INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

DEMAND ISCHEMIA, induced by rapid pacing in the presence of critical coronary stenosis, shifts the diastolic pressure-volume curve upward (3, 5-7, 21-23). The pathophysiological mechanism of this shift is not yet established, although several mechanisms have been proposed, generally relating to the level of Ca²⁺ within the sarcomere during diastole. The fall in Ca²⁺ level during diastole is due to Ca²⁺ reuptake into the sarcoplasmic reticulum and by Ca²⁺ extrusion through the sarcolemma by the sarcolemmal Ca²⁺ pump and by Na⁺/Ca²⁺ exchange. Treatment with agents that interfere with sarcoplasmic reticulum function, such as caffeine (1) and thapsigargin (17), did not affect the intracellular Ca²⁺ concentration ([Ca²⁺]_i) during hypoxia in isolated cells. Therefore, we hypothesized that agents that affect Na⁺/Ca²⁺ exchange, such as lidocaine, a class 1b-type Na⁺-channel blocker that decreases intracellular Na⁺ (33), could affect the diastolic pressure-volume relationship because of indirect effects on intracellular Ca²⁺. Lidocaine is a drug widely used to treat arrythmias in patients with myocardial ischemia. We studied the effects of lidocaine on diastolic dysfunction associated with demand ischemia. We compared diastolic (as represented by the shift in the diastolic pressure-volume relationship) and systolic function during demand ischemia before and after lidocaine injection. The upward shift of the diastolic pressure-volume curve with demand ischemia was significantly greater following lidocaine administration, suggesting that lidocaine could aggravate diastolic dysfunction in patients with ischemic heart disease.

	METHODS

Top Abstract Introduction Methods Results Discussion References

All animal work was approved by the University of California, San Francisco, Committee on Animal Research.

Surgical Preparation

Instrumentation

View larger version (36K): [in this window] [in a new window]   Fig. 1.   Instrumentation of pig heart. LAD, left anterior descending artery; LAP, left atrial pressure; LCX, left circumflex coronary artery; LVP, left ventricular pressure; IVC, inferior vena cava.

To measure LV volume, a 7-F, eight-electrode, single-field conductance catheter (Webster Laboratories) was inserted in the LV through the LV apex with the tip electrode above the aortic valve and with the most proximal electrode at the apex. The conductance catheter was connected to a signal processor (Stitching Leycom-Sigma-5DF), which converts the conductance signal into volume (2). The conductance catheter and signal processing electronics are designed to have the catheter introduced through the aortic valve, not the apex. To obtain correct volume results with the catheter introduced through the apex, we used a custom-made adapter that reversed the electrode sequence of the catheter. The catheter was checked for proper positioning by comparing the individual segment signals. Because we were interested in volume changes and not absolute volumes, we based our analysis on uncorrected total conductance volume signals (4).

To confirm the absence of impaired regional systolic function when creating a critical coronary stenosis, and to assess the pressure-segment length relationship before and after creating demand ischemia, pairs of ultrasonic crystals were implanted in the areas perfused by the LAD and left circumflex coronary artery (LCX). The crystals were implanted parallel to the LV short axis and were positioned in the circumferential plane. The distance between each pair of crystals was ~1.5 cm. The crystals were connected to a sonomicrometer (Triton Technology).

To measure the coronary blood flow in the LAD, the short segment of the LAD just distal to the bifurcation of the first diagonal branch was carefully dissected and a transit-time ultrasonic flow probe (Transonic Systems) was placed around the LAD.

All hemodynamic signals [LV pressure, LA pressure, LV volume, LAD coronary flow, electrocardiogram (ECG), and segment lengths in the areas perfused by the LAD and LCX] were monitored continuously on a multichannel oscillograph and digitized at 200 Hz with a personal computer. The peak rate of LV pressure development (dP/dt_max) was monitored throughout the experiment as an index of LV contractility.

To create a coronary stenosis on the LAD, an adjustable occluder (24, 28, 29) was placed around the LAD just proximal to the bifurcation of the first branch. The adjustable occluder consisted of a head made of metal and a body made of plastic; the head was designed like a screw, and the body was designed in a "C" shape to be placed around the vessel in a stable standing position.

To measure myocardial pH continuously, a pH sensor (Micro-combination pH Electrodes) was calibrated with the use of buffers with pH 4.0 and 7.0, and then a H⁺-selective polymer membrane pH electrode was inserted into the myocardium perpendicular to the surface of the myocardium in the region supplied by the LAD so that the tip of the sensor was firmly placed in the endocardium.

To pace the heart, electrodes were attached to the LA using a demand-type stimulator (Medtronic Pacing System Analyzer model 5309).

To produce a range of pressures and volumes to measure end-diastolic pressure-volume relationships (EDPVR), an occlusion balloon catheter (8-F Fogarty venous thrombectomy catheter) was inserted through a femoral venous sheath into the inferior vena cava.

The pigs were given intravenous heparin (5,000 U) as an anticoagulant after the instrumentation had been completed.

Protocol

After instrumentation was completed and a hemodynamic steady state was achieved, we measured the blood conductiv-ity () to convert conductance into volume.  was also measured before each data recording episode.

LV pressure, LA pressure, LV volume, ECG, LAD coronary flow, and the segment lengths in the areas perfused by the LAD and LCX were recorded with the respirator stopped at end expiration. Next, preload was reduced by inflating the balloon in the inferior vena cava during data acquisition, and data were recorded during the transient phase of the vena cava occlusion when pressure was falling. We analyzed EDPVR and ESPVR by combining the data collected during the steady-state and declining phases of the vena caval occlusion. Just after the hemodynamic data were recorded during baseline for over 30 s, we printed out the LV pressure-volume loop and the EDPVR from the personal computer used to record the data.

To create demand ischemia, we produced a stenosis by constricting the adjustable vessel occluder on the LAD just proximal to the bifurcation of the first diagonal branch to reduce the peak anterograde flow velocity in the distal LAD by ~30% (27) while preserving the rest level of systolic shortening of the segment perfused by the LAD. This degree of reduction in coronary flow corresponds to a 75-90% lumen stenosis (9).

After the LAD stenosis was adjusted to reduce the peak coronary flow velocity by ~30%, we paced the heart at 1.5 times the resting heart rate for 1.5 min. We recorded hemodynamic variables immediately after cessation of the rapid pacing and then released the occlusion of the LAD.

To compare the EDPVR during demand ischemia with that during baseline, we printed out both the LV pressure-volume loop and the EDPVR during demand ischemia and compared them with those obtained during baseline. We drew the regression lines for both EDPVRs, regarding the maximal end-diastolic volume points on the lines as the end points, and then defined the upward-shift index (U) as end-diastolic pressure during demand ischemia minus the pressure during baseline at the largest end-diastolic volume common to both conditions (Fig. 2A). If we did not obtain the upward shift (positive U) in EDPVR, we waited at least 10 min after the previous demand ischemia for hemodynamics to stabilize and then performed another demand-ischemia run in which we changed the extent of coronary stenosis, the rate of rapid pacing, or the duration of rapid pacing. If we observed a rightward shift instead of an upward shift, we reduced the occlusion to increase coronary flow and improve washout of metabolites. If we observed no shift, we occluded the LAD a little more tightly and then paced using the same rate and duration as before to increase the level of demand ischemia. If this procedure did not produce an upward shift, we increased the pacing rate (up to 1.8 times the resting heart rate) for the same duration as before. If neither occluding the LAD a little more tightly nor increasing the rate of rapid pacing to 1.8 times the resting heart rate produced an upward shift, we extended the duration of rapid pacing to 5 min. We repeated this procedure until an upward shift of the EDPVR was observed.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Definition of upward-shift index (U; A) and rightward-shift index (R; B) for end-diastolic pressure-volume relationship between baseline and demand ischemia.

Table 1 shows that we obtained an upward shift in EDPVR on the first try in one pig (no. 704), the second try in three pigs (nos. 707, 709, and 711), and the third (no. 713), fourth (no. 705), fifth (no. 703), and eighth (no. 706) try in one pig each. In three experiments (nos. 703, 705, and 706), we infused phenylephrine (4 mg/h) to keep peak systolic pressure >100 mmHg. In one experiment (no. 706), dP/dt_max was <1,300 mmHg/s during baseline, so we infused milrinone (900 µg/h) to keep dP/dt_max >2,000 mmHg/s. The baseline LAD flow before occlusion was 42.9 ± 7.0 ml/min (mean ± SE); LAD flow was reduced to 66.3 ± 3.5% of baseline when we obtained the upward shift of the diastolic pressure-volume curve (Table 1). The baseline heart rate was 90.5 ± 15.5 beats/min, the pacing rate was 1.61 ± 0.17 times the resting heart rate, and the pacing duration was 1.88 ± 0.69 min when we obtained the upward shift of the diastolic pressure-volume curve during demand ischemia.

We waited for at least 10 min after recording the data for baseline demand ischemia to allow hemodynamics to stabilize while the LAD was occluded. We then administered 1% lidocaine (5 mg/kg iv). We waited at least 30 min after lidocaine injection before measuring postlidocaine baseline. We used the 5 mg/kg lidocaine dose because Matsumura et al. (18) have previously shown that lidocaine at the dose of 5 or 10 mg/kg neutralized the myocardial acidosis induced by partial LAD occlusion in dogs ~30 min after the occlusion, but that lidocaine at the dose of 10 mg/kg decreased systolic pressure. Therefore, use of the 5 mg/kg dose avoided the situation in which lidocaine impaired systolic pump performance.

Finally, we created the same level of stenosis as before lidocaine was administered, defined as the level of stenosis that reduced LAD flow by the same percentage as before. Next, we paced the heart at the rate that corresponded to the same multiple of the heart rate observed during baseline after lidocaine injection, for the same duration used to create demand ischemia before the lidocaine injection (Table 1). We recorded the hemodynamics and the LV pressure-volume loop and the LV EDPVR immediately after terminating pacing, as before.

Data Acquisition

Data Analysis

Hemodynamic variables in steady-state beats. Hemodynamic variables from steady-state beats were calculated as the mean of the variables for the first seven beats in each recording.

/

Pressure-volume loop of single beats. To examine the shift of the LV pressure-volume loops, a single pressure-volume loop during steady-state beats for each episode was plotted and visually compared in each pig. To evaluate the magnitude of the upward shift during demand ischemia, we compared the pressures at the volume corresponding to end diastole during baseline for representative single loops during baseline and demand ischemia. We repeated this comparison after lidocaine injection.

Diastolic function. The LV EDPVR was defined as the end-diastolic pressure-volume points obtained during vena caval occlusion. First, end-diastolic pressure-volume points for each of the four episodes of the eight experiments were plotted and visually compared in each pig. We then conducted a linear regression of end-diastolic pressure against end-diastolic volume (Fig. 2). The shift of the EDPVR from baseline to demand ischemia was quantified by an upward-shift index, U, as the end-diastolic pressure during demand ischemia minus the pressure during baseline at the largest end-diastolic volume common to both conditions (Fig. 2A). We also computed a rightward-shift index, R, as the largest end-diastolic volume during demand ischemia minus the largest end-diastolic volume during baseline (Fig. 2B).

Systolic function. To define the LV ESPVR, we applied linear regression to the LV end-systolic points of the pressure-volume loops during vena caval occlusion. The linear ESPVR was defined as where P_es is the end-systolic pressure, V_d is the volume-axis intercept (26), and E_es is the end-systolic elastance. E_es and V_d were estimated by use of the iteration procedure described by Kono et al. (13).

Statistics

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Hemodynamic Variables

Lidocaine did not change the hemodynamic variables significantly, after we controlled for the presence of ischemia (see P values for main effect of lidocaine in 2-way ANOVA, Table 2). This result is consistent with earlier work by Matsumura et al. (18).

Lidocaine had a significant interaction effect with demand ischemia for LV end-diastolic pressure (P = 0.04) and myocardial pH (P < 0.01). The increase in LV end-diastolic pressure during demand ischemia was greater in the presence of lidocaine (Table 2). As expected (12, 18), lidocaine blunted the change in myocardial pH that occurred during demand ischemia (Table 2). (This result is not surprising, because H⁺ reduces Ca²⁺ sensitivity of myofilaments so that for any level of diastolic Ca²⁺, force will be higher at higher pH.) These results suggest that lidocaine worsens the effects of demand ischemia on diastolic, but not systolic, function.

Changes in Single Loops of LV Pressure-Volume Curve

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Continuous recordings of diastolic section of left ventricular pressure-volume loops for all pigs.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Lidocaine increases upward shift of diastolic pressure-volume (PV) loop. Symbols indicate results from individual pigs (n = 7).

Determinants of Shift of EDPVR

View larger version (34K): [in this window] [in a new window]   Fig. 5.   End-diastolic pressure-volume relationships for all pigs.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Lidocaine increases upward shift (A) and reduces rightward shift (B) of end-diastolic pressure-volume relationship. Symbols indicate results from individual pigs (n = 8).

The magnitude of the upward shift index, U, was significantly correlated with baseline contractility, quantified as dP/dt_max (r = 0.74, P = 0.036). This result is consistent with earlier work (21, 29) showing that the higher baseline contractility, the greater the upward shift with demand ischemia.

Lidocaine also affected the rightward shift of the EDPVR that accompanied demand ischemia (P = 0.039). Without lidocaine, the index of rightward shift of the EDPVR curve, R, shifted rightward with demand ischemia by 3.71 ± 5.87 ml (range: 5.75 to 12.1 ml); after lidocaine, demand ischemia was associated with a leftward shift, quantified as R = 3.19 ± 6.28 ml (range: 13.1 to 4.77 ml) (Fig. 6B).

Determinants of LV Contractility

Myocardial pH

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The primary finding of this study is that lidocaine, in doses small enough to avoid depression in systolic function, increases the upward shift in the diastolic pressure-volume curve produced by demand ischemia.

Several hypotheses have been proposed to explain the upward shift of the diastolic pressure-volume curve during demand ischemia: 1) increased diastolic Ca²⁺ secondary to reduced sarcoplasmic reticulum ATPase activity (5, 11, 21); 2) increased diastolic Ca²⁺ secondary to increased Ca²⁺ influx from Na⁺/Ca²⁺ exchange in the setting of increased intracellular Na⁺ due to ischemia; 3) increased diastolic Ca²⁺ secondary to increased inward Ca²⁺ current through L-type Ca²⁺ channels; 4) no increase in diastolic Ca²⁺, but impaired cross-bridge inactivation on an energetic basis (32); and 5) increased diastolic Ca²⁺ secondary to stretch-activated channels (24, 28). The results of our experiments with lidocaine are inconsistent with the first two hypotheses and consistent with the last three hypotheses.

In support of persistence of Ca²⁺ from the previous systole as a mechanism of the upward shift, Kihara et al. (11) demonstrated that during 5 min of hypoxia, both free Ca²⁺ concentration and LV isovolumic pressure increased during diastole and decreased during systole. Authors (20) who support the residual cross-bridge interaction theory propose that a buildup of tissue metabolites such as H⁺ and Na⁺ due to decreased washout by coronary flow reduced the upward shift and increased the rightward shift of the diastolic pressure-volume curve during ischemia. These metabolites, as well as intracellular acidosis (induced by accumulated H⁺), inhibit contractile activity by reducing the Ca²⁺ sensitivity of the myofilaments and thus induce contractile failure despite higher myoplasmic Ca²⁺. Lidocaine also reduces tissue metabolites such as Na⁺ and H⁺. There is also evidence that lidocaine reverses increases in cellular Ca²⁺ that occur during ischemia associated with ventricular fibrillation. Stefenelli et al. (25) subjected isolated rat hearts to 1 min of pacing-induced ventricular fibrillation and found that diastolic [Ca²⁺]_i increased above the end-systolic [Ca²⁺]_i of normal beats. Lidocaine reversed this effect; [Ca²⁺]_i during fibrillation returned to the levels observed during baseline conditions, and the hearts converted to sinus rhythm. In the present experiments, however, lidocaine increased the upward shift and reduced the rightward shift of the diastolic pressure-volume curve.

There is good evidence from studies of isolated muscle (14, 34) that Na⁺/Ca²⁺ exchange accounts for most of the sarcolemmal Ca²⁺ exchange [77% according to one study (35)] and that this exchange is more related to the resting tension than to the sarcoplasmic reticulum-mediated Ca²⁺ flux. Resting tension increases when cardiac muscle is put in a low-Na⁺ or Na⁺-free bath (14, 15) because there is a buildup of intracellular Ca²⁺ in exchange for the outward flux of Na⁺ driven by the Na⁺ concentration gradient. Conversely, lower levels of intracellular Na⁺ reduce the concentration gradient that drives Na⁺ out of the cell and Ca²⁺ into the cell via Na⁺/Ca²⁺ exchange. Thus intracellular Na⁺ accumulation is correlated both with Ca²⁺ levels and increases in diastolic pressure in isolated rat hearts (30, 31).

If the upward shift of the diastolic pressure-volume curve that accompanies demand ischemia was due to an increase in the background level of Ca²⁺ within the sarcomere, we would expect lidocaine to attenuate the effects of demand ischemia by blocking Na⁺ influx. We found that lidocaine had just the opposite effect. Thus our results are inconsistent with the hypothesis that the upward shift of the diastolic pressure-volume curve is due to an increase in the background level of Ca²⁺ within the sarcomeres during diastole.

Our results are consistent with, but do not prove, the stretch-activation hypothesis. In support of stretch activation of the myocardium as a mechanism, Shintani and Glantz (24) used an LV volume clamp to show that during demand ischemia, the LV diastolic pressure-volume curve did not shift upward without filling of the ventricle, i.e., without stretch of the myocardium during filling. According to this hypothesis, when the myocardium is ischemic, the sarcolemma and the sarcoplasmic reticulum may become more likely to release Ca²⁺ in response to stretch. The released Ca²⁺ increases actin-myosin interaction during diastole, resulting in the upward shift of the diastolic pressure-volume curve. Further support is that gadolinium, a stretch-activated channel blocker, attenuated the shift in the diastolic pressure-volume curve associated with demand ischemia (28).

Limitations

The dose of lidocaine we used was selected to be small enough to avoid depression of systolic function, so there is some concern that we were not producing a substantial effect on the Na⁺ channel. The fact that we found changes in diastolic function, however, is evidence that the dose we used was adequate to produce physiological changes.

The data for myocardial pH during baseline before and after lidocaine injection are different (7.57 ± 0.02 vs. 7.47 ± 0.05, P < 0.01). Lidocaine was administered after data were collected from the first period of demand ischemia. Before lidocaine was administered, we waited at least 10 min for hemodynamic variables (pressures, heart rate, and coronary flow) to return to baseline levels (Table 2). Tissue pH did not return to control levels before lidocaine was administered. In addition, the increase in pH may have counteracted and overwhelmed the effects of reduced intracellular Ca²⁺ in our experiments because the increase in pH is associated with increased troponin sensitivity to Ca²⁺, which would lead to increased force for any given Ca²⁺ concentration. We do not know what effect, if any, these modest differences in baseline pH could have had on our results. It was clear, however, that the lidocaine blunted the effects of ischemia on changing pH.

The nature of the protocol required us to create several episodes of ischemia, which could have produced stunning or preconditioning. Either of these effects could have introduced artifacts in the second demand ischemia period, in which we administered the lidocaine. We sought to avoid this problem by waiting at least 40 min after the first demand ischemia episode before recording the postlidocaine baseline data. The fact that both diastolic and systolic function, measured as diastolic and systolic pressures, segment shortening, and stroke volume and stroke work, were similar during the initial baseline and lidocaine baseline (Table 2) suggests that neither stunning nor preconditioning were a problem in our preparation.

These experiments were done in open-chest anesthetized pigs. This preparation may have led to changes in coronary flow, and the responses to ischemia that could be different from what would have been observed in conscious animals.

In conclusion, we found that the Na⁺-channel blocker lidocaine increased the magnitude of the shift of the diastolic pressure-volume curve that accompanies demand ischemia. This result is evidence against the hypothesis that this upward shift is due to an increase in the background level of Ca²⁺ in the myocardium. It also points to the need for clinical studies to assess the effects of lidocaine on diastolic function in patients with ischemic heart disease. The fact that lidocaine adversely affected the response of the heart to demand ischemia during diastole at a dose small enough to avoid adverse effects on systolic function is further evidence that diastolic dysfunction appears before systolic dysfunction.