Ischemic heart disease (IHD), especially acute myocardial infarction, is one of the leading causes of morbidity and mortality of the world populations (1). It is generally believed that timely and successful restoration of blood supply to the ischemic area is the most effective method to rescue the ischemic myocardium and improve clinical outcomes (2). In clinical practice, moreover, various treatment strategies for restoration of blood supply to the ischemic myocardium, such as primary percutaneous coronary intervention or coronary artery bypass grafting (CABG), have been widely used. As a “double-edged sword,” however, reperfusion of blood flow to ischemic myocardium can further worsen damage in the ischemic myocardium and even cause more deaths of cardiomyocytes, which are even attributable to about 50% of final myocardial infarct size (3, 4). This phenomenon is called as myocardial ischemia/reperfusion injury (IRI). For decades, abundant studies have been conducted, but the physiopathology mechanism underlying myocardial IRI remains elusive. The available literatures indicate that oxidative stress, apoptosis, and inflammation are extensively involved in the development of myocardial IRI (3, 5). Unfortunately, as an intractable clinical issue, there have been still no effective interventions against myocardial IRI (6). Hence, developing reasonable and practical countermeasures to provide protection against myocardial IRI are extremely essential.

Sevoflurane, an inhaled anesthetic with good pharmacological properties of stable induction and rapid recovery, is extensively applied for anesthesia of cardiac surgery in clinical practice (7). Early in 2007, a meta-analysis of small randomized clinical trials comparing inhaled anesthesia and total intravenous anesthesia (TIVA) has showed that both the sevoflurane and desflurane are associated with significant reductions of myocardial infarction (2.4 vs. 5.1%) and mortality (0.4 vs. 1.6%) (8). Furthermore, a RCT demonstrated a decreased 1-year mortality after CABG with cardiopulmonary bypass (CPB) in patients receiving sevoflurane anesthesia compared with TIVA (17.8 vs. 24.8%) (9). Similarly, animal experiments have also demonstrated that sevoflurane conditionings can provide a protection against myocardial IRI and the underlying mechanisms involve antioxidant, antiapoptosis, and anti-inflammation pathways (10, 11). However, a multicenter study found no difference in the occurrence of myocardial ischemia when comparing sevoflurane vs. TIVA in patients undergoing noncardiac surgery at risk for perioperative myocardial ischemia (12). These inconsistent findings have caused confusion of available evidence about cardioprotective benefits of sevoflurane conditionings.

Undeniably, there are still many differences between the preclinical and clinical studies due to the complexity of clinical situations. Actually, poor reporting of preclinical study protocols can potentially lead to unreliable experimental results and unnecessary clinical trials (13, 14). In contrast, well-designed animal studies can provide both the safe and effective information on relevant factors that may improve the possibility of success in future clinical trials (15). Furthermore, a systematic reviews of preclinical animal data may partly explain the underlying mechanisms of clinical diseases and indicate the directions of clinical treatments (16). Thus, well-designed experimental studies may provide a deep insight into the cardioprotective efficacy of sevoflurane and useful information on relevant factors that may influence outcomes. However, it must be noted that previous preclinical studies regarding cardioprotection of sevoflurane against myocardial IRI are usually small sample size and vary considerably in the quality of representation, resulting in a low power of evidence. Thus, this comprehensive systematic review and meta-analysis were conducted to pool the findings of previous animal studies regarding the cardioprotective effects of sevoflurane conditionings. The main aims of this analysis are to provide a comprehensive evaluation on cardioprotective efficacy of sevoflurane conditionings in in-vivo animal models of myocardial IRI and attempt to explore the possible reasons for controversial clinical benefits.

The main findings of this comprehensive systematic review and meta-analysis are that compared with control treatment, both the SPreC and SPostC can provide a significant protection against myocardial IRI, shown by a significant infarct size sparing. Furthermore, the robust effect is comparable whether sevoflurane conditioning is carried out before the onset of ischemia, during ischemia, or at the time of reperfusion based on in-vivo data over 800 animals. In addition, the benefits of sevoflurane conditionings on myocardial IRI are not significantly affected by countries, species, duration of ischemia or reperfusion, and timing regimens of drug administration.

Myocardial IRI is a common and serious complication in clinical practice and has aroused extensive attention. Whether with the help of mechanical or pharmacological interventions, it is essential to reestablish coronary blood flow to limit myocardial damage following ischemia (55). The pathogenesis underlying myocardial IRI is extraordinarily complex and multifactorial. Among the etiological factors of myocardial IRI, oxidative stress is a vital mechanism, which initiates at the onset of reperfusion and triggers subsequent series of pathophysiological processes. Essentially, oxidative stress is characterized as severe imbalance between exaggerated reactive oxygen species (ROS) generation and inhibited antioxidant defense systems. Thus, ROS is considered as major determinants for adverse ventricular remodeling via promoting myocardial interstitial fibrosis, cardiomyocyte hypertrophy, and induction of cell death (56). Meanwhile, excessive oxidative stress causes mitochondrial permeability transition pore (mPTP) opening and cytochrome C releasing, subsequently triggering the intrinsic apoptosis by activation of caspase-9/3 signaling pathway (57). After reperfusion injury, oxidative stress and apoptosis collectively contribute to the substantial loss of cardiomyocyte and consequently enlarged infarct area. These pathological mechanisms provide the rationale for therapeutic strategies targeted against myocardial IRI.

As sevoflurane is an inhalational anesthetic most commonly used in clinical practice, its cardioprotection has attracted wide concern in the past decades. A recent meta-analysis elaborates the importance of various mechanisms involved in preconditioning and postconditioning of halogenated gases for cardioprotection (58). Several signaling molecules and pathways have been shown to be involved in sevoflurane-induced cardioprotection, but detailed mechanisms are not fully understood yet. To better understand the cardioprotection of sevoflurane conditionings, this analysis also included the in-vivo evidence regarding the potential mechanisms of sevoflurane conditioning reported in available literatures (Supplementary Table 2). Substantial literatures indicate that cardioprotection of sevoflurane conditionings is closely related to its antioxidant, antiapoptosis, and anti-inflammation properties. For example, Zhao et al. (10) established mice model of myocardial IRI in vivo and adult mouse cardiomyocytes model of simulated ischemia and reoxygenation (SI/R) in vitro and demonstrated that SPreC compared with control treatment decreased superoxide generation by 43.6%. Furthermore, this antioxidant property was largely retained in the AMP-activated protein kinase dominant-negative (AMPK-DN) mice, but completely abolished in caveolin-3 knockout (Cav-3KO) mice. These results suggest that SPreC-mediated attenuation of superoxide generation is highly dependent on Cav-3 and only partially dependent on the AMPK signaling axis. In addition, this study showed that SPreC compared with control treatment significantly decreased caspase-3 activity, indicated inhibition of cardiomyocyte apoptosis involved in the cardioprotection of SPreC. Previous study had identified that microRNAs (miRNAs) were important targets for regulating myocardial reperfusion damage and were the valuable biomarkers of myocardial injury (5). The study of Tan et al. (49) on a rat model of myocardial IRI in vivo demonstrated that SPostC compared with control treatment significantly decreased myocardial malondialdehyde level, but increased myocardial glutathione and superoxide dismutase activities, indicating that SPostC can inhibit oxidative stress in ischemia/reperfused myocardium by activating miR-203. Besides, this study also showed that SPostC decreased serum proinflammation factors and cardiomyocyte apoptosis. Similarly, Qi et al. (11) found that SPostC significantly inhibited oxidative stress and apoptosis to provide a protect against myocardial IRI by upregulating miR-145 expression and downregulating granzyme K expression.

It must be emphasized that several interesting phenomena about cardioprotection of SPreC and SPostC deserve special attention. Qiao et al. (47) demonstrated that cardioprotection of SPostC is age dependent, i.e., a significant cardioprotection is achieved in young animals but not in old ones. Available evidence indicates that this age-dependent cardioprotection of SPostC is at least associated with the inability to activate Akt and Erk1/2 in old animals (41). Furthermore, Obal et al. (18) found that a combination of SPreC with SPostC can provide an additive protection against myocardial IRI, which is partly mediated by opening mitochondrial adenosine triphosphate-dependent potassium (mK_ATP) channels. In addition, various risk factors of cardiovascular diseases, such as obesity, hyperlipidemia, and diabetes, can significantly attenuate or even abolish the cardioprotective effects of sevoflurane conditionings (26, 50, 59). As a main aim of this analysis was to determine the cardioprotective effects of sevoflurane conditionings, the preclinical studies performed on animals with the risk factors of cardiovascular diseases were specifically excluded from this analysis. Nevertheless, it is advisable to consider the risk factors of cardiovascular diseases when determining the optimum protocols of sevoflurane conditionings.

Although many in-vivo experiments have shown that sevoflurane conditionings can provide an infarct size sparing, clinical trials in humans have yielded variable results regarding cardioprotection of sevoflurane conditionings. Several studies demonstrate that sevoflurane has the potential benefits of cardioprotection in patients undergoing coronary artery bypass surgery when it is administrated to maintain anesthesia at 1 minimum alveolar concentration (MAC), shown by improved clinical results and myocardial injury biomarkers after surgery (60, 61). A recent meta-analysis reports that halogenated agents including desflurane, isoflurane, and sevoflurane can reduce the occurrence of myocardial infarction, mortality rate, and mechanical ventilation need after cardiac surgery, especially for the patients undergoing CABG surgery (62). However, a meta-analysis of 79 randomized controlled trials (RCTs) including 6,219 patients with noncardiac surgery reports that cardioprotective benefits of sevoflurane are inconclusive (63). It should be noted that this meta-analysis uses perioperative myocardial infarction or deaths as the main endpoint and no case occurs in any of the included studies. Similarly, another systematic review and meta-analysis indicate that the use of volatile anesthetics such as sevoflurane, desflurane, or isoflurane is associated with decreased mortality and perioperative complications in patients undergoing cardiac surgery, but does not improve these outcomes in patients with noncardiac surgery (64). It is still unclear why cardioprotection of sevoflurane is significantly different between patients with cardiac and noncardiac surgeries. Thus, further studies are needed to determine whether cardioprotection of sevoflurane is really surgery type-dependent and which factors can impair the transformation of cardioprotective benefits provided by sevoflurane into improved clinical outcomes of patients.

Tracing the potential sources responsible for heterogeneity in a meta-analysis can guide preclinical and clinical study designs. As our analysis showed that there were high levels of heterogeneity between the included studies, both the meta-regression and stratified analysis were used to explore the potential sources of heterogeneity and determine whether some of the predefined experimental variables in this analysis could influence the observed effect size with cardioprotective efficacy of sevoflurane conditionings. In fact, this method for analyzing heterogeneity has been successfully applied to several promising preclinical interventions (65, 66). Our stratified analysis showed that animal species (dog, rabbit, rat, and mice) did not produce significant effects on either effect size or heterogeneity of both the SPreC and SPostC groups. In addition, meta-regression suggested that country difference might be a source of significant heterogeneity for SPreC. Further stratified analysis by countries showed that there was still significant heterogeneity between the groups. Compared to other countries, moreover, Netherlands showed a large WMD, with a wide 95% CI. This may partly explain the source of heterogeneity. Anyway, it should be aware that preclinical characterization of sevoflurane administration protocol in-vivo studies has not yet been identified and current studies have showed uneven quality. All of these are essential to determine the important parameters that will facilitate optimization of administration protocol in further clinical trials.

Certainly, both the poor methodological quality and publication bias can result in over- or under-estimation of effect size (67, 68). To determine the internal validity of included studies in this analysis, both the quality of study and publication bias were evaluated by the reporting quality assessment tools, i.e., the ARRIVE guidelines 2.0 and a 12-item quality score. Our results showed that the overall scores of the reporting quality assessment tools for the included studies were high, indicating the validity of this analysis. In fact, main contents of experimental process, such as confirmation of ischemia and monitoring of ECG, were well reported in the included studies. However, a number of aspects required by the ARRIVE guidelines 2.0 were poorly reported in the most included studies, particularly in sample size calculations (11%) and blinding (16%). These may inevitably raise concerns of statistical validity and result in exclusion of literature selection. Furthermore, appropriate monitoring of experimental animals, such as PaO₂ or SaO₂ (27%), depth of anesthesia (22%), and concentration of sevoflurane (38%), was also reported poorly in the included studies. According to a position article on improving the preclinical assessment of novel cardioprotective therapies (69), these factors are actually essential to ensure a high-quality study. Thus, failure to report high-quality studies may partly account for the observed heterogeneity of this analysis. Nevertheless, the effect sizes by sevoflurane conditionings were consistent and robust despite high heterogeneity was observed.

Finally, publication bias within the included studies was visually assessed by a funnel plot and further detected by the Egger's and Begg's tests. Notably, publication bias was displayed in the funnel plot and identified by Egger's test (P < 0.01) for the SPostC group. Visual analysis of the funnel plot suggested that neutral or negative studies might be underrepresented in the SPostC group. That is true, studies with neutral or negative data are often not given priority for publication. Fortunately, it is reassuring that the bias of our analysis does not result in a statistically significant impact on the overall effect.

This meta-analysis has included all the studies, which met our rigorous inclusion criteria. Besides, the studies that are not published and do not meet important quality criteria are not be included in our analysis. As the validity of a meta-analysis is highly depended on the quality of all the included studies, there are several limitations in this analysis that deserve special attention. First, there is only a small amount of data from large animals in our analysis, though large animals share more anatomical and physiological characteristics of heart with human. This may limit the interpretation and extension of our results. It is warranted that large animal experiments are needed to further confirm the favorable effects of sevoflurane conditionings on myocardial IRI in small animal models. Second, as mentioned in the discussion, this analysis only includes the studies that are conducted on normal animals without any risk factors of cardiovascular diseases, such as aging, diabetes, obesity, hyperlipidemia, hypertension, and others. Evidently, our results cannot be generalized to the animals with risk factors of cardiovascular diseases. Third, besides country differences may be a source of heterogeneity for SPreC studies, meta-regression fails to reveal any influence of other prespecified covariates on pooled estimates of infarct size. However, robustness of the data is evidenced by both the sensitivity analysis and stratified analysis, which confirm the benefits and reliability of sevoflurane conditionings in ameliorating myocardial IRI. Thus, the real reasons for high degree of heterogeneity among the included studies are unclear. To address above issues and confirm therapeutic effect of sevoflurane conditionings on myocardial IRI, we believe that more animal studies and RCTs are still required.

This comprehensive systematic review and meta-analysis demonstrate that sevoflurane conditionings can provide a robust and highly reproducible infarct size-sparing effect in animal models of myocardial IRI. Furthermore, this beneficial effect can be obtained when sevoflurane conditioning is carried out before the onset of ischemia, during ischemia, or during reperfusion, despite a high heterogeneity among the included studies is observed. Nonetheless, preclinical studies with a high-quality design, especially for those conducted on large animal models and the animals with risk factors of cardiovascular diseases, are still required to future determine the protection of sevoflurane conditionings against myocardial IRI and explore relevant mechanisms.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

BH and F-SX contributed to the study design, planning, data analysis, and wrote the manuscript. TT, P-PH, W-CL, Y-GC, and T-YJ contributed to data analysis and revision of the manuscript. All authors had seen and approved the final version of the manuscript.

This study was performed with the financial support of the National Natural Science Foundation of China (No. 81470019 to F-SX).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.