Author + information
- Received August 21, 2012
- Revision received January 29, 2013
- Accepted February 25, 2013
- Published online May 14, 2013.
- Spyridon Deftereos, MD⁎,
- Georgios Giannopoulos, MD⁎,†,⁎ (, )
- Vasileios Tzalamouras, MD⁎,
- Konstantinos Raisakis, MD⁎,
- Charalambos Kossyvakis, MD⁎,
- Andreas Kaoukis, MD⁎,
- Vasiliki Panagopoulou, MD⁎,
- Sofia Karageorgiou⁎,
- Dimitrios Avramides, MD⁎,
- Konstantinos Toutouzas, MD‡,
- George Hahalis, MD§,
- Vlassios Pyrgakis, MD⁎,
- Antonis S. Manolis, MD∥,
- Dimitrios Alexopoulos, MD§,
- Christodoulos Stefanadis, MD‡ and
- Michael W. Cleman, MD†
- ↵⁎Reprint requests and correspondence:
Dr. Georgios Giannopoulos, Cardiology Department, Athens General Hospital “G. Gennimatas,” 154 Mesogeion Avenue, 11527 Athens, Greece
Objectives The aim of the present study was to assess the efficacy of remote ischemic post-conditioning (RIPC) by repeated intermittent balloon inflations in preventing acute kidney injury (AKI) in patients with a non–ST-segment elevation myocardial infarction undergoing percutaneous coronary intervention (PCI).
Background AKI complicating PCI is associated with increased morbidity and mortality. Remote ischemic preconditioning, using cycles of upper limb ischemia-reperfusion as a conditioning stimulus, has been recently shown to prevent AKI in patients undergoing elective coronary angiography.
Methods Eligible patients were randomized to receive RIPC by cycles of inflation and deflation of the stent balloon during PCI or a sham procedure (control patients). The primary endpoint was AKI, defined as an increase of ≥0.5 mg/dl or ≥25% in serum creatinine within 96 h from PCI. The 30-day rate of death or re-hospitalization for any cause was one of the secondary endpoints.
Results A total of 225 patients were included (median age, 68 years; 36% female). The AKI rate in the RIPC group was 12.4% versus 29.5% in the control group (p = 0.002; odds ratio: 0.34; 95% confidence interval: 0.16 to 0.71). The number needed to treat to avoid 1 case of AKI was 6 (95% confidence interval: 3.6 to 15.2). The 30-day rate of death or re-hospitalization for any cause was 22.3% in the control group versus 12.4% in RIPC patients (p = 0.05).
Conclusions RIPC by serial balloon inflations and deflations during PCI was found to confer protection against AKI in patients with a non–ST-segment elevation myocardial infarction undergoing PCI. The reduction in the rate of AKI translated into a clear trend (of borderline significance) toward better 30-day clinical outcome.
Renal dysfunction complicating percutaneous coronary intervention (PCI) is associated with a marked increase in morbidity and mortality, both in the short and long term (1,2). This is especially true for patients with acute myocardial infarction treated with PCI (3). However, despite indications that the incidence of acute kidney injury (AKI) shows a temporal trend of decline, presumably due to increased clinician awareness and better risk stratification (4), well-defined measures of renoprotection, with documented efficacy, are lacking.
The notion of remote ischemic conditioning encompasses a group of procedures whose common denominator is protection of an organ against tissue damage by application of cycles of brief periods of ischemia and reperfusion in a remote site (usually a different organ or a limb) (5–7). It was recently reported that cycles of ischemia-reperfusion of the upper limb (by means of a manometer cuff) can prevent AKI in patients undergoing elective coronary angiography (8). These are very promising results, but data in other clinically relevant patient populations, including those undergoing PCI in the context of an acute coronary syndrome, are needed.
The aim of the present study was to assess the efficacy of remote ischemic post-conditioning (RIPC) by repeated intermittent balloon inflations to prevent AKI in patients with non–ST-segment elevation myocardial infarction (NSTEMI) undergoing PCI.
This was a prospective, single-blind, randomized, parallel-group study, conducted at 3 Greek teaching hospitals. Patients with NSTEMI undergoing PCI within 72 h from symptom onset were included. Eligible patients had to have at least 2 troponin I or C measurements above the laboratory-defined 99th percentile of a normal reference population plus 1 of the following: chest pain or electrocardiographic abnormalities consistent with ischemia or new echocardiographic wall-motion abnormality. Exclusion criteria were persistent ST-segment elevation (defined as lasting >20 min), left main disease, cardiogenic shock, PCI performed >72 h after symptom onset, on dialysis, or inability or unwillingness to provide consent. Patients with persistent ST-segment elevation were not included to have a more homogeneous population (they were enrolled in a separate study). The protocol was approved by the institutional review boards. All patients provided informed consent.
Patients with NSTEMI slated for cardiac catheterization were screened for eligibility, and informed consent was obtained. Complete review of the patient's history and physical examination was recorded on the patient case report form. Pre-procedural estimated risk of AKI (henceforth referred to as Mehran score for brevity) was calculated according to the report by Mehran et al. (9). Patients not undergoing PCI (e.g., no significant lesions or referrals for surgical treatment) were not included. Eligible patients were randomized to the RIPC treatment group or the control group. (Randomization was done by sealed serially numbered assignment envelopes created by an electronic random number generator and distributed to the recruiting centers. The operator opened the corresponding envelope before the catheterization procedure to know whether a true or sham RIPC procedure was to be performed, and the operator was the only one to know the patient's treatment assignment and was not involved further in any of the study procedures; unblinding was performed after conclusion of the study procedures and database locking.) After PCI of the presumed culprit lesion, the stent balloon was used to perform alternating inflations and deflations (i.e., the RIPC procedure was performed at only 1 lesion site per patient in all patients). In the RIPC group, four 1-min cycles were performed, each consisting of 30 s of inflation of the stent balloon to the nominal pressure and 30 s of deflation. In control patients, a sham 4-cycle RIPC procedure was performed, whereby balloons were inflated to no more than 3 atm.
All patients received standard of care, according to established clinical practice guidelines. The choice of materials was left to the operators, with the exception of the radiographic contrast medium, which had to be nonionic low osmolar. Adequate pre- and post-procedural hydration was carefully observed whenever possible (this requisite, as in real-life practice, could not be entirely fulfilled in the case of patients undergoing PCI within a few hours of presentation or in patients with dyspnea). The hydration plan included normal saline solution 1 ml/kg of body weight per hour (reduced to 0.5 mg/kg/h in patients with a left ventricular ejection fraction of ≤35%) for 12 h before PCI (whenever possible) and 24 h after the procedure. Potentially nephrotoxic drugs were withheld for at least 72 h or discontinued altogether, if possible.
The primary study endpoint was AKI, defined as an absolute increase in serum creatinine of ≥0.5 mg/dl or a relative increase of ≥25% compared with baseline within 96 h after PCI (the maximal measured concentration of serum creatinine during these 96 h was used). Secondary endpoints were the relative reduction in estimated glomerular filtration rate (eGFR), calculated with the Modification of Diet in Renal Disease formula, within 96 h after PCI, maximum serum cystatin C change compared with baseline within 96 h after PCI, as well as the 30-day rate of death or re-hospitalization for any cause. Patients were followed by telephone contact at 15 and 30 days after the index PCI. In case of hospitalization, the attending physicians were contacted to obtain information regarding the cause for admission and the patient's course. In case of death, the certifying physician was contacted. The personnel involved in data collection and handling were blinded as to each patient's randomization. Clinical endpoints were blindly adjudicated by a 3-member committee.
It was calculated that to detect a decrease in the incidence of AKI from 30% to 15% with a probability (power) of 0.80, at an alpha level of 0.05, 120 subjects per randomization group were needed. Analysis was performed on an intention-to-treat basis. Continuous variables were reported as median (interquartile range) and were compared using nonparametric tests (Mann-Whitney U and Wilcoxon for unpaired and paired observations, respectively), as most of them deviated significantly from the normal distribution. Categorical variables were expressed as percentages and counts and compared using the chi-square test (or Fisher exact test if the produced matrixes contained cells with expected count <5). Multivariate analysis of the independent effect of RIPC to the occurrence of AKI was performed using binary logistic regression analysis, where continuous variables were entered as dichotomous factors (above/below median). SPSS version 17.0 software (SPSS Inc., Chicago, Illinois) was used. Two-sided p values <0.05 were considered as indicative of statistical significance.
Study flow and baseline population characteristics
A total of 225 patients were finally included in the analysis (median age, 68 years; 81 women). From 332 initially screened patients, 231 were randomized, 225 of whom (113 in the RIPC group and 112 in the control group) underwent true or sham RIPC and were included in the analysis (Fig. 1). The 2 randomization groups were well balanced with similar epidemiological background. They had equivalent pre-procedural assessed risk for AKI, as indicated by similar Mehran score and baseline renal function parameters. There were no significant differences with regard to baseline medications, with a special focus on drugs with a potential effect on renal function (Table 1). There were also no significant differences in the procedural aspects of the index intervention, as illustrated in Table 2.
The overall incidence of the primary endpoint (AKI) was 21%. The AKI rate was significantly lower in the RIPC group compared with the control group (Fig. 2). The univariate odds ratio for patients of the RIPC group compared with the control group was 0.34 (95% confidence interval [CI]: 0.16 to 0.71), with relative risk reduction of 58% (95% CI: 27% to 76%). The number of patients needed to treat to avoid 1 case of AKI was 6 (95% CI: 3.6 to 15.2).
Overall, the eGFR decreased from a median of 75 (interquartile range: 64 to 88) ml/min/1.73 m2 to 64 (interquartile range: 49 to 85) ml/min/1.73 m2 (p < 0.001 for the difference). The median relative decrease in eGFR was 10% (interquartile range: 0% to 23%). The decrease in the eGFR was significantly more marked in control patients compared with the RIPC group (Fig. 2). Serum creatinine increased by 14.3% (interquartile range: 0.0% to 33.3%) in the control group versus 7.1% (interquartile range: 0.0% to 14.3%) in the RIPC group (p = 0.004). RIPC was also associated with smaller increases in cystatin C levels (Fig. 2).
With regard to the 30-day death or re-hospitalization rate, 25 of 112 control patients had an event (22.3%) compared with 14 of 113 patients of the RIPC group (12.4%). Almost all of the events were re-hospitalizations (23 of 25 in the control group and 13 of 14 in the RIPC group). The admission diagnoses were, in the control and the RIPC group, respectively, acute heart failure (9 and 5), nonspecific chest pain (5 and 2), persistent atrial fibrillation (3 and 2), unstable angina (2 and 1), presyncope (1 and 2), myocardial infarction (1 and 0), stroke (1 and 0), and trauma/bone fractures (1 and 1). Although the arithmetic difference in the clinical events between the 2 groups is substantial, its statistical significance is only borderline (p = 0.047 using the uncorrected chi-square test, whereas it becomes 0.054 if the Fisher exact test is used [odds ratio: 0.49; 95% CI: 0.22 to 1.01]); of note, the study was underpowered to assess clinical endpoints.
In the multivariate analysis, RIPC was an independent negative predictor of AKI after adjusting for age, body mass index, volume of administered contrast medium, and baseline eGFR. The multivariate hazard ratio for patients in the RIPC group was 0.23 (95% CI: 0.11 to 0.50; p < 0.001) (Fig. 3).
The principal finding of this prospective, randomized study is the decrease in AKI rate associated with RIPC in patients undergoing PCI in the context of an acute coronary syndrome. RIPC was implemented by intermittent inflations and deflations of the stent balloon during PCI. The reduction in AKI was accompanied by a clear, but marginally significant, trend toward fewer clinical events (death or re-hospitalization for any cause) in the 30-day follow-up after the index PCI. This observation regarding the clinical implications of decreasing AKI (which is a laboratory endpoint) is consistent with existing evidence that renal dysfunction associated with PCI is an important predictor of adverse outcomes (1–3).
Remote ischemic conditioning was originally described by Przyklenk et al. (7), who demonstrated that the application of brief occlusions and reperfusion of the circumflex coronary artery dramatically reduced the size of the infarct arising from a sustained occlusion of the left anterior descending coronary artery. An indication for the potential of protecting the kidney through RIPC on the myocardium in a relevant clinical setting has been provided by an insightful study of Whittaker and Przyklenk (10), who showed retrospectively, in 45 patients with acute myocardial infarction undergoing PCI, that those who received ≥4 balloon inflations and deflations during angioplasty exhibited preserved renal function, in terms of the eGFR, compared with those who received <4 balloon inflations. This study was retrospective and unpowered to demonstrate differences in the rate of AKI, but its results were hypothesis generating and are confirmed by the present trial. Importantly, Er et al. (8) recently reported that in patients with impaired renal function undergoing elective coronary angiography, remote preconditioning by cycles of limb ischemia-reperfusion (inflation and deflation of a manometer cuff placed at the arm) leads to reduction of AKI. The present study differs on several important points from the study of Er et al. (8), namely, in the conditioning stimulus and procedure, the timing of conditioning, and the patient population and clinical setting. Patients undergoing PCI in the context of an acute coronary syndrome represent a particularly relevant clinical population because they are at increased risk of renal injury compared with those undergoing elective angiography due to various additional risk factors, including hemodynamic instability, greater possibility of renal microembolization, and the inability to consistently observe an adequate hydration protocol (for example, in cases in which the patient is catheterized under emergent or semiemergent circumstances or has dyspnea/pulmonary congestion) (9,11–13). The present study and that by Er et al. (8) provide evidence that 2 different RIPC modalities can be effective in preventing AKI; however, given the different populations and designs of the 2 studies, no conclusions can be drawn on the relative effectiveness and merits of each one.
The notion of remote conditioning as a biological phenomenon, irrespective of its potential clinical significance, is well established (6,14–16). Ischemic conditioning for renoprotection, either local or remote, has shown evidence of effectiveness in various settings (16–21). However, the accumulated evidence should climax into demonstration of definitive clinical benefit if any of these ischemic conditioning methods are to become a part of clinical practice as an effective measure of renoprotection. The results of the study by Er et al. (8) are certainly pointing in this direction, as are the results of the present study, especially the demonstrated borderline significant benefit in terms of reduced 30-day clinical event rate.
With regard to its mechanistic substrate, remote conditioning may appear to be something like a “spooky action at a distance” (spukhafte Fernwirkung, a term that Einstein used to deride quantum entanglement). However, several experimental studies have identified a large part of the mechanisms underlying remote conditioning, although there remain a lot of unanswered questions. Three main mechanisms for transmitting the protective signal from the organ or tissue where the conditioning stimulus is applied to the target organ have been proposed: the neuronal pathway, the release of circulating humoral factors, and activation of a systemic protective effect (such as an antiapoptotic or anti-inflammatory response) (22). The neuronal pathway involves the release of endogenous autacoids, including neuropeptides such as calcitonin gene–related peptide (23), adenosine (21), and bradykinin (24), from the remotely conditioned organ to activate local afferent nerves. These in turn stimulate efferent nerves that terminate at the remote organ and tissue to mediate protection. Opioid, erythropoietin, endocannabinoid, angiotensin I, and prostaglandin receptors and the associated signaling pathways have been implicated in the protective effect of remote ischemic conditioning through humoral mechanisms (22). However, the actual circulating humoral mediators of remote conditioning remain as yet unidentified. These 2 mechanisms are apparently in a multifaceted cross-talk with system-level responses to the conditioning stimulus, including suppression of proinflammatory and proapoptotic genes and modulation of mitochondrial channel activity, leading to more efficient energy management and increased resistance to ischemia-reperfusion damage (25).
The present study was not designed to provide any mechanistic insight into the potential humoral or other mediators of remote conditioning. It would certainly be interesting to measure potential conditioning effectors and study their associations, if any, with clinical parameters and outcomes. Furthermore, although AKI is a valid surrogate marker of renal damage and has been described extensively in the literature, it is not a clinical endpoint. The trend toward a better 30-day outcome observed in the RIPC group, as a secondary endpoint, is of considerable import, but does not equate to a primary clinical outcome measure, for which a more powered study would be necessary.
RIPC, using intermittent myocardial ischemia-reperfusion cycles with serial balloon inflations and deflations, appears to confer protection against AKI in patients with NSTEMI undergoing PCI. The reduction in the rate of AKI translates to a clear trend toward better clinical outcome in the 30-day follow-up (lower rate of death or re-hospitalization for any cause). Of importance, the described RIPC protocol involves no additional materials or expenses and presents minimal, if any, technical or logistic challenges. However, more powered studies with longer follow-up are needed to assert that remote ischemic conditioning can indeed lead to clinical benefit and to compare different types of conditioning stimuli.
The authors thank Prof. Peter Whittaker, Wayne University School of Medicine, Detroit, Michigan, who, although not directly involved in the present study, provided the basic idea for its design.
The authors have reported that they have no relationships relevant to the content of this paper to disclose. Drs. Deftereos, Giannopoulos, and Tzalamouras contributed equally to this paper.
- Abbreviations and Acronyms
- acute kidney injury
- estimated glomerular filtration rate
- non–ST-segment elevation myocardial infarction
- percutaneous coronary intervention
- remote ischemic post-conditioning
- Received August 21, 2012.
- Revision received January 29, 2013.
- Accepted February 25, 2013.
- American College of Cardiology Foundation
- Rihal C.S.,
- Textor S.C.,
- Grill D.E.,
- et al.
- Wi J.,
- Ko Y.G.,
- Kim J.S.,
- et al.
- Cheung M.M.,
- Kharbanda R.K.,
- Konstantinov I.E.,
- et al.
- Rentoukas I.,
- Giannopoulos G.,
- Kaoukis A.,
- et al.
- Przyklenk K.,
- Bauer B.,
- Ovize M.,
- Kloner R.A.,
- Whittaker P.
- Er F.,
- Nia A.M.,
- Dopp H.,
- et al.
- Mehran R.,
- Aymong E.D.,
- Nikolsky E.,
- et al.
- McCullough P.A.
- Fukumoto Y.,
- Tsutsui H.,
- Tsuchihashi M.,
- Masumoto A.,
- Takeshita A.,
- Cholesterol Embolism Study (CHEST) Investigators
- Marenzi G.,
- Lauri G.,
- Assanelli E.,
- et al.
- Cheung M.M.,
- Kharbanda R.K.,
- Konstantinov I.E.,
- et al.
- Walsh S.R.,
- Sadat U.,
- Boyle J.R.,
- et al.
- Wever K.E.,
- Warlé M.C.,
- Wagener F.A.,
- et al.
- Schoemaker R.G.,
- van Heijningen C.L.