The Protective Effects of Myocardial Ischemic Postconditioning on Cardiac Surgry: A Systematic Review and Meta-Analysis of Rats and Clinical Trials

This is an open-access article, published by Evidence Based Communications (EBC). This work is licensed under the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium or format for any lawful purpose.To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. From Department of Anesthesiology, Renmin Hospital of Wuhan University, Wuhan, China.


I
schemic heart disease remains a leading cause of postoperative morbidity and mortality in the whole world.The definitive treatment for myocardial ischemia is reperfusion.However, ischemic-reperfusion (IR) injury occurs when ischemic myocardium is reperfused with oxygen and substrate rich blood, which causes additional reversible and irreversible damage to the myocardium (1).Ischemic postconditioning (IPC) is a method whereby brief cycles of ischemia-reperfusion immediately after a sustained occlusion of a coronary artery, which was firstly proposed by Zhao in 2003, and provides cardiac protection as ischemic preconditioning.It limits the sudden, unexpected IR injury, and is more applicable than ischemic preconditioning in clinical practice (2,3).
Over the past years, the role of IPC was evaluated in animal cardiac IR models, and it has been shown that IPC is a potent way to protect myocardium from IR injury.However, the role of IPC in animal models with pre-existing disease such as diabetes mellitus and hyperlipidemia is not clear-cut.Most of studies reported that the protective effects of IPC were abolished due to the underlying diseases (4)(5)(6)(7)(8), on the contrary, some scholars hold the opinion that preexisting disease such as hyperlipidemia does not prevent the cardioprotection brought by IPC (9).In clinical practice, IPC is wildly performed in patients undergoing primary percutaneous coronary intervention (PCI).Meanwhile, IPC, which is comprised of repeated episodes of IR by declamping and reclamping the aortic, also is fulfilled in patients undergoing cardiac surgery.It is an attractive strategy due to its simplicity, cost-effectiveness, and ease of performance.However, until now, IPC has not yet been translated universally into cardiac surgery.
Aiming to assess the protective effects of myocardial IPC on cardiac surgery and provide a new perspective for clinical practice, we performed a unique systematic review and metaanalysis with inclusion of several relevant studies in animal and clinical respectively.

MATERIALS AND METHODS
The review and meta-analysis was carried out in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement (10).

Search Strategy
Electronic literature databases including PubMed and Web of Science were searched.The last day of the literature search was on June 1, 2015, without any language limitation.The Medical Subject Heading (MeSH) terms were used in combination: "ischemic postconditioning" AND "heart" OR "myocardium".Hand research was also performed and the reference lists of all the retrieved articles were screened.

Inclusion and Exclusion Criteria
Inclusion of the studies was done by two of the researchers.We included studies to assess the role of IPC in rat models and clinical trials.
Researches on rat models were included in the systematic review if they fulfilled all of the following criteria: 1) the study assessed the effect of local IPC on myocardial IR injury; 2) the study was performed in animal hearts in vivo or in vitro; 3) the rats pre-existed with diabetes mellitus or hyperlipidemia or not; 4) the study was an original full paper which presented unique data.Studies were excluded if 1) the aim of the study was pharmacological or remote postconditioning; 2) studies were performed in animals rather than rats; 3) rats pre-existed with other diseases such as hyphenation; or 4) studies were performed only in genetically modified rats.
Clinical trials including the review met all of the following criteria: 1) randomized controlled trials (RCTs); 2) trials compared local IPC with IR absence of other protective measures.Trials were excluded if 1) patients were treated with pharmacological or remote IPC; 2) Patients underwent non-cardiac thoracic surgery such as PCI; or 3) they were case-only studies or case reports.

Data Extraction
Data were extracted if raw data or group averages, standard deviation (SD) or standard error (SE), and number of subjects per group (N) were reported, or could be recalculated.In animal studies, if a study contained separate groups for each gender, or several postconditioning protocols, these groups were analyzed as if they were separate studies.If data were presented only graphically, we used digital image analysis software to obtain the numerical values.Data extracted from the papers included: name of the first author, publication year, species of animal, number of subjects in IPC and IR groups, the method of reperfusion, the protocol of IPC, the design of the trials, the years of patients, the type of surgery, and definition of end points.The percentage of myocardium infarction area and apoptosis index were extracted in rat studies, and postoperative mortality, cardiac troponin I (cTnI), inotropes scores, time of ventilation and intensive care unit (ICU) stay were extracted in clinical trials, respectively.

Assessment of Methodological Quality
According to CAMARADES (11), we designed a 14-point rating system to assess the methodological quality of the included animal publications.It comprises 1) research question specified and clear, 2) animals randomized across groups, 3) outcome assessment randomized across group, 4) concealment of allocation, 5) group characteristics clearly described, 6) group characteristics described as equal, 7) correct control group used, 8) body temperature controlled within 3℃ variation, 9) blinded assessment of outcome, 10) IR treatment protocol clearly described, 11) IPC treatment protocol clearly described, 12) number of animals per group clear, 13) complete outcome data, 14) compliance with animal welfare regulations.
Clinical trials were evaluated using the guidelines recommended by the Cochrane Collaboration.(12) The risk of bias was evaluated in six categories: 1) randomization and sequence generation, 2) allocation concealment, 3) blinding method in performance and, 4) in outcome assessment, 5) incomplete outcome data, 6) selective outcome reporting.Every category was appraised according to three rulings: low risk, unclear risk, and high risk.
All data were reviewed by two people independently, and final inclusion of studies was determined by consensus.

Data Synthesis and Statistical Analyses
Data were analyzed using Review Manager 5.3 (The Cochrane Collaboration, Copenhagen).Meta-analysis of animal studies was performed for the outcome measured myocardium infarct size and apoptosis index; and in clinical trials, outcome measured postoperative mortality, cT-nI, inotropes scores, time of ventilation and ICU stay were analyzed.Standardized mean differences (SMD) and the corresponding 95% confidence intervals (CI) were calculated for continuous outcome data using a random effects model.In the case of dichotomous outcome data, risk ratios (RR) and their corresponding 95% CIs were calculated and analyzed using a random effects model.
To account for anticipated heterogeneity, we used random effect models in which some heterogeneity is permitted.Heterogeneity was investigated by means of subgroup analyses, which were used to evaluate the influence of variables on IPC efficacy, and meanwhile, to explore possible causes for heterogeneity.In animal studies, three subgrouping variables were postconditioning protocol (in vivo or langendorff), animal species (Sprague-Dawley [SD] or Wistar), and preexisting disease (diabetes mellitus or hyperglycemia).We assessed the possibility of publication bias by visually evaluating the possible asymmetry in funnel plots.

Study Characteristics
Table 1 and table 2 summarized the characteristics of all the included studies in rats and clinical respectively.
In all 37 animal studies, 3 researches (4-6) used diabetic rats, 4 researches (7-9, 43) focused on hyperlipidemia rats, and the residual 30 studies did research in animal without pre-existing disease.SD rats were used in 44% of the 37 included studies, and other researchers chose Wi- star rats as their study subjects.Male rats (89% ) were favored, 1 study (35) employed female rats, 1 trial (18) studied both male and female rats and 1 trial (33) studied rats without sexual limitation.Langendorff reperfusion system was used in 41% of the 37 included studies, and most of the studies employed in vivo reperfusion.Ischemia was performed by occluding left anterior descending coronary artery (LAD, 61% ), coronary artery (CA, 6% ) or aortic root (33% ).Ischemia lasted 20-45 minutes with a 30minute-24-hour reperfusion.IPC was fulfilled by 1-20 cycles of brief myocardium ischemia and reperfusion.The effects of IPC were assessed by myocardium infarction size and apoptosis index.
7 randomized clinical trials were included in this meta-analysis, comprising 421 participants.3 studies (44,45,50) included 203 patients underwent surgery for tetralogy of Fallo, 3 studies (47-49) included 130 patients underwent congenitally malformed heart repayment surgery, and 1 study (46) included 79 patients underwent coronary artery bypass grafting.5 trials (44,45,(48)(49)(50) studied the effects of IPC applied in patients less than 18 years old, including 1 study (48) focused on 1-5 years old chil-dren; and the patients in the other 2 clinical trials (46,47) were adults (more than 18 years old).IPC was implemented by 2 or 3 cycles of 30 seconds ischemia and 30 seconds reperfusion via reclamping and declamping the aortic.Data to evaluate the effect of IPC on postoperative cardiac function (cTnI, inotrope scores) and prognosis of surgery (mortality, time of ventilation and ICU stay) were collected.

Methodological Quality of Studies
The results of the quality assessment of the 37 animal studies included in this systematic review were shown in figure 2. Studies reported 9 out of 14 characteristics on average.Almost 90% of the studies reported randomization of the animals across treatment groups, yet no study reported randomization for outcome measurement across groups, neither did concealment of allocation.Although body temperature was important, only 30% of the studies reported whether the body temperature of the animals was controlled.
As shown in figure 3, all of the 7 clinical trials provided the methods to generate randomization sequence.As to the allocation concealment, only 1 study (45) but 3 trials (44,46,47) did not mention that.IPC was performed after the anesthesia, which made the blinding of patients easily, so the performance bias was low.Because the outcomes of the studies were objective indexes, so detection bias was low.
As the outcome of meta-analysis in myocardium infarct size, the role of IPC made no differences in different rat species and reperfusion protocols.Subgroup analysis showed that in healthy rats IPC reduced apoptosis index (SMD -2.66, 95% CI -3.35 to -1.97, P<0.0001).Two trials (8,43) studied apoptosis index in hyperlipdemia rat models, however, there was no significant difference between IR and IPC hyperlipidemia rat models (SMD -0.31, 95% CI -1.13 to 0.50, P=0.46).What's more, the heterogeneity was lower when the studies were grouped based on pre-existing disease.

Time of Ventilation
There were six trials (44-49) that recorded the time of ventilation after surgery.Half of the trials (47-49) came from the same team held negative outcomes.According to the analysis, IPC seemed to significantly shorten the time of ventilation (SMD -0.57, 95% CI -0.87 to -0.26, P= 0.0002), and the overall study heterogeneity was a little high (53%) (Figure 9).After excluded one trial (46), study heterogeneity decreased obviously (19% ).The trial excluded was done by Westerners, while other trials were done by Chinese.We speculated that the overall study heterogeneity was induced by the difference in human species.

Time of ICU Stay
Six studies (44)(45)(46)(47)(48)(49) reported the time of ICU stay, and different outcomes from different trials were extracted.In the analysis, IPC reduced the time of ICU stay significantly (SMD -0.57, 95% CI -0.84 to -0.30, P<0.0001) (Figure10).funnel plots for all outcome measures.The analysis revealed that there was obvious publication bias in animal studies; while for clinical trials, funnel plot analysis showed low risk of publication bias.

Sensitivity Analysis
Sensitivity analysis was undertaken if the overall or subgroup group studies heterogeneity was 50% .In the analysis of myocardium infarct size and apoptosis index, we performed sensitivity    analysis after subgroups were divided, however, this did not significantly alter the outcome of any of the outcome measures.Sensitivity analysis was also conducted in cTnI, and similar result was got.

DISCUSSION
We did a systematic review and meta-analysis for the protective effect of IPC in both animal studies and clinical trials respectively, for the first time.
In the meta-analysis of animal studies, we found that IPC showed a beneficial effect in reduction of myocardial injury or damage in healthy rats when compared with IR alone, however, the myocardium protection was decreased or abolished in diabetic or hyperlipidemia rat  (18) concluded that in langendorff perfused rat hearts, none of the IPC protocols improved myocardial tolerance to ischemia reperfusion injury nor decreased infarct size, while in vivo IPC did confer protection.The outcomes in diabetic rat models were consistent with each other primarily, but the data from studies of hyperlipidemia rat models appeared discordantly.Outcomes from 2 of the studies (9,43) showed that IPC was positive in protecting myocardium from IR injury in hyperlipidemia rat models, as well as in the rat models without pre-existing disease.Another 2 studies (7, 8) implied that the protective effect of IPC was abolished in hyperlipidemia rat models.In addition, we found the role of IPC has no difference in different rat species and myocardial perfusion protocols.
After a continuous ischemia, sudden reperfusion causes myocardial injuries by multiple mechanisms: generation of reactive oxygen species (ROS), endothelium damage, loss of calcium homeostasis, and so on.IPC contributes to decreasing these injuries.It involves initiators (adenosine, bradykinin, opioids) and their receptors coupled to G proteins.Many complex signaling pathways are involved, such as the reperfusion-induced salvage kinase pathway, protein kinase C and the Janus kinase-signal transducers and activator of transcription (JAK-STAT) pathway.Effectors are mainly represented by the mPTP, the mKATP and ROS (5,9,42).According to several relevant studies, IPC seems to lose its cardioprotective effect in subjects with diabetes and hyperlipidemia, while the underlying mechanism is largely unknown.At present, the accumulation of ROS led by aggravated oxidative stress is thought to account for the inefficiency of IPC in diabetes and hyperlipidemia rats models (4,6,8,42,43).
In our meta-analysis of animal trials, high heterogeneity was caused by multiple factors, such as pre-existing disease, rat species, reperfusion protocols, and so on.Our analysis indicated that the heterogeneity in myocardium apoptosis index was aroused by pre-existing disease, and subgroup analysis based on pre-existing disease dropped the heterogeneity between studies.
The evidence from the present meta-analysis of clinical trials showed that IPC provides myocardial protection after cardiac surgery, and it is also beneficial to postoperative prognosis.In recent decades, IPC has been translated successfully to intervention therapy, especially in patients with acute myocardium infarction for PCI therapy.Evidence from those clinical trials strongly supported that IPC played a key role in myocardial protection in ST-elevation myocardial infarction patients who underwent PCI (51)(52)(53).Nevertheless, negative results also exist.Freixa et al. (54) found that IPC lacks effects on infarct size in PCI, so did Dwyer et al. (55).
Clinical trials of IPC performed on cardiac surgery such as repairing of Tetralogy of Fallo and valve replacement are rare.In those trials, IPC protocol consisted of two or three 30 seconds cycles of clamping the ascending aorta, followed by 30 seconds of declamping just before the cessation of cardiopulmonary bypass (47)(48)(49)(50).But clamping/decamping the ascending aorta was lined with a high risk of embolic accidents, especially in adults with atherosclerotic lesions of the aortic wall.It is the reason for paucity of clinical trials in cardiac surgery, perhaps (56).
The evidence from our meta-analysis in both animal and clinical trials regarding cardioprotection and prognosis of cardiac surgery suggested significant benefit of IPC, however, the accuracy of the outcomes were weakened by rare trials and small sample size, especially in clinical trials.Furthermore, there were several shortcomings in this meta-analysis.Firstly, the analytical results were affected by the reviewers, although we attempted to overcome this drawback.Secondly, because of incomplete raw data and publication limitations, some related studies could not be included, which reduce the available sample further.So the results of the present metaanalysis should be considered attentively.

CONCLUSION
In animal studies, IPC provided cardiac protection in healthy rat models, but the cardioprotective effect was decreased or abolished in diabetic or hyperlipidemia rat models.In clinical trials, patients with IPC during cardiac surgery 10 seconds R & 10 seconds I 6 cycles of 10 seconds R & 10 seconds I 5 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I 6 cycles of 10 seconds R & 10 seconds I 2/3/6 cycles of 10 seconds R & 10 seconds I 6 cycles of 10 seconds R &10 seconds I 4 cycles of 15 seconds R & 15 seconds I 3 cycles of 30 seconds R & 30 seconds I 4 cycles of 20 seconds R & 20 seconds I 6 cycles of 10 seconds R & 10 seconds I 6 cycles of 10 seconds R & 10 seconds I 6 cycles of 10 seconds R & 10 seconds I 3 cycles of 30 seconds R & 30 seconds I 4 cycles of 30 seconds R & 30 seconds I 3 cycles of 10 seconds R & 10 seconds I 1 cycles of 2 minutes R & 2 minutes I 3 cycles of 30 seconds R & 30 seconds I 3 cycles of 10 seconds R & 10 seconds I 6 cycles of 10 seconds R & 10 seconds I 3 cycles of 30 seconds R & 30 seconds I 3 cycles of 30 seconds R & 30 seconds I 4 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 30 seconds R & 30 seconds I 6 cycles of 20 seconds R & 20 seconds I 6 cycles of 15 seconds R & 15 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 30 seconds R & 30 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I 3 cycles of 10 seconds R & 10 seconds I End points

Figure 3 .
Figure 3. Quality Assessment of Clinical Trials Included in Meta-Analysis.

Figure 4 .
Figure 4. Forest Plot of the Animal Studies that Analyzed Myocardium Infarction Size.

Figure 6 .
Figure 6.Forest Plot of the Clinical Studies that Analyzed Mortality.

Figure 9 .
Figure 9. Forest Plot of the Clinical Studies that Analyzed Time of Ventilation.

Figure 1. Screening Process of Studies Included in Meta-Analysis.
mentioned sealed envelopes,

Forest Plot of the Clinical Studies that Analyzed Inotrope Score.
Protective Effects of Myocardial Ischemic Postconditioning on Cardiac Surgry models.Most of the studies involvd in this review demonstrated that IPC had cardioprotective effect on both vivo and vitro healthy rat hearts.Lee et al.