Can cell-based therapies bridge the gap between research and reality in the treatment of myocarditis? A systematic review and meta-analysis

Figures

• 

  Fig. 1. Characteristics of included studies: quantity of published studies by year and the contribution of each country (a); by cell administration methods and cell types (b); by disease model and stage (c); and by animal species and four follow-up periods (d). EHT was generated from neonatal rat cardiomyocytes (5 × 106 cells) embedded in a three-dimensional extracellular matrix composed of type I collagen and Matrigel, cast into ring-shaped constructs and matured in vitro prior to transplantation [83]. iACS were constructed by differentiating syngeneic adipose tissue-derived stromal vascular fraction cells on temperature-responsive culture dishes, allowing spontaneous sheet detachment prior to implantation in rats with EAM [80]. QQMNCs + F-sheets consisted of Quality- and Quantity-controlled peripheral blood mononuclear cells co-cultured with syngeneic dermal fibroblasts on temperature-responsive dishes to form scaffold-free, multilayered cell sheets for transplantation [102]. AMI - acute myocardial injury; BMM2 - bone marrow derived M2-like macrophages; BMNCs - bone marrow mononuclear cells; CBTs - cell-based therapies; CDCs - cardiosphere derived cells, CMs - cardiomyocytes; CVB3 -…

• 

  Fig. 2. Risk of bias (RoB) assessment by SYstematic Review Centre for Laboratory animal Experimentation (SYRCLE) risk-of-bias tool for animal intervention studies. Six domains of bias were evaluated in each study. Almost all included studies demonstrated either unclear or high risk in the areas of Selection and Performance bias. Only one paper by Jin B. et al., 2010 [79] described sequence generation, while 5 out of 60 studies provided detailed baseline characteristics. Studies were heterogeneous in terms of the other four areas of bias: Detection, Attrition, Reporting, and Other sources of bias.

• 

  Fig. 3. Quality assessment across studies by the Collaborative Approach to Meta-Analysis and Review of Animal Data from Experimental Studies (CAMARADES) checklist. Ten criteria were used to assess the quality of the studies. Overall, the majority of studies (52 out of 60) were of moderate quality, rated between 4 and 6 points, while the remaining 8 studies demonstrated high (4 studies) and low (4 studies) quality. None of the studies reported sample size calculations or took measures to prevent publication bias.

• 

  Fig. 4. Effect of cell-based therapies (CBTs) on left ventricular fractional shortening (LVFS). CBTs improved LVFS (%) by 7.17 [5,8,99,100]. Despite significant heterogeneity, the 95% confidence interval (CI) of most studies did not cross the midline. The sample size varied from 3 to 17 animals in the treatment and control groups.

• 

  Fig. 5. Effect of cell-based therapies (CBTs) on left ventricular ejection fraction (LVEF). CBTs improved LVEF (%) by 9.00 [7.03, 10.97] with significant study heterogeneity I2 = 86%. Almost 2/3 of the studies showed statistically significant results as the corresponding 95% confidence intervals did not intersect the middle zero line.

• 

  Fig. 6. Effect of cell-based therapies (CBTs) on inflammatory cell infiltration rate (ICIR) and capillary density (CD). ICIR (a) and CD (b) were driven by CBTs towards a better pattern, as shown by the figures of −178.99 [95% CI: −225.91, −132.08] and 300.50 [95% CI: 45.01, 555.99], respectively, indicating fewer inflammatory cells and more capillaries per millimetre square of myocardium. As with the other results of interest, there was considerable heterogeneity.

• 

  Fig. 7. Effect of cell-based therapies (CBTs) on fibrotic area (FA). A similar positive therapeutic effect was observed during FA evaluation as well. CBTs could reduce FA up to 6% (−6.04 [95% CI: −6.83, −5.25]) compared with control groups. Heterogeneity was as high as 97%.

• 

  Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flowchart.

• 

  Subgroup analysis of left ventricular fractional shortening results by animal species.

• 

  Subgroup analysis of left ventricular fractional shortening results by disease stage.

• 

  Subgroup analysis of left ventricular fractional shortening results by disease model.

• 

  Subgroup analysis of left ventricular fractional shortening results by cell administration methods.