The search was done by using online libraries: Excerpta Medica database, Medline database, and Cochrane library (EMBASE/PubMed/Cochrane library). Medical subject headings (MeSH) terms and following keywords were applied: “Acute coronary syndrome,” “ACS,” “Acute myocardial infarction,” “AMI,” “microRNA,” “miRNA” and “miR.” Studies published in English, include specimens of humans, carried out from January 2010 onwards, and include patients with acute coronary syndrome (ACS) as an outcome are all included. Studies containing valvular heart disease or congenital heart disease are being excluded. Only case-control studies are included in this article, as many as 20 in number. Additionally, references of all publications from the initial search are also included for supplementary sources.
After a careful investigation, as many as 20 clinical studies were covered in this review. In total, 3560 subjects were included. The source of extraction of microRNA was plasma in 13 studies. The source of microRNA extraction was plasma in six studies, but only one study (Wang et al., 2011) used both plasma and whole blood as the source of microRNA extraction [6].
Table 1. A comparison table of clinical studies which investigated cardiac-specific microRNAs in acute coronary syndrome populations
No.
|
Author
|
Year
|
Study Population
|
Type
|
Outcome
|
1.
|
Wang et al. [7]
|
2010
|
AMI (n=33); Non AMI (n=33)
|
Case Control
|
ACS
|
2.
|
Adachi et al. [8]
|
2010
|
AMI (n=9); UAP (n= 5); CHF III (n=9); CHF II (n=6); Control (n=10)
|
Case Control
|
ACS
|
3.
|
Corsten et al. [9]
|
2010
|
AMI (n=32); Control (n=36)
|
Case Control
|
ACS
|
4.
|
D’Alessandra et al. [10]
|
2010
|
AMI (n=33); Control (n=17)
|
Case Control
|
ACS
|
5.
|
Ai et al. [11]
|
2010
|
AMI (n=93); Control (n=66)
|
Case Control
|
ACS
|
6.
|
Cheng et al. [12]
|
2010
|
AMI (n=31); Control (n=20)
|
Case Control
|
ACS
|
7.
|
Wang et al. [6]
|
2011
|
AMI (n=51); Control (n=28)
|
Case Control
|
ACS
|
8.
|
Widera et al. [13]
|
2011
|
STEMI (n=196); NSTEMI (n=131); Unstable angina (n=117)
|
Case Control
|
ACS
|
9.
|
Kuwabara et al. [14]
|
2011
|
ACS (n=29); Control (n=42)
|
Case Control
|
ACS
|
10.
|
Gidlof et al. [15]
|
2011
|
AMI (n=25); Control (n=11)
|
Case Control
|
ACS
|
11.
|
Devaux et al. [16]
|
2012
|
STEMI (n=397); NSTEMI (n=113); Control (n=87)
|
Case Control
|
ACS
|
12.
|
Olivieri et al. [17]
|
2012
|
NSTEMI (n=92); CHF (n=81); Control (n=99)
|
Case Control
|
ACS
|
13.
|
Orleman et al. [18]
|
2012
|
ACS (n=106); Non ACS (n=226)
|
Case Control
|
ACS
|
14.
|
Li et al. [19]
|
2013
|
AMI (n=117); AP (n=182); Control (n=100)
|
Case Control
|
ACS
|
15.
|
Gidlof et al. [20]
|
2013
|
AMI (n=319); Non AMI (n=88)
|
Case Control
|
ACS
|
16.
|
Li YQ et al. [21]
|
2013
|
AMI (n=67); Control (n=32)
|
Case Control
|
ACS
|
17.
|
Chen et al. [22]
|
2014
|
AMI (n=53); UA (n=20); Control (n=30)
|
Case Control
|
ACS
|
18.
|
Zhao et al. [23]
|
2015
|
AMI (n=59); Control (n=60)
|
Case Control
|
ACS
|
19.
|
Bialek et al. [24]
|
2015
|
STEMI (n=19); Stable CAD (n=12); Control (n=8)
|
Case Control
|
ACS
|
20.
|
Agiannitopoulos et al. [25]
|
2018
|
AMI (n=80); Control (n=50)
|
Case Control
|
ACS
|
Note: AMI = acute myocardial infarction; ACS = acute coronary syndrome; UAP = unstable angina pectoris; CHF = chronic heart failure; STEMI = ST elevation myocardial infarction; NSTEMI = non ST elevation myocardial infarction; AP = angina pectoris; UA = unstable angina; CAD = coronary artery disease.
Real-time polymerase chain reaction (RT-PCR) is the common technique used to extract microRNA in all involved studies. This systematic review concentrates mainly on cardiac-specific microRNAs, such as miR-1, miR-133a, and miR-208b, miR-499. In most available studies, these microRNAs sub-types were frequently observed to be up-regulated rapidly in the sample following myocardial necrosis. The majority of the studies further considered blood sampling timing once the symptoms started.
Four studies collectively investigated four miRNA, miR-1, miR-133a/b, miR-208a/b, and miR-499 [7,13,15,21]. Wang et al. in 2010 reported that each four microRNA levels were up-regulated, notably more increased than control (p <0.01), and positively associated with cTnI. At the same time, the mean interval of the sampling of the blood was 4.8+3.5 hours [7]. Compared to other miRNAs, a more prominent diagnostic significance for AMI is demonstrated in miR-208a, having 0.965 (95% CI, 0.920-1.000) as the area-under-curve (AUC). However, the diagnostic significance of miR-208a was weaker than CTnI, which employs an AUC of 0.987 (95% Cl, 0.966-1.000). In three AMI patients, they saw that the level of miR-208a was noticeable in one to four hours after the onset of chest pain, especially when the level of CTnI was identified to be under the cutoff value. Widera et al. in 2011 reported that compared to patients with unstable angina, patients with STEMI or NSTEMI were found to manifest with greater levels of miR-1, miR133a, and miR-208b [13]. miR-1, miR-133a, miR-133b, and miR-208b were found to be exclusively connected with CTnT levels in a multiple linear regression analysis which incorporated CTnT and clinical variables (p <0.001). In 2011, Gidlof et al. published that all miRNA levels were considerably more eminent in diseased populations than in those who are normal (p < 0.001) [15]. Although other miRNAs did not associate with EF or CTnT, miR-208b was negatively correlated with ejection fraction (EF) and positively correlated with cTnT [15]. Samples of blood were retrieved at 24 hours, 48 hours, and 72 hours. Within 12 hours of the onset of the symptoms in STEMI patients, the levels of miR-1, miR-133a, miR-208b, and miR-499-5p were increased. Li YQ et al. in 2013 revealed that miR-1, -133a, -208b, and -499 levels were considerably elevated in samples of blood or plasma that are retrieved within 12 hours of the onset of AMI. However, for the diagnosis of AMI, the four results of increased miRNAs levels were not remarkable to CTnT (p ˃0.05) [21].
As many as three studies investigated two microRNAs, miR-208b and miR-499, they found both upregulated [9,16,25]. In 2010, Corsten et al. identified a 1600-fold upregulation of miR-208b (p <0.005) and a 100-fold higher level of miR-499 (p <0.0005), and both biomarkers were significantly associated to CTnT (p=0.0005 in miR-208b and p=0.0001 in miR-499) when sampling time was less than 12 hours. In 2012, Devaux et al. recognized that the levels of miR-208b and miR-499 were significantly associated with CTnT and CK (p <10-9) with correlation coefficients of -0.18 (p <0.0008) and inversely associated with the Ejection Fraction with correlation coefficients of -0.17 (p <0.001). The study also found that miR-208b and miR-499 levels were elevated in STEMI than NSTEMI (p <0.0001) and elevated in AMI patients than in controls (p <0.0001) [16]. Agiannitopoulos et al. in 2018 identified that miR-208b and miR-499 were significantly elevated in AMI patients than in control patients (p <0.0001) and associated with CTnT (p <0.0001) [25].
The other three studies that investigated miR-499 individually found the biomarker to be upregulated [8,22,23]. In 2010, Adachi et al. observed that miR-499 was associated with CK-MB and meaningfully elevated in the AMI group compared to the other groups (p <0.0001). The blood sampling time was not beyond 48 hours after the onset of chest pain as a symptom [8]. The plasma concentration of miR-499 peaked at between six to 12 hours [8]. Chen et al. in 2015 published a study in which he took blood samples at 0 hours, 12 hours, 24 hours, three days, and seven days after the onset of chest pain [22]. The average duration of the onset of symptoms, which is chest pain, and the emergency room arrival was found to be 4.46±3.36 hours. Compared to those in unstable angina (UA) group and healthy control group, the relative plasma miR-499 level was meaningfully elevated in 53 patients with AMI (2.75±1.39 in UA group, 0.50±0.35 in the healthy control group, and 5.12±2.29 in AMI group). The discrepancies were statistically significant (p <0.01) and positively-correlated with CTnI and CK-MB (r=0.384 vs r=0.402, p <0.01 vs p <0.01, respectively) [22]. In 2015, Zhao et al. stated that in MI group, miRNA-499 was meaningfully elevated compared to controls (p <0.05) [23]. Three hours following the onset of chest pain in AMI, mRNA-499 could be identified in the serum, peaked after 12 hours, and progressively decreased after 15 hours [23]. In confirming AMI diagnosis, miR-499 (AUC of 0.915, 95% CI, 0.826-1.000) was inferior to CTnI (AUC of 0.971, 95% CI, 0.951-1.000) [23].
In two studies that examined miR-1, it was found that the biomarker is up-regulated [11,12]. In 2010, Ai et al. described that the level of miR-1 was remarkably elevated and positively associated with the cardiac troponin [11]. Cheng et al. in 2010 elaborated that the level of miR-1 was higher compared to healthy controls (p <0.05) and positively correlated with the level of CK-MB (r=0.68; p <0.05). The blood sampling average time was 8.5±3.82 hours [12].
In 2010, D’Alessandra et al. noticed the upregulation of miR-1, miR-133a, miR-133b, and miR-499-5p, and the downregulation of miR-122 and miR-375 [10]. The levels of miRNAs were remarkably altered as seen in the AMI group compared to the control group (p <0.01). miRNAs were also found to be correlated positively with CTnI (p <0.01). The average blood sampling time was 517+309 minutes following the onset of AMI. The plasma levels of miR-1, miR-133a, and miR-133b were formerly peaked at T0, or at a time point adjacent to the peak of CTnI. Reversely, miR-499-5p displayed a more gradual time progression and peaked behind the CTnI. In conclusion, the study summarized that after a three-day time development, the levels of miR-1, miR-133a, and miR-133b had recovered back to control levels.
Wang et al. in 2011 reported the upregulation of both miR-133 and miR-328 [6]. The 4.4-fold elevation of miR-133 in patients with AMI compared to the control group (p=0.006) in samples of whole blood was proportionate to plasma. The levels of miR-328 in patients with AMI were significantly elevated by 10.9-fold in plasma (p=0.033) and 16.1-fold whole blood compared to control (p <0.001). The samples were collected at T0 of 5.24±1.38 hours following AMI. There was an elevation of CTnI level at T0, stayed elevated till 20 hours, and was eventually returned to the normal amount seven days following the onset of symptoms. Nevertheless, the plasma or whole blood samples' levels of miR-133 and miR-328 were peaked at T0 already. The increase of miR-133 and miR-328 was reduced 20 hours following T0 and recovered to level of control seven days following T0. Still, the miR-133 and miR-328 displayed more rapid peaks compared to CTnI, and a positive association between miR-133 or miR-328 levels and the level of CTnI were also established.
Kuwabara et al. in 2011 inspected the upregulation of miR-1 (p <0.0005) and miR-133a (p <0.0001) [14]. A positive correlation with CTnT was also found in both miR-1 (p <0.005) and miR-133a (p <0.0001),
In 2012, Oerlemans et al. studied and elaborated the upregulation of miR-1 (OR of 1.44, 95% CI, 1.19-1.73), miR-208a (OR of 1.12, 95%CI, 0.95-1.35), miR-499 (OR of 1.38, 95% CI, 1.19-1.61), miR-21 (OR of 1.34, 95% CI, 1.15-1.55) and miR-146a (OR of 1.06, 95%CI, 0.97-1.15) [18]. miR-1, miR-499, and miR-21 was found to be superior compared to CTnT (p<0.001).
Gidlöf et al. in 2013 observed the upregulation of miR-1, miR-208b, and miR-499-5p [20]. miR-208b and miR-499-5p were markedly elevated in both patients with STEMI or NSTEMI compared to non-MI patients (p < 0.001). However, the result was inferior to CTnT, the modern gold standard cardiac marker. The average blood sampling time was 38.4 hours [20].
In 2013, Olivieri et al. came across that levels of plasma were upregulated in miR-1, miR-21, miR-133a, miR-208a, miR-423-5p and miR-499-5p [17]. Nonetheless, miR-499-5p displayed the greatest elevation contrasted with other miRNAs, all in NSTEMI versus control (p <0.001); NSTEMI versus congestive heart failure (CHF) (p <0.05), and CHF versus control (p <0.05). In the whole group, including the NSTEMI group, acute CHF group and control group, the level of miR-499-5p was markedly associated with CTnT (p <0.001) [17].
In 2013, Li et al. reported the upregulation of six miRNAs as opposed to the control. It was found in miRNAs as follow: miR-1, miR-223 and miR-499 (p <0.05), and miR-134, miR-186 and miR-208 (p <0.001) [19]. However, between these six miRNAs, only miR-208 and miR-499 have increased more prominent in patients with angina pectoris (AP) compared to patients with AMI. The values of AUC of the six-serum miRNAs as mentioned in the study were more prominent (0.830, 95% CI, 0.751-0.910) compared to the AUC of CTnT (0.768, 95% CI, 0.672-0.864) and the AUC of CK-MB (0.709; 95% CI, 0.606-0.812).
In a 2015 study conducted by Bialek et al., the level of miRNA-208a was elevated in patients with STEMI, especially during the admission time, and almost imperceptible in patients with CAD and the control group (p <0.001) [24]. The levels of miRNA-208a were heavily associated with the mass of CK-MB (p <0.05) and CTnI (p <0.05). The study also observed a notable elevation in miRNA-208a plasma level on the time of admission (time 0) in STEMI patients. The concentration of plasma miRNA-208a was elevated in the first three hours following the first symptoms and continued to elevate to 12 hours. The study noted that the concentrations of CK-MB and CTnI mass were beneath the cutoff value for MI during admission time. Both biomarkers elevated at a later time, having peaked at six hours following the admission, and remained continued to elevate at the time of observation for around 48 hours.
Table 2. Summary of clinical studies which investigated cardiac-specific microRNAs in acute coronary syndrome populations
No.
|
Author
|
miRNA Regulation
|
Source
|
Analysis Technique
|
Blood Sampling Timing
|
Results
|
Correlation with Biomarkers
|
1.
|
Wang et al. [7]
|
Upregulated: miR-1, miR-133a, miR-499, miR-208a
|
Plasma
|
qRT-PCR
|
4.8±3.5 hours
|
miRNA levels were markedly elevated than control (p< 0.01)
|
miRNAs were correlated with cTnI
|
2.
|
Adachi et al. [8]
|
Upregulated: miR-499
|
Plasma
|
qRT-PCR
|
48 hours
|
miR-499 values in the AM group were markedly elevated than other group (p<0.0001)
|
CK-MB
|
3.
|
Corsten et al. [9]
|
Upregulated: miR-208a, miR-499
|
Plasma
|
PCR
|
<12 hours
|
miR-208b and miR-499 were markedly elevated (p<0.005 and p<0.0005, respectively) in AMI compared to control
|
miR-208b and miR-499 significantly correlated with CTnT
|
4.
|
D’ Alessandra et al. [10]
|
Upregulated: miR-1, miR-133a/b,miR-499-5p; Downregulated: miR-122, miR-375
|
Plasma
|
qRT-PCR
|
517+309 minutes
|
miRNAs levels were markedly altered in AMI compared to control (p<0.01)
|
Correlated positively with CTnI p<0.01 compared to control
|
5.
|
Ai et al. [11]
|
Upregulated: miR-1
|
Plasma
|
PCR
|
Not given
|
miRNA levels were markedly elevated
|
Positive correlation with cardiac troponin
|
6.
|
Cheng et al. [12]
|
Upregulated: miR-1
|
Serum
|
qRT-PCR
|
8.5±3.82 hours
|
miR-1 was elevated in AMI patients than in control group (p< 0.05)
|
miR-1 &CK-MB were correlated positively (r=0.68; p<0.05)
|
7.
|
Wang et al. [6]
|
Upregulated: miR-133, miR-328
|
Whole blood
Plasma
|
RT-PCR
|
5.24 ± 1.38 hours, 20 hours, 7 days
|
miR-133 plasma levels in AMI patients were elevated compared to control (p=0.006). miR-328 plasma and whole blood levels of AMI patients were markedly elevated compared to control (p=0.033 and p<0.001)
|
Correlated with CTnI
|
8.
|
Widera et al. [13]
|
Upregulated: miR-1, miR-133a;
Insignificant: miR-133b, miR-208a, miR-208b, miR-499
|
Plasma
|
qRT-PCR
|
Not given
|
NSTEMI or STEMI patients presented with elevated levels of miR-1, miR133a, and miR-208b compared to unstable angina patients (p=0.001)
|
miR 133a (p<0.001), miR-1, miR-133a, miR-133b, and miR-208b were correlated independently with hsTnT levels (p<0.001)
|
9.
|
Kuwabara et al. [14]
|
Upregulated: miR-1, miR-133a
|
Serum
|
qRT-PCR
|
Not given
|
miR-1 p<0.0005, miR-133a p=0.001
|
miRNAs were correlated positively with cTnT
|
10.
|
Gidlof et al. [15]
|
Upregulated: miR-1, miR-133a, miR-208b, miR-499-5p
|
Plasma
|
RT-PCR
|
24, 48, and 72 hours
|
miR-1 (p<0.01); miR-133a (p<0.01); miR-208b (p<0.001); miR-499-5p (p<0.01) as compared to control
|
cTnT correlated positively with miR-208b (p=0.01, r 2=0.25); EF correlated negatively with miR-208b (p=0.01, r 2=0.32)
|
11.
|
Devaux et al. [16]
|
Upregulated: miR-208b, miR-499
|
Plasma
|
RT-PCR
|
Not given
|
miRNAs were more elevated in MI patients (p<0.001)
|
Correlation of miRNAs and CK-MB and CTnT were highly significant p<10¯9. miRNAs were inversely correlated with the EF
|
12.
|
Olivieri et al. [17]
|
Upregulated: miR-1, miR-21, miR-133a, miR-423-5p, miR-499-5p
|
Plasma
|
qRT-PCR
|
Not given
|
NSTEMI vs control p<0.05; NSTEMI vs CHF p<0.05; CHF vs CTR p<0.05
|
miR-499-5p and cTnT were correlated positively (p<0.001) in the total population and NSTEMI patients
|
13.
|
Orleman et al. [18]
|
Upregulated: miR-1, miR-208a, miR-499, miR-21, miR-146a
|
Serum
|
RT-PCR
|
Not given
|
Levels of circulating miRNAs were elevated in ACS patients. Circulating miR-21 and miR-146a levels markedly increased in ACS patients (p<0.001)
|
miRNA combined assay (miR-1, miR-499, and miR-21) was better than hs-CTnT
|
14.
|
Li et al. [19]
|
Upregulated: miR-1, miR-134, miR-186, miR-208, miR-223, miR-499
|
Serum
|
RT-PCR
|
Not given
|
Serum levels of the six miRNAs were increased in AMI than in control: miR-1, miR-223, and miR-499 (p<0.05); miR-134, miR-186, miR-208 (p<0.001)
|
Correlated to CKMB and CTnT
|
15.
|
Gidlof et al. [20]
|
Upregulated: miR-1, miR-208b, miR-499-5p
|
Plasma
|
qRT-PCR
|
Average time to sample: 38.4 hours
|
miR-1 was elevated (p<0.01), miR-133a (p<0.01), miR-208b (p <0.001), and miR-499-5p (p<0.01) compared to healthy controls
|
miR-208b and miR -499-5p were correlated strongly with TnT. The accuracy was inferior to Troponin T. miR-1 was weakly associated with TnT
|
16.
|
Li YQ et al. [21]
|
Upregulated: miR-1, miR-133a, miR-208b, miR-499
|
Plasma
|
qRT-PCR
|
Within 12 hours and day
|
All miRNAs were markedly elevated in AMI patients (p<0.001) than in healthy volunteers
|
miRNAs and cTnT were correlated positively. None of circulating miRNAs were better to cTnT for prompt diagnosis of AMI (p˃0.05)
|
17.
|
Chen et al. [22]
|
Upregulated: miR-499
|
Plasma
|
qRT-PCR
|
0, 12, 24, 72 hours and 7 days after AMI
|
miR-499 levels were markedly elevated immediately in AMI patients than in the UA and controls (p<0.01)
|
miR-499 correlated positively with CK-MB (p<0.01)
and cTnI (p<0.01)
|
18.
|
Zhao et al. [23]
|
Upregulated: miR-499
|
Plasma
|
qRT-PCR
|
3, 12, and 15 hours
|
miRNA-499 in AMI was markedly elevated than in controls (p<0.05)
|
The sensitivity and specificity of microRNA-499 in AMI diagnosis were still inferior than cTnI
|
19.
|
Bialek et al. [24]
|
Upregulated: miR-208a
|
Serum
|
qPCR
|
0, 3, 6, 12, 24, and 48 hours
|
miR-208a was elevated in STEMI patients (p<0.001)
|
miR-208a was related with CK-MB and CTnT mass
|
20.
|
Agiannito-poulos et al. [25]
|
Upregulated: miR-208b, miR-499
|
Plasma
|
PCR
|
Not given
|
miR-208b (p<0.0001), miR-499 (p<0.0001) compared to controls
|
All miRNAs were correlation with CTnT
|
Note: miR = microRNA; qRT = quantitative real time; PCR = polymerase chain reaction; RT-PCR = real time polymerase chain reaction; CTnI = cardiac troponin I; CTnT = cardiac troponin T; CK-MB = creatine kinase-muscle/brain; STEMI = ST elevation myocardial infarction; NSTEMI = non ST elevation myocardial infarction; hsTnT = high sensitive troponin T; AMI = acute myocardial infarction; UA = unstable angina; EF = ejection fraction; CHF = congestive heart failure.