The intermediary metabolite pyruvate attenuates stunning and reduces infarct size in in vivo porcine myocardium

Gentian Kristo, Yukihiro Yoshimura, Jianli Niu, Byron J. Keith, Robert M. Mentzer Jr., Rolf Bünger, Robert D. Lasley


The intermediary metabolite pyruvate has been shown to exert significant beneficial effects in in vitro models of myocardial oxidative stress and ischemia-reperfusion injury. However, there have been few reports of the ability of pyruvate to attenuate myocardial stunning or reduce infarct size in vivo. This study tested whether supraphysiological levels of pyruvate protect against reversible and irreversible in vivo myocardial ischemia-reperfusion injury. Anesthetized, open-chest pigs (n = 7/group) underwent 15 min of left anterior descending coronary artery (LAD) occlusion and 3 h of reperfusion to induce stunning. Load-insensitive contractility measurements of regional preload recruitable stroke work (PRSW) and PRSW area (PRSWA) were generated. Vehicle or pyruvate (100 mg/kg iv bolus + 10 mg·kg–1·min–1 intra-atrial infusion) was administered during ischemia and for the first hour of reperfusion. In infarct studies, pigs (n = 6/group) underwent 1 h of LAD ischemia and 3 h of reperfusion. Group I pigs received vehicle or pyruvate for 30 min before and throughout ischemia. In group II, the infusion was extended through 1 h of reperfusion. In the stunning protocol, pyruvate significantly improved the recovery of PRSWA at 1 h (50 ± 4% vs. 23 ± 3% in controls) and 3 h (69 ± 5% vs. 39 ± 3% in controls) reperfusion. Control pigs exhibited infarct sizes of 66 ± 1% of the area at risk. The pyruvate I protocol was associated with an infarct size of 49 ± 3% (P < 0.05), whereas the pyruvate II protocol was associated with an infarct size of 30 ± 2% (P < 0.05 vs. control and pyruvate I). These findings suggest that pyruvate attenuates stunning and decreases myocardial infarction in vivo in part by reduction of reperfusion injury. Metabolic interventions such as pyruvate should be considered when designing the optimal therapeutic strategies for limiting myocardial ischemia-reperfusion injury.

  • energetics
  • myocardial ischemia
  • reperfusion

pyruvate is a key glycolytic intermediate and natural antioxidant that can exert beneficial effects on both cytosolic and mitochondrial energy and redox metabolism. Supraphysiological concentrations (≥5 mM) of pyruvate have been shown to increase the cellular energy state, lower the cytoplasmic NADH/NAD+ ratio, and enhance the glutathione antioxidant system in numerous cell types, including in vitro and in vivo cardiac preparations (5, 6, 2427, 35, 43). These properties, as well as the ability of pyruvate to decrease reactive oxygen species (ROS) and improve cardiomyocyte Ca2+ homeostasis, likely contribute to the beneficial effects of pyruvate in hypoxic and ischemic-reperfused myocardium (1, 38, 10, 15, 19, 21, 28, 29, 41, 43, 45, 47, 48). It has also been reported (13, 14, 16) that pyruvate exerts a positive inotropic effect in papillary muscle from failing human myocardium.

In addition to its energetic and antioxidant effects, pyruvate has also been shown to reduce cell death (both necrosis and apoptosis) in various tissues. Pyruvate decreases cell death in various neural cell lines, in the bovine aortic endothelium, lymphoid cells, renal cells, and murine thymocytes (2, 17, 18, 23, 31, 35, 38, 40, 42). In isolated perfused hearts there is evidence that increased pyruvate levels reduce the opening and promote the closure of mitochondrial permeability transition pores, an event believed to play an important role in apoptosis (19).

Despite numerous in vitro studies of the beneficial effects of pyruvate in mammalian myocardium, there have been few reports documenting its in vivo efficacy. In a porcine hemorrhage-resuscitation model, intravenous pyruvate proved to be an antioxidant and alkalizer, which enhanced survival, markedly prolonged electrical brain activity, improved cerebral and cardiac energy metabolism, reduced hepatic apoptosis, and improved overall cardiovascular stability (32, 33). In patients with congestive heart failure, intracoronary pyruvate infusions have been shown to improve parameters of global cardiac performance and reduce pulmonary capillary wedge pressure and heart rate while increasing cardiac output (14). We (30, 48) have previously reported that brief intracoronary infusions of pyruvate elicited positive inotropic effects in intact normal and stunned porcine and canine myocardium. However, in these studies, only transient metabolic and inotropic effects of pyruvate were tested. No studies have addressed whether pyruvate actually attenuates in vivo myocardial stunning.

In addition, there appear to be only two studies examining the potential beneficial effects of pyruvate in irreversibly injured myocardium. Regitz et al. (39) reported that intracoronary pyruvate infusion for 10 min before occlusion and during 90 min of reperfusion significantly reduced infarct size in dogs. Unfortunately, the animals were reperfused for only 90 min and coronary collateral blood flow was not measured. In contrast, Gutterman et al. (12) reported that intracoronary pyruvate infusion initiated 15 min before a prolonged 3-h occlusion, or at the onset of 90 min of reperfusion, did not reduce infarct size in pigs. Thus the findings regarding the infarct sparing effect of pyruvate are controversial.

The purpose of the present study was to determine whether exogenous pyruvate infusion could attenuate myocardial stunning and decrease infarct size in an in vivo porcine model of regional myocardial ischemia. Additional studies in rat isolated ventricular myocytes were conducted to determine whether pyruvate loading attenuates oxidative stress.


All animals in this study received humane care according to the guidelines set forth in The Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 86-23, Revised 1996). In addition, animals were used in accordance with the guidelines of the University of Kentucky Institutional Animal Care and Use Protocol.

Animal preparation. Domestic pigs of either sex weighing 20–30 kg were used in this study. Anesthesia was induced with ketamine (20 mg/kg im), followed by pentobarbital sodium (15–20 mg/kg iv) through an ear vein. Anesthesia was then maintained with additional pentobarbital sodium (1.5–2 mg/kg iv) every 15 min. A tracheotomy was performed, and the animals were mechanically ventilated with a mixture of room air and 100% O2. Tidal volume, respiratory rate, and percent O2 in the inspired air were adjusted to maintain normal arterial blood gas and pH values. The right femoral artery was cannulated for the withdrawal of arterial blood samples and to monitor arterial blood pressure. The femoral vein was cannulated for the infusion of fluids and for the maintenance of anesthesia. Normal saline solution was administered through a peripheral vein at 7–10 ml·kg–1·h–1 after an initial bolus of 400–500 ml. Core body temperature was monitored with an esophageal temperature probe and maintained with a heating pad between 37° and 37.5°C.

The heart was exposed by median sternotomy and suspended in a pericardial cradle. Left ventricular (LV) pressure was measured with a 5-Fr high-fidelity pressure-sensitive tip transducer (Millar Instruments; Houston, TX) placed in the LV cavity through the apex and secured with a purse-string suture. A segment of the left anterior descending coronary artery (LAD) distal to the origin of the first diagonal branch was dissected free of surrounding tissue. The area at risk (AAR) was delineated by a brief (<20 s) occlusion of the LAD with a small vascular occluding clamp. A transit time perivascular flow probe (Transonic Systems; Ithaca, NY) was placed around this segment to measure coronary blood flow. The preparation was allowed to stabilize for 30 min after all instrumentation was complete before the experimental protocols were initiated.

Experimental protocols. The porcine regional myocardial ischemia models of stunning and infarction used in this study have been previously described in detail (22). In the stunning protocol, pigs (n = 7 per group) were subjected to 15 min of LAD occlusion, followed by 3-h reperfusion. Vehicle (normal saline) or Na-pyruvate (100 mg/kg bolus iv + 10 mg·kg–1·min–1 intra-atrial infusion) was infused during the 15-min ischemic period and for the first hour of reperfusion. We (33) showed that this administration regimen raised plasma pyruvate levels from ∼0.1 to 3–6 mM, while the plasma lactate-to-pyruvate ratio decreased from ∼10–15 to near 1. Sodium pyruvate was made as a 30% stock solution (pH adjusted to 7.4 with NaOH). Vehicle-treated animals received the same total volume of saline as pyruvate-treated pigs.

Myocardial infarction was induced with a 60-min LAD occlusion, followed by 3-h reperfusion. The following groups were studied: 1) saline controls (n = 6), 2) pyruvate I (n = 6), and 3) pyruvate II (n = 6), in which pigs were administered pyruvate (100 mg/kg bolus iv + 10 mg·kg–1·min–1 intra-atrial infusion). In pyruvate I, the pyruvate treatment was initiated 30 min before and throughout the 60-min LAD occlusion. In the pyruvate II group, the infusion of pyruvate was extended throughout the first 60 min of reperfusion. Lidocaine (2 mg/kg iv bolus) was administered immediately before occlusion in the infarct studies and just before reperfusion in the stunning studies. After 3-h reperfusion, the ischemic AAR was determined by LAD reocclusion and infusion of a 0.5% Evans blue solution into the left ventricle while the aorta was occluded. The AAR was devoid of the Evans blue stain. While under deep anesthesia, the animals were euthanized with an overdose of pentobarbital sodium and saturated KCl solution and the hearts were excised. In the stunning experiments, crystal placement in the ischemic and nonischemic beds was verified after excision of the heart.

Regional myocardial function. In the stunning experiments, pairs of piezoelectric segment-shortening crystals (Crystal Biotech; Houston, TX) were placed in the LAD and left circumflex coronary artery perfused beds to measure segment shortening (SS) by sonomicrometry. Crystals were placed in the midmyocardium (∼4–6 mm deep) 5 to 15 mm apart and aligned in a manner such that the intercrystal axis was parallel to the direction of myocardial fiber shortening.

All hemodynamic and sonomicrometry signals were fed through a 32-bit analog-to-digital converter into an online data-acquisition computer with customized software (Augury; Coyote Bay Instruments, Manchester, NH). End diastole was defined as the onset of pressure increase over time (+dP/dt), and end systole was defined as 20 ms before peak pressure decrease over time (–dP/dt). SS was defined as end-diastolic length (EDL)–end-systolic length (ESL), and percent SS (%SS) was calculated as (EDL–ESL/EDL) × 100%. All hemodynamic data were continuously displayed on a computer monitor. Stroke work (SW) was calculated by quantifying the area within the pressure-segment-length loops generated during each cardiac cycle. Measurements of load-insensitive parameters of regional cardiac contractility, preload-recruitable SW (PRSW) and PRSW area (PRSWA), were generated from the segment length, and LV pressure data collected during brief (7 s) vena cava occlusions. The inferior vena cava was occluded by gradual tightening of a snare formed of umbilical tape around its supradiaphragmatic portion. During data acquisition, ventilation was held at end expiration to avoid the effects of varying venous return on preload. PRSW and PRSWA were calculated according to the methods of Glower and colleagues (11). PRSW was based on linear regression of the relationship between SW and the end-diastolic segment length calculated by the equation SW = MSW (EDL–LW), were MSW is the slope of PRSW and LW is the x-axis intercept. PRSWA was determined by the formula PRSWA = MSW/2 (1.2 LWmaxLW)2, where LWmax was the maximum x-axis intercept during the entire experiment. Baseline and caval occlusion data were saved at specific time points in the protocol for off-line analysis. An average of 9–11 beats was used in each calculation.

AAR and infarct size measurement. The isolated left ventricles were cut into four slices of equal thickness in a plane parallel to the atrioventricular groove. Each slice was compressed between two transparent Plexiglas acrylic plates (Rohm and Hass; Philadelphia, PA) separated by a distance of 8 mm to achieve uniform thickness. The cross-sectional surface and ischemic areas of each slice were traced onto a transparency sheet. The slices were then incubated in a 1% triphenyl tetrazolium chloride solution (Sigma) in phosphate-buffered saline solution at 37°C for 15 min. The presence of a brick-red stain indicated viable tissue, whereas nonviable tissue (infarcted) remained pale. If any infarct was present, the tissue slices were again compressed between the Plexiglas acrylic plates and retraced. The areas were quantified with a digitizer (model 1200 III, parallel postscanner at 200 dpi; Mustek; Irvine, CA) and graphic analysis software (Sigma Scan Pro Automated Image Analysis Software; Jandel Scientific, SPSS; San Rafael, CA). The percent AAR was calculated for each slice by dividing the AAR by the total slice area. The sum of the AAR of all slices was divided by the sum of the areas of all slices to obtain the percentage of the LV that was ischemic. The infarct size was expressed as the percentage of the AAR and AAR as the percentage of LV.

Isolated ventricular myocyte studies. Ventricular myocytes were isolated as described by this laboratory with minor modifications (20, 28). Male Sprague-Dawley rats (250–300 g) were heparinized and anesthetized with pentobarbital sodium (50 mg/kg). The heart was then rapidly excised and perfused with a HEPES medium containing (in mM) 118 NaCl, 4.7 KCl, 1.2 KH2PO4, 1.2 MgSO4, 10 HEPES, 11 glucose, 5 creatine, 2 carnitine, and 5 taurine (pH 7.2, 37°C). Hearts were perfused with this calcium-free solution for 5 min in a nonrecirculating mode. Perfusion was then switched to a recirculating mode, and 0.14 mg/ml collagenase (Liberase, Roche Biochemicals; Indianapolis, IN) was added into the medium. Calcium was then incrementally introduced to a final concentration of 200 μM. The heart was removed from the cannula, minced, and the myocytes were mechanically dispersed. Dissociated myocytes were stored in the above HEPES buffer (pH 7.40, 1 mM CaCl2) at room temperature for at least 1 h before use. Three myocyte isolations were performed with HEPES medium supplemented with 10 mM pyruvate. After isolation, these myocytes were stored in pyruvate-supplemented HEPES for at least 1 h before use.

Oxidative stress. Intracellular oxidative stress was detected using the fluorescent dye dichlorofluorescein (DCF) diacetate (DCFDA), as previously described by this laboratory (34). Myocytes were loaded in the dark with DCFDA (5 μM, Molecular Probes; Eugene, OR) for 15 min at room temperature. The myocytes were washed with HEPES medium to remove extracellular dye and incubated for an additional 30 min to allow for intracellular deacetylation of DCFDA, which results in the formation of the reduced and weakly fluorescent DCF moeity. On oxidation by ROS and reactive nitrogen species reduced dichlorofluorescin is converted to the highly fluorescent DCF (37). An aliquot of dye-loaded myocytes was then allowed to settle on laminincoated coverslips placed in a 300-μl, temperature-controlled recording chamber (RC-24 chamber, TC-324B temperature controller, Warner Instrument; Hamden, CT) on the stage of an inverted microscope (model IX-70; Olympus America; Melville, NY). Cells were suffused with normal HEPES buffer (pH 7.4 at 37°C) at a flow rate of 1 ml/min. Two baseline DCF fluorescence values were recorded after 5-min and 20-min suffusion before exposing the cells to H2O2 (150 μM), which induced oxidative stress that was maintained for 10 min. All cell fluorescence values were background corrected, and increases in DCF during H2O2 were expressed relative to each cell's baseline fluorescence.

An excitation wavelength of 490 nm was used, and emission fluorescence was collected at 520 ± 5 nm. To reduce excitation-dependent oxidation of the dye, a neutral density filter (ND12) was placed in the path of the excitation source (75-W xenon arc lamp) and sampling times were limited to 160 ms. Epifluorescence was collected with the use of a charge-coupled device camera attached to a side port of the microscope, and fluorescence intensity analysis was made on a Pentium-based computer using custom-designed software (Coyote Bay Instruments).

Statistical analysis. Results are expressed as means ± SE. In the infarct studies, differences between groups were determined by one-way ANOVA, followed by Tukey's correction. Effects of pyruvate on blood chemistries were analyzed by one-way ANOVA with repeated measures. The stunning experimental results were analyzed by two-way ANOVA for treatment and time. A P value <0.05 was considered statistically significant. The myocyte DCF results were analyzed by two-way ANOVA for treatment and time.


Stunning protocol. The systemic hemodynamic and global ventricular function parameters are presented in Table 1. There were no significant differences in heart rate, mean arterial blood pressure, LV +dP/dt, or LAD coronary blood flow between the groups at baseline, during ischemia, or at any time during the course of the reperfusion. Table 2 shows the blood gas and plasma sodium levels in the two groups. Although pyruvate infusion was associated with a tendency toward higher pH and increased Na+ and Math levels, the differences between the groups were not statistically significant at any time during the experimental protocol. There were also no differences in K+ and Ca2+ concentrations, hematocrit, or Po2 levels (data not shown).

View this table:
Table 1.

Systemic hemodynamics and global ventricular function in stunning protocol

View this table:
Table 2.

Arterial blood pH, Pco2 Na+, and Formula in stunning protocol

The ischemic AAR were similar in both groups (24 ± 2% of the left ventricle in the control group and 21 ± 1% in the pyruvate-treated group). As expected, there was no evidence of infarction (based on triphenyl tetrazolium chloride-positive staining) in any of the pigs.

Figure 1 illustrates the effects of pyruvate treatment on LAD %SS. There was no difference in preischemic %SS values between the two groups. The recovery of SS at 1-h reperfusion in the control group was only 19 ± 2%, reflecting marked depression of myocardial segment work, indicating severe myocardial stunning. In the pyruvate group, recovery of segment shortening after 1-h reperfusion was considerably improved; 50 ± 6% of preischemic values (P < 0.05 vs. run). At 2 and 3 h of reperfusion, i.e., 1 to 2 h after termination of pyruvate infusion, LAD %SS in the pyruvate group remained somewhat elevated relative to the time controls, but this effect was not significant.

Fig. 1.

The recovery of segment shortening (SS) during reperfusion (RP) after 15 min of left anterior descending coronary artery (LAD) occlusion. Results are expressed as the percentage of preischemic values. Values are means ± SE. Pyruvate was infused through the first 60 min of reperfusion.

Figure 2 summarizes the effects of pyruvate treatment on LAD PRSW, an index of ventricular contractile state. The preischemic PRSW values were similar in both groups. The percent recoveries of PRSW in the control group at 1- and 3-h reperfusion were 54 ± 3%, and 78 ± 7%, respectively, reflecting decreased regional ventricular contractility. In the pyruvate-treated group PRSW recovery was greatly accelerated and complete already at 1 h of reperfusion (102 ± 5%) and continued to rise above controls after 3 h of reperfusion (130 ± 12%).

Fig. 2.

RP recovery of LAD bed preload-recruitable stroke work (PRSW). Results are expressed as the percentage of preischemic values. Values are means ± SE.

Markedly improved recovery of contractile work due to pyruvate was also apparent when the PRSWA was analyzed. The effects of pyruvate treatment on LAD PRSWA are shown in Fig. 3. Preischemic PRSWA values in the control and pyruvate group were similar. The recovery of PRSWA in the control group at 1-, 2-, and 3-h reperfusion was severely depressed, 23 ± 4%, 30 ± 3%, and 39 ± 3%, respectively, indicating poor segmental work performance consistent with severe myocardial stunning. On the other hand, in the pyruvate-treated group recovery of PRSWA at 1-, 2-, and 3-h reperfusion was markedly, albeit not completely enhanced, 53 ± 5%, 57 ± 6%, and 73 ± 6%, respectively. These differences in segment work performance between the two groups were statistically significant throughout reperfusion.

Fig. 3.

The effect of 15-min occlusion and 3-h RP on LAD bed PRSW area (PRSWA). Results are expressed as the percentage of preischemic values. Values are means ± SE.

As shown in Fig. 4, regional function in the nonischemic, left circumflex coronary beds in both groups remained stable throughout the LAD stunning protocol. In two additional pigs, sodium l-lactate was administered in exactly the same manner as pyruvate to determine its effect in stunned myocardium. After 1 and 3 h of reperfusion, the recovery of PRSWA in one animal was 18% and 26%, respectively, and in the second pig, recovery was 40% and 45%, respectively. These observations indicate that l-lactate exerted no measurable functional benefit in stunned myocardium.

Fig. 4.

PRSW and PRSWA in the nonischemic left circumflex (LCX) bed before LAD occlusion (baseline) and after 3-h RP.

Infarction protocol. As shown in Table 3, there were no significant differences in any hemodynamic and global ventricular function parameters between the groups throughout the protocol. After 3 h of reperfusion, MAP and LV +dP/dt tended to be lower than preischemia values, but there were no statistically significant differences between the groups. Arterial acid base and electrolyte data are presented in Table 4. Pyruvate administration was not associated with any statistically significant increases in pH, Pco2, Na+, or Math levels. There were also no differences in K+ and Ca2+ concentrations, Po2, or hematocrit (data not shown).

View this table:
Table 3.

Systemic hemodynamics and global ventricular function in infarction protocol

View this table:
Table 4.

Arterial blood pH, Pco2, Na+, and Formula in infarction protocol

Figure 5 illustrates the infarct size and AAR determined after 3 h of reperfusion. In control pigs, 22 ± 2% of the left ventricle was ischemic, and pyruvate I and pyruvate II groups had ischemic zones of 20 ± 2% and 20 ± 1%, respectively. The infarct size in the control group was 66 ± 1% of the region at risk. When pyruvate was administered before and throughout the 60-min LAD occlusion, infarct size was reduced to 49 ± 3%. When the pyruvate treatment was extended through the first 60 min of reperfusion, infarct size was further decreased to 29 ± 2% of the region at risk. Two pigs treated with sodium lactate through 1 h of reperfusion exhibited infarct sizes of 84% and 69% of the AAR.

Fig. 5.

Left ventricular (LV) infarct size expressed as percentage of the area at risk (AAR). Hearts were submitted to 60 min LAD occlusion and 3-h reperfusion. Infarct size was determined by triphenyltetrazolium chloride (TTC) staining.

Myocyte oxidative stress. The beneficial effects of prolonged pyruvate exposure in isolated rat cardiomyocytes exposed to oxidative stress are shown in Fig. 6. Myocytes stored with standard glucose containing HEPES medium exhibited a 5.0 ± 0.7-fold increase in DCF fluorescence (relative to baseline values) after 5-min exposure to H2O2. Pyruvate-treated myocytes exhibited a 50% (2.3 ± 0.1-fold increase) reduction in DCF fluorescence at this time point, although this difference was not statistically significant via two-way ANOVA. After 10 min of H2O2 exposure, the increase in DCF fluorescence in pyruvate-treated myocytes (6.8 ± 0.8-fold increase) was greatly blunted compared with myocytes isolated and stored in the presence of glucose (19.5 ± 1.7-fold increase).

Fig. 6.

The effect of pyruvate loading on H2O2-induced oxidative stress in adult rat ventricular myocytes. Oxidative stress was assessed by dichlorofluorescein (DCF) fluorescence. *P < 0.05 vs. glucose control group.


The results of the present study indicate that increasing plasma pyruvate levels into the millimolar range attenuates myocardial stunning and limits myocardial infarction in vivo. In the stunning protocol, pyruvate treatment significantly improved regional PRSW and PSRWA throughout reperfusion indicating that both LV contractile work and contractile state were enhanced when the stunned hearts received therapeutic doses of pyruvate during reperfusion. In the infarction protocol, pyruvate infusion exerted a time-dependent reduction of infarct size, being most effective when it was administered during reperfusion. Isolated rat ventricular myocytes incubated with pyruvate for ≥60 min exhibited significantly reduced oxidative stress during H2O2 exposure. These findings indicate that supraphysiological circulating pyruvate levels exert significant protection against both reversible and irreversible myocardial ischemia-reperfusion injury possibly by decreasing reperfusion cardiomyocyte oxidative stress and improving myocardial energetics.

The intermediary metabolite pyruvate has been shown to exert numerous beneficial effects in in vitro cardiac models of oxidative stress and ischemia-reperfusion injury (3, 58, 10, 19, 21, 26, 27, 30, 41, 4345, 47, 48). Observations of enhanced cardiac function after hypoxia, ischemia, and oxidative stress have been associated with improved cellular energetics, phosphorylation potential, and cytosolic redox state, and reduced adenosine nucleoside loss or production (3, 5, 6, 26, 4345, 48). It has also been reported that elevated pyruvate levels improve myocardial intracellular calcium handling in several species, including a report in human cardiac muscle (1, 13, 24, 28). Despite these numerous reports, there have been very few studies assessing the potential therapeutic benefits of pyruvate in in vivo myocardium. We (30, 48) have previously reported that brief intracoronary infusions of pyruvate in intact porcine and canine models increased myocardial phosphorylation potential index and increased regional ventricular function in both normal and stunned myocardium. Similarly, Hasenfuss at al. (14) reported that intracoronary pyruvate infusions increased cardiac index and decreased pulmonary capillary wedge pressures and heart rate in heart failure patients. In these studies, however, the effects of pyruvate were only transient and cardiac function returned to prepyruvate levels within 15–20 min after termination of treatment.

Although there is significant evidence of the ability of pyruvate to decrease apoptosis and necrosis in various tissues (2, 17, 18, 23, 31, 35, 38, 40, 42), there appear to be only two studies addressing the effects of pyruvate in irreversibly injured myocardium, and the findings were contradictory. Regitz et al. (39) reported that intracoronary pyruvate infusion significantly reduced infarct size in the dog after 90 min of ischemia. The pyruvate infusion was started before occlusion and continued during reperfusion. However, the animals were reperfused for only 90 min and coronary collateral blood flow was not measured. In contrast, Gutterman et al. (12) reported that intracoronary pyruvate infusion administered either before occlusion and during reperfusion or just during reperfusion failed to limit infarct size after a 3-h coronary occlusion.

The results of the present study indicate that systemic pyruvate administration during the first hour of reperfusion significantly improved regional segment shortening and the load-insensitive measurements of regional cardiac contractility, PRSW and PRSWA, in a porcine model of regional myocardial stunning. Although pyruvate was discontinued after the initial hour of reperfusion, the improved recovery of regional PRSW and PRSWA persisted throughout reperfusion. In the infarct protocol, pyruvate infusion before ischemia and during the first hour of reperfusion was associated with a 50% reduction in myocardial infarct size.

One of the limitations of using pyruvate in in vivo preparations is the large amount of pyruvate needed to increase plasma levels to the therapeutic range of ∼5 mM. In the present study, we used a pyruvate dosing regimen that has been shown to increase arterial pyruvate levels in the intact pig from ∼100 μM to 4–6 mM within 30 min (32, 33). On the basis of these study results, we initiated the pyruvate infusion in the stunning studies before the 15-min occlusion to ensure that plasma pyruvate levels were elevated early in reperfusion. The improved postischemic regional function at the 1-h time point is consistent with elevated circulating pyruvate levels. However, the increased regional ventricular function and contractility at this time point could have simply reflected the positive inotropic effect of pyruvate that we have previously observed during intracoronary pyruvate infusions in normal and stunned myocardium (30, 48). To differentiate between the inotropic and antistunning effects of pyruvate in the present study, we terminated the pyruvate infusion after 1 h of reperfusion. It has been reported (32, 33) that plasma pyruvate levels in the pig decreased to ∼1 mM within 30 min of discontinuing this same treatment protocol. Thus it is likely that in the present study plasma pyruvate levels decreased to below the therapeutic range by at least 2 h of reperfusion. The observation that pyruvate-treated animals continued to exhibit improved postischemic regional contractility for the duration of reperfusion suggests that the pyruvate administration actually attenuated stunning and was not merely acting as a positive inotrope.

The same dosing regimen of pyruvate also decreased infarct size in the intact pig, the extent of which depended on the duration of the infusion. When pyruvate was administered before and during the LAD occlusion (protocol pyruvate I), infarct size was reduced 26% compared with saline-treated controls. In fact, the pretreatment phase of this pyruvate protocol did not appear to add much benefit, because in three additional pigs when pyruvate was infused systemically only during the LAD occlusion, infarct size was reduced by 23% (to 51 ± 2% of the AAR). When pyruvate was administered before and throughout the 60-min LAD occlusion and for the first 60-min reperfusion period (protocol pyruvate II), infarct size was reduced from 66% to 29% of the AAR, a 56% reduction compared with saline-treated controls. Given that sedentary pigs have very little native coronary collateral circulation, the reduction in infarct size in these two groups is likely due to elevated plasma pyruvate in the early reperfusion period.

One of the common components of reperfusion after both reversible and irreversible ischemic injury is oxidative/nitrosative stress to myocytes, endothelial cells, and vascular smooth muscle. The results of our rat myocyte studies indicate that pyruvate loading can significantly reduce oxidative/nitrosative stress imposed by exogenous H2O2. Myocytes isolated and stored in medium supplemented with 10 mM pyruvate exhibited significantly reduced increases in DCF fluorescence during H2O2 exposure compared with myocytes isolated in medium containing only glucose. Because pyruvate has been reported to directly scavenge H2O2 in solution (8), we did not include pyruvate in the suffusion medium that contained the H2O2. The present findings of reduced oxidative stress are thus a result of loading the myocytes with pyruvate over a period of ∼2 h, a time course similar to that used in the in vivo infarction protocol. Our findings are consistent with other studies examining the beneficial effects of pyruvate in myocardial oxidative stress. Bassenge et al. (3) reported that pyruvate infusion dose-dependently decreased postischemic ROS formation in isolated guinea pig hearts. Mallet et al. (26) reported that pyruvate treatment of isolated working guinea pig hearts after H2O2 exposure significantly improved cardiac contractility, levels of reduced glutathione and NAD+, and myocardial energetics. Thus supraphysiological pyruvate appears capable of protecting the myocardium from oxidative stress, which improves outcome in terms of reduced reperfusion stunning and markedly reduced infarct size.

Although the present in vivo results support the numerous in vitro studies on the beneficial effects of pyruvate, there are limitations to the clinical use of pyruvate. These relate principally to the volume of pyruvate solution needed to increase plasma pyruvate levels to the effective therapeutic range and associated effects on plasma sodium and potassium levels in patients with compromised renal function. However, these factors could be minimized by the co-application of diuretics and/or by coinfusion of potassium salts such as K+ gluconate. An alternative approach would be the use of a nonsodium salt version of pyruvate, such as ethyl pyruvate. This agent has also been shown to exert beneficial effects in vivo (46).

In summary, the results of the present study indicate that the intermediary metabolite pyruvate can significantly protect both reversibly and irreversibly injured in vivo myocardium. Furthermore, these effects appear to be most effective during reperfusion. Research and development of new cardioprotective therapies should continue to target myocardial metabolism.


The authors thank Dr. Salik Jahania for assistance with the load-independent cardiac function analyses.


This work was supported by National Heart, Lung, and Blood Institute Grant RO1-HL-34759 (to R. M. Mentzer, Jr.).


  • 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