| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2007;116:1386-1395.)
© 2007 American Heart Association, Inc.
Molecular Cardiology |
From the Vascular Physiology Unit, Institute of Child Health (S.P.L., R.W., A.T.P., A.D., J.E.D.), MRC Centre of Epidemiology for Child Health, Institute of Child Health (T.J.C.), Hatter Cardiovascular Institute and Centre for Cardiology (D.M.Y.), and Centre for Clinical Pharmacology and Therapeutics (S.P.L.), University College London; and Department of Cardiothoracic Surgery, UCL Hospitals NHS Foundation Trust (S.K.K.), London, United Kingdom.
Correspondence to Dr S.P. Loukogeorgakis, Vascular Physiology Unit, Institute of Child Health, University College London, 34 Great Ormond St, London, WC1N 3JE, UK. E-mail s.loukogeorgakis{at}ich.ucl.ac.uk
Received August 15, 2006; accepted July 5, 2007.
| Abstract |
|---|
|
|
|---|
Methods and Results— Endothelial function was assessed by flow-mediated dilation before and after IR (20 minutes of arm ischemia followed by reperfusion). RIPC was induced by conditioning cycles of 5 minutes of ischemia and reperfusion on the contralateral arm or leg before IR. For RPostC induction, conditioning cycles were administered during the ischemic phase of IR. Oral glibenclamide was used to determine the dependence of RIPC and RPostC on KATP channels. IR caused a significant reduction in flow-mediated dilation in healthy volunteers (baseline, 9.3±1.2% versus post-IR, 3.3±0.7%; P<0.0001) and patients with atherosclerosis (baseline, 5.5±0.6% versus post-IR, 2.3±0.5%; P<0.01). This reduction was prevented by RIPC (post-IR+RIPC: healthy volunteers, 7.2±0.5% [P<0.0001 versus post-IR]; patients, 4.5±0.3% [P<0.01 versus post-IR]) and RPostC (post-IR+RPostC: 8.0±0.5%; P<0.0001 versus post-IR). The protective effects of RIPC and RPostC were blocked by glibenclamide.
Conclusions— This study demonstrates for the first time in humans that RPostC can be induced by transient limb ischemia and is as effective as RIPC in preventing endothelial IR injury. RIPC and RPostC share mechanistic similarities, with protection being dependent on KATP channel activation. These results suggest that remote conditioning stimuli could be protective in patients with acute ischemia about to undergo therapeutic reperfusion.
Key Words: ischemia ischemic preconditioning potassium channels reperfusion injury
| Introduction |
|---|
|
|
|---|
Editorial p 1344
Clinical Perspective p 1395
The clinical applicability of IPC and PostC is limited by logistics; preconditioning requires advance warning of ischemic events, a circumstance that is unlikely to arise given their unpredictable nature in clinical practice. PostC can be performed after an ischemic event has occurred, but effecting intermittent reperfusion of vital organs requires mechanical intervention, which is possible in ST-elevation myocardial infarction, improbable in other acute coronary syndromes, and implausible in stroke.
However, systemic effects of ischemic conditioning stimuli exist that might be used to harness their protection in acute ischemic syndromes. Preconditioning the intestine,6 kidney,7 or limb8 in advance of an ischemic insult protects the myocardium and other tissues from IR injury, a facet of IPC called remote IPC (RIPC) and extensively investigated in a range of animal models9–11 and by our group in humans.8,12 Remote PostC (RPostC) is a hybrid of RIPC and PostC in that the conditioning ischemic stimulus is applied at a remote site but contemporaneously with a prolonged, injurious ischemic insult.13 The remote effects of an ischemic conditioning stimulus applied to an organ that is relatively resistant to ischemia (such as the limb) might protect tissues that are more sensitive to ischemia (such as the heart and brain).
We have used IR injury to the vascular endothelium as a model to characterize IPC and PostC in humans; in this model, IR injury causes transient endothelial dysfunction that is prevented by IPC14 and PostC.15 We also have used this model to demonstrate remote effects of IPC in humans.12 The aims of the present study were to determine whether RPostC occurred in humans and, if so, whether RIPC and RPostC shared similar mechanisms.
| Methods |
|---|
|
|
|---|
Induction of IR
The nondominant arm was made ischemic by inflating a 9-cm-wide blood pressure cuff placed around the upper part of the arm to a pressure of 200 mm Hg for 20 minutes, as described previously.12
Induction of RIPC and RPostC
RIPC and RPostC were induced by inflating a 9-cm-wide blood pressure cuff placed around the upper part of the contralateral arm or the contralateral leg. The cuff was inflated to 200 mm Hg for 5 minutes (ischemia of the arm or leg, respectively), followed by a 5-minute deflation (reperfusion). This constituted a conditioning cycle, and 2 to 3 cycles were used in the protocols described below. The timing of the conditioning cycle with respect to the IR stimulus determined whether RIPC or RPostC was induced; for RIPC, the conditioning cycle was completed in advance of IR, whereas for RPostC, the conditioning cycle was induced during IR.
Assessment of Conduit Vessel Function
Conduit vessel endothelial function was assessed by measuring flow-mediated dilation (FMD) of the brachial artery in the nondominant arm, as described previously.15,16 The dilator response of the brachial artery to glyceryl trinitrate (GTN; 25 µg sublingually) was used to assess endothelium-independent dilation.
Experimental Protocols
Effect of IR on Vascular Dilator Function
To determine the effect of IR on endothelial function, FMD was assessed before ischemia (baseline) and 20 minutes after reperfusion (n=11; Figure 1, protocol a). Similarly, the effect of IR on smooth muscle function was determined in separate studies by assessing the dilation of the brachial artery in response to sublingual GTN (25 µg) before and after IR (n=6), as described previously.12
|
Effect of RIPC on Endothelial IR Injury in Healthy Volunteers
FMD was assessed before and after IR but immediately preceded by RIPC. The following RIPC stimuli were tested for their potential to induce protection against IR-induced endothelial dysfunction: (1) 2 RIPC cycles applied on the arm (RIPCArm2C; n=9; Figure 1, protocol b), (2) 3 RIPC cycles applied on the arm (RIPCArm3C; n=9; Figure 1, protocol c), (3) 2 RIPC cycles applied on the leg (RIPCLeg2C; n=9; Figure 1, protocol b), and (4) 3 RIPC cycles applied on the leg (RIPCLeg3C; n=9; Figure 1, protocol c). In control studies, the dilation of the brachial artery in response to GTN 25 µg was measured before and after RIPC (3 cycles applied on the contralateral arm; RIPCArm3C) alone to determine whether RIPC had a direct effect on vascular smooth muscle function (n=6). We have demonstrated previously that such an RIPC protocol does not directly affect brachial artery endothelial function.12
Effect of RIPC on Endothelial IR Injury in Patients With Atherosclerosis
To determine whether the protective effects of RIPC against endothelial IR injury remain intact in patients with atherosclerosis, we recruited patients who had undergone coronary artery bypass graft surgery within the last 6 months with no evidence of active myocardial ischemia (angina) after surgery. Patients using nitrates, KATP channel openers (eg, nicorandil), or sulfonylureas (eg, glibenclamide) were excluded. FMD was measured at baseline and after IR alone (n=6; Figure 1, protocol a) or IR preceded by RIPC (RIPCArm3C; n=6; Figure 1, protocol c).
Mechanism of Protection by RIPC: Role of KATP channels
Ten healthy volunteers underwent repeat assessment of the effects of IR alone (Figure 1, protocol a) and IR immediately preceded by RIPC (RIPCArm3C; Figure 1, protocol c) on endothelial function. To assess the role of KATP channels, on a different study day, the same group of volunteers received 5 mg of the nonselective KATP channel blocker glibenclamide orally. To achieve peak blood concentration of the drug before the start of the study, glibenclamide was administered 45 minutes before baseline FMD assessment.17 This was followed by the application of a 3-cycle RIPC stimulus on the contralateral arm, IR, and a final FMD assessment (Figure 2, protocol a). To exclude a direct effect of glibenclamide on endothelial IR injury, the protocol was repeated without RIPC (Figure 2, protocol b). In all studies, a high-carbohydrate meal (carbohydrates, 59 g; fat, 9 g; protein, 5 g; energy, 340 kcal) was given immediately and 3 hours (carbohydrates, 115 g; fat, 20 g; protein, 27 g; total energy, 750 kcal) after the administration of glibenclamide.18 Blood glucose was monitored throughout the study.
|
Effect of RPostC on Endothelial IR Injury
Nine healthy volunteers underwent assessment of the effects of IR on endothelial function (Figure 1, protocol a). To determine whether transient limb ischemia induces remote protection by modifying the reperfusion phase of endothelial IR injury (RPostC), 2 cycles of 5 minutes of ischemia and reperfusion were applied on the contralateral leg during the 20-minute ischemic insult (index ischemia) to the nondominant arm (RPostCLeg2C; n=9; Figure 1, protocol d). This stimulus was sufficient to prevent endothelial injury when applied in advance of IR (see the Effect of RIPC on Endothelial IR Injury in Healthy Volunteers section) and was chosen because it allowed the completion of the RPostC protocol during index ischemia. To establish whether it was necessary for the RPostC stimulus to be applied in full before the onset of reperfusion, a 3-cycle stimulus was administered on the contralateral arm during ischemia and early reperfusion (RPostCArm3C; n=9; Figure 1, protocol e). This protocol allowed the completion of 2 cycles of RPostC on the arm during index ischemia, but the final cycle was completed after reperfusion had begun.
Mechanism of Protection by RPostC: Role of KATP Channels
To establish whether protection by RPostC is dependent on KATP channel activation, the RPostCLeg2C protocol was repeated in the same group of healthy volunteers in the presence of systemic glibenclamide (n=7; Figure 2, protocol c). Glibenclamide (5 mg) was administered orally 45 minutes before baseline assessment of endothelial function, and high-carbohydrate meals were given to prevent hypoglycemia, as mentioned previously.
Calculations and Statistical Analysis
All data are expressed as mean±SEM unless otherwise stated. Brachial artery diameter was measured in millimeters, and dilation was expressed as percentage increase from baseline diameter. Data were compared by use of a paired Student t test or 1-way ANCOVA as appropriate. Comparisons by ANCOVA were performed between post-IR values adjusted for baseline FMD as the covariate. Adjustment for baseline FMD led to a nonsignificant between-subject term in the repeated-measures analysis. For multiple comparisons, probability values by ANCOVA were Scheffé adjusted. In all cases, values of P<0.05 using 2-tailed tests were considered statistically significant.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
| Results |
|---|
|
|
|---|
Effect of IR on Vascular Dilator Function
IR reduced brachial artery FMD (Figure 3a) but had no effect on arterial dilation in response to GTN (Figure 3b). A significant correlation existed between baseline and post-IR FMD (Pearsons r=0.8, P<0.01). This correlation justified using ANCOVA to adjust post-IR FMD values for baseline FMD in subsequent analyses.
|
Effect of RIPC on Endothelial IR Injury in Healthy Volunteers
RIPC (RIPCArm3C or RIPCLeg3C) immediately before IR preserved endothelial function (Table 1 and Figure 4). RIPC administered as 2 cycles was effective only when applied to the leg (Table 1 and Figure 4). RIPC had no direct effect on brachial artery smooth muscle function (dilation to GTN: baseline, 11.8±1.5% versus post-RIPC, 11.5±0.9%; P=0.8, paired t test; n=6).
|
|
Effect of RIPC on Endothelial IR Injury in Patients With Atherosclerotic Disease
IR caused a significant reduction in FMD in the patient group (baseline, 5.5±0.6% versus post-IR, 2.3±0.5%; P<0.01, paired t test; n=6). The IR-induced reduction in FMD was prevented by RIPC (RIPCArm3C) (Table 2).
|
Mechanism of Protection by RIPC: Role of KATP Channels
IR resulted in brachial artery endothelial dysfunction (Figure 5a). Glibenclamide did not affect the endothelial response to IR (Table 3 and Figure 5b). RIPC (RIPCArm3C) prevented endothelial dysfunction when applied immediately before IR but had no effect in the presence of systemic glibenclamide (Table 3 and Figure 5b).
|
|
Effect of RPostC on the Reperfusion Phase of Endothelial IR Injury
IR caused a significant reduction in brachial artery FMD (Figure 6a). RPostCLeg2C protected the endothelium of the brachial artery against IR injury (Table 4 and Figure 6b). Protection by RPostC required the completion of a sufficient stimulus during index ischemia (before the onset of reperfusion) because RPostCArm3C did not prevent an IR-induced reduction in FMD when applied during ischemia and early reperfusion (Table 4 and Figure 6b). Systemic glibenclamide blocked the protective effects of RPostC (RPostCLeg2C) against endothelial IR injury (Table 4 and Figure 6b).
|
|
| Discussion |
|---|
|
|
|---|
Vascular endothelial dysfunction is a possible pathogenic mechanism in organ dysfunction precipitated by IR injury, with potential to promote vasoconstriction and thrombosis through the loss of endothelium-derived factors, including nitric oxide and prostacyclin.19,20 In this study, the transient endothelial dysfunction of the brachial artery was used to model the protective effects of RIPC and RPostC. In agreement with previously published data from our group,8,12 IR injury reduced FMD of the brachial artery by
50% and was largely prevented by RIPC. RIPCArm3C induced protection, but shortening the stimulus to 2 cycles was not protective. Tissue volume exposed to the preconditioning stimulus also determined the degree of protection because 2-cycle RIPC was effective against endothelial IR injury when applied to the leg. These data are consistent with the requirement for the preconditioning stimulus to cross a threshold21–23 that may be determined by endogenous mediators, some of which (eg, catecholamines) are released during IR injury. Nonetheless, our small sample sizes cannot exclude a subtle beneficial effect of a subthreshold RIPC protocol on endothelial function after IR. Despite such uncertainties, 4-cycle RIPC of the arm has recently been shown to reduce myocardial injury caused by cardiac bypass surgery in children,24 raising the possibility that when an ischemic event can be predicted, a threshold remote conditioning stimulus might have a clinically demonstrable benefit.
Investigating the clinical utility of RIPC requires a demonstration that the phenomenon is intact in patients with atherosclerosis, who remain the principal group to benefit from tissue protection. To minimize the effect of drug therapy on the biology of preconditioning, we recruited patients who were asymptomatic of ischemic episodes after coronary artery bypass grafting. RIPC was as effective in this group as in healthy volunteers in preventing endothelial IR injury, demonstrating for the first time the preservation of RIPC in patients with established atherosclerosis.
However, most ischemic events are unpredictable, which limits the therapeutic potential of ischemic conditioning stimuli (remote or otherwise) that need to be applied in advance of such episodes.25,26 This has stimulated interest in conditioning stimuli that are protective during or after ischemia. Recent studies affirm that much of the reversible tissue damage occurring during IR injury is sustained during early reperfusion rather than during ischemia; modifying the conditions of reperfusion (gradual27,28 or intermittent reperfusion [PostC]4,5) reduces experimental IR injury to a degree similar to ischemic preconditioning stimuli. Beneficial effects of PostC in IR injury in humans have recently been demonstrated by our group in healthy volunteers15 and by Staat et al29 in patients undergoing coronary angioplasty who showed that 3 cycles of reinflation of the angioplasty balloon catheter in the first minute of reperfusion effected intermittent restoration of blood flow to the ischemic myocardium and reduced infarct size.
Nonetheless, mechanical intervention at the onset of reperfusion will be feasible in only a minority of ischemic events. An alternative is RPostC, defined as "an ischemic conditioning stimulus applied concurrently with an injurious ischemic episode but at a remote site,"13 which combines facets of both RIPC and PostC. We observed that a conditioning stimulus consisting of 2 cycles of leg ischemia followed by reperfusion applied during index (20-minute) arm ischemia prevented brachial artery endothelial dysfunction caused by IR. These data demonstrate for the first time in humans that RPostC reduces experimental IR injury to the vascular endothelium. The degree of protection was similar to that seen with RIPC, provided that the ischemic conditioning stimulus crossed a threshold for protection before reperfusion occurred; a subthreshold conditioning stimulus (2 cycles on the contralateral arm during index ischemia) did not induce protection.
Emerging evidence from animal studies suggests that IPC/RIPC and PostC share common signaling pathways, including triggers (adenosine receptor stimulation7,30), mediators (protein kinase C activation31,32), and end effectors (opening of mitochondrial KATP channels,7,33 activation of prosurvival kinases [PI3K-Akt, Erk1/2],34,35 and inhibition of mitochondrial permeability transition pore opening36,37). In this study, we chose to investigate the role of KATP channels in the mechanisms of protection by RIPC and RPostC. Opening of KATP channels, particularly the mitochondrial subtype, has been shown to be a prerequisite for the induction of protection against IR injury by IPC,38 RIPC,7 PostC,33 and RPostC.39 Glibenclamide, a nonselective KATP channel blocker, abolished protection induced by RIPC and RPostC, the first such demonstration in humans of the involvement of KATP channels in these phenomena. These data do not identify the cellular location (sarcolemmal versus mitochondrial) or the site in the conditioning pathway of the channels involved in protection by RIPC or RPostC. KATP channel blockade did not affect baseline FMD and did not exacerbate IR injury. Thus, the effect of glibenclamide on RIPC and RPostC could not be attributed to direct actions of glibenclamide to alter baseline endothelial function or the endothelial response to IR. These findings using systemic glibenclamide are convergent with our previous results using local infusions of glibenclamide,40 suggesting that minimal effects of systemic blockade of KATP channels on endothelial function are present in this model.
Uncertainties remain regarding the mechanism of transfer of protection from one limb to the other. If humoral factors are involved,7,9,13,41,42 they need to be active during the critical early stages of reperfusion to induce protection.13,43 An alternative, a neurogenic pathway, could be activated during conditioning ischemia and induce protection in tissues that are cut off from the circulation.6,31,44,45 These are issues for further study. Regardless of these and other mechanistic uncertainties, this study demonstrates that ischemic protection can be induced in humans by the application of contemporaneous ischemic conditioning stimuli at remote sites. Such remote conditioning schedules could be used as adjuncts to current reperfusion strategies to protect from more substantial injury in patients with acute myocardial infarction or stroke.
| Acknowledgments |
|---|
The study was funded by the British Heart Foundation (MB PhD studentship FS/03/050).
Disclosures
None.
| References |
|---|
|
|
|---|
Related Article:
Circulation 2007 116: 1337.
This article has been cited by other articles:
![]() |
D. J. Hausenloy and D. M. Yellon Remote ischaemic preconditioning: underlying mechanisms and clinical application Cardiovasc Res, May 20, 2008; (2008) cvn114v2. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Downey and M. V. Cohen Bypassing Big Pharma Circulation, September 18, 2007; 116(12): 1344 - 1345. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |