ei(wpspvwdtlTr

ZCv

ScZ

0d

Factors Influencing Local and Systemic Levels of PlasmaMyeloperoxidase in ST-Segment Elevation Acute

Myocardial Infarction

Catriona J. Marshall, MBChBa, Mark Nallaratnam, MBChBa, Tessa Mocatta, BScb,David Smyth, MDa, Mark Richards, MD, PhDa, John M. Elliott, MBChB, PhDa,

James Blake, MBChBa, Christine C. Winterbourn, PhDb, Anthony J. Kettle, PhDb, andDougal R. McClean, MDa,*

Myeloperoxidase (MPO) is associated with risk in acute coronary syndromes. However, theprecise role it plays in ST-elevation myocardial infarction (STEMI) remains unclear. In thisstudy we tested the hypothesis that levels of MPO in plasma after a myocardial infarctionare affected by its ability to bind to the endothelium and there is local release of the enzymeat the culprit lesion. We measured plasma MPO in systemic circulation and throughout thecoronary circulation in patients with STEMI undergoing primary percutaneous coronaryintervention (PCI). MPO levels at the femoral artery were higher (p <0.001) in patientswith STEMI (n � 67, median 45 ng/ml, interquartile range 34 to 83) compared to controlpatients (n � 12, 25 ng/ml, 19 to 30) with chronic stable angina undergoing elective PCI.After administration of the anticoagulant bivalirudin in 13 patients with STEMI, plasmaMPO was increased only at the culprit coronary artery lesion before PCI (178 ng/ml, 91 to245) versus all other sites (femoral artery 86 ng/ml, 54 to 139, p � 0.019). Administrationof heparin caused a marked increase of plasma MPO. Even so, it was still possible to detectan increase of plasma MPO at culprit lesion in patients with STEMI (n � 54, 171 ng/ml,122 to 230) versus controls (n � 12, 136 ng/ml, 109 to 151, p <0.05) after heparin and beforePCI. MPO levels were higher at the culprit lesion in patients with STEMI who presentedearly and in those with restricted flow (p <0.05). In conclusion, our results demonstratethat, in addition to a systemic increase of MPO in patients presenting early with STEMI,levels of this leukocyte enzyme are increased at the culprit coronary lesion before

PCI. © 2010 Elsevier Inc. All rights reserved. (Am J Cardiol 2010;106:316–322)

cs

M

HrfSeeh�tctw

spacew

We hypothesized that in patients with acute ST-segmentlevation myocardial infarction (STEMI) who present early,n addition to a systemic effect, plasma myeloperoxidaseMPO) levels may be increased locally at the culprit lesionithin the occluded coronary artery. Blood was samplederipherally before administration of anticoagulant and atites throughout the coronary circulation during primaryercutaneous coronary intervention (PCI). Initially we in-estigated levels of plasma MPO in a subgroup of patientsith STEMI using the direct thrombin inhibitor bivaliru-in, which does not affect release of MPO from the endo-helium.1 We then assessed levels of plasma MPO in aarger group of patients with STEMI who received heparin.o control for the effects of heparin, we compared the

esults in the STEMI group to a group of patients with

aCardiology Department, Christchurch Hospital, Christchurch, Newealand; and bFree Radical Research Group, University of Otago,hristchurch, New Zealand. Manuscript received December 7, 2009; re-ised manuscript received and accepted March 11, 2010.

Mr. Nallaratnam was supported by the New Zealand 2006 Cardiacociety/MSD Research Fellowship, Wellington, New Zealand. Ms. Mo-atta was supported in part by the Health Research Council of Newealand, Auckland, New Zealand.

*Corresponding author: Tel: 64-3-364-1122; fax: 64-3-364-1415.

sE-mail address: dougal.mcclean@cdhb.govt.nz (D.R. McClean).

002-9149/10/$ – see front matter © 2010 Elsevier Inc. All rights reserved.oi:10.1016/j.amjcard.2010.03.028

hronic stable angina (CSA) undergoing elective PCI, afterimilar doses of weight-adjusted heparin.

ethods

The study was approved by the New Zealand Ministry ofealth Upper South A regional ethics committee (ethics

eference URA/05/08/097). All patients gave written in-ormed consent before enrollment. Patients with acuteTEMI presenting to Christchurch Hospital were consid-red eligible if symptoms suggesting acute myocardial isch-mia lasted �30 minutes, onset of symptoms was �12ours previously, ST-segment elevation was �0.1 mV in2 leads on electrocardiogram, and had no contraindica-

ions to primary PCI. Exclusion criteria were left mainoronary artery disease, stent thrombosis, rescue PCI afterhrombolysis, or cardiogenic shock. Patients were pretreatedith aspirin 300 mg and clopidogrel 600 mg.Blood was drawn peripherally from the femoral artery

heath before administration of anticoagulant. Further sam-ling was carried out before angioplasty and stenting in thescending aorta at the coronary ostium through a guidingatheter and from the coronary sinus using a separate cath-ter introduced through the femoral vein. The culprit lesionithin the coronary artery was sampled after careful pas-

age of the guidewire across the lesion and use of a low-

www.ajconline.org

pBosr((ncwstPwaalacs

ccwIicd

ipgpppm

dpctwtltibPcsam

awntl3MsHtwlc

rsiapuslaes

R

mfg

h4wdiwgbfl3(

Fcgpn

317Coronary Artery Disease/Myeloperoxidase in ST-Elevation Myocardial Infarction

rofile sampling catheter (multifunctional probing catheter;oston Scientific Corp., Natick, Massachusetts) at the sitef occlusion, before further mechanical intervention. Beforeampling, baseline flow in the culprit coronary artery wasecorded using Thrombolysis In Myocardial InfarctionTIMI) grade flow, with patients grouped into TIMI 0 to 1absent or little) flow or TIMI 2 to 3 (almost normal orormal) flow.2 Figure 1 depicts the multifunctional probingatheter sampling blood from the culprit coronary lesionithin the coronary artery. After the culprit lesion was

ampled, standard primary PCI using aspiration thrombec-omy and coronary stenting was performed. After successfulCI, further blood samples were drawn at the culprit lesionith the same sampling catheter, coronary sinus, and aorta

t the coronary ostium. A follow-up sample was drawn fromforearm vein 24 hours after PCI. All samples were col-

ected into tubes containing ethylenediaminetetra-aceticcid and placed on ice immediately. Whole blood wasentrifuged at 1,200 g at 4°C for 10 minutes. Plasma waseparated and stored at �80°C for subsequent analysis.

Twelve patients with Canadian Cardiovascular Societylass II to III angina symptoms and documented stableoronary disease undergoing elective single-vessel PCIere studied as a control. All patients had a plasma troponinlevel �0.01 before PCI. The blood sampling protocol was

dentical to that for the STEMI group. Patients were ex-luded if they had active infection, acute inflammatoryisease, malignancy, or chronic kidney disease.

Plasma MPO was measured by sandwich enzyme-linkedmmunosorbent assay and results expressed in nanogramser milliliter. Based on a molecular weight of 145,000/mol for MPO, these values can be converted to picomoleser liter by multiplying them by a factor of 7. Control andatient samples were included on each plate.3 Plasma sam-les were diluted 1:10 and MPO was captured using a

igure 1. Illustration of sampling technique used to obtain blood from theulprit coronary lesion. (A) An occlusive culprit coronary lesion with TIMIrade 0 flow before aspiration thrombectomy and stenting. The fine sam-ling catheter minimizes plaque or thrombus disruption. (B) Culprit coro-ary lesion with nonocclusive thrombus and TIMI grade 3 flow.

onoclonal antibody (Abcam, Cambridge, United King- w

om). It was detected using a rabbit polyclonal antibodyroduced in house, and goat anti-rabbit immunoglobulin Gontaining a biotin conjugate, which bound avidin coupledo alkaline phosphatase. When 5 different plasma samplesere spiked with purified MPO 60 ng/ml, 99 � 6% of the

otal was detected. Detection of MPO in plasma by enzyme-inked immunosorbent assay is not affected by administra-ion of heparin to patients because previously we found thatn a separate group of patients that there was no associationetween MPO levels in plasma and heparin treatment.3

lasma leukocyte elastase was measured using a commer-ial kit (Immunodiagnostik, Bensheim, Germany). Plasmaamples were diluted 1:100 with assay buffer then bound,nd unbound elastase was measured by enzyme-linked im-unosorbent assay.Whole blood was obtained from healthy controls (n � 5)

nd treated with heparin or bivalirudin in vitro to determinehether either anticoagulant caused release of MPO fromeutrophils. Concentrations of anticoagulant varied overheir pharmacologic ranges (heparin 0 to 15 IU/ml, biva-irudin 0 to 0.1 mg/ml). After 30 minutes of incubation at7°C, whole blood was centrifuged, plasma extracted, andPO assayed using the enzyme-linked immunosorbent as-

ay described earlier. Neutrophils were isolated by Ficoll-ypaque centrifugation, dextran sedimentation, and hypo-

onic lysis of red blood cells as previously described.4 Theseere also treated with the anticoagulants as described ear-

ier. After 30-minute incubation, cells were pelleted byentrifugation and supernatants analyzed for MPO.

Data were analyzed using nonparametric statistics andeported as median and interquartile range unless otherwisetated. Mann-Whitney rank sum test was used for compar-ng 2 groups. Friedman repeated measures analysis of vari-nce on ranks was used to determine differences for multi-le comparisons. Student-Newman-Keuls method was thensed to identify groups that differed from each other. Pear-on chi-square test was used for categorical variables. Re-ation between demographic variables and plasma MPO wasssessed using forward stepwise linear regression. Differ-nces with a p value �0.05 were accepted as statisticallyignificant.

esults

Sixty-seven patients with acute STEMI treated by pri-ary PCI received anticoagulant after sampling from the

emoral artery sheath. Baseline demographics of the 3roups are listed in Table 1.

Femoral artery plasma MPO, before anticoagulants, wasigher (p �0.001) in patients with STEMI (n � 67, median5 ng/ml, interquartile range 34 to 83) compared to controlith CSA (n � 12, 25 ng/ml, 19 to 30; Figure 2). Toetermine whether femoral artery plasma MPO levels werenfluenced by baseline flow in the culprit coronary artery,e categorized patients with STEMI into those with TIMIrade 0 to 1 flow and those with TIMI grade 2 to 3 flowefore PCI. Patients with STEMI and TIMI grade 0 to 1ow had higher (p �0.05) plasma MPO (n � 42, 50 ng/ml,7 to 87) compared to those with TIMI grade 2 to 3 flown � 11, 37 ng/ml, 24 to 48) (Figure 2). Furthermore, patients

ith STEMI who presented within 0 minute to 90 minutes

fpw

aSaw

ccftHcs

TCw

V

MADHHSPSA�BC

T

B

FH

Fim(i(

318 The American Journal of Cardiology (www.ajconline.org)

rom onset of symptoms of chest pain had higher levels oflasma MPO (p � 0.002) compared to those presentingithin 90 to 180 minutes (Figure 2).Bivalirudin was given after sampling from the femoral

rtery. Before bivalirudin, this subgroup of patients withTEMI had increased plasma MPO (p �0.01) at the femoralrtery (n � 13, 86 ng/ml, 54 to 140) compared to controls

able 1linical, demographic, and angiographic characteristics of patients with Sith ST-segment elevation myocardial infarction and patients with electiv

ariable STEMBival(n �

en 10 (7ge (years) (age range) 63 (4iabetes mellitus 3 (2ypertension 8 (6ypercholesterolemia‡ 9 (6moker 9 (6receding angina 5 (3tatin therapy 1 (8ngiotensin-converting enzyme inhibitor therapy 3 (2-blocker therapy 3 (2aseline leukocyte count 109/L, mean � SD 11.35 �ulprit arteryLeft anterior descending coronary artery 6 (4ime to presentation (minutes)0–90 4 (390–180 2 (1�180 7 (5aseline Thrombolysis In Myocardial Infarction grade flow0–1 13 (12–3 0inal Thrombolysis In Myocardial Infarction grade 2–3 flow 13 (1eparin (IU/kg), median (interquartile range) N

Percentages are of total unless otherwise stated.* Patients with STEMI treated with bivalirudin compared to those treat† Patients with STEMI given heparin compared to those with elective P‡ Total cholesterol �200 mg/dl or treated with medication.

igure 2. Levels of MPO in plasma from the femoral artery in patients witn arterial plasma from patients with STEMI were compared to those froeasured by enzyme-linked immunosorbent assay. Data are presented as

whiskers). Subgroup analysis of patients with STEMI was dependent onntervention and (C) time from onset of symptoms in patients with acute Sn � 8) or 90 to 180 minutes (n � 21) before blood sampling.

ith CSA (n � 12, 25 ng/ml, 19 to 29). This result was p

onsistent with their baseline TIMI grade 0 to 1 culpritoronary artery flow. Levels of MPO in plasma sampledrom the aorta and coronary sinus before PCI were similaro those in plasma from the femoral artery (Figure 3).owever, plasma MPO levels at the site of the culprit

oronary lesion (n � 13, 178 ng/ml, 91 to 245) wereignificantly increased (p � 0.019) compared to those in

ent elevation myocardial infarction treated with bivalirudin and patientstaneous coronary intervention treated with heparin

p Value* STEMI anHeparin

Elective PCI andHeparin

p Value†

(n � 54) (n � 12)

NS 35 (65%) 8 (67%) NSNS 61 (36–88) 68 (49–79) 0.11NS 15 (28%) 4 (33%) NS0.07 21 (39%) 11 (92%) 0.001NS 26 (48%) 11 (92%) 0.009NS 34 (63%) 4 (33%) 0.14NS 14 (26%) 12 (100%) �0.001NS 7 (13%) 10 (83%) �0.0010.18 5 (9%) 8 (67%) �0.001NS 8 (15%) 9 (75%) �0.001NS 10.53 � 4.51 6.28 � 0.90 0.01

NS 23 (43%) 10 (83%) 0.006

0.06 4 (7%)NS 26 (48%)NS 24 (44%)

0.03 39 (72%)15 (28%)

NS 52 (96%) 12 (100%) NS85.3 (77–89) 83.3 (83–86) NS

heparin.trol group) given heparin.

I and effect of baseline flow in a culprit coronary artery. (A) MPO levelsols with elective PCI before administration of anticoagulants. MPO was

and 25th and 75th percentiles (box plots) and 5th and 95th percentilesseline TIMI grade flow in the culprit coronary artery before mechanicalnd TIMI grade 0 to 1 flow who presented with chest pain 0 to 90 minutes

T-segme percu

I andirudin

13)

6%)2–85)3%)2%)9%)9%)8%)%)3%)3%)

4.56

6%)

1%)5%)4%)

00%)

00%)A

ed withCI (con

h STEMm contrmedian(B) ba

TEMI a

lasma from the femoral artery (n � 13, 86 ng/ml, 54 to

1dw

pthc

essfia

ubmpwwf

hrdliciHM

cwCdtc

ah1(oSc

iMwbf

Fwiwf others

FaiPamSNts

319Coronary Artery Disease/Myeloperoxidase in ST-Elevation Myocardial Infarction

39) and all other sites (p �0.05; Figure 3). This resultemonstrates that levels of plasma MPO at the culprit lesionere higher before PCI than after the intervention.Leukocyte elastase (LE), like MPO, is derived from the

rimary granules of neutrophils.5 Its plasma levels mirroredhose of MPO in patients receiving bivalirudin and wereigher at the culprit coronary lesion before PCI (p � 0.014)ompared to all other sampling sites (Figure 3).

In subgroups of patients with STEMI and those withlective PCI, heparin was administered after blood wasampled from the femoral artery or aorta. There was noignificant difference between levels of MPO in plasmarom the femoral artery and aorta in patients with STEMI orn control patients with elective PCI when heparin was

igure 3. Blood was collected from the femoral artery (FA) before adminisas collected from the aorta (Ao), coronary sinus (CS), and culprit coronar

n the order shown. (A) MPO and (B) neutrophil elastase were measured inere made between different sampling sites using Friedman repeated mea

or post hoc analysis to identify sites that were significantly different from

igure 4. Blood was sampled from the femoral artery and aorta beforedministration of heparin and from the aorta after administration of heparinn 9 patients with STEMI and 6 controls with CSA undergoing electiveCI. MPO levels were measured by enzyme-linked immunosorbent assaynd comparisons between different sampling sites were made using Fried-an repeated measures of analysis of variance on ranks for patients withTEMI (p �0.001) and controls with elective PCI (p �0.008). Student-ewman-Keuls method was used for post hoc analysis, which showed that

he plus-heparin groups were significantly different from the femoral arteryamples (p �0.05). Abbreviations as in Figure 3.

dministered after blood was sampled from the aorta (Fig- h

re 4). However, when heparin was administered afterlood was sampled from the femoral artery, there was aarked and significant increase in the level of MPO in

lasma from the aorta in patients with STEMI and controlsith elective PCI (p �0.05). These results are consistentith previous studies showing that heparin mobilizes MPO

rom the endothelium.1

In addition to mobilizing MPO from the endothelium,eparin may affect neutrophils directly and cause them toelease their intracellular stores of MPO.6 Therefore, weetermined whether anticoagulants could promote the re-ease of MPO from neutrophils in whole blood and fromsolated cells. Addition of bivalirudin or heparin at pharma-ologic doses to whole blood in vitro did not cause anncrease in levels of MPO detected in plasma (Figure 5).owever, heparin but not bivalirudin caused release ofPO from isolated neutrophils (Figure 5).Before PCI, MPO levels in plasma from the culprit

oronary artery lesion were higher (p �0.05) in patientsith STEMI (171 ng/ml, 121 to 230) than in controls withSA (136 ng/ml, 109 to 151) who received comparableoses of weight-adjusted heparin (Figure 6). However, inhese patients levels of MPO were similar in the aorta andoronary sinus before PCI.

Patients with STEMI who received heparin and hadbsent or little flow (TIMI grade 0 to 1) had significantlyigher (p � 0.015) plasma MPO levels (n � 39, 192 ng/ml,36 to 241) at the culprit lesion than those with brisk flowTIMI grade 2 to 3, n � 15, 136 ng/ml, 97 to 158). Levelsf MPO at the culprit coronary lesion in patients withTEMI and TIMI grade 0 to 1 flow were also higher than inontrols with CSA and elective PCI (p � 0.012; Figure 6).

Despite our finding that heparin increased levels of MPOn plasma, there was no significant difference in plasma

PO at the culprit coronary lesion before PCI in patientsith STEMI treated with heparin (171 ng/ml, 121 to 230) orivalirudin (178 ng/ml, 91 to 245). This finding remainedor patients with TIMI grade 0 to 1 flow who received

of bivalirudin. After administration of anticoagulant but before PCI, bloodlesion (CA). After PCI (arrow), blood was collected from these locationsfrom 13 patients by enzyme-linked immunosorbent assays. Comparisonsanalysis of variance on ranks. Student-Newman-Keuls method was used(�p �0.05).

trationy arteryplasma

sures of

eparin (n � 39, 192 ng/ml, 136 to 241) or bivalirudin (n �

1Mlc

ToctF1ib

0s

pSc0ttPwc

Fcc

Ftaasa1ws

320 The American Journal of Cardiology (www.ajconline.org)

3, 178 ng/ml, 91 to 245). At 24 hours after PCI, plasmaPO in patients with STEMI had decreased to baseline

evels but remained significantly higher (p � 0.003) than inontrols with elective PCI.

Patients with acute STEMI treated with heparin who hadIMI grade 0 to 1 flow and presented within 0 to 90 minutesf chest pain had higher plasma MPO (p � 0.029) at theulprit coronary lesion than those who presented within 90o 180 minutes with TIMI grade 0 to 1 flow (Figure 6).urthermore, in patients with STEMI and TIMI grade 0 toflow, plasma MPO at the culprit lesion before PCI was

ndependently predicted by preceding angina (p � 0.03),

igure 5. Effects of (A) bivalirudin or (B) heparin on release of MPO fromoncentrations of each anticoagulant. MPO was measured by enzyme-linollected from different patients.

igure 6. (A) MPO was measured in plasma by enzyme-linked immunoo 12 patients with CSA and elective PCI (clear box plots). Blood wdministration of anticoagulant but before PCI, blood was collected frofter PCI (arrow), blood was collected from these sites and then fromites between patients with STEMI and controls using Mann-Whitney rat the culprit coronary artery lesion with (B) TIMI flows of grade 0 to2, all with TIMI grade 2 to 3 flow) and (C) time from onset of symptoith chest pain 0 to 90 minutes (n � 5) or 90 to 180 minutes (n � 15)

ample (�p �0.05). Abbreviations as in Figure 3.

ut was not predicted by age (p � 0.71), diabetes (p � f

.44), gender (p � 0.67), hypertension (p � 0.15), ormoking (p � 0.78).

Mechanical intervention with stenting had no effect onlasma MPO at the culprit coronary lesion after PCI in theTEMI or elective control group (Figure 6). After PCI,ulprit lesion plasma MPO was not significantly higher (p �.05) in patients with STEMI treated with heparin comparedo controls with elective PCI. All patients had TIMI grade 2o 3 flow at the time of coronary artery sampling after PCI.lasma MPO levels at the aorta and coronary sinus after PCIere higher in patients with STEMI (p �0.05) than in

ontrols with CSA and elective PCI (Figure 6). These dif-

d neutrophils (PMNs) or those in whole blood were determined at variousmunosorbent assay. Data are means � SDs of 5 experiments on blood

t assays from 54 patients with STEMI (gray box plots) and comparedcted from the femoral artery before administration of heparin. Afteraorta, coronary sinus, and culprit coronary artery lesion. Immediatelyheral vein at 24 hours. Comparisons were made at different samplingtest (�p �0.05). Subgroup analysis was determined based on sampling39) or 2 to 3 (n � 14) versus controls undergoing elective PCI (n �

patients with acute STEMI and TIMI grade 0 to 1 flow who presentedblood sampling. All patients received heparin after the femoral artery

isolateked im

sorbenas collem thea peripnk-sum1 (n �ms inbefore

erences were not apparent before PCI.

D

tmiptttaostM

iCiogmticpMa

dgcirafniwwwiMiPlcaa

Mbcaa

sitta

ppsrle

dehtdhpptiThcapw

ctwvPh

uMmrmstcms

321Coronary Artery Disease/Myeloperoxidase in ST-Elevation Myocardial Infarction

iscussion

Numerous recent studies have indicated a role for neu-rophils in the early pathology of MI and that MPO is aarker or contributor to subsequent adverse effects includ-

ng death.7–10 The findings in this study support previousroposals that neutrophils are activated during acute MI andhat this leads to their release of MPO throughout the sys-emic circulation.7,8 Furthermore, it reinforces the concepthat MPO should be a useful marker of neutrophil activationnd that its levels may be associated with various clinicalutcomes after an MI. Several of our findings, however,uggest that measurement of MPO in plasma may underes-imate the true extent of neutrophil activation and the impact

PO itself may have on the subsequent pathology of MI.Although the level of MPO in peripheral circulation was

ncreased in patients with STEMI compared to patients withSA, the degree of increase was influenced by baseline flow

n the culprit coronary artery and the time of sampling afternset of chest pain. These findings suggest that there is areater degree of neutrophil degranulation when flow isarkedly restricted and that MPO is rapidly cleared from

he circulation. Clearance of MPO most likely results fromts adherence to the endothelium, as discussed later. Aonsequence of these findings is that future studies aimed atrobing associations between MPO and clinical outcomes inI should stratify patients based on time of presentation

nd TIMI grade flow.In addition to the widespread activation of neutrophils

uring MI, our results indicate that there may be an exag-erated activation of these inflammatory cells at the culpritoronary artery lesion. This was evident from the significantncrease of MPO and elastase at this site only in patientseceiving bivalirudin as an anticoagulant. These 2 enzymesre contained in azurophilic granules of neutrophils.5 There-ore, their concomitant increase supports the activation ofeutrophils. There was also a significant, although small,ncrease of MPO at the culprit coronary lesion in patientsho were given heparin. Plasma MPO levels at this siteere unaffected by type of anticoagulant administered,hich suggests that neutrophil degranulation was the dom-

nant source of MPO. These results imply that levels ofPO at the culprit coronary lesion are potentially better

ndexes of inflammatory sequels than peripheral values.resence of active MPO at this site has important physio-

ogic implications because it has the potential to produceytotoxic hypochlorous acid, which will chlorinate proteinsnd lipids. It will also consume nitric oxide catalytically,ffecting endothelial function and vasoconstriction.11–13

In accord with our data for peripheral blood, increase ofPO at the culprit coronary artery lesion was influenced by

aseline coronary flow and time of sampling from onset ofhest pain. Hence, measurement of MPO at this site shouldlso take these variables into account when assessing thessociation of this enzyme with clinical outcomes.

Symptoms of preceding angina were independently as-ociated with higher culprit lesion MPO levels. Ongoingnflammation, plaque instability with rupture, and intermit-ent transient occlusion and recanalization before occlusivehrombus may involve progressive neutrophil recruitment

nd activation with release of MPO. Alternatively, neutro-

hil degranulation at time of plaque vulnerability may de-osit MPO within the thin fibrous cap of the plaque. Onceequestered within the plaque MPO may precipitate itsupture by producing hypochlorous acid, which activatesatent matrix metalloproteinases and promotes apoptosis ofndothelial cells.14,15

In agreement with the early work from Baldus et al,1 weemonstrated that heparin causes a marked increase in lev-ls of MPO in plasma. Recently, this same group found thateparin caused an increase of peripheral MPO only in pa-ients with CSA and not those with acute coronary syn-romes.16 Interestingly, our results showed that the effect ofeparin was more marked in patients with elective PCI,resumably because their baseline levels were lower than inatients with STEMI. Negatively charged heparin is likelyo compete with negatively charged proteoglycans for pos-tively charged MPO and displace it from the endothelium.his has not been conclusively demonstrated, however, andeparin may liberate MPO from plasma proteins such aseruloplasmin and apolipoprotein A1.17,18 It is unlikely toct by promoting degranulation of neutrophils, as pro-osed recently,6 because we found no evidence for this inhole blood.After PCI there was no indication that MPO increased. In

ontrast, for patients receiving bivalirudin, MPO levels ac-ually decreased. It is unlikely, therefore, that the proceduree used caused activation of neutrophils as proposed pre-iously. Previous studies showing increases in MPO afterCI with stable angina but not MI did not control foreparin administration between groups.19

Our findings with heparin clearly indicate that studiessing this anticoagulant in conjunction with measurement ofPO should be interpreted with caution. This is because theajority of MPO may have arisen from the endothelium

ather than activated neutrophils. In patients with CSA,edian MPO level increased by fivefold (Figure 4), which

uggests that 80% of the MPO measured after administra-ion of heparin was from the endothelium. How this MPOontributes to cardiovascular disease versus what is nor-ally measured in plasma is particularly intriguing and

hould be investigated further.

1. Baldus S, Rudolph V, Roiss M, Ito WD, Rudolph TK, Eiserich JP,Sydow K, Lau D, Szocs K, Klinke A, Kubala L, Berglund L, SchrepferS, Deuse T, Haddad M, Risius T, Klemm H, Reichenspurner HC,Meinertz T, Heitzer T. Heparins increase endothelial nitric oxidebioavailability by liberating vessel-immobilized myeloperoxidase.Circulation 2006;113:1871–1878.

2. TIMI Study Group. The Thrombolysis In Myocardial Infarction(TIMI) trial. Phase I findings. N Engl J Med 1985;312:932–936.

3. Mocatta TJ, Pilbrow AP, Cameron VA, Senthilmohan R, FramptonCM, Richards AM, Winterbourn CC. Plasma concentrations of my-eloperoxidase predict mortality after myocardial infarction. J Am CollCardiol 2007;49:1993–2000.

4. Böyum A. Isolation of mononuclear cells and granulocytes from hu-man blood. Isolation of mononuclear cells by one centrifugation; andof granulocytes by combined centrifugation and sedimentation at 1 g.Scand J Clin Lab Invest 1968;21:77–89.

5. Nauseef WM. How human neutrophils kill and degrade microbes: anintegrated view. Immunol Rev 2007;219:88–102.

6. Li G, Keenan AC, Young JC, Hall MJ, Pamuklar Z, Ohman EM,Steinhubl SR, Smyth SS. Effects of unfractionated heparin and gly-coprotein IIb/IIIa antagonists versus bivalirudin on myeloperoxidase

release from neutrophils. Arterioscler Thromb Vasc Biol 2007;27:1850–1856.

1

1

1

1

1

1

1

1

1

1

322 The American Journal of Cardiology (www.ajconline.org)

7. Naruko T, Ueda M, Haze K, van der Wal AC, van der Loos CM, ItohA, Komatsu R, Ikura Y, Ogami M, Shimada Y, Ehara S, YoshiyamaM, Takeuchi K, Yoshikawa J, Becker AE. Neutrophil infiltration ofculprit lesions in acute coronary syndromes. Circulation 2002;106:2894–2900.

8. Buffon A, Biasucci LM, Liuzzo G, D’Onofrio G, Crea F, Maseri A.Widespread coronary inflammation in unstable angina. N Engl J Med2002;347:5–12.

9. Dominguez-Rogriguez A, Samimi-Fard S, Abreu-Gonzalez P, Garcia-Gonzalez MJ, Kaski JC. Prognostic value of admission myeloperoxi-dase levels in patients with ST-segment elevation myocardial infarc-tion and cardiogenic shock. Am J Cardiol 2008;101:1537–1540.

0. Khan SQ, Kelly D, Quinn P, Davies JE, Ng LL. Myeloperoxidase aidsprognostication together with N-terminal pro-B-type natriuretic pep-tide in high-risk patients with acute ST elevation myocardial infarc-tion. Heart 2007;93:826–831.

1. Galijasevic S, Saed GM, Hazen SL, Abu-Soud HM. Myeloperoxidasemetabolizes thiocyanate in a reaction driven by nitric oxide. Biochem-istry 2006;45:1255–1262.

2. Eiserich JP, Baldus S, Brennan ML, Ma W, Zhang C, Tousson A,Castro L, Lusis AJ, Nauseef WM, White CR, Freeman BA. Myelo-peroxidase, a leukocyte-derived vascular NO oxidase. Science 2002;296:2391–2394.

3. Vita JA, Brennan M-L, Gokee N, Mann SA, Goormastic M, Shishebor

MH, Penn MS, Keaney JF, Hazen SL. Serum myeloperoxidase levels

independently predict endothelial dysfunction in humans. Circulation2004;110:1134–1139.

4. Baldus S, Eiserich JP, Mani A, Castro L, Figueroa M, Chumley P, MaW, Tousson A, White CR, Bullard DC, Brennan ML, Lusis AJ, MooreKP, Freeman BA. Endothelial transcytosis of myeloperoxidase confersspecificity to vascular ECM proteins as targets of tyrosine nitration.J Clin Invest 2001;108:1759–1770.

5. Hazen SL. Myeloperoxidase and plaque vulnerability. ArteriosclerThromb Vasc Biol 2004;24:1143–1146.

6. Goldmann BU, Rudolph V, Rudolph TK, Holle AK, Hillebrandt M,Meinertz T, Baldus S. Neutrophil activation precedes myocardial in-jury in patients with acute myocardial infarction. Free Radic Biol Med2009;47:79–83.

7. Griffin SV, Chapman PT, Lianos EA, Lockwood CM. The inhibitionof myeloperoxidase by ceruloplasmin can be reversed by anti-my-eloperoxidase antibodies. Kidney Int 1999;55:917–925.

8. Zheng L, Nukuna B, Brennan ML, Sun M, Goormastic M, Settle M,Schmitt D, Fu X, Thomson L, Fox PL, Ischiropoulos H, Smith JD,Kinter M, Hazen SL. Apolipoprotein A-I is a selective target formyeloperoxidase-catalyzed oxidation and functional impairment insubjects with cardiovascular disease. J Clin Invest 2004;114:529–541.

9. Aminian A, Boudjeltia KZ, Babar S, Van Antwerpen P, Lefebvre P,Crasset V, Leone A, Ducobu J, Friart A, Vanhaeverbeek M. Coronarystenting is associated with an acute increase in plasma myeloperoxi-dase in stable angina patients but not in patients with acute myocardial

infarction. Eur J Intern Med 2009;20:527–532.