Rational Route of Delivery Mesenchymal Stem Cell Therapy for Acute Myocardial Infarction (AMI) and Chronic Ischemic Cardiomyopathy (ICM): a Systematic Review and Meta-analysis

Background: Recent studies suggest that mesenchymal stem cells (MSCs) may have therapeutic potential for both acute myocardial infarction (AMI) and chronic ischemic cardiomyopathy (ICM). However, the rational route of delivery MSC therapy has not reached consensus. We performed a systematic review of clinical trials evaluating the rational route of delivery MSCs for AMI or ICM. Methods: Databases including Embase, PubMed, and Cochrane Central Register of Controlled Trials were searched from inception to February 2021. Studies that examined the use of MSCs in adults with AMI or ICM were eligible. Bias of included studies were assessed by the Cochrane risk of bias tool. The primary outcome was cardiac function assessed by left ventricular ejection fraction (LVEF) and the secondary outcome was cardiac remodeling which was assessed by left ventricular end-systolic volume (LVESV) and left ventricular end-diastolic volume (LVEDV), we also explored the safety between different routes. Results: 18 studies fullled eligibility criteria, which consist of 11 studies evaluated AMI and 7 studies evaluated ICM. In AMI group, only when patients received intracoronary infusion(IC) can improve LVEF (SMD 0.88, 95% CI 0.64-1.12), and there was a decrease in LVEDV&LVESV when administered IC or intravenous infusion (IV). While in ICM group, no signicant difference in LVEF was noted no matter administered which route, and transendocardial stem cell injection(TESI) seems to be effective in decreasing LVEDV&LVESV. Safety appeared no difference between different routes. Conclusions: Results from our systematic review suggest that intracoronary infusion seems more effective for MSC’s delivery in AMI group, while in ICM group, TESI better. .Recent

region of the heart, which means that nding an optimal route can lead to better clinical outcomes for cell therapy. Thus, we try to explore the optimal route of Mesenchymal Stem Cell(MSC) delivery.

Search strategy
We searched for trials in databases including PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL) registry of the Cochrane Collaboration until February 2021. The following search terms were used alone or in combination: mesenchymal stromal cells, mesenchymal stem cells, MSCs, mesenchymal stem cell transplantation, multipotent mesenchymal stromal cells, bone marrow mesenchymal stem cells, adipose-derived mesenchymal stem cells, Wharton's Jelly Cells, Myocardial Infarction, Myocardial Ischemia, Cardiovascular Stroke, Heart Attack, Ischemic Heart Disease. Two review authors identi ed all studies independently. The search was limited to controlled studies with a comparator arm. No language limit was applied. Any disagreement relating to the eligibility of a particular study was settled through discussion with senior reviewers.

Eligibility criteria
Inclusion criteria for this analysis were as follows: (1) Only unmodi ed MSCs were included, any pretreated, genetically engineered, or transfected cell were excluded; (2)The resources(bone marrow, umbilical cords, or adipose tissue) and administration route were unrestricted; (3) studies that were conducted in patients who had Ischemic cardiomyopathy; (4) Both cell groups and control arms should receive standard therapy, the control arm did not receive stem cells.
Data extraction and quality assessment A standardized data extraction form was used to extract data from the included studies. Information extracted included: study setting (publication year and country); demographics and baseline characteristics of the study population; details of the intervention and control conditions, such as cell delivery route, resource, dose, time interval between PCI to injection, change in LVEF, LVESV, LVEDV, follow-up duration; information for assessment of the risk of bias. Data was extracted independently by two review authors, discrepancies identi ed were settled by discussion with senior reviewers.

Summary measures
Change in LVEF is our primary outcome measures, many methods were used measuring LVEF(% EF), echocardiography data were used in preference unless MRI data were available. Change in LVEDV and LVESV were used as secondary outcome measures. In addition, we explore safety within different routes.

Method for meta-analysis
The data were analysed using Stata 16.0 (Stata Corporation, College Station, TX).The mean difference (MD) was used for continuous variables, while Forest plots were used to present the results of our metaanalysis. When studies that contained more than two treatment arms, only control and MSC groups were analyzed Values pooled together using mean, standard deviation, and size for studies conducting more than one experimental group containing MSCs. P < 0.05 was considered statistically signi cant and twosided 95%CI were reported throughout the study. Chi-square-based Q test and I2 tests were used to assess the quantity of heterogeneity among these studies. When I2 < 50 %, p > 0.1, the pieces of evidences were thought to be acceptable heterogeneity, the xed-effects model would be used. Otherwise, the randomeffects model was applied.
Using a sensitivity analysis, signi cant heterogeneity was found in the trials reported by Wang [22] and Gao et al [18] . I 2 decreased from 95.7% to 68.7% when the data from Wang [22] and Gao et al. [18] were excluded.
Secondary outcomes--cardiac remodeling LVEDV & LVESV A total of 7 AMI studies and 4 ICD studies provided data on LVEDV. Patients who underwent intravenous or intracoronary decreased LVEDV signi cantly in AMI clinical trials. While in ICD clinical trials, the route included in was only TESI way, which limited our assessment of other pathways ,and TESI displayed an effective role on reducing cardiac remodeling (SMD -0.23,95% CI -0.43--0.03). The above results were also appeared in LVESV.
Tertiary Outcomes--safety Mortality 9 AMI studies and 5 ICM studies reported mortality. Patients no matter AMI or ICM showed no difference in mortality(RR 0.90,95%CI 0.56-1.43), the same result was observed no matter administered which route. Sever adverse event 3 AMI studies and 4 ICM studies reported Sever adverse event. No signi cant difference in the risk of mortality between MSC and control groups, no matter in which group or administered which route. Interestingly, there was seems a trend to reduce sever adverse event when TESI was applied in ICM group, though it was not statistically signi cant.
Rehospitalization 6 AMI studies and 3 ICM studies provided data on readmission. In these trials, AMI group included IC and IV way, while only TESI is incorporate in ICM group, which makes it di cult to analysis the optimal route for reducing readmission between AMI or ICM group. Under such limited condition, we found that TESI showed a subtle advantage on reducing rehospitalization(SMD 0.6,95%CI 0.37-0.96).
Other outcomes WMSI In AMI group, WMSI was more signi cantly decreased when use IC than TESI injection, and there were no signi cant differences between MSC-and placebo-treated patients when applied IV injection(data not shown).WMSI was rarely reported in ICD group, thus we can't know its real e cacy.

6min-walk
A total of 4 studies reported 6min-walk in ICD group, there were no signi cant differences between MSCand placebo-treated patients when use DI or TESI way. 6min-walk was rarely reported in AMI group and it limit our conclusion.
No comparisons are made between other outcomes like infarct size, myocardial perfusion, NYHA class, quality of life due to its insu cient reports in studies included.

Risk of Bias Assessment
Two studies met all seven criteria for low risk of bias. Two studies ful lled six of seven risk of bias criteria. Seven studies described a low risk of bias in randomization procedures. Four studies underwent allocation concealment with low risk of bias. For Double-blinding procedures, six studies met low risk of bias. One study had a high risk of bias due to incomplete outcome data reporting and three study had an unclear risk of bias for selective reporting. None of the studies were considered to be at high risk for other biases.

Discussion
During the past decades, mesenchymal stem cell(MSC) have been extensively used in various forms of restorative and preventive medicine, which have shown con icting and inconclusive effects in ischemic cardiomyopathy, reasons for this owing to vast differences within various clinical trials, like patient pro les, cell phenotypes, dosing, routes of delivery, study endpoints and design, making it challenge to gure out an optimal intervention. Our systematic review try to assess the optimal route of MSC therapy for AMI and ICM. We major focused our review on cardiac performance and cardiac remodeling, and we found that In AMI group, only when patients received intracoronary infusion(IC) can improve LVEF (SMD 0.88, 95% CI 0.64-1.12), and there was a decrease in LVEDV&LVESV when administered IC or intravenous infusion (IV). While in ICM group, no signi cant difference in LVEF was noted no matter administered which route, and transendocardial stem cell injection(TESI) seems to be effective in decreasing LVEDV&LVESV. Safety appeared no difference between different routes. Results from our systematic review suggest that intracoronary infusion seems more effective for MSC's delivery in AMI group, while in ICM group, TESI better.
There is a large heterogeneity in our study, I 2 decreased from 95.7% to 68.7% when the data from Wang [22] and Gao et al. [18] were excluded. When we tried to gure out the source of heterogeneity in these studies, we found that these trials were different from other studies in the transplantation timing and dosage. The cell dosage used in most studies ranged from 10 6~1 0 7 , which was the rational cell dosage suggested from Zi Wang [33] 's meta-analysis, while Wang et al. [22] used 2X10 8 cells, this may be one of the reasons why this study had little effect on the improvement of LVEF. The time interval from PCI to injection in most studies is 15~30 days, Gao et al. [18] only used 5 days, the best transplantation timing reported by Zi Wang [33] , this may can explain why the improvement in cardiac function from this studies was more noticeable.
The nding above is different from kanelidis [34] et al. study, one potential explanation for these results is that only 6 clinical trials were included, which makes it di cult to make rm conclusions. In addition to including more studies, we took chronic ischemic myopathy patients into account, and it turn out to be a completely different result on cardiac function and cardiac remodeling when compared with AMI clinical trials, which seems to indicate distinct mechanisms between both AMI and ICD when administered different route of MSC delivery.
Intramyocardial injection(IM) consists of DI and TESI. DI requires directly injecting stem cells into and around the infarcted area of the heart under thoracotomy, which would increase risks for complications, morbidity and mortality. TESI means injecting stem cells into the myocardium directly through the endocardium, which is a minimally invasive surgery. All studies included use the NOGA system (Biosense Webster) for imaging. IV refers to the cells infused from peripheral vein, cells will migrate toward the injured myocardium depending on preponderance of physiological homing signals. IC means delivering MSCs through the coronary vessel to the infarcted myocardial regions. The route of delivery makes a major in uence on grafting e ciency and survival of transplanted cells in the infarcted region of the heart.
For patients with acute myocardial infarction, Forest VF et al. [35] reported a signi cant cell fraction retained within the heart after intracoronary injection, whereas no cardiac homing was observed in IV group. In addition, study from Hayashi M [36] showed that the survival of BMCs in the infarcted area was signi cantly higher in the IM group than in the IV groups. Therefore, IV route means the least cell implantation and retention. In theory, retention of BMCs in the IM group should be higher than IC group [37] , however, Fukushima S et al. [38] found that the IM and IC groups showed a similar survival of donor cell, which may due to the harsh microenvironment in infarcted heart. The hypoxic-ischemic and in ammatory microenvironments in the ischemic myocardium lowers survival and viability of MSCs [39] ,besides, exosomes from injured cardiomyocytes accelerates transplanted bone marrow mesenchymal stem cells injury [40] . Distinct cell clusters were found from immunohistochemistry when stem cell injection through intramyocardial, containing donor-derived cells and accumulated host-derived in ammatory cells in the infarct border zone [41,42] showed that the cells were engulfed in macrophages that had in ltrated the injection areas. In contrary, IC cell injection provided more homogeneous donor cell dissemination with less in ammation and without disrupting the native myocardial structure. [38] ,indicating that IC cell injection results in less mechanical injury or biochemical stress to donor cells than IM cell injection does.
For patients with chronic ischemic myopathy, mesenchymal stem cell therapy has no signi cant effect on increasing cardiac function probably due to different pathophysiological mechanism from this vulnerable population. MSCs are thought to work primarily because of their anti-in ammatory effects, so it is perhaps not surprising that MSCs had little impact in chronic disease such as ICM, where the level of in ammatory burden is much lower. In the aspect of cardial remodeling, exosomes derived from Mesenchymal Stem Cells can promote Fibroblast-to-Myo broblast differentiation in In ammatory Environments and bene t cardioprotective effects [14] .Interestingly, our studies found that TESI used in ICM group show a tendency in reducing mortality when compared to DI, which may due to its minimally invasive.

Limitations
Of note, the studies we included mainly focused on small randomized controlled studies, there are few study compared all the route directly, limiting our conclusions. Therefore, further experimental and clinical studies, which contain large-scale, rigorously conducted, randomized, adequately powered, placebocontrolled, blinded designs with outcome measures based on clinically relevant markers, are required to explore direct comparison of transmission pathways.
Apart from that, outcomes were not available in all studies included, like adverse reactions, readmission rates, which led us can not compare the safety of the different routes more comprehensively. We hope that studies in the future can provide more data on safety and provide longer follow-up data so that we can understand MSC therapy's safety over time.

Conclusions
Our study reveal that in AMI clinical trials, IC appears to be the most rational route of delivery due to its reduction in cardiac remodeling and improvement of LVEF, while in ICD clinical trials, none of any route can affect cardial function, but TESI can reduce cardial remodeling. Availability of data and materials Available data in this study exist in the above clinical trials.

Competing interests
The authors declare that they have no competing interests. Authors' contributions HYT and LZ contributed to the conception and design of the study. YXB and CSQ were responsible for the collection of data. HYT and LWZ performed the statistical analysis and manuscript preparation. GK,LZS,XML and SYW were responsible for checking the data. All authors were responsible for the drafting of the manuscript and read and approved the nal version. Figure 1 Impact of route of delivery MSC on LVEF. Forest plot of standardized mean difference(SMD) on LVEF compared with control. IC intracoronary, TESI transendocardial stem cell injection, IV intravenous, DI directly intramyocardia, CI con dence interval Figure 2