comparative effects of hypoxia and ischemia in the...

121

Comparative Effects of Hypoxia and Ischemiain the Isolated, Blood-Perfused Dog Heart:

Evaluation of Left Ventricular DiastolicChamber Distensibility and Wall Thickness

R. Michael Wyman, Eli R. Farhi, Oscar H.L. Bing, Robert G. Johnson,

Ronald M. Weintraub, and William Grossman

To compare the effects of hypoxia and ischemia on left ventricular (LV) diastolic function, westudied 17 isolated, isovolumic dog hearts by measuring LV diastolic chamber distensibility (LVend diastolic pressure at constant volume), wall thickness, and myocardial pH in response tohypoxia at constant coronary flow or pressure versus global ischemia (zero coronary bloodflow). Hypoxic perfusates consisted of methemoglobin-containing red blood cells suspended inlactated Ringer's solution. Brief cross-damping of the coronary perfusion line was used to assessthe contribution of coronary turgor to chamber distensibility and wall thickness. With hypoxia,left ventricles showed a significant early (5 minutes) decrease in diastolic distensibility and anincrease in wall thickness, at either constant coronary perfusion pressure or flow. The increasein wall thickness was independent of hypoxia-induced changes in coronary turgor. In contrast,global ischemia produced an early increase in LV diastolic chamber distensibility and a decreasein wall thickness. When global ischemia was continued beyond 60 minutes, a decrease in LVchamber distensibility developed. This diastolic contracture was not associated with an increasein LV wall thickness. Myocardial pH decreased slightly during 15 minutes of hypoxia andmarkedly with 15 minutes of global ischemia. Thus, LV diastolic chamber distensibility decreasedduring 15 minutes of hypoxia, while an increase in distensibility was seen during global ischemiaof similar duration. During hypoxia, these changes were associated with increased LV wallthickness, at either constant coronary perfusion pressure or constant coronary flow. Prolongedischemia led to diastolic contracture without an increase in wall thickness. The differences in wallthickness with hypoxia versus ischemk contracture suggest differing mechanisms for these twotypes of diastolic dysfunction. (Circulation Research 1989;64:121-128)

Left ventricular (LV) diastolic dysfunction isnow a well-documented explanation for theclinical and hemodynamic observations of

increased filling pressures during angina. 1~9 In exper-imental preparations, alterations in both myocardialrelaxation and left ventricular diastolic distensibil-ity are seen during demand ischemia10-12 andhypoxia.11-15 However, the pathophysiological mech-

From the Charles A. Dana Research Institute and the Harvard-Thomdike Laboratory of Beth Israel Hospital, and the Depart-ments of Medicine (Cardiovascular Division) and Surgery, BethIsrael Hospital and Harvard Medical School, Boston, Massachu-setts.

Supported in part by National Institutes of Health GrantHL-28939.

Address for reprints: William Grossman, MD, CardiovascularDivision, Beth Israel Hospital, 330 Brookline Avenue, Boston,MA 02215.

Received April 14, 1988; accepted July 6, 1988.

anisms underlying these acute changes are contro-versial and appear to involve a complex interplaybetween mechanical (e.g., geometric, pericardial,or coronary turgor) and intracellular (e.g., pro-longed and persistent contractile element interac-tion) factors. In fact, ischemia and hypoxia do nothave uniformly concordant effects on these poten-tial mechanisms in an experimental setting, partic-ularly in the isolated heart model.16 While hypoxiccrystalloid perfusion of an isolated heart is associ-ated with slowed relaxation, decreased LV cham-ber distensibility, and increased coronary turgordue to vasodilation,1314 ischemia produced by areduction in coronary blood flow may have quitedifferent effects. Acute low-flow ischemia in theabsence of increased myocardial oxygen demanddecreases coronary turgor and increases diastolicdistensibility, with an associated increase in localaccumulation of metabolites and hydrogen ion.14-17-18

by guest on June 22, 2018http://circres.ahajournals.org/

Dow

nloaded from

http://circres.ahajournals.org/

122 Circulation Research Vol 64, No 1, January 1989

More prolonged (30-60 minutes) global ischemiacauses an essentially irreversible "rigor" state ofthe myocardium, the cellular mechanism of which isuncertain.19-20 It has been suggested21 that the patho-physiological basis of these reversible and irrevers-ible types of diastolic dysfunction is related todecreased sequestration of calcium by sarcoplasmicreticulum (SR), changes in coronary turgor, changesin myocardial pH, and actin-myosin rigor bondformation. The purpose of the present study is toexamine in further detail the relation between acutehypoxia and global ischemia, particularly withregards to changes in left ventricular wall thickness,an important determinant of diastolic function thathas received little attention in the aforementionedstudies. We performed a series of experiments in acanine isovolumic, isolated heart preparation withsimultaneous assessment of LV diastolic distensi-bility and wall thickness changes in response toacute hypoxia, global ischemia, and changes incoronary turgor. Our results support the conceptthat there is an inherent mechanistic differencebetween the diastolic dysfunction of acute hypoxiaand prolonged ischemic contracture.

Materials and MethodsIsolated Heart Preparation

Isolated perfused hearts from 17 mongrel dogswere prepared as previously described22 and shownin Figure 1. Briefly, in each experiment a supportdog was anesthetized with chloralose and urethaneand placed on mechanical ventilation. Catheterswere placed in the femora! artery and vein, and theanimal was anticoagulated with heparin (500 units/kg i.v.). The heart from a second dog was preparedby insertion of cannulas into the right ventricle andthe carotid and subclavian arteries; the heart wasthen isolated by ligation of the superior and inferiorvenae cavae and descending thoracic aorta and wasremoved from the thorax. Coronary perfusion of theisolated heart was maintained by retrograde perfu-sion of the coronary arteries, via the carotid cannula,with blood obtained from the femoral artery of thesupport dog and circulated by a roller pump througha bubble trap, flowmeter, and heat exchanger tomaintain blood temperature at 37° C. Coronary venousblood was drained from the right ventricle (in orderto minimize ventricular interactive effects) throughanother flowmeter into the femoral vein of the sup-port dog. A left ventricular drain was inserted througha stab wound in the LV apex to drain Thebesian veinflow, and an intramyocardial thermistor was placedadjacent to it in the apex.

Ventricular fibrillation was then induced in theisolated perfused heart, and the left atrium wasopened, Ultrasonic crystals were placed across theLV free wall, avoiding papillary muscles, to allowdirect measurement of wall thickness. An inflatablelatex balloon was placed in the left ventricle throughthe mitral anulus and secured with sutures in the

Perfusion linePressure Transducer

RV Drain

Pressure Transducer

Balloon Mount

LV Drain

FIGURE 1. Isolated heart preparation. Retrograde cor-onary perfusion is via the cannulated carotid artery. Thepulmonary artery (PA) is ligated and a right ventricular(RV) drain inserted to collect coronary venous effluentand minimize ventricular interactive effects. The leftventricular (LV) latex balloon is secured through themitral anulus and is connected via fluid filled tubing to apressure transducer and syringe. The apical LV draincollects Thebesian drainage. Ao, aorta.

anulus. A balloon size was chosen in each case suchthat balloon volume would be greater than LVdiastolic volume, and thus pressure incrementswould reflect LV rather than balloon tensionchanges. In eight dogs, a hydrogen-ion selectivepolymer membrane pH electrode (Life-Span 100TM Ph monitor, Biochem International, Waukesha,Wisconsin) was implanted in the anterior wall suben-docardium to permit continuous measurement ofintramyocardial pH. The heart was then submergedin a blood bath, and any remaining air was removedfrom the LV cavity. Following ventricular defibril-lation and reinstitution of sinus rhythm, the isolatedheart was allowed to equilibrate for 30 minutes at acoronary perfusion pressure of 100 mm Hg and aheart rate of approximately 120 beats/min (main-tained by atrial pacing if necessary) at 37° C.

Control MeasurementsTo evaluate LV systolic and diastolic function,

the LV balloon was inflated with saline until asystolic pressure of approximately 100 mm Hg wasgenerated with each contraction. To allow compar-isons between hearts of different sizes, this balloonvolume (measured in milliliters) was arbitrarilyassigned a value of 100%, and subsequent measure-ments are reported as percentages of this volume.Since left ventricular contractions are isovolumic in


Dow

nloaded from


Wyman et al Diastolic Wall Thickness During Hypoxia and Ischemia 123

m

30

FIGURE 2. The coronary turgor or erectile effect on left ventricular (LV) diastolic pressure. LV pressure, perfusionpressure, and coronary flow are shown at baseline and during transient crossclamping of the aorta (mark). Coronaryperfusion pressure falls rapidly, and coronary flow ceases. LV diastolic pressure decreases 6 mm Hg (the "erectileeffect"), to 18 mm Hg (the "residual component").

this model, changes in LV end-diastolic pressure(EDP) directly reflect changes in diastolic distensi-bility. We prefer the term diastolic distensibility tothe more traditional diastolic compliance. The latterrefers strictly to changes in the slope of the pressure-volume relation, while changes in distensibility indi-cate only that a higher diastolic pressure is neededto fill the ventricle to the same volume.

After the 30-minute equilibration period, weobtained two sets of baseline measurements ofsystolic and diastolic pressure, as well as LV wallthickness (at 100% volume), coronary blood flow,and myocardial pH, at 25%, 50%, 75%, and 100% ofthe baseline volume. The coronary perfusion linewas then clamped for 30 seconds (at 75% balloonvolume) to determine the coronary turgor or erec-tile contribution to diastolic distensibility.13 An exam-ple of a determination of the turgor, or erectile,contribution to LVEDP is shown in Figure 2, whichdemonstrates the fall in LVEDP associated with areduction of the coronary perfusion pressure. Atthe mark, the perfusion line was clamped, andcoronary perfusion pressure fell until coronary flowceased. This decrease in pressure and flow wasassociated with a drop in the LV diastolic pressure,in this example, from 24 to 18 mm Hg. We desig-nated 18 mm Hg the residual component of diastolicpressure, as opposed to the erectile (or coronaryturgor) component, in this case the 6 mm Hg drop inpressure seen during flow cessation.

InterventionsRepeat control measurements were obtained at

least 5 minutes after restoration of coronary blood

flow. The perfusion system was then switched to areservoir containing the perfusate described below,which was perfused either at constant coronaryperfusion pressure or at constant coronary flow.Coronary flow and perfusion pressure, as well asLV diastolic pressure-volume curves and wall thick-ness, were recorded at 5, 10, and 15 minutes ofperfusion. After the 10-minute curve, the coronaryperfusion line was clamped once again for 30 sec-onds to determine the erectile contribution to dia-stolic pressure and wall thickness at 75% volume.After the 15-minute curve, the perfusion systemwas switched back to oxygenated blood from thesupport dog. The isolated heart was allowed 30minutes to recover, and the experimental protocolwas then repeated at constant coronary perfusionpressure or flow, whichever had not been usedpreviously. The sequence of perfusate administra-tion was varied randomly between experiments.After a second recovery period, global ischemiawas produced by clamping of the coronary perfu-sion line, and measurements of left ventricularpressure, wall thickness, and myocardial pH weremade at 1, 15, 30, 45, 60, and 75 minutes.

Two different myocardial perfusates were used inthese experiments. During control and recoveryperiods, the coronary perfusate was arterial bloodfrom the support dog. The hypoxic perfusate con-sisted of methemoglobin-containing red blood cellssuspended in lactated Ringer's solution to obtain ahematocrit of 20%. The preparation of thesemethemoglobin-containing red blood cells has been


Dow

nloaded from



WALLTHICKNESS

(mm)

WALLTHICKNESS

(mm)

8.6*±1.18.5*

±1.1

8.2*±10

67±0 9

3U

40

30

20

10

Q

-B ri i i

17.3 •±0.8

172*±0.9

17 2 *±0.9

16.7±08

25% 50% 75%

LV VOLUME

IOO% SO% 75%

LV VOLUME

FIGURE 3. Left ventricular end-diastolic pressure (LVEDP)-volume relations during control(•) and at 5 (A), 10 (O), and 15(A) minutes of hypoxia. End-diastolic volume is expressedas a percentage of maximal bal-loon volume. At constant coro-nary perfusion pressure Qeftpanel), there is decreased LVdiastolic chamber distensibilityat 5 minutes, with progressivechanges at 10 and 15 minutes.End diastolic wall thickness at100% volume (to the right ofeach curve) increases progres-sively. At constant coronaryflow (right panel), there is aless marked, but significant,decrease in distensibility andincrease in wall thickness.(*p<0.05 vs. control).

described previously.23 Briefly, washed red bloodcells were placed in 0.02 M sodium nitrite (NaNO^)to oxidize all the hemoglobin to methemoglobin. Thesecells were then washed four times with normal salineto remove the NaN02 and were suspended in lactatedRinger's.

Relative coronary blood flow was measured withan in-line electromagnetic flow probe (BiotronexLaboratory, Kensington, Maryland).

Statistical AnalysisAll data are expressed as the mean±SEM. Com-

parisons of paired data were made using Student's ttest. Analysis of variance was used when compar-ing three or more conditions in the same heart.

ResultsHypoxic Perfusion

As seen in Figure 3, hearts perfused withmethemoglobin-containing hypoxic red blood cellssuspended in lactated Ringer's solution at constantcoronary perfusion pressure showed a significantdecrease in LV diastolic chamber distensibility(LVEDP at 100% volume increased from 14±2 to46±5 mm Hg at 15 minutes of hypoxia, p<0.01)associated with a significant increase in LV wallthickness (16.7±0.9 to 18.6±1.1 mm, p<0.01).Both of these changes were apparent as early as 5minutes into hypoxia. Changes in end diastolicpressure relative to volume were much more prom-inent at higher diastolic volumes: 25% and 50%volumes showed no significant increase in LVEDPduring hypoxia.

Cessation of coronary flow resulted in a signifi-cant drop in both LVEDP and wall thickness (h)during the control period (the erectile effect), asshown in Figure 4. In addition, the erectile effect on

distensibility was greater during hypoxia than dur-ing the control period (LVEDP, 2±0.4 to 11 ±2mm Hg, p<0.01). The "residual" component ofLV-diastolic distensibility (LVEDP-ALVEDP)was also significantly altered compared with control(LVEDP at 75% volume 6±1 to 10±2 mm Hg,p<0.05). The erectile effect on wall thickness (i.e.,the decrease in wall thickness induced by coronaryflow cessation) did not change significantly duringhypoxia (h, 0.4±0.1 to 0.5±0.3 mm, p=NS). Thusthe increase in wall thickness during hypoxia atconstant pressure was not related to changes incoronary turgor but rather was due solely to a"residual" effect (h-Ah). During reoxygenation,LVEDP, the erectile effect on LVEDP, and wallthickness all returned to baseline.

Perfusion with the same solution at constantcoronary flow necessitated reduction in coronaryperfusion pressure to 41 ±10 mm Hg. Even withdecreased perfusion pressure, however, this hypox-ic perfusate caused a significant decrease in dia-stolic chamber distensibility (LVEDP at 100% vol-ume, 13± 1 to 29±6 mm Hg, p<0.01) and increase inwall thickness (16.7±0.8 to 17.3±0.8 mm, p<0.05),as shown in Figure 3. Again, these changes occurearly and predominantly at larger ventricular vol-umes. The erectile effect, although significant dur-ing both oxygenated and hypoxic perfusion periods,did not significantly influence the decrease in dis-tensibility (ALVEDP, 2±0.4 to 5±2 mm Hg, p=NS)or increase in wall thickness (Ah, 0.4±0.2 to 0.4±0.1mm) seen with hypoxia (Figure 4). In fact, the"residual" component of diastolic distensibility andwall thickness at constant flow is not significantlydifferent from that at constant pressure.

Global IschemiaAbrupt cessation of coronary flow resulted in an

immediate increase in diastolic distensibility (LVEDP


Dow

nloaded from


30 r

xE

20

15

10

Wyman et al Diastolic Wall Thickness During Hypoxia and Ischemia 125

30

25

-119

5EE

20

i s r - I

io17 I

_ i

I 19

18

17 •

00tuzoI

Control Hypoxia Recovery

LVEDP AT75% VOLUME

Control Hypoiia Recovery

END DIASTOLICWALL THICKNESSAT 75% VOLUME

1 . 1 ERECTILE| | RESIDUAL


LVEDP AT7 5 % VOLUME


END DIASTOLICWALL THICKNESSAT 75% VOLUME

FIGURE 4. The effect of coronary turgor on left ventricular end diastolic pressure (LVEDP) and end diastolic wallthickness during control and 10 minutes of hypoxia. Shaded areas represent the change in LVEDP or wall thickness uponcoronary vascular collapse ("erectile effect"). At constant perfusion pressure (left panel), there is a marked increase inthe erectile contribution to LVEDP during hypoxia although the "residual" LVEDP is still elevated compared withcontrol. Wall thickness decreases significantly during cross clamping, but there is no increase in the erectile contributionto wall thickness during hypoxia. During reoxygenation, all parameters (distensibility, erectile effect on distensibility,wall thickness) return to baseline. At constant coronary flow (right panel), the erectile contribution is not significantlyaffected by hypoxia for either LVEDP or wall thickness. (*p<0.05 vs. control).

at 100% volume fell from 19±3 to 12±2 mm Hg,p<0.05) and decrease in LV wall thickness (17.1 ± 1.6to 16.4± 1.6 mm, p<0.05) in the isolated heart (Figure5). After prolonged ischemia (>60 minutes), how-ever, diastolic distensibility decreased substantially(LVEDP at 100% volume, 54±8 mm Hg at 75minutes). In contrast, LV wall thickness did notchange during this time period, never significantlyincreasing even over the immediate postischemiclevel. Thus, although similar decreases in distensibil-ity occurred over time during this ischemic contrac-ture and during acute hypoxia, an analogous increasein wall thickness did not occur.

Relation Between Intramyocardial pH and LVDiastolic Distensibility

Figure 6 demonstrates the changes in intramyo-cardial pH during hypoxia and global ischemia.Hypoxic coronary perfusion (with methemoglobin-containing red blood cells) at either constant coro-nary perfusion pressure or flow was associated witha decrease in intramyocardial pH of approximately0.2 units, while global ischemia was associated witha pH decrease approximately twice as great. At 15minutes of hypoxic perfusion at constant coronaryperfusion pressure, LVEDP at 100% volume hadrisen considerably while intramyocardial pH hadfallen from 7.01+0.13 to 6.82±0.15. In contrast, at15 minutes of global ischemia, a significantly greaterfall in intramyocardial pH (6.91 ±0.11 to 6.55±0.8)

was associated with a decrease in LVEDP at 100%volume (18±3 to 11±1 mm Hg).

DiscussionThe pathophysiological mechanisms underlying

increased left ventricular diastolic pressure relativeto volume during angina pectoris have proven dif-ficult to define. The complex interaction of suchfactors as normal versus ischemic wall segment,pericardial constraint, and right ventricular pres-sure changes have led to the use of experimentalmodels to study the theoretically "pure" myocar-dial effects of ischemia. The effects of global isch-emic injury on the isolated heart have been found todiffer substantially from hemodynamic changes seenduring angina pectoris in man.171824 Global isch-emia causes either no change or an increase in LVdiastolic distensibility acutely, with a subsequentstiffening of the left ventricle only after 30 to 60minutes of continued severe ischemia. This latediastolic stiffening is thought to be due to markedlyreduced ATP levels20 and perhaps actin-myosinrigor bond formation. Whether this late "contrac-ture" development of the isolated heart is mecha-nistically similar to or distinct from diastolic dys-function during angina pectoris in man'-9 is unclear.

In our experiments, the use of hypoxic perfusatesin the isolated heart produced a much earlier changein diastolic function than did global ischemia. Thistemporal sequence is more analogous to the changesseen in man1-9 and in the dog coronary stenosis


Dow

nloaded from



WALLTHICKNESS

(mm)

16711.3

I61±l 617 I ± I 6

I6 4 ± I 61644:1.6

23% 50% 75% 100%

LV VOLUME

FIGURE 5. The effect of global ischemia on left ventric-ular end-diastolic distensibility and wall thickness. • ,control; A, / minute of global ischemia; O, 30 minutes; A,60 minutes; a, 75 minutes. Aortic cross-clamping pro-duces an early increase in distensibility (downward shiftof the pressure-volume relation) and decrease in wallthickness (end diastolic, measured at 100% volume, toright of each curve). A decrease in distensibility occursonly after 60 minutes of global ischemia ("ischemiccontracture") and is not associated with a significantincrease in wall thickness. (*p<0.05 vs. control).

model,10 wherein LVEDP is seen to rise virtuallycoincident with the onset of angina and ischemia.While hypoxia did not produce this immediate (i.e.,within several beats) rise in LVEDP, there was agradual and sustained increase to the physiologi-cally and statistically significant level noted at 5minutes. The reason for this less abrupt change maybe related to the differences in oxygen supply anddemand. During pacing-induced angina, and in thedog coronary stenosis-pacing model, there is both a

decrease in myocardial oxygen supply and anincrease in myocardial oxygen demand. Duringglobal hypoxia, there is a decreased oxygen supply,yet there is no increase (and perhaps even adecrease, due to a fall in left ventricular systolicpressure) in demand. This may explain the moregradual time course of changes in distensibility.

Other investigators13-13 have demonstrated therelatively rapid effects of hypoxia on diastolic dis-tensibility as well. While a hypothetical mechanismfor this effect involves decreased sequestration ofintracellular calcium by the SR, and thus slow and/or incomplete relaxation,9-12 there are again othercomplicating factors involved, most notably that ofcoronary turgor. The effects of variations in cor-onary pressure and flow on diastolic chamber dis-tensibility (the "erectile effect") have been studiedin some d e t a i l . - Specifically, the increase incoronary turgor associated with hypoxia-inducedvasodilation plays a significant role in compliancechanges. Global ischemia, by definition, involvesthe loss of such turgor, thus providing at leastpartial explanation for the observed acute increasein distensibility. Yet another factor to be consid-ered in these preparations is that of intracellularhydrogen ion accumulation. The increase in intra-cellular H associated with low-flow ischemia could,via effects on myofilament-Ca binding sensitivity,lead to a blunting of the diastolic stiffening seenwith hypoxia alone, since flow is maintained dur-ing hypoxic perfusion and H is washed out.

Thus, intracellular calcium changes, coronaryturgor, and pH are all potential factors that mightexplain the differences between hypoxia and isch-emic contracture. However, another significant fac-tor in diastolic functional abnormalities that hasreceived little attention in the analysis of hypoxicversus ischemic alterations is left ventricular wallthickness. Wall thickness has been shown to be animportant determinant of left ventricular diastolicstiffness and pressure, independent of LV mass orthe presence of LV hypertrophy.32 While wall thick-ness is known to decrease acutely with global

7.2

7.0Io._) 6.8

£ 6.6o° 64Z

2 6.2z~ 6.0

5.8

o:

FIGURE 6. Changes in intramyocardial pH dur-ing hypoxia and global ischemia (n=8). Intra-myocardial pH falls slightly during 15 minutes ofhypoxia at both constant coronary perfusionpressure (A) and flow (•), while at 15 minutes ofglobal ischemia (O) there is a marked drop inpH, with a further decline during more pro-longed ischemia. (*p<0.05 vs. control).

10 30 6 0

TIME (mln)


Dow

nloaded from


Wyman et al Diastolk Wall Thickness During Hypoxia and Ischemia 127

ischemia,14 direct measurements of thickness withhypoxia at varying coronary flows (including tran-sient vascular collapse), and the comparison of thiswith prolonged ischemia, have not been reportedpreviously.

In our studies hypoxic perfusion led to an earlyand sustained decrease in LV diastolic chamberdistensibility at both constant coronary pressureand flow. Wall thickness increased significantly andsimultaneously with the decrease in distensibility.Direct measurement of LV wall thickness in thepresence and absence of coronary turgor demon-strated that the increase in diastolic wall thicknessduring hypoxia, at both constant pressure and flow,was independent of hypoxia-induced changes in thevascular compartment; that is, the increase in wallthickness was due primarily to a "residual" myo-cardial effect. In addition, wall thickness changeswere reversible upon reoxygenation.

Global ischemia in our model produced a fall inLVEDP at 100% volume, with increases in LVEDPdeveloping only after prolonged (>60 minutes) globalischemia. Wall thickness, similarly, decreased imme-diately upon aortic cross-clamping. However, in-creases in wall thickness did not parallel the subse-quent increases in LVEDP, as had been the casewith hypoxia. At 75 minutes of ischemia, whendiastolic contracture had produced marked stiff-ening of the ventricle, wall thickness had not in-creased to control values.

Thus, in addition to coronary turgor and pH, wallthickness, a significant determinant of diastolic func-tion, is affected quite differently by acute hypoxiaand global ischemia. The mechanism of these dif-ferences is not entirely clear. Wall thickness in theabsence of the pericardium is determined by vascu-lar, myocardial cellular, and extracellular compo-nents. We have demonstrated that hypoxia-inducedchanges in the vascular compartment, while signif-icant, do not effect the changes in wall thickness.Myocardial cellular changes in hypoxia and isch-emia could conceivably be due to cell swelling and/or interstitial edema3334 or to incomplete inactiva-tion of systolic tension development. As notedpreviously, persistent actin-myosin interaction intodiastole may account for the changes in relaxationand chamber compliance seen during acute hypoxiaand prolonged ischemia. How might wall thicknessdifferences between the two be explained in accord-ance with this hypothesis? The mechanism of rigorformation during prolonged ischemia is presumablydue to severe ATP depletion,20 but whether thisresults in decreased removal of calcium from thetroponin-tropomyosin complex or in direct bindingof actin to myosin is unclear. The latter has beendocumented in skeletal muscle,33 and isolated heartexperiments employing a "quick stretch" to tem-porarily relieve ischemic contracture have sug-gested this mechanism as well.21"36 The cause ofdiastolic stiffening in hypoxia, on the other hand,has been attributed to impairment of calcium uptake

by the SR, resulting in increased availability ofcalcium to the myofilament apparatus, with persis-tent cross-bridge cycling.1537 These differences inintracellular mechanisms of contracture could con-ceivably lead to differences in wall thickness, withcontinued cross-bridge cycling causing a prominentincrease in wall thickness (analogous to systolicwall thickening). On the other hand, direct actin-myosin bond formation, without the physiologicalcross-bridge cycling that is associated with normaltension development, might result in a static wallthickness even with markedly increased myocardialstiffness. Additional support for this concept comesfrom studies on papillary muscle contracture,38-39 inwhich changes in the stiffness-tension relation wereshown to be linearly related to resting force, sug-gesting that the type of stiffness characterized bythis contracture was different from that participat-ing in active force development.

Changes in the extracellular matrix could alsoaccount for wall thickness differences. Although therole of edema in the physiological setting of isch-emia and hypoxia is controversial,40 interstitialedema could conceivably result in increased wallthickness during hypoxia (especially in the settingof a perfusate with decreased oncotic pressure).However, as noted above, whether the rapid incep-tion and resolution of compliance changes in thisand other studies could be the result of edemaformation is unclear. Of interest, other changes inthe extracellular matrix (primarily collagen struc-ture and cross-linking) have recently been noted inassociation with repetitive bursts of ischemia andreperfusion41 although it is uncertain whetherchanges of this nature could occur with brief epi-sodes of hypoxia, such as the 5-15-minute hypoxicperfusion used in this study.

In summary, our studies on diastolic function inthe isolated heart model have demonstrated that thehypoxia-induced decreases in left ventricular cham-ber distensibility are associated with a substantialincrease in end-diastolic wall thickness that is inde-pendent of changes in coronary turgor. In addition,there are important differences in the time courseand degree of wall thickening during acute hypoxiaand prolonged ischemia. These results support theconcept that there is a fundamental mechanisticdifference between the diastolic dysfunction of acutehypoxia and prolonged ischemia in this preparation,despite their similar effects on LV diastolic pressurerelative to volume.

AcknowledgmentThe authors wish to acknowledge Alvin Franklin

for his expert technical assistance in the perfor-mance of this experimental protocol.

References1. McLaurin LP, Rolett EL, Grossman W: Impaired left ven-

tricular relaxation during pacing induced ischemia. Am JCardiol 1973;32:752-757


Dow

nloaded from



2. Barry WH.BrookerJZ, Alderman EL, Harrison DC: Changesin diastolic stiffness and tone of the left ventricle duringangina pectoris. Circulation 1974;49:255-263

3. Mann T, Brodie BR, Grossman W, McLaurin LP: Effect ofangina on the left ventricular diastolic pressure-volumerelationship. Circulation 1977 ;55:761-766

4. Bourdillon PD, Lorell BH, Mirsky I, Paulus WJ, Wynne J,Grossman W: Increased regional myocardial stiffness of theleft ventricle during pacing-induced angina in man. Circula-tion 1983;67:316-323

5. Aroesty JM, McKay RG, Heller GV, Royal HD, Als AV,Grossman W: Simultaneous assessment of left ventricularsystolic and diastolic dysfunction during pacing-inducedischemia. Circulation 1985;71:889-900

6. Carroll JD, Hess OM, Hirzel HO, Krayenbuehl HP: Exercise-induced ischemia: The influence of altered relaxation onearly diastolic pressures. Circulation 1983;67:521-527

7. Carroll JD, Hess OM, Krayenbuehl HP: Diastolic functionduring exercise-induced ischemia in man, in Grossman W,Lorell BH (eds): Diastolic Relaxation of the Heart, Boston,Martinus Nijhoff, 1988, pp 217-229

8. Sasayama S, Nonogi H, Miyazaki S, Sakurai T, Kawai C,Eiho S, Kuwahara M: Changes in diastolic properties of theregional myocardium during pacing-induced ischemia in humansubjects. J Am CoU Cardiol 1985 ;5:599-606

9. Grossman W: Why is the left ventricular diastolic pressureincreased during angina pectoris? J Am Coll Cardiol 1985;5:607-608

10. Serizawa T, Carabello BA, Grossman W: Effect of pacing-induced ischemia on left ventricular diastolic pressure-volume relations in dogs with coronary stenosis. Circ Res1980;46:430-439

11. Momomura SI, Bradley AB, Grossman W: Left ventricularpressure-segment length relations and end-diastolic distensi-bility in dogs with coronary stenoses: An angina physiologymodel. Circ Res 1984^5:203-214

12. Paulus WJ, Grossman W, Serizawa T, Bourdillon PD, Pasi-poularides A, Mirsky I: Different effects of two types ofischemia on regional left ventricular systolic and diastolicfunction. Am J Physiol 1985;248:H719-H728

13. Serizawa T, Vogel WM, Apstein CS, Grossman W: Com-parison of acute alterations in left ventricular relaxation anddiastolic chamber stiffness induced by hypoxia and isch-emia. J Clin Invest 1981;68:91-102

14. Vogel WM, Apstein CS, Briggs LL, Gaasch WH, Ahn J:Acute alterations in left ventricular diastolic chamber stiff-ness. Role of the "erectile" effect of coronary arterialpressure and flow in normal and damaged hearts. Circ Res1982^1:465-478

15. Nayler WG, Buckley DJ, Elz JS: Hypoxia and relaxation, inGrossman W, Lorell BH (eds): Diastolic Relaxation of theHeart, Boston, Martinus Nyhoff, 1988, pp 67-72

16. Apstein CS, Grossman W: Opposite initial effects of supplyand demand ischemia on left ventricular diastolic compli-ance. The ischemia-diastolic paradox. J Mol Cell CardiolI987;19:119-128

17. Gaasch WH, Bing OHL, Franklin A, Rhodes D, BernardSA, Weintraub RM: The influence of acute alteration incoronary blood flow on left ventricular diastolic complianceand wall thickness. Eur J Cardiol 1978;7(suppl): 147-161

18. Palacios I, Johnson RA, Newell JB, Powell WJ Jn Leftventricular end-diastolic pressure volume relationships withexperimental acute global ischemia. Circulation 1976^53:428-436

19. Katz A, Tada M: The "stone heart": A challenge to thebiochemist. Am J Cardiol 1972^9:578-580

20. Hearse DJ, Garlick PB, Humphrey SM: Ischemic contrac-ture of myocardium. Mechanisms and prevention. Am JCardiol 1977^9:986-993

21. Wexler LF, Weinberg EO, Ingwall JS, Apstein CS: Acutealterations in diastolic left ventricular chamber distensibility:Mechanistic differences between hypoxemia and ischemia in

isolated perfused rabbit and rat hearts. Circ Res 1986;59:515-528

22. Gaasch WH, Bing OHL, Pine MB, Franklin A, Clement J,Rhodes D, Phear WP, Weintraub RM: Myocardial contrac-ture during prolonged ischemic arrest and reperfusion. Am JPhysiol 1978;235:H619-H627

23. Bing OHL, La Raia PJ, Franklin A, Stoughton J, WeintraubRM: Mechanism of myocardial protection during blood-potassium cardioplegia: A comparison of crystalloid red celland methemoglobin solutions. Circulation 1984;70(suppl I):I-84-I-90

24. Apstein CS, Mueller M, Hood WB Jn Ventricular contrac-ture and compliance changes with global ischemia andreperfusion and their effect on coronary resistance in the rat.Circ Res 1977 ;41:206-217

25. Salisbury PF, Cross CE, Rieben PA: Influence of coronaryartery pressure upon myocardial elasticity. Circ Res 1960;8:794-800

26. Olsen CO, Attarian DE, Jones RN, HOI RC, Sink JD, Lee KL,Wechsler AS: The coronary pressure-flow determinants of leftventricular compliance in dogs. Circ Res 1981;49:856-865

27. Verrier ED, Bristow JD, Hoffman JIE: Coronary vasodila-tion shifts the diastolic pressure-dimension curve of the leftventricle. J Mol Cell Cardiol 1986;18:579-594

28. Vogel MW, Briggs LL, Apstein CS: Separation of inherentdiastolic myocardial fiber tension and coronary vascularerectile contributions to wall stiffness of rabbit hearts dam-aged by ischemia, hypoxia, calcium paradox and reperfu-sion. J Mol Cell Cardiol 1985;17:57-70

29. Blanchard EM, Solaro RJ: Inhibition of the activation andtroponin calcium binding of dog cardiac myofibrils by acidicpH. Circ Res 1984^5:382-391

30. Bing OHL, Brooks WW, Masser JV: Heart muscle viabilityfollowing hypoxia: Protective effect of acidosis. Science1973;180:1297-1298

31. Momomura SI, Ingwall J, Parker JA, Sahagian P, FergusonJJ, Grossman W: The relationships of high energy phos-phates, tissue pH and regional blood flow to diastolic disten-sibility in the ischemic dog myocardium. Circ Res1985^7:822-835

32. Grossman W, McLaurin LP, Moos SP, Stefadouros M,Young DT: Wall thickness and diastolic properties of the leftventricle. Circulation 1974,49:129-135

33. Leaf A: Cell swelling: A factor in ischemic tissue injury.Circulation 1973;48:455-458

34. Brachfeld N, Christodoulou J, Erlandson R, Smithen C: Theeffects of hyperosmolal coronary perfusion on the hemody-namic, metabolic, and ultrastructural changes of myocardialanoxia. Postgrad Med J 1975^1:299-310

35. Weber A, Murray JM: Molecular control mechanisms inmuscle contraction. Physiol Rev 1973,53:612-673

36. Apstein CS, Ogilby JD: Effects of "paradoxical" systolicfiber stretch on ischemic myocardial contracture, compli-ance and contractility in the rabbit. Circ Res 1980;46:745-754

37. Grossman W, Barry WH: Diastolic pressure-volume rela-tions in the diseased heart. Fed Proc 1980;39:148-155

38. Lewis MJ, Housmans PR, Claes VA, Brutsaert DL, Hend-erson AH: Myocardial stiffness during hypoxic and reoxy-genation contracture. Cardiovasc Res 1980;14:339-344

39. Henderson AH: Ischemic/hypoxic and reperfusion/reoxy-genation contractures: Mechanisms, in Grossman W, LorellBH (eds): Diastolic Relaxation of the Heart, Boston, Mar-tinus Nyhoff, 1988, pp 73-82

40. Pine MB, Caulfield JB, Bing OHL, Brooks WW, AbelmannWH: Resistance of contracting myocardium to swelling withhypoxia and glycolytic blockade. Cardiovasc Res 1979;13:215-224

41. Zhao M, Zhang H, Robinson TF, Factor SM, SonnenblickEH, Eng C: Profound structural alterations of the extracel-lular collagen matrix in postischemic dysfunctional("stunned") but viable myocardium. J Am Coll CardiolI987;1O:1322-1334

KEYWORDS • diastole • isolated heart • hypoxia • ischemia


Dow

nloaded from


R M Wyman, E R Farhi, O H Bing, R G Johnson, R M Weintraub and W Grossmanevaluation of left ventricular diastolic chamber distensibility and wall thickness.

Comparative effects of hypoxia and ischemia in the isolated, blood-perfused dog heart:

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1989 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.64.1.1211989;64:121-128Circ Res.

http://circres.ahajournals.org/content/64/1/121World 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:


Dow

nloaded from

http://circres.ahajournals.org/content/64/1/121

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

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


comparative effects of hypoxia and ischemia in the...

Documents