j Mol Cell Cardio117, 973-980 (1985)

Lipid Accumulation in the Myocardium During Acute Regional Ischaemia in Cats

Hara ld Jodalen , Lodve Stange land, Ke t i l Grong, H a r a l d Vik-Mo and J o n Lekven

Departments of Anatomy, Surgery and Clinical Physiology, University of Bergen, Norway

(Received 22 May 1984, accepted in revised form 21 December 1984)

H. JODALEN, L. STANGELAND, K. GRONG, H. Vm-Mo ANDJ. LEKVEN. Lipid Accumulation in the Myocardium During Acute Regional Ischaemia in Cats. Journal of Molecular and Cellular Cardiology (1985) 17, 973-980. Accumulation of lipid material in the myocardium was studied in cat hearts with acute regional ischaemia of 3 h duration. The fractional volume of lipid droplets in cytosol was analysed by electron microscopy of myocardial biopsies using a quantitative stereologic technique. Ischaemic and normally perfused myocardium were identi- fied by fluoresceine injection, and tissue blood flow measurements were performed with labelled microspheres. In normal myocardium only small amounts of lipid droplets were found. A marked accumulation of lipid droplets occurred in borderline tissue between the two types of myocardium, whereas lipid accumulation in ischaemic myocardium was less pronounced. The arterial concentration of nonesterified fatty acids was clearly" increased during the 3 h coronary artery occlusion period. Increased trigtyceride synthesis from arterial fatty acids, or redistribution of intracellular lipids, are suggested as possible explanations for lipid accumulation during acute myocardial ischaemia.

KEY WORDS : Lipid droplets ; Microspheres ; Morphometric analysis; Myocardial infarction; Plasma fatty acids ; FFA ; NEFA.

In troduc t ion Myocardia l accumulat ion of lipid droplets has been described in ischaemic tissue [8, 17, 23]. Accumula t ion appears shortly after the onset of ischaemia and reaches a m a x i m u m degree after 3 to 6 h [22]. Beyond this time cellular damage is severe and accurate morphometr ic analysis is difficult in severely ischaemic tissue. Earlier quali tat ive and quant i ta t ive studies of this process have focused on the central ischaemic zone [7, 17, 31]. Biochemical and morphological data, however, suggest an accumula t ion of neutral lipids in tissue sur- round ing the ischaemic myocard ium after 6 to 24 h of ischaemia [2]. The concept of a border zone is controversial as to whether it is a t ran- sition zone or a sharp demarcat ion between ischaemic and normal myocard ium [11]. Based on determinat ion of local tissue blood flow, our emphasis was to apply a sampling method that provides representative biopsies from the central ischaemic, borderline, and

normal myocardial tissue after coronary occlusion of 3 h durat ion. The morpho- logical pa t te rn of myocardial lipid accumu- lation was examined in these three specific tissue regions of the myocardium, and the plasma concentra t ion of nonesterified fatty acids (NEFA) dur ing the ischaemic period was measured.

Mater ia l s and M e t h o d s

Animal preparation Eleven adul t cats of either sex with an average weight of 3.1 kg were fasted overnight and anaesthetized with sodium pentobarbi ta l i.m. (40 mg/kg). The cats were tracheotomized and venti lated with a mixture of 60% N 2 0 and 40% 0 2 conta in ing 5% CO 2 through a positive pressure venti lator ( L O O S C O Infan t Venti la tor MK2, Amsterdam), and addi- tional doses of 10 mg of sodium pentobarbi ta l were adminis t ra ted i.v. if and when needed.

Address for correspondence and reprint requests: Lodve Bergen, Haukeland Hospital, N-5016 Bergen, Norway.

0022 2828/85/010973+ 08 $03.00/0

Stangeland, Surgical Research Laboratory, University of

�9 1985 Academic Press Inc. (London) Limited

974 It. Jodalen et. al.

Arterial blood gas analyses were performed to ensure adequate ventilation throughout the experiment. The heart was exposed through a midline thoracotomy and a wide pericar- diotomy. Left ventricular blood pressure was continuously measured by a polyethylene catheter introduced through the ventricular apex and connected to a Statham P23De pressure transducer (Hato Rey, Puerto Rico). The first derivative of left ventricular press- ure, dP/dt, was obtained by a differentiating circuit (Hewlett Packard 8814A, Waltham, MA) connected to the pressure channel. Body temperaturewas measured by a rectal therm- istor and maintained at 37.0~ by an adjust- able heat pad and coverage of the opened chest. The left atrium was cannulated with a short polyethylene catheter for injection of microspheres and the left femoral artery was cannulated for collection of reference blood samples at a constant rate of 1.5 ml/min for 2 min (Sage Intruments 351, Cambridge, MA); the blood aliquots were later weighed and the exact withdrawal rate calculated.

The left anterior descending coronary artery (LAD) was carefully dissected just dis- tally to the left main stem, and a 4-0 silk thread was loosely encircled. After stable hae- modynamic conditions were reached and recorded, the LAD artery was permanently occluded by tightening the ligature. By this procedure, c. 20% to 35% of left ventricular muscle mass becomes ischaemic. Myocardial ischaemia was evident from impaired contrac- tion in the LAD-supplied tissue, reduced ven- tricular dP/dt, and raised end-diastolic press- ure. The cats were then observed for 3 h and kept in a stable haemodynamic and nor- mothermic condition. Four cats without LAD occlusion served as a control group for evalu- ation of the morphometric data.

Three minutes before killing, a population of labelled microspheres was injected into the left atrium. In order to visually identify the ischaemic tissue, 0.5 cm 3 of 10% fluoresceine was injected into the left atrium immediately before killing the animal. Exactly 3 h follow- ing LAD occlusion, the heart was perfusion fixed by injection of 40 ml ice-cold Kar- novsky's fixative into the left ventricle through the left apical catheter [15]. The heart was quickly removed, the left ventricular wall kept in cold Karnovsky's fixative to which 5%

EM

Flow measurements

Border line

itral ischaemic ocardium

Normally perfused myocardium

F I G U R E 1. Schematic representation of specimen col- lection from central ischaemic, borderline and normally perfused tissues. The right part shows how specimens were divided into two subsamples for tissue blood flow measurements with a central biopsy for electron micros- copy (EM). The figure also shows the location of EM biopsy in relation to the fluoresceine demarcation line in specimens from the borderline tissue.

sucrose was added, and cut under u.v. illumi- nation to identify ischaemic and normally perfused tissue. Nine specimens were collected from each cat heart (Fig. 1). All specimens were subdivided into two subsamples ofc. 300 mg, and these were used for tissue blood flow measurements. Biopsies of 0.7 mg for electron microscopy were collected from the mid- myocardial cut surface between the two sub- samples. Specimens which crossed from ischaemic to nonischaemic tissue showed a sharp demarcation in the cut surface, called the border line, observed under u.v. light where only normally perfused tissue shows fluorescence. In these specimens subdivision was made exactly along the demarcation line, giving one normally perfused and one isch- aemic subsample. Five border specimens were collected from each heart, and the two most representative biopsies, i.e. the specimens showing the largest difference in flow values in adjacent subsamples from the ischaemic and the normally perfused side of the border line, were selected for morphometric analysis. The biopsy sites for electron microscopy from the borderline tissue were thus verified by blood flow measurements.

All samples were transmural sections. Sub- division into endocardial and epicardial halves were not performed because of too small sample size for proper flow analysis, and

Myocardial Lipid Accumulation 975

because the cat heart shows little or no endocardial/epicardial differences in isch- aemic blood flow distribution. This has become evident from regional blood flow determinations in the cat following LAD liga- tion [9, 10]. Therefore the mid-wall portion of the transmural samples were selected for elec- tron microscopy.

Tissue blood flow measurements The regional myocardial tissue blood flow was measured by the distribution of carbonised microspheres labelled with either 141Ce or 113Sn (New England Nuclear, Boston). Approximately 2.3 x 106 spheres with an average diameter of 15.0 __+ 1.1 (S.D.) /~m were used in a randomized sequence. All tissue sub- samples were analysed for radioactive emis- sion in a multichannel 7-counting detector set at the appropriate isotope energy photopeaks (ICN Instruments SC772, Oakland). Blood flow rate and cardiac output were calculated according the comprehensive description pro- vided by Heymann et al. [12].

Lipid analysis Perfusion-fixed biopses taken for electron microscopy were kept in ice-cold Karnovsky's fixative until further processing [15]. Biopsies were washed overnight in a cacodylate- buffered 1% sucrose solution and treated with 1% cacodylate-buffered OsO4 solution for 1.5 h. After one wash in distilled water, the tissue was stained for 1.5 h in 2% uranyl acetate solution before dehydration through an acetone series.

Ultrathin sections (500 to 1000/~) were grid-stained with lead citrate [21] and there- after one randomly selected section from each biopsy was examined in the electron micro- scope. Five fields (19.4 /tm x 25.7 /tm) were selected at random from each section and photographed. In each micrograph all lipid droplets were counted. The lipid droplets were spherical, not membrane-bound, vaguely opaque and found between myofibrils, often in close contact to mitochondria. The frac- tional volume of lipid droplets, defined as the ratio of droplet area to cytoplasma area, was calculated according to the Delesse principle [32] by a morphometric analyser (CBM Inc., Model 8032, Santa Clara, USA).

To confirm that the droplets contain lipid material, additional biopsies were prepared for light microscopy. These biopsies were cry- ofixed and shock-frozen in liquid N 2. Ten micrometres of frozen sections were cut, fixed in a 10% formalin solution and rinsed in dis- tilled water, thereafter stained in a saturated solution of Sudan Black B in 70% ethanol, rinsed in running water and embedded for microscopy.

Arterial blood samples were drawn into precooled tubes, and plasma was separated by centrifugation at 4~ Blood samples were taken before, 30 min, 1 h and 3 h after coro- nary artery occlusion. Plasma concentrations of NEFA were measured by a spectrophoto- metric method [1].

Statistics A one-way analysis of variance (ANOVA) was used to calculate statistical probabilities of the morphometric data [26]. The Wilcoxon nonparametric test for paired data was used to evaluate the haemodynamic and metabolic measurements [33].

Resul t s The haemodynamic and metabolic measure- ments are presented in Table 1. Immediately following coronary occlusion, cardiac contrac- tility (dP/dt) regularly decreased and left ven- tricular end-diastolic pressure increased. However, cardiac function gradually reco- vered during the occlusion period. Three hours postocclusion, heart rate and cardiac contractility reached levels clearly above pre- occlusion values, whereas left ventricular end- diastolic pressure remained permanently elevated. Arterial plasma concentration of NEFA increased gradually during the occlusion period and was doubled after 3 h.

Coronary occlusion markedly reduced tissue blood flow in myocardium supplied by the LAD artery (Fig. 2). Normally perfused myocardium had an average tissue flow of 1.83 ml/min per g wet wt, and the average value for the central ischaemic tissue was 0.08 ml/min per g wet wt. In the selected border specimens, tissue flow in subsamples from the normal side were, on average, 93% of flow in normally perfused tissue of the same heart.

976 H . J o d a l e n et. al.

T A B L E 1. H a e m o d y n a m i c and metabo l i c m e a s u r e m e n t s . M e a n values __ S.E.M. for seven cats wi th L A D occ lus ion

Preocc lus ion 3 h pos tocc lus ion

Hear t rate (beats /min) 166 __ 12 185 _+ 15" L V S P ( m m H g ) 116 • 5 119 _-t- 5 s's" L V E D P ( m m H g ) 4 .4 _+ 0.5 7.4 __ 1.0" dP/dt ( m m H g / s ) 3143 __ 459 3964 + 338* N E F A (/~mol/1) 245 _+ 83 498 +__ 173"

LVSP = left ventricular systolic pressure. LVEDP = left ventricular end-diastolic pressure, dP/dt = first derivative of left ventricular press- ure. NEFA = arterial plasma concentration ofnonesterified fatty acids.

* P < 0.05 versus preocclusion. N.S. = not significant.

Thus, these subsamples had very little contri- bution of ischaemic tissue. Conversely, sub- samples from the ischaemic side of the border line had little contribution of normally per- fused tissue. Biopsies for electron microscopy taken from between these subsamples are thus representative for the borderline tissue.

Figure 3 shows the fractional volume ( x 10-3) of lipid droplets in ischaemic, bor- derline and normally perfused tissue; mean values were 3.25 ___ 1.08, 9.79___ 2.11 and 1.79 _+ 0.59, respectively. One-way ANOVA showed a highly significant difference between the three tissue types. There was a marked lipid accumulation in the borderline tissue compared with both the ischaemic and nor- mally perfused tissues. The fractional volume

2 .0

Q A

m.-- ~ 1.5

~.__. ~ Lo

0.5

0.1

I SE.M.

n J

CIT BLT

2 i i i i ! !

N

.L H ~

+ > H .

H ~

i i i ~ i i !

H ~

H ,

H ~

NPT

F I G U R E 2. Average myocardial blood flow in central ischaemic tissue (CIT), borderline tissue (BLT; I = ischaemic subsample, N = normal subsample) and in normally perfused tissue (NPT). Total number of flow measurements was 84.

of lipid droplets was much lower in central ischaemic tissue than in the borderline tissue, but still significantly higher than in normally perfused myocardium. In the four control cats without LAD ligation, the amount of lipid droplets averaged 1.26__ 0.31 (x 10-3), based on 24 micrographs, two biopsies from each cat. This was not different from normally perfused tissue in hearts with coronary occlusion (P > 0.3).

Figure 4 illustrates the morphology of lipid droplets, Biopsies from the borderline tissue prepared for light microscopy showed accu- mulation of Sudan Black B positive lipid drop- lets (Fig. 5).

? C)

x i/)

_~ ~o

" O

~6

L t -

. o < 0 . 0 5 ** P < 0.01

I S.E.M

CIT BLT NPT

I , _ _ J

Total number of micrographs: 210

F I G U R E 3. Schematic representation of lipid droplet accumulation in the different myocardial regions. Abbre- viations as in Figure 2.

Myocardial Lipid Accumulation 977

FIGURE 4. Micrographs of cardiac muscle from (a) normally perfused tissue, (b) borderline tissue and (c) central ischaemic tissue of the left ventricular wall 3 h after coronary occlusion. Numerous lipid droplets (arrowheads) are present in (b), less in (c). x 6900.

978 H. Jodalen et. al.

FIGURE 5. Myocardial cells showing Sudan Black B positive lipid droplets in borderline tissue. Nuclei counter- stained with haematoxylin, x 1000.

D i s c u s s i o n

This study shows that lipid accumulation occurs in severely ischaemic myocardium after coronary artery ligation of 3 h duration. Our observation using an electron micro- scopic technique is in accordance with earlier investigations in dogs using biochemical tech- niques [2 4, 6, 7, 13, 17], but in contrast to one recent study showing no effect on triglyceride content in ischaemic myocardium and lower- ing, compared to controls, in non-ischaemic myocardium in dogs with 2 h of partial coro- nary artery occlusion [27]. Most striking in our study was, however, that biopsies from the borderline tissue showed a much higher frac- tional volume of lipid droplets than from the central ischaemic tissue. Borderline tissue therefore appears to be morphologically dif- ferent from the central ischaemic tissue in this respect. I t is emphasized that the term 'bord- erline tissue' in this text refers to the tissue as close to the fluoresceine demarcation line as possible and with clearly different blood flow values in adjacent subsamples.

We have identified Sudan Black B positive lipid droplets in biopsies prepared for light microscopy, especially in borderline tissue (Fig. 5). Our results correspond well with the findings of Bilheimer et al. [2], who showed

that after 6 h of LAD occlusion the most pro- nounced accumulation of labelled lipids were located in the tissue surrounding the isch- aemic myocardium. A similar tendency was observed, but not quantified, by light and electron microscopy in their study. Our inter- pretation of the vacuoles observed by electron microscopy as lipid droplets is in accordance with earlier investigators [2, 8, 17], although direct evidence for the true nature of vacuole content has not conclusively been demon- strated.

The mechanism by which lipid material accumulates in ischaemic tissue is still poorly understood. Lipid droplets are believed to represent deposits of neutral fat in the cell [7]. Tissue accumulation of triglycerides might be due to increased synthesis from plasma NEFA, or reduced lipolysis of endogenous tri- glycerides within the ischaemic myocardium, or both. Plasma NEFA are still extracted in the reversibly ischaemic myocardium, although at a reduced rate [20], but since beta-oxidation is impaired, the main fate of extracted NEFA is esterification to tri- glycerides [18, 20]. During acute myocardial ischaemia the incorporation of 14C-palmitate into triglycerides is increased, suggesting a stimulation of triglyceride synthesis in isch- aemic tissue [24]. Since extraction of NEFA in

Myocardial Lipid Accumulat ion 979

i schaemic myoca rd ium is dependen t on the p l a sma N E F A concent ra t ion besides the residual blood flow [29], the e levat ion o f p l a sma N E F A concent ra t ion found in this s tudy dur ing coronary occlusion would favour N E F A up take in the ischaemic myoca rd ium and thus would tend to enhance t r iglyceride synthesis. O n the other hand, lipolysis of myo- ca rd ia l t r iglycerides is also enhanced dur ing acute myoca rd ia l i schaemia [5, 30], but our s tudy indicates that despite this there is net t r ig lycer ide synthesis 3 h after coronary ar tery l igat ion. Based on biochemical analysis of ischaemic myoca rd ium in dogs, Jesmok et al. found evidence for a biphasic effect of coro- na ry occlusion on endogenous tr iglycerides wi th a lowering after 30 min of i schaemia fol- lowed by a g radua l increase dur ing the follow- ing hours [13].

Lip id drople t accumula t ion has been repor ted in cul tured cardiocytes dur ing anoxia in med ium without exogenous N E F A [-25]. Thus, l ipid accumula t ion might not be solely dependen t of exogenous N E F A supply bu t m a y also represent redis t r ibut ion o f in t r a - cel lular lipids. This concept also receives suppor t from earl ier investigations at our ins t i tu t ion where lipid d rop le t format ion dur ing i soproterenol - induced myocard ia l d a m a g e was found not to be solely dependen t of the p lasma N E F A level [14].

Accumula t ion of l ipid in ischaemic and border l ine ischaemic myoca rd ium is believed to have toxic effects. Ra t hearts with high tri-

g lycer ide contents exposed to ischaemic per- fusion exhibit a h igher oxygen requi rement and an increased incidence of a r rhy thmias c o m p a r e d with hearts with no rma l tri- glycer ide content [3], Whe the r toxici ty of accumula ted tr iglycerides is dependen t on the t r iglycerides per se, or is re la ted to the well documen ted effect of high N E F A levels on the ischaemic myoc a rd ium [16, 19, 28], is not clear. Lack of vent r icu lar ectopic act ivi ty in our s tudy does not indicate a re la t ionship be tween a r rhy thmia and l ipid accumulat ion .

O u r results show tha t there is an extensive in t race l lu la r accumula t ion of l ipid droplets in the border l ine ischaemic myocard ium, and indicate that this pa r t of the myoc a rd ium is morphologica l ly and metabol ica l ly different from both ad jacent normal and ischaemic myoca rd ium. Whe the r this is a t empora ry phenomenon , ana tomica l ly fixed or dynamic , and whether it is of significance for protec t ion of the ischaemic myoca rd ium, are questions tha t mer i t fur ther investigation.

Acknowledgements W e are indebted to E. K. Olsen, T. Henr ich- sen, I. Vikoyr , E. Sohvedt , L, H. Andreassen and T. Fiskeseth for skilled technical assist- ance. F inancia l suppor t was received from the Norwegian Research Council for Science and the Humani t ies ( N A V F grants C13.21.30- 026, C13.31.36-048 and -045), and the Nor- wegian Council on Card iovascula r Diseases.

R e f e r e n c e s l BERGMAN, S. R., CARLSON, E., DANNEN, E., SOBEL, B. E. An improved assay with 4-(2-thiazolylazo)-recorcinol for

non-esterified fatty acids in biological fluids. Clin Chim Acta 104, 53 63 (1980). 2 BILttEI~ER, D. W., BUJA, L. M., PARREY, R. W., BONTE, F.J., WILL~RSON, J. T. Fatty acid accumulation and

abnormal lipid deposition in peripheral and border zones of experimental myocardial infarcts. J Nucl Med 19, 276 283 (1978).

3 BROWNSEY, R. W., BRUNT, R. V. The effect of adrenaline-induced endogenous lipolysis upon the mechanical and metabolic performance ofischaemically perfused rat hearts. Clin Sci Mol Med 53, 513 521 (1977).

4 BRUCE, T. A., Myers, J. T. Myocardial lipid metabolism in ischemia and infarction. Rec Adv Cardiac Struct Metab3,773 780 (1973).

5 CHRISTIAN, D. R., KILSHEIMER, G. S., PETTETT, G., PARADISE, R., ASHMORE, J. Regulation of lipolysis in cardiac muscle: a system similar to the hormone-sensitive lipase of adipose tissue. Adv Enzyme Regul 7, 71 82 (1969).

6 CRASS lI1, M. F., STERRETT, P. R. Distribution of glycogen and lipids in the ischemic canine left ventricle: biochemical and light and electron microscopic correlates. Res Adv Cardiac Struct Metab 10, 251 263 (1975).

7 CRASS III, M. F. Myocardial triacylglycerol metabolism in ischemia. Tex Rep Biol Med 39, 439 452 (1979). 8 DUSEK, J., RONA, G., KAHN, D. S. Healing process in the marginal zone of an experimental myocardial infarct.

Findings in the surviving cardiac muscle cells. AmJ Patho162, 321 338 ( 1971). 9 GRONG, K., STANGELAND, L., ANDERSEN, K. S., LEKVEN, J. Effect of timolol on blood flow distribution in the

myocardium during acute regional ischaemia in cats. Cardiovasc Res 15, 430- 435 (1981).

980 H. Jodalen et. al.

10 GRONG, K., STANGELAND, L., ANDERSEN, K. S., LEXVEN, J. Effects of timolol on blood flow distribution in the feline myocardium with acute regional ischaemia during controlled haemodynamic conditions. Cardiovasc Res 16, 269-275 (1982).

11 HEARSE, D. J., YELLON, D. M. The 'boderzone' and myocardial protection: a time for reassessment? Acta Med Scand [Suppl 651] 37-46 (1981).

12 HEYMANN, M. A., PAYNE, B. D., HOFFMAN, J. I. E., RUDOLPH, A. M. Blood flow measurements with radionuclide- labeled particles. Progr Cardiovasc Dis 20, 55-79 (1977).

13 JESMOK, G. J., WARLTIER, D. C., GROSS, G. J., HARDMAN, H. F. Transmural triglycerides in acute myocardial ischaemia. Cardiovasc Res 12, 659-665 (1978).

14 JODALEN, H., LIE, R., ROTEVATN, S. Effect of isoproterenol on lipid accumulation in myocardial cells. Res Exp Med 181,239-244 (1982).

15 KARNOVSRY, M. J. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 27, 137A (1965).

16 LIEDTKE, A. J. Alterations of carbohydrate and lipid metabolism in the acutely ischemic heart. Prog Cardiovasc Dis 23, 321-336 (1981).

17 NAYLER, W. G., SLADE, A. M. The cardioprotective effect of verapamil. Clin Exp Pharmacol Physiol 6, 75 87 (1982).

18 NEELV, J. R., MORGAN, H. E. Relationship between carbohydrate and lipid metabolism and the energy balance of heart muscle. Ann Rev Physio136, 413 459 (1974).

19 OLIVER, M. F. The influence of myocardial metabolism on ischemic damage. Circulation 53 [Suppl I], 168 170 (1976).

20 OPIE, L. EI., OWEN, P., RIEMERSMA, R. A. Relative rates of oxidation of glucose and free fatty acids by ischaemic and non-ischaemic myocardium after coronary artery ligation in the dog. EurJ Clin Invest 3, 419-435 (1973).

21 REYNOLDS, E. S. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J Cell Biol 17, 208-212 (1963).

22 SAKURAI, I. Pathology of acute ischaemic myocardiuin. Acta PatholJap 27, 587 603 (1977). 23 SCHAPER, J. Structural characteristics in regional versus global ischemia. Adv Myocardiol 2, 311 314 (1980). 24 SCHEUER, J., BRACttFELD, N. Myocardial uptake and fractional distribution ofpalmitate-1-C 14 by the ischemic dog

heart. Metabolism 15, 945-954 (1966). 25 SCHWARTZ, P., PIPER, H. M., SPAHR, R., SmECKERMANN, P. G. Ultrastructure of cultured adult myocardial cells

during anoxia and reoxygenation. AmJ Pathol 115, 349 361 (1984). 26 SNEDECOR, G. W., COCHRAN, W. G. Statistical Methods, 6th edn. Ames, Iowa: The Iowa State University Press,

(1967). 27 VAN DER VUSSE, G.J., ROEMEN, Th. H. M., PmNZEN, F. W., COUMANS, W. A., RENEMAN, R. S. Uptake and tissue

content of fatty acids in dog myocardium under normoxic and ischemic conditions. Circ Res 50, 538 546 (1982). 28 Vm-Mo, H., Mjos, O. D. Influence of free fatty acids on myocardial oxygen consumption and ischemic injury.

AmJ Cardio148, 361-365 (1981). 29 VIK-Mo, H., Mjos, O. O., RIEMERSMA, R. A., OLIVER, M. F. Importance ofischemia-induced myocardial lipolysis

in dogs. Adv Physiol Sci 8, 121 128 (1980). 30 VIK-Mo, H., RIEMERSMA, R. A., MJOS, O. D., OLIVER, M. F. Effect of myocardial ischaemia and antilipolytic

agents on lipolysis and fatty acid metabolism in the in situ dog heart. ScandJ Clin Lab Invest 39, 559-568 (1979). 31 WARTMAN, W. B.,JENNm~S, R. B., YOROYAMA, H. O., CLABAVGH, G. F. Fatty change of the myocardium in early

experimental infarction. Arch Patho162, 318 323 (1956). 32 WEIBEL, E. R., KISTLER, Q. S., SCHERLE, W. F. Practical stereological methods for morphometric cytology. J Cell

Bio130, 23-38 (1966). 33 ZAR, J. H. Biostatistical Analysis. Englewood Cliffs, N.J. : Prentice-Hall Inc. (1974).