Journal of Applied Physiology

Limiting sarcolemmal Na+ entry during resuscitation from ventricular fibrillation prevents excess mitochondrial Ca2+ accumulation and attenuates myocardial injury

Sufen Wang, Jeejabai Radhakrishnan, Iyad M. Ayoub, Julieta D. Kolarova, Domenico M. Taglieri, Raúl J. Gazmuri


Background: intracellular Na+ accumulation during ischemia and reperfusion leads to cytosolic Ca2+ overload through reverse-mode operation of the sarcolemmal Na+-Ca2+ exchanger. Cytosolic Ca2+ accumulation promotes mitochondrial Ca2+ (Ca2+m) overload, leading to mitochondrial injury. We investigated whether limiting sarcolemmal Na+ entry during resuscitation from ventricular fibrillation (VF) attenuates Ca2+m overload and lessens myocardial dysfunction in a rat model of VF and closed-chest resuscitation. Methods: hearts were harvested from 10 groups of 6 rats each representing baseline, 15 min of untreated VF, 15 min of VF with chest compression given for the last 5 min (VF/CC), and 60 min postresuscitation (PR). VF/CC and PR included four groups each randomized to receive before starting chest compression the new NHE-1 inhibitor AVE4454B (1.0 mg/kg), the Na+ channel blocker lidocaine (5.0 mg/kg), their combination, or vehicle control. The left ventricle was processed for intracellular Na+ and Ca2+m measurements. Results: limiting sarcolemmal Na+ entry attenuated cytosolic Na+ increase during VF/CC and the PR phase and prevented Ca2+m overload yielding levels that corresponded to 77% and 71% of control hearts at VF/CC and PR, without differences among specific Na+-limiting interventions. Limiting sarcolemmal Na+ entry attenuated reductions in left ventricular compliance during VF and prompted higher mean aortic pressure (110 ± 7 vs. 95 ± 11 mmHg, P < 0.001) and higher cardiac work index (159 ± 34 vs. 126 ± 29 g·m·min−1·kg−1, P < 0.05) with lesser increases in circulating cardiac troponin I at 60 min PR. Conclusions: Na+-limiting interventions prevented excess Ca2+m accumulation induced by ischemia and reperfusion and ameliorated myocardial injury and dysfunction.

  • calcium
  • cardiopulmonary resuscitation
  • myocardial ischemia
  • sodium

increased sarcolemmal Na+ influx with consequent cytosolic Na+ accumulation due to inability of the Na+-K+-ATPase to extrude Na+ represents an important pathophysiological mechanism responsible for cell injury during ischemia and reperfusion (3, 28). Main routes for sarcolemmal Na+ entry include the sodium-hydrogen exchanger isoform-1 (NHE-1), Na+ channels, and the Na+-HCO3 cotransporter. However, NHE-1 and Na+ channels appear to be the preferred routes for Na+ entry during ischemia and reperfusion (21, 38, 54). NHE-1 is activated by the intense intracellular acidosis that accompanies ischemia, initiating an electroneutral sarcolemmal Na+-H+ exchange. Na+ channels are activated following sarcolemmal depolarization. However, they inactivate slowly during ischemia and contribute to cytosolic Na+ overload (7, 54).

Cytosolic Na+ accumulation is believed to worsen ischemic injury mainly as a result of increased Ca2+ entry through the sarcolemmal Na+-Ca2+ exchanger isoform-1 (NCX-1) operating in reverse mode (1). Cytosolic Ca2+ overload, in turn, leads to mitochondrial Ca2+ overload, which can worsen cell injury by disrupting mitochondrial function (32, 61). Cytosolic Ca2+ overload can also favor reperfusion arrhythmias through delayed afterdepolarizations, causing ventricular arrhythmias (13, 60). Despite the prevailing belief that Ca2+ drives injury associated with cytosolic Na+ overload, Iwai et al. (30, 31) reported that cytosolic Na+ overload may directly alter mitochondrial function by depolarizing its inner membrane and reducing the rate of oxidative phosphorylation.

Most of the mechanistic knowledge gained on the beneficial effects of limiting cytosolic Na+ overload has resulted from work in isolated cardiac myocytes, isolated heart preparations, and intact animal models of global or regional ischemia. We have focused our work on understanding the effects of NHE-1 inhibition during resuscitation from cardiac arrest precipitated by VF. In this setting, the effects of ischemia (cardiac arrest) and reperfusion (resuscitation) are compounded by VF, which intensifies ischemic injury and prompts additional Na+ entry through activation of Na+ channels. We previously reported that during resuscitation from VF administration of NHE-1 inhibitors 1) ameliorates or prevents decreases in left ventricular compliance, 2) attenuates reperfusion arrhythmias eliminating recurrent episodes of VF, and 3) lessens postresuscitation myocardial dysfunction (4, 5, 16, 18, 37). We designed the current studies to investigate during resuscitation from VF 1) whether the effects of NHE-1 inhibition are in fact associated with lesser increases in cytosolic Na+, 2) whether similar or additive effects can be elicited by blockade of Na+ channels, 3) whether attenuation of cytosolic Na+ overload limits mitochondrial Ca2+ increases, and 4) whether these cellular effects result in less myocardial injury and dysfunction in an intact rat model of VF and closed-chest resuscitation.

Hearts were harvested at various time events representative of baseline, untreated VF, closed-chest resuscitation, and postresuscitation, while sarcolemmal Na+ entry was limited by administration of the new NHE-1 inhibitor AVE4454B, the Na+ channels blocker lidocaine, or both. Left ventricular tissue was processed to determine cytosolic Na+ content (using Co-EDTA as marker of the extracellular space) and total mitochondrial Ca2+ content. These findings were related to myocardial function and circulating levels of cardiac troponin I (cTnI). After observing that intracellular Na+ remained elevated postresuscitation, we measured activity of the sarcolemmal Na+-K+-ATPase in a separate series of experiments. The studies showed that limiting sarcolemmal Na+ entry during resuscitation from VF attenuates cytosolic Na+ increases, prevents excess mitochondrial Ca2+ accumulation, attenuates increases in cTnI, and ameliorates functional myocardial manifestations of ischemic injury.


The studies were approved by our Institutional Animal Care and Use Committee and conducted in accordance with institutional guidelines.

Animal Preparation

Adult male Sprague-Dawley rats (463–570 g) were anesthetized using pentobarbital sodium (45 mg/kg ip for induction and 10 mg/kg iv for maintenance every 30 min). A 5-Fr. cannula was orally advanced into the trachea and used for positive pressure ventilation during cardiac resuscitation and the postresuscitation interval. Proper placement was verified using an infrared CO2 analyzer (CO2SMO model 7100, Novametrix Medical Systems). A conventional lead II ECG was recorded through subcutaneous needles. PE-50 catheters were advanced through the right femoral vein into the right atrium and from the left femoral artery into the abdominal aorta for pressure measurement and blood sampling. A thermocouple microprobe (IT-18, Physitemp) was advanced through the right femoral artery into the thoracic aorta and used for measuring cardiac output and core temperature. A PE-50 catheter was advanced through the left external jugular vein into the right atrium and used exclusively for injection of thermal tracer. A 3-Fr. catheter (C-PUM-301J, Cook) was advanced through the right external jugular vein into the right atrium. A precurved guide wire was then fed through its lumen, advanced into the right ventricle, and used for electrical induction of VF. Core temperature was maintained between 36.5 and 37.5°C using an infrared heating lamp.

VF and Resuscitation Protocols

VF was induced by delivering a 60-Hz alternating current (0.1–0.6 mA) to the right ventricular endocardium for an uninterrupted interval of 3 min, after which the current was turned off and VF was allowed to continue until completion of a predetermined interval (described below). Chest compression was then started using an electronically controlled and pneumatically driven (50 PSI) chest compressor (CJ-80623, CJ Enterprises) set to deliver 200 compressions/min with a 50% duty cycle. The depth of compression was adjusted within the first minute to attain an aortic diastolic pressure between 26 and 28 mmHg and ensure a coronary perfusion pressure above the resuscitability threshold of 20 mmHg in rats (57). The depth of compression was adjusted during the remaining interval of chest compression to maintain the aortic diastolic pressure within the target range. Positive pressure ventilation with 100% oxygen was provided using a volume controlled ventilator (model 683, Harvard Apparatus) programmed to deliver 6 ml/kg body wt at 25 breaths/min unsynchronized to chest compression. Defibrillation was attempted after 5 min of chest compression by delivering a maximum of two 3-J transthoracic shocks using a biphasic waveform defibrillator (Smart Biphasic Heartstream XL M4735A, Agilent Technologies). If VF persisted or an organized rhythm with a mean aortic pressure of ≤25 mmHg ensued, chest compression was resumed for 30 s. The defibrillation-compression cycle was repeated up to three additional times, increasing the energy of individual shocks if VF persisted to 5-J and then to 7-J for the last two cycles. Successful resuscitation was defined as the return of an organized cardiac rhythm with a mean aortic pressure ≥60 mmHg for ≥5 min. After return of spontaneous circulation, the ventilation rate was increased to 60 breaths/min. Resuscitated rats were ventilated initially with 100% oxygen for 15 min and then continued with 50% oxygen for the remaining postresuscitation interval. Rats were monitored for a maximum of 60 min postresuscitation. Successfully resuscitated rats that died before 60 min postresuscitation were excluded.

Experimental Groups

Rats were randomized after completion of surgical preparation to one of 10 groups of six rats each (Fig. 1). In group 1, hearts were harvested at baseline (BL); in group 2, hearts were harvested after 15 min of untreated VF (VF); in groups 36, hearts were harvested after 15 min of VF with chest compression and ventilation provided during the last 5 min of VF (VF/CC); and in groups 710, hearts were harvested at 60 min postresuscitation (PR) following a VF and resuscitation protocol as in VF/CC groups. Rats in VF/CC groups that spontaneously defibrillated and restored spontaneous circulation before the designated harvest time were reassigned to the corresponding PR group. Rats in VF/CC and PR groups were randomized to one of four interventions. Three groups received Na+-limiting interventions, namely, lidocaine vehicle followed by AVE4454B, lidocaine followed by AVE4454B vehicle, and lidocaine followed by AVE4454B. One group served as control and received the vehicles of lidocaine and AVE4454B. Randomization proceeded by blocks with the investigators blind to the treatment assignment. Each block included one representative of each experimental group for a total of six blocks.

Fig. 1.

Experimental groups depicting the various intervention and the time at which hearts were harvested (AD). The interval between Co-EDTA injection and harvesting was set at 10, 20, or 30 min in baseline (BL) and postresuscitation (PR) groups and at 25, 35, or 45 min in ventricular fibrillation (VF) and VF with chest compression (VF/CC) groups to account for the 15-min interval of VF. Rats in VF/CC and PR groups were randomized to AVE4454B and lidocaine vehicle, AVE4454B vehicle and lidocaine, AVE4454B and lidocaine, and vehicles immediately before starting chest compression (Drug in figure).

Experimental Drugs and Vehicle Controls

AVE4454 hydrochloride (AVE4454B; kindly donated by Sanofi-Aventis) was used for NHE-1 inhibition. AVE4454B is a selective NHE-1 inhibitor newly developed with the intent of circumventing the adverse effects of cariporide reported in the EXPEDITION trial (41). Fluorometric image plate reader assay performed by the manufacturer using human NHE subtypes demonstrated high potency and selectivity for NHE-1 (IC50, 0.051 μM) compared with NHE-2 (IC50, 7.6 μM) and NHE-3 and -5 (no inhibition at 10 μM). In a rat model of coronary occlusion and reperfusion, AVE4454B dose dependently (0.1–3 mg/kg iv and 3 mg/kg and 10 mg/kg oral) reduced infarct size (unpublished). The AVE4454B dose and preparation followed the manufacturer's specifications. Accordingly AVE4454B was dissolved in 1.8% glycine buffer (pH 4.00) to a concentration of 1.0 mg/ml and administered in bolus dose of 1 mg/kg (1 ml/kg). For AVE4454B vehicle control, mannitol was dissolved in glycine buffer to a final concentration of 1.3 mg/ml and given in bolus dose of 1 ml/kg. Lidocaine hydrochloride was purchased from Sigma and used for Na+-channel blockade. Lidocaine was dissolved in 0.9% NaCl to a concentration of 5 mg/ml and administered in bolus dose of 5 mg/kg (1 ml/kg). The dose was chosen based on previous studies reporting myocardial protective effects in pig and rabbit models of ischemia and reperfusion injury using 2–10 mg/kg (24, 26, 56). In preliminary studies, lidocaine administered in bolus dose of 5 mg/kg to a 472 g and to a 490 g rat during spontaneous circulation elicited no electrocardiographic or hemodynamic effects. For lidocaine vehicle control, 0.9% NaCl was administered in bolus dose of 1 ml/kg.

Hemodynamic Measurements

Cardiac output was measured after right atrial bolus injection of 200 μl of 0.9% NaCl at room temperature. The dilution curves were analyzed using custom-developed LabVIEW-based software. Cardiac index (CI) was calculated dividing cardiac output by body weight and reported as milliliters per minute per kilogram. Cardiac work index (CWI) was calculated multiplying CI by the difference between mean aortic and mean right atrial pressures and reported as gram times meters per minute per kilogram (after converting to work units multiplying by 1.36 ×10−3 in kg·cm−2·mmHg−1). The coronary perfusion pressure during closed-chest resuscitation was defined as the pressure difference between the aorta and right atrium immediately before chest compression. Depth of compression was measured using a displacement transducer (DSPL, World Precision Instruments) enabling continuous recording throughout the interval of chest compression, and manually with a ruler at the end of such interval. Both data were reported given a few instances in which the displacement transducer failed. The ratio between coronary perfusion pressure and compression depth (CPP/depth) was used to assess changes in left ventricular compliance during chest compression as reported previously (37).

Co-EDTA Space.

Co-EDTA was used to measure the left ventricular extracellular space (ECS) based on techniques previously described by Holman (27) and Goldberg (19). Co-EDTA is presumed to cross cell membranes in negligible quantities (50) such that concomitant measurement of Co3+ in plasma and in tissue after intravenous administration of Co-EDTA allows estimation of the ECS according to the following equation: Math(1) where MCot is the molar amount of Co-EDTA in tissue (mmol), [Co]p is Co-EDTA concentration in plasma (mM), Wt is tissue weight (kg), and 1.053 is density of rat heart tissue (kg/l) (55).

For this purpose, a bolus of Co-EDTA (100 mg/kg; prepared by dissolving 100 mg of Na[Co-EDTA]·2H2O in 1.0 ml of 0.9% NaCl) was given intravenously (1.0 ml/kg) before harvesting. We confirmed previous studies (50) reporting no measurable physiological effects elicited by such Co-EDTA dose during spontaneous circulation (n = 9 rats). To monitor for possible leak of Co-EDTA to the intracellular space, the interval between Co-EDTA injection and heart harvesting was varied within groups injecting Co-EDTA at 10, 20, or 30 min before harvest in BL and PR groups and 25, 35, or 45 min in VF and VF/CC groups to account for the 15-min interval of untreated VF. Intracellular leak would manifest as time-dependent increases in the Co-EDTA space.

Left Ventricular Tissue Analysis for Determination of Intracellular Na+ and Mitochondrial Ca2+


Trace metal grade nitric acid (HNO3), perchloric acid (HClO4), sulfuric acid (H2SO4), American Chemical Society grade cobalt chloride hexahydrate (CoCl2·6H2O), sodium acetate (CH3COONa), EDTA, 30% hydrogen peroxide (H2O2), and HPLC grade alcohol were purchased from Fisher. Highest purity (SigmaUltra) grade sodium chloride (NaCl), calcium carbonate (CaCO3), sucrose, HEPES, EGTA, HPLC grade (+)-cis-diltiazem hydrochloride (diltiazem), and technical grade ruthenium red (98%) were purchased from Sigma. Co-EDTA sodium salt (Na[Co-EDTA]·2H2O), which was prepared according to the method reported by Scheufler and Peters (50). A mixture of CoCl2·6H2O (8 g), CH3COONa (20 g), and EDTA (10 g) in water (60 ml) was heated to near boiling temperature. H2O2 3% (30 ml) was added gradually to obtain a deep red solution, after which the solution was allowed to equilibrate with room temperature and alcohol added to crystallize and wash. Na[Co-EDTA]·2H2O crystals were obtained after four or five crystallization cycles.

Tissue Processing.

Immediately before the heart was removed, arterial blood (≈0.5 ml) was withdrawn into a heparinized syringe, transferred to a 1.5-ml Eppendorff tube, centrifuged at 2,300 g for 10 min and the plasma fraction (≈0.3 ml) stored at −20°C until processing for Na+ and Co-EDTA. The heart was then rapidly excised through a midline sternotomy and placed in an ice-chilled Petri dish. The right ventricle and both atria were removed and ≈600 mg of the left ventricle was apportioned for measuring intracellular Na+ (posterolateral wall and posterior portion of the septum) and ≈100 mg for measuring mitochondrial Ca2+ (anterolateral wall and anterior portion of septum). The fraction for intracellular Na+ measurement was kept at −20°C until processing, whereas the fraction for mitochondrial Ca2+ measurement was processed immediately (see below under Mitochondrial Ca2+).

Co-EDTA and Na+ Measurements.

The frozen left ventricular tissue was lyophilized for 24 h at −50°C and <30 millibars (Lyph-Lock 4.5 Liter Freeze Dry System, Labconco) determining its water content by differential weight measurement before and after complete lyophilization. The lyophilized tissue was then powderized using a mixer mill (MM200, Retsch) set at 30 Hz for 5 min. Plasma (0.2 ml) and the powderized tissue (≈30 mg dry wt) was acid digested using a 3:1:1 (vol:vol:vol) solution of HNO3, HClO4, and H2SO4 mixture (0.8 ml) at 80°C overnight. Using deionized water, the plasma and tissue digestates were diluted 400 and 200 times for Na+ measurement and 100 and 20 times for Co-EDTA measurement, respectively. A Varian SpectrAA·640 system (Varian) equipped with modules for flame atomic absorption spectrometry (FAAS) and for high sensitive graphite furnace atomic absorption spectrometry (GFAAS) was used with the aid of SpectrAA 5 PRO software (Varian). Na+ was measured using FAAS at 589.6 nm, whereas Co-EDTA was measured using GFAAS detecting Co3+ at 242.5 nm. A Na+ and Co3+ stock solution was prepared by dissolving 101.7 mg NaCl and 80.8 mg CoCl2·6H2O in 2 liters deionized water. The stock solution was diluted in deionized water to obtain 10-ml aliquots of 0.1, 0.2, 0.5, 1.0, and 2.0 μg/ml of Na+ standard solutions and 1.0 ml aliquots of 10, 20, 50, and 100 ng/ml of Co3+ standard solutions and used to construct Na+ and Co3+ standard curves. Samples were measured in aliquots of 5.0 ml for Na+ and 15 μl for Co3+ with standard Na+ (0.5 μg/ml) and Co3+ (50 ng/ml) solutions measured every 10 samples for quality control.

Intracellular Na+ ([Na]i) in millimoles per liter was calculated according to the following equation: Math(2) where MNat is molar amount of Na+ in tissue (mmol), [Na]p is Na+ concentration in plasma (mM), and ECS is extracellular space determined using Co-EDTA (Eq. 1).

Mitochondrial Ca2+.

The 100-mg sample of left ventricular tissue separated after heart harvesting was quickly minced into small pieces and immersed into ice-cold inhibitor buffer as described by Pepe et al. (47). The inhibitor buffer was composed of 250 mM sucrose (to maintain osmotic pressure and preserve mitochondrial structural integrity), 3.2 mM ruthenium red (to block the mitochondrial Ca2+ uniporter), 30 mM diltiazem (to block mitochondrial Na+-Ca2+ exchange), 2 mM EGTA (to chelate free extra-mitochondrial Ca2+), and 10 mM HEPES; the last two were adjusted to pH 7.40 using KOH. The sample was then hand homogenized using a Dounce homogenizer at a temperature between 0 and 4°C to minimize Ca2+ redistribution. Mitochondria were then separated by differential centrifugation as described by Hansford et al. (23). Briefly, the left ventricular homogenate was diluted to 5 ml with inhibitor buffer and centrifuged at 1,000 g for 5 min to pellet nuclei and debris. The supernatant was further centrifuged at 27,000 g for 2.5 min and the pellet (mitochondrial fraction) resuspended in 0.2 ml of inhibitor buffer and frozen at −80°C until analysis. The frozen mitochondrial fraction was thawed on ice and 0.05 ml of the suspension added to 0.2 ml of high purity HNO3 for overnight digestion at 80°C. The digestate was diluted 100 times with 0.5% HNO3 (to prevent phosphate interference) before measurement using GFAAS at 422.7 nm. Ca2+ standard stock solution (5.0 μg/ml) was prepared by dissolving 25 mg CaCO3 in 2.0 liters 0.5% HNO3 solution (instead of deionized water to completely dissolve CaCO3). The stock solution was then diluted 5,000, 2,500, 1,000, and 500 times using 0.5% HNO3 solution to obtain 1.0 ml aliquots of 1.0, 2.0, 5.0, and 10.0 ng/ml of Ca2+ standard solutions, respectively. Samples were measured in aliquots of 1.0 ml with standard Ca2+ solution (5.0 ng/ml) measured every 10 samples for quality control. Mitochondrial total protein was estimated using a protein assay kit (Micro BCA Protein Assay Kit, Pierce). Values of mitochondrial Ca2+ were expressed as nanomoles per milligram protein.

Cardiac Troponin I

Plasma obtained at baseline and after resuscitation in the PR groups were assayed for cTnI using a commercially available one step “sandwich” enzyme immunoassay method developed for human cTnI (Dimension Clinical Chemistry System using Cardiac Troponin-I Flex Reagent Cartridge, Dade Behring). The method had previously been shown to have excellent reactivity and specificity for rat cTnI (92.8% homology; Refs. 6, 12, 45, 46).

Na+-K+-ATPase Activity in Left Ventricular Tissue

Activity of the Na+-K+-ATPase was measured in a small subset of additional experiments after observing that intracellular Na+ remained elevated at 60-min postresuscitation, using the same VF and resuscitation protocol as in the main studies. The technique previously described by Schwinger et al. (52) and Fuller et al. (14) was used with minor modifications. Hearts were harvested at 60 min postresuscitation and rinsed in ice-cold Tris-buffered saline buffer. The right ventricle and both atria were removed and the left ventricle rapidly frozen by immersion in liquid N2. The frozen left ventricular tissue was then weighed, diced, and agitated in 10 ml of a high salt solution (2 M NaCl, 20 mM HEPES, pH 7.40) for 30 min at 4°C to depolymerize myofilaments. The tissue was then rinsed and homogenized using a Polytron homogenizer in a buffer containing 20 mM HEPES, 250 mM sucrose, 2 mM EDTA, 1 mM MgCl2, 0.2 mM PMSF (pH 7.40) to a final concentration of 10 ml/g wet wt of left ventricular tissue. Aliquots of ≈3 ml were stored at −80°C until analysis. Protein concentration was determined by the BCA method. The Na+-K+-ATPase activity was determined by measuring the ouabain-inhibitable generation of Pi in the presence of excess ATP. All reagents and solutions were prepared in phosphate-free glassware.

Assay reaction:

Two aliquots of total homogenate (500 μg protein, ≈65 μl) were mixed in 1.5 ml tubes with 65 μl of reaction buffer (200 mM Tris·HCl, 30 mM MgCl2, 200 mM NaCl, 60 mM KCl, 10 mM EGTA, 0.2 mM PMSF, and protease inhibitor cocktail, pH 7.50) with ouabain (2 mM) in one aliquot and without ouabain in the other. The aliquots and buffer were mixed gently and kept on ice for 1 h to achieve maximal Na+-K+-ATPase inhibition. The assay reaction was started by adding 14.4 μl of 0.1 M ATP (pH 7.00) to a 10 mM final concentration followed by incubation at 37°C. The reaction was stopped after 10 min by adding 13 μl of 100% (wt/vol) ice-cold trichloroacetic acid. The samples were left on ice for 1 h to facilitate precipitation of proteins and then centrifuged at 20,000 g for 30 min. The supernatant (60 μl) was assayed for Pi by a colorimetric method described by King (34). The supernatant (60 μl) was transferred to tubes containing 1.5 ml of 0.5% TCA and incubated at room temperature for 10 min with 150 μl of 5% ammonium molybdate and 60 μl of ANS reagent [0.25% aminonaphtholsulfonic acid (ANSA) in 15% NaHSO3 and 6% Na2SO3]. Absorbance was measured at 660 nm and converted to micromoles of phosphate using a standard curve prepared with KH2PO4. The Na+-K+-ATPase activity was expressed as micromoles Pi per milligram protein per hour.

Data Analysis

A primary analysis was performed to determine whether limiting sarcolemmal Na+ entry attenuated cytosolic Na+ increases and whether such effect influenced mitochondrial Ca2+ content and myocardial injury and function. For this analysis, the data from all three Na+-limiting interventions were pooled together. A secondary analysis was then performed to detect differences among the three Na+-limiting interventions.

Statistical Analysis

For continuous variables, Student's t-test was used when comparing differences between Na+-limiting interventions and control and one-way ANOVA when comparing differences among the three Na+-limiting interventions and among the various time events (applying Dunnett's method for multiple comparisons if overall differences were detected). In addition, two-way ANOVA was used in VF/CC and PR groups to simultaneously test for treatment and time factors and to examine potential interactions between these factors. The strength of association between variables was analyzed using Pearson's product moment correlation test. Alternative nonparametric tests were used if the data failed tests for normality or equal variance. The data were presented as means ± SD unless otherwise stated. A two-tail value of P < 0.05 was considered significant.


Hemodynamic Effects

Baseline measurements and cumulative dose of pentobarbital sodium given before induction of VF were comparable among all groups. During closed-chest resuscitation, adjustments in compression depth successfully maintained the CPP above 20 mmHg in each rat in the VF/CC and PR groups. Yet the required depth of compression was less in the 36 rats treated with Na+-limiting interventions compared with the 12 control rats (1.40 ± 0.11 vs. 1.50 ± 0.19 cm using the manually measured depth; P = 0.028 by Student's t-test). There were no statistically significant differences among the Na+-limiting interventions despite a lower depth favoring the AVE4454B and lidocaine combination (AVE4454B, 1.42 ± 0.11 cm; lidocaine, 1.41 ± 0.12 cm; and AVE4454B/lidocaine, 1.38 ± 0.11 cm). The time course of compression depth and the CPP/depth ratio are shown in Fig. 2.

Fig. 2.

Depth of chest compression (Depth, A) and ratio between the coronary perfusion pressure (CPP) and depth of compression (CPP/Depth, B) during closed-chest resuscitation in rats treated with Na+-limiting interventions (•, n = 36) or vehicle control (○, n = 12). Failure of the displacement transducer for the initial 3 min of chest compression in experimental groups (1 rat) and spontaneous defibrillation with return of circulation before completion of the 5-min interval of chest compression in 3 experimental and 1 control explain fewer observations (in brackets). Right panels depict effects of individual Na+-limiting interventions at the 5th min of chest compression. A, AVE4454B and lidocaine vehicle; L, AVE4454B vehicle and lidocaine; A/L, AVE4454B and lidocaine. Means ± SE. *P = 0.030 vs. control by Student's t-test.

Efforts to terminate VF and reestablish spontaneous circulation in the PR groups are shown in Fig. 3. Spontaneous defibrillation with return of spontaneous circulation [previously reported by us associated with administration of the NHE-1 inhibitor cariporide in rats (16)] occurred before completion of the 5-min interval of chest compression or shortly thereafter in one control, two AVE4454B-, four lidocaine-, and three AVE4454B/lidocaine-treated rats. As a result, fewer shocks and less cumulative energy were required to terminate VF in Na+-limiting interventions. Recurrence of VF in the early postresuscitation period (<15 min) required delivery of additional electrical shocks in control and in AVE4454B groups (unexpectedly) but not in the lidocaine and AVE4454B/lidocaine groups (Fig. 3).

Fig. 3.

A: shocks required in each rat to terminate VF during closed-chest resuscitation (R). B: shocks required in each rat to terminate episodes of refibrillation during the PR interval. C, vehicle control. The number in brackets denotes number of shocks delivered. *P < 0.05 for differences among the 4 groups for Kruskal-Wallis one-way ANOVA on ranks. Dunn's method for multiple comparisons failed to identify specific group differences.

After return of spontaneous circulation, statistically significant reductions in cardiac index and cardiac work index along with statistically insignificant decreases in mean aortic pressure occurred in all groups, consistent with postresuscitation myocardial dysfunction (Fig. 4). However, Na+-limiting interventions attenuated decreases in mean aortic pressure and cardiac work index, although no statistically significant differences were detected among the individual interventions.

Fig. 4.

A: mean aortic pressure (MAP) at BL, at 15 min of untreated VF, at 15 min of VF with chest compression for the last 5 min (VF/CC), and at 60 min PR. B: cardiac work index (CWI) and cardiac index (CI) at BL and PR. Open bars denote control rats and closed bars rats treated with Na+-limiting interventions. Shaded bars denote effects of individual Na+-limiting interventions. Hatched bars denote BL and VF. Means ± SE. *P < 0.05 vs. BL by Kruskal-Wallis one-way ANOVA on ranks using Dunn's method for multiple comparisons; ‡P < 0.001 vs. control by Student's t-test; §P < 0.05 vs. BL by one-way ANOVA using Dunnett's Method for multiple comparisons; †P < 0.05 vs. control by Student's t-test.

Left Ventricular Co-EDTA Space

The left ventricular Co-EDTA space at baseline and during VF corresponded to 21.1% and 20.7% of the total left ventricular volume without dependency on the time of Co-EDTA administration and was consistent with values reported by other investigators using Co-EDTA (40), TmDOTP5−, inulin, and mannitol (10). During VF/CC, the left ventricular Co-EDTA space decreased to 18.1% in control hearts and to 18.3% in Na+-limiting interventions hearts [not significant (NS)], without differences among interventions (Fig. 5A). The Co-EDTA space, however, markedly increased in the PR group to 29.2% in control hearts and to 26.6% in hearts subjected to Na+-limiting interventions, without statistically significant differences among specific interventions. The increases in Co-EDTA space occurred with minimal changes in left ventricular water content, which varied between 75.9% at baseline and 77.9% at VF/CC (Fig. 5B), suggesting that increases in the Co-EDTA space after resuscitation were unlikely the result of increases in the extracellular space. Such effect would require concomitant and proportional reductions in the intracellular space and would be inconsistent with the reported increases (rather than decreases) in the intracellular space during and after ischemia (2). We therefore elected to apply the baseline measurements of extracellular space (21.1%) to the estimations of intracellular Na+ for all groups.

Fig. 5.

Left ventricular Co-EDTA distribution space (A) and water content (B) at BL, at 15 min of untreated VF, at 15 min of VF/CC, and at 60 min PR. Open bars denote control rats and closed bars rats denote those treated with Na+-limiting interventions. Shaded bars denote effects of individual Na+-limiting interventions. Hatched bars denote BL and VF. Means ± SE. *P < 0.05 vs. BL by Kruskall-Wallis 1-way ANOVA on ranks using Dunn's Method for multiple comparisons; #2-way ANOVA using time factor (VF/CC vs. PR) and treatment factor (control vs. Na+-limiting interventions) was significant for time factor (P < 0.001).

Left Ventricular Intracellular Na+

Left ventricular intracellular Na+ increased from 11.8 ± 3.0 mM to 16.5 ± 1.5 mM after 15 min of VF (NS). In control hearts, intracellular Na+ additionally increased to 18.5 ± 1.8 mM when VF included chest compression and to 22.6 ± 11.1 mM postresuscitation (P < 0.05 for each group vs. BL by Kruskal-Wallis one-way ANOVA on ranks using Dunn's method for multiple comparisons). Na+-limiting interventions prevented such increases in intracellular Na+, yielding levels of 16.6 ± 3.7 mM in VF/CC groups and 15.7 ± 4.0 mM in PR groups without statistically significant differences among the specific interventions despite lower levels in the AVE4454B/lidocaine group (Fig. 6A).

Fig. 6.

Left ventricular intracellular Na+ ([Na+]i) (A) and mitochondrial Ca2+ ([Ca2+]m) (B) at BL, at 15 min of VF, at 15 min of VF/CC, and at 60 min PR. Open bars denote control rats and closed bars denote rats treated with Na+-limiting interventions. Shaded bars denote effects of individual Na+-limiting interventions. Hatched bars denote BL and VF. Means ± SE. *P < 0.05 vs. BL by Kruskal-Wallis 1-way ANOVA on ranks using Dunn's method for multiple comparisons; †P < 0.05 vs. control by Student's t-test in PR groups; #2-way ANOVA using time factor (VF/CC vs. PR) and treatment factor (control vs. Na+-limiting interventions) was significant for treatment factor (P = 0.013) for [Na+]i and for both time factor (P = 0.045) and treatment factor (P = 0.021) for [Ca2+]m.

Left Ventricular Mitochondrial Ca2+

Left ventricular mitochondrial Ca2+ remained relatively unchanged during the 15-min interval of untreated VF and during the interval of VF with chest compression. However, mitochondrial Ca2+ increased to ≈140% of baseline during the postresuscitation interval in control hearts. In rats subjected to Na+-limiting interventions, mitochondrial Ca2+ corresponded to 77% and 71% of control hearts at VF/CC and PR, without statistically significant differences among the specific Na+-limiting interventions (Fig. 6B). The changes in mitochondrial Ca2+ were correlated with changes in intracellular Na+ levels at PR (r = 0.44, n = 24, P < 0.04) but not at VF/CC (r = 0.11, n = 24, P = 0.63). Similar analysis within individual groups at PR and VF/CC failed to disclose significant correlations (data not shown). This was likely the result of small sample size and reduced variability within groups given the controlled nature of the experiments.


Plasma cTnI increased in control hearts from 0.7 ± 0.5 at baseline to 48.6 ± 30.2 ng/ml at 60 min postresuscitation. In rats treated with Na+-limiting interventions, plasma cTnI increased from 0.5 ± 0.3 at baseline to 27.5 ± 13.0 ng/ml at 60 min postresuscitation (P < 0.05 vs. control by Student's t-test), without statistically significant differences among the specific interventions (Fig. 7). Plasma cTnI levels at 60 min postresuscitation were positively correlated with left ventricular intracellular Na+ and left ventricular Co-EDTA distribution space and inversely correlated with cardiac work index (Fig. 7).

Fig. 7.

Plasma cardiac troponin I (cTnI) at BL and at 60 min PR in rats subjected to VF and resuscitation. Bar graphs: open bar denotes control rats; closed bar denotes rats treated with Na+-limiting interventions. Shaded bars denote effects of individual Na+-limiting interventions. Hatched bar denotes BL. Means ± SE. *P < 0.05 vs. control by Student's t-test. Scatterplots: depict the correlations and linear regressions between cTnI and cardiac work index (CWI), left ventricular [Na+]i, and left ventricular Co-EDTA distribution space at 60 min PR. •, AVE4454B and lidocaine vehicle; ▾, AVE4454B vehicle and lidocaine; ▴, AVE4454B and lidocaine; ○, AVE4454B vehicle and lidocaine vehicle (control).

Na+-K+-ATPase Activity

In separate studies (including 2 subjects per group), the Na+-K+-ATPase activity measured at 60 min postresuscitation in control rats was reduced to ≈40% of baseline, from 2.82 ± 0.18 to 1.16 ± 0.39 μmol Pi·mg protein−1·h−1 (P < 0.05 by Student's t-test). Compared with the control group, treatment with AVE4454B was associated with an additional reduction in Na+-K+-ATPase activity (0.69 ± 0.27 μmol Pi·mg protein−1·h−1, P < 0.05), treatment with lidocaine was associated with increased Na+-K+-ATPase activity (4.43 ± 0.43 μmol Pi·mg protein−1·h−1, P < 0.05), and treatment with the AVE4454B/lidocaine combination preserved the Na+-K+-ATPase activity at baseline levels (2.69 ± 0.32 μmol Pi·mg protein−1·h−1, NS) by one-way ANOVA and Bonferroni t-test for multiple comparisons against control.


The present studies demonstrate that limiting sarcolemmal Na+ entry during resuscitation from VF 1) prevents left ventricular intracellular Na+ increases, 2) maintains lower mitochondrial Ca2+ levels during and after resuscitation, 3) attenuates reductions in left ventricular compliance during chest compression, 4) lessens myocardial injury, and 5) ameliorates postresuscitation myocardial dysfunction. No statistically significant differences could be demonstrated among the various Na+-limiting interventions except for failure of AVE4454B to prevent recurrence of VF.

Na+ Limiting Interventions

We previously (18) reported comparable increases in left ventricular intracellular Na+ in an isolated rat heart model of VF in which clinical resuscitation was simulated by providing 10 min of no-flow ischemia followed by 15 min of low-flow reperfusion. In these studies, increases in intracellular Na+ were halved if VF was absent despite identical periods of no-flow and low-flow reperfusion, highlighting the importance of VF for promoting cytosolic Na+ overload. VF may contribute to intracellular Na+ overload through Na+ channel activation followed by slow inactivation during ischemia (7). However, VF could also contribute by intensifying ischemic injury (17) and exacerbating determinants of Na+ overload such as energy depletion with further inhibition of the Na+-K+-ATPase and by intensifying intracellular acidosis increasing the rate of Na+-H+ exchange.

The present study points also to reperfusion as a factor that can intensify intracellular Na+ overload. For the same duration of VF (i.e., 15 min), intracellular Na+ was higher when chest compression (i.e., reperfusion) was provided during the last 5 min of VF (Fig. 6). Reperfusion has been shown to increase the rate of Na+ entry, and attributed to the wash out of protons accumulated in the extracellular space during the period of no-flow ischemia (28). These observations are applicable to the setting herein modeled in which ischemia and reperfusion were of short duration, global, and compounded by the presence of VF. The findings, however, may not be directly extrapolated to other settings, such a regional ischemia without VF.

In addition to the (expected) aforementioned effects, we found (unexpectedly) that the intracellular Na+ levels were substantially higher at 60 min postresuscitation than during chest compression, challenging the presumption that restoration of aerobic metabolism could reverse the processes leading to intracellular Na+ accumulation. The possibility that determinants of intracellular Na+ entry and overload such as intracellular acidosis and decreased sarcolemmal Na+-K+-ATPase activity persisted at 60 min postresuscitation was considered and additional exploratory experiments were conducted to assess the Na+-K+-ATPase activity after resuscitation. These studies demonstrated reductions in Na+-K+-ATPase activity at 60 min postresuscitation to ≈40% of baseline, suggesting that persistent intracellular Na+ increases could have been in part attributed to decreased removal of intracellular Na+ excess. Na+-limiting interventions had intriguing effects on Na+-K+-ATPase activity. AVE4454B reduced, lidocaine increased, and the AVE4454B/lidocaine combination maintained the Na+-K+-ATPase activity at baseline levels. These observations suggest the possibility of additional effects beyond those related to ion homeostasis. Understanding these effects would require additional experiments beyond the scope of the present studies.

Another factor that may affect intracellular Na+ is changes in NHE-1 activity. Studies in isolated rat hearts have shown sevenfold increases in NHE-1 messenger RNA early after ischemia and reperfusion (15). NHE-1 activity has been shown to be increased by α1-adrenergic stimulation, which can occur as part of the neuroendocrine response to cardiac arrest (62, 63).

Regardless of mechanisms, administration of Na+-limiting interventions halted increases in left ventricular intracellular Na+ during the interval of chest compression and following return of spontaneous circulation, yielding levels comparable to those observed during VF in the absence of reperfusion. This effect was associated with lower mitochondrial Ca2+ levels.

Changes in mitochondrial Ca2+ are directionally coupled with changes in cytosolic Ca2+ via the mitochondrial Ca2+ uniporter (Ca2+ entry) and various antiporters, including the mitochondrial Na+-Ca2+ and H+-Ca2+ antiporters (Ca2+ efflux) such that increases in cytosolic Ca2+ would be expected to increase mitochondrial Ca2+ and vice versa (9, 29, 35, 42, 48). Accordingly, attenuation in mitochondrial Ca2+ increases by Na+-limiting interventions in the present studies is best explained by reductions in Na+-induced cytosolic Ca2+ entry (and overload). Mitochondrial Ca2+ overload during ischemia and reperfusion has been shown to worsen mitochondrial injury, compromising its capability to sustain oxidative phosphorylation (61) and promoting the release of proapoptotic factors (8).

The functional benefits associated with Na+-limiting interventions were as previously reported in various animal models of VF and closed-chest resuscitation associated with NHE-1 inhibition (5, 16, 37). First, they enabled hemodynamically more effective chest compression such that less depth of compression was required to attain a predetermined coronary perfusion pressure. This effect is best explained by amelioration (or prevention) of reductions in left ventricular compliance that leads to preload-dependent reductions in forward blood flow (5, 36, 37). Second, Na+-limiting interventions (with the exception of AVE4454B as discussed below) prevented episodes of refibrillation, eliminating the need for additional electrical shocks during the early postresuscitation interval. Finally, Na+-limiting interventions reduced myocardial cell injury, as evidenced by lower postresuscitation plasma cTnI increases, and attenuated postresuscitation myocardial dysfunction, yielding a higher cardiac work index with a higher mean aortic pressure.

Postresuscitation myocardial injury and dysfunction was accompanied by a higher Co-EDTA distribution space. We interpreted these findings as indicative of disruption of sarcolemmal membrane enabling Co-EDTA to enter the intracellular space and cTnI to exit it and be measured in plasma. Consistent with this explanation, the Co-EDTA distribution space was lower in rats treated with Na+-limiting interventions, which also had lower plasma cTnI and higher cardiac work index. Likewise, the Co-EDTA distribution space was positively correlated with intracellular Na+ and plasma cTnI and negatively correlated with cardiac work index. We also observed lack of time dependency between the time of tracer injection and Co-EDTA distribution space, favoring a mechanism of injury affecting a small percentage of cells, leading to rapid intracellular Co-EDTA equilibration rather than a diffuse alteration of plasma membrane permeability with slow Co-EDTA intracellular “leakage.”

Differences Among Interventions

There were no statistically significant differences among the three Na+-limiting interventions except that AVE4454B failed to prevent recurrent episodes of VF during the early postresuscitation interval. This contrasted with the prominent and consistent suppression or reperfusion arrhythmias reported using other NHE-1 inhibitors, including cariporide (5, 16, 20, 59). Although the present data do not provide direct clues on the mechanisms, previous observations using the more broadly known inhibitor cariporide might shed some light. Cariporide given in bolus dose during closed-chest resuscitation in pigs produced plasma levels that far exceeded the IC90 of NHE-1 (4) and reached levels that were previously shown to inhibit slowly inactivating Na+ currents (51). These observations suggested that some of the benefits of cariporide could be mediated through Na+ channel blockade. Lidocaine, which characteristically inhibits these channels, minimizes reperfusion arrhythmias (39, 53). Accordingly, greater selectivity of AVE4454B (lacking nonselective effects on slowly inactivating Na+ currents) could have been a potential factor. Work beyond the scope of the present studies would be required to resolve this issue.

Lidocaine showed cardioprotective effects comparable to those of AVE4454B with the added benefits on reperfusion arrhythmias. Lidocaine has been shown in several preclinical studies to be cardioprotective (26, 44, 49) through mechanisms that involve attenuation of cytosolic Na+ and Ca2+ overload (21, 53). These findings, however, contrast with the poor performance shown in clinical trials in patients suffering out-of-hospital cardiac arrest (11). In these trials, however, lidocaine was administrated 25 min after ambulance dispatch. Yet, one retrospective study showed higher resuscitability and hospital admission rates in patients who had received lidocaine (25).

Whether an additive beneficial effect occurred between AVE4454B and lidocaine cannot be established based on the present data. Numerical differences favoring the combination were observed with regards to depth of compression and intracellular Na+ during chest compression and intracellular Na+ and cTnI at 60 min postresuscitation. These differences, however, were not statistical significance. Larger sample sizes per group would be required (corresponding to 141 for depth of compression, 35 for intracellular Na+, and 25 for cTnI) to demonstrate statistical significance for the reported numerical differences using a power of 90%.

Limitations of Study


Extrapolation of the present findings to clinical settings is constrained by several factors. VF was induced electrically after prolonged electrical stimulation, whereas clinically VF typically occurs in patients with underlying coronary artery disease and is often precipitated by coronary occlusion. In addition, we cannot exclude that alterations in membrane composition occurred consequent to electrical stimulation. However, in a previous study using a Langendorff rat heart preparation, similar electrical stimulation at normal perfusion flows elicited only transient myocardial depression after return of sinus rhythm without changes in myocardial oxygen consumption (17). Inherent to animal models is the use of anesthesia, which exerts independent myocardial protective effects (33). However, the rat model herein used has been highly effective in the development and testing of novel concepts with subsequent confirmation in larger animal models, particularly when examining highly conserved mechanisms that transcend species differences (i.e., Na+-H+ exchange, Na+ channel kinetics, and mitochondrial function).


We examined changes in mitochondrial Ca2+ by measuring total instead of the physiologically more relevant matrix free Ca2+. However, because previous studies had shown a good linear correlation between total and matrix free mitochondrial Ca2+ in the isolated rat heart (22) such compromise appeared reasonable given the limitations for measuring matrix free Ca2+ in intact animals. For measuring intracellular Na+ concentration, we concomitantly determined the Co-EDTA space as marker of the extracellular space based on previous studies indicating negligible plasma membrane permeability to Co-EDTA. However, we found postresuscitation increases in Co-EDTA space that could not be explained by increases in extracellular space (discussed in results). We elected to apply the baseline measurement of extracellular space for the calculations of intracellular space in all groups. The possibility that prominent changes in extracellular space occurred and were not detected seemed unlikely given the very minor changes in total left ventricular water. Even if small changes occurred, the effects of ischemia and reperfusion and the effects of Na+ limiting interventions were compared with proper controls.

Potential Clinical Significance

The effects of Na+-limiting interventions reported here and in previous publications (5, 37) are physiologically relevant and would be expected to yield higher resuscitation and survival rates in patients provided the effects could be replicated. Beyond the differences between animal models and clinical resuscitation already mentioned, the timing of drug delivery is likely to be a critical determinant of efficacy. It is widely accepted that the maximal efficacy of NHE-1 inhibitors is attained when given before the onset of ischemia (58), with the efficacy diminishing as delivery is delayed during reperfusion (43). We developed a protocol to simulate drug delivery at the earliest possible time in a clinical scenario (i.e., after the onset of VF but before the start of cardiopulmonary resuscitation). Implementation of such approach in humans would require a paradigm shift, prompting drugs to be delivered before starting chest compression. From this perspective, the intraosseous route for vascular access is available in both children and adults without the limitations and delays imposed by cannulation of peripheral or central veins in a crisis situation.


We confirmed that limiting sarcolemmal Na+ entry represents an important therapeutic goal during resuscitation from VF that helps attenuate ischemia and reperfusion injury and preserve left ventricular function. Given that NHE-1 inhibition and Na+ channel blockade yielded comparable effects and given that NHE-1 inhibitors are not yet clinically available, use of lidocaine for the purpose of limiting sarcolemmal Na+ entry by early administration during resuscitation from VF should be considered.


Work supported by National Heart, Lung, and Blood Institute Grant R01-HL-71728-01 entitled “Myocardial Protection by NHE-1 Inhibition” and a VA Merit Review Grant entitled “Myocardial Protection During Ventricular Fibrillation.”


The authors thank Linda C. Dowell, Laboratory Manager at the North Chicago VA Medical Center, for conducting the determinations of plasma cardiac troponin I.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract