Does mean platelet volume predict the location of the infarct-related artery in non-ST-segment elevation myocardial infarction patients?

• Infarct-related artery
• mean platelet volume
• myocardial infarction
• NSTEMI
• thrombus

To the editor

Mean platelet volume (MPV) is known to predict recurrent myocardial infarction and is related to an unfavorable prognosis in patients with acute coronary syndrome [1, 2]. Moreover, Liu et al. recently reported that MPV was an independent predictor of the presence of a culprit lesion in the left anterior descending (LAD) artery in non-ST-elevation myocardial infarction (NSTEMI) patients [3]. We read the article with great interest. Identification of an infarct-related artery is often challenging in the setting of NSTEMI as electrocardiogram findings become less reliable compared with that in the setting of ST-elevation myocardial infarction. NSTEMI resulting from a LAD artery lesion is associated with higher short- and long-term mortality rates [4]. Therefore, a simple blood test providing additional information to identify the infarct-related artery would be of great clinical value. In this context, we retrospectively analyzed the records of our diverse population of NSTEMI patients to assess the validity of previous findings and to assess the possible association between MPV and other angiographic findings.

Our retrospective analysis included 763 patients with NSTEMI, who had undergone coronary angiography from January 2013 to June 2014 at our facility. The present study complied with the Declaration of Helsinki. The institutional review board approved the study, and written informed consent was waived because of the retrospective study design. Myocardial infarction was diagnosed according to the European Society of Cardiology and American College of Cardiology criteria [5]. Inclusion criteria included 1) troponin value greater than the 99^th percentile reference value before cardiac catheterization; 2) either chest pain (or the anginal equivalent), or ischemic change on the electrocardiogram including horizontal or down-sloping ST-segment depression (≥0.05 mV) or T-wave inversion (≥0.1 mV) in two or more contiguous leads; and 3) absence of ST elevation and new left bundle branch block on the electrocardiogram. Exclusion criteria were 1) coronary catheterization more than five days after presentation (n = 143); 2) elective procedure (n = 22); 3) in-hospital onset (n = 13); 4) a history of coronary artery bypass grafting (CABG) (n = 61); 5) non-obstructive coronary artery disease (n = 100); 6) left main culprit lesion (n = 11) and multiple territory ischemia (n = 34); 7) other identifiable causes for troponin elevation (n = 64); and 8) insufficient data for analysis (n = 40).

Patients’ demographic data, risk factors, and clinical characteristics on admission, including hemodynamic parameters, were obtained. The Thrombolysis In Myocardial Infarction (TIMI) risk score was calculated, and patients were classified into three groups: low-risk (0–2), medium-risk (3–4), and high-risk groups (5–7). Peripheral venous blood samples were drawn in the emergency department at the time of presentation and placed in tubes containing dipotassium ethylenediaminetetraacetic acid. Blood samples were processed within one hour after sample collection in the majority of the cases. MPV was analyzed using a Beckman Coulter LH 750® (Beckman Coulter Inc., Brea, CA). Cardiac troponin I levels were measured using the second-generation VITROS® (Ortho-Clinical Diagnostics Inc., Raritan, NJ). The upper limit of normal for cardiac troponin I was 0.034 µg/l, which represented the 99th percentile reference value. Transthoracic echocardiography was performed during hospitalization, and left ventricular ejection fraction was calculated using either the Teichholz or biplane Simpson’s method.

All patients underwent cardiac catheterization within five days after presentation. An independent cardiologist blinded to the clinical data reviewed all coronary angiography findings, and the assessment was compared with the primary assessment by the treating cardiologist. In case of discrepancy in the assessments, a third investigator made the final decision. Obstructive coronary artery disease was defined as stenosis ≥ 70% (50% in the left main coronary artery). Revascularization procedures including percutaneous coronary intervention (PCI) and CABG were performed at the discretion of the treating physician. Angiographic findings including the location of the culprit lesion, number of diseased vessels, angiographic thrombus (TIMI grades 1–5), collateral circulation to the infarct-related artery (Rentrop grade 2–3), and revascularization procedures were recorded. The culprit lesion was determined based on the electrocardiogram, echocardiogram, and angiography findings including previous angiography findings, if available. Multiple territory ischemia was diagnosed when stenosis ≥ 90% in more than one territory of three major territories (LAD, left circumflex, and right coronary arteries) without a single identifiable culprit lesion. Patients were categorized into two groups: those with a LAD culprit lesion and those with a non-LAD culprit lesion. In-hospital major adverse cardiac events (MACE) including death, recurrent myocardial infarction, and target vessel revascularization along with in-hospital heart failure, in-hospital cardiogenic shock, and length of stay were recorded.

Normally distributed continuous variables are expressed as mean ± standard deviation. Non-normally distributed continuous variables are expressed as median [interquartile range]. Dichotomous variables are expressed as number (percentage). For continuous variables, we used Shapiro–Wilk analysis to check the normality of the distribution. Continuous variables were compared using either Student’s t-test or Wilcoxon rank-sum test, as appropriate. Dichotomous variables were compared using the Chi-squared test or Fisher’s exact test. Multivariate linear regression analysis was performed to assess the effect of a LAD culprit lesion on MPV. MPV was entered into the model after natural logarithmic transformation to achieve a normal distribution. Variables including age, sex, hypertension, diabetes, hyperlipidemia, smoking status and previous myocardial infarction were entered into the model. A two-sided p value < 0.05 was considered statistically significant. All statistical analyses were performed with R software, version 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria).

Of the 763 patients who met the inclusion criteria, 488 patients were excluded according to the pre-specified criteria. In all, 275 patients were included in the final analysis. Baseline characteristics are summarized in Table I. Among the 275 patients, 113 (41.1%) had a LAD culprit lesion and 162 (58.9%) had a non-LAD culprit lesion (78 and 84 patients had a culprit lesion in the left circumflex artery and right coronary artery, respectively). There was no significant difference in baseline characteristics and risk factors between the two groups. The median MPV of the overall study population was 8.5 fl [interquartile; 7.9–9.3]. The MPV was comparable between the two groups. Although MPV values were also compared in several subgroups, no significant difference was found between patients with a LAD culprit lesion and a non-LAD culprit lesion. Patients with a LAD culprit lesion had a significantly lower left ventricular ejection fraction. Patients with a LAD culprit lesion had a significantly lower rate of TIMI grade 0 flow, angiographic thrombus, and collateral circulation to the infarct-related artery. There was no difference in the rate of PCI, CABG, and in-hospital clinical outcomes between groups. In addition, MPV was compared in the several subgroups in relation to angiographic findings including culprit lesions in the right coronary or left circumflex artery, or TIMI grade 0 or 0–1, or angiographic thrombus (TIMI grade 1–5), angiographic thrombus (TIMI grade 4–5), or collateral circulation (Table II). There was no statistically significant difference in the MPV value in these subgroups. On multivariate linear regression analysis, a LAD culprit lesion was not significantly associated with MPV after adjusting for other variables (β coefficient −0.016, 95% confidence interval −0.046 to +0.015, p = 0.31).

Table I. Patient characteristics.

	LAD group (n = 113)	Non-LAD group (n = 162)	p Value
Age (years)	66 ± 12	65 ± 12	0.53
Men	67 (59)	108 (67)	0.21
Hypertension	89 (79)	112 (69)	0.08
Diabetes	45 (40)	53 (33)	0.23
Hyperlipidemia	64 (57)	89 (55)	0.78
Current smoking	24 (21)	49 (30)	0.1
Previous MI	18 (16)	27 (17)	0.87
Previous PCI	39 (35)	56 (35)	0.99
TIMI risk score			0.97
Low risk 0–2	20 (18)	30 (19)
Medium risk 3–4	63 (56)	91 (56)
High risk 5–7	30 (27)	41 (25)
Hemodynamic and laboratory data
Killip class >1 on admission	15 (13)	15 (9)	0.29
Systolic blood pressure (mmHg)	144 ± 25	145 ± 26	0.88
Heart rate (beats/min)	82 [72–94]	78 [69–90]	0.16
White blood cell (10^3/µl)	8.8 [7.1–10.6]	8.3 [6.5–10.1]	0.11
Hemoglobin (g/dl)	13.1 [11.8–14.5]	13.6 [12.3–14.3]	0.4
Platelet (10^3/µl)	227 [171–266]	200 [171–234]	0.06
Mean platelet volume (fl)	8.6 [7.9–9.2]	8.4 [7.9–9.5]	0.72
Peak troponin I (µg/l)	0.81 [0.13–4.3]	1.79 [0.24–8.7]	0.02
Echocardiography data
LVEF	55 [43–60]	60 [50–65]	0.02
Angiographic findings
Three-vessel disease	21 (19)	40 (25)	0.23
TIMI grade 0 flow	10 (9)	30 (19)	0.03
TIMI grade 0–1 flow	17 (15)	51 (31)	0.002
Thrombus (TIMI grade 1–5)	33 (29)	96 (59)	<0.001
Thrombus (TIMI grade 4–5)	14 (12)	42 (26)	0.006
Collateral (Rentrop grade 2–3)	11 (10)	36 (22)	0.007
In-hospital events
In-hospital PCI	82 (73)	127 (78)	0.27
In-hospital CABG	7 (6)	17 (10)	0.21
In-hospital MACE	2 (2)	3 (2)	1
In-hospital heart failure	17 (15)	16 (10)	0.19
Cardiogenic shock	3 (3)	3 (2)	0.69
Length of stay	4.2 [2.5–6.4]	4.0 [2.3–7.6]	0.82
Mean platelet volume in subgroups
Subgroup
Age ≥65 years (n = 147)	8.5 [7.9–9.1]	8.4 [7.9–9.5]	0.64
Age <64 years (n = 128)	8.8 [8.0–9.2]	8.4 [7.9–9.5]	0.99
Men (n = 175)	8.6 [7.8–9.1]	8.3 [7.9–9.4]	1.00
Women (n = 100)	8.6 [8.1–9.3]	8.8 [8.2–9.8]	0.35
No previous MI (n = 230)	8.6 [7.9–9.2]	8.4 [7.9–9.5]	0.83

Data are expressed as number (percentage), mean ± SD, or median [interquartile range].

MI, myocardial infarction; PCI, percutaneous coronary intervention; TIMI, thrombolysis in myocardial infarction; LVEF, left ventricular ejection fraction; CABG, coronary artery bypass grafting; MACE, major adverse cardiac events.

Table II. Mean platelet volume in relation to angiographic findings.

Variable	Numbers of present	Present vs. absent	p value
LAD culprit	113	8.6 [7.9–9.2] vs. 8.4 [7.9–9.5]	0.72
LCX culprit	78	8.4 [7.9–9.6] vs. 8.5 [7.9–9.2]	0.91
RCA culprit	84	8.4 [8.0–9.4] vs. 8.5 [7.9–9.3]	0.78
TIMI grade 0 flow	40	8.3 [7.9–9.2] vs. 8.6 [7.9–9.4]	0.34
TIMI grades 0–1 flow	68	8.4 [7.9–9.3] vs. 8.6 [8.0–9.3]	0.69
Thrombus (TIMI grade 1–5)	129	8.5 [7.9–9.2] vs. 8.5 [8.0–9.4]	0.4
Thrombus (TIMI grade 4–5)	56	8.3 [7.8–9.2] vs. 8.6 [8.0–9.4]	0.13
Collateral (Rentrop grade 2–3)	47	8.2 [7.7–9.2] vs. 8.6 [8.0–9.4]	0.12

Data are expressed as median [interquartile range].

LAD, left anterior descending artery; LCX, left circumflex artery; RCA, right coronary artery; TIMI, thrombolysis in myocardial infarction.

Our study showed that the MPV did not differ significantly between the patients with a LAD culprit lesion and non-LAD culprit lesion. We performed a multivariate analysis because various factors including hypertension, diabetes, hyperlipidemia, obesity, and smoking can potentially affect MPV [6]. On multivariate analysis, there was no significant association between a LAD culprit lesion and MPV after adjusting for baseline characteristics. Therefore, MPV had no predictive value in identifying the infarct-related artery in our cohort of patients. This result, however, was not consistent with the previous finding reported by Liu et al. that demonstrated that MPV was significantly smaller in the group with a LAD culprit lesion, and that a smaller MPV value was an independent predictor of the presence of a culprit lesion in the LAD artery. The exclusion criteria in the present study were similar to that of their study, and patients with multiple territory ischemia, left main culprit lesion, and non-obstructive coronary artery disease were excluded. There are several differences between the two studies. The incidence of heart failure in their study was 29%, which was higher than that in our study population (12%). In addition, the mortality rate in their study was 3.7%, which was also higher than that in our study population (1.5%). These differences suggest that their study population included patients at higher risk than those in our study. Although these differences may have affected the study results, the discrepancy between the results of the two studies may be because of the following factors, in the previous study: 1) among patients with a non-LAD culprit lesion, 15 had TIMI grade 0–1 flow, although 37 patients had a 100% occluded culprit vessel; 2) the data of the majority of the clinical variables in the two tables were not consistent (e.g. the number of previous myocardial infarctions was 23 in Table I, but was 7 in Table II among the same study population); and 3) the list of pre-hospitalization medications was exactly the same in the LAD culprit and in-hospital event groups. These probable errors in data collection and analysis might have affected the results.

Patients with a LAD culprit lesion had a lower rate of TIMI grade 0 flow than patients with a non-LAD culprit lesion. This finding was consistent with that in a previous report [7]. In addition, we found that the rate of angiographic thrombus was also lower in patients with a LAD culprit lesion. We thought that the higher rate of thrombus in patients with a non-LAD culprit lesion might lead to greater platelet activation, which would affect the MPV. In this context, we evaluated the association between MPV and angiographic thrombus, and TIMI flow grade. However, we did not find any significant difference in MPV value depending on these angiographic findings. This result was similar to a recent report conducted in patients with ST-elevation myocardial infarction, which showed that the MPV was comparable between patients with high-thrombus burden and those with a low-thrombus burden [8]. These results suggested that the MPV was not directly related with ongoing thrombus burden, though a high MPV has been shown to predict future thromboembolic events.

Our study has several limitations. First, our study was subjected to the usual limitations associated with a retrospective observational study including missing information and selection bias. It should be noted that we excluded patients with left main culprit lesions and those with more than two probable culprit lesions. In addition, we excluded patients who underwent angiography after five days of admission, since angiography performed later is less likely to represent the actual coronary flow at the time of the event. The large number of excluded patients may limit the generalizability of our study results. Second, the incidence of in-hospital MACE was small, and our study did not have enough statistical power to assess the prognostic value of MPV.

In conclusion, our study demonstrated that MPV does not have a predictive value in identifying culprit lesions in NSTEMI patients. This result, however, was not consistent with that in a previous study by Liu et al., and a further larger study is warranted to draw a definitive conclusion. In contrast, the role of MPV as a predictor of recurrent myocardial infarction and worse prognosis in patients with acute coronary syndrome has been established by accumulating evidence [1, 2].

Declaration of interest

We have no conflict of interest to declare.