and ticagrelor had more previous PCI and myocardial infarction (MI) and more frequent use of ACEI. In the laboratory data acquired after PCI, the levels of hsCRP, ALT and AST were markedly higher in the group given aspirin and ticagrelor. Interestingly, patients in the aspirin + clopidogrel group had a significantly higher value of MPV (11.4±1.4 vs 10.2±0.9 fL, P < 0.001). Nevertheless, no significant difference was found in PLT count between the two groups.
Table 1. Demographics of the study population.
Variables
|
Aspirin + clopidogrel group
(n = 46)
|
Aspirin + ticagrelor group
(n = 66)
|
P
|
Age, yrs.
|
58.1±9.1
|
56.3±11.6
|
0.401
|
Male, n (%)
|
36(78.3)
|
55(83.3)
|
0.499
|
BMI, kg/m2
|
25.9±3.3
|
25.9±3.2
|
0.963
|
LVEF, %
|
63.0(60.0, 66.0)
|
60.0(52.8, 65.0)
|
<0.001
|
Risk factors
|
|
|
|
Hypertension, n (%)
|
33(71.7)
|
31(47.0)
|
0.009
|
Diabetes, n (%)
|
9(19.6)
|
26(39.4)
|
0.026
|
Hyperlipidemia, n (%)
|
43(93.5)
|
58(89.2)
|
0.441
|
Past history
|
|
|
|
Previous PCI, n (%)
|
9(19.6)
|
28(42.4)
|
0.011
|
Previous MI, n (%)
|
17(37.0)
|
56(84.8)
|
<0.001
|
Previous stroke, n (%)
|
10(22.2)
|
4(6.1)
|
0.012
|
Medications
|
|
|
|
Statins, n (%)
|
46(100.0)
|
66(100.0)
|
1.000
|
ACEI, n (%)
|
11(24.4)
|
29(46.0)
|
0.022
|
ARB, n (%)
|
8(18.2)
|
7(11.1)
|
0.300
|
β-blocker, n (%)
|
38(84.4)
|
60(93.8)
|
0.112
|
Laboratory
|
|
|
|
Hemoglobin, g/L
|
147.7±14.0
|
143.5±12.4
|
0.095
|
LDL-C, mmol/L
|
2.0(1.7, 2.8)
|
2.2(1.5, 2.7)
|
0.708
|
HsCRP, mg/L
|
1.1(0.5, 3.0)
|
2.6(1.2, 7.2)
|
0.006
|
AST, U/L
|
22.0(19.0, 27.5)
|
27.0(20.0, 85.0)
|
0.007
|
ALT, U/L
|
27.0(17.0, 39.2)
|
32.5(22.8, 52.5)
|
0.022
|
Platelet, 109 cells/L
|
223.1±60.2
|
244.7±69.7
|
0.091
|
MPV, fL
|
11.4±1.4
|
10.2±0.9
|
<0.001
|
Values are mean±SD if the distribution is normal; median (interquartile range) if skewed; number, n (proportions, %) for categorical variables.
ACEI: angiotensin converting enzyme inhibitor; ALT: alanine transaminase; AST: aspartate transaminase; ARB: angiotensin receptor blocker; BMI: body mass index; CABG, coronary artery bypass graft; hsCRP: highly sensitive C-reactive protein; LDL-C, low-density lipoprotein cholesterol; LVEF: left ventricular ejection fraction; MI, myocardial infarction; MPV, mean platelet volume; PCI, percutaneous coronary intervention.
3.2 Platelet function testing in patients of the two DAPT groups
As measured by LTA, the maximal light transmittance (LT) changes in response to 5 µM ADP in the aspirin + clopidogrel group was dramatically higher than that in the aspirin + ticagrelor group (38.8 ± 16.2 % vs 18.5 ± 9.5%, P < 0.001; Fig. 1A). According to the results of the VASP test, the aspirin + ticagrelor group presented a significantly lower PRI level than the aspirin + clopidogrel group (23.9 ± 15.9 % vs 63.5 ±21.5%, P < 0.001; Fig. 1B).
Using the criteria which have been reported in the literature[14, 27], the HPR rates were compared between the different DAPT groups and platelet function results (Fig. 2). In the LTA assay, the HPR rate differed significantly between the two groups (aspirin + clopidogrel: aspirin + ticagrelor = 21.7:1.5, P < 0.001; Fig. 2A). Similar results could also be observed in the VASP assay, with a higher HPR rate in those given aspirin and clopidogrel (aspirin + clopidogrel: aspirin + ticagrelor = 73.9:7.6, P < 0.001; Fig. 2B).
In addition, as shown in Additional file 2 (Fig. S2), a significant positive correlation was observed between the results for LTA and VASP assay (r = 0.660, P < 0.001). Multivariate regression analysis for prediction of HPR by the LTA and VASP assay is presented in Additional file 3 (Table S1). After multivariate adjustment for previous percutaneous coronary intervention, previous stroke, MPV, hsCRP, left ventricular ejection fraction, and aspartate transaminase, dual antiplatelet therapy remained independently significant in predicting HPR in the LTA and VASP assay.
3.3 Association between mean platelet volume and high on-treatment platelet reactivity
To investigate whether MPV could indicate the level of HPR in patients receiving DAPT, the participants were divided into three groups, according to MPV tertiles (< 10.0 fL; 10.0-11.0 fL; ≥ 11.0 fL). Table 2 shows the main clinical characteristics and biochemistry results according to the MPV values. Larger platelets were associated with higher LVEF (P = 0.026), higher Hb (P = 0.011), lower AST levels (P = 0.010) and lower PLT count (P = 0.001), and with less history of previous MI (P = 0.031). Notably, more patients in the highest MPV tertile received aspirin and clopidogrel (P < 0.001). With regard to the platelet function results, ADP-mediated platelet aggregation (P = 0.006) and PRI (P <0.001) was much higher in patients of the highest MPV tertile. A higher percentage of HPR was observed at VASP assay in patients with higher MPV (P < 0.001), and this was also true for ADP-induced aggregation at LTA test (P = 0.048). Furthermore, analysis showed that there existed a positive correlation between MPV levels and PRI (r = 0.488, P < 0.001), as well as maximal LT changes in response to 5 µM ADP (r = 0.343, P < 0.001) (Fig. 3). There was also an inverse relationship between MPV and platelet count (r = - 0.413, P < 0.001).
Table 2. Clinical characteristics and chemistry results according to mean platelet volume.
Variables
|
I tert
< 10.0 fL
(n = 37)
|
II tert
10.0-11.0 fL
(n = 42)
|
III tert
≥11.0 fL
(n = 33)
|
P
|
Age, yrs.
|
58.2±10.3
|
55.7±12.1
|
57.4±9.2
|
0.566
|
Male, n (%)
|
33(89.2)
|
31(73.8)
|
27(81.8)
|
0.216
|
BMI, kg/m2
|
26.1±3.5
|
25.9±3.4
|
25.5±2.6
|
0.758
|
LVEF, %
|
60.0(53.0, 65.0)
|
63.0(60.0, 65.0)
|
65.0(58.0, 68.0)
|
0.026
|
Risk factors
|
|
|
|
|
Hypertension, n (%)
|
20(54.1)
|
20(47.6)
|
24(72.7)
|
0.083
|
Diabetes, n (%)
|
14(37.8)
|
14(33.3)
|
7(21.2)
|
0.304
|
Hyperlipidemia, n (%)
|
34(91.9)
|
38 (90.5)
|
29(87.9)
|
0.850
|
Past history
|
|
|
|
|
Previous PCI, n (%)
|
15(40.5)
|
12(28.6)
|
10(30.3)
|
0.489
|
Previous MI, n (%)
|
29(78.4)
|
28(66.7)
|
16(48.5)
|
0.031
|
Previous stroke, n (%)
|
4(10.8)
|
4(9.5)
|
6(18.2)
|
0.494
|
Medications
|
|
|
|
|
Statins, n (%)
|
37(100.0)
|
42(100.0)
|
33(100.0)
|
1.000
|
ACEI, n (%)
|
17(45.9)
|
14(33.3)
|
9(27.3)
|
0.245
|
ARB, n (%)
|
6(16.2)
|
3(7.1)
|
6(18.2)
|
0.313
|
β-blocker, n (%)
|
36(97.3)
|
34(81.0)
|
28(84.8)
|
0.078
|
Laboratory
|
|
|
|
|
Hemoglobin, g/L
|
140.0±13.0
|
147.0±13.2
|
148.8±11.9
|
0.011
|
LDL-C, mmol/L
|
2.4(1.7, 2.8)
|
2.2(1.6, 2.8)
|
2.0(1.7, 2.2)
|
0.457
|
HsCRP, mg/L
|
3.1(1.2, 8.4)
|
1.8(0.9, 4.7)
|
1.5(0.5, 6.7)
|
0.232
|
AST, U/L
|
25.0(20.5, 85.0)
|
25.5(20.8, 38.2)
|
21.0(16.5, 28.0)
|
0.010
|
ALT, U/L
|
26.0(22.0, 42.5)
|
32.0(21.5, 52.0)
|
29.0(17.0, 40.0)
|
0.474
|
Platelet, 109 cells/L
|
257.3±73.4
|
243.9±63.2
|
201.4±48.3
|
0.001
|
Maximal LT changes in response to 5μM ADP, %
|
17.0(12.4, 25.7)
|
20.6(12.1, 36.5)
|
32.0(16.4, 51.5)
|
0.006
|
PRI, %
|
21(11.5, 28.0)
|
31.0(23.0, 66.5)
|
61.0(30.5, 79.0)
|
<0.001
|
Percentage of HPR-LTA, %
|
0(0)
|
6(14.3)
|
5(15.2)
|
0.049
|
Percentage of HPR-VASP, %
|
4(10.8)
|
15(35.7)
|
20(60.6)
|
<0.001
|
Dual antiplatelet therapy
|
|
|
|
|
Traditional therapy, n (%)
|
6(16.2)
|
18(42.9)
|
22(66.7)
|
<0.001
|
New therapy, n (%)
|
31(83.8)
|
24(57.1)
|
11(33.3)
|
<0.001
|
Values are mean±SD if the distribution is normal; median (interquartile range) if skewed; number, n (proportions, %) for categorical variables.
ADP, adenosine diphosphate; HPR, high on-treatment platelet reactivity; LT, light transmittance; PRI, platelet response index. Others were as Table 1.
As can be seen in Table 3, the independence of MPV and MPV tertiles in predicting HPR at VASP assay was subsequently determined using univariate and multivariable logistic regression analysis. Results indicated that there was an independent association between MPV level and the increased prevalence of HPR in the VASP assay, as well as MPV tertiles. As the increasing trend of MPV tertiles, it also presented higher OR values which indicating a higher prevalence of HPR. After adjustment for percutaneous coronary intervention, previous stroke, dual antiplatelet therapy, left ventricular ejection fraction and aspartate transaminase, MPV was significantly and independently associated with HPR (OR=2.105, 95% CI:1.175–3.771, P = 0.012) whereas MPV tertiles were no longer statistically significant. Nevertheless, MPV and MPV tertiles did not present significant independence in predicting HPR at LTA testing. In addition, analysis of ROC curves showed a potent ability of MPV in predicting the presence of HPR at VASP assay in patients undergoing PCI and receiving DAPT (area under curve [AUC] = 0.788, 95% CI: 0.701-0.875, P < 0.001) (Fig. 4A). The cut-off value of MPV in predicting HPR was 10.55 fL. Furthermore, analysis of ROC curves was also conducted in patients given aspirin and clopidogrel (Fig. 4B). MPV could predict HPR at VASP assay for those patients (area under curve [AUC] = 0.729, 95% CI: 0.552-0.906, P = 0.019) with a cut-off value of 11.65 fL.