666

Coronary Venous Perfusion of the IschemicMyocardium during Acute Coronary Artery Occlusion

in Isolated Rat Hearts

Yuji Taira, Hideo Kanaide, and Motoomi NakamuraFrom the Research Institute of Angiocardiology and Cardiovascular Clinic, Faculty of Medicine, Kyushu University,

Fukuoka, Japan

SUMMARY. Effects of retrograde coronary venous perfusion on oxygen supply and energymetabolism of ischemic myocardium of the isolated perfused rat heart were examined by meansof NADH fluorescence photography. Occlusion of the left coronary artery produced regionalischemia of the left ventricular free wall, as evidenced by the sharply demarcated increase inNADH fluorescence. During ischemia, a narrow area of minimal fluorescence (140 ± 10 jim),indicating sufficient oxygenation for oxidative phosphorylation, was observed around the epicar-dial coronary veins in the ischemic lesion. Retrograde perfusion was introduced through thecoronary vein (left cardiac vein) that drained off the ischemic area, which resulted in a markedreduction of the area of increased NADH fluorescence in the epicardial surface. In the cross-sectional view, although the myocardium of the entire ischemic area induced by left coronaryartery occlusion could be perfused by venous retroperfusion, the effect on reduction of the areaof increased NADH fluorescence was seen only in the epicardial half of the myocardium.Retrograde coronary venous perfusion also resulted in a small increase in tension development(P < 0.05), a decrease in resting tension (P < 0.01), and partial preservation of myocardial highenergy phosphate content (P < 0.01). We propose that coronary venous retroperfusion improvesoxygenation, partially preserves oxidative phosphorylation in the epicardium, and improvescontractile function in the ischemic region. (Circ Res 56: 666-675, 1985)

EXPERIMENTS involving retrograde perfusion onarterialization of the coronary veins were first re-ported by Pratt (1893), who was able to sustain thecontraction of the cat heart for several hours bycoronary sinus perfusion. Since then, attempts havebeen made to restore the impaired myocardial cir-culation in cases of ischemic heart disease. The firstclinical application of this attempt was accomplishedby Beck (1948). He performed a carotid-coronarysinus anastomosis or an aortocoronary sinus anas-tomosis, using a vein graft. However, the mortalitywith the procedure remained high. These studieshave been overshadowed by the development ofdirect coronary artery revascularization (Berger andStary, 1971). Recently, there is renewed interest inretrograde coronary venous perfusion, in selectivesituations such as diffuse coronary arteriosclerosis,and in cases of temporal support for patients withacute myocardial infarction. Recent efforts of Meer-baum et al. (1976) showed the usefulness of a syn-chronized diastolic retroperfusion system to de-crease damage in the acutely ischemic myocardium.

There have been numerous studies on the useful-ness of retrograde coronary venous perfusion; yet,metabolic events associated with the possible in-crease in oxygenation to the tissue are poorly under-stood. The objective of our study was to obtainevidence for direct oxygen delivery to the ischemic

lesion, by retrograde coronary venous perfusion(RCVP), using NADH fluorescence photography(Barlow and Chance, 1976; Steenbergen et al., 1977;Harken et al., 1978). Nicotinamide adenine dinucle-otide, reduced form (NADH), fluorescence increaseswhen the oxygen supply is deficient, and, as a result,oxidative phosphorylation is inhibited (Chance,1976). Fluorescence photography was used to deter-mine the area involved by antegrade aortic perfusionand RCVP, and for this, a fluorescent dye was addedto the perfusate.

Methods

Perfusion TechniqueMale Wistar rats weighing 250-350 g were given free

access to a commercial diet and tap water. Thirty minutesbefore excision of the hearts, 500 U of heparin were givenintraperitoneally. The isolated hearts were perfused at apressure of 80 cm H2O by the Langendorff procedure(Langendorff, 1895), with a minor modification. The per-fusate was Krebs-Ringer bicarbonate buffer solution con-taining 2.5 mM calcium and 5.5 mM glucose, aerated witha 95% 02-5% CO2 mixture for the aerobic perfusion anda 95% N2-5% CO2 mixture for the anaerobic perfusion.The perfusate was allowed to pass through the heart onlyonce. The hearts were perfused at 30°C and were pacedelectrically at 200 beats/min. Under these conditions, weobtained a stable preparation for almost 60 minutes. Myo-

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Taira et al./Coronary Venous Perfusion in the Ischemic Rat Heart 667

cardial oxygen consumption under these conditions was1.08 ± 0.08 mmol O2/hr per g.

A strain gauge (SB-IT, Nihon Kohden) was attachedvia a nylon ligature to the ventricular apex for the meas-urement of tension development of the whole heart. Atthe start of each experiment, the resting tension was setat about 1 g.

To induce regional ischemia in the left ventricular wall,we surgically occluded the left coronary artery (LCA) at apoint 5 mm from its origin, using the techniques developedbySelyeetal. (1960).

Arterial and venous effluent Po2 during the perfusionin aerobic, anaerobic, and LCA occlusions was measuredwith a Corning blood gas analyzer. Coronary flow rates(flow rates of venous effluent) were measured with a dropcounter.

In all experiments, preconditioning of the preparationinvolved a 20-minute perfusion to wash out the red bloodcells and to stabilize the heart rate and coronary flow.

Coronary Venous RetroperfusionFor retrograde coronary venous perfusion (RCVP), a

polyethylene catheter, 300 ^m in diameter, was insertedinto the left cardiac vein (Halpern, 1953) via the rightatrium. The distal end of the catheter was carefully locatednear the margin of the ischemic region, and retrogradeflow rate through the catheter was controlled by a rollerpump. To diminish the possible fluctuation of the flowrate, a 10-ml air space was made within the perfusionroute. Since the outside diameter of the catheter was lessthan the inside diameter of the vein, obstruction of ante-grade venous drainage of the perfusate from the heartduring systole and resultant possible tissue edema wasprevented (Farcot et al., 1978). Since coronary flow ratesduring the aerobic control period and during LCA ligationwere 8.8 ± 0.3 ml/min and 5.7 ± 0.4 ml/min, respectively(indicating that there was a reduction of at least 3 ml/minduring ischemia), retrograde flow rate was set at 3 or 5ml/min; with these flow rate levels, there was no tissueedema. To monitor pressure in the retrograde-perfusedvein, we punctured a branch of the vein near the ischemic

margin, inserted a polyethylene catheter 300 fim in di-ameter, and connected it to a catheter-tip transducer (Mil-lar; pc-350). For simultaneous determination of the areaof arterial antegrade perfusion and the area of venousretrograde perfusion by fluorescence photography,fluorescein sodium (Wako-Junyaku) and thioflavin S(Chroma) were perfused from the aortic and venous roots,respectively.

Fluorescence PhotographsPhotographs of NADH fluorescence were obtained by

a system we developed in collaboration with the UnionGiken Company, Osaka, Japan, the principles and thetechniques of which were described initially by Barlowand Chance (1976). A Canon A-l camera with a 50-mmfocal length microlens and extension tube (FD-25) wasused. Tashiba Y-45 and B-46 filters were placed beforethe camera lens to allow for transmission in the 440- to500-nm region. NADH excitation was provided by two300J xenon flash tubes (FS-403, Union Giken), onemounted on either side of the camera lens, to providefairly uniform illumination on the left ventricular freewall. The xenon flash tubes were covered with ToshibaUV-D36C and UV-35 filters to provide 340-380 nm ex-citation. To avoid possible fluctuations of fluorescenceduring the cardiac cycle, the flash was synchronized tooccur during diastole, with an electrical pacing spike as atrigger.

All heart preparations for the cross-sectional NADHfluorescence photography were immersed immediately inliquid nitrogen, and then freeze-dried. Cross-sectioningwas performed with a razor blade, at a right angle to thelong axis of the freeze-dried heart, at a level 10 mm fromthe apex. For detailed microscopic examinations of thecross-sectional distribution of NADH fluorescence,fluoromicrophotography was carried out with a reflectedfluoromicroscope (Zeiss-standard 18F, filter set 01).

For the fluorescence photography of both thioflavin S(emission: 440 nm) and fluorescein sodium (emission: 530nm), 340- to 380-nm wavelength was also used for exci-tation. Both thioflavin S and NADH emit at 440 nm.

FIGURE 1. Epicardial NADH fluorescence photographs under conditions of aerobic perfusion (panel A), hypoxia (panel B), and regional ischemia(panel C).

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

668 Circulation Research/Vol. 56, No. 5, May 1985

When we examined the distribution of NADH and thio-flavin S fluorescence in the same heart, NADH fluores-cence was photographed before the venous retroperfusionof thioflavin S. The fluorescence intensity of thioflavin Sis so high that the photographic condition required todetect the fluorescence of added dyes is not sensitiveenough to be disturbed by NADH fluorescence of lowintensity. Fluorescence of NADH and added dyes wasclearly recorded in characteristic color (emission) on KodakTri X film (ASA 400, monochrome) and Kodak colornegative film (ASA 400).

Biochemical AnalysisAfter 20 minutes of regional ischemia, with or without

venous retrograde perfusion, the left ventricular free wallsubjected to ischemia assessed by NADH fluorescencewas quickly excised and frozen, using a Wollenberger'stong precooled in liquid nitrogen. In addition, the controlleft ventricular free wall was obtained from the heartperfused with aerobic solution for 20 minutes, following20 minutes of perfusion for preconditioning (control).Tissue concentrations of adenosine 5'-triphosphate (ATP)and creatine phosphate (CP) were assayed according to

the method of Lamprecht et al. (1975). Adenosine diphos-phate (ADP) and adenosine monophosphate (AMP) wereassayed by the procedures described by Jaworek et al.(1975). Tissue concentrations of nicotinamide adenine di-nucleotide (NAD) and its reduced type (NADH) wereassayed according to the method of Williamson and Cor-key (1969).

Statistical AnalysisResults were expressed as mean ± SEM. Statistical sig-

nificance was determined by analysis of variance. Valuesof P < 0.05 were considered significant.

Results

The Area of Increased NADH FluorescenceInduced by LCA Occlusion

All control photographs of each normoxic-preis-chemic heart showed uniform distributions of min-imal NADH fluorescence (Fig. 1A). When the heart,initially perfused with aerobic perfusate (Fig. 1A),was perfused with an anaerobic one, the entire heart

FIGURE 2. Photographs of NADH fluo-rescence, indicating the area of ischemialacking oxidative phosphorylation(panel A), the fluorescence of fluoresceinsodium perfused via the aorta after oc-clusion, indicating the area of normalperfusion (panel B), and the cross-sec-tional fluorescence of both NADH (right:blue) and fluorescein sodium (left: green)(panel C). Panels A and B were takenfrom the same heart in which LCA li-gation was carried out 5 mm from origin.Panel C: ligation of LCA was carried outat its origin.

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Taira et ah /Coronary Venous Perfusion in the Ischemic Rat Heart

showed an increase of epicardial NADH fluores-cence that was relatively uniform in intensity (Fig.IB). Upon occlusion of LCA, increased NADH flu-orescence was clearly visible distal to the occlusion(Fig. 1C). The ischemic myocardium showed a zoneof NADH fluorescence relatively low in intensityaround the epicardial vein, which was broader (1.4times) than that observed under conditions of an-aerobic perfusion. Figure 2A also shows a typicalNADH fluorescence photograph of regional is-chemia induced by LCA occlusion. In this experi-ment, the relatively distal portion of the LCA wasligated to demonstrate clearly the border betweenthe ischemic and normal areas, in a photograph ofthe left ventricular free wall. Because of the mini-mum NADH fluorescence, both in and around theepicardial veins, the areas appeared distinctly asnetworks of dark lines in the ischemic area of in-creased fluorescence. When fluorescein sodium wasperfused by admixture with the aortic perfusatefollowing LCA ligation, we obtained a photographof fluorescein sodium with a reversed impression ofthe NADH fluorescence photograph (Fig. 2B). Thearea of minimal NADH fluorescence in Figure 2Acoincided with the area of fluorescein sodium inFigure 2B. The epicardial veins in the ischemic areaclearly showed evidence of fluorescein sodium,which may have derived from the perfused nor-

669

TABLE 1Changes in Coronary Flow and Po2 during Normoxia,

Anoxia, and Regional Ischemia

Normoxia Anoxia Regional ischemia

Coronary flow 8.8 ± 0.3 17.5 ± 0.5 5.7 + 0.4(ml/min)

Arterial Po2 595 ± 12 39 ± 2 597 ± 13(mm Hg)

Effluent Po2 273 ± 27 23 ± 2 251 ± 12(mm Hg)

moxic area because the dye was absent in the is-chemic area and was present only in the epicardialveins.

Figure 2C shows a cross-sectional fluorescencephotograph of both NADH and fluorescein sodiumperfused from the aorta during LCA occlusion.NADH fluorescence (blue color) demonstrates ananoxic area in regional ischemia, and fluoresceinsodium (green color) demonstrates a perfusion areafrom the aorta. Between these two fluorescent areas,there was a narrow band of no fluorescence approx-imately 180 ± 30 nm wide. It is emphasized that theNADH fluorescence area was clearly demarcated,and transition from minimal to full NADH fluores-cence was abrupt.

Cross-sectional fluoromicrophotographs of

lOOjjm

FIGURE 3. Fluoromicrophotographs around the epicardial large vein under conditions of hypoxia (panel A) and regional ischemia (panel B) PanelsA and B: cross-sectional views.

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

670 Circulation Research/Vo/. 56, No. 5, May 1985

NADH of the subepicardial myocardium in the per-fusional anoxia and regional ischemia are shown inFigure 3. In cases of global anoxia, an increase inNADH fluorescence around the subepicardial veinwas fairly uniform (Fig. 3A), whereas there was anarrow less-fluorescent zone (140 ± 10 j*m in 10heart experiments) around an epicardial vein of asize approximately equal to that in Figure 3A inregional ischemia (Fig. 3B).

Changes in coronary flow rate, arterial P02 inaerobic perfusion, hypoxic perfusion, and perfusionunder LCA occlusion are shown in Table 1. Therewas an increase in coronary flow rate in cases ofhypoxic perfusion and a decrease in regional is-chemia. Venous effluent Po2 in LCA occlusion was251 ± 12 mm Hg, which is significantly lower thanin aerobic control perfusion, but much higher thanin cases of perfusional anoxia.

Effect of RCVP on Venous PressureFigure 4 shows changes in the pressure of the

distal branch of the epicardial vein, under conditionsof RCVP. The mean venous pressure in the aerobiccontrol perfusion was 8.2 ± 0.2 mm Hg. Venouspressure decreased by LCA occlusion to 3.6 ± 0.4mm Hg. During RCVP, there was a flow-dependentincrease in the venous pressure. We observed an

30

20

EE

jfjUJOH

a.

ouj 10

VENOUS RETROPERFUSION FIDW RATE

( ml / min )

FIGURE 4. Effect of coronary venous retroperfusion (RCVP) on theperfusion pressure of left cardiac vein. O: aerobic control perfusion• : during LCA occlusion (mean ± SEM of 10 heart preparations).

almost direct correlation between the RCVP flowrate and the venous pressure.

Perfusion Area of RCVPAssessment of the area perfused by RCVP (5

ml/min) during LCA occlusion was carried out byadmixing thioflavin S with the venous-retroperfus-ate. We found that the area of increased NADHfluorescence (Fig. 5A) during occlusion coincidedalmost completely with the area perfused by RCVP(Fig. 5B). In the normally perfused normoxic area,however, only a network of large epicardial veinswas stained with thioflavin S of RCVP (Fig. 5B).Thus, the dye had reached the normoxic area byRCVP, but it remained at the large epicardial veinsand did not enter the smaller veins or capillariesand, as a result, no area of thioflavin S was observedin the normoxic area. A cross-sectional view of thedistribution of the thioflavin S and fluorescein so-dium, indicating the area of RCVP and the area ofarterial antegrade perfusion, respectively, is shownin Figure 5C. The area of thioflavin S and the areaof fluorescein sodium adjoined, and both the borderzone and the overlapping of the two fluorescentdyes were negligible.

Effect of RCVP on the Area of Increased NADHFluorescence during LCA Occlusion

Figure 6 represents a typical sequence of epicardialNADH fluorescence photographs taken during aero-bic perfusion, LCA occlusion, and LCA occlusionwith subsequent RCVP at 3 ml/min and 5 ml/min.When LCA of the aerobic perfused heart (Fig. 6A)was ligated at 5 mm from its origin, a sharplydemarcated area of the increased NADH fluores-cence, relatively uniform in intensity, was observedin the left ventricular free wall (Fig. 6B). Upon RCVPat 3 ml/min, a heterogeneous reduction in the inten-sity of NADH fluorescence occurred, especiallyaround the epicardial veins, and the NADH fluores-cence had a patchy appearance (Fig. 6C). When theRCVP flow rate was increased to 5 ml/min, reduc-tion in the area and intensity of NADH fluorescencebecame more prominent (Fig. 6D).

Typical cross-sectional NADH fluorescence pho-tographs of aerobic control, LCA occlusion, and LCAocclusion with subsequent RCVP (5 ml/min) of sep-arate hearts are shown in Figure 7. There was min-imal NADH fluorescence in the aerobic control heart(Fig. 7A). Upon LCA occlusion, a transmural in-volvement of an increase in NADH fluorescencewas observed in the left ventricular free wall (Fig.7B). In accordance with a heterogeneous reductionof increased epicardial surface NADH fluorescence,RCVP after LCA occlusion induced patchy areas ofreduced intensity in NADH fluorescence in the leftventricular free wall, especially in the myocardiumadjacent to the epicardium (Fig. 7C). These patchyareas were clearly demarcated from the surroundingareas with increased NADH fluorescence, because

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Taira et al. /Coronary Venous Perfusion in the Ischemic Rat Heart 671

FIGURE 5. NADH fluorescence pho-tograph (panel A) indicates the areaof ischemia lacking oxidative phos-phorylation. Fluorescence photo-graph of thioflavin S (panel B) ad-mixed with the buffer solution forretroperfusion indicates the area ofRCVP. Panel C indicates the cross-sectional distribution of thioflavin S(right side), in the same heart.

the intensities of the NADH fluorescence of theseareas were almost equal to those of oxygenated areasof the heart. We divided the left ventricular wallinto endo- and epicardium of an equal width bydrawing a line along the midpoint between theendo- and epicardial margin. When the areas ofincreased NADH fluorescence in an epicardial halfand an endocardial half of the myocardium of theleft ventricular wall were measured separately (thevalues in the ischemic state prior to RCVP beingabout 50% for each), RCVP induced a decrease inthe area of an increased NADH by 15% in theepicardial half (P < 0.01 vs. no-RCVP) and a negli-gible change in the endocardial half (Fig. 8).

Effects of RCVP during LCA Occlusion on theMyocardial Function and MetabolicIntermediates

Changes in developed tension and resting tensionof aerobic control, and in LCA occlusion, with andwithout subsequent RCVP, are shown in Figure 9.LCA occlusion produced a rapid decrease in devel-

oped tension and a gradual increase in the restingtension. Subsequent RCVP resulted in prevention ofthese deleterious changes induced by LCA occlusion(P < 0.05 in developed tension, P < 0.01; restingtension).

Tissue concentrations of adenine nucleotides,creatine phosphate, and pyridine nucleotides after20 minutes of regional ischemia, with or withoutRCVP (5 ml/min), are shown in Figure 10. UponLCA ligation, myocardial stores of ATP, total aden-ine nucleotides, creatine phosphate (CP), and NAD+

markedly decreased (vs. normal control; P < 0.01).RCVP had a significant preserving effect on thereduction of these constituents (P < 0.01). The myo-cardial concentrations of AMP and NADH whichsignificantly increased during LCA occlusion (vs.normal control; P < 0.01), were partially preventedby RCVP (vs. no-RCVP, P < 0.01).

DiscussionIn the present study, the effect of RCVP on myo-

cardial oxygenation was evaluated by both epicar-

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

672 Circulation Research/Vol. 56, No. 5, May 1985

FIGURE 6. A representative series of epicardiaiNADH fluorescence photographs under (panel A)aerobic perfusion, (panel B) LCA occlusion, (panelC) LCA occlusion with subsequent coronary venousretroperfusion at 3 mm/min, and (panel D) at 5m\j mm.

dial and cross-sectional NADH fluorescence photog-raphy.

Barlow and Chance (1976) first used NADHfluorescence photography to evaluate regional is-chemia. When most of the NADH fluorescence wasattributed to mitochondrial NADH (Chance, 1976),there was a linear relation between increase of epi-cardiai surface fluorescence and increase of tissueNADH content (Chance, 1976; Kanaide et al., 1982),and the oxidation-reduction potential of NADH wasproportional to the state of phosphorylation (Owenand Wilson, 1974). If the intensity of NADH fluo-rescence under conditions of anaerobic perfusionwas taken as 100%, and that of the aerobic perfusion

as 0%, the increase in NADH fluorescence report-edly reached a 50% increase at approximately10~8 M intracellular oxygen tension (Chance, 1976).Thus, it is well accepted that NADH fluorescence isan excellent indicator of tissue anoxia (Chance et al.,1965). In the present study, in addition to the surfacefluorophotography, cross-sectional observations ofthe rapidly freeze-dried tissues were made, by bothfluorophotography and fluoromicroscopy, to evalu-ate three-dimensional changes in the ischemic areasand the NADH fluorescence of the perfused rathearts. The propriety of the freeze-drying techniquewas supported by the following findings obtainedfrom preliminary studies: the global shapes of the

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Taira et a/./Coronary Venous Perfusion in the Ischemic Rat Heart 673

FIGURE 7. Typical cross-sectional NADH fluorescence photographs in (panel A) aerobic perfusion, in (panel B) regional ischemia, and in (panelC) regional ischemia with coronary venous retroperfusion at 5 ml/min, in separate hearts.

heart were not changed by freeze-drying and tissueconcentrations of pyridine nucleotides before andafter freeze-drying were equal.

In contrast to the homogeneous increase in surface

no-RCVP RCVP

e p i endo e p i endo

no-RCVP RCVPFIGURE 8. Relative areas of increased NADH fluorescence in theepicardial half and endocardial half of the myocardium. The leftventricular wall was divided by drawing a line along the midpointbetween endo- and epicardium. Percent NADH fluorescence area wascalculated separately. Each column represents data on 10 heartpreparations; no-RCVP: LCA occlusion only; RCVP: LCA occlusionwith subsequent coronary venous retroperfusion (5 ml/min). **P <0.01 vs. epicardial half in no-RCVP.

NADH fluorescence in perfusional global anoxia,occlusion of LCA 5 mm from its origin under aerobicperfusion induced a nonfluorescent zone about 140nm wide around the epicardial large veins in theischemic region of increased NADH fluorescence.We mixed the fluorescent dye with aerobic-aorticperfusate and found that the epicardial large veins

o

•5 2.0

1.5

1.0

3 5 10

TIME ( min )

15 20

FIGURE 9. Changes in developed tension (above) and resting tension(below) with aerobic perfusion (O), regional ischemia (U), and re-gional ischemia with subsequent coronary venous retroperfusion (9).Coronary venous retroperfusion was introduced (5 ml/min) at thetime indicated by the arrow ([). (n — 8 heart preparations in eachgroup). *P < 0.05, " P < 0.01 DS. no-RCVP.

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

674 Circulation Research/Vol. 56, No. 5, May 1985

Aerobic control no-RCVP RCVP

FIGURE 10. Tissue concentration of high energyphosphates and pyridine nucleotides in left ven-tricular free wall of aerobic control and inregional ischemia, with and without subsequentcoronary venous retroperfusion, Ei, ATP; B,ADP; • , AMP; D, total adenine nucleotides;M, creatine phosphate; S, NAD; M, NADH; (n= eight heart preparations in each group). +P< 0.05, ++P < 0.01 vs. aerobic control; *P <0.05, "P < 0.01 vs. no-RCVP.

Regional ischemia

(left cardiac vein) in the ischemic area seem to drainaway perfusate, not only from the ischemic area,but also from the area of normal perfusion. A narrowzone 140 ^m wide around the epicardial veins,which receives an oxygen supply from these veins,may maintain myocardial oxygenation; hence, therewould be no increase in NADH fluorescence. Themaintenance of myocardial oxygenation may con-tribute to the preservation of a thin layer of intactmyocardium adjacent to the epicardial large veinsas reported in the studies of myocardial infarctionin rats (Seyle et al., 1960; Fishbein et al., 1978).

The vascular system of the rat heart differs ana-tomically from those of dogs or humans. The leftcircumflex artery is poorly developed, and the leftcoronary artery is not accompanied by large veins(Selye et al., 1960). The drainage of most of the leftventricular free wall is through the left cardiac vein,which traverses the left ventricular surface andopens in the right atrium or right anterior vena cavawithout forming a coronary sinus (Halpern, 1953).In the present study, RCVP was carried out throughthe left cardiac vein cannulated with a fine polyeth-ylene tube via the right atrium. The area of RCVPestimated by admixing thioflavin S with the retro-perfusion fluid coincided fairly well with that ofNADH fluorescence, both in the epicardial surfaceand cross-sectional views. These findings are similarto those of Gardner et al. (1974), who used 131Imacroaggregated albumin in retrograde perfusion toassess the effects of coronary circulation and con-cluded that the retrograde flow was largely distrib-uted in the ischemic zone in the in situ dog heart.

Coronary venous retroperfusion under conditionsof regional ischemia demonstrated the improvementof S-T changes in epicardial electrocardiograms(Bhayana et al., 1974) and a reduction in the in-farcted area (Chiu and Mulder, 1975). Meerbaum etal. (1976) developed a synchronized diastolic coro-

nary venous retroperfusion system with a ECG trig-gered balloon pumping unit, and noted the improve-ment of myocardial force, decrease in epicardial S-T elevation, and increase of O2 uptake, as assessedfrom blood samples. Using ultrasonic crystals tomeasure myocardial contractility and an endocardialelectrocardiogram to assess changes in the suben-docardium, Gundry (1982) demonstrated the bene-ficial effect of coronary venous retroperfusion onmyocardial contractility and also ECG changes inthe subendocardium, both in normal and hypertro-phied canine hearts. More recently, Meerbaum et al.(1982) applied hypothermic synchronized retroper-fusion in the closed-chest dog after acute coronaryocclusion and demonstrated a favorable effect oncardiac function after reperfusion. Despite the in-creasing number of reports indicating the beneficialeffects of RCVP, its effect on the nutritional supplyto the ischemic region has not been determined.

In the present study, oxygen delivery by RCVP tothe ischemic area was assessed directly by NADHfluorescence photography. RCVP induced a markedreduction of the area of increased epicardial NADHfluorescence during LCA occlusion. Although theRCVP flow was distributed over the entire ischemicarea, oxygen delivery and the resultant restorationof oxidative phosphorylation were limited to theepicardial half of the myocardium. Inhibition ofdiastolic tissue pressure reduction in the subendo-cardial myocardium in acute ischemia (Sabbah andStein, 1982) or high rate of energy utilization andresultant high rate of oxygen consumption in thesubendocardial myocardium in the reversible phaseof ischemia (Dunn and Griggs, 1975), or both, mayexplain the insufficient oxygen supply to the sub-endocardial layer by RCVP.

When Chiu and Mulder (1975) included 15-jimmicrospheres in the retrograde perfusate and esti-mated flow distribution, only a small portion of

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Taira et al./Coronary Venous Perfusion in the Ischemic Rat Heart 675

retrograde flow, even in the excised heart, providednutritional benefits to the ischemic lesion. Since theretroperfusion technique of their study is similar inprinciple to ours, the small microsphere flow in-creases in their study may be related to the patchyand modest epicardial increases in "oxygenation" inthe present study.

Although we used a relatively low work heartwith a high flow rate, and the perfusion mediumfor antegrade arterial perfusion and for RCVP wasof low viscosity, RCVP resulted in a relatively mod-est increase in tension development, a decrease inresting tension, and a partial preservation of myo-cardial high energy phosphate content. Thus, wepropose that RCVP improves oxygenation, partiallypreserves oxidative phosphorylation in the epicar-dium, and improves contractile function in the is-chemic region.

Farcot et al. (1978) reported that, in addition tothe oxygenation of the tissue, one of the importanteffects of RCVP is the acceleration of secretion oftoxic metabolites in the ischemic lesion. This factorwas not given attention in the present study.

We are grateful to M. Okada and R. Maeda for technical assistance,M. Funato of Union Ciken Co. for help in the design and developmentof the fluorescence photography systems, M. Ohara for critical read-ings of the manuscript, and N. Hayashi and A. Nishi for secretarialservices.

This study was supported by Crants-in-Aid 56570337, 57570354,and 544049 for Scientific Research from the Ministry of Education,Science and Culture, japan.

Address for reprints: Hideo Kanaide, M.D., Research Institute ofAngiocardiology, and Cardiovascular Clinic, Faculty of Medicine,Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812, japan.

Received June 27, 1983; accepted for publication December 28,1984.

ReferencesBarlow CH, Chance B (1976) Ischemic areas in perfused rat hearts:

Measurement by NADH fluorescence photography. Science193:909-910

Beck CS, Stanton E, Batinchok W, Leifer E (1948) Revasculariza-tion of the heart by graft of systemic artery into the coronarysinus. JAMA 137: 436-442

Berger RL, Stary HC (1971) Assessment of operability by thesaphenous vein bypass operation. N Engl J Med 285: 248-252

Bhayana JN, Olsen BB, Byrne JP, Kolff WJ (1974) Reversal ofmyocardial ischemia by arterialization of the coronary vein. JThorac Cardiovasc Surg 67: 125-132

Chance B(1976) Pyridine nucleotide as an indicator of the oxygenrequirements for energy-linked functions of mitochondria. CircRes 38 (suppl I): 31-38

Chance B, Williamson JR, Jamieson D, Schoener B (1965) Prop-erties and kinetics of reduced pyridine nucleotide fluorescence

of the isolated in vivo rat heart. Biochem Z 341: 357-377Chiu CJ, Mulder DS (1975) Selective arterialization of coronary

veins for diffuse coronary occlusion. ] Thorac Cardiovasc Surg70: 177-182

Dunn RB, Griggs DM (1975) Transmural gradients in ventriculartissue metabolites produced by stopping coronary blood flowin the dog. Circ Res 37: 438-445

Farcot JC, Meerbaum S, Lang TW, Kaplan L, Corday E (1978)Synchronized retroperfusion of coronary veins for circulatorysupport of jeopardized ischemic myocardium. Am J Cardiol 41:1191-1201

Fishbein MC, Maclean D, Maroko PR (1978) Experimental myo-cardial infarction in the rat. Am J Pathol 90: 57-70

Gardner RS, Magovern GJ, Park SB, Dixon CM (1974) Arteriali-zation of coronary veins in the treatment of myocardial is-chemia. J Thorac Cardiovasc Surg 68: 273-282

Gundry SR (1982) Modification of myocardial ischemia in normaland hypertrophied hearts utilizing diagnostic retroperfusion ofthe coronary veins. J Thorac Cardiovasc Surg 83: 659-669

Halpern MG (1953) Extra coronary cardiac veins in rat. Am ]Anat 92: 309-329

Harken AH, Barlow CH, Harden WR, Chance B (1978) Two andthree dimensional display of myocardial ischemic "border zone"in dogs. Am J Cardiol 42: 954-959

Jaworek K, Gruber W, Bergmeyer HU (1975) Adenosine-5'-di-phosphate and adenosine-5'-monophosphate. In Methods ofEnzymatic Analysis, edited by HU Bergmeyer. New York,Academic Press, pp 2127-2131

Kanaide H, Yoshimura R, Makino N, Nakamura M (1982) Re-gional myocardial function and metabolism during acute cor-onary artery occlusion. Am J Physio! 242: H980-H989

Lamprecht W, Stein P, Heinz F. Weisser H (1975) Creatinephosphate. In Methods of Enzymatic analysis, edited by HUBergmeyer. New York, Academic Press, pp 1777-1781

Langendorff O (1895) Untersuchungen am uberlebenden getier-herzen. Pfiir Ges Physiol Menschen Tier 61: 191-332

Meerbaum S, Lang TW, Osher JV, Hashimoto K, Lewis GW,Feldstein C, Corday E (1976) Diastolic retroperfusion of acutelyischemic myocardium. Am J Cardiol 37: 588-598

Meerbaum S, Haendchen RV, Corday E, Povzhitkov M, FishbeinMC, Y-Rit J, Lang TW, Uchiyama T, Aosaki N (1982) Hypo-thermic coronary venous phased retroperfusion: A closed-chesttreatment of acute regional myocardial ischemia. Circulation65: 1435-1445

Owen CS, Wilson DF (1974) Control of respiration by the mito-chondrial phosphorylation state. Arch Biochem Biophys 161:581-591

Pratt FH (1893) The nutrition of the heart through the vessels ofThebesius and the coronary veins. Am J Physiol 1: 86-103

Sabbah HN, Stein PD (1982) Effect of acute regional ischemia onpressure in the subepicardium and subendocardium. Am JPhysiol 242: H240-H244

Selye H, Bajusz S, Gransso S, Mendell P (1960) Simple techniquefor the surgical occlusion of coronary vessels in the rat. An-giology 11: 398-407

Steenbergen C, Delleuw G, Barlow CH, Chance B, Williamson JR(1977) Heterogeneity of the hypoxic state in perfused rat heart.Circ Res 41: 606-615

Williamson JR, Corkey BE (1969) Assays of intermediates of thecitric acid cycle and related compounds by fluorometric enzymemethods. Methods Enzymol 13: 434-514

INDEX TERMS: Myocardial ischemia • Coronary venous per-fusion • NADH fluorescence photography • Energy metabolism

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Y Taira, H Kanaide and M Nakamuraocclusion in isolated rat hearts.

Coronary venous perfusion of the ischemic myocardium during acute coronary artery

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1985 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.56.5.6661985;56:666-675Circ Res. 

http://circres.ahajournals.org/content/56/5/666World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on February 13, 2018http://circres.ahajournals.org/

Dow

nloaded from