Author + information
- Received October 4, 2001
- Revision received May 7, 2002
- Accepted June 27, 2002
- Published online October 2, 2002.
- Satoshi Genda, MD*,
- Tetsuji Miura, MD, PhD, FACC*,* (, )
- Takayuki Miki, MD, PhD*,
- Yoshihiko Ichikawa, MD* and
- Kazuaki Shimamoto, MD, PhD*
- ↵*Reprint requests and correspondence:
Dr. Tetsuji Miura, Second Department of Internal Medicine, Sapporo Medical University School of Medicine, South-1 West-16, Chuo-ku, Sapporo 060-8543 Japan.
Objectives This study aimed to clarify the role of adenosine triphosphate-sensitive K+ (KATP) channels in the no-reflow phenomenon and in its extension by hypercholesterolemia.
Background The no-reflow phenomenon is an important target of therapy in patients with acute myocardial infarction, but its mechanism remains unclear.
Methods The left circumflex coronary artery was occluded for 30 or 60 min and reperfused in rabbit hearts in situ. The no-reflow zone, area at risk, and infarct size were determined by thioflavin-S, Evans blue, and tetrazolium staining, respectively. No-reflow zone size was expressed as a percentage of infarct size (%NR/IS). Hypercholesterolemia was induced by two weeks of cholesterol-enriched diet.
Results A KATP channel blocker, glibenclamide (0.3 mg/kg), increased %NR/IS after 30-min ischemia/90-min reperfusion from 33.6 ± 1.9% to 45.9 ± 1.6% and %NR/IS after 60-min ischemia/90-min reperfusion from 32.8 ± 3.4% to 46.1 ± 1.7%. However, NG-monomethyl-l-arginine (l-NMMA), a nitric oxide (NO) synthase inhibitor, and nicorandil, a hybrid of KATP channel opener and nitrate, failed to significantly modify %NR/IS. Hypercholesterolemia increased %NR/IS to 61.6 ± 0.6%, which was not further enlarged by glibenclamide, and delayed infarct healing during the subsequent five days of reperfusion. These effects of hypercholesterolemia were significantly suppressed by nicorandil. Neither glibenclamide, l-NMMA, nicorandil, nor hypercholesterolemia modified infarct size.
Conclusions The KATP channel activation, but not NO, is a major mechanism of protection against microvascular injury, causing the no-reflow phenomenon in the heart. Suppression of KATP channel opening may underlie the hypercholesterolemia-induced extension of no-reflow, which delays infarct healing.
When a coronary artery has been occluded for a long period of time, reopening of the artery fails to restore perfusion in the entire ischemic myocardium and leaves some areas without reperfusion (1–4). Even after recanalization of the occluded coronary artery, the no-reflow phenomenon is observed in 25% to 30% of patients (3,5). Furthermore, this phenomenon has been shown to be associated with ventricular remodeling and poor functional recovery after infarction (3,5–7). Therefore, the no-reflow phenomenon appears to be an important target of therapy, but the mechanism of this phenomenon and methods for its prevention still remain elusive.
Although it has been suggested that plugging of capillaries by leukocytes and platelet activation contribute to no-reflow (1,2,8), these blood cell elements are not necessary for the development of this phenomenon, because no-reflow has been observed in buffer-perfused hearts as well (9,10). It is not clear how ischemia induces irreversible damage in the endothelium and vascular smooth muscle cells after ischemia/reperfusion and whether there is any endogenous mechanism against development of ischemic microvascular injury, causing the no-reflow phenomenon.
In the present study, we examined the roles of adenosine triphosphate-sensitive K+ (KATP) channels and nitric oxide (NO) in the no-reflow phenomenon by using a KATP channel blocker, an inhibitor of nitric oxide synthase (NOS), and nicorandil, which is a hybrid drug of KATP channel opener (11) and nitrate. In addition, we examined the possibility that reported extension of the no-reflow zone by hypercholesterolemia is due to suppressed function of the KATP channel. Rationales for this hypothesis are previous findings that chronic hypercholesterolemia impairs functions of endothelial pertussis toxin-sensitive G proteins (GPTX) (12) and that GPTX mediates a signal to the KATP channel in coronary arterioles (13,14). However, to avoid the influence of atherosclerosis on ischemic vascular injury, we assessed the effects of a relatively short period (i.e., two weeks) of hypercholesterolemia on the no-reflow phenomenon.
This study was conducted in accordance with “The Guide for the Care and Use of Laboratory Animals,” published by the National Institutes of Health (NIH publication No. 85-23, revised 1996) and was permitted by the Animal Use Committee of Sapporo Medical University.
Experiment 1: effects of KATP channel blockade and inhibition of NOS on the no-reflow phenomenon
Male albino rabbits (Japanese White, 11 to 13 weeks old) were prepared to induce myocardial infarction in vivo as in our previous studies (15,16). In brief, rabbits were anesthetized with intravenous (IV) pentobarbital (40 mg/kg), intubated, and ventilated by a Harvard respirator (model 683, Harvard Apparatus, South Natick, Massachusetts) using room air and oxygen supplement. The heart was exposed, and a coronary snare was placed around the marginal branch of the left coronary artery. Fluid-filled catheters were placed in a carotid artery and a jugular vein for monitoring of blood pressure and for drug injection, respectively. An electrocardiogram was monitored using precordial bipolar electrodes.
The left circumflex coronary artery of each rabbit was occluded for 30 min (30-min ischemia protocol) or 60 min (60-min ischemia protocol) and reperfused for 90 min. In both protocols, rabbits were subjected to one of the following four pretreatments before myocardial ischemia: 1) no pretreatment; 2) IV injection of glibenclamide, a selective KATP channel blocker; 3) IV injection of NG-monomethyl-l-arginine (l-NMMA), an NOS inhibitor; and 4) infusion of nicorandil. Glibenclamide (0.3 mg/kg) was administered at 20 min before coronary occlusion. The l-NMMA (30 mg/kg) was injected 5 min before ischemia and 5 min after reperfusion in the 30-min ischemia protocol and 5 min before and 35 min after coronary occlusion and 5 min after reperfusion in the 60-min ischemia protocol. Nicorandil infusion (10 μg/kg/min IV) was commenced 5 min before ischemia and was continued until the end of reperfusion. In pilot experiments, this dose of l-NMMA inhibited hypotensive response (i.e., ∼20 mm Hg of blood pressure fall) to 1 μg/kg/min of acetylcholine for at least 40 min. After 90 min of reperfusion, the rabbits were heparinized with 2,000 U of heparin sodium, and 1 ml/kg of 4% thioflavin-S in saline was injected as a bolus to stain the vascular endothelium in vivo.
Postmortem analysis of the heart
After excision of the heart, the circumflex artery was reoccluded to avoid wash-out and diffusion of thioflavin-S from capillaries. The heart was then mounted onto a Langendorff apparatus, and Evans blue dye was infused into the perfusion line to negatively mark the territory of the occluded artery (i.e., area at risk). The heart was frozen and sectioned into six slices (each of 2-mm in thickness) from the apex to the base. Every other slice was stained with triphenyltetrazolium chloride to visualize infarcts (Fig. 1A) or illuminated with ultraviolet light to determine the no-reflow zone (Fig. 1B), but the areas of infarct, risk zone, and no-reflow zone in the same cross sections could be measured by tracing those areas on the facing aspects of two adjacent slices. The trace images of these areas were read by a Macintosh G3 computer and measured using NIH image, an image analysis software. To confirm no-reflow zone assessment by thioflavin-S, histology slides were prepared from the slices used for no-reflow analysis and stained with hematoxylin-eosin.
Experiment 2: effects of hypercholesterolemia on the no-reflow phenomenon
Preparation of hypercholesterolemic rabbits and their controls
Eleven-week-old rabbits were fed a 1% cholesterol-enriched diet (Oriental Yeast Co., Tokyo, Japan) or a standard laboratory chow for two weeks before the experiment. Rabbits were prepared as in Experiment 1, and blood samples were taken on the day of the experiment for measurements of plasma total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), and triglyceride (TG) by enzymatic assay.
After a 15-min stabilization period, all rabbits underwent 30-min coronary occlusion and 90-min reperfusion. Rabbits that were fed a standard chow received no pretreatment and served as controls. Hypercholesterolemic rabbits received injection of saline (vehicle) or one of the following three treatments: glibenclamide injection, nicorandil infusion, and nitroglycerin infusion. Glibenclamide (0.3 mg/kg) was intravenously injected at 20 min before coronary occlusion. Nicorandil (10 μg/kg/min) infusion and nitroglycerin (1 μg/kg/min) infusion were commenced at 5 min before ischemia and continued until the end of reperfusion. Determination of no-reflow zone size and risk zone size was performed as in Experiment 1. Using separate groups of hypercholesterolemic and normocholesterolemic rabbits (n = 4 for each group), ventricular muscle samples without ischemia were excised and quickly frozen in liquid nitrogen. The protein levels of SUR2, Kir6.1, and Kir6.2 in the tissue samples were assessed by Western blotting.
Western blotting of SUR2, Kir6.1, and Kir6.2
Ventricular tissue samples were homogenized, and particulate fractions were separated from cytosolic fractions and processed for Western blotting as previously reported (17). In brief, the particulate samples were electrophoresed on a 12.5% polyacrylamide gel and then electroblotted onto polyvinylidine difluoride membranes (Millipore, Bedford, Massachusetts). After blocking with buffer containing 5% nonfat dry milk, the blots were then incubated with 1,000-fold diluted antibodies against SUR2, Kir6.1, and Kir6.2 (Santa Cruz Biotechnology, Santa Cruz, California). These proteins were then visualized using secondary antibodies and an ECL Western blotting detection kit (Amersham, Little Chalfont, Minnesota). Protein levels of SUR2, Kir6.1, and Kir6.2 were quantified by using SigmaGel, gel analysis software (SPSS Inc., Chicago, Illinois).
Experiment 3: effects of KATP channel blockade and hypercholesterolemia on infarct healing
Surgical preparation and experimental protocols
Because a KATP channel blocker and hypercholesterolemia enlarged no-reflow zone size (see Results section), the impact of such extension of the no-reflow zone on infarct healing was examined in these experiments. Rabbits with hypercholesterolemia and normal controls were prepared as in Experiments 1 and 2. All rabbits underwent 60-min coronary occlusion and reperfusion. Normal control rabbits received injection of saline or glibenclamide (0.3 mg/kg) at 20 min before coronary occlusion. Hypercholesterolemic rabbits received saline or nicorandil infusion (10 μg/kg/min) from 5 min before ischemia until 90 min after reperfusion.
After stabilization of hemodynamic parameters following reperfusion, the surgical wounds were repaired, and rabbits were returned to their cages for recovery. At five days after the surgery, rabbits were heparinized and hearts were then excised and processed for postmortem analysis. In these experiments, surgical procedures were performed under sterile conditions, and a combination of 100 mg ampicillin and 100 mg cloxacillin was injected intramuscularly for prophylaxis of infection. After the surgery, all rabbits were fed a standard chow.
Excised hearts were immersion-fixed in 10% neutrally buffered formalin for at least two days and sectioned into 2.5-mm-thick slices. Histology slides were prepared from each heart slice by standard techniques and stained with hematoxylin-eosin and Mallory’s connective tissue stain modified by Heidenhaim. The slide images were enlarged 9.2 times using a CanoScan 600 scanner (Canon, Tokyo, Japan) and a Macintosh G3 computer. Infarcted zone and the organized zone (i.e., granulation tissue and fibrosis) of the infarct (Fig. 2) were traced and their areas were determined by NIH image. Based on our previous studies (15,16), we used the volume of organized infarct as a percentage of whole infarct size (%OZ/IS) as an index to assess the extent of infarct healing. Because %OZ/IS inversely correlates with absolute volume of infarct (indicating that infarct healing is slower in larger infarcts), delay of infarct healing was assessed as a shift of infarct size-%OZ/IS relationship. The validity of this approach for determining the effects of reperfusion and corticosteroids on the rate of infarct healing has been demonstrated (15,16).
Glibenclamide, l-NMMA, and thioflavin-S were purchased from Sigma (St. Louis, Missouri). Nitroglycerin was obtained from Nihon Kayaku Co. (Tokyo, Japan). Nicorandil was provided by Chugai Pharmaceutical (Tokyo, Japan).
One-way analysis of variance (ANOVA) combined with the Scheffé post hoc test was used to assay intergroup differences. Differences between time-related hemodynamic parameters in treatment groups were tested by two-way repeated-measures ANOVA. Differences in regression lines were compared using analysis of covariance. The difference was considered significant at a level of p < 0.05. SigmaStat (SPSS Inc.) was used to perform ANOVA. All data are presented as means ± SEM.
Mortality and hemodynamic parameters
A total of 54 rabbits were used in Experiment 1, and none of them died during the coronary occlusion and reperfusion. As shown in Table 1, baseline heart rate and mean blood pressure were comparable among the groups. Administration of l-NMMA slightly increased mean blood pressure and decreased heart rate, but these differences did not reach statistical significance. After ischemia/reperfusion, mean blood pressure was decreased in all groups without any significant intergroup differences.
Effects of glibenclamide, l-NMMA, and nicorandil on no-reflow zone size
As shown in Table 1, 60-min ischemia resulted in larger infarcts than did 30-min ischemia in the controls: infarct sizes as a percentage of area at risk (%IS/AR) were 89.3 ± 1.4% and 49.0 ± 4.6%, respectively (p < 0.05). However, sizes of the no-reflow zone as a percentage of infarct size (%NR/IS) were similar after 30-min and 60-min coronary occlusion (33.6 ± 1.9% vs. 32.8 ± 3.4%). There was also no difference between histological findings of the no-reflow zone (i.e., plugging of capillaries with red cells, interstitial hemorrhage, and contraction-band necrosis of myocytes) in the 30-min ischemia and 60-min ischemia protocols. The %IS/AR was not modified by glibenclamide, l-NMMA, or nicorandil. However, glibenclamide increased %NR/IS after 30-min ischemia and after 60-min ischemia by approximately 1.5-fold to 45.9 ± 1.6% and 46.1 ± 1.7%, respectively (p < 0.05 vs. controls). The l-NMMA treatment tended to increase %NR/IS after 30-min and 60-min ischemia, but the differences from untreated controls did not reach a statistical significance. Nicorandil failed to reduce %NR/IS in both 30-min ischemia and 60-min ischemia protocols.
Effects of cholesterol-enriched diet on plasma lipid profile
Levels of plasma TC and LDL-C were significantly higher in the rabbits fed a cholesterol-enriched diet than in the control rabbits (641 ± 48 vs. 36 ± 2 mg/dl, 364 ± 31 vs. 12 ± 1 mg/dl), whereas plasma TG level tended to be lower in the hypercholesterolemic rabbits than in the controls (46.1 ± 13.1 vs. 81.0 ± 14.5 mg/dl). No significant differences existed between plasma lipid profiles in hypercholesterolemic rabbits that received nicorandil, nitroglycerin, or glibenclamide and those that received no drug treatment (data not shown).
Mortality and hemodynamic parameters
Thirty-six rabbits were used in this series of experiments, and 8 rabbits (1 normal control, 3 hypercholesterolemic controls, 2 hypercholesterolemic rabbits that received nicorandil infusion and 2 hypercholesterolemic rabbits that received nitroglycerin) died from ventricular fibrillation during the coronary occlusion. At baseline, mean blood pressure was comparable among the groups, as shown in Table 2. Infusion of nicorandil and nitroglycerin tended to reduce mean blood pressure by a few mm Hg, but the changes were not statistically significant. After coronary occlusion and reperfusion, mean blood pressure was decreased in all groups without any significant intergroup differences. Heart rates were comparable in the study groups throughout the experiments.
Effects of hypercholesterolemia on no-reflow zone size
The sizes of the risk area were comparable among the study groups, and both %IS/AR and %NR/IS in normal controls were similar to those in Experiment 1. In hypercholesterolemia controls, %NR/IS was almost two-fold larger than that in the normal controls (61.6 ± 0.6% vs. 33.7 ± 4.2%, p < 0.05), while there was no significant difference between infarct sizes in these groups (Table 2). Administration of glibenclamide did not induce additional enlargement of the no-reflow zone in hypercholesterolemic rabbits (%NR/IS = 67.4 ± 3.3%). Nicorandil infusion, but not nitroglycerin infusion, significantly reduced the no-reflow zone size in hypercholesterolemic rabbits (%NR/IS = 46.5 ± 3.7%, 60.9 ± 1.6%, respectively). No significant difference was seen between infarct sizes in all treatment groups (Table 2).
Western blotting of SUR2, Kir6.1, Kir6.2
The densitometric analysis showed that SUR, Kir6.1, and Kir6.2 levels were slightly higher in hypercholesterolemic rabbits than in normal rabbits (919 ± 63 arbitrary units [au] vs. 897 ± 103 [au], 879 ± 77 [au] vs. 694 ± 56 [au], and 1,330 ± 34 [au] vs. 1,108 ± 58 [au]), though only the difference between Kir6.2 levels was statistically significant.
Mortality and hemodynamic parameters
Of the 34 rabbits used in Experiment 3, one rabbit in each study group died after the surgery, presumably due to heart failure after myocardial infarction. Alterations of hemodynamic parameters by ischemia/reperfusion in Experiment 3 were similar to those in Experiments 1 and 2 (data not shown).
Relationship between infarct size and extent of infarct healing
As previously reported (15,16), %OZ/IS was inversely correlated with infarct size in all study groups, indicating that larger infarcts healed more slowly (Fig. 3). However, slopes of the regression lines in hypercholesterolemic rabbits and glibenclamide-treated rabbits were more negative compared with those in the controls, though the difference between the glibenclamide-treated and control rabbits was not statistically significant. Administration of nicorandil in hypercholesterolemic rabbits significantly shifted the regression line upward compared with that for hypercholesterolemia controls. These findings suggest that infarct healing was delayed by hypercholesterolemia and possibly by glibenclamide and that nicorandil prevented hypercholesterolemia-induced delay in the infarct healing process.
Roles of the KATP channel and NO in the no-reflow phenomenon
Nitric oxide and the KATP channel are important regulators of coronary circulation, though NO dilates mainly larger arterioles and the KATP channel dilates smaller arterioles (14,18). In the present study, glibenclamide, a blocker of the KATP channel, enlarged the no-reflow zone by 40%, and this effect was observed after 30 min and also after 60 min of ischemia, resulting in different sizes of infarct. The no-reflow zone sizes in l-NMMA-treated rabbits tended to be enlarged, but differences between untreated controls and l-NMMA-treated groups were not statistically significant in either the 30-min or 60-min ischemia protocol. These findings suggest that activation of the KATP channel plays a major role in protection of the microvasculature from irreversible damage during ischemia/reperfusion, while the possibility of some contribution of NO cannot be excluded. Furthermore, levels of KATP channel activation and NO production during ischemia/reperfusion in normal rabbits appear to be above their thresholds for protection, because administration of nicorandil, a hybrid of KATP channel opener and nitrate, failed to modify the no-reflow zone size. It is also notable that this protective effect by endogenous activation of the KATP channel is likely to be specific to the microvasculature, as myocardial infarct size was not enlarged by glibenclamide.
In the present study, we selected 90 min after reperfusion as the time point for no-reflow zone determination; this is because we previously found that the size of the no-reflow zone in the same rabbit model of infarction increased threefold from 10 to 60 min after reperfusion (2). We cannot exclude the possibility that the size of the no-reflow zone increases further after this time point. However, intergroup differences in no-reflow zone size at 90 min after reperfusion inversely correlated with those in %OZ/IS, which was measured 5 days after reperfusion. These findings suggest that alterations in no-reflow zone size at 90 min after reperfusion indeed reflect the changes in ultimate size of the no-reflow zone in our rabbit preparation.
Effects of hypercholesterolemia on the no-reflow phenomenon
Extension of the no-reflow zone size by hypercholesterolemia in the present study was consistent with an earlier observation by Golino et al. (19,20), who fed rabbits a cholesterol-enriched diet for three days and induced infarction by 30-min coronary occlusion. We hypothesized that impairment of KATP channel function during ischemia/reperfusion is an important mechanism for extension of no-reflow by hypercholesterolemia. This hypothesis was supported by two lines of evidence in the present experiments. First, a KATP channel blocker, glibenclamide, did not further enlarge the no-reflow zone in hypercholesterolemic rabbits, in contrast with its effect in rabbits with normal cholesterol levels. Second, nicorandil, a hybrid of KATP opener and nitrate, significantly blunted hypercholesterolemia-induced extension of no-reflow, but this effect was not mimicked by a nitrate donor, nitroglycerin.
How hypercholesterolemia impaired KATP channel function in the microvasculature during ischemia/reperfusion remains unclear. The SUR2, Kir6.1, and Kir6.2 levels were not reduced in the hypercholesterolemic rabbits in the present study. Although we could not selectively assess the protein level of smooth muscle type SUR2 (i.e., SUR2B), these results argue against the possibility that hypercholesterolemia reduces expression of vascular KATP channels. Conversely, recent studies have provided circumstantial evidence of the involvement of G protein dysfunction in altered function of the KATP channel induced by hypercholesterolemia (12–14,21). The GPTX protein plays an important role in regulation of KATP channels (13,14), and its function is impaired by chronic hypercholesterolemia (12). A recent study by Pongo et al. (21) demonstrated dysfunction in protein kinase C–KATP channel coupling by hypercholesterolemia in rabbit coronary arteries. Furthermore, this hypercholesterol-induced impairment was restored by treatment with farnesol, suggesting that failed association of G protein with the membrane may underlie the disturbance in signal-mediated regulation of the KATP channel. Nevertheless, further investigation is needed to clarify the mechanism by which hypercholesterolemia inhibits function of the KATP channel during myocardial ischemia/reperfusion.
Infusion of nicorandil prevented 55% of the extension of the no-reflow zone induced by hypercholesterolemia. We did not use large doses of nicorandil to minimize its effects on systemic hemodynamics, and we cannot exclude the possibility that a high dose may be necessary to totally prevent the effect of hypercholesterolemia. However, other mechanisms such as hypersensitivity of platelets and dysfunction of the Ca2+-activated K+ (KCa) channel may also contribute to aggravation of the no-reflow phenomenon by hypercholesterolemia (20,22,23). Golino et al. (20) reported that depletion of platelets by anti-platelet serum prevented enlargement of the no-reflow zone size and infarct size by hypercholesterolemia. A recent study by Jeremy and McCarron (22) demonstrated that hypercholesterolemia impaired KCa-mediated vasodilation in systemic arteries in rabbits. Such a KCa channel dysfunction in the coronary arteries, if any, would accelerate ischemia-induced injury of microvessels, because the KCa channel plays a critical role in coronary vasodilation under the condition of hypoperfusion (23).
Effect of the no-reflow phenomenon on infarct healing
An earlier study from this laboratory (15) demonstrated that reperfusion after myocardial infarction substantially accelerates replacement of necrotic myocytes with granulation tissues, even when reperfusion was too late to salvage the myocardium. Because the no-reflow phenomenon compromises reperfusion of the infarcted myocardium, it should offset the reperfusion-induced acceleration of infarct healing. Owing to methodological limitation, we could not measure no-reflow zone and extent of infarct healing in the same animal. However, enlargement of the no-reflow zone size by glibenclamide and hypercholesterolemia and its reversal by nicorandil corresponded to their effects on the extent of infarct healing (Table 2, Fig. 3). Effects of glibenclamide on no-reflow zone size and %OZ/IS were modest compared with those of hypercholesterolemia. Furthermore, the effects of hypercholesterolemia on both no-reflow zone size and %OZ/IS were significantly attenuated by nicorandil. These findings strongly support the notion that the no-reflow phenomenon delays the healing process of infarcts.
Possible clinical implications
Clinical studies (6,24) have shown that reduction in no-reflow zone size suppressed subsequent ventricular remodeling and dysfunction a month later. However, earlier studies (1,2,25) and the present experiments (Table 2) have failed to demonstrate that suppression of no-reflow results in infarct size limitation. A possible explanation for this apparent discrepancy is that a clinical benefit of suppression of no-reflow may be acceleration of infarct repair, which would prevent expansion of the infarcted region and possibly preserve the contractile function of the salvaged subepicardial myocardium.
In contrast with the present rabbit experiments, hypercholesterolemia is unlikely to be a requisite for beneficial effects of nicorandil on no-reflow in human hearts (6). A plausible explanation for this difference is that the coronary microvasculature in humans, particularly in patients with coronary artery disease, may be vulnerable to ischemia/reperfusion injury. In fact, coronary endothelial dysfunction has been demonstrated in patients with coronary artery disease (26), though alteration in vascular smooth muscles has not been fully characterized (27).
☆ This study was supported by a grant from Sapporo Medical University Foundation for Medical Research, Sapporo, and a grant from Chugai Pharmaceutical, Tokyo, Japan.
- analysis of variance
- pertussis toxin-sensitive G protein
- infarct size as a percentage of area at risk
- KATP channel
- adenosine triphosphate-sensitive K+ channel
- KCa channel
- Ca2+-activated K+ channel
- low density lipoprotein-cholesterol
- nitric oxide
- nitric oxide synthase
- no-reflow zone size as a percentage of infarct size
- volume of organized infarct as a percentage of whole infarct size
- total cholesterol
- Received October 4, 2001.
- Revision received May 7, 2002.
- Accepted June 27, 2002.
- American College of Cardiology Foundation
- Endoh A.,
- Miura T.,
- Iimura O.
- Ito H.,
- Maruyama A.,
- Iwakura K.,
- et al.
- Asanuma T.,
- Tanabe K.,
- Ochiai K.,
- et al.
- Wu K.C.,
- Zerhouni E.A.,
- Judd R.M.,
- et al.
- Ito H.,
- Taniyama Y.,
- Iwakura K.,
- et al.
- Gerber B.L.,
- Rochitte C.E.,
- Melin J.A.,
- et al.
- Manciet L.H.,
- Poole D.C.,
- McDonagh P.F.,
- Copeland J.G.,
- Mathieu-Costello M.
- Shimokawa H.,
- Flavahan N.A.,
- Vanhoutte P.M.
- Komaru T.,
- Tanikawa T.,
- Sugimura A.,
- et al.
- Tanikawa T.,
- Kanatsuka H.,
- Koshida R.,
- et al.
- Ohnuma Y.,
- Miura T.,
- Miki T.,
- et al.
- Komaru T.,
- Lamping K.G.,
- Eastham C.L.,
- Dellsperger K.C.
- Golino P.,
- Maroko P.R.,
- Carew T.E.
- Golino P.,
- Maroko P.R.,
- Carew T.E.
- Jeremy R.W.,
- McCarron H.
- Node K.,
- Kitakaze M.,
- Kosaka H.,
- et al.
- Taniyama Y.,
- Ito H.,
- Iwakura K.,
- et al.
- Kloner R.A.,
- Rude R.E.,
- Carlson N.,
- et al.