Abstract
Over the course of the last 2 decades, the concept of remote ischemic conditioning (RIC) has attracted considerable research interest, because RIC, in most of its embodiments offers an inexpensive way of protecting tissues against ischemic damage inflicted by a number of medical conditions or procedures. Acute kidney injury (AKI) is a common side effect in the context of various medical procedures, and RIC has been suggested as a means of reducing its incidence. Outcomes regarding kidney function have been reported in numerous studies that evaluated the effects of RIC in a variety of settings (eg, cardiac surgery, interventions requiring intravenous administration of contrast media). Although several individual studies have implied a beneficial effect of RIC in preserving kidney function, 3 recently published randomized controlled trials evaluating more than 1000 patients each (Effect of Remote Ischemic Preconditioning in the Cardiac Surgery, Remote Ischaemic Preconditioning for Heart Surgery, and ERICCA) were negative. However, AKI or any other index of renal function was not a stand-alone primary end point in any of these trials. On the other hand, a range of meta-analyses (each including thousands of participants) have reported mixed results, with the most recent among them showing benefit from RIC, pinpointing at the same time a number of shortcomings in published studies, adversely affecting the quality of available data. The present review provides a critical appraisal of the current state of this field of research. It is the opinion of the authors of this review that there is a clear need for a common clinical trial framework for ischemic conditioning studies. If the current babel of definitions, procedures, outcomes, and goals persists, it is most likely that soon ischemic conditioning will be “yesterday’s news” with no definitive conclusions having been reached in terms of its real clinical utility.
Keywords
Introduction
When Einstein stumbled upon the—later acknowledged as—nonlocal nature of quantum correlations (one form of which is what is more widely known today to the lay public as “quantum entanglement”), he conceived it as a paradox and used it to demonstrate the apparent incompatibility of quantum mechanics with “reality,” coining the term “spooky action at a distance” (spukhafte Fernwirkung). 1 Not much less disbelieving has been the approach of some clinicians toward the concept of remote ischemic conditioning (RIC). In 1993, Przyklenk et al 2 suggested that ischemic conditioning may not only be beneficial for the myocardial area in which ischemia is directly provoked, but distant cardioprotection effects do exist as well. This notion soon evolved to the concept of RIC. 3 According to this concept, ischemic conditioning could exert protective effects against ischemia/reperfusion injury to tissues and/or organs that are remote to the ones that reversible ischemia is applied (interorgan protection). 4,5 Although still not fully elucidated, experimental data suggest that neuronal and humoral pathways are involved in this endogenous protective process. 3,6
The effects of RIC on kidney function have been reported in a variety of patient subsets, undergoing procedures with or without intravenous contrast agent administration. Recently, the findings of 3 key randomized controlled trials (RCTs) regarding RIC during cardiac surgery were published, namely, The Effect of Remote Ischemic Preconditioning in the Cardiac Surgery (RIPC), 7 the Remote Ischaemic Preconditioning for Heart Surgery (RIPHeart) 8 study, and the Effect of Remote Ischaemic Preconditioning on Clinical Outcomes in CABG Surgery (ERICCA) study. 9 Their findings have challenged the potential beneficial effect of RIC on kidney function, in contrast to what has been suggested by a number of previous studies. 10
The goal of the present review is to provide a critical synopsis of the current state of affairs in this field.
Pathophysiologic Background
Several experimental studies have investigated the mechanisms underlying RIC, but a lot of unanswered questions remain. Three main ways of transmission of 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 humoral pathway, and activation of a systemic protective effect (such as an antiapoptotic or anti-inflammatory response). 11
The neuronal pathway begins with the release of endogenous autacoids, including certain neuropeptides, 12 ∓14 in response to the ischemic insult from the remotely conditioned organ to activate local afferent nerves, which in turn stimulate efferent nerves terminating at the remote tissue to mediate protection.
Angiotensin-1, endocannabinoid, erythropoietin, prostaglandin, and opioid receptors, together with the associated signaling pathways, have been implicated in the protective effect of RIC through humoral mechanisms. 11,15 However, the actual circulating humoral mediators of remote conditioning remain as yet unidentified.
These 2 mechanisms seem to also trigger system-level responses, including inhibition of proinflammatory and proapoptotic genes and modulation of mitochondrial channel activity, leading to more efficient energy management and increased resistance to ischemia–reperfusion damage. 16
Current Clinical Data
Randomized Controlled Trials
Although numerous studies regarding the effects of RIC are available, the majority of them evaluated a total (experimental plus control group) of less than 500 participants. However, 3 randomized control trials have been recently published (Table 1), involving more than 1000 patients each.
Key Randomized Trials Involving More than 500 Participants Evaluating Ischemic Conditioning Effects on Acute Kidney Injury.
Abbreviations: A/R, anesthetics or muscle relaxation; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass, IC, ischemic conditioning; INH, inhalation agent; IV, intravenous; P/O, primary outcome; RIPC, Remote Ischemic Preconditioning in the Cardiac Surgery; RIPHeart, Remote Ischaemic Preconditioning for Heart Surgery; SBP, systolic blood pressure.
aIncluded in a binary composite end point that was considered met if any of the parameters was positive.
bRenal failure: “new requirement of dialysis treatment”; Renal dysfunction: “renal dysfunction as postoperative serum creatinine of ≥2.0 mg/dL accompanied by an increase of creatinine of ≥0.7 mg/dL from the preoperative baseline.”
cAcute renal failure: “an increase in the serum creatinine level by a factor of 2 or more from baseline, a urine output of no more than 0.5 mL per kilogram per hour for 12 hours, the use of renal replacement therapy, or evidence of renal failure on autopsy.”
The RIPC study (Hong et al 7 ) evaluated 1280 patients who were scheduled for elective on- or off-pump cardiac surgery in 2 tertiary centers in South Korea. Participants in the experimental study arm underwent remote pre- and postconditioning courses. Outcomes on postoperative kidney function were assessed by investigator-defined criteria (Table 1). Renal failure or dysfunction was included in the composite primary end point, but the study was not primarily designed to assess potential differences on postoperative kidney function. Neither renal failure nor renal dysfunction differed between the RIC and the control group.
The RIPHeart study (Meybohm et al 8 ) evaluated 1385 patients in 14 academic hospitals in Germany. Patients who were scheduled for elective on-pump cardiac surgery were recruited. Remote ischemic conditioning protocols were implemented after the induction of anesthesia. Kidney function was assessed by renal failure, which was characterized by investigator-defined criteria (expanded RIFLE definition; Table 1). Renal failure was included in the composite primary end point and was also considered as a separate secondary end point at 30, 90, and 360 days after the index procedure. No significant differences regarding renal failure were shown.
In the same issue of the New England Journal of Medicine, the results of another large randomized controlled trial were published. The ERRICA study (Hausenloy et al 9 ), which was conducted in 30 cardiac surgery centers in the United Kingdom, evaluated a total of 1612 patients who underwent on-pump coronary artery bypass grafting (with or without concurrent heart valve surgery). Remote ischemic conditioning courses were conducted after anesthesia induction and prior to any surgical stimuli. Acute kidney injury (AKI) was evaluated only as a secondary outcome and was defined according to the Kidney Disease Improving Global Outcomes (KDIGO) 17 criteria. Additionally, neutrophil gelatinase-associated lipocalin (NGAL) was assessed, as an AKI surrogate, prior to the surgery and at 6-, 12-, and 24-week postsurgery. Neither AKI occurrence nor NGAL levels differed between the experimental and the control group (Table 1).
In contrast to these negative studies in the surgical setting (where the use of concurrent anesthetic/sedative medication may have interfered with RIC effects 18,19 ), a number of smaller, but randomized and controlled, studies in the setting of coronary angiography and percutaneous coronary intervention (PCI) or after contrast-enhanced computed tomography (CT) have provided more encouraging results.
Deftereos et al 4 randomized 225 patients with non-ST-elevation myocardial infarction (STEMI) undergoing PCI to receive RIC by cycles of inflation and deflation of the stent balloon or a sham procedure. The primary end point was AKI, defined as an increase of ≥0.5 mg/dL or ≥25% in serum creatinine within 96 hours from the index procedure. Acute kidney injury rate in the RIC group was 12.4% versus 29.5% in the control group (odds ratio: 0.34; 95% confidence interval: 0.16-0.71). The number needed to treat to avoid 1 case of AKI was 6. The 30-day rate of death or rehospitalization for any cause was 22.3% in the control group versus 12.4% in RIC patients (P = .05).
Crimi et al 20 reported in 95 patients with STEMI undergoing primary PCI (from the randomized RemPostCond study) that RIC by means of intermittent lower limb ischemia–reperfusion was not associated overall with lower postprocedural peak serum creatinine, but there was a significant interaction with preprocedural renal function—patients with baseline estimated glomerular filtration rate (eGFR) <77 mL/min/1.73 m2 were at significantly lower risk for renal dysfunction with RIC compared to controls. More encouraging were the findings of Yamanaka et al 21 who evaluated 94 patients with STEMI. Half of the patients underwent three 5-minute RIC cycles (at 5-minute intervals) on the upper arm (initiated with diagnosis and continued intraprocedurally and prior to reperfusion). The RIC group was found to be protected against contrast-induced AKI (10% vs 36%; P = .003). Additionally, in a recent study by Olafiranye et al, 22 it was shown that RIC applied in patients with STEMI during transportation from non-PCI-capable to PCI-capable centers had beneficial effects on postprocedural AKI occurrence (8.7% vs 18.5%; unadjusted odds ratio: 0.42, 95% confidence interval: 0.18-0.94; P = .036).
In a different clinical setting, that of elective coronary angiography, Er et al 23 randomized 100 patients with impaired renal function to standard care or standard care plus RIC by intermittent arm ischemia through 4 cycles of 5-minute inflation and 5-minute deflation of a blood pressure cuff. Acute kidney injury was defined as an increase in serum creatinine ≥25% or ≥0.5 mg/dL above baseline at 48 hours after contrast medium exposure. The incidence of AKI was 40% in the control group and 12% in the RIC group (odds ratio: 0.21; 95% confidence interval: 0.07-0.57). In search for the potential benefits of RIC in different patient risk profiles, Iragashi et al 24 studied 60 patients with moderately impaired renal function (eGFR between 30 and 60 mL/min/1.73 m2) who were scheduled for coronary angiography (and/or intervention). Acute kidney injury was defined by means of liver-type fatty acid-binding protein levels and was observed in 26.9% of the control group versus 7.7% of the RIC group (P = .038). With a focus on a more expanded patient subset in terms of underlying clinical conditions, Healy et al 25 evaluated, in a randomized fashion, 87 hospitalized patients who underwent abdominopelvic CT with intravenous contrast administration. Patients in the experimental arm received four 5-minute upper limb RIC cycles. Remote ischemic conditioning was found to be protective against a rise in serum creatinine (at a 48-hour follow-up) only in the subgroup with baseline eGFR less than 90 mL/min/1.73 m2.
Meta-Analyses
We performed a search in the PubMed database for published meta-analyses on the effect of RIC on renal function. We identified 17 meta-analyses that evaluated the effects of RIC on AKI in human participants (Sukkar et al, 26 Wang et al, 27 Sardar et al, 28 Zhang et al, 29 Zhou et al, 30 Hu et al, 31 Li et al, 32 Zhang et al, 10 Zhao et al, 33 Deng et al, 34 Zuo et al, 35 Bei et al, 36 Pei et al, 37 Niu et al, 38 Yang et al, 39 Li et al, 40 and Alreja et al 41 ), which are summarized in Table 2.
Recent Meta-Analyses of the Effects of Remote Ischemic Conditioning in Acute Kidney Injury.
Abbreviations: AAAr, abdominal aortic aneurysm repair; AKIN, Acute Kidney Injury Network; CAA, coronary artery angiography; CABG, coronary artery bypass grafting; CarSur, any cardiac surgery; CEA, carotid endarterectomy; CECT, contrast-enhanced computed tomography; ContrAdm, contrast administration; CPB, cardiopulmonary bypass; EVAR, endovascular aneurysm repair, KT, kidney transplantation; LT, liver transplantation; PCI, percutaneous coronary intervention; s-AVR, aortic valve replacement without CABG; Rad-Proc, radiological procedure (interventional or diagnostic); VascSur, vascular surgery; V-R, valve replacement.
aYear of e-publication or full publication; whichever comes first.
bAnalysis of the studies according to AKIN criteria reported in Supplementary Data of the published manuscript.
cOnly studies reporting AKI were included.
dSubstudies of studies with a total of >1000 patients.
It is of note that, although most of the meta-analyses involve several thousands of patients in total, the majority of included studies were relatively small. The results of these meta-analyses are evidently rather inconsistent, although it appears that the larger among them (ie, those including a larger number of participants) tend to show a renoprotective effect of RIC. Obviously, the evident variability in the underlying procedures on top of which RIC was applied and evaluated is an important source of inconsistency among analyses. A related issue and an important source of variability is the use of contrast agents in the procedures, and it appears that RIC may be more beneficial when utilized with procedures involving contrast administration (Table 2).
In the most recent meta-analysis (Sukkar et al 26 ), which included the largest population in aggregate (8493 participants), a significant reduction of AKI with RIC was observed (risk ratio: 0.83; 95% confidence interval: 0.71-0.97; P = .02). Still, the authors note that this finding is largely based on evidence of low quality, based on “allocation concealment (5 trials), blinding of outcome assessors (2 trials), selective outcome reporting (20 trials), and incomplete data reporting (19 trials).” Furthermore, in a supplemental analysis, wherein the authors tried to use a homogenized definition of AKI (with all problems and limitations that such a post hoc manipulation of data may entail), results were neutral in terms of benefit from RIC.
On the contrary, another recent meta-analysis, by Wang et al, 27 which evaluated 2077 patients in aggregate, is the first to report a negative effect of RIC on AKI (risk ratio: 1.53; 95% confidence interval: 1.12-2.10; P < .05). The authors note that this finding was only evident after the exclusion of a study by Zarbock et al, 42 which was identified as a cause of heterogeneity. Of note, the study by Zarbock et al 42 evaluated 240 patients and has been included in all other available meta-analyses published after 2015. Moreover, neither of the available key RCTs (RIPC, RIPHeart, ERRICA) have suggested any safety issues 43 ; therefore, the findings of Wang et al should be interpreted with caution.
Methodological Issues
A close look at the existing evidence on the renoprotective effects of RIC reveals that there is a large degree of variability and inconsistency, in terms of both study design and reported results. There are some evident methodological problems, which should be resolved if definitive answers are to be acquired as to the true clinical value of RIC.
The problem of acute kidney injury definition
While physicians are deeply aware and prepared to confront acute kidney function deterioration in the clinical setting, the lack of uniformity in the definition of this entity has been remarkable. 44 Predictably, this has been reflected in the numerous clinical registries and/or studies producing a “data conundrum.” Over the last decade, the following definitions of acute kidney injury (AKI) have been mostly utilized by clinicians and researchers—(1) Acute Dialysis Quality Initiative (RIFLE), 45 (2) Acute Kidney Injury Network, 46 (3) KDIGO, 17 and (4) a variety of investigator-defined criteria.
Study design issues
Inconsistency of observations between available studies has engendered a cycle of criticism. Observations from systematic reviews and meta-analyses imply that study outcomes could have been influenced by improper investigator blinding (eg, utilization of a deflated cuff instead of a “dummy arm”), by varying definitions of the research object (eg, investigator defined AKI) or even by study size and statistical power. In addition, in most of the published studies, AKI was not evaluated as a primary objective but has been reported as a secondary finding. Furthermore, while the majority of the studies evaluated the effects of preconditioning, a number of studies assessed the effects of post-, peri-, or combined RIC stimuli. This has, largely, not been taken into account in most meta-analyses, and data were analyzed as a whole. It is difficult to determine the significance of this issue, especially considering that in the case of kidneys, the timing of the “ischemic insult” is not always easy to pinpoint.
The “significant others” issue
Numerous studies have evaluated the effects of RIC under various settings (ie, cardiac surgery, 7 ∓9,42,47,48 liver 49 or kidney 50 transplantation, laparoscopic partial nephrectomy, 51 abdominal aortic aneurysm repair, 52 ∓54 PCI, 4,55,56 coronary arteriography, 57 and other radiologic procedures 58 ). Undoubtedly, this variety of clinical conditions may be an unpredictable confounding factor regarding kidney function. The nature of surgery (noncardiac vs cardiac), hydration status, perioperative medication (eg, anesthetics, vasopressors) regimens, and preexistence of chronic kidney disease may all be implicated with the occurrence of AKI. 44 Therefore, each and every different underlying condition should be individually assessed in terms of pathophysiology. For example, in a patient undergoing primary PCI for STEMI, both prerenal (potential hemodynamic instability) and renal (contrast media) factors converge synergistically to inflict kidney injury, while acute sympathetic and renin–angiotensin–aldosterone axis activation may also play an unfavorable role. Such a background does not apply to cardiac surgery or elective diagnostic procedures involving contrast media administration. Consequently, divergent study findings may be interpreted keeping in mind that “one size does not fit all” when it comes to the evaluation of RIC effects.
In addition, medication regimens per se (ie, anesthetics, analgesics, nitrates) have been reported to interfere with the potential favorable effects of ischemic conditioning. 6,19 Propofol, in particular, has been pinpointed as a potential source of confounding. Extensive criticism has been applied on the (almost) exclusive propofol utilization in 2 of the recent landmark trials (100% in RIPHeart, 90% in ERRICA) for anesthetic purposes. 43,59 To date, pathophysiologic mechanisms have been proposed to support the notion that propofol may both (1) attenuate the protective effects of RIC in the experimental group and (2) act protectively in the control group. 59 Of note, such data were available at a time when incorporation in the protocols of the 2 studies would have been theoretically possible. 43
Moreover, AKI is currently considered as a syndrome with a multivariable pathophysiologic substrate, rather than a single-factor-induced state. 60 Consequently, the findings of each study may not be easily extrapolated to other settings, while inclusion of such different results in a single meta-analysis should be done with caution. Even concomitant diseases (which, as a rule, are absent in animal experimental models and are sometimes inadequately reported in clinical studies) have been pinpointed as one of potential explanations for the problematic translation of experimental findings on RIC to the clinical setting. 61
Issues regarding RIC protocols
The diversity of the applied RIC protocols has been implicated in the contradictory findings between various clinical trials. 6 Τhe number of courses, the duration of each course, the duration of the interspersed reperfusion courses, the topography (eg arm, limb) and temporal variants (in relation to the index ischemia/reperfusion) of the remote ischemic stimulus are all parameters that are variably modified in the available experimental and clinical studies. Indeed, “hyperconditioning” (ie, application of “excessive” conditioning courses) could probably be deleterious. 3 On the other hand, the potential beneficial effects of daily ischemic conditioning sequences for a specific period of time are under investigation. 62 ∓64
Critical Perspective—Future Considerations
Despite the publication of 3 large randomized studies that showed no benefit from RIC, the possibility that RIC may have clinically significant renoprotective effects may not yet be dismissed, considering that all 3 studies were conducted in the same clinical setting, namely cardiac surgery, while several meta-analyses have shown benefit from RIC, especially in settings where contrast-induced nephropathy is the primary damaging insult. It is true that these meta-analyses have been based on studies with relatively small patient populations, while there are also issues regarding the RIC procedure per se, blinding, variability of the index procedure, comorbidities, and coadministered medications.
Taking existing evidence into consideration, there is a clear need for prospective randomized studies in clinical settings where RIC has performed the best. This is an important issue, considering that there is tantalizing evidence that conditioning may be effective under specific circumstances—most possibly in cases where specific potent noxious stimuli come into play. However, in our opinion, there is an even clearer need for a common clinical trial framework—or, to phrase it differently, a set of guidelines —for these kinds of studies, in a way more or less similar to other fields of cardiovascular research (eg, definitions of bleeding by research consortia 65 ).
Obviously, if the current modus operandi continues (characterized by a babel of definitions, procedures, outcomes, and goals), it is most likely that soon ischemic conditioning will be “yesterday’s news” with no definitive conclusions having been drawn in terms of its real clinical utility.
Footnotes
Authors Contributions
Georgios Giannopoulos and Dimitrios A. Vrachatis contributed equally to this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.