Functional heterogeneity of oxygen supply-consumption ratio in the heart

Cardiovascular Research 44 (1999) 488–497www.elsevier.com/ locate /cardiores

www.elsevier.nl / locate /cardiores

Review

Functional heterogeneity of oxygen supply-consumption ratio in the heart

*C.J. Zuurbier , M. van Iterson, C. InceDepartment of Anaesthesiology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

Received 17 March 1999; accepted 8 July 1999

Abstract

In this review, the regional heterogeneity of the oxygen supply-consumption ratio within the heart is discussed. This is an importantfunctional parameter because it determines whether regions within the heart are normoxic or dysoxic. Although the heterogeneity of thesupply side of oxygen has been primarily described by flow heterogeneity, the diffusional component of oxygen supply should not beignored, especially at high resolution (tissue regions <1 g). Such oxygen diffusion does not seem to take place from arterioles or venuleswithin the heart, but seems to occur between capillaries, in contrast to data recently obtained from other tissues. Oxygen diffusion mayeven become the primary determinant of oxygen supply during obstructed flow conditions. Studies aimed at modelling regional bloodflow and oxygen consumption have demonstrated marked regional heterogeneity of oxygen consumption matched by flow heterogeneity.Direct, non-invasive indicators of the balance between oxygen supply and consumption include NADH videofluorimetry (mitochondrialenergy state) and microvascular PO measurement by the Pd-porphyrin phosphorescence technique. These indicators have shown a2

relatively homogeneous distribution during physiological conditions supporting the notion of regional matching of oxygen supply withoxygen consumption. NADH videofluorimetry, however, has demonstrated large increases in functional heterogeneity of this ratio incompromised hearts (ischemia, hypoxia, hypertrophy and endotoxemia) with specific areas, referred to as microcirculatory weak units,predisposed to showing the first signs of dysoxia. It has been suggested that these weak units show the largest relative reduction in flow(independent of absolute flow levels) during compromising conditions, with dysoxia initially developing at the venous end of thecapillary. 1999 Elsevier Science B.V. All rights reserved.

Keywords: Regional blood flow; Oxygen consumption; Microcirculation, Ischemia; Coronary circulation

1. Introduction demonstrating marked regional variations in flow withinthe heart [1,2]. However, because oxygen is delivered by

The balance between oxygen supply and oxygen de- convection and diffusion, the heterogeneity of oxygenmand is of paramount importance for the heart since it supply is expected to differ from that of blood flow,determines whether the tissue is healthy or dysoxic (where especially when spatial resolution improves below 1-goxygen need exceeds oxygen supply). Yet the assessment tissue pieces. Such a discrepancy between flow supply andof this balance at a regional level has been exceedingly oxygen supply is supported by the observation of increaseddifficult to study. This is mainly due to the fact that both homogeneity of tissue oxygenation when diffusion be-the regional oxygen supply to the myocyte, which is tween capillaries is incorporated in a mathematical simula-provided by diffusion, and the regional oxygen consumption of the capillary circulation of the myocardium [3].tion are each difficult to assess separately, let alone Regional oxygen consumption of the heart has primarilysimultaneously. been determined by combining quick-freeze techniques

Oxygen supply is to a large part determined by convec- with microspectrometric determinations of arterial andtive blood flow and many excellent studies have been venous haemoglobin saturation [4]. Promising new de-

15 13carried out on the heterogeneity of cardiac blood flow, velopments such as O-oxygen PET [5] and C-NMR [6]obtain estimates of regional cardiac oxygen consumptiontogether with regional measurements of flow. However,*Corresponding author. Tel.: 131-20-566-5259; fax: 131-20-697-

9004.E-mail address: [email protected] (C.J. Zuurbier) Time for primary review 36 days.

0008-6363/99/$ – see front matter 1999 Elsevier Science B.V. All rights reserved.PI I : S0008-6363( 99 )00231-X

by guest on June 1, 2013http://cardiovascres.oxfordjournals.org/

Dow

nloaded from

http://cardiovascres.oxfordjournals.org/

C.J. Zuurbier et al. / Cardiovascular Research 44 (1999) 488 –497 489

these techniques are very elaborate, expensive and difficult hypoxia and ischemia on flow heterogeneity are criticallyto execute. Therefore, techniques which can give direct dependent on sample size [15] which may be one explana-measures of the balance between oxygen supply and tion for the conflicting results concerning effects ofconsumption are valuable tools, since such techniques will hypoxia and ischemia on flow heterogeneity: the hypoxiaprovide direct information about whether tissue cells are studies have used sample sizes ,0.01 g, whereas in thewell oxygenated or dysoxic. Regional NADH video- ischemia studies sample sizes were .0.01 g. Thus, re-fluorimetry [7,8] and microvascular oxygen pressures gional flow heterogeneity is strongly dependent on samplemeasured by Pd-porphyrin phosphorescence [9] are two size and condition of the heart (normal, hypoxia, ischemia).such direct indicators of the oxygen supply-consumption The important question remains, however, as to how thisratio. Since these indicators have been predominantly convective flow heterogeneity translates into the distribu-applied to the epicardium, heterogeneity in this review will tion of oxygen supply by diffusion.be discussed in terms of regional heterogeneity of theepicardium, without discussing chamber or layer (endo-cardium versus epicardium) heterogeneity. 3. Oxygen diffusion

The aim of this paper is to review the current knowledgeconcerning both the regional supply side of oxygen as well 3.1. Direct measurements of oxygen diffusion from theas the consumption side of oxygen within the heart and vasculaturecompare this information with that obtained by direct

thindicators of the oxygen supply-consumption ratio. The In the first half of the 20 century, it was thought thatheterogeneity of this ratio will be examined during nor- diffusion of oxygen from the vasculature system occurredmoxic and oxygen compromised conditions. mainly from the capillaries [22]. More recently, evidence

has shown that oxygen diffuses to and from most elementsof the vasculature system (thus not only from capillaries),

2. Flow heterogeneity and accounts for a significant portion of oxygen transportto the tissue cells [23]. When going from large to small

Micro-heterogeneity of flow has been established to a vessels in the arteriolar network, a longitudinal decrease inlarge extent by the use of microspheres since the early PO has been observed in the cheek pouch [24], cremaster2

studies of Yipintsoi et al. [1], and confirmed by others (for muscle [24], mesenteric microvascular network [25], ileumreview see Ref. [2]). An important feature of flow hetero- [26] and dorsal skin preparation [27]. In addition, vasculargeneity is that it increases as tissue piece size decreases PO was also found to increase with increased vessel size2

[10]. This can be accounted for by the tree-like coronary in the venular network, with capillaries taking up oxygenanatomy of vessels, from the large supplying arteries up to from overlying venules [28,29].the terminal arterioles [11]. Such fractal nature may All the aforementioned studies, however, have beenexplain the dependency of flow heterogeneity on piece size conducted in tissue beds other than the heart, and very few[12,13]. Being fractal, i.e. showing self similarity upon studies can be found that have directly examined oxygenscaling, suggests increased flow heterogeneity as resolution diffusion from large vessels within the heart. Weiss andis increased due to just one, similar, recurrent determinant Sinha [30] have observed no correlation between vesselof heterogeneity (such as branching patterns) for all piece size and arterial oxygen content, suggesting negligiblesizes. However, the non-tree-like topology of capillaries oxygen diffusion from these vessels. Honig and Gayeski[14] results in a breakdown of the fractal nature of flow [31] found no change in haemoglobin saturation as aheterogeneity and consequently in a different dependency function of vessel size between 20 and 200 mm forof heterogeneity on piece size [15]. This may explain why arterioles and venules in frozen tissue pieces of rat and dog

2flow heterogeneity measured at a resolution ,1 mm is heart. It was concluded that due to high red blood cellmuch smaller than predicted from the relation between velocity and short arteriolar length not enough time isheterogeneity and piece size for resolutions much larger available to allow a significant amount of oxygen to

2than 1 mm [15]. Although coronary anatomy is a primary diffuse across the vessel wall. Their point is well taken:determinant of flow heterogeneity, vascular tone also assuming the length of the cardiac arteriolar bed to beinfluences blood flow distribution [16]. In this respect, it is approximately 2 mm, while the diameter decreases fromimportant to note that flow heterogeneity (determined by 200 to 20 mm [32], a decrease of only 6% in Hb saturationdeposition of molecular markers or indicator-dilution or 6 mm Hg in vascular PO can be calculated when2

experiments) decreases with reductions in arterial O extrapolating the measured amount of oxygen that diffuses2

tensions at constant flow, probably as a result of decreased from arterioles in the dorsal skin preparation [27]. Thisvascular tone [15,17]. In contrast, most studies directed at anticipated small decrease would even be less when thethe effects of ischemia on heterogeneity have observed an higher flow rates in the arterial compartment of the heart,increase in heterogeneity (determined by microspheres) as compared to that in the dorsal skin preparations, arewith decreases in flow [18–21]. However, effects of considered. Data shows, thus, that oxygen diffusion from


Dow

nloaded from


490 C.J. Zuurbier et al. / Cardiovascular Research 44 (1999) 488 –497

large vessels is minimal for the heart due to the short flow perfused rat heart appear at higher effluent PO during2

path lengths and high flow rates. ischemia (low flow, normal PO perfusate) as compared to2

hypoxia (normal flow, low PO perfusate). Although this2

may be explained by differences in pH, there is also the3.2. Diffusive shunting of oxygen within the heart possibility that due to the lower flow during ischemia,

there is more time during ischemia than during hypoxia forInsight into the properties of O diffusion within the diffusional shunting to occur. This ‘shunted’ oxygen is not2

heart has been limited, due to the lack of appropriate used by the myocyte (anoxic zones) and increases thetechniques. The major available information concerning venous effluent PO .2

the nature and extent of diffusion within the heart has been Analysis of emergence functions in whole heart studies,obtained from experiments that have focused on a conse- however, does not have the sensitivity to exclude thequence of diffusion, i.e. diffusive shunting of compounds presence of shunting that bypasses only a small part of thewithin the heart. vasculature system, i.e. between capillaries at less extreme

Diffusive shunting is defined as the bypassing of a part ends of their length. That such diffusion probably takesof the vascular system by diffusion, e.g. arteriovenous place has been suggested by Wolpers et al. [39], where it(direct diffusion from the arteriole to the venule) or inter- was found that a model which allowed bypassing of aendcapillary (release and uptake of oxygen by different small part (,300 mm) of the capillary functional lengthcapillaries). This type of shunting can be envisaged by (500 mm) best described the outflow curves of inert gasescomparing the emergence function (for peak height or time in closed-chest dogs. In comparison, models that complete-of arrival) of compounds with different diffusibilities. For ly bypassed the capillary bed (thus arteriovenous shunting)example, diffusive shunting is indicated when a diffusible generated poor descriptions of the outflow curves.compound has a larger peak height or earlier time of In this context it is interesting to note that modelarrival in the venous effluent than a non-diffusible com- analysis of the transport of water or oxygen through thepound, which remains intravascular. heart, that does not take diffusional shunting into account

One of the first studies to examine diffusional shunting, between capillaries, resulted in low permeability estimatesin an isolated blood-perfused dog heart, found that the ratio for water and oxygen, which decreased as flow wasof the venous dilution peak of the compound with the decreased [40,41]. These low permeabilities were ex-highest diffusibility (tracer water) to the compound of plained by the suggestion that the capillary and thelower diffusibility (antipyrine) increased with lower flow- sarcolemmal membranes constitute a significant barrier to

21 21rates (,1 ml min g ) [33]. This data demonstrated the water and oxygen [36,42], in contrast to previous findingspresence of a diffusional shunt, even though the shunting which suggest water transport to be mainly flow-limitedfraction of water amounted to less than 3%. Roth & Feigl [33]. However, when diffusional interactions between[34] examined diffusional shunting in closed-chest dogs capillaries is allowed in the models, by either using highwith controlled coronary blood flow (0.21 to 1.17 diffusion constants in the model [43] or actually construct-

21 21ml min g ) and compared venous appearance times of ing interactions between capillaries [44], normal per-hydrogen with intravascular indicators. At low flow (0.2– meabilities were sufficient to describe the experimental

21 210.8 ml min g ), less than 0.1% of injected hydrogen data.preceded the intravascular, non-diffusible indicators. Be- In conclusion, most data indicate that diffusive shuntingcause the diffusion constant of hydrogen is approximately of oxygen within the heart primarily takes place betweentwice that of oxygen, this data indicated, indeed, a very capillaries and not between larger vessels. That diffusionsmall degree of oxygen shunting. This finding has been between large vessels does not seem to take place withincorroborated by the observation that the transport of the heart, is in agreement with the limited data whichoxygen always lags behind the transport of an intravascu- shows no detectable oxygen loss from these vessels [31].lar, non-diffusible compound during normal or high flow The diffusion of oxygen between capillaries causes oxygenconditions in rabbit and dog hearts [35,36]. In a subsequent supply at the capillary level to be more homogeneous thanstudy by Bassingthwaighte et al. [37] it was concluded oxygen supplied by convective flow alone, and may havethat, while the shunting of heat in isolated blood-perfused important functional consequences during restricted oxy-dog heart is great (diffusion constant of heat is two orders gen supply conditions.of magnitude higher than oxygen), the shunting of oxygenwill not be large. It can be concluded from the abovestudies that arteriovenous or end-capillary shunting of 4. Discrepancy between blood flow and oxygen supplyoxygen within the heart is negligible at normal, physiologi-

21 21cal flows (| 1 ml min g ), and increases only marginal- Blood flow can be considered as the convective trans-21 21ly at low, pathological flows (,0.5 ml min g ). These port of blood through the vascular system, and perfusion

findings may also explain the earlier findings by Steen- as the transport of solutes (such as oxygen) by blood flowbergen et al. [38] that regional anoxic zones in the isolated and diffusion. As such, regional blood flow has been


Dow

nloaded from



conventionally estimated by infusion and counting of the ples of the dog heart [4]. Subepicardial oxygen consump-tion was found to be approximately 20% lower thannumber and distribution of particles such as microspheressubendocardial oxygen consumption. These techniquesthat plug a minor part of the vasculature. Only small biaseshave however not been used to examine microhetero-exist toward higher flow in high flow regions, and lowergeneity of oxygen supply-demand. It should also be notedflows in low flow regions, especially when sample regionthat this technique is not without criticism: tissue samplesize decreases [12]. During normal physiological flowfreezing takes at least several seconds and is probablyconditions, diffusible tracers also give approximately simi-heterogeneous which may cause regional changes inlar perfusion heterogeneities as determined by micro-venous oxygen content as a result of irregular beatingspheres, although there is a tendency for higher hetero-tissue where metabolism continued and venous bloodgeneity with microspheres as compared with diffusiblemight redistribute [6].tracers [1]. Under conditions of restricted flow, however,

Only recently has the direct examination of regionalperfusion as determined by diffusible tracers is larger thanmyocardial oxygen consumption been pursued, the inter-flow determined by microspheres [45,46]. These findingsruption probably related to the difficulties of developingare corroborated by studies examining the transition zonereliable techniques. Regional oxygen consumption hasbetween the normoxic zone with normal blood flow

15been obtained for the in situ dog heart, combining O-(measured by fluorescein angiography) and the anoxicpositron emission tomography with an axial distributedzone (marked by NADH fluorescence) with no flow [47].blood–tissue exchange model [5]. Following intravenousThis transition zone was still normoxic despite the lack of

15injection of O-water for blood flow determination, theblood flow and had a length of about 300 mm. Thusmodel estimates regional oxygen consumption from thethrough diffusion alone, oxygen was supplied over a

15regional time–activity curve of inhaled O-oxygen fordistance of 300 mm. Together, this data indicates thatregions as small as 0.5 g. Correlations of r50.5–0.7 wereoxygen supply is probably more homogeneous and largerobserved among regional blood flow and oxygen consump-than blood flow in areas of restricted blood flow, due to thetion. Other studies have also demonstrated significantdiffusional aspect of oxygen supply.correlations between regional blood flow and metabolism:Wieringa et al. developed a network model of thebetween regional flow (microspheres) and regional oxygenmyocardial microcirculation to evaluate microheterogen-consumption calculated from the tricarboxylic acid cycleeity of tissue or capillary PO [3]. An assumption in the2

13rate using C-NMR and mathematical modelling in frozenmodel was the presence of homogeneous oxygen consump-tissue samples of isolated rabbit hearts [6]; betweention, so that effects of blood flow and different types ofregional blood flow (microspheres) and regional glucoseoxygen diffusion could be studied. This model demon-

3strated that intercapillary diffusion (an aspect not incorpo- uptake measured by H-deoxyglucose deposition in frozenrated in the classical Krogh’s cylindrical tissue model) tissue samples of in situ dog hearts [48]); between regionalresulted in more homogeneous capillary and tissue oxygen- flow (microspheres) and regional O consumption esti-2

18ation. These simulations showed larger heterogeneity for mated from H O residue counting in frozen tissue sam-2

flow than for oxygen and indicated that oxygen supply at ples of isolated rabbit hearts [49].15 13the level of the capillaries cannot only be determined by Both the O-PET and C-NMR techniques are state-

convective blood flow. of-the-art, highly developed, analysis tools to obtainSummarising, the diffusion component of perfusion quantitative data on regional myocardial oxygen consump-

results in oxygen supply being more homogeneous than tion. However, most of the new techniques are veryblood flow at the high spatial resolution of capillaries elaborate, time-consuming methods, that are also either

13during normal conditions. During restricted flow condi- expensive (PET) or destructive ( C-NMR, deoxyglucose-18tions, the impact of diffusion is probably extended to large and H O-accumulation), thus not allowing continuous2

tissue areas such that these areas may still be perfused measurement of regional oxygen consumption. It is alsoalthough minimal blood flow is present. This illustrates difficult with these techniques to obtain direct informationthat blood flow is not always a good indicator of oxygen concerning the functional heterogeneity (whether tissuesupply, and shows the difficulties in determining the regions are ischemic). In addition, no complete determi-supply side of oxygen. nation of the oxygen supply-consumption ratio is feasible

with these techniques, because in most cases (studies usingmicrospheres) only the convective supply side of oxygen is

5. Regional myocardial oxygen consumption measured. The omission of the diffusive component willbecome increasingly important during restricted oxygen

One of the first studies to provide quantitative in- supply conditions and when piece sizes smaller than 1 gformation on transmural myocardial oxygen consumption are examined: e.g. assuming that oxygen can travel adetermined haemoglobin saturation of arterial and venular distance of 300 mm by diffusion, this oxygen can affectvessels (20–200 mm) in combination with regional blood 11% of the volume of a 0.5 g tissue piece. Nevertheless,flow determination by microspheres in frozen tissue sam- these techniques have collectively shown that oxygen


Dow

nloaded from



consumption is heterogeneously distributed throughout the crocirculation of rat intestine, it primarily mirrors capillaryand venular PO values [26]. Therefore, the PO de-heart, which is, to a large part, matched to flow hetero- 2 2

termined with the fibre phosphorimeter has been calledgeneity: low flow regions having low oxygen consumption,microvascular PO , mPO . Most importantly, we directlyhigh flow regions having high oxygen consumption. 2 2

demonstrated here that the mPO is an out measure of the2

balance between oxygen supply and consumption inLangendorff-perfused pig hearts (Fig. 1). An increase of6. The oxygen supply and consumption ratioheart rate (↑O consumption) at constant flow (constant O2 2

6.1. Direct measurements of the oxygen supply- supply) decreased the mPO , whereas a decrease in flow2

consumption ratio (↓O supply) at constant mechanical performance (con-2

stant O consumption) also decreased the mPO (Fig. 1).2 2

When supply becomes limiting compared to consump- Although other techniques that determine myocardialtion, and no downregulation of metabolism occurs such as oxygenation may also partly reflect the balance betweenseen during hibernation, the myocyte becomes dysoxic and O supply and consumption, such as surface [60] and2

the PO in the capillary /venous compartment stabilises at tissue penetrating oxygen electrodes, the undefined origin2

a very low or even zero level. In that case, parameters that (vascular, cellular, interstitial) of these electrode PO ’s2

reflect the transition to the dysoxic state of the myocyte are hampers a clear interpretation of these signals and, thus,of functional importance. The term dysoxia, which is will not be further discussed here.defined as O -limited cytochrome turnover, is preferred A limitation of using these non-invasive surface illumi-2

because the prefix ‘dys’ describes both oxygen supply and nation techniques is the restricted penetration depth of theconsumption [50]. This in contrast to terms such as used wavelengths (NADH: |0.36 mm [61]; Pd-porphyrin:ischemia (literally meaning ‘restrained flow’) or hypoxia(meaning low oxygen), which have been shown to causeconfusion through the definitions [51].

A parameter that is viewed as the gold standard for thecondition of dysoxia within cells is the mitochondrial

1NADH/NAD redox state [7,8]. NADH is visible due toits fluorescence upon excitation with ultraviolet light,

1whereas NAD is not [52,53]. The redox state can bemeasured in vitro by videofluorimetry and after suitablecorrection for the optical properties of blood in vivo bydual wavelength videofluorimetry [54]. That NADH re-flects the balance between supply and consumption hasbeen shown by Steenbergen et al. [38]: NADH fluores-cence increased in isolated perfused rat hearts with tachy-cardia during constant low flow. It should be noted,however, that NADH levels have a low sensitivity tochanges in oxygen supply or demand during normoxicconditions, along with a strong dependency on substrateusage [55,56]. NADH fluorimetry is, however, a sensitiveindicator for the O balance during compromised O2 2

supply conditions.The recent development of an optical, non-invasive

technique now makes the determination of vascular oxygentension possible [9,57]. This technique is based on oxygen-dependent quenching of the phosphorescence of Pd-por- Fig. 1. Dependency of epicardial microvascular PO (mPO ) determined2 2

by the phosphorescence technique on oxygen consumption and supply.phyrin, a compound intravascularly administered to theThe mPO is determined on the surface of the left ventricle of an isolated2animal. Following excitation by a flash of green light, theLangendorff-perfused pig heart [59]. The heart was perfused at constant

measured half-life of the phosphorescence signal can be flow with a blood–Tyrode’s mixture (Hct620%) containing 220 mMquantitatively related to the oxygen tension, using proper Pd-porphyrin, gassed with room air /5% CO (arterial PO 5110 mm Hg).2 2

calibration of the probe [58]. Binding Pd-porphyrin to The coronary vasculature was maximally vasodilated with nitroprusside.(A) The mPO as a function of O consumption at constant O supplyalbumin has the advantages of confining the probe initially 2 2 2

(flow). O consumption was changed by varying the pacing frequency2to the vascular compartment and increasing the half-life offrom 100 to 180 beats per min (bpm) as indicated in the figure. (B) The

phosphorescence such that physiological oxygen tensionsmPO as a function of O supply at constant O consumption. O supply2 2 2 2

can be measured. We have shown that when this phosphor- was changed by varying the retrograde flow using a roller pump. Flow21 21escence PO is measured with optic fibres in the mi- was varied between 3.4 and 2.1 ml min g as indicated in the figure.2


Dow

nloaded from



|0.5 mm [62]). However, studies have shown that the These areas have been called microcirculatory weak unitsheterogeneous epicardial NADH pattern is characteristic of [71]. Using surface NADH fluorescence photography, inthe whole ventricle, for both small hearts [63] and large the isolated working rat heart, it has been observed thathearts [64,65]. In addition, these techniques may ideally be anoxic zones 100–300 mm wide developed during thesuited for the smaller heart size if most of the heart tissue progression of both hypoxia and ischemia [38]. Theneeds to be investigated, such as the mouse heart, and will authors suggested from the size of the anoxic regionsthus be an important analysis tool in the realm of molecu- (larger than intercapillary distances), that oxygen supply belar physiology. regulated at the level of the arterioles. However, Ince et al.

[67] concluded from the use of microspheres of different6.2. NADH fluorescence imaging during dysoxia sizes that the regulatory unit determining the patchy

pattern during compromised oxygen supply was located atThe first NADH fluorescence measurements demon- the level of the capillary (Fig. 2A and 2B). In an elaborate

strated the possibility of visualising dysoxic areas in study using an in situ rat heart preparation in combinationperfused rat hearts [53]. In a classical study [38] it was with quick freezing techniques and determination ofsubsequently shown that these high fluorescence, dysoxic NADH fluorescence in 5-mm thick sections, Vetterlein etareas, were heterogeneously distributed over the surface of al. showed that the size of the anoxic zones ranged from athe heart during high flow graded hypoxia, graded is- few myocytes to several hundred microns [70]. The size ofchemia and respiratory acidosis. In addition, the location the anoxic zones increased with further restriction of bloodand size of the dysoxic areas was similar during high flow flow through the left anterior coronary artery. It was foundhypoxia and ischemia (but not during acidosis) and did not that the dysoxic regions had more venous capillary seg-vary over a period of 15 min under constant flow con- ments than arterial capillary segments, as compared toditions. Repetitive periods of identical ischemia generate non-dysoxic regions. This data would indicate that dysoxiaidentical distribution patterns for a given heart [66]. It was mainly develops along the venous end of capillaries. Infurther demonstrated that the location of these patchy addition, it has been suggested by these authors [70] thatdysoxic areas was also identical during recovery from the large areas of organ surface NADH fluorescence thateither total global ischemia or high flow anoxia or tachy- have been observed [38] may be due to the interferencecardia for a given heart [67]. This heterogeneity in dysoxic from the fluorescence arising from layers beneath theareas has also been shown during inhibition of NO focused plane of observation, which would obscure thesynthesis in endotoxemic rat Langendorff-hearts [68], in visualisation of small areas of NADH fluorescence. Inhypertrophied rat Langendorff-hearts [69] and during summary, data suggests that hypoxia / ischemia result inhaemorrhagic shock in in situ pig hearts [65]. dysoxia, which starts at the level of the capillary [67,70],

Although NADH measurements have mostly been used and which occurs primarily at the venous end of theto evaluate dysoxia in compromised hearts, most studies capillary [70].have also reported that normoxic perfusion shows a Hypertrophy [69,72] and pH [38,72] have been shown torelatively homogeneous tissue fluorescence [38,66,67,70]. generate a different distribution and larger areas of highThese results should be cautiously interpreted due to the fluorescence than interventions using decreased oxygenlow sensitivity of NADH to the O balance during supply (Fig. 2C–F). Interestingly, superoxide dismutase2

normoxic conditions, but they do indicate matching of was able to reduce these large areas of high fluorescenceoxygen supply to oxygen consumption, in agreement with [72], suggesting that such large areas are determined by arecent measurements of regional oxygen consumption with vasoregulation mechanism at the level of the arterioles.flow [5,6,48,49]. This is in accordance with the observation that the larger

areas of fluorescence occur during embolization with6.3. Size and origin of the microcirculatory (weak) unit microspheres .10 mm which occlude arterioles [67]. Thus,

acidosis and hypertrophy are associated with dysoxia at theThe benefit of visualisation of indicators of the oxygen arteriole level, generating dysoxia in the relatively large

supply-consumption ratio is that the structure and size of area of the capillary bed of the arteriole.the dysoxic areas during the progress of dysoxia can beobserved continuously, and may thus provide insight into 6.4. Microvascular PO measurements2

the main determinant of dysoxia at each level of dysoxia(capillary ↔ arterioles, convection ↔ diffusion). An In the first report of applying the phosphorescenceinteresting observation is that the high NADH areas are technique to the heart, lifetime images of the phosphor-always in the same location, independent of whether escence signal were obtained through a microscope fromregional dysoxia was induced by hypoxia, ischemia or the epicardial surface of new-born piglets [73]. Theincreased atrial pacing [38,66,67]. This strongly suggests microvascular PO of 0.3-mg regions was 16.864.2 mm2

that certain areas within the heart are predisposed to Hg, at an arterial PO of 106 mm Hg. The coefficient of2

becoming the first dysoxic when oxygen supply is limited. variation (CV5SD/mean), which is an index of mPO2


Dow

nloaded from



Fig. 3. Anterior view of three open-chest pig hearts showing theheterogeneity of mPO on the epicardial surface, as determined by the2

phosphorescence technique. The mPO was measured in nine different22regions (|1 cm ) for each heart. The animals were ventilated with 33%

O , resulting in an arterial PO of 6180 mm Hg. Preparation as in van2 2

Iterson et al. [74,75]. Abbreviations used: RA5right auricle; LA5leftauricle; LAD5left anterior descending artery.

heterogeneity, was 25%, much smaller than the CVobtained for flow heterogeneity in such small tissuesamples [10]. In a study directed at resuscitation [74,75] ina model of shock in the pig, the mPO was determined2

with the phosphorescence technique by a fibre phosphor-2imeter [58,62] in nine different areas (|1 cm or 0.04–0.08

g of tissue) on the surface of the ventricles (Fig. 3). TheCV for these mPO values was between 11–12%, which2

again is considerably smaller than the CV of flow hetero-geneity (30%) for equally sized tissue samples [10]. Thus,the smaller CV’s indicate a matching of O supply with2

consumption within the heart. Very few studies have usedthe phosphorescence technique to evaluate myocardialmPO and its heterogeneity during conditions of restricted2

O supply. It has been shown, however, that during the2

transition from high-flow anoxia to normoxia, a veryheterogeneous mPO distribution occurs on the surface of2

the isolated rat heart [67]. The mPO patterns coincided2

with the NADH patterns that were also recorded, support-Fig. 2. Heterogeneous NADH fluorescence images of left ventricle areaof isolated Langendorff-perfused rat hearts, with apex of the heart always ing the notion that both NADH and mPO reflect a similar2at the bottom of the figure: (A) NADH fluorescence patterns elicited phenomenon, i.e. the O supply-consumption ratio.2during transition from anoxia (95% N ) to normoxia (95% O ) are, as2 2

shown in (B), the same as those elicited by embolization with 5.9-mmdiameter microspheres [66]. To aid comparison of the two images,

7. Outstanding issues concerning oxygen supply-corresponding areas in the two images are numbered. (C) Low homoge-neous NADH fluorescence during pH 7.5 perfusion, whereas large NADH consumptionfluorescence areas (indicated by arrows) are visible during pH 7.0perfusion (D). (E) and (F) show examples of hypertrophied hearts Various studies have shown that oxygen diffuses to cellsdemonstrating the large areas (indicated by arrows) of high NADH

from arterioles, capillaries and venules. Within the heart,fluorescence within 5 min after the start of perfusion [74]. (Figs. 3A andboth direct, but limited, data on oxygen diffusion from3B are reprinted with permission from Am J Physiol, Figs. 3C–F are

reprinted with permission from Biochim Biophys Acta). large vessels and indirect data on diffusional shunting


Dow

nloaded from



suggest that the release or uptake of oxygen by large and consumption is compared at different time points ofvessels is insignificant. However, more research examining the reduced flow condition, to study whether at some timethe PO along a considerable length of artery or arteriole point the heterogeneity of O supply-consumption ratio2 2

within the heart is needed to answer this question conclu- equals the heterogeneity observed during the controlsively. Furthermore, it also remains uncertain whether the condition.regions that show the first signs of dysoxia have anyrelation with the normal local level of flow, flow reserve,O diffusion or O consumption. Two recent studies have Acknowledgements2 2

shown that the resting normal, low and high blood flowregions are equally vulnerable to dysoxia during partial We gratefully acknowledge the skillful assistance ofcoronary stenosis [18,76], and that the relative flow Charles N.W. Belterman and Joris de Groot of the Depart-reduction during stenosis was independent of low or high ment of Experimental Cardiology of the University ofblood flow region [76]. This data is corroborated by the Amsterdam with the isolated blood-perfused pig heartobservation that no correlation exists between the intensity experiment.of NADH fluorescence and the distance to the nextperfused capillary [70], thus there seems to be no depen-dence of dysoxia on the absolute level of convective flow. ReferencesAssuming matching between flow and O consumption,2

this would also indicate equal vulnerability for regions of [1] Yipintsoi T, Dobbs WA, Scanlon PD, Knopp TJ, Bassingthwaightelow and high O consumption. Dysoxia was mainly JB. Regional distribution of diffusible tracers and carbonized2

microspheres in the left ventricle of isolated dog hearts. Circ Respredicted by the relative flow reduction upon stenosis [18],1973;33:573–587.implying a vasoregulatory (at the level of the arterioles or

[2] Hoffman JIE. Heterogeneity of myocardial blood flow. Basic Reshigher) mechanism as a primary determinant of dysoxia.Cardiol 1995;90:103–111.

However, other studies have shown that these dysoxic [3] Wieringa PA, Stassen HG, Van Kan JJIM, Spaan JAE. Oxygenareas can be best generated by obstructing capillaries, but diffusion in a network model of the myocardial microcirculation. Intnot arterioles, with microspheres [67]. It remains to be J Microcirc Clin Exp 1993;13:137–169.

[4] Weiss HR, Neubauer JA, Lipp JA, Sinha AK. Quantitative de-examined whether the obstructed capillaries mainly occurtermination of regional oxygen consumption in the dog heart. Circin the regions with the largest relative reductions in flow,Res 1978;42:394–401.

indicating dysoxia develops at the capillary level of those [5] Li Z, Yipintsoi T, Caldwell JH et al. Correlation between regionalareas. flow and metabolism in dog hearts in vivo. Ann Biomed Eng

The studies reviewed in this paper emphasise that it 1996;24 (suppl 1):S–28, abstract.[6] Van Beek JHGM. Intraorgan heterogeneity of blood flow, oxygenremains unknown whether a dependency may exist be-

consumption and tissue oxygenation. In: Sibbald WJ, Messmer K,tween dysoxia and oxygen supply by diffusion. AssumingFink MP, editors, Tissue oxygenation in acute medicine. update in

diffusive shunting of oxygen within the capillary bed, for intensive care and emergency medicine, Berlin, Heidelberg: Spring-which evidence is present, it may be hypothesised that er-Verlag, 1998, pp. 14–35.capillary regions where oxygen is shunted away from their [7] Chance B. Pyridine nucleotide as an indicator of the oxygen

requirements for energy-linked functions of mitochondria. Circ Resmicrocirculation become the first dysoxic. The observation1976;38(Suppl.):I31–I38.that the NADH fluorescent zones appear at higher venous

[8] Chance B. Mitochondrial redox state as a ‘Gold Standard’ of tissuePO during ischaemia as compared to high-flow hypoxia2 hypoxia. Adv Exp Med Biol 1994;361:226.[38] is in support of this hypothesis. Why certain capillary [9] Vanderkooi JM, Maniara G, Green TJ, Wilson DF. An opticalregions are more prone to diffusional shunting is unknown, method for measurement of dioxide concentration based upon

quenching of phosphorescence. J Biol Chem 1987;262:5476–5482.however, and may be related with capillary geometry[10] Bassingthwaighte JB, King RB, Roger SA. Fractal nature of regional(capillary length or number of cross-connections?).

blood flow heterogeneity. Circ Res 1989;65:578–590.The evidence in support of regional matching between [11] Van Bavel E, Spaan JAE. Branching patterns in the porcine coronary

oxygen supply and oxygen consumption within the heart is arterial tree. Estimation of flow heterogeneity. Circ Resrather strong. The question that arises is how such match- 1992;71:1200–1212.

[12] Bassingthwaighte JB, Malone MA, Moffett TC et al. Molecular anding occurs. One interesting explanation may be related toparticulate depositions for regional myocardial flows in sheep. Circthe recently observed adaptation of oxygen consumption toRes 1990;66:1328–1344.

extracellular oxygen tensions in hepatocytes and em- [13] Van Beek JHGM, Roger SA, Bassingthwaighte JB. Regionalbryonic cardiomyocytes [77,78]. Lowering the oxygen myocardial flow heterogeneity explained with fractal networks. Amtension in suspensions of isolated cardiomyocytes de- J Physiol 1989;257:H1670–H1680.

[14] Kassab GS, Yung YCB. Topology and dimensions of pig coronarycreased oxygen consumption within 5 min in these em-capillary network. Am J Physiol 1994;267:H319–H325.bryonic cells, offering a mechanism for regional matching

[15] Matsumoto T, Goto M, Tachibana H et al. Microheterogeneity ofof oxygen supply-consumption within the heart. This myocardial blood flow in rabbit hearts during normoxic and hypoxichypothesis may be tested in experiments directed at states. Am J Physiol 1996;270:H435–H441.hibernation: the relation between regional oxygen supply [16] Austin Jr. RE, Smedira NG, Squiers TM, Hoffman JIE. Influence of


Dow

nloaded from



cardiac contraction and coronary vasomotor tone on regional [42] Rose CP, Goresky CA, Bach GG. The capillary and sarcolemmalmyocardial blood flow. Am J Physiol 1994;266:H2542–H2553. barriers in the heart. An exploration of labelled water permeability.

[17] Wolpers HG, Hoeft A, Korb H, Lichtlen PR, Hellige G. Hetero- Circ Res 1977;41:515–533.geneity of myocardial blood flow under normal conditions and its [43] Zuurbier CJ, Yipintsoi T, Li Z, Deussen A, Bassingthwaighte JB.dependence on arterial PO . Am J Physiol 1990;258:H549–H555. High diffusional water conductance obtained by modelling water2

[18] Loncar R, Flesche CW, Deussen A. Coronary reserve of high- and transport in the heart using a distributive model. Ann Biomed Englow-flow regions in the dog heart left ventricle. Circ 1998;98:262– 1996;24 (suppl 1):S–44, abstract.270. [44] Beard DA, Bassingthwaighte JB. Advection and diffusion of oxygen

[19] Polansky L, Weiss HR. Effect of flow reduction on coronary blood in biological systems containing complex vascular networks. Annflow heterogeneity. Proc Sec Exp Biol Med 1993;202:97–102. Biomed Eng 1996;24 (suppl 1):S–40, abstract.

[20] Standa M, Iversen PO, FlatebØ T, Nicolaysen G. Regional hetero- [45] Yoshida S, Akizuki S, Gowski D, Downey JM. Discrepancygeneity of myocardial blood flow within the rabbit left ventricle between microsphere and diffusible tracer estimates of perfusion toduring haemorrhagic hypotension. J Physiol Pharmacol ischemic myocardium. Am J Physiol 1985;249:H255–H264.1992;43:209–218. [46] Fluck DS, Etherington PJE, O’Hare D, Winlove P, Sheridan DJ.

[21] Kleen M, Welte M, Lackermeier P et al. Myocardial blood flowMyocardial tissue perfusion determined by particulate and diffusible

heterogeneity in shock and small-volume resuscitation with coronarytracers during ischaemia: what is measured? Cardiovasc Res

stenosis. J Appl Physiol 1997;83:1832–1841.1996;32:869–878.

[22] Krogh A. The anatomy and physiology of capillaries, New York:[47] Simson MB, Harden W, Barlow C, Harken AH. Visualization of theHatner, 1959.

distance between perfusion and anoxia along an ischemic border.[23] Ellsworth ML, Ellis CG, Popel AS, Pittman RN. Role of mi-Circ 1979;60:1151–1155.crovessels in oxygen supply to tissue. News in Physiol Sci

[48] Sonntag M, Deussen A, Schultz J et al. Spatial heterogeneity of1994;9:119–123.blood flow in the dog heart. 1 Glucose uptake, free adenosine and[24] Duling BR, Berne RM. Longitudinal gradients in periarteriolar

¨oxidative /glycolytic enzyme activity. Pflugers Arch 1996;432:439–oxygen tension. Circ Res 1970;27:669–678.450.[25] Itoh T, Yaegashi K, Kosaka T, Kinoshita T, Morimoto T. In vivo

[49] Schipke JD, Arnold G, Schwanke U. Regional myocardial activityvisualization of oxygen transport in microvascular network. Am Jcorrelates with regional coronary flow. Proc Exp Biol 1998;A1028,Physiol 1994;267:H2068–H2078.abstract.[26] Sinaasappel M, Donkersloot C, Van Bommel J, Ince C. Measure-

[50] Connett RJ, Honig CR, Gayeski TEJ, Brooks GA. Defining hypoxia:ments in the rat intestinal microcirculation. Am J Physiola systems view of VO , glycolysis, energetics and intracellular PO .1999;276:61515–61520. 2 2

J Appl Physiol 1990;68:833–842.[27] Torres Filho IP, Kerger H, Intaglietta M. PO measurements in2

[51] Hearse DJ. Myocardial ischaemia: can we agree on a definition forarteriolar networks. Microvasc Res 1996;51:202–212.st[28] Shonat RD, Johnson PC. Oxygen tension gradients and hetero- the 21 century. Cardiovasc Res 1994;28:1737–1744.

geneity in venous microcirculation: a phosphorescence quenching [52] Chance B, Thorell B. Localization and kinetics of reduced pyridinestudy. Am J Physiol 1997;272:H2233–H2240. nucleotides in living cells by microfluorometry. J Biol Chem

[29] Ellsworth ML, Pittman RN. Arterioles supply oxygen to capillaries 1959;234:3044–3050.by diffusion as well as by convection. Am J Physiol [53] Barlow CH, Chance B. Ischemic areas in perfused rat hearts;1990;258:H1240–H1243. measurements by NADH fluorescence photography. Science

[30] Weiss HR, Sinha AK. Regional oxygen saturation of small arteries 1976;193:909–910.and veins in the canine myocardium. Circ Res 1978;42:119–126. [54] Coremans JMCC, Ince C, Bruining HA, Puppels GJ. (Semi-)

[31] Honig CR, Gayeski TEJ. Precapillary O loss and arteriovenous Quantitative analysis of reduced nicotinamide adenine dinucleotide2

diffusion shunt are below limit of detection in myocardium. Adv fluorescence images of blood-perfused rat heart. Biophys JExp Med Biol 1989;247:591–599. 1997;72:1849–1860.

[32] Kassab GS, Rider CA, Tang NJ, Fung Y-C. Morphometry of pig [55] Chapman JB. Fluorometric studies of oxidative metabolism incoronary arterial trees. Am J Physiol 1993;265:H350–365. isolated papillary muscle of the rabbit. J Gen Physiol 1972;59:135–

[33] Yipintsoi T, Bassingthwaighte JB. Circulatory transport of iodoan- 154.tipyrine and water in the isolated dog heart. Circ Res 1970;27:461– [56] Scholz TD, Laughlin MR, Balaban RS et al. Effect of substrate on477. mitochondrial NADH, cytosolic redox state, and phosphorylated

[34] Roth AC, Feigl EO. Diffusional shunting in the canine myocardium. compounds in isolated hearts. Am J Physiol 1995;268:H82–H91.Circ Res 1981;48:470–480. [57] Wilson DF, Pastuszko A, Digiacomo JE et al. Effect of hyperventi-

[35] Van Beek JHGM, Elzinga G. Diffusional shunting of oxygen in lation on oxygenation of the brain cortex of newborn piglets. J App¨saline-perfused isolated rabbit hearts is negligible. Pflugers Arch Physiol 1991;70:2691–2696.

1987;410:263–271. [58] Sinaasappel M, Ince C. Calibration of Pd-porphyrin phosphor-[36] Rose CP, Goresky CA. Limitations of tracer oxygen uptake in the escence for oxygen concentration measurements in vivo. J Appl

canine coronary circulation. Circ Res 1985;56:57–71. Physiol 1996;81:2297–2303.[37] Bassingthwaighte JB, Yipintsoi T, Knopp TJ. Diffusional arterioven- [59] Coronel R, Fiolet JWT, Wilms-Schopman FJG et al. Distribution of

ous shunting in the heart. Microvasc Res 1984;28:233–253. extracellular potassium and its relation to electrophysiologic changes[38] Steenbergen C, Deleeuw G, Barlow C, Chance B, Williamson JR. during acute myocardial ischaemia in the isolated perfused porcine

Heterogeneity of the hypoxic state in perfused rat heart. Circ Res heart. Circ 1988;77:1125–1138.1977;41:606–615. [60] Hobbhahn J, Conzen PFM, Goetz A et al. Myocardial surface

[39] Wolpers HG, Hoeft A, Korb H, Lichtlen PR, Hellige G. Transport of PO -an indicator of myocardial tissue oxygenation? Cardiovasc Res2

inert gases in mammalian myocardium: comparison with a convec- 1989;23:529–540.tion–diffusion model. Am J Physiol 1990;259:H167–H173. [61] Chance B, Barlow C, Nakase Y et al. Heterogeneity of oxygen

[40] Yudilevich DL, Alvarez OA. Water, sodium and thiourea trans- delivery in normoxic and hypoxic states: a fluorometer study. Am Jcapillary diffusion in the dog heart. Am J Physiol 1967;213:308– Physiol 1978;235:H809–H820.314. [62] Sinaasappel M, van Iterson M, Ince C. Microvascular oxygen

[41] Bergman SR, Fox KA, Rand AL et al. Quantification of regional pressure in the pig intestine during haemorrhagic shock and resusci-15myocardial blood flow in vivo with H O. Circ 1984;70:724–733. tation. J Physiol 1999;514:245–253.2


Dow

nloaded from



[63] Eerbeek O, Ince C. Weak spots inside the myocardium of a [72] Hulsmann WC, Ashruf JF, Bruining HA, Ince C. Imminent ischemialangendorff rat heart observed by NADH videofluorometry. Adv in normal and hypertrophic Langendorff rat hearts; effects of fattyExp Med Biol 1997;411:347–351. acids and superoxide dismutase monitored by NADH surface

[64] Harken AH, Barlow CH, Harden WR, Chance B. Two and three fluorescence. Biochim Biophys Acta 1993;1181:273–278.dimensional display of myocardial ischemic ‘border zone’ in dogs. [73] Rumsey WL, Pawlowski M, Lejavardi N, Wilson DF. OxygenAm J Cardiol 1978;42:954–959. pressure distribution in the heart in vivo and evaluation of the

[65] Miyazaki K, Hori S, Inoue S et al. Characterization of energy ‘ischemic border’. Am J Physiol 1994;266:H1676–H1680.metabolism and blood flow distribution in myocardial ischemia in [74] Van Iterson M, Sinaasappel M, Burhop K, Trouwborst A, Ince C.haemorrhagic shock. Am J Physiol 1997;273:H600–H607. Low volume resuscitation with a haemoglobin-based oxygen carrier

[66] Harden WR, Barlow CH, Simson MB, Harken AH. Heterogeneity of after hemorrhage improves gut microvascular oxygenation in swine.the coronary microcirculation during low-flow ischemia: A model J Lab Clin Med 1998;132:421–431.for the heart in shock. Adv Shock Res 1980;3:239–250. [75] Iterson van M, Siegemund M, Dries J, Trouwborst A, Ince C.

[67] Ince C, Ashruf JF, Avontuur JA et al. Heterogeneity of the hypoxic Microvascular oxygenation of the heart and gut during haemorrhagestate in rat heart is determined at capillary level. Am J Physiol in pigs. Int Care Med 1998;24(suppl 1):190, abstract.1993;264:H294–H301. [76] Bussemaker J, Groeneveld ABJ, Teerlink T et al. Low- and high-

[68] Avontuur JAM, Bruining HA, Ince C. Inhibition of nitric oxide blood flow regions in the normal pig heart are equally vulnerable to¨synthesis causes myocardial ischemia in endotoxemic rats. Circ Res ischaemia during partial coronary stenosis. Pflugers Arch

1995;76:18–425. 1997;434:785–794.[69] Ashruf JF, Ince C, Bruining HA, Hulsmann WC. Ischemic areas in [77] Schumacker PT, Chandel N, Agusti AGN. Oxygen conformance of

hyperthropic Langendorff rat hearts visualized by NADH videofl- cellular respiration in hepatocytes. Am J Physiol 1993;265:L395–uorimetry. Adv Exp Med Biol 1994;345:259–262. L402.

[70] Vetterlein F, Prange M, Lubrich D et al. Capillary perfusion pattern [78] Budinger GRS, Chandel N, Shao ZH et al. Cellular energy utiliza-and microvascular geometry in heterogeneous hypoxic areas of tion and supply during hypoxia in embryonic cardiac myocytes. Amhypoperfused rat myocardium. Am J Physiol 1995;268:H2183– J Physiol 1996;270:L44–L53.H2194.

[71] Ince C, Sinaasappel M. Microcirculatory oxygenation and shuntingin sepsis and shock. Crit Care Med 1999: in press.


Dow

nloaded from


Functional heterogeneity of oxygen supply-consumption ratio in the heart

Documents