790 lACC Vol. 17, No.3 March I, 1991:790-6

Pressure-Length Loop Area: Its Components Analyzed During Graded Myocardial Ischemia AMIRA SAFWAT, MD,* BRUCE J. LEONE, MD,t ROBIN M. NORRIS, MD,:j:

PIERRE FOEX, MD, DPHIL§

Davis, California, Durham, North Carolina, Auckland, New Zealand, Oxford, England

The changes in total pressure-length loop area were compared with changes in effective shortening area, systolic lengthening area and postsystolic shortening area (defined with respect to end­diastolic and end-systolic lengths) of the pressure-length loop during myocardial ischemia in seven anesthetized dogs instru­mented for measurement of left ventricular pressure and regional segmental wall motion (sonomicrometry) in the minor axis of the apical region of the left ventricle.

Ischemia was induced by gradual tightening of a micrometer­controlled snare around the left anterior descending coronary artery, which supplied the apical myocardium. Data were ob­tained at normal flow, after critical constriction (loss of pulsatile coronary flow), mild ischemia (ischemia 1: onset of regional dysfunction, i.e., postsystolic shortening and mild hypokinesia) and moderate ischemia (ischemia 2: marked hypokinesia). At each stage, acute afterloading was performed by partially occluding the descending thoracic aorta.

The pressure-length loops were analyzed in terms of four areas: total loop area, effective shortening area, postsystolic

The area of the pressure-volume loop is often used as a measure of external work of the right or left ventricle (1-3). Similarly, the area of the pressure-length loop is an index of the contribution of a ventricular segment to the total work of the ventricle (4-7). When myocardial ischemia is present, systolic lengthening and postsystolic shortening develop (8-11) and the pressure-length loop becomes distorted (9,12-14). Acute elevation of afterload may affect regional myo­cardial function as well, and further distort the pressure­length loop. Thus, the area of the pressure-length loop may

From the 'Department of Anesthesiology, University of California, Davis, California; t§Nuffield Department of Anaesthetics, Oxford, England: tDepartment of Anesthesiology, Duke University Medical Center, Durham, North Carolina; and Koronary Care Unit, Green Lane Hospital and Univer· sity of Auckland School of Medicine, Auckland, New Zealand. This study was supported by Grants GSOO2204SA and GS306S50SA from the Medical Research Council of Great Britain, London, England. Dr. Norris was sup· ported by the Medical Research Council of New Zealand, Auckland, New Zealand, and the Wellcome Trust, London, England.

Manuscript received May 14, 1990: revised manuscript received Septem· ber IS, 1990, accepted September 2S, 1990.

Address for reprints: Bruce J. Leone, MD, Department of Anesthesiol· ogy, Duke University Medical Center, P.O. Box 3094, Durham, North Carolina 27710.

© 1991 by the American College of Cardiology

shortening area and systolic lengthening area. Total loop area decreased only when marked hypokinesia was present (176 ± 18.3 mm Hg x mm at ischemia 2 versus 245.1 ± 26.9 mm Hg x mm at ischemia 1, p < 0.05). However, effective shortening area (98.2 ± 0.8% of total loop area at baseline; 93.8 ± 2.4% at critical constriction; 76.3 ± 7.2% at ischemia 1; 51.9 ± 12.2% at ischemia 2) and postsystolic shortening area (1.8 ± 0.8% of total loop area at baseline; 5.2 ± 1.9% at critical constriction; 14.3 ± 3/4% at ischemia 1; 23.8 ± 5.1 % at ischemia 2) changed signifi­cantly with each progressive stage of ischemia. Afterloading significantly decreased total loop area at all stages compared with normal loading, whereas effective shortening and postsystolic shortening areas changed only during ischemia (ischemia 1 and ischemia 2).

Thus, the individual component areas of the pressure-length loop, in particular the effective shortening area, are more sensitive to ischemic changes than is the total loop area.

(J Am Coil CardioI1991;17:790-6)

no longer be a valid index of the contribution of the ischemic segment to the external work effected by the ventricle (4,15,16).

This study was designed to test the hypothesis that the effective area-the area delineated by ventricular pressure and end-diastolic and end-systolic lengths-is a more sensi­tive index of function than is the total loop area when the myocardium is rendered ischemic and during abrupt eleva­tions in afterload.

Methods Experimental preparation. This study conforms to the

Animals (Scientific Procedures) Act of 1986 (U .K.) and to the position of the American Heart Association on research animal use. Seven mongrel dogs (average weight 18 kg, range 16 to 20) were premedicated with morphine sulfate and anesthetized with thiopental (15 mg/kg), and the trachea was intubated. Anesthesia was maintained with halothane (1% to 2%) in an oxygen (40%) and nitrogen mixture throughout the surgical preparation. Ventilation was controlled to maintain an end-tidal Peo2 of 35 to 40 mm Hg. Temperature, moni­tored at the mid-esophagus, was maintained between 37° and

0735·1097/91/$3.50

lACC Vol. 17, No.3 March 1, 1991:790-6

38°C by a heating element incorporated into the operating table.

The left common carotid artery was exposed and a rigid SF fluid-filled polyethylene catheter (outer diameter 2.76 mm) inserted and advanced into the aorta to within 1 cm of the aortic valve, and connected to a Druck® pressure transducer (Druck). An intravenous cannula was threaded via the femoral vein into the inferior vena cava for infusion of 0.9% saline solution at a constant rate of 4 ml/kg·h. A left thoracotomy was performed, the fifth and sixth ribs were excised and a segment of the thoracic aorta was carefully dissected free. A snare was placed around the aorta and used for temporary aortic constriction. The heart was then ex­posed and suspended in a pericardial cradle. A second rigid SF fluid-filled cannula was inserted into the left ventricle through a stab wound in the apical dimple and connected to a pressure transducer. The frequency response characteris­tics of this fluid-filled catheter system are such that there is <5% signal distortion at 32 Hz; thus, defined conditions for measurement of left ventricular relaxation indexes were met (17). After dissection of the fat pad surrounding the ascend­ing aorta, a cuffed electromagnetic flow transducer (Trans­flow 601, Skalar Medical) was placed around this vessel and attached to an electromagnetic flowmeter (S.E.M. 275, S. E. Medics).

A segment of the left anterior descending coronary artery was dissected free distal to the second major diagonal branch. A 3-0 woven Dacron suture was placed loosely about the artery and connected to a micrometer-controlled spring-suspended snare that could be tightened or loosened in increments of 0.25 mm. A 2 mm electromagnetic flow transducer (Transflow 601, Skalar Medical) was placed around the vessel proximal to the snare and attached to an electromagnetic flowmeter. An occluding snare was posi­tioned distal to the probe to obtain zero flow readings. This snare was used to obtain zero flow readings only at the beginning and at the end of the experimental protocol in order not to cause further deterioration in regional function during progressive ischemia. A pair of 5 MHz ultrasonic piezoelectric crystals was inserted into the subendocardium, oriented parallel to the minor axis of the left ventricle, in the apical area supplied by that portion of the left anterior descending coronary artery distal to the micrometer­controlled snare. During the first occlusion of the left ante­rior descending coronary artery, deterioration of apical regional myocardial function was also checked; abrupt changes in regional myocardial function confirmed the place­ment of the apical crystals in a region supplied by the artery.

Regional myocardial function was assessed by obtaining continuous measurement of segment length between each pair of crystals based on measurements of ultrasound transit times (1S,19). A continuous analog signal of dynamic seg­ment length was provided by a repetitive stimulation rate of 1 kHz. The transit time was calibrated in steps of 1 JLS generated by a quartz-controlled oscillator.

The signals were recorded on a Mingograf-S1 eight-

SAFW AT ET AL. 791 PRESSURE-LENGTH LOOP AREA AND ISCHEMIA

channel recorder (Elema Schoenander) at a paper speed of 250 mm/s. Aortic and left ventricular blood pressures, limb lead II of the electrocardiogram, airway pressure, myocar­dial segment length, left anterior descending coronary artery blood flow and aortic blood flow were recorded; stroke volume (from the integral of the aortic flow signal) and the first derivative of left ventricular pressure with respect to time (left ventricular dP/dt) were derived in real time and recorded.

Protocol. After the surgical preparation was completed, conditions were allowed to stabilize for 1 h. During this time, instrument calibrations were performed. Dextran (average molecular weight 70,000 daltons) was administered intrave­nously as was physiologic saline solution to maintain left ventricular end-diastolic pressure at >5 mm Hg. The end­tidal halothane concentration was adjusted to O.S%. Arterial blood gas analysis was performed and ventilation was ad­justed accordingly. Sodium bicarbonate was given intrave­nously to maintain arterial pH within the range of 7.35 to 7.45, with a base excess <5.0 mmoilliter as appropriate.

The protocol was designed to examine changes in total loop area and the components of the total loop area during four levels of regional function obtained by graded reduc­tions in regional epicardial coronary blood flow: 1) baseline; 2) critical constriction; 3) ischemia 1; and 4) ischemia 2.

Critical constriction of the left anterior descending coro­nary artery was accomplished by tightening the micrometer snare until there was an obvious decrease in pulsatile pattern of the coronary flow tracing, and was not confirmed by the absence of reactive hyperemia after 10 s of complete coro­nary occlusion, because we were concerned about inducing regional myocardial ischemia with repetitive occlusions. Regional myocardial function, however, was unchanged from baseline conditions with critical constriction.

After baseline observation, abrupt afterload of left ven­tricle was produced by constricting the snare around the descending thoracic aorta to produce an approximately 50% increase in systolic arterial blood pressure. This acute aortic constriction was maintained for six cardiac cycles, while data were continuously recorded. The second or third car­diac cycle with afterloading was analyzed, and in each dog the identical cardiac cycle after application of afterload was analyzed.

Changes in regional myocardial function were based on changes in the configuration of the pressure-length loop. Pressure-length loops, with the parameters of left anterior descending-supplied apical minor axis length (x axis) and left ventricular pressure (y axis) were continuously displayed in real time on an oscilloscope. The critical constriction stage was observed for :::::20 min before data collection to ensure that regional myocardial function did not deteriorate. After data collection during critical constriction, ischemia 1 was produced by gradual tightening of the snare until the appear­ance of postsystolic shortening with only a mild decrease in systolic shortening as seen on the pressure-length loop. After 3 min, during which time loop configuration remained

792 SAFWA T ET AL. PRESSURE-LENGTH LOOP AREA AND ISCHEMIA

stable, ischemia I data were recorded. Immediately after collection of data during abrupt afterloading with ischemia I, ischemia 2 was produced by further tightening of the snare until the appearance of marked hypokinesia on the pressure­length loop. This stage was observed for 3 min to ensure stable loop morphology and no further deterioration of regional function. If further deterioration of regional func­tion occurred, the snare was loosened slightly and retight­ened.

Data analysis and computations. Instantaneous left ven­tricular pressure and apical length signals were digitized at 4 ms intervals for each stage of the study, including a normally loaded beat and a beat during afterload. End-systole was defined as the point of return to 0 of the aortic flow signal. End-diastole was defined as the point of the initial upslope of the first derivative of left ventricular pressure (dP/dt). The data were entered into a Hewlett-Packard 9825A desktop calculator to generate pressure-length loops.

The total loop area was measured by planimetry and, in addition, the loop area was divided into three component areas (Fig. 1): 1) Effective shortening area, which is the "rectangular" area delineated horizontally by the upper and lower portions of the pressure-length loop and vertically by lines drawn through the end-diastolic pressure-length point and the end-systolic pressure-length point. 2) Postsystolic shortening area, which is the area of the pressure-length loop to the left of the vertical line through the end-systolic pressure-length point. 3) Systolic lengthening area, which is the area of the pressure-length loop to the right of the vertical line through the end-diastolic pressure-length point. Systolic shortening was defined as the difference between end-diastolic length and end-systolic length and calculated as a percent of end-diastolic length (14). Changes in segment shortening are also expressed as a percent of the baseline values.

Postsystolic shortening was defined as any decrease in segment length below end-systolic length occurring after the end of systole, and postsystolic shortening was expressed as a percent of total segment shortening:

(End-systolic length - Lmin)

Postsystolic shortening = x 100, End-diastolic length - Lmin

where Lmin is the minimal segment length observed. Any increase in segment length greater than end-diastolic length during systole was defined as systolic bulging; this was expressed as a percent of end-diastolic length:

(Lmax - End-diastolic length)

Systolic lengthening = x 100, End-diastolic length

where Lmax is the maximal segment length attained during systole.

Statistics. Results were analyzed for statistical signifi­cance using Friedman two-way analysis of variance, with Wilcoxon matched-pairs signed-rank test to determine if changes were significant between progressive stages only.

JACC Vol. 17, No.3 March 1. 1991:790-6

Normal and afterloaded beats were compared within each stage of the protocol by Wilcoxon matched-pairs signed-rank test. A p value < 0.05 was considered significant. Results are presented as mean values ± SEM.

Results Hemodynamics during ischemia. Constriction of the left

anterior descending coronary artery resulted in significant reductions of coronary flow to 70 ± 8.7% of baseline (critical constriction), 60 ± 12% of baseline (ischemia I), and 41.5 ± 9.3% of baseline (ischemia 2). Heart rate was unchanged during coronary artery constriction. Systolic and diastolic arterial pressures were also within 5% of baseline throughout the study. Left ventricular dP/dtmax decreased to 91.0 ± 3.5% of baseline with ischemia 2 (p = NS).

The increase in systolic arterial pressure during tempo­rary aortic constriction ranged from 28.3 ± 1.6 to 33.8 ± 2.1 mm Hg for all stages, with a mean heart rate of 128 ± 1.4 beats/min for the normally loaded beats and 128.4 ± 1.2 beats/min for the afterloaded beats.

Ischemia and regional function (Table 1). Coronary artery constriction caused significant reductions in systolic short­ening at ischemia I and ischemia 2 stages. These reductions were accompanied by increases in postsystolic shortening and systolic lengthening at ischemia 1 and ischemia 2 (Table I).

Total loop area was significantly reduced only during progression from ischemia 1 to ischemia 2. The effective shortening area showed significant reductions from baseline with each successive reduction in coronary flow. Similarly, postsystolic area showed significant increases from baseline with each successive reduction in coronary flow. The sys­tolic lengthening area was significantly increased with pro-

Figure 1. Pressure-length loop generated by plotting left ventricular pressure (y axis) and length of the myocardial segment supplied by the left anterior descending coronary artery (x axis). This loop was obtained during substantial hypokinesia (ischemia 2 stage). Note the characteristic change in the configuration of the loop seen with ischemia. End-diastole (ED), defined by the initial upslope of the first derivative of left ventricular pressure (dP/dt), and end-systole (ES), defined by the return of aortic flow to 0, are indicated by vertical lines. ES = effective shortening area; PSS = postsystolic shortening area; SL = systolic lengthening area.

-a 175 I ES ED E g 150

~ 125 OJ :a ~ 100 n.

0 75 "S u ~ 50

~ 25 ;I:: ~ 0

10 11 12 13 14 15

Segment Length (mm)

JACC Vol. 17, No.3 March 1, 1991:790-6

SAFW AT ET AL. 793 PRESSURE-LENGTH LOOP AREA AND ISCHEMIA

Table 1. Comparison of Loop Areas in Seven Dogs With Progressive Ischemia During Normal Loading and With Aortic Constriction

Critical Baseline Constriction Ischemia I Ischemia 2

Normal Loading Systolic shortening (% of end-diastolic length) 20.8 :!: 2.4 19.9:!: 2.2 13.0 :!: 1.9* 3.3 :!: \.7* PSS (% of total segment shortening) 6.2 :!: 2.8 8.6 :!: 3.1 32.3 :!: 5.7* 78.4 :!: 9.2* Systolic lengthening (% of end-diastolic length) 0.2 :!: 0.1 03. :!: 0.1 2.3 :!: 0.6* 6.9:!: 1.2* LAD blood flow (% of baseline) loo:!: 0.0 70.0 :!: 8.7* 59.6 :!: 12.0* 41.5 :!: 9.3* Total loop area (mm Hg x mm) 283.3 :!: 36.0 273.1 :!: 38.9 245.1 :!: 26.9 176.0:!: 18.3* Effective shortening area (% of total loop area) 98.2 :!: 0.8 93.8 :!: 2.4* 76.3 :!: 7.2* 51.9 :!: 12.2* PSS area (% of total loop area) 1.8 :!: 0.8 5.2 :!: 1.9* 14.3 :!: 3.4* 23.8:!: 5.1* Systolic lengthening area (% of total loop area) o :!: 0 1.0 :!: 0.7 9.3 :!: 4.4* 24.3 :!: 8.0*

Afterloading Total loop area (mm Hg x mm) 232.6 :!: 53.1 t 215.4 :!: 40.6t 148.0 :!: 2\.7*t 72.6 :!: 16.0*t Effective shortening area (% of total loop area) 90.0 :!: 5.2 84.2 :!: 4.8 39.5 :!: 12.3*t 9.6 :!: 9.I*t PSS area (% of total loop area) 7.9:!: 3.8 14.3 :!: 4.8 42.5 :!: 7.3*t 47.2 :!: 3.5t Systolic lengthening area (% of total loop area) 2.1 :!: 2.1 1.5 :!: 1.0 18.0 :!: 6.3t 43.2 :!: 7.8t

*p < 0.05 versus previous stage by Friedman two-way analysis of variance and Wilcoxon matched-pairs rank-sum test: tp < 0.05 versus normal loading by Wilcoxon matched-pairs rank-sum test. All values are expressed as mean values:!: SEM. LAD = left anterior descending coronary artery; PSS = postsystolic shortening.

gression of regional ischemia from critical constriction to ischemia 1 and from ischemia 1 to ischemia 2 (Table I; Fig. I and 2).

Effect of aortic constriction (Table 1). The increase in afterload produced significant reductions (p < 0.02) in total loop area at all stages of the study. Effective shortening area was significantly smaller and postsystolic area significantly larger during ischemia 1 and ischemia 2 stages. Similarly, the increases in systolic lengthening brought about by aortic constriction were significant during ischemia I and ischemia 2 stages (Fig. 3 and 4).

Discussion During normal contractions, the area of the pressure­

length loop is a useful index of segment work (4-7). It has

Figure 2. Left ventricular pressure-left anterior descending coro­nary artery myocardial segment length loops obtained in the same dog during baseline and ischemia 2 stages. Left, Loop obtained during baseline conditions. Right, Loop recorded during ischemia 2 stage. Note the characteristic change in loop configuration seen with ischemia. The shaded area indicates postsystolic shortening and the striped area indicates systolic lengthening.

a 175 a 175 :I: :I: E 150 E 150 S S ~ 125 ~ 125

11 j 100 ~ ~ 100 ~

.Q 75 .Q 75 OJ OJ 0 0 .., 50 ~ 50 C II) II)

> 25 > 25 ::: ::: II) 0 ~ 0 --'

10 11 12 13 14 15 10 11 12 13 14 15

Segment Length (mm) Segment Length (mm)

been suggested that decreases in the total area sub tended by the pressure-length loop are sensitive to regional myocardial ischemia (16). The present study demonstrates that only when substantial hypokinesia (i.e., ischemia 2) is present does total pressure-length loop area decrease significantly during regional ischemia.

Changes in pressure-length loop configuration during re­gional ischemia. Thus, the usefulness of total pressure­length loop area in examining segment work during ische­mia, when abnormal wall motion is present, must be questioned (15). Changes in pressure-length loop configura­tion and configuration caused by graded reductions in coro­nary flow have been previously reported (9,12). With ische­mia. the pressure-length loop develops an area to the right of the end-diastolic pressure-length point, representing systolic lengthening, and an area to the left of the end-systolic pressure-length point. representing postsystolic shortening

Figure 3. Changes in the pressure-length loop during baseline stage with acute afterloading from the same dog whose data are shown in Figure 2. With acute afterloading, a minimal amount of postsystolic shortening (shaded area) develops.

175 175

Oi ISO ~1SO :r E E .s 125 .s 125

~ ., OJ 100 :; 100 II) II) II) II) .,

75 ~ Ii: 75 ~

:;; SO :;;

SO :; :; .Q .g ~ 25 C 25 II) Q) > 0 > 0 a; 10 " 12 13 14 15 4i 10 II 12 13 14 IS ..J ..J

Segment Length (mm) Segment Length (mm)

794

Ci 175 J: E lSO E. ~ 125 ::> en ::l 100 ct iu 75 :; .g SO C Q)

> 25

i 0 --'

SAFW A T ET AL. PRESSURE-LENGTH LOOP AREA AND ISCHEMIA

Ci 175 J: E lSO E. ~ 125 ::> en en 100 Q)

a: iu 75 :; .g SO C Q)

> 25 a;

0 --' 10 11 12 13 14 15 10 11 12 13 14 15

Segment length (mm) Segment length (mm)

Figure 4. Changes in the pressure-length loop during ischemia with acute afterloading. Left, Loop obtained during ischemia 2 stage. Right, Loop recorded after acute constriction of the descending thoracic aorta. The white area, representing the effective segmental shortening area, is abolished with afterloading. The shaded area indicates postsystolic shortening; the striped area indicates systolic lengthening.

(Fig. 2). Theoretically, the area of systolic lengthening must represent work done by normally contracting segments of the heart on the ischemic segment, causing the latter to lengthen during isovolumetric contraction. As normal seg­ments contract, they expend energy to produce a pressure that requires tensile strength to oppose. The ischemic seg­ments, unable to generate sufficient tension, passively bulge. Thus, some of the energy expended by the normal segments to produce this pressure passively dilates the ischemic segment; the energy involved in this process is wasted because it contributes not to ejection but to the displacement of blood within the ventricle. Because postsystolic shorten­ing is believed to be an active phenomenon (l0, Il), the area delineated by it must represent work done by the ischemic segment, which is ineffective because it occurs after aortic valve closure. Therefore, postsystolic shortening does not contribute to the external work and pump function of the ventricle. Thus, the systolic lengthening and postsystolic shortening areas represent ventricular contractile energy that is wasted.

Effective shortening area of the pressure-length loop. If one specifically defines the area of the pressure-length loop that represents segmental contraction contributing to ejec­tion, as suggested by Forrester et al. (4), this effective shortening area may be far more representative of impair­ment of systolic function in the ischemic myocardial seg­ment. Indeed, the effects of worsening ischemia on the effective shortening area of the pressure-length loop were much more pronounced than the effects on the total loop area in the present study. Critical constriction, associated with a 30% reduction in coronary blood flow, decreased the effective shortening area significantly but had no effect on the total loop area. Further reductions in coronary flow resulted in substantial decreases in the effective shortening area, but the total pressure-length loop area decreased only when marked hypokinesia (that is, ischemia stage 2) was

lACC Vol. 17. No.3 March I. 1991:790-6

present. Thus, the effective shortening area is more sensitive to myocardial ischemia than is the total pressure-length loop area.

Postsystolic shortening area of the pressure-length loop. Postsystolic shortening area increased significantly at all stages of coronary constriction, whereas systolic lengthen­ing area increased significantly only at ischemia I and 2 stages. Like the effective shortening area, postsystolic short­ening area was therefore more sensitive than was total loop area to ischemic changes in regional function. However, postsystolic shortening area represents a different phenom­enon from systolic lengthening area: it is an index of inef­fective shortening performed by the ischemic segment and thus of contractile energy wasted by the ischemic segment. That the systolic lengthening area appears less sensitive to ischemia than does postsystolic shortening area may be a reflection of the relative sensitivity of early diastolic ventric­ular relaxation and systolic function to ischemia (20). As systolic lengthening area represents contractile work that has been performed by other normally contracting myocar­dium on the ischemic region, this area quantifies the degree of inefficiency the ischemic region introduces to the contrac­tile performance of the normal myocardial regions.

Comparison with changes in global indexes of ischemia. Alterations in regional myocardial function may be reflected in global indexes of left ventricular contraction. In this regard, it was interesting that the marked regional myocar­dial ischemia caused by the reduction in left anterior de­scending coronary artery blood flow occurred without alter­ation in global indexes of myocardial function. Specifically, heart rate, systolic and diastolic arterial blood pressures, and left ventricular dP/dtmax did not change even at the ischemia 2 stage.

The lack of change in left ventricular dP/dtmax is in contrast with a previous report that this variable is a sensi­tive indicator of regional myocardial ischemia (21). Other investigators (22-24), however, using preparation similar to that in the present study, did not detect significant alter­ations in LV dP/dtmax despite profound apical myocardial ischemia. This may be explained by the relation between global indexes of left ventricular contraction and the amount of ischemic myocardium: a greater amount of ischemic myocardium may produce significant changes in global in­dexes of contraction (24). In the present study, the ischemic territory constituted roughly 25% (wet weight) of the mass of the left ventricle, similar to the preparation used by Akaishi et al. (24); in both studies, left ventricular dP/dtmax was unaltered despite profound decreases in left anterior de­scending coronary artery blood flow and regional myocardial function. Thus, global indexes of left ventricular function are not sensitive to profound ischemia in smaller regions of myocardium; however, indexes such as left ventricular dP/dtmax may be significantly altered during regional ische­mia if the ischemic region constitutes a large portion of the left ventricle.

lACC Vol. 17. No.3 March I. 1991:790-6

CD . !: q; en as IX!

E ,g ~ ~

0

-20

-40

-60

-80

-100

B

STAGE

CC 2

Figure 5. Changes in the total pressure-length loop area (black columns) and the effective shortening loop area (hatched columns) with afterloading caused by acute descending thoracic aortic con­striction. Values are expressed as a percent of the values obtained with normal loading. With afterloading. both areas decreased. and these decreases were significant for both areas during ischemia 1 and ischemia 2 stages. However, afterloading resulted in a significantly greater reduction in effective shortening area than total loop area during ischemia. *p < 0.05 by two-way analysis of variance and Duncan's test, tp < 0.05 by paired t test. B = baseline; CC = critical constriction, 1 = ischemia 1; 2 = ischemia 2.

Effects of afterloading. Afterloading, induced by abruptly constricting the descending thoracic aorta, resulted in de­creases in the total pressure-length loop area. At baseline, in the absence of coronary flow reductions, afterloading in­duced little change in the effective shortening area (Fig. 3). With the onset of ischemic dysfunction caused by coronary stenosis (ischemia I and 2), a profound decrease in the effective shortening area occurred with aortic constriction, indicating a marked reduction in the amount of effective work done by the ischemic segment when afterloaded. Indeed, at ischemia 2 stage, afterloading frequently obliter­ated any effective shortening area (Fig. 4). Decreases in total loop area were exaggerated during afterloading with progres­sion of ischemia, yet decreases in effective shortening area induced by afterloading were substantially greater than the decreases in total loop area brought about by afterloading (Fig. 5).

These observations highlight the deleterious effects of afterloading on systolic and diastolic performance of ische­mic myocardium, as previously reported (20). Because total loop area may be decreased by afterloading in the absence of ischemia, while the effective shortening area decreases only when afterloading is combined with ischemia, reductions in the effective shortening area are more specific than reduc­tions in the total loop area for detection of regional ischemia.

The acute aortic constrictions caused postsystolic short­ening area to increase during ischemia in contrast to the opposite effects of afterloading on effective shortening area. The observations of an increase in postsystolic shortening area, and thus an increase in wasted energy, and a decrease in effective shortening area, and thus a decrease in effective contractile energy, suggest that afterloading increases the inefficiency of left ventricular contraction in the presence of

SAFWAT ET AL. 795 PRESSURE-LENGTH LOOP AREA AND ISCHEMIA

regional ischemia. Afterloading also increased the area of systolic lengthening, as previously reported (25), further contributing to left ventricular inefficiency .

Acute elevations in afterload can cause alterations in myocardial contractility, producing the "homeometric auto­regulation" (26) or "Anrep effect" (27). This effect was described first in isolated, supported heart preparations after several minutes of afterloading, and does not appear in the first 20 to 30 s of acute afterloading (28). It is more pro­nounced with intravenous catecholamine infusion, suggest­ing involvement of neurohumoral mechanisms (26,27). The lack of such an effect in the present study may be due to the short duration (six cardiac cycles) of the afterloading and the effect of the volatile anesthetic halothane, a known depres­sant of baroreceptor and catecholamine response (29).

There is evidence that acute cross-clamping of the ab­dominal aorta in patients with impaired left ventricular function results in the development or worsening of regional wall motion abnormalities (30), indicating that increases in afterload cause or augment myocardial ischemia. Con­versely, reductions in afterload were recently shown to diminish ischemia-induced alterations in left ventricular re­laxation in experimental (31) and clinical (32) studies. Al­though one could speculate that afterload reductions during ischemia would increase the effective shortening area and decrease the postsystolic shortening area, the present study did not address this question and further studies are war­ranted.

Study limitations. This study was conducted with pro­gressive ischemia by gradual coronary constriction. Progres­sion of left anterior descending coronary artery stenosis without reperfusion may have resulted in a cumulative ischemic effect on regional myocardial function. A reperfu­sion phase between each stage may have yielded different results. The protocol was conducted so as to minimize the time taken between protocol steps yet ensure a stable level of ischemia before data collection, minimizing any possible cumulative ischemic effect. Application of afterload, causing further ischemia, could have exacerbated this ischemic effect. However, regional myocardial function did not ap­pear to have changed after release of the afterloading; also, the duration of acute afterioading was only six cardiac cycles. Therefore, although the protocol design may have allowed some cumulative ischemic effect to influence the results, this influence was probably minimal.

The definition of ischemia 1 and ischemia 2 stages was based on visual, real-time inspection of the left ventricular pressure-apical segment length loop during the experimental protocol. Ischemia I was defined in the original, intended protocol as the appearance of postsystolic shortening with no discernible decrease in systolic shortening in the apical segment. In fact, a decrease in systolic shortening during ischemia I was apparent when the data were analyzed (Table I). Ischemia 2, defined as the point at which a decrease in systolic shortening became obvious on the pressure-length loop, resulted in a marked reduction in systolic shortening

796 SAFWAT ET AL. PRESSURE-LENGTH LOOP AREA AND ISCHEMIA

when compared with that recorded during ischemia 1 (Table 1). Clearly, ischemia 1 and ischemia 2 represented two separate gradations of myocardial ischemia.

Conclusions. The pressure-length loop, in the presence of regional myocardial ischemia, can be separated into three components: the systolic lengthening area, the area of effec­tive shortening and the postsystolic shortening area. Both the effective shortening area and the post systolic shortening area were more sensitive than the total pressure-length loop area to regional ischemia induced by progressive constric­tion of the left anterior descending coronary artery. After­load substantially altered the systolic lengthening, effective shortening and postsystolic shortening areas, in addition to decreasing the total loop area. Measurement of the effective shortening area of the pressure-length loop may be a more sensitive and specific indicator than total pressure-length loop area of effective regional myocardial function in the presence of ischemia.

We express our gratitude to W. Allen Ryder for skilled assistance with all technical aspects of this study and to Averyl L. Nunn for excellent assistance in the preparation of this manuscript.

References I. Suga H. Hayashi T, Shirahata M. Ventricular systolic pressure-volume

area as a predictor of cardiac oxygen consumption. Am 1 Physiol (Heart Circ Physioi) 1981;9:H39-44.

2. Khalatbeigui F, Suga H. Sagawa K. Left ventricular systolic pressure­volume area correlates with oxygen consumption. Am 1 Physiol (Heart Circ Physiol) 1979;6:H566-9.

3. Suga H. Total mechanical energy of a ventricle model and cardiac oxygen consumption. Am 1 Physiol (Heart Circ Physiol) 1979;5:H498-505.

4. Forrester lS, Tyberg lV. Wyatt HL. Goldner S. Parmley WW, Swan H1C. Pressure-length loop: a new method for simultaneous measurement of segmental and total cardiac function. 1 Appl Physiol 1974;37:771-5.

5. Pagani M. Vatner SF. Baig H. Braunwald E. Initial myocardial adjust­ments to brief periods of ischemia and reperfusion in the conscious dog. Circ Res 1978;43:83-91.

6. Stowe DF, Mathey DG, Moores WY. et al. Segment stroke work and metabolism depend on coronary blood flow in the pig. Am 1 Physiol (Heart Circ Physiol) 1978;3:H597-607.

7. Tyberg lV, Forrester lS, Wyatt HL. Goldner Sl. Parmley WW. Swan HJC. An analysis of segmental ischemic dysfunction utilizing the pres­sure-length loop. Circulation 1974;49:748-54.

8. Vatner SF, Baig H. Manders WT, Ochs H. Pagani M. Effect of propran­olol on regional myocardial function, electrograms. and blood flow in conscious dogs with myocardial ischemia. 1 Clin Invest 1977;60:353-60.

9. Doyle RL, Foex P. Ryder WA. 10nes LA. Differences in ischaemic dysfunction after gradual and abrupt coronary occlusion: effects on isovolumic relaxation. Cardiovasc Res 1987;21:507-14.

10. Brown MA, Norris RM, Takayama M, White HD. Post-systolic shorten­ing: a marker of potential for early recovery of acutely ischaemic myocardium in the dog. Cardiovasc Res 1987;21:703-16.

II. Takayama M, Norris RM. Brown MA. Armiger L. Rivers IT. White HD.

JACC Vol. 17. No.3 March I. 1991:790-6

Post-systolic shortening of acutely ischemic canine myocardium predicts early and late recovery of function following coronary reperfusion. Circulation 1988 ;78:994-1007.

12. Forrester lS. Wyatt HL, da Luz PL. Tyberg lV, Diamond GA, Swan H1C. Functional significance of regional ischemic contraction abnormal­ities. Circulation 1976;54:64-70.

13. Geffin GA. Drop LJ. Newell JB, et al. Effect of preload on ischaemic and non-ischaemic left ventricular regional function. Cardiovasc Res 1986;20: 415-27.

14. Theroux P, Ross 1, Franklin D, Kemper WS, Sasayama S. Regional myocardial function in the conscious dog during coronary occlusion and responses to morphine. propranolol. nitroglycerin. and lidocaine. Circu­lation 1976;53:302-14.

15. Foex P. Francis CM. Cutfield GR, Leone B. The pressure-length loop. Br 1 Anaesth 1988;6O:65-7IS.

16. Sasayama S. Nonogi H, Fujita M. et al. Analysis of asynchronous wall motion by regional pressure-length loops in patients with coronary artery disease. 1 Am Coli Cardiol 1984;4:259-67.

17. Gersh Bl, Hahn CEW, Prys-Roberts C. Physical criteria for measurement of left ventricular pressure and its first derivative. Cardiovasc Res 197I:5:32-40.

18. Bugge-Asperheim B. Leraand S. Kiil F. Local dimensional changes of myocardium measured by ultrasonic technique. Scand 1 Clin Lab Invest 1969;24:361-71.

19. Hagl S. Heimisch W, Meisner H. Erben R, Baum M, Mendler N. The effect of hemodilution on regional myocardial function in the presence of coronary stenosis. Basic Res Cardiol 1977;72:344-64.

20. St 10hn Sutton MG. Ritman EL. Effects of progressive reduction in coronary blood flow on regional and global left ventricular contraction and relaxation during normal and increased afterload: a roentgen video­metric study. Cardiovasc Res 1982;16:535-45.

21. Lekven 1, Mjj}s OD. Kjekshus lK. Compensatory mechanisms during graded myocardial ischemia. Am 1 CardioI1973;31:467-73.

22. Theroux P. Franklin D, Ross 1 lr, Kemper WS. Regional myocardial function during acute coronary artery occlusion and its modification by pharmacologic agents in the dog. Circ Res 1974;35:896-908.

23. Waters DD. da Luz P, Wyatt HL. Swan H1C. Forrester lS. Early changes in regional and global left ventricular function induced by graded reduc­tions in regional coronary perfusion. Am 1 Cardiol 1977;39:537-43.

24. Akaishi M, Schneider RM, Mercier Rl, et al. Relation between left ventricular global and regional function and extent of myocardial ischemia in the canine heart. 1 Am Coli CardioI1985;6:104-12.

25. Noma S. Askenase AD. Agarwal JB. Helfant RH. The effect of changes in afterload on systolic bulging. Circulation 1988;77:221-6.

26. Sarnoff Sl. Mitchell JH. Gilmore lP. Remensnyder JP. Homeometric autoregulation in the heart. Circ Res 1960;8:1077-91.

27. von Anrep G. On the part played by the suprarenals in the normal vascular reactions of the body. J Physiol (Lond) 1912-1913;45:307-17.

28. Wilcken DEL. Charlier AA, Hoffman JIE, Guz A. Effects of alterations in aortic impedence on the performance of the ventricles. Circ Res 1964;14: 283-93.

29. Duke PC. Fownes D. Wade JG. Halothane depresses baroreflex control of heart rate in man. Anesthesiology 1977:46:184-7.

30. Roizen MF. Beaupre PN. Alpert RA, et al. Monitoring with two­dimensional transesophageal echocardiography: comparison of myocar­dial function in patients undergoing supraceJiac. suprarenal-infraceliac. or infrarenal aortic occlusion. 1 Vase Surg 1984;1:300-5.

31. Applegate RJ, Walsh RA, O'Rourke RA. Effects of nifedipine on diastolic function during brief periods of flow-limiting ischemia in the conscious dog. Circulation 1987;76:1409-21.

32. Nakamura Y, Sasayama S. Nonogi H, et al. Effects of pacing-induced ischemia on early left ventricular filling and regional myocardial dynamics and their modification by nifedipine. Circulation 1987;76:1232-44.