BrHeartj 1994;71:135-140

Fatty-acid-binding protein as a plasma marker forthe estimation of myocardial infarct size inhumans

Jan F C Glatz, Appie H Kleine, Frans A van Nieuwenhoven, Wim T Hermens,Marja P van Dieijen-Visser, Ger J van der Vusse

AbstractBackground-There are substantialamounts of cytoplasmic heart-type fatty-acid-binding protein (FABP) (15 kDa) inmyocardial tissue. The rapid release ofFABP into plasma during ischaemiaindicates the possibility of using this pro-tein as a biochemical marker forischaemic myocardial miury.Objective-To study the completeness ofthe release ofFABP from damaged tissuein patients with acute myocardial infarc-tion (AMI) and the suitability of serialplasma FABP concentrations for estima-tion ofmyocardial infarct size.Methods-Immunochemically assayedFABP and enzymatically assayedcreatine kinase isoenzyme MB (CK-MB)and a-hydroxybutyrate dehydrogenase(HBDH) were determined serially inplasma samples from 49 patients withAMI who had been treated with throm-bolytic agents within six hours after theonset of AMI. Previously validatedcirculatory models and a value of2*6 h-' for the fractional clearance rate ofFABP from plasma were used to calcu-late cumulative protein release intoplasma.Results-Release ofFABP was completedearlier (24-36 h) after AMI than that ofCK-MB (50-70 h) and that of HBDH (>70 h). However, infarct size estimatedfrom the cumulative release of the pro-teins and expressed as gram equivalentsof healthy myocardium per litre ofplasma yielded a comparable value of 4-6for both FABP and the two enzymes.Conclusion-The data indicate thatFABP released from the heart after AMIis quantitatively recovered in plasma andthat FABP is a useful biochemicalplasma marker for the estimation ofmyocardial infarct size in humans.

(Br HeartJ' 1994;71:135-140)

Acute myocardial infarction (AMI) in humansis usually assessed or excluded by themeasurement in plasma of the activities ofcardiac enzymes, such as creatine kinaseisoenzyme MB (CK-MB) and a-hydroxybu-tyrate dehydrogenase (HBDH)1 2 or of theconcentrations of cardiac proteins such asmyoglobin3 4 and tropinin T.5 6 BecauseMyoglobin appears in substantial quantities inplasma within 2-3 h of the onset of AMI,

whereas other proteins take at least fourhours, it has been suggested that myoglobinmay be a biochemical marker that is especiallysuitable for the early assessment of AMI,78and may also allow reperfusion to be distin-guished from persistent occlusion after throm-bolytic therapy.910 In addition, both cardiacenzymes and myoglobin released from theheart after AMI are completely recovered inplasma.1' 12 Hence the cumulative release ofeach of these markers can be used to estimateinfarct size in gram equivalents of healthymyocardium per litre of plasma.

Recently, another cardiac protein, heart-type fatty-acid-binding protein (FABP), hasbeen proposed as an early plasma marker forAMI."'-5 This small (15 kDa) cytoplasmicprotein is abundant in cardiomyocytes and isthought to be involved in myocardial lipidhomoeostasis.16 17 We"8 and others'9 found thatFABP is released in substantial amounts fromhuman hearts after AMI and that, like myo-globin, plasma FABP concentrations areincreased considerably within three hours ofAMI and return to normal within 24 hours. Itis not yet known, however, whether all FABPis released from damaged tissue and whetherthis protein too can be used as biochemicalmarker to estimate the size of myocardialinfarcts.We compared the release of FABP, CK-

MB, and HBDH into plasma after AMI andestimated infarct size from the correspondingplasma curve areas.

Patients and methodsPATIENTS AND BLOOD SAMPLINGWe studied 50 patients (six women, 44 men;age 29-71) with chest pain and ST segmentelevation typical of AMI. Patients were eligi-ble for this study if they were admitted to thecoronary care unit of the hospital within sixhours after the first onset of infarct-relatedsymptoms. Patients were participating in apilot study of treatment with a new inhibitorof platelet function by selective blockade ofthromboxane A, synthetase (Ridogrel, JanssenPharmaceutics, Beerse, Belgium) and werealso treated with alteplase (BoehringerIngelheim, Germany) and heparin. Coronaryangiography, performed at 90 minutes afterstart of treatment, showed successful reperfu-sion in 40 patients. Subsequent percutaneoustransluminal coronary angioplasty (PTCA), inthe remaining cases also showed reperfusion.Ischaemia was located anteriorly in 13patients and inferioposteriorly in 37 patients.

Department ofPhysiology, UniversityofLimburg,Maastricht, TheNetherlandsJ F C GlatzA H KleineF A van NieuwenhovenG J van der VusseCardiovascularResearch InstituteMaastricht,Maastricht, TheNetherlandsW T HermensDepartment ofClinical Chemistry,De Wever Hospital,Heerlen, TheNetherlandsM P van Dieijen-VisserCorrespondence to:Dr Jan F C GlatzDepartment of Physiology,CARIM, University ofLimburg, P.O. Box 616,NL-6200 MD MaastrichtThe NetherlandsAccepted for publication1 1 October 1993

135

on February 10, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.71.2.135 on 1 February 1994. D

ownloaded from

Glatz, Kleine, van Nieuwenhoven, Hermens, van Dieijen-Visser, van der Vusse

None of the patients had cardioversion.Plasma samples were incomplete for onepatient who died: the results of the remaining49 patients are shown. Further details onpatient selection and clinical treatment aredescribed elsewhere.20

Blood samples were taken upon admissionto the hospital (2-8 (1-3) h, range 1 1-6 0 h)after the onset of symptoms mean (SD) for n =49) and 3, 6, 9, 12, 24, 36, 48, 72, and (for 35patients only) 96 h thereafter, resulting innine or 10 samples per patient. Samples werecollected in glass tubes containing dryheparin. After routine centrifugation plasmawas stored in several aliquots at - 20°C untilanalysis.

ANALYTICAL TECHNIQUESFABP was measured in plasma and tissuesamples by a sensitive non-competitiveenzyme-linked immunosorbent assay of theantigen capture type (sandwich ELISA) asdescribed elsewhere.'8 Samples were dilutedwith phosphate buffered saline (pH 7-4) con-taining 0-1% bovine serum albumin and0-05% Tween-20. The detection limit of theassay was 0 5 ,ug/l. The recovery (mean (SD))of purified human heart FABP added invarious quantities to control human plasmawas 94 (12)% (n = 11) and the interassaycoefficient of variation was 6-5%.The activities of creatine kinase isoenzyme

MB (CK-MB) and HBDH were measuredspectrophotometrically at 25°C in a centri-fugal analyser (Cobas Bio System, HoffmannLa Roche, Basel, Switzerland) with commer-cially available test kits. For CK-MB we usedan enzyme assay kit that is based onimmunoinhibition of the predominant M unitin creatine kinase (Boehringer Mannheim,Germany). For HBDH, which reflectsmainly the activity of lactate dehydrogenaseisoenzyme-1, we used an assay kit with 2-oxobutyrate as the substrate (BoehringerMannheim, Germany). Activities areexpressed in,umol substrate converted perminute (units) per litre of plasma.

PLASMA REFERENCE VALUESReference values (upper normal concentra-tion or activity) for FABP, CK-MB, andHBDH in plasma were estimated in non-haemolytic blood samples obtained from 72healthy blood donors. Mean (SD) plasmaFABP concentration was 9 (5) ,ug/l, CK-MBactivity 4 (3)U/1, and HBDH activity 90(35)U/l. The reference values (mean plasmaconcentration or activity plus twice the stan-dard deviation) were 19 ,ig/l for FABP, 10 U/lfor CK-MB, and 160 U/l for HBDH.

CALCULATION OF CUMULATIVE PROTEINRELEASECumulative release of cardiac proteins fromthe onset of AMI (t = 0) up to time t, Q(t),was calculated for a two-compartmentmodel2' as follows:

t t

Q(t) = C(t) +TER I exp [ERR(T-t)] C(T) dT +FCR I C(r) dT0 0

where the three terms are the actual proteinconcentration (or enzyme activity) in plasma,the extravascular concentration, and theamount of protein eliminated from plasma,respectively, each expressed per litre ofplasma. TER, ERR, and FCR are the frac-tional rate constants for transcapillary escape,extravascular return, and catabolism (elimina-tion) of protein, respectively.

For CK-MB the values used were: TER =0-014 h-', ERR = 0-018 h-', and FCR = 0 34h-'.' 21 For HBDH the values for TER andERR were equal to those of CK-MB, whereasFCRwas 0-015 h-'.2'

Calculation of the cumulative release ofFABP was hampered by the fact that its frac-tional clearance rate is not known. However,turnover studies in humans of other smallproteins such as myoglobin (17'2 kDa),222'lysozyme (14 kDa)24 and retinol binding pro-tein (21 kDa)25 all showed elimination fromplasma predominantly by rapid renal clear-ance with a half life of 10-20 min. Hencethe cumulative release of myoglobin intoplasma after AMI was calculated in a one-compartment (that is plasma volume)model'2 26 where the second term in the aboveformula was left out, so that the cumulativeprotein release equals the integrated plasmacurve, multiplied by FCR. Because the meanhalf life of myoglobin in AMI patients is 16min23 the FCR amounts to (ln 2)/(tI,2) =2-6 h-'. This approach was validated in a sepa-rate set of 10 patients with clinically con-firmed AMI in whom plasma samples weretaken every hour (fig 1). The ratio of theplasma concentrations of myoglobin andFABP remained constant throughout theperiod that plasma values were raised. Themean plasma myoglobin: FABP ratio differedless than 16% from the plasma reference ratioof 591. In addition a one-compartment modelfor circulating FABP was validated in dogs.27

For each patient measured plasma FABPconcentrations and enzyme activities werecorrected by subtraction of the normal steady-state values. For this we used the respectiveplasma reference values (give above) or theconcentration or activity measured in the firstplasma sample taken from the patient whenthis value was lower than the reference value.

1OrO-mLL

- 50

E0

0 3 6 9 12Time after first onset of AMI (h)

15

Figure 1 Plasma myoglobin: FABP ratio as a function oftime afterAMI in 10 patients from whom hourly plasmasamples were obtained (mean (SEM). AMI, acutemyocardial infarction; FABP, fatty-acid-binding protein.

136

on February 10, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.71.2.135 on 1 February 1994. D

ownloaded from

FABP as markerfor myocardial infarction

Inclusion of patients who presented up tosix hours after onset of pain introduced anerror into the calculation of the cumulativerelease of FABP because it ignored priorrelease. However, only one patient enteredthe study more than 4-8 h after the onset ofAMI, and in the absence of thrombolysis-induced reperfusion the release of cardiacproteins up to 4-8 h is relatively small.

MYOCARDIAL INFARCT SIZEThe cumulative release of each cardiac pro-tein per litre of plasma was divided by themyocardial content of the specific protein pergram of wet weight tissue, so that myocardialinjury could be expressed as gram equivalentsof healthy heart muscle per litre of plasma.The myocardial content of FABP wasmeasured in parts of intact human heart (leftventricular tissue) obtained from either theDe Wever Hospital in Heerlen or theAcademic Hospital Maastricht after necropsy(performed within 12 hours after death) onpatients who died from non-cardiac causes.FABP content, assayed by the same methodas used for plasma, was 0-56 (007)mg/g wetweight (mean (SD) for 17 individuals).Regional and transmural differences in the leftventricle were not significant (data notshown). Myocardial enzyme content, mea-sured under the same assay conditions asplasma in the present study, were 132 U/g forCK-MB28 and 123 U/g for HBDH.29

VALIDATION OF SAMPLING PROTOCOLWe validated the present sampling protocolfor calculation of the cumulative release ofFABP in the 10 patients (one woman, ninemen; age 45-75) referred to earlier, in whomhourly plasma samples were obtained up to 12hours after admission to hospital. (About 80%of total FABP release is completed at thistime). The cumulative protein release calcu-lated on the basis of samples obtained at zero,three, six, nine and 12 hours after admissionto hospital correlated significantly with therelease calculated on the basis of all 13samples (fig 2) (r = 0-97, p < 0-05), indicatingthat the less frequent sampling protocol willyield good estimates of cumulative proteinrelease.

STATISTICAL ANALYSISThe release curves for proteins into plasmaare presented as means (SEM). Statistical

Figure 2 Relationbetween the cumulativerelease over 14 hours(Q14) ofFABP intoplasma as cakulatedfromplasma concentrations from5 (3-hourly) or 13(hourly) blood samples.The line of identity is given(dotted line). Thecalculated regression line(solid line) isy = 0-96x+0 07. FABP, fatty-acid-binding protein.

CAn

_- 'a 4-

E co_ 3>m 20 ° 2-

2 3 4

Q14A(mg/1), hourly samples

analysis of differences (between groups) wasperformed with Student's t-test. The level ofsignificance was set at p < 0 05.

ResultsMean plasma concentration or activities of thethree proteins examined as a function of timefor 49 patients (fig 3) showed a large differ-ence between the plasma kinetics of FABPand those of CK-MB and HBDH. The peakplasma concentration FABP was reached 5-7(1I 4) h after AMI, whereas that of CK-MBwas reached 117 (4A4) h after AMI and thatof HBDH 28-2 (13-5) h after AMI (mean(SD), n = 49). Within 24 hours the plasmaconcentration of FABP had returned to nor-mal, whereas CK-MB took 50-70 hours andHBDH more than 70 hours (fig 3).

In one patient a recurrent myocardialinfarction developed soon (<10 h) after theinitial AMI. The appearance of this recurrentinfarction is reflected clearly in the plasmacurve for FABP but is less apparent from theCK-MB and HBDH plasma curves (fig 4).The cumulative release patterns of the

three proteins, expressed in gram equivalentsof tissue per litre of plasma, also show a differ-ence between FABP and CK-MB and HBDH

1000

C FABP0

< 750OOt500 - < CK-MB

0 500-C.'-~ ~ ~ ~ ~ HD

250

0 12 24 36 48 60 72Time after first onset of AMI (h)

Figure 3 Plasma concentration ofFABP (x 5) andplasma activities ofCK-MB (x 10) andHBDH as afunction of time afterAMI in forty-nine patients (mean(SEM). AMI, acute myocardial infarction; CK-MB,creatine kinase isoenzyme-MB; FABP, fatty-acid-bindingprotein; HBDH, hydroxybutyrate dehydrogenase.

' 1500 CK-MB0

c 1000 FABP

500 S.0

EC"-

0- -0 12 24 36 48 60 72Time after first onset of AMI (h)

Figure 4 Plasma concentration ofFABP (x 5) andplasma activities ofCK-MB ( x 10) and HBDH as afunction of time after initial AMI in a patient in whomAMI recurred. AMI, acute myocardial infarction; CK-MB, creatine kinase isoenzyme-MB; FABP, fatty acid-binding protein; HBDH, hydroxybutyrate dehydrogenase.

137

on February 10, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.71.2.135 on 1 February 1994. D

ownloaded from

Glatz, Kleine, van Nieuwenhoven, Hermens, van Dieijen-Visser, van der Vusse

;d 6 - * '

E 0-

12 24 36 48 60 72

Time after first onset of AMI (h)

(fig 5). Release of FABP seemed to occur andbe completed much earlier than that of eitherCK-MB and HBDH. For instance, 12 h afterAMI about 75% of the total amount of FABPthat was eventually released had appeared inplasma, but only 42% of CK-MB and HBDH(fig 5). Total release of FABP was virtuallycomplete 24-36 h after AMI, but that of CK-MB took about 50 h and that of HBDH more

than 70 h. Despite this kinetic difference, foreach of the three proteins the released totalquantities yielded comparable estimates of themean extent of injury when evaluated at 72 hafter the onset ofAMI (fig 5).

DiscussionThe rationale for using fatty-acid-binding pro-

tein (FABP) as a plasma marker for myocar-dial injury is based on this soluble proteinbeing present in the myocardium in largeamounts and its virtual confinement to thecytoplasmic space. FABP has also beendetected in the matrix of bovine heart mito-chondria,30"3 but total mitochondrial FABPrepresents less than 1% of the cellular con-tent.30 Earlier studies"5 18 19 showed that in

humans plasma FABP concentrationsincreased significantly with ischaemic myocar-dial injury and, hence, that FABP was a usefulqualitative index for the assessment or exclu-sion of acute myocardial infarction. In thepresent study we found that the release ofFABP from damaged tissue was essentiallycomplete within 36 hours after the onset ofAMI and that the cumulative release ofFABPinto plasma can be used to estimate myocar-dial infarct size.

PLASMA KINETICS OF FABPFor the calculation of cumulative release we

assumed that the plasma kinetics of FABPwere similar to those of myoglobin. Thisimplies that the elimination of these proteinsfrom plasma can be described with a one-

compartnent (that is, a plasma volume)model, because their plasma kinetics are dom-inated entirely by the high rate of proteinelimination from plasma so that extravasationcan be neglected.2' Even if there were signifi-cant extravasation of FABP, this would affectonly the shape of the release curve and, pro-vided that a full curve is recorded, not thecumulative release over time, which is thevariable of interest.

The conclusion that FABP, like myoglobin,is eliminated from plasma predominantly byrapid renal clearance is supported by thereported detection of FABP in urine samplescollected as early as two hours after onset ofchest pain.19

RECOVERY OF FABP IN PLASMA AFTERISCHAEMIC MYOCARDIAL DAMAGEThe total quantities of CK-MB and HBDHdepleted from dog heart after coronary occlu-sion equal the calculated release of these pro-teins into plasma.32 The recovery of FABP inplasma after ischaemic myocardial injury hasnot yet been measured in experimentalstudies. In view of the stability ofCK-MB andHBDH29 it is generally assumed that inhumans too these cardiac proteins are quanti-tatively recovered in plasma after AMI. Thefact that in the present study the estimate ofinfarct size (expressed in gram equivalents ofhealthy myocardium) based on cumulativerelease of FABP resemble estimates based on

cumulative release of CK-MB or HBDH sug-gests that the recovery in plasma of FABPdepleted from the heart is also essentiallycomplete.

Release of FABP was completed earlierthan that of the two enzymatic markers. Therelative delay in the appearance of theenzymes in plasma may relate either to a

slower release of enzymes from damagedmyocardial cells or a slower transport fromthe interstitial space to the vascular space bylymph drainage and by increased trans-endothelial transport in the diseased tissue.There is evidence of binding of cytoplasmiccreatine kinase to structural elements in heartmuscle.3 However, a delay in the interstitialor transendothelial transport of enzymesappears most likely, because experimentalstudies with isolated rat hearts subjected tolow-flow ischaemia and reperfusion showedno differences in the release pattern of variouscardiac proteins.'4 Because CK-MB (80 kDa)

and HBDH (130 kDa) are much larger thanFABP (15-0 kDa) it is tempting to suggestthat molecular size may well be a main deter-minant of the rate of protein transport fromthe interstitial space to the plasma space.

ESTIMATION OF INFARCT SIZEEstimates of infarct size 72 h after the onset ofAMI based on the various markers were notsignificantly different (fig 5). However, infarctsize calculated from the cumulative release ofFABP tended to be higher. This tendencymay have been caused by an underestimationof the myocardial content of FABP, but thepresently measured value (0 56 mg/g) was

assayed by the same method as used forplasma and is similar to that published byothers (06 mg/g.35) The difference may alsobe due to an overestimation of the fractionalclearance rate of FABP, which was taken toequal that of myoglobin. The difference in theisoelectric points of FABP (pI 5.1) andmyoglobin (pI 7.0) means that at physio-logical pH FABP is more negatively charged,causing FABP to be cleared less rapidly by

Figure 5 Cumulativerelease ofFABP, CK-MB,andHBDH in plasmaafterAMI in forty-ninepatients. Data areexpressed in gramequivalents of healthymyocardium per litre ofplasma (mean (SEM)).*Cumulative release ofFABP significantly higher(p < 005) than that ofboth CK-MB andHBDH. AMI, acutemyocardial infarction;CK-MB, creatine kinaseisoenzyme-MB; FABP,fatty-acid-binding protein;HBDH, hydroxybutyratedehydrogenase.

138

on February 10, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.71.2.135 on 1 February 1994. D

ownloaded from

FABP as markerfor myocardial infarction

the kidneys and hence to stay longer in thecirculation than myoglobin.'6 This would leadto a lower fractional clearance rate and in turnto lower values of calculated infarct size.

Alternatively, infarct sizes calculated fromthe cumulative release of CK-MB and HBDHin plasma might have been underestimatedbecause these enzymes were measured asactivity whereas FABP was measured by animmunochemical assay (that is, as proteinmass). Measurement of CK-MB in plasma asprotein mass was more reliable than measure-ment as activity.37

Despite these uncertainties we concludethat, provided frequent blood samples aretaken, FABP gives a clinically useful estimateof myocardial infarct size, because 72 h afterthe onset of AMI we found no significant dif-ferences between the estimates based on thevarious markers.

CLINICAL APPLICATIONAs a plasma marker for ischaemic myocardialinjury FABP shares several characteristicswith other cardiac proteins (myoglobin, tro-ponin T) and with cardiospecific enzymes(CK-MB, HBDH), but it also shows someunique differences that will enhance thedetection and evaluation of an acute myocar-dial infarction. Like myoglobin, the rapidappearance of FABP in plasma after tissuedamage permits the early assessment of exclu-sion of AMI'8 as well as the immunohisto-chemical confirmation of very recentmyocardial infarction.'8 This application isalso enhanced by the relatively rapid elimina-tion of FABP from plasma because this keepsthe steady state plasma concentration of thisprotein at a low level. A further example ofthis concept is the fact that the ratio of thecytoplasmic to the vascular concentration ofFABP (amounting to about 2 x 105) is oneorder of magnitude higher than that of any ofthe cardiac enzymes (CK-MB, 6 x 1 04;HBDH, 5 x 103). The consequence of such ahigh ratio is that the release of only minuteamounts of FABP from myocardial cells withsignificantly raise its plasma concentration,making FABP a diagnostic marker with highsensitivity. Furthermore, the rapid clearanceof FABP also allows recurrent infarctions tobe more easily identified.

Despite the occurrence of various types ofFABP, with some types found solely in a singletissue-such as intestinal FABP in interstitialepithelial cells39 40-heart-type FABP is foundnot only in cardiac muscle but, like myoglo-bin, also in striated skeletal muscle.41 42However, the myoglobin:FABP ratio ofhuman skeletal muscle is 32-70 (dependingon the muscle fibre type composition) andthat of heart 4 9 (1.2),42 making FABP morecardiospecific than myoglobin. Because FABPand myoglobin show similar patterns ofrelease from tissue and of elimination fromplasma, the plasma myoglobin:FABP ratiowill allow discrimination between cardiac andskeletal muscle damage.42

Another condition that may cause erro-

neous values of plasma FABP (and of myoglo-

bin) is a decreased glomerular filtration rate(GFR). In such cases the FABP released fromdamaged myocardial tissue will accumulate inthe plasma,'8 thus leading to a possible overes-timation of infarct size.The rapid release of FABP into plasma

makes possible a reliable measure of myocar-dial infarct size within 24 hours of AMI.However, this application must meet two con-ditions. First, sufficient plasma samples haveto be obtained during the first day of hospitaladmission. Secondly, the use of the FABPplasma concentration as an early diagnostictool of AMI requires a fast assay system. Thisis not yet available (the sandwich ELISA usedin the present study takes a few hours tocomplete), but recent developments withimmunodiagnostic tests43 should lead to a sen-sitive assay for plasma FABP that will givequantitative data within minutes. In addition,immunochemical detection of biochemicalplasma markers is generally considered tohave the advantage of being free from prob-lems inherent to the measurement of enzymeactivity of proteins.

We thank Prof Dr M L Simoons, Department of Cardiologyand Thoraxcenter, University Hospital Dijkzigt, Rotterdam,The Netherlands, for his interest and valuable advice. Thisstudy was supported by StiPT, Executive Agency forTechnology Policy (MTR 88002). J F C G is an EstablishedInvestigator of the Netherlands Heart Foundation.

1 Willems GM, Muijtjens AMM, Lambi FHH, HermensWTh. Estimation of circulatory parameters in patientswith acute myocardial infarction. Significance for calcu-lation of enzymatic infarct size. Cardiovasc Res 1979;13:578-87.

2 Van der Veen FH, Visser R, Willems GM, Kop-KlaassenB, Hermens WTh. Myocardial enzyme depletion ininfarcted human hearts: infarct size and equivalent tissuemass. Cardiovasc Res 1988;22:611-9.

3 Stone MJ, Waterman MR, Harimoto D, Murray G,Wilson N, Platt MR, et al. Serum myoglobin level asdiagnostic test in patients with acute myocardial infarc-tion. BrHeartJ 1977;39:375-80.

4 McComb JM, McMaster EA, MacKenzie G, Adgey AAJ.Myoglobin and creatine kinase in acute myocardialinfarction. BrHeartj 1984;51:189-94.

5 Katus HA, Remppis A, Looser S, Hallermeier K,Scheffold T, Kubler W. Enzyme-linked immuno assay ofcardiac troponin T for the detection of acute myocardialinfarction in patients. Jf Mol Cell Cardiol 1989;21:1349-53.

6 Katus HA, Looser S, Hallermeier K, Remppis A,Scheffold T, Borgya A, et al. Development and in vitrocharacterization of a new immunoassay of cardiactroponin T. Clin Chem 1992;38:386-93.

7 Hangaard J, Rasmussen 0, Norregaard Hanssen K,Jorgensen N, Simonsen EE, Norgaard-Pedersen B. Earlydiagnosis of acute myocardial infarction with a rapidlatex agglutination test for semi-quantitative estimationof serum myoglobin. Acta Med Scand 1987;221:343-8.

8 Mair J, Artner-Dworzak E, Lechleitner P, Morass B,Smidt J, Wagner I, et al. Early diagnosis of acutemyocardial infarction by a newly developed rapidimmunoturbidimetric assay for myoglobin. Br Heart Jf1992;68:462-8.

9 Ellis KA, Litle T, Masud Z, Liberman HA, Morris DC,Klocke FJ. Early noninvasive detection of successfulreperfusion in patients with acute myocardial infarction.Circulation 1988;78:1352-7.

10 Katus HA, Diederich KW, Scheffold T, Uellner M,Schwarz F, Kubler W. Non-invasive assessment ofinfarct reperfusion: the predictive power of the time to

peak value of myoglobin, CK-MB and CK in serum. EurHeartJ 1988;9:619-24.

11 De Zwaan Ch, Willems GM, Vermeer F, Res J, VerheugtFWA, Van der Laarse A, et al. Enzyme tests in theevaluation of thrombolysis in acute myocardial infarc-tion. BrHeartJ 1988;59:175-83.

12 Delanghe JR, De Buyzere ML, Cluyse LP, Thierens HM,Clement DL. Acute myocardial infarction size andmyoglobin release into serum. Eur J Clin Chem ClinBiochem 1992;30:823-30.

13 Glatz JFC, Van Bilsen M, Paulussen RJA, Veerkamp JH,Van der Vusse GJ, Reneman RS. Release of fatty acid-binding protein from isolated rat heart subjected toischemia and reperfusion or to the calcium paradox.Biochim Biophys Acta 1988;961:148-52.

14 Knowlton AA, Apstein CS, Saouf R, Brecher P. Leakage

139

on February 10, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.71.2.135 on 1 February 1994. D

ownloaded from

Glatz, Kleine, van Nieuwenhoven, Hermens, van Dieijen-Visser, van der Vusse

of heart fatty acid-binding protein with ischemia andreperfusion in the rat. J Mol Cell Cardiol 1989;21:577-83.

15 Knowlton AA, Burrier RE, Brecher P. Rabbit heart fattyacid-binding protein. Isolation, characterization, andapplication of a monoclonal antibody. Circ Res 1989;65:981-8.

16 Glatz JFC, Van der Vusse GJ. Cellular fatty acid-bindingproteins: Current concepts and future directions. MolCell Biochem 1990;98:237-5 1.

17 Veerkamp JH, Peeters RA, Maatman RGHJ. Structuraland functional features of different types of cytoplasmicfatty acid-binding proteins. Biochim Biophys Acta1991;1081:1-24.

18 Kleine AH, Glatz JFC, Van Nieuwenhoven FA, Van derVusse GJ. Release of heart fatty acid-binding proteininto plasma after acute myocardial infarction in man.Mol Cell Biochem 1992;116:155-62.

19 Tanaka T, Hirota Y, Sohmiya K-I, Nishimura S,Kawamura K. Serum and urinary heart fatty acid-bindingprotein in acute myocardial infarction. Clin Biochem1991;24:195-201.

20 Van der Wieken LR, Nijssen KM, Simoons ML. Ridogrelas an adjunct to thrombolysis in acute myocardial infarc-tion. Eur HeartJ (in press).

21 Willems GM, Visser MP, Krill MTA, Hermens WTh.Quantitative analysis of plasma enzyme levels based onsimultaneous determination of different enzymes.Cardiovasc Res 1982;16:120-31.

22 Sylven C. The kinetics of myoglobin in old volunteers andin patients with acute myocardial infarction. ScandJ7 ClinLab Invest 1979;38:561-5.

23 Groth T, Sylven C. Myoglobin kinetics in patients sufferingfrom acute myocardial infarction in its early phase-asstudied by the single injection method. Scand J Clin LabInvest 1981;41:79-85.

24 Hansen NE, Karle H, Andersen V, 0lgaard K. Lysozymeturnover in man. Jf Clin Invest 1972;51:1146-55.

25 Vahlquist A, Peterson PA, Wibell L. Metabolism of thevitamin A transporting protein complex. Eur J ClinInvest 1973;3:352-62.

26 Ellis AK, Saran BR. Kinetics of myoglobin release andprediction of myocardial depletion after coronary arteryreperfusion. Circulation 1989;80:676-83.

27 Tsuji R, Tanaka T, Somiya K, Yoshimoto K, Hirota Y,Kawamura K. Canine heart fatty acid-binding protein(ch-FABP). Purification and kinetic studies [abstr].J Mol Cell Cardiol 1992;24:S183.

28 Hermens WTh, Willems GM, Davids HA, Hollaar L, Vander Laarse A. Enzymatic assessment of myocardialinjury after infarction or cardiac surgery. Is isozymeanalysis required? Clin Chim Acta 1986;156:235-46.

29 Van der Laarse A, Dijkshoorn NJ, Hollaar L, Caspers T.The (iso-)enzyme activities of lactate dehydrogenase,alphahydroxybutyrate dehydrogenase, creatine kinaseand aspartate aminotransferase in human myocardialbiopsies and autopsies. Clin Chim Acta 1980;104:381-91.

30 Borchers T, Unterberg C, Rudel H, Robenek H, Spener F.Subcellular distribution of cardiac fatty acid-bindingprotein in bovine heart muscle and quantitation with anenzyme-linked immunosorbent assay. Biochim BiophysActa 1989;1002:54-61.

31 Unterberg C, Borchers T, Hojrup P, Roepstorff P,Knudsen J, Spener F. Cardiac fatty acid-binding pro-teins. Isolation and characterization of the mitochondrialfatty acid-binding protein and its structural relationshipwith the cytosolic isoforms. 7 Biol Chem 1990;265:16255-61.

32 Hermens WT, Van der Veen FH, Willems GM, Mullers-Boumans ML, Schrijvers-Van Schendel A, RenemanRS. Complete recovery in plasma of enzymes lost fromthe heart after permanent coronary occlusion in the dog.Circulation 1990;81:649-59.

33 Ottaway JH. Evidence for binding of cytoplasmic creatinekinase to structural elements in heart muscle. Nature1967;215:521-2.

34 Vork MM, Glatz JFC, Surtel DAM, Van der Vusse GJ.Protein release from isolated rat heart during normoxia,low-flow ischemia and reperfusion. Can J PhysiolPharmacol (in press).

35 Paulussen RJA, Van Moerkerk HTB, Veerkamp JH.Immunochemical quantitation of fatty acid-binding pro-teins. Tissue distribution of liver and heart FABP typesin human and porcine tissues. Int J Biochem 1990;22:393-8.

36 Guyton AC. Glomerular filtration and glomerular filtrate.In: Wonsiewicz MJ, ed. Textbook of medical physiology.Philadelphia: Saunders, 1991:290-7.

37 Eisenberg PR, Shaw D, Schaab C, Jaffe AS. Concordanceof creatine kinase-MB activity and mass. Clin Chem1989;35:440-3.

38 Kleine AH, Glatz JFC, Havenith MG, Van NieuwenhovenFA, Van der Vusse GJ, Bosman FT. Immuno-histochemical detection of very recent myocardial infarc-tions in humans with antibodies against heart-type fattyacid-binding protein. Cardiovasc Pathol 1993;2:63-9.

39 Sacchettini JC, Hauft SM, Van Camp SL, Cistola DP,Gordon JI. Development and structural studies of anintracellular lipid binding protein in the ileal epithelium.J' Biol Chem 1990;265:19199-207.

40 Kanda T, Nakatomi Y, Ishikawa H, Hitomi M, MatsubaraY, Ono T, et al. Intestinal fatty acid-binding protein as asensitive marker of intestinal ischemia. Digest Dis Sci1992;37:1362-7.

41 Peeters RA, Veerkamp JH, Van Kessel AG, Kanda T, OnoT. Cloning of the cDNA encoding human skeletal musclefatty acid-binding protein, its peptide sequence andchromosomal localization. Biochem J 199 1;276:203-7.

42 Van Nieuwenhoven FA, Kleine AH, Keizer HA, VanDieijen-Visser MP, Van der Vusse GJ, Glatz JFC.Comparison of myoglobin and fatty acid-binding proteinas plasma markers for muscle damage in man [abstr].EurJ Physiol 1992;421:R40.

43 Newman JD. Turner APF. Biosensors: principles andpractice. Essays Biochem 1992;27:147-59.

140

on February 10, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.71.2.135 on 1 February 1994. D

ownloaded from