This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pike, M. M. Articles by Anderson, P. G. Search for Related Content PubMed PubMed Citation Articles by Pike, M. M. Articles by Anderson, P. G.

(Circulation Research. 1995;77:394-406.)
© 1995 American Heart Association, Inc.

Articles

²³Na and ³¹P Nuclear Magnetic Resonance Studies of Ischemia-Induced Ventricular Fibrillation

Alterations of Intracellular Na⁺ and Cellular Energy

M. M. Pike, C. S. Luo, S. Yanagida, G. R. Hageman, P. G. Anderson

From the Department of Medicine (M.M.P., C.S.L., S.Y.), Division of Cardiovascular Disease; the Department of Physiology and Biophysics (G.R.H.); and the Department of Pathology (P.G.A.), Division of Molecular/Cellular Pathology; University of Alabama at Birmingham.

Correspondence to Dr M.M. Pike, Department of Medicine, Division of Cardiovascular Disease, 703 S 19th Street, ZRB 308, Birmingham, AL 35294-0007.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract To clarify the role of Na⁺_i, pH_i, and high-energy phosphate (HEP) levels in the initiation and maintenance of ischemia-induced ventricular fibrillation (VF), interleaved ²³Na and ³¹P nuclear magnetic resonance spectra were collected on perfused rat hearts during low-flow ischemia (51 minutes, 1.2 mL/g wet wt). When untreated, 50% of the hearts from normal (sham) rats and 89% of the hypertrophied hearts from aortic-banded (band) rats (P<.01 versus sham) exhibited VF. Phosphocreatine content was significantly higher in sham than band hearts during control perfusion (53.3±1.6 versus 39.8±2.0 µmol/g dry wt). Before VF at 20 minutes of ischemia, Na⁺_i accumulation was greater in hearts that eventually developed VF than in hearts that did not develop VF for both band and sham groups (144% versus 128% of control in sham; P<.005) and was the strongest metabolic predictor of VF; ATP depletion was also greater for VF hearts in the sham group. Infusion of the Na⁺-H⁺ exchange inhibitor 5-(N,N-hexamethylene)-amiloride prevented VF in sham and band hearts; reduced Na⁺_i accumulation but similar HEP depletion were observed compared with VF hearts before the onset of VF. Rapid changes in Na⁺_i, pH_i, and HEP began with VF, resulting in intracellular Na⁺_i overload (300% of control) and increased HEP depletion. A delayed postischemic functional recovery occurred in VF hearts, which correlated temporally with the recovery of Na⁺_i. In conclusion, alterations in Na⁺_i were associated with spontaneous VF transitions, consistent with involvement of excess Na⁺_i accumulation in VF initiation and maintenance and with previously reported alterations in Ca²⁺_i with VF. Hypertrophied band hearts exhibited enhanced susceptibility to ischemia-induced VF, possibly linked to a lower HEP reserve.

Key Words: ventricular fibrillation • intracellular Na⁺ • ischemia • nuclear magnetic resonance • cardiac hypertrophy

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Ventricular fibrillation (VF) is a lethal arrhythmia characterized by rapid unsynchronized depolarizations, resulting in ineffective contractile activity. It is often associated with myocardial infarction, heart failure, or myocardial reperfusion. Hypertrophied myocardium may have enhanced susceptibility to VF initiation.^{1 2 3 4} An understanding of the cellular basis of this life-threatening arrhythmia, in terms of both initiation and maintenance, is critical in terms of potential strategies for prevention or treatment. Nevertheless, the mechanisms underlying VF are still controversial. The classic mechanisms involve reentrant excitation, which is caused by areas of altered action potential conduction. However, various reports have indicated that the initiation and maintenance of VF could also be related to membrane-triggered activity activated by elevated cytosolic free Ca²⁺ (Ca²⁺_i).^{5 6 7 8 9 10 11 12 13} Direct measurements of Ca²⁺_i using the ¹⁹F NMR indicator 5F-BAPTA,¹⁴ the fluorescent indicator indo 1,¹⁵ and the bioluminescent indicator aequorin¹⁶ have revealed rapid Ca²⁺_i increases on electrically induced VF in perfused hearts. Aequorin measurements have also indicated Ca²⁺_i increases preceding VF in perfused ferret hearts exposed to acetylstrophanthidin¹⁶ ; these increases were followed by larger increases during VF. In a multicellular myocyte model, fura 2 measurements of Ca²⁺_i have indicated that Ca²⁺_i elevations preceded the development of arrhythmias and progressed further with induction of chaotic beating activity.¹³ These studies provide evidence for a role for Ca²⁺_i overload in the metabolic changes leading to^{13 16} and sustaining^{13 14 15 16} VF. Energy supply/demand imbalance can be impaired during VF, as indicated by increased oxygen consumption, lactate production, and ATP depletion as well as, in some cases, reduced functional recovery.^{14 17 18} However, to understand VF as it normally occurs in vivo, metabolic measurements investigating VF during a physiological model of ischemia are required. By use of NMR spectroscopy and other techniques, various studies have reported increases in Ca²⁺_i and/or Na⁺_i during myocardial ischemia^{19 20 21 22 23 24 25 26 27} and are generally consistent with the hypothesis that there is substantial coupling between the Na⁺_i and Ca²⁺_i levels during ischemia/reperfusion via Na⁺-Ca²⁺ exchange.^{28 29} However, most of these measurements have been performed with the global zero-flow ischemia model, which decreases excitability sufficiently to prevent VF. In the present study, we have used a low-flow ischemia model. The study used the ²³Na NMR shift reagent Tm(DOTP)^5- to resolve the intracellular and extracellular ²³Na NMR signals.³⁰ This reagent enables excellent resolution of the Na⁺_i resonance in perfused tissues and has the added benefit of allowing acquisition of excellent ³¹P NMR spectra as well. We have made use of this advantage by acquiring rapidly interleaved ²³Na and ³¹P NMR spectra, continuously monitoring Na⁺_i, pH_i, and high-energy phosphate levels in the same preparation.²³ The goal was to determine whether alterations in these parameters are associated with the initiation and/or maintenance of VF during ischemia. The hypothesis that excess Na⁺_i accumulation is associated with, or required for, VF induction during ischemia was tested by inhibition of a major Na⁺ influx pathway, Na⁺-H⁺ exchange. The relative susceptibility of normal and hypertrophied myocardium to VF was assessed under controlled ischemic conditions. The studies have provided new information concerning the relation between Na⁺_i, cellular energy levels, and ischemia-induced VF.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Heart Preparations
A total of 57 perfused hearts were included in the present study: 51 were used for NMR and biochemical measurements, and 6 were used for electrophysiological measurements. Both normal (sham, n=30) and aortic-banded (band, n=27) male Sprague-Dawley rats were used. Band rats were surgically banded between the innominate and left subclavian arteries at 3 weeks of age; aortic blood flow was restricted progressively with animal growth. Sham rats were subjected to the same anesthesia (Brevital, 5 mg IP) and surgery excepting band placement. All animals were killed at 8 weeks of age. The rats were heparinized (2000 U/kg IP), anesthetized with pentobarbital (60 mg/kg IP), and weighed. The hearts were removed, put in cold perfusate, and quickly trimmed, weighed, and Langendorff-perfused. The right atrium was removed, and atrioventricular conduction was surgically blocked. All hearts were stimulated at 150 beats per minute throughout the experiments by using a stimulator connected to the right ventricle via polyethylene tubing filled with KCl-saturated agar and tipped with thin platinum wires. The hearts were initially perfused at 80 mm Hg by using a previously described perfusion system²³ consisting of a modified Krebs-Henseleit perfusate aerated with 95% O₂/5% CO₂ (32°C, pH adjusted to 7.4 with NaOH) and containing the following (mmol/L): NaCl 112, NaHCO₃ 24, KCl 4.7, MgSO₄ 1.2, glucose 11, pyruvate 5, CaCl₂ 1.5, and EDTA 0.5 (Sigma Chemical Co). Hearts were then switched to the shift reagent perfusate, and coronary flow was then pump-controlled and adjusted to a constant rate of 12 mL/g wet wt. The shift reagent perfusate was identical to that described above, except for the inclusion of 4 mmol/L of the ²³Na NMR shift reagent Na₅Tm(DOTP); NaCl was adjusted to keep the total Na⁺ content unchanged, and total Ca²⁺ content was adjusted to 5.5 mmol/L to adjust for binding by the shift reagent, resulting in a free ionized Ca²⁺ of 1.07 mmol/L as measured by using a Ca²⁺ ion–selective electrode (Orion) and perfusate (32°C, 95% O₂/5% CO₂) calibration solutions (excluding shift reagent or EDTA). Tm(DOTP)^5- was synthesized, purified, and recovered after the experiment, as described previously.²³ All perfusate was filtered through 5-µm filters before the experiment and with in-line filters during the experiment. Heart function was monitored with a fluid-filled left ventricular balloon in line with a Spectramed p23XL transducer and an Astro-Med MT9500 multichannel recorder. Left ventricular end-diastolic pressure was set at 4 to 5 mm Hg. The functional data were recorded both on paper and digitally. NMR spectra were acquired continuously during a control period, 51 minutes of 10% low-flow ischemia (1.2 mL/g wet wt), and 60 minutes of reperfusion at the control flow rate. After the experiment, all hearts were weighed after drying to a constant weight at 80°C. All procedures involving animals conformed to University of Alabama at Birmingham Institutional Animal Care and Use Committee guidelines.

Pharmacological Interventions
The Na⁺-H⁺ exchange inhibitor HMA was dissolved in perfusate (617 µmol/L stock) and infused into the aortic line just above the heart (final concentration, 10 µmol/L) for 15 minutes before ischemia in subgroups of sham and band (sham-HMA and band-HMA). Infusion was continued during ischemia (same final concentration) and was terminated on reperfusion.

NMR Spectroscopy
The ²³Na and ³¹P NMR spectra were obtained with a Bruker AM-360 spectrometer with a homebuilt temperature-controlled switchable NMR probe, which could collect spectra from both nuclei without retuning. Data collection alternated automatically between the collection of ²³Na (1 minute) and ³¹P (2 minutes) NMR spectra. The ²³Na NMR spectral acquisition parameters and the signal area measurement methodology were as previously described.²³ Gaussian multiplication of the ²³Na free induction decays was used for signal-to-noise noise improvement and (minimal) resolution enhancement; the NMR1 GM subroutine was used with parameters G1=0, G2=20, and G3=0.1 (NMR1 software, New Methods Research, Inc).

NMR-visible Na⁺_i was quantified in the experiments by comparison with spectra obtained after the experiment from an NaCl solution contained within a heart-sized glass sphere. However, for all protocols, Na⁺_i was reported as percent control to focus on changes in the Na⁺_i rather than absolute differences and to reduce experimental variability resulting from quantification errors. Also, the reporting of Na⁺_i content requires assumptions concerning the ²³Na NMR visibility, which has not been directly determined for these conditions. Comparison between the experimental groups using percent control is valid because absolute control values for sham and band (19.1±0.5 and 18.1±1.0 µmol NMR visible Na⁺_i per g dry wt, respectively) were not significantly different. Also, no significant differences were found between subgroups of sham or of band.

The ³¹P NMR spectral acquisition parameters, plotting parameters, and signal area measurement methodology were as previously described.²³ Phosphorus metabolites were quantified by comparing the signal areas with those from the 100 mmol/L PPA standard solution in the left ventricular balloon; linear regression was performed on signal areas from several spectra obtained at the end of the experiment after incremental balloon volume increases. For each experiment, a fully relaxed spectrum, using a recycle time of 10 seconds, was acquired before the control period. From comparison with the control spectra, empirical saturation factors for the various resonances were derived: 1.58, 1.40, and 1.17 for PPA, PCr, and ATP, respectively. Low control P_i levels necessitated calculating a value (1.10) from the reported spin-lattice relaxation time (T₁) for P_i in rat heart³¹ by using Equation 17 from Becker et al.³² pH_i was calculated from the shift of the P_i resonance.³³

Lactate Measurements
Lactate efflux (in micromoles per minute per gram dry weight) was measured by analysis of the coronary effluent (assay kit No. 826, Sigma Chemical Co). During ischemia, the entire effluent was collected at 17-minute intervals, taking care to remove coronary effluent remaining in the NMR tube by external flushing. During reperfusion, the effluent was collected at 2, 3, 4, 5, 10, 15, 30, and 45 minutes of reperfusion.

Electrogram Measurements
Simultaneous electrogram and NMR measurements were not possible because of the introduction of radiofrequency noise. However, several sham (n=3) and band (n=3) hearts were perfused with shift reagent perfusate exactly as described above without collecting NMR data. Three bipolar plunge electrodes constructed of polytetrafluoroethylene (Teflon)–coated wire were placed in the ventricular myocardium. By use of an Astromed 9500 multichannel recorder, multiple local electrograms were simultaneously recorded from different positions on the ventricles.

Statistics
Values were reported as mean±SEM. An NCSS software package was used for statistical analysis. General linear models ANOVA was generally used; paired t tests were also used as indicated. Fishers exact test was used to compare the incidence of VF between groups. Results were considered significant at P<.05. To test VF predictability, the Number Cruncher Statistical System (NCSS) discriminant analysis routine was used as described.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Functional and Electrophysiological Measurements
Fig 1 shows portions of typical functional traces from two sham hearts subjected to ischemia and reperfusion. During early ischemia, both hearts maintained contraction at the paced heart rate at reduced developed pressure. In the heart depicted in the top trace, diminished contractions continued steadily throughout ischemia; contraction amplitude was quickly restored after reperfusion. In contrast, the heart depicted in the lower trace exhibited abrupt contractile failure, which continued until reperfusion despite stimulation. Unlike the heart in the top trace, diastolic tension rapidly increased on reperfusion, accompanied by chaotic oscillations; it decreased markedly on the restoration of contraction. These patterns suggested ischemia-induced VF.¹⁸ Fig 2 shows a local electrogram from one of several electrodes placed on a sham heart. The recordings taken minutes before ischemic contractile failure show the typical synchronized electrical activity of contracting myocardium. Seconds before contractile failure, the pressure and electrogram recordings indicated the presence of additional contractions between the pacing stimuli. After contractile failure, the heart was not quiescent but indicated the chaotic electrical activity typical of VF, which was not synchronized with that recorded from the other electrodes (not shown). Electrogram recordings from each of the hearts exhibiting contractile failure (two of three sham hearts, three of three band hearts) were quite similar, with initiation of sustained VF beginning at the time of contractile failure. In each case, VF continued into reperfusion and terminated at, or shortly before, the return of normal contraction. Like the lower trace in Fig 1, the pressure traces indicated abrupt contractile failure and recovery patterns as well as elevated diastolic tension and chaotic oscillations during early reperfusion; these were also the typical patterns for all of the hearts in the study that experienced contractile failure. This, combined with the electrogram data, strongly suggests that in this model of ischemia, contractile failure during ischemia and reperfusion is caused by VF.

View larger version (18K): [in this window] [in a new window]   Figure 1. Portions of functional traces indicating left ventricular pressure are shown for two sham hearts (at different chart recorder speeds). The traces indicate function during control perfusion, the midischemic period, and early and late reperfusion; the arrows indicate the onset of ischemia or reperfusion.

View larger version (32K): [in this window] [in a new window]   Figure 2. Representative LVDP (top traces) and local electrograms (lower traces) were recorded in myocardium from a band heart during low-flow ischemia and 2.5-Hz pacing at 10 minutes before contractile failure (left), 20 seconds before contractile failure (middle), and 20 seconds after contractile failure (right).

The percentage of all untreated sham hearts (n=24, including those used for NMR and electrogram measurements) exhibiting such contractile failure/VF during the ischemic period was 50%, a proportion significantly less than the 89% observed for untreated band hearts (n=18, P<.01). The mean ischemia times at which fibrillating sham (sham-VF) hearts and fibrillating band (band-VF) hearts went into VF were 35.8±2.3 and 29.8±2.1 minutes, respectively. Most interestingly, the preparations treated with HMA during ischemia, sham-HMA and band-HMA groups, all continued contracting throughout the protocol and did not exhibit VF, representing significantly lower incidences of VF than were observed in the untreated sham and band groups. This is consistent with previous reports that Na⁺-H⁺ inhibition reduces reperfusion arrhythmias.^{23 34} The mean heart dry weights, body weights, and heart dry weight–to–body weight ratios were 0.213±0.004 g, 357±7 g, and 6.01±0.13x10^-4, respectively, for the sham group and 0.281±0.008 g (P<.0001 versus sham), 320±10 g (P<.005 versus sham), and 8.92±0.28x10^-4 (P<.0001 versus sham), respectively, for the band group. The data clearly indicate the development of substantial hypertrophy in band hearts. There were no significant differences or different trends evident between the sham or the band subgroups for these parameters (data not shown). To eliminate from the study hearts from animals that may have been banded ineffectively, band hearts were not included in the study if the total heart wet weight–to–body weight ratio was lower than the mean value by more than 1 SD; only two of 29 band hearts were eliminated from the study in this fashion.

Table 1 summarizes LVDP during (pre-HMA) control and end reperfusion for all sham and band subgroups on which NMR measurements were obtained, including untreated hearts that contracted throughout ischemia (sham-C, n=11; band-C, n=2), HMA-treated hearts (sham-HMA, n=6; band-HMA, n=9), and hearts exhibiting contractile failure/VF (sham-VF, n=10; band-VF, n=13). As expected, the band hearts had significantly higher ventricular pressures than did the sham hearts because of hypertrophy of the ventricular wall. No significant differences were found between the sham subgroups or between the band subgroups in either control or end reperfusion LVDP. During reperfusion, a complete functional recovery was observed in all hearts, confirming that the VF was ischemia-induced. However, the recovery for the sham- and band-VF hearts was delayed compared with the sham- and band-C and -HMA groups. Steady contraction in the range of 10 mm Hg LVDP was maintained in all band- and sham-C and -HMA groups during ischemia, with diastolic tension remaining relatively stable throughout the protocol. In contrast, between early and late ischemia in sham-VF, diastolic tension increased from 3±1 to 14±1 mm Hg. On reperfusion, diastolic tension markedly increased to 47±6 mm Hg (at 2 minutes reperfusion). The mean diastolic tension gradually decreased over the next 10 minutes, although in individual experiments the decreases were invariably abrupt and associated with the restart of contraction. The band-VF group also showed increasing diastolic tension during ischemia (increased from 4±1 to 22±5 mm Hg), which also sharply increased to 53±10 mm Hg at 2 minutes of reperfusion before returning to the control level. The LVDP values at the end of preischemic HMA infusion for sham-HMA and band-HMA were 89±15 and 159±9 mm Hg, respectively, values that were virtually unchanged from the pre-HMA control values reported in Table 1.

View this table: [in this window] [in a new window]   Table 1. LVDP During Control Perfusion (Preceding HMA Administration) and During End Reperfusion for Sham- and Band-C, -VF, and -HMA Subgroups

NMR Measurements
Fig 3A displays ²³Na NMR spectra obtained from a sham-C heart during control and at end ischemia. Note that the Tm(DOTP)^5- shift reagent affords excellent resolution of the Na⁺_i resonance even in the presence of the large extracellular Na⁺ resonance, derived largely from the coronary effluent surrounding the preparation. The frequency of the Na⁺_i and extracellular Na⁺ resonances as well as the area of the extracellular Na⁺ resonance were extraordinarily stable throughout the experiments. The balloon reference signal area was similarly stable, but its frequency exhibited instability and heterogeneity upon the introduction of ischemia. Macroscopic geometry considerations make the bulk magnetic susceptibility shift component of the balloon resonance sensitive to the vascular changes that alter the shift reagent content of the surrounding tissue.³⁵ The figure indicates an increase in the Na⁺_i resonance during ischemia to 144% of control, as was typical for the contracting hearts. In Fig 3B, analogous ²³Na NMR spectra are shown from a band-VF heart. In contrast to the heart depicted in Fig 3A, the spectra indicate a substantial increase in Na⁺_i to 350% of control, revealing that in the fibrillating heart, Na⁺_i homeostasis was significantly altered. Fig 4 shows the interleaved ³¹P NMR spectra from the same two hearts shown in Fig 3 and reveals that the fibrillating band-VF heart also indicated substantially more depletion of PCr, ATP, and accumulation of P_i than the contracting sham-C heart.

View larger version (15K): [in this window] [in a new window]   Figure 3. A, 23Na NMR spectra obtained on a sham heart in which contraction was maintained throughout ischemia (sham-C), shown during control and at end ischemia. The unshifted resonance at 0 ppm is from Na+i. The resonance at 2.5 ppm is from Na+o, shifted downfield with 4 mmol/L Tm(DOTP)5-. At 8 ppm is the left ventricular balloon reference, also shifted with 4 mmol/L Tm(DOTP)5- (larger shift due to absence of Ca2+). B, 23Na NMR spectra obtained from a band-VF heart (fibrillating band) during control and at end ischemia.

View larger version (15K): [in this window] [in a new window]   Figure 4. 31P NMR spectra from the same two hearts shown in Fig 3, obtained during control and at end ischemia. A, Sham-C heart, with the intracellular resonances: Pi, PCR, -, ß-, and -ATP, and the PPA reference from the left ventricular balloon, as indicated. B, Analogous spectra from the band-VF heart.

In Fig 5, mean values of Na⁺_i and pH_i are shown across time for the various experimental groups. Consistent with the spectra in Fig 3, the data indicate that in the contracting groups Na⁺_i increased moderately during ischemia and then became relatively stable. In contrast, Na⁺_i increased to 300% of control in the sham-VF and band-VF groups, a much greater increase compared with their respective sham- and band-C and -HMA groups (P<.005, end ischemia). Although the pH_i changes were not nearly as severe as are observed with zero-flow ischemia models,²³ the sham-VF and band-VF groups also indicated greater decreases in pH_i compared with their respective sham- and band-C and -HMA contracting groups (P<.05 at end ischemia, except for band-C). For the sham-HMA and band-HMA groups, Na⁺_i decreased during the 15-minute preischemic loading period, decreasing to 78±2% and 87±3% of the control value, respectively. Even during early ischemia, Na⁺_i in the HMA groups was maintained at lower levels than in the VF groups. In the sham-HMA group, HMA treatment attenuated Na⁺_i accumulation, even when compared with the sham-C group (P<.05 versus sham-HMA at end ischemia). The divergence of the cationic changes in the VF and contracting groups became particularly marked during latter ischemia, when changes in the VF groups accelerated. No differences were evident when comparing sham groups with analogous band groups; the changes were all remarkably similar. An essentially complete recovery of both Na⁺_i and pH_i during reperfusion was observed for all groups; at end reperfusion (pooled the final 10 minutes), Na⁺_i in the band-VF group did remain at 10% above control levels (P<.05). The HMA groups tended to recover to levels below their respective control levels (P<.05 for sham-HMA).

View larger version (36K): [in this window] [in a new window]   Figure 5. Mean values of Na+i (A) and pHi (B) are shown across time during control, ischemia, and reperfusion for sham-C, -VF, and -HMA groups. Similarly, Na+i (C) and pHi (D) are shown for the analogous band groups, with band-C only displayed at -4-, 20-, 50-, and 110-minute time points for clarity of presentation.

Fig 6 shows mean PCr, ATP, and P_i content plotted across time for sham and band subgroups. There were no significant differences in phosphorus metabolites between the sham or the band subgroups during control conditions. However, the figure indicates lower control levels of PCr in band compared with sham groups. The PCr, ATP, and P_i for combined sham subgroups in the control condition (before HMA) were as follows (µmol/g dry wt): 53.3±1.6, 29.2±1.0, and 1.9±0.5, respectively, and 39.8±2.0 (P<.0001 versus sham), 26.4±1.2 (P=.08 versus sham), and 3.3±0.5 (P<.05 versus sham), respectively, for combined band subgroups. During ischemia, greater PCr and ATP depletion and P_i accumulation occurred in the sham-VF and band-VF groups than in their respective sham- and band-C and -HMA groups, particularly during latter ischemia (P<.05 at end ischemia). Deterioration of cellular energy in the sham- or band-HMA and -C groups was moderate but not necessarily equal: P_i accumulation was greater in the sham-HMA group than in the sham-C group (P<.05 at end ischemia). In all groups, PCr recovered completely during reperfusion. In contrast, ATP remained significantly below control levels at end reperfusion (pooled the final 10 minutes) in all groups and also was lower in VF groups than in analogous contracting groups at that time (sham-VF versus sham-C and sham-HMA, P<.005; band-VF versus band-HMA, P<.05). In the VF groups, P_i remained significantly above control levels at end reperfusion.

View larger version (35K): [in this window] [in a new window]   Figure 6. Mean values of PCr (A), ATP (B), and Pi (C) are shown across time during control, ischemia, and reperfusion for sham-C, -VF, and -HMA groups. Similarly, PCr (D), ATP (E), and Pi (F) are indicated for the analogous band groups, with band-C only displayed at -4-, 20-, 50-, and 110-minute time points for clarity of presentation.

Fig 7 shows mean lactate efflux rates for sham and band subgroups. Lactate efflux increased gradually throughout ischemia to rather substantial levels, near 20 µmol/g dry wt per minute. On reperfusion, lactate efflux increased transiently and then rapidly decreased toward control levels. No significant differences in lactate efflux were detected between any groups during ischemia. However, during early reperfusion, lactate efflux markedly increased for sham- and band-VF groups to levels much higher than were observed in the sham- and band-C groups. There was also evidence for elevated reperfusion lactate efflux in the HMA groups. Calculation of total lactate efflux during the first 15 minutes of reperfusion confirmed that reperfusion lactate efflux was greater in the sham- and band-VF than in the sham- and band-C groups (µmol/g dry wt): sham-VF, 135±25; sham-C, 53±13 (P=.012 versus sham-VF); sham-HMA, 87±13; band-VF, 109±17; band-C, 28±19; and band-HMA, 118±29. Both the combined sham- and band-VF groups (P<.003) and the combined sham- and band-HMA groups (P<.02) demonstrated greater total reperfusion lactate efflux than the combined sham- and band-C groups.

View larger version (22K): [in this window] [in a new window]   Figure 7. A, Mean lactate efflux values are indicated across time during control, ischemia, and reperfusion for sham-C, -VF, and -HMA groups. Insets show detail of early reperfusion period. *P<.05 vs sham-C. B, Mean lactate efflux values for analogous band groups. *P<.05 vs band-HMA.

Fig 8 shows sham-VF and band-VF Na⁺_i and pH_i data obtained during the ischemic period only that are plotted versus the time from VF. Ischemia times vary at the onset of VF, so the observed changes require careful interpretation. However, the plots reveal that rapid Na⁺_i and pH_i changes started to occur at the onset of VF. The abrupt change in the slopes indicates that exactly at that time, net Na⁺ influx dramatically increased and resulted in levels near 300% of control within a few minutes. Panels B and D in Fig 8 indicate that these rapid changes were paralleled by decreases in pH_i, which decreased from values near 7.0 before VF to values near 6.8 after 10 minutes of VF. Table 2 indicates values of the metabolic parameters 3 minutes before and 3 minutes after the onset of VF and reveals that rapid changes were also occurring in energetic parameters. Although only 6 minutes separated these time points, all of the cationic and energetic parameters were found to be significantly different (paired t test).

View larger version (26K): [in this window] [in a new window]   Figure 8. Mean values of Na+i (A) and pHi (B) are shown during ischemia for the sham-VF group plotted vs the time from VF, with zero time at VF initiation. For comparison, data from the sham-C and -HMA groups obtained during latter ischemia are plotted vs ischemia duration, referenced to the average time of VF initiation in the sham-VF group as the zero time point. Similarly, Na+i (C) and pHi (D) are indicated for the band-VF group plotted vs the time from VF. Data from the band-C and -HMA groups are also shown, with zero time at the average time of VF initiation in the band-VF group.

View this table: [in this window] [in a new window]   Table 2. Metabolic Variables, Na+i, pHi, PCr, ATP, and Pi 3 Minutes Before and 3 Minutes After VF

In addition to showing Na⁺_i and pH_i data versus time from VF, Fig 8 also indicates data from the contracting groups (sham- or band-C and -HMA) obtained during the latter ischemic period; the data were referenced to ischemia duration, but with the average time of VF initiation in the analogous VF group taken as the zero time point. The data illustrate the relative stability of Na⁺_i and pH_i during latter ischemia in the contracting groups. Comparison with VF groups clearly indicates that substantial alteration of cation homeostasis occurs after VF and is not due to ischemia alone but is largely due to the presence of sustained ischemia-induced VF. However, the figure also suggests that although not nearly as severe, alteration in cationic homeostasis before the onset of VF in the VF groups was greater than that occurring during ischemia in the contracting groups. The Na⁺_i levels 3 minutes before VF initiation in the sham-VF and band-VF groups (see Table 2) were higher than those in their respective sham- and band-C and -HMA groups at 3 minutes before the average time of VF (P<.001, excluding band-C). The Na⁺_i level for both VF groups was at 160% of control. In contrast, it was 129±3% and 116±3% for sham-C and -HMA groups, respectively, and 126±18% and 120±5% for band-C and -HMA groups, respectively, indicating a Na⁺_i accumulation equal to or less than half of that observed in the VF groups. A similar comparison with pH_i indicated small but measurable differences (P<.05, excluding band-C).

In addition to comparing data referenced to VF onset with that referenced to ischemia duration, a rigorous analysis was done comparing data obtained at the same duration of ischemia. The 20-minute time point was chosen because it is a substantial period of ischemia, the metabolic parameters in the contracting groups were stabilizing, and importantly, none of the hearts were in VF. Although the 20-minute time point long preceded the average time of VF in both VF groups, differences in certain metabolic parameters were apparent between the VF and contracting groups. Notably, although Na⁺_i at 20 minutes did not approach the extreme levels occurring after VF, Na⁺_i was significantly greater at 20 minutes in the sham-VF group (144±4%) than in both the sham-C and sham-HMA groups (128±4% and 117±3%, respectively; P<.005). Similarly, the mean Na⁺_i level in the band-VF group (148±3%) increased significantly more at 20 minutes of ischemia than in the band-C group (122±19%; P<.05) and the band-HMA group (118±4%; P<.005). Consistent with Fig 8, the data indicate that Na⁺_i accumulation in the contracting groups was approximately half of that observed in the VF groups. In the sham-VF group, ATP was significantly lower and the P_i was higher than in the sham-C group. Statistical differences in phosphorus metabolites were not detected between band-VF and band-C groups. Interestingly, there were no significant differences in any phosphorus metabolites detected between the VF and HMA subgroups in both sham and band groups. Consistent with this observation was that PCr and ATP were significantly lower and P_i was higher in the sham-HMA group than in the sham-C group at the 20-minute time point. Hence, the data suggest a relatively specific attenuation of Na⁺_i accumulation during ischemia in band and sham groups with Na⁺-H⁺ exchange inhibition; HMA did not offer any measurable protection from high-energy phosphate depletion.

Data at the 20-minute ischemia time point were analyzed further by discriminant analysis to determine the ability to predict VF from the metabolic variables. This statistical technique simultaneously considers a profile of variables, thus improving predictive power by considering variable interaction and patterns. The analysis models an equation, which calculates the probability of a certain experimental outcome (VF), based on the variables. Table 3 summarizes the results of this analysis. The parameter PR is the percent improvement in classification accuracy over random classification and is lower than (or equal for PR=100%) the absolute classification accuracy; PR of 80% indicates that 90% of the experiments were correctly classified. The F-PROB values indicate the significance of the variable in the prediction, with lower values indicating higher significance (range, 0 to 1). Table 3 summarizes VF prediction analyses using P_i, PCr, ATP, pH_i, and Na⁺_i (at 20 minutes of ischemia) with three different group combinations: sham-VF versus sham-C, sham-VF versus sham-HMA, and band-VF versus band-HMA. An analysis of band-VF versus band-C alone was not possible because of the low experimental number of band-C. Combined band and sham analyses were not performed because they started with inherently different energy patterns at control. The PR values indicated in the table demonstrate excellent VF prediction from the metabolic variables, and the F-PROB parameters indicate that Na⁺_i is clearly the most powerful VF predictor. This is especially true for the analyses involving HMA groups, which indicate 100% predictive accuracy and very low F-PROB values for Na⁺_i. The analyses were repeated after removing Na⁺_i or both Na⁺_i and pH_i from consideration (not shown). Interestingly, for the sham-VF versus sham-C groups, prediction accuracy did not decrease; PR remained at 80%, and P_i became the most significant variable. When VF prediction was tested from individual variables alone, Na⁺_i gave the lowest F-PROB values, but Na⁺_i, P_i, and ATP each still succeeded in predicting VF with a PR of 60% (other variables, 30%). Contrasting results were obtained with sham-VF versus sham-HMA and band-VF versus band-HMA groups. Removal of Na⁺_i from consideration resulted in low prediction accuracy (PR, 30%). Only Na⁺_i could predict VF by itself, with by far the lowest F-PROB values and with PR values of 88% and 62% for sham and band groups, respectively (30% for the other variables). Analyses that tested VF prediction within all three sham groups and within all three band groups were also performed. In these cases, prediction accuracy was still high, and under all conditions, Na⁺_i was the only predictive variable; the inclusion of sham-HMA apparently reduced the prediction significance of the energy variables within the sham subgroups.

View this table: [in this window] [in a new window]   Table 3. F-PROB and PR Values From the Discriminant Analysis Indicating the Predictability of VF From the Metabolic Variables

In addition to using the metabolic parameters at 20 minutes of ischemia, extensive discriminant analyses were performed using the control data. Little predictive accuracy was found at that time, and PR remained 20% in all cases. Additionally, dry weight, the dry weight–to–body weight ratio, and control LVDP were found to be poor predictors of VF within sham or band groups. No combination of these variables with the metabolic variables resulted in increased predictive accuracy.

In comparing the functional and Na⁺_i reperfusion data in the VF groups, a significant correlation (P<.0001) was found between the reperfusion time required to reestablish paced contractions with an LVDP of 10 mm Hg (T_func) and that required for Na⁺_i to decrease below 160% of control (T_Na⁺), the mean level at which VF was initiated (see Table 2) during ischemia. Linear regression was used to fit the following function (in minutes): T_func=0.94 · T_Na⁺+0.44; R²=.62. Also of note is that the mean values for these parameters were essentially identical in both VF groups: mean T_func and mean T_Na⁺ were 9.0±1.0 and 8.8±0.7 minutes, respectively, for the sham-VF group and 7.6±1.3 and 8.0±1.2 minutes, respectively, for the band-VF group. Consistent with this, an examination of the Na⁺_i levels in each heart using the ²³Na NMR spectrum closest to the time of contractile recovery indicated mean values of 148±8% and 159±10% for sham-VF and band-VF groups, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Metabolic Alterations Preceding VF
Table 2 shows that the metabolic states immediately preceding VF in sham- and band-VF groups were strikingly similar, despite the different control energy reserve in those groups. Fig 8 indicates that this pre-VF metabolic state, and in particular Na⁺_i homeostasis, was different from that observed in the analogous contracting groups during latter ischemia. A comparison of all parameters at 20 minutes of ischemia (preceding all VF) revealed that although relatively moderate, the increases in Na⁺_i in both sham- and band-VF groups were consistently greater than those occurring in their respective sham- and band-C and -HMA groups. However, significant differences in pH_i were not detected. Differences in the cellular energy state were detected between sham-VF and sham-C groups but not in the three other comparisons between VF and contracting groups. The observations in the untreated sham subgroups provide evidence that the metabolic changes resulting in ischemia-induced VF could involve both Na⁺_i and cellular energy. However, the only measured metabolic factor that was consistently different between the VF groups and virtually all the contracting groups was that of the Na⁺_i level, suggesting that its involvement may be more direct. Normally, the Na⁺_i level could well be coupled to the cellular energy state,³⁶ but in hearts exposed to HMA, Na⁺_i accumulation was attenuated irrespective of the pattern of energetic deterioration. Exemplifying this, phosphorus metabolite changes in the sham-HMA group at 20 minutes of ischemia were similar to those in the sham-VF group. Nevertheless, Na⁺_i accumulation was attenuated and VF was prevented. In addition, the discriminant analysis indicated that the Na⁺_i level before VF was invariably the most predictive metabolic variable and was in fact the only consistent VF predictor between the VF and contracting groups. The retention of predictive accuracy when Na⁺_i and pH_i were removed from the analysis with the untreated sham groups did suggest that a different pattern of phosphorus metabolites existed in those two subgroups. However, VF prediction was excellent between analogous VF and HMA groups, and it depended solely on Na⁺_i.

Prevention of VF via HMA was not likely to result from effects unrelated to Na⁺-H⁺ inhibition. HMA is a specific inhibitor, with a K_i (0.16 µmol/L) >500 times lower than that of amiloride.³⁷ Unlike with amiloride, experiments using HMA can employ an effective inhibitory concentration that is an order of magnitude lower than the K_i reported for Na⁺-Ca²⁺ exchange.³⁷ The lack of functional effects after 15 minutes of preischemic perfusion with HMA argues against nonspecific effects, especially in regard to direct effects on Ca²⁺_i handling. The lack of an effect on pH_i during ischemia is consistent with a previous report from our laboratory²³ and with other studies.^{25 38 39 40} This observation is likely to be related to the activity of alternative H⁺ extrusion mechanisms such as lactate-H⁺ cotransport and CO₂ efflux.⁴⁰ Consistent with a previous study,²³ Na⁺-H⁺ exchange inhibition succeeded in reducing Na⁺_i levels during low-flow ischemia. This reduction was associated with prevention of VF. Although a cause-and-effect relation between VF initiation and increased Na⁺_i cannot be conclusively determined, such a role would be consistent with the data and with recent reports indicating involvement of Ca²⁺_i levels.^{5 6 13 16} Electrogenic Na⁺-Ca²⁺ exchange is the primary sarcolemmal Ca²⁺ extrusion mechanism, and its activity exhibits a steep dependence on the Na⁺ gradient.²⁸ This is imposed by the 3:1 exchange stoichiometry, which at exchange equilibrium would theoretically magnify Na⁺_i changes to the third power in terms of responding Ca²⁺_i changes. These considerations make it likely that Ca²⁺_i would increase in response to an increase in Na⁺_i. Possible mechanisms by which increased Ca²⁺_i may initiate VF include induction of triggered activity in the form of afterdepolarizations, which often precede VF.^{5 6} Afterdepolarizations may be triggered by a "transient inward current," reportedly linked to excess Ca²⁺_i accumulation.^{8 9 10 13} Ca²⁺_i accumulation can induce oscillations in Ca²⁺_i via spontaneous sarcoplasmic reticulum Ca²⁺ release⁷ and thus potentiate the inward current by activation of electrogenic Na⁺-Ca²⁺ exchange¹¹ or nonspecific cation channels.¹² The reported antiarrhythmic effect of inhibiting sarcoplasmic reticulum function with ryanodine supports such a role for Ca²⁺_i oscillations.^{13 16 41} In support of such a role for Ca²⁺_i overload in ischemia-induced VF, increases in Ca²⁺_i have been well documented in myocardium exposed to zero-flow ischemia or hypoxia.^{20 21 22 25 26 27} Low-flow ischemia is more complex, since with mild coronary flow reductions insufficient to cause cellular energy depletion (flow, >50% of control), systolic Ca²⁺_i reportedly decreases; this could be an important mechanism to downregulate function and energy consumption.⁴² However, as flow reductions become severe and approach zero flow, Ca²⁺_i increases are likely. Using indo 1 and a 10% flow glucose-perfused rat heart model like the one in the present study, Camacho et al¹⁹ reported increases in diastolic and systolic Ca²⁺_i levels in contracting hearts. Also, the reported high-energy phosphate and pH_i changes were very consistent with those in the present study. The 40% increase in diastolic Ca²⁺_i reported during 10% flow¹⁹ is similar to the increase reported to precede spontaneous VF initiated with partial Na⁺-K⁺ ATPase inhibition in the ferret heart¹⁶ ; it is also consistent with changes reported for a rat heart before spontaneous VF.¹⁵ Importantly, these moderate pre-VF elevations of Ca²⁺_i are in a range consistent with the magnitude of the pre-VF Na⁺_i changes observed in the present study, in terms of likely effects from Na⁺-Ca²⁺ coupling and exchange.²⁸

In the VF groups, remarkably few arrhythmias were observed during early ischemia, but they were usually observed before the onset of VF. It was found that the time duration between the point the arrhythmias started and that of VF initiation (mean, 8.1±1.4 minutes) correlated significantly (P<.0001) with the Na⁺_i levels 3 minutes before the onset of VF in individual hearts in the combined sham-VF and band-VF groups. Linear regression was used to fit the following function: arrhythmia duration=26 · Na⁺_VF-34 (R²=.56), where Na⁺_VF is the Na⁺_i level 3 minutes before the onset of VF. This correlation is consistent with the concept that Na⁺_i plays a causal role in the induction of arrhythmias that precede and precipitate VF.^{1 5 6 13 16} Notably, however, Na⁺ influx during ischemia is not restricted to Na⁺-H⁺ exchange, because continued depolarization and contraction ensures a contribution via voltage-activated Na⁺ channels. The magnitude depends on conditions such as the pacing rate, and the presence of ventricular extrasystoles could potentially accelerate Na⁺_i accumulation itself. In this context, it is important to note that the 20-minute ischemia time point not only preceded all VF, but in sham-VF, it also preceded the development of all arrhythmias in most cases. When the one exception was excluded, the remaining (n=9) sham-VF hearts still exhibited higher Na⁺_i at 20 minutes than did sham-C (P=.01) and sham-HMA (P<.0001) hearts. Consistent with the effects observed with Na⁺-H⁺ exchange inhibition, this suggests that the excess Na⁺_i accumulation preceding VF is related to more than simply the presence of extrasystoles and is likely to be more than just a marker for the arrhythmias that lead to VF. A more interactive role for Na⁺_i would be consistent with the data.

Metabolic Alterations After VF
The Na⁺_i versus time from VF data shown in Fig 8 indicate that at a time coinciding with VF initiation, the slopes changed by an order of magnitude. This indicates a dramatic increase in net Na⁺ influx to rates far exceeding those normally observed during zero-flow ischemia.²³ This is consistent with reports that ventricular cells are rapidly excited at the onset of VF, with substantial fast channel (voltage-activated Na⁺ channel) activity occurring during its early stages.^{43 44} The Na⁺_i increases and associated alterations of Na⁺-Ca²⁺ exchange²⁸ are more than adequate to explain the fivefold increases in Ca²⁺_i reported by Koretsune and Marban,¹⁴ which occurred within 5 to 10 minutes of VF initiation. Other studies have also reported extraordinary increases in Ca²⁺_i after VF, occurring on the same time scale as the Na⁺_i changes in the present study,^{15 16} suggesting that Na⁺_i is an important factor to be considered in the acceleration and maintenance of the Ca²⁺_i overload and hence VF itself. The cellular energy and pH_i changes may also be accelerated by the Na⁺_i overload. Increased Na⁺_i increases energy demand by stimulating Na⁺,K⁺-ATPase activity.⁴⁵ Accompanying increases in Ca²⁺_i would also increase energy demand via contractile apparatus activation, consistent with the post-VF increase in diastolic tension observed in the present study. The post-VF pH_i decreases in the present study may also be related to Ca²⁺_i, because increases in Ca²⁺_i reportedly depress pH_i.⁴⁶ Severe Ca²⁺ overload can result in mitochondrial Ca²⁺ accumulation and thus decrease energy production, which is already slowed during low-flow ischemia.⁴⁷ Such an effect could promote cycles of further metabolic deterioration by slowing Na⁺,K⁺-ATPase activity and thus increasing Na⁺_i further. Consistent with this, both increased energy utilization and decreased energy production have been reported during VF initiated via partial Na⁺,K⁺-ATPase inhibition.⁴⁸ Severe Ca²⁺_i overload and cellular energy depletion also occurred.^{16 18 48} With VF initiated with electrical burst pacing, Ca²⁺_i overload is of lesser severity, minimal energy depletion occurs, and in contrast to the other model, spontaneous defibrillation often occurs.^{14 16 48} With either model, effective defibrillation occurs with low Ca²⁺ perfusion.⁴¹ In combination with the present findings, these results are consistent with the hypothesis that VF is maintained, in part, by an inability to regain control of ion homeostasis because of compromised energy production and/or Na⁺,K⁺-ATPase activity.

The general association of Na⁺_i accumulation and VF is consistent with a previous study that investigated hypoxia-induced contractile failure in the perfused rat heart.⁴⁹ However, the previous report implied that most of the increase in Na⁺_i occurred well before the onset of contractile failure/VF. In contrast, the present study indicates that severe Na⁺_i overload occurred only after VF initiation, a pattern that tracks reported Ca²⁺_i changes with spontaneous VF.^{15 16} A large Na⁺_i increase preceding VF might then imply a decoupling of the Na⁺_i and Ca²⁺_i time courses. These apparent differences may be related to the different hypoxic and ischemic VF models but may also be related to the difficulty of direct data comparison. Different ²³Na NMR shift reagent methodology was used, and the previous report did not plot data versus the time from VF. The Na⁺_i increase before the mean time of VF may have been largely attributable to hearts already in VF, as is clearly the case in the present study (see Figs 5 and 8, and mean times of VF initiation). Although in the previous study the parameters curve-fitted to the pre-VF Na⁺_i data indicated a divergent time course, the magnitude of the pre-VF Na⁺_i changes was not indicated.

Functional and Metabolic Alterations During Reperfusion
Functional recovery was delayed for the VF groups, and end reperfusion ATP was also lower compared with the contracting groups, indicating greater metabolic injury. However, a complete reperfusion functional recovery was observed in all groups, which is consistent with other studies using low-flow ischemia in perfused rat hearts supplied with glucose.^{23 27 50} That the restart of contraction during reperfusion coincided closely in time with VF termination was strongly suggested by the functional and electrogram data. Hence, the observed correlation between the reperfusion recovery times for Na⁺_i (T_Na⁺) and function (T_func), as well as the association of VF initiation during ischemia, and termination during reperfusion with a similar Na⁺_i elevation, are consistent with a role for Na⁺_i in VF initiation and maintenance. The data suggest that the recovery of Na⁺_i and Ca²⁺_i levels during reperfusion, enabled by the reactivation of energy production and Na⁺,K⁺-ATPase activity, is likely to be an important factor in the termination of VF during reperfusion. This would be consistent with observations in hearts that defibrillated spontaneously¹⁶ or with lidocaine infusion^{14 15} ; Ca²⁺_i levels were observed to rapidly approach control levels before or at the time of conversion to sinus rhythm.

In the present study, a stimulation in reperfusion lactate efflux was observed in sham- and band-VF compared with sham- and band-C groups. This could be related to the greater ion imbalance in the VF groups, given the reported role for glycolytic energy production in ion homeostasis⁵¹ and reperfusion Ca²⁺_i recovery.⁵² Alternatively, the observations could to some degree be related to the occurrence of poorly perfused regions in the VF hearts, perhaps related to the gradual increases in diastolic tension/left ventricular balloon pressure that occurred after VF. Notably, however, these increases were relatively moderate, and the high ischemic flow rates maintained in this crystalloid-perfused low-flow model should ensure a minimum of vascular washout throughout the tissue. Also, the pH_i-sensitive P_i resonance frequency normally indicates the presence of regional infarcts or perfusion heterogeneity by indicating multiple shift (pH_i) components⁵³ ; this was not observed in the experiments at any time, and all hearts exhibited an excellent functional recovery. This suggests that tissue perfusion and the various metabolic parameters are not likely to vary to any great extent across the preparation.

Increased Susceptibility of Hypertrophied Hearts to VF
The increased susceptibility of the band hearts to ischemia-induced VF is consistent with previous reports,^{2 3} but the present study is the first to document increased VF in hypertrophied myocardium in a globally ischemic model that normalized vascular differences by using the same ischemic flow per gram heart weight. Consistent with other studies in normal and hypertrophied rat heart models,^{2 3} heart weight parameters were not consistently associated with VF incidence. Band hearts were large, but neither the dry weight nor dry weight–to–body weight ratio differed significantly between subgroups of sham or band or were predictive of VF within sham or band. Band hearts did have a lower high-energy phosphate reserve compared with sham hearts, the origin of which is potentially complex. In the context of the similarity of the ischemic lactate production and identical flow/oxygen delivery per gram tissue, the altered energy reserve could increase the probability of band hearts reaching a critical energy-depleted metabolic state during ischemia, sufficient to permit excess cation accumulation. However, control levels of phosphorus metabolites were similar between subgroups of sham or band and were not predictive of VF within sham or band. In some studies, reduced Na⁺,K⁺-ATPase activity in models of myocardial hypertrophy has been reported,^{54 55} although not in the aortic-banded models.⁵⁶ The present study contains no evidence for a difference in inherent Na⁺ efflux capacity in band hearts. Also, the present study cannot address the possibility that altered electrophysiological parameters could play a role in increased VF susceptibility in band hearts.⁴

In summary, the present study demonstrated that initiation of ischemia-induced VF was consistently preceded by a moderate degree of excess Na⁺_i accumulation, the prevention of which also prevented VF. This deterioration in Na⁺_i homeostasis may be coupled with the cellular energy state under normal conditions. However, increased cellular energy depletion preceding VF was not consistently observed. The present study is consistent with a causal role for excess Na⁺_i accumulation in VF initiation. Furthermore, the severe Na⁺_i overload that progressed rapidly after VF initiation strongly argues an important role for Na⁺_i in VF maintenance and the associated deterioration of cellular energy. The association between Na⁺_i and VF was further strengthened by a correlation between the reperfusion Na⁺_i and contractile recoveries. Given the likely coupling of Na⁺_i and Ca²⁺_i homeostasis and the previous reports linking Ca²⁺_i and VF, the present study provides new and important information regarding the metabolic deterioration leading to and sustaining ischemia-induced VF.

	Selected Abbreviations and Acronyms

 

HMA = 5-(N,N-hexamethylene)-amiloride LVDP = left ventricular developed pressure NMR = nuclear magnetic resonance PCr = phosphocreatine Pi = inorganic phosphate PPA = phenylphosphonic acid Tm(DOTP)5- = thulium 1,4,7,10- tetraazacyclododecane-N,N',N'',N'''- tetramethylenephosphonate VF = ventricular fibrillation

	Acknowledgments

 
Dr Pike is an Established Investigator of the American Heart Association. We thank Dr Michael Hardin and Brett Thorn for valuable advice concerning the statistical analysis, Rebecca Lynn for secretarial assistance, and Marla Scogin for assistance with graphics.

Received February 17, 1995; accepted April 21, 1995.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Brandenburg RO. Cardiomyopathies and their role in sudden death. J Am Coll Cardiol. 1985;5:185B-189B.
2. Kohya T, Kimura S, Myerburg RJ, Basset AL. Susceptibility of hypertrophied rat hearts to ventricular fibrillation during acute ischemia. J Mol Cell Cardiol. 1988;20:159-168. [Medline] [Order article via Infotrieve]
3. Robertson E, Hof RP, Zierhut W. Effect of hypertrophy induced by pressure overload or volume overload on reperfusion induced arrhythmias in anaesthetized rats. Cardiovasc Res. 1993;27:515-519. [Abstract/Free Full Text]
4. Kowey PR, Friehling TD, Sewter J, Wu Y, Sokil A, Paul J, Nocella J. Electrophysiological effects of left ventricular hypertrophy: effect of calcium and potassium channel blockade. Circulation. 1991;83:2067-2075. [Abstract/Free Full Text]
5. Wit AL, Rosen MR. Afterdepolarization and triggered activity. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System. New York, NY: Raven Press Publishers; 1986:1449-1490.
6. Clusin WT, Buchbinder M, Harrison DC. Calcium overload, `injury current,' and early ischaemic cardiac arrhythmias: a direct connection. Lancet. 1983;1:272-273. [Medline] [Order article via Infotrieve]
7. Orchard CH, Eisner DA, Allen DG. Oscillations of intracellular Ca²⁺ in mammalian cardiac muscle. Nature. 1983;304:735-738. [Medline] [Order article via Infotrieve]
8. Lederer WJ, Tsien RW. Transient inward current underlying arrhythmogenic effects of cardiotonic steroids in Purkinje fibers. J Physiol (Lond). 1976;263:73-100. [Abstract/Free Full Text]
9. Kass RS, Lederer WJ, Tsien RW, Weingart R. Role of calcium ions in transient inward currents and aftercontractions induced by strophanthidin in cardiac purkinje fibers. J Physiol (Lond). 1978;281:187-208. [Abstract/Free Full Text]
10. Kass RS, Tsien RW, Weingart R. Ionic basis of transient inward current induced by strophanthidin in cardiac Purkinje fibers. J Physiol (Lond). 1978;281:209-226. [Abstract/Free Full Text]
11. Fedida D, Noble D, Rankin AC, Spindler AJ. The arrhythmogenic transient inward current i_T1, and related contraction in isolated guinea-pig ventricular myocytes. J Physiol (Lond). 1987;392:523-542. [Abstract/Free Full Text]
12. Colquhoun D, Neher E, Reuter H, Stevens CF. Inward current channels activated by intracellular Ca in cultured cells. Nature. 1981;294:752-754. [Medline] [Order article via Infotrieve]
13. Thandroyen FT, Morris AC, Hagler HK, Ziman B, Pai L, Willerson JT, Buja LM. Intracellular calcium transients and arrhythmia in isolated heart cells. Circ Res. 1991;69:810-819. [Abstract/Free Full Text]
14. Koretsune Y, Marban E. Cell calcium in the pathophysiology of ventricular fibrillation and in the pathogenesis of postarrhythmic contractile dysfunction. Circulation. 1989;80:369-379. [Abstract/Free Full Text]
15. Stefenelli T, Wikman-Coffelt J, Shao T, Wu ST, Parmley WW. Intracellular calcium during pacing-induced ventricular fibrillation. J Electrocardiol. 1992;25:221-228. [Medline] [Order article via Infotrieve]
16. Kihara Y, Morgan JP. Intracellular calcium and ventricular fibrillation: studies in the aequorin-loaded isovolumic ferret heart. Circ Res. 1991;68:1378-1389. [Abstract/Free Full Text]
17. Grover FL, Fewel JG, Ghidoni JJ, Norton JB, Arom KV, Trinkle JK. Effects of ventricular fibrillation on coronary blood flow and myocardial metabolism. J Thorac Cardiovasc Surg. 1977;73:616-624. [Abstract]
18. Kusuoka H, Jacobus WE, Marban E. Calcium oscillations in digitalis-induced ventricular fibrillation: pathogenic role and metabolic consequences in isolated ferret hearts. Circ Res. 1988;62:609-619. [Abstract/Free Full Text]
19. Camacho SA, Figueredo VM, Brandes R, Weiner MW. Ca²⁺ dependent fluorescence transients and phosphate metabolism during low-flow ischemia in rat hearts. Am J Physiol. 1993;265:H114-H122. [Abstract/Free Full Text]
20. Marban E, Kitakaze M, Koretsune Y, Yue DT, Chacko VP, Pike MM. Quantification of Ca²⁺_i in perfused hearts: critical evaluation of the 5F-BAPTA and nuclear magnetic resonance method as applied to the study of ischemia and reperfusion. Circ Res. 1990;66:1255-1267. [Abstract/Free Full Text]
21. Kihara Y, Grossman W, Morgan JP. Direct measurement of changes in intracellular calcium transients during hypoxia, ischemia, and reperfusion of the intact mammalian heart. Circ Res. 1989;65(suppl 4):1029-1044.
22. Lee H, Mohabir R, Smith N, Franz M, Clusin W. Effect of ischemia on calcium-dependent fluorescence transients in rabbit hearts containing indo 1 correlation with monophasic action potentials and contraction. Circulation. 1988;78:1047-1059. [Abstract/Free Full Text]
23. Pike MM, Luo CS, Clark MD, Kirk KA, Kitakaze M, Madden MC, Cragoe EJ Jr, Pohost GM. NMR measurements of Na⁺ and cellular energy in the ischemic rat heart: role of Na⁺/H⁺ exchange. Am J Physiol. 1993;265:H2017-H2026.