Decrease in Forces Responsible for Diastolic Suction During Acute Coronary Occlusion
Background The production of left ventricular (LV) restoring forces generated during contraction, which are responsible for diastolic suction, is dependent on end-systolic volume (ESV) and systolic transmural and 3D deformation. We tested the hypothesis that acute coronary occlusion would result in loss of forces that cause suction.
Methods and Results Ten open-chest dogs were subjected to a 10-minute acute coronary occlusion (proximal left anterior descending coronary artery). A servomotor connected to the left atrium (LA) was used to rapidly clamp LA pressure during systole below the level of the succeeding LV diastolic pressure, resulting in nonfilling diastoles during which the LV fully relaxed at its ESV. LA clamps at multiple ESVs (conductance catheter) allowed delineation of positive and negative portions of the fully relaxed LV pressure-volume relation (FRPVR). A negative fully relaxed pressure (FRP) indicates the presence of restoring forces. After 10 minutes of acute coronary occlusion, there was an upward shift of the FRPVR. Thus, for example, at matched ESVs before and during coronary occlusion, FRP was −1.1±1.1 (±SD) mm Hg before versus 0.2±1.2 mm Hg after 10 minutes of coronary occlusion (P<.05).
Conclusions Acute coronary occlusion results in a rapid decrease in forces responsible for suction. This phenomenon is independent of the level of ESV and may contribute to ischemic diastolic dysfunction.
Acute coronary occlusion resulting in “supply” ischemia has been reported to have complex effects on diastolic LV function. In the myocyte, diastolic calcium concentration is thought to be elevated but counterbalanced by myofilament calcium desensitization, such that diastolic tension is not substantially increased.1 At the level of the ventricle, slowed relaxation,2 loss of turgor,3 4 and alterations in external constraints5 have been noted. One determinant of filling that may be altered during coronary occlusion but has not previously been studied is the ability of the LV to fill by suction. Suction occurs when the ventricle contracts below its Veq6 7 8 9 and generates restoring forces, whose magnitude is inversely related to ESV. A component of the potential energy of restoring forces is thought to reside in the complex, transmural systolic deformation of the LV and resultant 3D deformation, including twist, or counterclockwise torsional rotation followed by diastolic untwisting.10 11 12 13 Another component results from passive stretch of elastic elements in the thickened wall when ESV is below Veq and is manifested in the fully relaxed LV as a negative transmural pressure.9 Coronary occlusion alters regional systolic transmural deformation by virtue of both impaired fiber shortening and decreased regional turgor. Recently, Gibbons Kroeker et al14 measured apex rotation (an index of twist) during acute LAD occlusion. Within 2 minutes, they observed decreases in the magnitude of rotation and complex alterations during isovolumic relaxation and filling. Thus, coronary occlusion also modifies 3D deformation of the LV. Taken together, these observations suggest that acute coronary occlusion may alter generation of restoring forces and filling by suction.
Recently, we and others have described an LA servomotor system that rapidly clamps LAP at a specified value during systole and maintains it at that value during the subsequent diastole.15 16 We used this device to produce nonfilling diastoles by clamping LAP at a level below the LV diastolic pressure. This allowed measurement of the FRP at its end-systolic configuration and delineation of the relation between FRP and ESV and Veq. A negative FRP indicates that a restoring force is present. In the present study, we tested the hypothesis that acute coronary occlusion decreases the ability of the LV to generate a negative FRP and therefore impairs filling by suction.
Ten adult mongrel dogs (18 to 27 kg) of either sex were anesthetized with sodium pentobarbital (30 mg/kg IV) after premedication with morphine sulfate (1 mg/kg IM). The dogs were laid on their right sides on a heating pad to maintain body temperature, intubated, and ventilated with a mixture of 100% O2 and room air (Harvard Apparatus). Physiological pH (7.35 to 7.45 mm Hg) and Pco2 (35 to 45 mm Hg) were maintained with intravenous sodium bicarbonate and adjustments of the rate and tidal volume of ventilation. A surgical plane of anesthesia was confirmed, and succinylcholine (0.5 mg/kg IV) was given. A left thoracotomy and pericardiotomy were performed to expose the heart. Vascular constrictors were placed around both caval vessels and the ascending aorta. The left carotid artery was dissected through a medial ventral cervical incision. After administration of heparin (300 U/kg IV), the carotid artery was cannulated with a 7F catheter-tip micromanometer with fluid port (SPC 471A, Millar Instrument Co), which was advanced into the LV. A volume conductance catheter (Leycom, Sigma 5, CardioDynamics)17 was inserted through the apex of the LV and its tip positioned just below the aortic valve. The conductance catheter was not calibrated, because we did not require knowledge of absolute volume. A large-bore (8-mm ID) rigid cannula was placed in the LA via the appendage, positioned just above and at the midpoint of the mitral valve, and attached to the servomotor. A 5F high-fidelity pressure transducer (MPC 500, Millar) inserted into a rigid sheath with multiple side holes and a fluid-filled catheter attached to the large-bore cannula were used to measure LAP. A circumferentially oriented pair of midwall ultrasonic crystals was positioned about 1 cm apart in the anterior LV free wall, in the distribution of the LAD. Pacing electrodes were attached to the LA and connected to a stimulator (SD-9, Grass Instruments) that was controlled by a computer (486/33 MHz, Gateway 2000). A ligature was placed around the LAD proximal to the first major diagonal branch. The ends of the suture were passed through a soft rubber tube that could be advanced against the artery, causing complete occlusion.
After instrumentation was completed, the calcium channel blocker zatebradine (UL-FS 49, 1 mg/kg IV, Boehringer Ingelheim) was administered. Zatebradine slows the sinus rate but does not significantly depress cardiac contractility.18 At this dose, we were able to consistently pace the LA at 90 bpm with 1:1 atrioventricular conduction and maintain heart rate constant at this value throughout the study. We then applied brief partial caval constrictions and aortic constrictions to generate a variety of steady-state ESVs. At each ESV, data were recorded with respiration suspended at end expiration during four or five steady-state beats followed by a nonfilling beat. To produce nonfilling diastoles, the servomotor system was used to rapidly clamp the LAP below the LV diastolic pressure during ventricular systole, as described earlier.15 The LV was allowed to fully relax at its ESV, and the FRP was measured as the plateau value after completion of relaxation. To ensure that no LV filling occurred during LAP clamp beats, we required that there be no change in the conductance catheter volume signal after the time of the LV-LA crossover pressure (determined from the filling beat preceding the LAP clamp beat) and that LV pressure decline monotonically to a plateau value during the LAP clamp. Using color Doppler, we previously showed that beats meeting these criteria do not demonstrate transmitral flow during the LAP clamp.15 The LAD was then occluded, and after 10 minutes, measurements of the FRP were repeated at several different ESVs. In each experiment, we required that the midwall segment display systolic bulging during coronary occlusion. Delineation of as complete a range of ESVs as those obtained before coronary occlusion was not attempted because of the time required to alter the steady-state ESV and measure the FRP during this non–steady-state intervention.
All data were digitized on-line at a 200-Hz sampling frequency and analyzed with custom-designed software for heart rate, peak systolic LV pressure, peak positive and negative LV dP/dt, LVEDP (LVP when LV +dP/dt reached 10% of its maximum value), LV-LA crossover pressure, T1/2 (time for LVP to decrease by 50% from its end-systolic value, defined as LV pressure at 30 ms before minimum dP/dt), and FRP during the LAP clamp. We also calculated the average rate of LVP fall over 10 ms after the LV-LA pressure crossover (average dP/dt) for the filling beat immediately preceding the LAP clamp and for the clamped beat after the LV pressure dropped below the LV-LA crossover value of the preceding filling beat. The uncalibrated conductance catheter signal17 was used to determine ESV as the smallest volume preceding peak negative dP/dt. The sonomicrometer signals were used to ensure that there was dyskinesis in the ischemic region during coronary occlusion.
Nikolic et al19 20 21 have previously fitted the relation between FRP and ESV (the FRPVR) to different logarithmic equations for its positive and negative portions. (Above Veq, they included fully relaxed values from both filling and nonfilling beats.) In their preparation, an electronically controlled mitral valve prosthesis was used to occlude the mitral orifice and produce nonfilling diastoles.19 20 Negative FRPs in the −5 to −10 mm Hg range were recorded. In our preparation15 and another22 used to produce nonfilling diastoles in which the native mitral valve has been intact, less negative FRPs (typically −2 to −3 mm Hg) have been recorded. In consequence, we usually recorded FRPs over a very narrow negative range. As a result, we could not consistently fit the coronary occlusion FRPVR data to curvilinear equations of any form. We therefore report representative examples of FRPVR data sets obtained before and during coronary occlusion. In view of our inability to fit the FRPVR data, we refer to changes in “apparent” Veq. To statistically analyze our data, we compared FRP at matched values of ESV before and during coronary occlusion by the following algorithm. We first selected a control beat with a negative FRP. Then we selected a beat during coronary occlusion whose ESV was smaller but otherwise closest to the ESV of the control beat. Since the magnitude of FRP is inversely proportional to ESV, comparison of the control beat with a beat during occlusion with a smaller ESV biased against our hypothesis that FRP is less negative during occlusion. We also compared segment lengths during nonfilling diastoles before and during coronary occlusion. As we previously reported,15 the segment undergoes an initial deformation during nonfilling diastoles but then assumes a constant value after completion of relaxation (Fig 1⇓). In each dog, we compared this plateau segment length at an ESV before occlusion that was identical to that used for FRP comparisons with the value present during occlusion, at an ESV matched as closely as possible to that present before coronary occlusion.
A paired t test was used to compare each parameter. For all statistical comparisons, P<.05 was considered significant. Data are reported as mean±SD.
Fig 1⇑ shows representative tracings for filling and nonfilling beats before (left) and during (right) coronary occlusion. Vertical lines indicate the initiation of LA volume withdrawal to reduce the LAP to a preset level below the LV diastolic pressure, preventing diastolic LV filling. Absence of filling during the clamped beat was confirmed by use of the criteria described earlier. In this example, ESV was virtually identical for control and coronary occlusion beats. FRP was −0.3 mm Hg for the control and 2.4 mm Hg for the coronary occlusion beat. Note that the segment reaches a plateau value during the nonfilling diastole both before and during coronary occlusion and that the segment is larger during coronary occlusion despite the virtually identical ESV.
Representative examples of FRPVR data for control and coronary occlusion conditions are shown in Fig 2⇓. An upward shift of the FRPVR over the range of our observations is evident during coronary occlusion. In the case shown on the left, none of the FRPs during coronary occlusion were negative, and the lowest values occurred at ESVs associated with slightly negative control FRPs. By interpolation, apparent Veq was shifted to the left (smaller) during coronary occlusion. In the case shown on the right, the FRPs during coronary occlusion were recorded over a larger range of ESVs, including one negative value, and apparent Veq was again shifted to the left. In 7 of 10 dogs, there was a clear upward shift of both the negative- and positive-pressure portions of the FRPVR in association with a leftward shift of apparent Veq to a smaller value during coronary occlusion. In 1 dog, there was a definite upward shift that appeared to be confined to the negative range, in 1 a small upward shift confined to the negative range, and in 1 no apparent shift. In the Table⇓, baseline hemodynamic data before (control) and during acute coronary occlusion are shown in the left two columns (baseline refers to conditions present in the absence of any load manipulation). During coronary occlusion, EDP and T1/2 were slightly increased (the latter was statistically significant only for nonfilling beats), whereas average dP/dt was substantially reduced during occlusion for both filling and nonfilling beats. LA-LV crossover and LV minimum pressure were higher during coronary occlusion, but only the latter was statistically significant. FRP was significantly increased by an average of 2.1 mm Hg during coronary occlusion. Data at matched ESV before and during coronary occlusion are shown in the two right columns of the Table⇓. LV peak systolic pressure tended to be reduced (P<.06) during coronary occlusion because caval constriction was required to produce matched ESV points. Under these conditions, LVEDP was not significantly different. Because of the ESV matching algorithm, ESV was very slightly but significantly smaller during coronary occlusion. Changes in T1/2, while similar to those during baseline, were not significant, whereas average dP/dt remained substantially decreased. Neither LA-LV crossover nor LV minimum pressure was significantly changed during coronary occlusion. Under these conditions, FRP increased significantly from an average value of −1.1 to 0.2 mm Hg.
Comparisons of segment length values at matched ESV during nonfilling diastoles revealed a significantly larger mean value during coronary occlusion (10.1±2.4 mm before versus 11.0±2.4 mm during coronary occlusion, P<.02). For these comparisons, ESV averaged 76.7 mL before and 77.3 mL during coronary occlusion (P=.8).
LV restoring forces are most likely present under normal resting conditions.15 20 22 They result in diastolic suction, such that filling occurs at a lower level of diastolic pressure than would otherwise be the case. Our results indicate that acute LAD occlusion causes an upward shift of the FRPVR with a reduction in apparent Veq and consequently a decrease in the net force causing suction at any ESV. Under control operating conditions, FRP was negative, whereas under operating conditions during coronary occlusion, FRP was positive. Thus, coronary occlusion resulted in a situation in which a force causing suction was no longer present under operating conditions. Data at matched ESVs provide a quantitative estimate of the change in the FRPVR, which averaged 1.3 mm Hg.
The ability to generate restoring forces depends on ESV, the position of the negative FRPVR, and Veq. A positive inotropic agent (dobutamine) has been reported to lower the position of the negative FRPVR without changing Veq.8 20 With coronary occlusion, the position of the negative FRPVR was shifted upward, usually in association with a decrease in Veq. Thus, during coronary occlusion, suction would be expected to be impaired due to the combination of the change in position of the FRPVR (a less negative or a positive pressure at any ESV <control Veq), decreased Veq per se, which makes it more difficult to contract to an ESV <Veq, and depression of overall contractile function, which also makes it more difficult to contract to an ESV <Veq.
Although the FRPVR was shifted upward, this does not necessarily mean that a reduced ability to generate restoring forces was the cause. An important assumption underlying equating a negative FRP with the magnitude of restoring forces is that myofilament inactivation and as a result diastolic tension are normal. If inactivation were impaired such that diastolic tension were elevated, FRP would increase independently of any effect on restoring forces. (Obviously, the term FRP would then also be a misnomer.) Although diastolic calcium concentration may be elevated in supply ischemia, it has generally been considered that diastolic tension is not as a result of concomitant myofilament calcium desensitization.1 It would also be surprising if an ischemia-related alteration in cellular calcium handling so severe that it resulted in incomplete inactivation were to cause marked effects on late LV pressure fall but minimal effects on isovolumic pressure fall. Thus, although we cannot exclude it, an increase in FRP due to incomplete inactivation seems unlikely.
If the change in FRP resulted from a decrease in restoring forces, a straightforward explanation would be reduced blood volume in the perfusion zone of the occluded LAD, with a decrease in regional wall thickness and a resultant decrease in passive stretch of elastic elements in the wall at ESVs below Veq. If the only factor responsible for a change in FRP during coronary occlusion were decreased turgor, the predicted result would be an upward shift at ESVs below Veq and a downward shift at ESVs above Veq3 4 but no change in Veq itself. Thus, the fact that Veq was reduced during coronary occlusion suggests that additional factors influence the FRP. One possibility is interference with deformations that normally accompany contraction below Veq and are thought to be partly responsible for suction, for example, differences in subepicardial and subendocardial contraction and relengthening and twist, which allow storage of the potential energy of restoring forces before it is converted to elastic recoil and suction.11 12 13 14 However, during the course of a nonfilling diastole, such contraction-related deformations should revert to the undeformed state and associated restoring forces should dissipate as relaxation progresses and therefore not contribute to the FRP. Our finding that the segment length during nonfilling diastoles was longer during coronary occlusion at matched ESV suggests another possibility. “Creep” of ischemic segments at end diastole has been recognized for many years23 and attributed to repeated stretching of the ischemic tissue. Moreover, a longer segment length is exactly the opposite of what would be anticipated on the basis of a decrease in blood volume in the wall. Thus, in addition to decreased turgor, complex alterations in regional wall stress distribution due to creep in the ischemic region could influence the FRP and provide an explanation for the decrease in apparent Veq. Further studies will be required to fully understand the factors that alter FRP during coronary occlusion. Regardless of the mechanism, however, the change in FRP once again serves to reduce the net force responsible for suction.
During coronary occlusion, there was a decrease in the rate of late LVP fall (average dP/dt) for both normal filling and nonfilling beats. A decrease in or loss of restoring forces in effect reduces instantaneous LV chamber compliance during the time that restoring forces are dissipated, ie, during early filling. However, it is unknown whether dissipation of restoring forces influences rate of LVP fall. Our results are consistent with this possibility, but there are other explanations. A decrease in myocyte relaxation rate due to a reduced rate of calcium removal from the myofilaments has already been mentioned. As has been discussed, complex transmural and 3D deformations normally occur during relaxation and filling. During nonfilling diastoles, regional deformations remain despite the absence of filling.15 (Indeed, as suggested earlier, the latter may be a manifestation of partial dissipation of restoring forces generated during contraction.) These deformations represent normally occurring regional inhomogeneities and must also be associated with changes in loads imposed on the myofibers during relaxation.24 Thus, slowing of late pressure fall during coronary occlusion could reflect an alteration in the load dependence of myofilament inactivation related to increased regional inhomogeneity most marked during the later phase of relaxation.24 This would also fit with the very minimal effect on isovolumic pressure fall, since these inhomogeneities would be most marked during filling, when volume is unconstrained. Last, a coronary occlusion–related increase in viscous resistance to filling is another explanation,25 26 27 but this is unlikely because average dP/dt was decreased by a similar extent during coronary occlusion even in the absence of filling.
As discussed, because of a limited range of data points, we could not fit these data to mathematical models, which would have allowed a more complete and rigorous description of the changes in the FRPVR that were observed. However, we purposely analyzed the data in a way that would bias against an upward shift.
Our data were recorded at low to low-normal EDPs; therefore, it is legitimate to question their relevance to clinical coronary occlusion and myocardial infarction, in which EDP is typically normal or elevated. In our previous study in an identical preparation,15 we showed that restoring forces are normally present over approximately the lower half of the physiological range of LVEDP, with Veq occurring when EDP is on average ≈8 mm Hg. In many cases of coronary occlusion and/or myocardial infarction, EDP remains normal or is reduced to normal by treatments such as inotropic drugs, vasodilators, or diuretics. These same treatments also reduce ESV. Thus, our results may be most applicable to these situations. Moreover, our open-chest preparation with barbiturate anesthesia undoubtedly resulted in some depression in baseline contractile performance, tending to increase ESV and minimize restoring forces compared with more physiological conditions. Last, the actual changes in FRP in this study were quantitatively small. However, ventricular suction operates via its effects on the early diastolic transmitral pressure gradient, and the magnitude of change we observed is not small in relation to the usual magnitude of the gradient.
Selected Abbreviations and Acronyms
|FRP||=||fully relaxed LV pressure|
|FRPVR||=||fully relaxed LV pressure-volume relation|
|LA||=||left atrial, left atrium|
|LAD||=||left anterior descending coronary artery|
|LAP||=||left atrial pressure|
|LV||=||left ventricular, left ventricle|
This research was supported by NIH grant HL-51201.
Presented in part at the 68th Annual Scientific Sessions of the American Heart Association, Anaheim, Calif, November 13-16, 1995, and published in abstract form (Circulation. 1995;92[suppl I]:I-451).
- Received December 12, 1996.
- Revision received May 12, 1997.
- Accepted May 20, 1997.
- Copyright © 1997 by American Heart Association
Barry WH. Mechanical dysfunction of the heart during and after ischemia. Circulation. 1990;82:652-654.
Jean GF, Bronzwaer MD, de Bruyne B, Ascoop CAPL, Paulus WJ. Comparative effects of pacing-induced and balloon coronary occlusion ischemia on left ventricular diastolic function in man. Circulation. 1991;84:211-222.
Wong BYS, Toyama M, Reis RL, Goodyer AVN. Sequential changes in left ventricular compliance during acute coronary occlusion in the isovolumic working canine heart. Circ Res. 1978;43:274-286.
Kass DA, Midei M, Brinker J, Maughan WL. Influence of coronary occlusion during PTCA on end-systolic and end-diastolic pressure-volume relations in humans. Circulation. 1990;81:447-460.
Brecher GA. Critical review of recent work on ventricular diastolic suction. Circ Res. 1958;6:554-566.
Ebstein E. Die Diastole des Herzens. Ergeb Physiol. 1904;3:123-194.
Hori M, Yellin EL, Sonnenblick EH. Left ventricular diastolic suction as a mechanism of ventricular filling. Jpn Circ J. 1982;4:124-129.
Yellin EL, Hori M, Yoran C, Sonnenblick EH, Gabby S, Frater RWM. Left ventricular relaxation in the filling and non-filling intact canine heart. Am J Physiol. 1986;250:H620-H629.
Ingels NB Jr, Hansen DE, Daughters GT II, Stinson EB, Alderman EL, Miller DC. Relation between longitudinal, circumferential, and oblique shortening and torsional deformation in the left ventricle of transplanted human heart. Circulation. 1989;64:915-927.
Rademakers FE, Buchalter MB, Rogers WJ, Zerhouni EA, Weisfeldt ML, Weiss JL, Shapiro EP. Dissociation between left ventricular untwisting and filling. Circulation. 1992;85:1572-1581.
Gibbons Kroeker CA, Tyberg JV, Beyar R. Effects of load manipulations, heart rate, and contractility on left ventricular apical rotation: an experimental study in anesthetized dogs. Circulation. 1995;92:130-141.
Gibbons Kroeker CA, Tyberg JV, Beyar R. Effects of ischemia on left ventricular apex rotation: an experimental study in anesthetized dogs. Circulation. 1995;92:3539-3548.
Bell SP, Fabian J, Higashiyama A, Chen Z, Tischler MD, Watkins MW, LeWinter MM. Restoring forces assessed with left atrial pressure clamps. Am J Physiol. 1996;270:H1015-H1020.
Ingels NB Jr, Daughters GT II, Nikolic SD, DeAnda A, Moon MR, Bolger AF, Komeda M, Derby GC, Yellin EL, Miller DC. Left atrial pressure-clamp servomechanism demonstrates LV suction in canine hearts with normal mitral valves. Am J Physiol. 1994;267:H354-H362.
Boltwood CM, Appleyard RF, Glantz SA. Left ventricular volume measurement by conductance catheter in intact dogs: parallel conductance volume depends on left ventricular size. Circulation. 1989;80:1360-1377.
Chen Z, Slinker BK. The sinus node inhibitor UL-FS 49 lacks significant inotropic effect. J Cardiovasc Pharmacol. 1991;19:264-271.
Nikolic S, Yellin EL, Tamura K, Tamura T, Frater RWM. Effects of early diastolic loading on myocardial relaxation in the canine left ventricle. Circ Res. 1990;66:1217-1226.
Nikolic S, Yellin EL, Tamura K, Vetter H, Tamura T, Meisner JS, Frater RWM. Passive properties of the canine left ventricle: diastolic stiffness and restoring forces. Circ Res. 1988;62:1210-1222.
Nikolic SD, Feneley MP, Pajaro OE, Rankin JS, Yellin EL. Origin of regional pressure gradients in the left ventricle during early diastole. Am J Physiol. 1995;268:H550-H557.
Ohtani M, Nikolic SD, Glantz SA. A new approach to in situ left ventricular volume clamping in dogs. Am J Physiol. 1991;261:H1335-H1343.
Theroux P, Franklin D, Ross JR Jr, Kemper WS. Regional myocardial function during acute coronary arterial occlusion and its modification by pharmacologic agents in the dog. Circ Res. 1974;35:896-908.
Brutsaert DL, Sys SU. Relaxation and diastole of the heart. Physiol Rev. 1989;69:1228-1315.
Pouleur H, Karliner JS, LeWinter MM, Covell JW. Diastolic viscous properties of the intact canine left ventricle. Circ Res. 1979;45:410-419.
Nikolic SD, Tamura K, Tamura T, Dahm M, Frater RWM, Yellin EL. Diastolic viscous properties of the intact canine left ventricle. Circ Res. 1990;67:352-359.