INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HYPERTHERMIA (HT) (<41°C) increases cardiac output and myocardial oxygen consumption (MO₂) due to increasing heart rate and cardiac work in the isolated and in situ heart (4, 6, 10, 13, 16). A few investigators (5, 35), in contrast, reported that HT adversely reduced cardiac function. Although these hemodynamic effects suggest that HT may be unfavorable for patients with cardiovascular diseases, the exact cardiac effect of HT remains controversial. These controversies might have resulted from the uncontrolled heart rate and cardiac loading and the evaluation of left ventricular (LV) contractility by load-dependent indexes. To clarify this issue, a precise control of cardiac loading conditions and appropriate utilization of a load-independent index of contractility would be necessary.

In addition, variation of temperature has been reported to change the activities of many enzymes in the myocardium (14-15, 24). This effect of varied temperature on the myocardium suggests that the key processes in myocardial contraction [i.e., basal metabolism, excitation-contraction (E-C) coupling, and cross-bridge cycling] are also affected. In previous studies (10, 16, 24), however, these fractions in cardiac energetics have not been differentiated. Moreover, it is still unclear whether and how HT influences the cardiac chemomechanical efficiency and the oxygen cost of contractility.

To assess the mechanoenergetic effects of HT, we used the LV contractility index [E_max; the slope of the end-systolic pressure-volume relation (ESPVR)] as a relatively load-independent index of LV contractility (26) and LV pressure-volume (PV) area (PVA) as a measure of the total mechanical energy generated by the LV (26). The relation between MO₂ and PVA is proven to be linear and independent of various ventricular loading conditions in a stable contractile state (26). The reciprocal of the slope of the linear MO₂-PVA relation is considered to reflect the contractile efficiency of the energy conversion from MO₂ to PVA via ATP (26). The MO₂ intercept of this relation reflects the MO₂ fraction of E-C coupling and basal metabolism (26). With the use of this mechanoenergetic framework of the E_max-PVA-MO₂ relation, we could precisely analyze and characterize the relation between changes in LV contractility and mechanoenergetic cost during HT.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Surgical Preparation

A thin latex balloon with a miniature pressure gauge (P-6, Konigsberg) was placed in the LV. The balloon was primed with water and connected to a volume servo pump, which accurately controlled and measured LV volume. A LV epicardial electrocardiogram was recorded to trigger the volume command signal of the servo pump. Heart rate was held constant by left atrial pacing throughout the experiments.

Coronary blood flow was measured with an electromagnetic flowmeter (MVF-2100, Nihon Kohden) in the coronary venous drainage tube from the right heart. We neglected LV thebesian venous blood flow because of its small fraction (<3%) in the total coronary flow (20). Coronary arteriovenous oxygen content difference (AVO₂D) was measured continuously with a custom-made oxygen content difference analyzer (PWA-2000S, Erma) (27). This analyzer was calibrated against an IL-286 CO-oximeter in each experiment.

The temperature of the heart was measured with a thermister probe (TF-DNP-1, Termo) placed between the endocardium and the balloon via the apical stab wound. The heart temperature was gradually increased from 36-37 to 40-42°C by warming the arterial cross-circulation tube in a thermostat bath.

Mean arterial blood pressure of the support dog served as the coronary perfusion pressure of the heart preparation. This pressure was maintained above 80 mmHg by infusing either fresh blood, which had been collected from the heart donor dog, or 10% Dextran 40 solution as needed. Arterial pH, PO₂, and PCO₂ were maintained within their physiological ranges by using supplemental oxygen and intravenous sodium bicarbonate if necessary.

Experimental Protocol

Volume run under normothermia (normothermia-volume run; n = 12). In a stable, baseline contractile state under normothermia (NT) (heart temperature 36.3 ± 0.3°C), we produced isovolumic contractions at 6-10 different LV volumes to obtain a baseline relation between MO₂ and PVA. We waited 2-3 min after each change in LV volume until the cardiac variables reached a new steady state. In the present study, we used isovolumic contractions to avoid confounding effects of ejection on the cardiac contractility and energetics (26).

Calcium run under NT (NT-Ca²+ run; n = 8). In this run, LV volume was fixed at a moderate level (23.0 ± 1.5 ml), where peak isovolumic pressure was ~100 mmHg before calcium (1% CaCl₂) infusion. Calcium was continuously infused into the coronary arterial perfusion tube with an infusion pump (STC-521, Termo) at a starting rate of 0.01 meq · l¹ · min¹. The infusion rate was increased in steps every 5 min until E_max was nearly doubled to obtain six to eight sets of mechanoenergetic data at the preset volume. The calcium infusion was then stopped, and 10-15 min was allowed to elapse for the return of contractility to the precalcium baseline level.

HT run (n = 8). We fixed LV volume at the same LV volume as in the NT-Ca²⁺ run under NT. Heart temperature was then gradually increased from 36 to 41°C. During this procedure, all data were repeatedly obtained.

Volume run under HT (HT-volume run; n = 12). After the heart temperature reached the target temperature (40-42°C), we obtained the MO₂-PVA relation and other variables in a manner similar to the NT-volume run.

Calcium run under HT (HT-Ca²+ run; n = 8). At the highest heart temperature (40-42°C), LV volume was fixed at the same LV volume as in the NT-Ca²⁺ run. We then repeated the calcium infusion in a manner similar to the NT-Ca²⁺ run until E_max was enhanced to the baseline level observed in the NT-volume run.

Calcium volume run under HT (HT-Ca²+ volume run; n = 6). When E_max was steadily enhanced to the NT level during the HT-Ca²⁺ run, we obtained another MO₂-PVA relation in a manner similar to the NT-volume and HT-volume runs.

Post-HT volume run (n = 6). After the hyperthermic protocols,we readjusted the heart temperature to 36-37°C. In a stable contractile state, we obtained the MO₂-PVA relation during NT.

KCl arrest under NT and HT (n = 6). Six hearts were arrested at the volume at which peak isovolumic pressure was zero (V₀) by injecting KCl (5- to 6-ml bolus dose of 0.75 mol/l) into the coronary arterial tube. After both coronary blood flow (CBF) and AVO₂D were stabilized under NT, we determined MO₂ as the basal metabolic MO₂ under NT. As heart temperature was then increased to the same temperature as in the previous HT run, we obtained another basal metabolic MO₂ under KCl arrest during HT.

Data Analysis

Contractility Index

1

1

1

Pressure-Volume Area

1

1

View larger version (16K): [in this window] [in a new window]   Fig. 1.   A: schematic illustration of left ventricular (LV) systolic pressure-volume (P-V) area (PVA) for isovolumic contraction in P-V diagram. PVA is area surrounded by end-systolic P-V relation (ESPVR), end-diastolic P-V relation (EDPVR), and systolic P-V trajectory. Slope of ESPVR is defined as ventricular contractility index [maximum elastance (Emax)]. B: relation between myocardial oxygen consumption (MO2) and PVA. The MO2-PVA relation is linear and represented as MO2= a · PVA + b, where a · PVA indicates PVA-dependent MO2, which increases in proportion to PVA. Slope (a) of this relation reflects oxygen cost of PVA. Intercept (b) of this relation indicates PVA-independent MO2, which consists of MO2 for excitation-contraction (E-C) coupling and basal metabolism. C: MO2-PVA data points during an enhancement of Emax during a calcium run. MO2-PVA data points () shift right and upward with an increase in Emax. Four diagonal solid lines represent the MO2-PVA relations in the respective Emax levels. The MO2 intercepts () of these MO2-PVA relations, which indicate PVA-independent MO2 for 4 different Emax levels, also increase. D: relation between Emax and PVA-independent MO2 derived from C. Slope of this relation is designated as oxygen cost of contractility.

Myocardial Oxygen Consumption

Oxygen Cost of Contractility

We also calculated the ratio of the change in () PVA-independent MO₂ to E_max from a set of data in the HT, NT-Ca²⁺, and HT-Ca²⁺ runs. In these calculations, E_max was matched among the three runs.

Statistics

A value of P < 0.05 was considered statistically significant. Data are presented as means ± SD.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of HT on LV function

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Simultaneous recordings of left ventricular pressure (LVP), LV volume (LVV), coronary blood flow (CBF), and arteriovenous oxygen content difference (AVO2D) and an electrocardiogram (ECG) in normothermic (NT; A) and hyperthermic (HT; B) conditions in a representative heart.

Table 1 compares the mean values of cardiac variables between NT and HT conditions at the fixed LV volume. Heart rate was not significantly different between both conditions; in three experiments, heart rate under HT conditions exceeded the cardiac pacing rate. During HT, LV peak pressure at the same LV volume, and hence E_max, significantly decreased by 36 and 38%, respectively (both P < 0.001), whereas dP/dt_max decreased by 24%. Both CBF and AVO₂D slightly decreased in HT, but these were not significantly different between the two conditions. The two energetic variables, PVA and MO₂, significantly decreased during HT by 38 and 19%, respectively (both P < 0.001). LV diastolic function was assessed by the LV end-diastolic pressure and relaxation time constant. Both the LV end-diastolic pressure and time constant decreased during HT, indicating that HT augmented the rate of relaxation in LV.

Figure 3A shows a representative example of ESPVR during NT and HT in a heart. Both ESPVRs were linear over a wide range (both r > 0.98). The slope of ESPVR (E_max ) was decreased in HT (NT: 10.8 vs. HT: 6.1 mmHg · ml¹ · 100 g LV), indicating that LV contractility decreased in the HT condition over the full range of LV volume. In 12 hearts, E_max was significantly decreased in HT by 35.2 ± 10.7% (P < 0.001 by paired t-test; Table 2).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Graphs showing representative examples of ESPVR (A) and the relation between the MO2 and PVA (B) obtained in the NT and HT volume runs in a heart.

Effect of HT on Cardiac Energetics

5

5

1

1

1

1

In all hearts, both relations were highly linear under both the NT and HT conditions, and there was no significant difference in the slope of the MO₂-PVA relation by ANCOVA. In 4 of 12 hearts, there was a significant elevation difference by ANCOVA between the NT and HT conditions, but these differences in the MO₂ intercept were practically small. Averaged data for the slope and MO₂ intercept of the MO₂-PVA relation were not significantly different by paired t-test between the two conditions (Table 2).

Figure 4A shows the MO₂-PVA data points and MO₂-PVA regression lines in the NT-volume and NT-Ca²⁺ runs in the same heart as shown in Fig. 3. As E_max was enhanced by calcium infusion, the MO₂-PVA data points shifted right and upward from the MO₂-PVA regression line in the NT-volume run (NT-Ca²⁺: MO₂ = 3.1 × 10⁵ PVA + 0.018, r = 0.997; NT-volume: MO₂ = 1.3 × 10⁵ PVA + 0.024, r = 0.995).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   A: MO2 and PVA data points and the regression line obtained from the NT-Ca2+ run and NT-volume run. B: MO2 and PVA data points and the regression lines obtained in the HT-volume, HT-Ca2+, and HT-Ca2+ volume runs. Note that the regression line in the HT-Ca2+ volume run shifts upward in a parallel manner from that in the HT-volume run.

Figure 4B shows the MO₂-PVA data points and regression lines during the HT-volume, HT-Ca²⁺, and HT-Ca²⁺ volume runs. When depressed contractility during HT was restored to the level of the NT condition with calcium in the HT-Ca²⁺ volume run (E_max: from 6.2 to 10.8 mmHg · ml¹ · 100 g LV), the slope of the MO₂-PVA relations remained unchanged (from 1.32 × 10⁵ to 1.45 × 10⁵ ml O₂ · mmHg¹ · ml¹, P = NS by ANCOVA), but the MO₂ intercept increased from 0.024 to 0.033 ml O₂ · beat¹ · 100 g LV¹ (P < 0.01 by ANCOVA). These indicate that the MO₂-PVA relation was elevated in a parallel manner in the HT-Ca²⁺ volume run. In each heart, ANCOVA showed that the difference in the slope of the MO₂-PVA relation between the HT-volume and HT-Ca²⁺ volume runs was not significant, whereas the elevation difference was significant. In all six hearts, the paired t-test indicated no significant difference in the slope of the MO₂-PVA relation between the HT-volume and HT-Ca²⁺ volume runs (1.64 ± 0.32 vs. 1.66 ± 0.25 × 10⁵ ml O₂ · mmHg¹ · ml¹, P = NS), but a significant difference in the MO₂ intercept (0.022 ± 0.005 vs. 0.028 ± 0.006 ml O₂ · beat¹ · 100 g LV¹, P < 0.01) was found. The MO₂-PVA data points in the HT-Ca²⁺ run shifted in the same manner as those in the NT-Ca²⁺ run (HT-Ca²⁺: MO₂ = 3.6 × 10⁵ PVA + 0.018, r = 0.995).

The amount of infused calcium to achieve the same E_max was greater in the HT than NT conditions (calcium/E_max: NT 4.8 ± 1.7 vs. HT 5.8 ± 2.7 mg · ml · mmHg¹, P < 0.05 by paired t-test). This result demonstrates that, in the HT condition, a greater calcium delivery is required to enhance LV contractility to the same extent.

Effect of HT on Oxygen Cost of Contractility

1

1

2

1

1

1

1

2

1

1

View larger version (17K): [in this window] [in a new window]   Fig. 5.   A: relations between the PVA-independent MO2 and Emax obtained in the NT-Ca2+ and HT-Ca2+ runs in the same heart as in Fig. 3. The slope of each regression line indicates oxygen cost of contractility. Note that the slope of the regression line is greater in HT than in NT. B: relations between the PVA-independent MO2 and Emax obtained in the HT run. Both regression lines in the NT-Ca2+ and HT-Ca2+ runs are the same as shown in A.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Comparison of the slope (oxygen cost of contractility; A) and intercept (B) of the relation between PVA-independent MO2 and Emax in NT-Ca2+ and HT-Ca2+ runs. Group data are presented as means ± SD. NS, not significant.

Figure 7 shows average responses of MO₂ /PVA (i.e., the oxygen cost of PVA) and PVA-independent MO₂/E_max (i.e., the oxygen cost of E_max) among the HT, NT-Ca²⁺, and HT-Ca²⁺ runs. The slope of MO₂ /PVA was greater in the HT-Ca²⁺ run than in the NT-Ca²⁺ run. These responses of MO₂/PVA in both runs shifted right and upward from the average regression line in the HT-volume run. In contrast, MO₂ /PVA in the HT run changed along the average regression line of the HT-volume run (Fig. 7A). The slope of PVA-independent MO₂/E_max was greater in the HT-Ca²⁺ run than in the NT-Ca²⁺ run (NT-Ca²⁺: 0.0013 ± 0.0002 vs. HT-Ca²⁺: 0.0025 ± 0.0008 ml O₂ · ml · mmHg¹ · beat¹ · 100 g LV²). The slope in the HT run showed a near-zero oxygen cost of contractility (HT: 0.5 × 10⁵ ± 0.0009 ml O₂ · ml · mmHg¹ · beat¹ · 100 g LV²; Fig. 7B). These results in the HT condition (i.e., the HT and HT-Ca²⁺ runs) indicated that the oxygen cost of contractility under the HT condition was greater than that in the NT condition.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Graphs showing the changes in () MO2 and PVA (MO2/PVA; A) and the changes in PVA-independent MO2 and Emax (PVA-independent MO2/Emax; B) from their respective baseline values in HT, NT-Ca2+, and HT-Ca2+ runs. In A, the dashed line indicates averaged data of the MO2-PVA relation in the HT-volume run.

Recovery From HT to NT

1

5

1

1

1

1

Effect of HT on Basal Metabolism

1

1

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This study investigates the effects of HT on LV mechanoenergetics in the isolated, blood-perfused canine heart. Our findings demonstrate that HT depresses LV contractility and shortens time to peak contraction. Despite the depressed LV contractility, PVA-independent MO₂ remains unchanged in HT. When the depressed contractility under HT is restored to the baseline level (NT condition) by calcium infusion, the PVA-independent MO₂ under HT is markedly higher than that under NT. Moreover, the oxygen cost of contractility under HT is significantly higher than that under NT.

Effect of HT on LV Contractility

D'ambra et al. (5) reported that the regional LV function assessed by the shortening fraction depressed by 42% when myocardial temperature was increased from 38 to 42°C; the pressure-length loop area was inversely related with myocardial temperature. Templeton et al. (35) showed that HT from 37 to 40°C depressed LV stiffness, assessed by sinusoidal forcing function, by 17% in a heart-lung bypass preparation. Our present results, assessed by E_max and dP/dt_max at the same LV volume, clearly demonstrate that HT reduces LV contractility by itself. Our results have confirmed that HT depresses LV contractility, as was found in those previous studies (5, 35).

Effect of HT on Cardiac Energetics

Varied heart temperature affects myosine ATPase activity and the cross-bridge cycling rate (both the reported rates of change with a 10°C increase in temperature, Q₁₀ = 2-3) (1, 7). This finding suggests that HT increases the oxygen consumption in cross-bridge cycling and hence the PVA-dependent MO₂. The present study, however, shows that HT does not change the PVA-dependent MO₂ and contractile efficiency. Simultaneously, we observe that HT shortens the time to peak contraction (i.e., T_max) and time constant despite the negative inotropic effect, suggesting that HT enhances the attachment and detachment rates of cross-bridge cycling. Thus the increased myosine ATPase activity and cross-bridge cycling rate by HT seems to cause the reduced T_max and time constant rather than affecting the PVA-dependent MO₂ and contractile efficiency.

In the present study, the most interesting point is the unchanged PVA-independent MO₂ despite the depressed contractility in HT. The MO₂ component for basal metabolism under KCl arrest increases during HT. This suggests that the unchanged PVA-independent MO₂ results from a combined effect of the increased basal metabolism and a decreased E-C coupling energy, such as those under propranolol. In our previous observations (29), propranolol lowered the PVA-independent MO₂ by 25% without a change in basal metabolism when E_max decreased by 48% in the same experimental preparation. In contrast, HT does not change the PVA-independent MO₂ but increases the basal metabolic MO₂ by only 7% of the PVA-independent MO₂, whereas HT decreases E_max by 38%. Therefore, this small increment of basal metabolic MO₂ cannot explain the unchanged PVA-independent MO₂ despite the marked decrease in contractility during HT. Thus HT should have substantially increased the MO₂ component for E-C coupling for a given LV contractility.

Oxygen Cost of Contractility

In the stunned myocardium (22) and during acidosis (8-9), the oxygen cost of contractility has been shown to be higher by 120 and 50%, respectively, than in the normal heart. Therefore, HT is similar to the acidotic hearts. This similarity may predict that intracellular acidosis is one of the important mechanisms of the negative inotropism in HT. Hypothermia has been reported to result in intracellular alkalosis in cardiac muscle (25). In contrast, Kusuoka et al. (11) demonstrated that hypothermia (31-38°C) did not alter intracellular pH and energy-related phosphorus compounds. The shortened T_max and time constant in the present study are not observed in the acidotic heart. Therefore, intracellular acidosis may not be the substantial mechanism of the negative inotropism in HT.

Thus the mechanism of the negative inotropism in HT remains unclear. Our observation demonstrates that, in a given LV contractility, the HT condition needs calcium more than the NT condition. This suggests that the mechanism of the negative inotropism is either a decreased responsiveness of the contractile protein to Ca²⁺, a decreased intracellular Ca²⁺ level, or both.

The first possible mechanism of the negative inotropism in HT is a decreased responsiveness of the contractile protein to Ca²⁺. Kusuoka et al. (11) demonstrated that the maximal Ca²⁺-activated pressure was inversely related with decreased temperature (30-37°C), indicating that the responsiveness of the contractile protein to Ca²⁺ decreased at a high temperature range. In contrast, Brandt et al. (2) observed that an increased temperature (20-29°C) enhanced calcium sensitivity in ventricular muscle. However, the temperature ranges in this study are low and not physiological. No study in muscle and skinned preparation has been performed around 38-40°C. Recently, Mikane et al. (18) reported that HT decreased the Ca²⁺ responsiveness of cross-bridge force development and shifted the force-pCa²⁺ curve to the right in a simulation model. Therefore, a decreased responsiveness of the contractile protein to Ca²⁺ should play an important role in the negative inotropism in HT.

The second possible mechanism of the negative inotropism in HT is a decreased intracellular Ca²⁺ level. In hypothermic hearts, decreases in the active transport of Na⁺, Ca²⁺ efflux (33), and release and sequestration of Ca²⁺ by the sarcoplasmic reticulum (12) have been reported. These findings indicate that HT reduces the intracellular Ca²⁺ level and handling. In the present study, however, little change in the MO₂ components for E-C coupling was observed during HT, indicating that HT does not influence the intracellular Ca²⁺ handling. Thus this mechanism is considered to be a small part of the negative inotropism in HT.

The third possible mechanism is that a reduced amount of released Ca²⁺ during systole is due to the dysfunction of the Ca²⁺ release channel of the sarcoplasmic reticulum. In our previous observations (34), low-dose ryanodine (30-40 nM) decreased E_max but disproportionately induced high MO₂ for E-C coupling by increasing the open probability of the Ca²⁺ release channel. Inappropriate leak of Ca²⁺ from the sarcoplasmic reticulum should result in similar mechanoenergetic findings to the present study. However, the time to peak contraction and time constant were prolonged by ryanodine, whereas they were shortened by HT. Therefore, this mechanism induced by ryanodine is different from that of HT.

Limitations

In summary, we have assessed the direct effect of HT by 5°C from 36°C on LV mechanoenergetics, fully utilizing the MO₂-PVA-E_max framework in the isolated cross-circulated canine heart. HT depresses E_max, shortens T_max, and decreases both PVA and MO₂. The MO₂-PVA relation in HT is superimposable to that in NT, and the apparent oxygen cost of contractility obtained during the course of an increase in heart temperature is virtually zero, like that in hypothermia (28) but unlike the conventional negative inotropism. However, the true oxygen cost of contractility obtained with calcium is 1.5 times greater in HT than in NT. These results suggest that the depressed LV contractility during HT is largely due to a decreased Ca²⁺ responsiveness of the contractile protein.