INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HYPOTHERMIA is a key to successful cardiac preservation because it greatly decreases the metabolic rate and delays degradation of intracellular enzymes if the heart is not perfused. Most enzymatic activity decreases 50% for each 10°C fall in temperature (1), so even at 17°C myocardial O₂ consumption (MO₂) is maintained at ~25% of that at 37°C. Mild and moderate hypothermia as low as 17°C is widely used to protect hearts during open heart surgery. In normal ionic solutions, contractile force increases during mild and moderate hypothermia (10, 38). The force of contraction of cardiac myofilaments is highly dependent on the free intracellular Ca²⁺ concentration ([Ca²⁺]_i). Hypothermia may alter sarcolemmal Ca²⁺ flux via Ca²⁺ channels and other ion channels and ion exchangers to cause a change in Ca²⁺-induced Ca²⁺ release by the sarcoplasmic reticulum (SR) (23, 26, 41). Ca²⁺ homeostasis may be disrupted by slowed release or attenuated reuptake of Ca²⁺ by the SR during hypothermia. A rise in diastolic [Ca²⁺]_i and attenuated myofilament ATP hydrolysis during hypothermia may dissociate contractility from [Ca²⁺]_i and diastolic contracture at normal heart rates.

There are no published reports regarding the effects of mild (27°C) and moderate (17°C) hypothermia on phasic Ca²⁺ transients and myocardial sensitivity to [Ca²⁺]_i in the intact perfused heart. We hypothesized that the paced heart can adapt mechanically to mild hypothermia by prolonging the period of contraction for a given amount of Ca²⁺ availability but cannot adapt with more intense hypothermia because Ca²⁺ loading and impaired relaxation impedes contractility. The model selected was the paced, constantly perfused guinea pig heart loaded with the Ca²⁺ fluorescence indicator dye indo 1. Ca²⁺ transients were recorded from the left ventricular (LV) free wall, and contractility was measured isovolumetrically with an LV balloon and pressure transducer. Measured variables were heart rate (HR), coronary flow (CF), percent oxygen extraction (%O₂E), MO₂, and cardiac work efficiency.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Langendorff Isolated Heart Preparation and Measurements

1

Left ventricular pressure (LVP) was measured isovolumetrically with a transducer connected to a thin, saline-filled latex balloon, inserted into the LV through the mitral valve from a cut in the left atrium. Balloon volume was adjusted to maintain a diastolic LVP of 0 mmHg during the initial control period so that any increase in diastolic LVP would reflect an increase in LV wall stiffness or diastolic contracture. A pair of bipolar electrodes was placed in the right atrial appendage to monitor spontaneous HR and in the right ventricular free wall to pace hearts at 220 beats/min at two times the minimal stimulus threshold current pulses of 10 ms in duration. Coronary sinus effluent was collected by placing a cannula into the right ventricle through the pulmonary artery after both vena cava were ligated. Coronary flow (aortic inflow) was measured at constant temperature and constant perfusion pressure (55 mmHg) by a self-calibrating, in-line, ultrasonic flowmeter (Transonic T106X, Ithaca, NY) placed directly into the aortic inflow line. Adenosine (0.2 ml of 200 µM stock solution) was injected into the aortic root cannula to assess the adequacy of coronary flow reserve before the study was continued. Adenosine increased average flow by 95 ± 3%.

Coronary arterial O₂ tension was measured off-line with an intermittently self-calibrating analyzer system (Radiometer ABL-2, Medtron Chicago, Des Plaines, IL) together with measurements of pH and PCO₂. Coronary sinus effluent was collected by placing a small catheter into the right ventricle through the pulmonary artery after both venae cavae were ligated. Coronary sinus venous O₂ tension was measured continuously on-line with an O₂ Clark-type electrode (model 203B, Instech, Plymouth Meeting, PA). %O₂E was calculated as 100 × (Pa_O2  Pv_O2)/Pa_O2 (where Pa_O2 and Pv_O2 are arterial and venous PO₂, respectively), MO₂ as (CF/g) × (Pa_O2  Pv_O2) × (24 µl of O₂/ml) at 760 mmHg, and cardiac work efficiency as (developed LVP × HR)/MO₂.

Measurement of Cytosolic Free Ca²⁺ in Intact Hearts

Calculation of dissociation constant at different temperatures. Eight hearts (1.5 ± 0.1 g) were perfused with standard perfusate (see below) for 20 min to wash out blood. Hearts were then removed from the perfusion apparatus, immersed in 15 ml of a mixture of 5 mM HEPES buffer (pH adjusted to 7.0) with 20 mM NaCl and 115 mM KCl, and homogenized using a polytron blender. The soluble protein fraction was collected after centrifugation at 5,000 g for 20 min and ultracentrifugation at 100,000 g for 1 h at 25°C. With all liquid retained, the final concentration of soluble heart protein was 0.4 g/ml. Fluorescence was measured with a modified luminescence spectrophotometer (SLM Aminco-Bowman II, Spectronic Instruments, Urbana, IL). Peak intensities of paired test homogenates, 75 µM free indo 1 and an indo 1 free blank, were measured in a quartz cuvette at 37°C at 456 nm for minimal [Ca²⁺] (R_min; 20 mM EGTA to chelate Ca²⁺) and at 385 nm for a range of maximal [Ca²⁺] from 0.7 to 160 µM (R_max). Emission scans were conducted over a range of excitation wavelengths from 370 to 550 nm in 1-nm increments. The excitation wavelength was changed every 5 s. The monochrometer of the emission photomultiplier tube alternated between 385 and 456 nm every 2.5 s. Total emission scan duration was 15 min. Homogenate [Ca²⁺] was measured with a Ca²⁺-selective electrode (Orion Research, Cambridge, MA). Emission scans done at 0, intermediate, and maximal [Ca²⁺] revealed that the dissociation constant (K_d) of indo 1 was 149 ± 8 nM at 37°C. R_max was calculated as 5.986 and R_min as 0.059.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effect of temperature on free indo 1 fluorescence intensity in 1 homogenate heart study. For a given scanning emission wavelength, fluorescence intensity decreased progressively at lower temperatures for maximal (1 mM) and minimal (EDTA) calcium in homogenate. For 4 hearts, dissociation constants (Kd) were calculated as 149, 205, and 254 nM at 37°C, 27°C, and 17°C, respectively.

Loading fluorescent probe indo 1 and recording Ca²⁺ transients. Experiments were carried out in an ambient light-free apparatus. The heart was partially immobilized by being hung from the aortic cannula, the pulmonary artery catheter, and the LV balloon catheter. The heart was immersed in a bath. The distal end of the trifurcated-fiber silica fiber-optic cable (optic surface area 38.5 cm²) was placed against the LV epicardial surface through a hole in the bath, and a rubber O-ring was placed between the ferrule and the heart to reduce cardiac motion at the contact point of the fiber-optic tip. For each heart, background autofluorescence was determined after initial perfusion and equilibration at 37°C, 27°C, and 17°C. Thereafter, hearts were loaded with indo 1-AM at room temperature (25 ± 0.6°C) for 20-40 min with 165 ml of a recirculated, modified KR solution containing 6 µM indo 1-AM (Sigma Chemical, St. Louis, MO). Indo 1-AM was initially dissolved in 1 ml of dimethyl sulfoxide containing 16% (wt/vol) Pluronic I-127 and diluted to 165 ml with modified KR solution. Loading was stopped when F₃₈₅ and F₄₅₆ intensities increased 10-fold. Residual indo 1-AM was washed out by perfusing hearts with standard perfusate for at least another 20 min and then rewarming them to 37.5 ± 0.2°C before initiating the study.

Calculation of compartmental Ca²⁺ concentration from Ca²⁺ transients. The Ca²⁺ transient obtained from F₃₈₅/F₄₅₆ is nonlinearly proportional to [Ca²⁺]_i (16). Calibration curves were derived according to previously published protocols by Brandes et al. (6, 7) using modifications of a standard equation for fluorescence indicators (16). Total [Ca²⁺]_i ([Ca²⁺]_i,tot) was calculated from the total fluorescence ratio (F_385,tot/F_456,tot = R_tot), R_max (S_r/bH for >100 µM Ca²⁺), R_min [R_max × (S₃₈₅/S₄₅₆) for 0 Ca²⁺], S₄₅₆ (R_max/R_min at 456-nm emission), and K_d according to

(1)

where S_r = (1  S₄₅₆)/(1  S₃₈₅) and bH is the slope (b) of F_385,tot as a function of F_456,tot.

(2)

(3)

(4)

(5)

Experimental Protocols

Data Presentation and Interpretation

1

Excitation-contraction coupling was measured by plotting the relationship of 1) developed LVP to peak systolic Ca²⁺ and total released (area) Ca²⁺, 2) the LVP time integral to peak systolic [Ca²⁺]_i and released (area) Ca²⁺, and 3) maximal +dLVP/dt (velocity of isovolumetric global contraction) to peak systolic [Ca²⁺]_i and released Ca²⁺. A leftward shift in the sigmoid plots was considered as a sensitizing effect and a rightward shift as a desensitizing effect (5, 8, 14, 17, 29, 44). Shifts with no changes in slope have been shown to result from a change in Ca²⁺ binding to troponin C (central effect). Changes in the slope of the relationship have been shown to be related to changes in myofilament cross-bridge cycling (downstream effect). Changes in peak developed (systolic-diastolic) LVP and the LVP time integral at maximal systolic-diastolic [Ca²⁺]_i were considered to reflect changes in maximal Ca²⁺-activated force, i.e., a change in the total number of myofilament cross-bridge formations.

The relationships of diastolic LVP and maximal dLVP/dt (velocity of isovolumetric global relaxation) to diastolic [Ca²⁺]_i were also plotted. A leftward shift was considered to be related to improved performance by the SR Ca²⁺ pump and a rightward shift to be a sign of impaired performance by the Ca²⁺ pump. Noncytosolic [Ca²⁺] was not plotted against systolic or diastolic LV performance, but alterations were interpreted in the setting of the plotted excitation-contraction curves. An increase in noncytosolic [Ca²⁺] was considered to be a sign of impaired myocardial Ca²⁺ handling and a decrease to be a sign of improved Ca²⁺ handling. A change in noncytosolic [Ca²⁺] may alter the efficiency of Ca²⁺ pumping from the cytosol during diastole, so changes in noncytosolic [Ca²⁺] were considered to reflect changes in function of the SR (and sarcolemmal) Ca²⁺ pump (inverse noncytosolic Ca²⁺-to-LVP relationship).

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of Indo 1 Loading and Time Controls

Indexes of Isovolumetric Contraction and Relaxation and Metabolism as a Function of [CaCl₂]_e at Three Temperatures

Arterial and coronary sinus pH, PO₂, and PCO₂ at 37°C were, respectively, 7.43 ± 0.01 and 7.29 ± 0.01, 495 ± 9 and 150 ± 6 mmHg, and 21 ± 2 and 28 ± 2 mmHg; at 27°C, these values (analyzed at 37°C) were 7.46 ± 0.01 and 7.38 ± 0.01, 470 ± 12 and 240 ± 8 mmHg, and 21 ± 2 and 24 ± 3 mmHg; at 17°C, these values (analyzed at 37°C) were 7.49 ± 0.02 and 7.45 ± 0.02, 465 ± 9 and 288 ± 11 mmHg, and 21 ± 2 and 22 ± 1 mmHg. Coronary sinus pH and PO₂ were higher at 27°C and 17°C, and PCO₂ was lower at 17°C compared with values at 37°C (P < 0.05).

Ca²⁺ and LVP Transients as a Function of [CaCl₂]_e at Three Temperatures

View larger version (41K): [in this window] [in a new window]   Fig. 2.   Effect of temperature on phasic left ventricular pressure (LVP) and cytosolic Ca2+ transients in 1 guinea pig heart. Ca2+ transients were mathematically calibrated and converted to nanomolar intracellular Ca2+ concentrations ([Ca2+]i). Note elevation and flattening of LVP and Ca2+ tracings at a constant pacing rate. See text and Fig. 3 for details.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Single-beat tracings of cytosolic [Ca2+]i and LVP from 1 heart recorded at 3 temperatures during perfusion with solution containing 4.5 mM CaCl2. Note that both systolic and diastolic Ca2+ rise with increasing hypothermia but that diastolic LVP increases and systolic LVP decreases. Rates of contraction and relaxation are decreased with increasing hypothermia for both variables. However, the amount of released Ca2+ (area of systolic-diastolic Ca2+ transient) increases whereas the LVP work per beat (area of LVP transient) decreases. See Figs. 4-10 for quantitative data and relationships between characteristics of LVP and [Ca2+]i.

Indexes of [Ca²⁺]_i as a Function of [CaCl₂]_e at Three Temperatures

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Plots of systolic (sys), diastolic (dia), and noncytosolic (primarily mitochondrial) [Ca2+]i as a function of increasing extracellular CaCl2 concentration ([CaCl2]e) in 5 isolated hearts exposed in random fashion to hypothermia for 20 min. [CaCl2]e was increased incrementally once per minute at each temperature. Data represent means ± SE. Note that the various indexes of [Ca2+]i attained a maximum at 4.5 [CaCl2]e and that hypothermia to 17°C increased these indexes the most. See text for details.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Plots of systolic-diastolic (sys-dia) [Ca2+]i and area Ca2+ as a function of increasing [CaCl2]e in 5 isolated hearts exposed in random fashion to hypothermia for 20 min. [CaCl2]e was increased incrementally once per minute at each temperature. Data represent means ± SE. Note that the various indexes of [Ca2+]i attained a maximum at 4.5 [CaCl2]e and that hypothermia to 17°C increased these indexes the most. See text for details.

Indexes of Contraction and Relaxation as a Function of Systolic, Diastolic, and Noncytosolic [Ca²⁺]_i: Effect of Temperature on ED₅₀

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Plots of various indexes of contractility as a function of sys[Ca2+]i and sys-dia[Ca2+]i during increases in [CaCl2]e at steady-state periods of hypothermia. Data represent means ± SE. Note the reduced maximal contractile effort and rightward shift in peak rates of LVP development (+dLVP/dt) and sys-diaLVP curves. See text for details.

Figure 7 plots diastolic LVP and peak dLVP/dt as a function of diastolic [Ca²⁺]_i and plots peak dLVP/dt as a function of noncytosolic [Ca²⁺]_i. Diastolic LVP_max increased from near 0 mmHg at 37°C to 52 ± 11 mmHg at 27°C and to 70 ± 12 mmHg at 17°C (P < 0.01), whereas peak dLVP/dt_max (rate of relaxation) decreased significantly by 61 ± 8% at 27°C (P <0.01) and by 86 ± 9% at 17°C (P < 0.01) compared with values at 37°C. At 50% peak dLVP/dt_max, diastolic [Ca²⁺]_i ED₅₀ was significantly higher (P <0.05) at both 27°C (302 ± 60 nM, mean ± CI_95%) and 17°C (552 ± 44 nM) than at 37°C (196 ± 70 nM); the results obtained at 27°C and 17°C were also different from each other (P < 0.05). Similar, but higher, noncytosolic [Ca²⁺]_i ED₅₀ values were obtained at 50% peak dLVP/dt_max. The plots in Fig. 7 demonstrate that, compared with responses at 37°C, maximal hypothermia-induced diastolic contracture (diastolic LVP_max) was equivalently increased at 27°C and 17°C and the maximal rate of relaxation (peak dLVP/dt_max) as a function of either diastolic or noncytosolic [Ca²⁺]_i was markedly reduced at 27°C and more so at 17°C.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Plots of various indexes of relaxation and contracture as a function of dia[Ca2+]i and noncytosolic (mitochondrial) [Ca2+]i during increases in [CaCl2]e at steady-state periods of hypothermia. Data represent means ± SE. Note the reduced maximal relaxation effort and rightward shift in diastolic LVP and peak rate of relaxation (dLVP/dt) curves. See text for details.

Indexes of Contractility as a Function of Area [Ca²⁺]_i: Effect of Temperature on Curve Slopes

1

1

1

1

1

1

1

1

1

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Plots of various indexes of contraction as a function of total released Ca2+ (area [Ca2+]i) during increases in [CaCl2]e at steady-state periods of hypothermia. Data represent means ± SE. Note the reduced maximal contractile effort and rightward shift in peak +dLVP/dt and LVParea curves, but not in sys-diaLVP curve, when expressed as a function of area under LVP curve (LVParea). See text for details.

Figure 9 plots 100% normalized systolic-diastolic LVP as a function of systolic-diastolic [Ca²⁺]_i as well as area [Ca²⁺]_i. For 100% systolic-diastolic LVP_max versus systolic-diastolic [Ca²⁺]_i, the slopes at ED₅₀ were 11.5% (37°C), 24.2% (27°C), and 6.4% (27°C) per 100 nM [Ca²⁺]_i, each of which were different from each other (P < 0.05). When plotted against area [Ca²⁺]_i, the slopes at ED₅₀ were 10.6% (37°C), 15.3% (27°C), and 4.6% (27°C) per 1,000 nM · s · beat¹ Ca²⁺, where only the value at 17°C was different from the others (P < 0.05).

View larger version (30K): [in this window] [in a new window]   Fig. 9.   Plots of sys-diaLVP normalized to 100% of maximal sys-diaLVP as a function of sys-dia and area [Ca2+]i during increasing [CaCl2]e at steady-state periods of hypothermia. Note apparent leftward shift and increased slope of 27°C curve and rightward shift and decreased slope of 17°C curve when expressed as a function of sys-dia[Ca2+]i. When expressed as a function of area [Ca2+]i, there was no significant shift of 27°C curve.

Figure 10 plots 100% normalized LVP_area and peak +dLVP/dt at each temperature as a function of area [Ca²⁺]_i. For 100% LVP_area, the ED₅₀ was significantly greater at 17°C (12,860 ± 760 nM · s · beat¹, mean ± CI_95%) but not different at 27°C (6,815 ± 515 nM · s · beat¹) and 37°C (6,810 ± 450 nM · s · beat¹), and there were no differences among the slopes at their ED₅₀ values. However, for 100% peak +dLVP/dt_max, the ED₅₀ values after normalization were not significantly different at 27°C (8,080 ± 680 nM · s · beat¹) or at 17°C (8,310 ± 540 nM · s · beat¹) from the value at 37°C (8,920 ± 480 nM · s · beat¹), and there were no differences among the slopes at their ED₅₀ values.

View larger version (26K): [in this window] [in a new window]   Fig. 10.   Plots of LVParea and peak +dLVP/dt normalized to 100% of maximal LVParea as functions of area [Ca2+]i during increasing [CaCl2]e at steady-state periods of hypothermia. Data represent means ± SE. Note apparent rightward shift of 17°C curve when contractility is expressed as LVParea. When expressed as peak +dLVP/dt, there was no significant shift in 17°C curve. See text for details.

The plots of Figs. 8-10 demonstrate that contractile indexes plotted as a function of total Ca²⁺ release per beat (area [Ca²⁺]_i) furnish different information than when plotted as a function of systolic-diastolic [Ca²⁺]_i. For a given amount of released (area) Ca²⁺, contractility was progressively reduced at 27°C and 17°C, but the ED₅₀ was not affected at 27°C and was reduced at 17°C compared with that at 37°C. Hypothermia to 17°C produced rightward shifts in the curves of 100% normalized systolic-diastolic LVP and LVP_area indexes as a function of released Ca²⁺ but did not shift the curve of 100% normalized +dLVP/dt as a function of released Ca²⁺. Slopes were unaffected by temperature.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This is the first report on the effects of hypothermia on myoplasmic [Ca²⁺]_i and contractile function in the intact heart. This study documents the complex effects of mild and moderate hypothermia to alter the dependence of beat-to-beat contractility on phasic myoplasmic [Ca²⁺] at physiological heart rates and temperature. The major findings of this study are as follows. At 27°C, the paced heart adapts functionally to the greater accumulation of Ca²⁺ during diastole by reducing the rate of release of Ca²⁺, which effectively maintains the net systolic release of Ca²⁺ per beat so that the prolonged systolic time interval maintains the contractile work over each beat. At 17°C, however, despite the very prolonged release and reuptake of Ca²⁺ over the cardiac cycle (increased effective systolic area Ca²⁺), diastolic Ca²⁺ loading becomes so marked that relaxation is impaired, and contractility decreases. Interestingly, at both 27°C and 17°C, the normalized maximal rate of contractility (+dLVP/dt) for a given amount of released (area) Ca²⁺ is not altered.

More specifically, we observed the following. 1) Hypothermia to 27°C increases maximal diastolic (and noncytosolic) [Ca²⁺]_i by up to 1.5- to 2-fold but does not change maximal systolic [Ca²⁺]_i or released (area) [Ca²⁺]_i; however, hypothermia to 17°C markedly increases each index of [Ca²⁺]_i by up to 2- to 3-fold (Figs. 3 and 4). 2) Hypothermia to 27°C and 17°C similarly increases diastolic LVP, does not increase systolic LVP, and decreases phasic (systolic-diastolic) LVP by up to 50%, but it causes large temperature-dependent decreases in the rates of contraction and relaxation (Figs. 6-8). 3) An apparent increase in Ca²⁺ sensitivity (left curve shift and increased slope), observed at 27°C when normalized phasic LVP values are plotted against phasic [Ca²⁺]_i (Fig. 9), is not observed when normalized LVP_area values [isovolumetric pressure work (mmHg · s)] are plotted against [Ca²⁺]_i area values; only a rightward shift is observed at 17°C for this relationship (Fig. 10). 4) LVP_area is reduced at 17°C but not at 27°C (Fig. 10, top); however, the rate of pressure development per beat, normalized to a 100% maximum, is not altered at 27°C or 17°C for a given release of Ca²⁺ per beat (Fig. 10, bottom).

Is Myofilament Sensitivity to Ca²⁺ Altered by Mild and Moderate Hypothermia?

By using interpretations applied from intact and skinned muscle preparations (5, 8, 14, 17, 29, 46, 47), our study implies that cooling from 37°C to 27°C and 17°C at a fixed heart rate effectively reduces the number or frequency of actin-myosin interactions for a given net release of [Ca²⁺]_i on the basis of the slowed development of contraction and reduced peak contraction with each beat at maximal [CaCl₂]_e (upstream effect). This is likely caused by an increase in diastolic Ca²⁺ that could result from increased sarcolemmal Ca²⁺-induced SR Ca²⁺ release or reduced SR Ca²⁺ ATP-dependent uptake, which impairs relaxation during the shortened diastolic phase. Hypothermia may also prolong the interaction of Ca²⁺ with troponin C at 27°C (but not at 17°C) on the basis of enhanced peak (systolic-diastolic LVP) contraction or may maintain the contractile curve area normalized for a given peak (systolic-diastolic) [Ca²⁺]_i or area [Ca²⁺]_i. This effect implies impeded inhibitory function of the tropomyosin molecule to block actin-myosin interaction (central effect). Thus a reduced number of myofilament interactions that occur at 27°C appear to be balanced by a prolongation of these interactions. At 17°C, both the number and duration of new interactions may become markedly reduced; an exception would be that the rate of contraction normalized to the area of the Ca²⁺ transient does not appear to be altered at 17°C. This may be evidence to suggest that at 17°C the strength or rate (cooperativity) of individual actin-myosin cross-bridge interactions is impaired (downstream effect) in terms of phasic LVP but not in terms of LVP_area or peak +dLVP/dt.

The present study supports the idea that mild hypothermia may prevent desensitization of myofilaments to Ca²⁺ by prolonging the permissive effect of Ca²⁺ on troponin C. Because a change in Ca²⁺ sensitivity of the contractile apparatus depends on the temporal relation between the time for the cross bridge to complete a cycle and the time Ca²⁺ remains bound to troponin C, the relative maintenance of contractility observed during hypothermia could be explained by the decreased (slowed) intracellular kinetics of Ca²⁺. Prolonged availability of Ca²⁺ for binding to the subunits of troponin C prolongs the contractile state. Furthermore, the dissociation rate constant of Ca²⁺ from troponin C could be slowed, which would also lead to a prolonged contraction, but the present study cannot answer this. If diastolic relaxation predicts overall efficiency of contraction, then the rate of contraction is relatively less efficient during hypothermia. By prolonging the relaxation phase, moderate hypothermia may also reduce diastolic elastance and oppose the apparent usefulness of enhanced inotropic responsiveness during mild hypothermia (15). It is important to note that if hearts were not paced and allowed to beat at a lower rate during hypothermia, the added time for contraction would likely be accompanied by greater relaxation and lower diastolic Ca²⁺. Slowing would offset the reduced kinetics of actinomyosin interaction and SR Ca²⁺ flux.

Possible Factors to Increase Diastolic and Noncytosolic Ca²⁺ During Hypothermia

Changes in cardiac cell action potential characteristics give indirect insight into voltage-dependent ion mechanisms at play during graded hypothermia (19, 20, 25). At 27°C and 17°C, myocytes appear to maintain a relatively normal Na⁺ and K⁺ homeostasis because resting membrane potential (E_m) and phase 0 voltage changes are maintained (25). However, because action potential duration (APD) is markedly increased, net inward Ca²⁺ current flux is increased. The ATP-sensitive K⁺ (K_ATP) channel opener bimakalim (25) and Ca²⁺ channel blockers (3, 4) markedly attenuate or reverse mild hypothermia-induced increases in APD. Cooling from 37°C to ~20°C increases APD and accentuates APD effects at shorter cycle lengths (3, 4, 19, 20, 25, 42).

There are several possibilities for increased Ca²⁺ loading during hypothermia. Increasing hypothermia could slow sarcolemmal Ca²⁺-induced SR Ca²⁺ release or uptake of Ca²⁺ by Ca²⁺ ATP-dependent pumps (41). Moderate hypothermia reduces, and severe hypothermic storage abolishes, membrane-bound Ca²⁺-pump activity necessary for Ca²⁺ reuptake into SR or extrusion through the sarcolemma (22, 26). Reduced Ca²⁺ uptake caused by impeded Ca²⁺-pump activity would disrupt Ca²⁺ homeostasis and thereby interrupt excitation-contraction coupling and relaxation. However, mild hypothermic perfusion may not greatly alter Ca²⁺-pump function at 27°C because there were equivalent increases in both systolic and diastolic [Ca²⁺]_i. SR Ca²⁺ handling by the protein kinase phospholamban system could also be altered by a reduced cytosolic energy level during hypothermia (28).

All voltage-dependent ion exchangers and ion channels are dependent on maintenance of the Na⁺-K⁺-ATP pump. Mild to moderate hypothermia could increase [Ca²⁺]_i by exchange for excess Na⁺ that accumulates during the longer repolarization phase or via slowed Na⁺-pump activity (9). Increasing hypothermia likely retards Na⁺-pump activity so that efflux of K⁺ and influx of Na⁺ down their concentration gradients may not be matched by Na⁺ extrusion and K⁺ uptake mediated by the Na⁺ pump. We could not control for the possible effect of a change in intracellular pH on Na⁺/H⁺ exchange during hypothermia. Coronary sinus pH increased with hypothermia at 2.5 mM CaCl₂ and likely reflects a mild intracellular alkalosis caused by reduced cardiac metabolism; the increased cell pH effect, per se, would be expected to slightly decrease contractility. Although mild hypothermia may cause a small decrease in [H⁺]_i, severe hypothermia may cause a much larger decrease in [H⁺]_i because the metabolic rate and Na⁺-pump activity are markedly reduced and fewer hydrogen ions are produced; this promotes Ca²⁺ loading. When isolated embryonic myocytes were exposed to 10°C hypothermia for up to 6 h, total cell Na⁺ (2-3 times) and total cell Ca²⁺ (1.5 times) increased (21), but extracellular acidosis, or a Na⁺/H⁺ exchange inhibitor, prevented the rise in total cell Na⁺. This strongly suggests that severe hypothermia increases Na⁺ influx mostly via Na⁺/H⁺ exchange. Hypothermia may increase [Na⁺]_i caused by a longer Na⁺ channel opening or reduced Na⁺-pump activity. However, because hypothermia to only 27°C markedly prolongs action potentials (AP) but does not alter resting E_m (25), Na⁺-pump activity may be slowed, but is not likely abolished, during mild hypothermia.

Mild hypothermia could enhance [Ca²⁺]_i by increasing [Na⁺]_i via voltage-gated Na⁺ influx when the cardiac action potential is prolonged but resting E_m or AP amplitude are unchanged (3, 4, 19, 20, 25, 42). Slowed Na⁺ channel inactivation allows more channels to remain open longer during the plateau phase so that APD increases. Enhanced Na⁺ entry may help to maintain a positive inotropic effect during moderate hypothermia. Evidence that hypothermia increases total Na⁺ influx during the depolarized plateau phase of the AP is suggested by a decreased peak voltage-dependent conductance but a shift in the inactivation curve toward more depolarized potentials (2, 31).

Hypothermia could also slow activation and attenuate inactivation of K⁺ channels. This could contribute, with greater Na⁺ and Ca²⁺ influx, to a marked delay in AP repolarization (25). A decrease from 34°C to 24°C decreases both the delayed rectifier and the inward rectifying K⁺ currents (I_K) (20). As the outward repolarizing current becomes reduced because of the longer depolarized state, APD would further increase as more Na⁺ and Ca²⁺ accumulate (20). Giving a K_ATP channel opener during mild and moderate hypothermia may be an effective way to attenuate Na⁺ and Ca²⁺ loading (25).

The present study supports other data showing that hypothermia prolongs AP depolarization as more Na⁺ and Ca²⁺ enters myocytes during the plateau phase in part because Na⁺ and Ca²⁺ channels slowly inactivate. The increase in [Na⁺]_i and reduced outward I_K conductance shifts the reversal potential for Na⁺/Ca²⁺ exchange toward less negative membrane potentials and so promotes greater Ca²⁺ influx by the exchanger to increase [Ca²⁺]_i, i.e., prolonged "reverse" mode operation. It has been reported recently that the relative contributions of the SR Ca²⁺ pump and the Na⁺/Ca²⁺ exchanger to contractility, while depressed, are equivalent during cooling (34). However, the contribution of Ca²⁺ release to Ca²⁺-induced Ca²⁺ release triggered by Na⁺/Ca²⁺ exchange may be reduced more than that of the inward Ca²⁺ current with cooling (45).

In summary, it seems plausible that with the longer phase 2 action potential plateau during mild and moderate hypothermia, slowed Na⁺, Ca²⁺, and K⁺ channel kinetics, reversed Na⁺/Ca²⁺ exchange, and Na⁺/H⁺ exchange each contribute in part to [Ca²⁺]_i loading and altered contractile responses to a given amount of Ca²⁺ released. Noncytosolic Ca²⁺ loading, like cytosolic Ca²⁺ loading, occurs as cytosolic [Ca²⁺]_i increases with increasing levels of hypothermia. [Ca²⁺]_i likely plays a regulatory role in the link between cardiac mechanics and energy production. Mitochondrial Ca²⁺ overload could damage mitochondrial membranes to impair ATP synthesis and contribute to contractile dysfunction (32).

This study contributes to our knowledge of the effects of clinically used levels of hypothermia on the intact, paced heart. The results indicate that the naturally homeothermic, mammalian heart, if paced, can adapt quite well to a 10°C fall in temperature to maintain contractility by prolonging the period of contraction for a given amount of net available [Ca²⁺] despite a large increase in diastolic [Ca²⁺]. A 20°C fall in temperature, however, results in reduced contractility because of excessive [Ca²⁺] loading and a markedly reduced capacity for relaxation.