Extra-Corporeal Membrane Oxygenation (ECMO) for Postcardiotomy Cardiogenic Shock (PCCS) in Asian Patients: A Systematic Review and Meta-Analysis

Background Extra-corporeal membrane oxygenation (ECMO) is an important extracorporeal life support system to treat patients with postcardiotomy cardiogenic shock (PCCS). But its effectiveness and safety are still inconclusive. Existing systematic reviews and meta-analysis have heterogeneity challenges such as the inclusion of different races in the same study. Objective The impact of ECMO treatment on the survival rate of Asian adult patients with PCCS was evaluated by searching the literature and using the method of systematic review and meta-analysis. Study Design and Methods PubMed, Web of Science, Cochrane Library, EMBASE, CNKI, WANFANG MED ONLINE, SinoMed were searched to nd relevant research on the use of ECMO on PCCS patients in Asia. Outcomes included survival rate to hospital discharge, long-term survival rate, complications. Results A total of 32 articles were selected, including observational studies and one RCT, involving China and Japan covering 4,278 PCCS patients. The pooled rate of survival to hospital discharge was 43.0% (95% CI 38% ~47%, I 2 = 87%), the pooled 1-year, 2-year, and 3-year survival rate were 34% (95% CI 26% ~ 42%, I 2 = 85%), 29% (95% CI 18% ~39%, I 2 = 93%), 25% (95%CI 16% ~35%, I 2 =93%). The pooled rate of bleeding, neurologic complications, rethoracotomy, leg ischemia, renal failure, renal replacement therapy, and infection were 18% (95% CI 13% ~24%, I 2 =91%), 13% (95% CI 9% ~17%, I 2 = 90.0%), 36% (95% CI 27% ~46%, I 2 = 97.0%), 11% (95% CI 9% ~13%, I 2 = 55%), 40% (95% CI 27% ~54%, I 2 = 90%), 45% (95% CI 35% ~ 56%, I 2 = 98%), 26% (95%CI (20%~31%, I 2 =95%). increased complications. In addition, we found these systematic reviews and meta-analyses were faced with heterogeneity challenges, for they pooled patients of different races together. We present hereby a systematic review and meta-analysis that attempt to evaluate the ecacy of ECMO for Asian PCCS patients.


Abstract
Background Extra-corporeal membrane oxygenation (ECMO) is an important extracorporeal life support system to treat patients with postcardiotomy cardiogenic shock (PCCS).But its effectiveness and safety are still inconclusive.Existing systematic reviews and meta-analysis have heterogeneity challenges such as the inclusion of different races in the same study.
Objective The impact of ECMO treatment on the survival rate of Asian adult patients with PCCS was evaluated by searching the literature and using the method of systematic review and meta-analysis.
Study Design and Methods PubMed, Web of Science, Cochrane Library, EMBASE, CNKI, WANFANG MED ONLINE, SinoMed were searched to nd relevant research on the use of ECMO on PCCS patients in Asia.Outcomes included survival rate to hospital discharge, long-term survival rate, complications.
Results A total of 32 articles were selected, including observational studies and one RCT, involving China and Japan covering 4,278 PCCS patients.
Conclusion ECMO can provide survival bene ts for Asian adult patients with PCCS.However, attention should be paid to reducing the incidence of complications, especially renal failure.More high-quality clinical studies are needed to con rm the survival bene t of ECMO.
1 Background Postcardiotomy cardiogenic shock (PCCS) is a severe complication after cardiac surgery.Intra-aortic balloon pump (IABP) and inotropic drugs are conventional treatment for these patients, but some patients still experience low cardiac function and insu cient cardiac output with the above supports.For these patients, effective mechanical circulation support (MCS) measures must be taken to prevent death.Extra-corporeal membrane oxygenation (ECMO) is an important mechanical circulation life support system to treat patients who do not respond well to high-dose vasoactive drugs and IABP.In recent years, studies have shown that about 1% of patients need ECMO due to cardiogenic shock after cardiac surgery [1][2][3] .Hu RTC et al [3] retrospectively analyzed adults treated with veno-arterial ECMO (VA-ECMO) for PCCS between 2001 and 2016 in Australia, they found that VA-ECMO improved survival rate, but morbidity was also increased.Lehmann S et al. [4] evaluated the clinical outcomes, 30 day and long-term follow-up in patients undergoing ECMO for refractory PCCS (n = 1040) and concluded that ECMO is an acceptable option for patients with PCCS and is justi ed by good long-term outcomes of hospital survivors.
Researchers also conducted systematic reviews and meta-analyses to evaluate the e cacy of ECMO for PCCS patients.However, their conclusions are inconsistence.
Khorsandi M et al. [5] including 24 retrospective cohort studies, concluded that VA-ECMO could provide survival bene ts for patients with refractory PCCS.Harahwa TE et al. [6] found that the ECMO could improve the survival rate of PCCS patients for whom positive inotropic therapy was ineffective.This systematic review also found that without ECMO support, the death risk of patients would be increased.On the contrary, a systematic review by Wang L et al. [7] found that neither the short-term nor mid-term survival rates of PCCS patients treated with ECMO improved, accompanied with increased complications.In addition, we found these systematic reviews and meta-analyses were faced with heterogeneity challenges, for example, they pooled patients of different races together.We present hereby a systematic review and meta-analysis that attempt to evaluate the e cacy of ECMO for Asian PCCS patients.

Objective
The impact of ECMO treatment on survival rate of Asian adult patients with PCCS was systematically reviewed, and the meta-analysis was used when data are available.
Speci c search strategies are shown in Table 1.

Selection of Studies and Data Extraction
Two researchers (LD, YT) independently carried out literature retrieve, screening, data extraction and quality evaluation.If there is a disagreement, it is resolved through discussion.The data extraction includes the baseline data of the patients included in the study, the number of IABP assistance, the type and incidence of complications.

Quality Assessment
The modi ed Newcastle-Ottawa scale was used to evaluate the quality of observational studies, [7] the Modi ed Jadad's Scores Scale was used to evaluate the quality of RCT [8] .

Statistical Analyses
Data analysis was performed using R version 4.0.3software (the R Foundation) with the package of "meta" [9] .The results were expressed as pooled proportions (%) with 95% con dence interval (95% CI).Statistical heterogeneity was evaluated using the I 2 test (I 2 50% was seen a high degree of statistically signi cant heterogeneity).Random-effect models were used because these are mainly observational studies.Publication bias was evaluated using the Egger test.Sensitivity analysis was performed to examine the stability of the results.A p value 0.05 was considered to be statistically signi cant.

Search Results and Study Characteristics
Through literature retrieval using the strategy we reported above, six-hundred and twenty-ve records were found.After screening and evaluation, thirty-two articles were nally included for systematic review and meta-analysis  . The litrature retrieval and screening process is shown in Fig. 1.

Publication Bias
To assess publication bias, Egger's test was performed (Table 5).Survival rate to hospital discharge (Egger's p = 0.2809) and renal replacement therapy (Egger's p = 0.9895) have low possibility of publication bias, bleeding, neurological complications, rethoracotomy, leg ischemia and infection have higher publication bias.

Sensitivity Analyses
In order to assess the stability of the results, we conducted sensitivity analyses.The omission of each study of the following outcomes showed no signi cant difference, con rming the stability of the results: survival rate to hospital discharge, 2-year survival rate, 3-year survival rate, bleeding, rethoracotomy, renal replacement therapy, infection.In the 1-year survival rate and neurological complications, the omission of the Chen SW study showed a signi cant difference, the omission of the Wu F study in leg ischemia showed a signi cant difference, and the omission of the Hsu PS study in renal failure showed a signi cant difference.The results of sensitivity analysis are shown in Supplementary 2.

Discussion
Cardiogenic shock is a serious complication after cardiac surgery.For patients who do not respond well to inotropic drugs or IABP support, mechanical circulatory support should be used in order to reduce mortality rate.ECMO is one of the extracorporeal life support systems, and relative research started in the 1960s [42,43] .The application of ECMO for cardiogenic shock after cardiac surgery have emerged since 1980s [44] .
The difference of these outcomes between our study and previous studies could be the result of racial differences.Unlike other studies, our study only included Asian populations.Race can have an impact on the e cacy of the use of ECMO.Vallabhajosyula S et al. [45] conducted a retrospective cohort to evaluate the e cacy of ECMO for acute myocardial infarction (AMI) between 2000 and 2014 in the USA.One of the conclusions was that compared with whites, non-whites were related with higher in-hospital mortality with ECMO patients.One of the reasons was that non-whites often have insu cient insurance and come from lower socio-economic status.Our study included research mainly from China and Japan, in which ECMO treatment can only be carried out in the economically developed areas with high medical standards.Patients in these areas are more likely to have insurance and higher socio-economic status.Thus, our conclusion can have practical signi cance for PCCS patients in Asian with regard to the use of ECMO.
In this study, it is found that renal failure is the most common complication in all morbidity, and the rate of renal replacement therapy is quite high.This conclusion is similar to previous meta-analysis.Research on the timing of the application of RRT has found that early RRT treatment for PCCS patients with ECMO is bene cial to the recovery and improves the prognosis [26] [28] .
The subgroup analysis was performed on the studies with the number of patients ≥ 50, and the pooled discharge survival rate was 45% (95% CI 40%-50%, I 2 = 90%).It is speculated that as the sample size increases, the medical staff will become more pro cient in the indications and operating speci cations of the ECMO technology, thereby improving the survival rate of discharge from the hospital.However, the discharge survival rate of study by Chen SW [25] , which sample size was 1137, was signi cantly lower than the pooled discharge survival rate (38% VS. 43%).The possible reason was that the study included patients between 2000 and 2011, with a long time span.In the early stage, the immaturity of the technology and the inappropriate selection of indications for PCCS with ECMO may reduce the survival rate.At the same time, we found that the two systematic reviews by Khorsandi M et al. [5] and Wang L et al. [7] were mostly enrolled patients before 2015, and there were many studies from the 1990s (45.83% and 35%).Comparing with these two systematic reviews, most studies enrolled by our systematic review were conducted after 2015.It can reasonably conclude that with the development of technology and the improvement of medical staff's skills, the outcome of patients de nitely will be improved.
The study by Wu F et al. [21] showed the lowest survival rate (16%) and the study of Zhang BW et al. [26] showed the highest survival rate (81%).Wu F et al. [21] included patients after valvular surgery, which meant these patients had abnormal heart structures or other injuries.Hou JF et al. [15] indicated that ECMO was better for patients with PCCS after heart transplantation.The PCCS of these patients mainly caused by cold ischemia and ischemia-reperfusion injury in the donor heart, rather than structural abnormalities or other injuries.Zhang BW et al. [26] included patients, whose treatment time with ECMO were over 48h.In the early stage of ECMO application (within 12 hours), if the patient's hemodynamics is not improved, death is very likely to occur.The study may arti cially exclude potentially dying patients, resulting in a higher survival rate.
There are some limitations to this study.First of all, the included studies are mostly observational studies, and the level of evidence are not high, and there is a high risk of selection bias and publication bias.The methodological quality of the included studies was evaluated, and 17 studies were of high quality.The publication of outcomes such as bleeding, neurological complications, rethoracotomy, leg ischemia, and infection are likely to be biased.Second, there are differences in the characteristics of patients in the included studies.Three of the studies included mainly older patients [11] [18] [31]   ; the patients included in the studies all used IABP and ECMO at the same time [15] [37] ; two studies included patients treated with renal replacement therapy [28] .The underlying diseases of the patients are also different.In the Hsu PS study [40] , 31% of patients had renal insu ciency before surgery, which was higher than that reported in other studies.The omission of this study in the sensitivity analysis showed the renal failure rate was signi cantly reducing in pooled result, which indicated those who suffered from renal dysfunction before operation should be under intensive monitoring.In addition, there are differences in the surgical methods adopted among enrolled studies.For example, we found that omitted the valvular surgery patients [21] from pooled result led to reduce of heterogeneity and leg ischemia occurrence.The number of patients included is also quite different.The minimum number of patients included is 12 [37] , and the maximum number of patients is 1137 [25] , which is far higher than the other studies.In the sensitivity analysis, the omissions of the study with 1137 patients about 1-year survival rate and neurological complications showed greatly decreased in I 2 (62.9%VS. 85%, 66% VS.90%).As we mentioned above, this largest study contained many patients in early stage of ECMO development that re ex the learning curve and development curve of this technology, and cannot represent current status.

Conclusion
ECMO can provide survival bene ts for Asian adult patients with PCCS, including short-term and long-term survival.However, attention should be paid to reducing the incidence of complications, especially renal failure.More high-quality clinical studies are needed to con rm these bene ts.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Figures Figure 1 Flow
Figures

Table 2
Summary of Detailed Information about Included Studies

Table 3
Modi ed Jadad's Scores Scale for RCT