It is well known that heart failure results in a significantly reduced quality of life, is a major societal burden, and has become a leading cause of mortality and morbidity [1,2,3]. Ischemic heart diseases account for two thirds of all cases of systolic heart failure [4]. Poor outcomes are determined by extensive myocardial remodeling and chamber enlargement, and currently there is lack of effective treatment [5,6,7]. Cardiac transplantation and mechanical circulatory support as destination therapy are high-risk therapeutic options that are limited by donor availability, patient eligibility, and costs [8].

Cell-based therapies have become the important paradigm-shifting alternatives [9,10,11]. However, clinical studies report the inconsistent efficacy of cell-based therapies, possibly ascribed to the unpredictable potency of cell products and their limited retention. Methods of cell therapy optimization include myocardial priming to improve cell homing, exploiting resident cell populations, and leveraging combined cell regimens [12,13,14,15]. Mesenchymal stem cells (MSCs) are reported to enhance cardioreparative functionality and induce a restorative response in failing hearts [16]. The Congestive Heart Failure Cardiopoietic Regenerative Therapy trial confirmed the efficacy and safety of cardiopoietic cells delivered via a retention-enhanced catheter for advanced symptomatic heart failure of ischemic etiology [17, 18].

However, the use of MSC treatment for heart failure has not been well established. Recently, several studies on the topic have been published, and the results have been conflicting [19,20,21,22]. Considering these inconsistent effects, we therefore conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) to evaluate the efficacy of MSC treatment in patients with heart failure.

Methods

Ethical approval and patient consent were not required since this was a systematic review and meta-analysis of previously published studies. The systematic review and meta-analysis were conducted and reported in adherence to PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses; [23]).

Search strategy and study selection

Two investigators independently searched the following databases (inception to May 2018): PubMed, Embase, and the Cochrane Register of Controlled Trials. The electronic search was performed using the following keywords: mesenchymal stem cell or MSC, and heart failure. We also checked the reference lists of the screened full-text studies to identify other potentially eligible trials.

The following selection criteria were applied: (a) population: patients with heart failure; (b) intervention: MSC treatment; (d) comparison: placebo treatment; and (d) study design: RCT. The exclusion criteria were acute coronary syndrome, valvular heart disease, and malignant tumor.

Data extraction and outcome measures

We used a piloted data-extraction sheet, which covered the following information: first author, number of patients, age, male, body mass index (BMI), New York Heart Association (NYHA) class, details of methods used in two groups. Data were extracted independently by two investigators, and discrepancies were resolved by consensus. We contacted the corresponding author to obtain data when necessary. No simplifications and assumptions were made.

The primary outcome was cardiovascular death. Secondary outcomes included left ventricular ejection fraction (LVEF), rehospitalization, myocardial infarction, the recurrence of heart failure, and total death.

Quality assessment in individual studies

The Jadad Scale was used to evaluate the methodological quality of each RCT included in this meta-analysis [24]. This scale consists of three evaluation elements: randomization (0–2 points), blinding (0–2 points), dropouts and withdrawals (0–1 points). One point is allocated to each element if they are mentioned in the article, and another one point is given if the methods of randomization and/or blinding are appropriately described. If the methods of randomization and/or blinding are inappropriate, or dropouts and withdrawals are not recorded, then one point is deducted. The Jadad Scale score varies from 0 to 5 points. An article with a Jadad score of ≤ 2 is considered to be of low quality. If the Jadad score is ≥ 3, the study is thought to be of high quality [25].

Statistical analysis

We estimated mean differences (MDs) with 95% confidence intervals (CIs) for continuous outcomes (LVEF), and risk ratios (RRs) with 95% CIs for dichotomous outcomes (cardiovascular death, rehospitalization, myocardial infarction, heart failure, and total death). Heterogeneity was tested using the Cochran Q statistic (p < 0.1) and quantified with the I2 statistic, which describes the variation of effect size that is attributable to heterogeneity across studies. An I2 value greater than 50% indicates significant heterogeneity. The value of the I2 statistic is used to select the appropriate pooling method: fixed-effects models are used for I2 < 50% and random-effects models for I2 > 50%. If significant heterogeneity was present, we searched for the potential sources of heterogeneity. Sensitivity analysis was performed to detect the influence of a single study on the overall estimate via omitting one study in turn when necessary. Owing to the limited number (<10) of included studies, publication bias was not assessed. Results were considered statistically significant at p < 0.05. All statistical analyses were performed using Review Manager Version 5.3 (The Cochrane Collaboration, Software Update, Oxford, UK).

Results

Literature search, study characteristics, and quality assessment

A detailed flowchart of the search and selection results is shown in Fig. 1. In total, 511 potentially relevant articles were identified initially. Finally, six RCTs that met our inclusion criteria were included in the meta-analysis [19,20,21,22, 26, 27].

Fig. 1
figure 1

Flowchart of study search and selection process. RCT randomized controlled trial

Full size image

The main characteristics of the six RCTs included in the meta-analysis are presented in Table 1. The six studies were published between 2013 and 2017, and sample sizes ranged from 30 to 371 with a total of 625. The approaches for MSC application included intravenous infusion, intracoronary transplantation, and intramyocardial injection.

Table 1 Characteristics of studies
Full size table

Among the six RCTs, three studies reported cardiovascular death [19, 20, 22], two studies reported LVEF [20, 21], three studies reported rehospitalization [20,21,22], two studies reported myocardial infarction [19, 20], two studies reported the recurrence of heart failure [20, 26], and five studies reported total death [19,20,21, 26, 27]. The Jadad scores of the six included studies vary from 3 to 5, and all six studies are considered to be of high quality according to the quality assessment.

Primary outcome: cardiovascular death

The primary outcome data were analyzed with the fixed-effects model. The pooled estimate of the three included RCTs suggested that compared with control group for heart failure, MSC intervention had no significant influence on cardiovascular death (RR = 0.76; 95% CI = 0.38–1.52; p = 0.43), with low heterogeneity among the studies (I2 = 32%, heterogeneity p = 0.23, Fig. 2).

Fig. 2
figure 2

Forest plot for the meta-analysis of cardiovascular death. M-H Mantel–Haenszel method, MSC mesenchymal stem cells

Full size image

Sensitivity analysis

Low heterogeneity was observed among the included studies regarding cardiovascular death. Thus, we did not perform sensitivity analysis by omitting one study in turn to detect the source of heterogeneity.

Secondary outcomes

Compared with the control group, MSC treatment was associated with significantly increased LVEF (MD = 9.64; 95% CI = 7.56–11.71; p < 0.00001; Fig. 3) and reduced rehospitalization rates (RR = 0.41; 95% CI = 0.23–0.73; p = 0.003; Fig. 4), but it had no significant impact on the incidence of myocardial infarction (RR = 0.72; 95% CI = 0.10–5.02; p = 0.74; Fig. 5), the recurrence of heart failure (RR = 0.88; 95% CI = 0.28–2.81; p = 0.83; Fig. 6), and total death (RR = 0.68; 95% CI = 0.37–1.25; p = 0.21; Fig. 7).

Fig. 3
figure 3

Forest plot for the meta-analysis of left ventricular ejection fraction (%). IV independent variable, MSC mesenchymal stem cells

Full size image
Fig. 4
figure 4

Forest plot for the meta-analysis of rehospitalization rate. M-H Mantel–Haenszel method, MSC mesenchymal stem cells

Full size image
Fig. 5
figure 5

Forest plot for the meta-analysis of myocardial infarction. M-H Mantel–Haenszel method, MSC mesenchymal stem cells

Full size image
Fig. 6
figure 6

Forest plot for the meta-analysis of the recurrence of heart failure. M‑H Mantel–Haenszel method, MSC mesenchymal stem cells

Full size image
Fig. 7
figure 7

Forest plot for the meta-analysis of total death. M‑H Mantel–Haenszel method, MSC mesenchymal stem cells

Full size image

Discussion

Stem cell therapy has been explored for the treatment for heart failure for more than a decade [20], but different stem cell populations and evaluation methods remain a challenge to fully understanding the efficacy of stem cell administration for clinical treatment [28]. Improvements in cardiac function and regeneration of damaged heart tissue are observed through various mechanisms including transdifferentiation, cell fusion, and paracrine modulation [29, 30]. Stem cell therapy is reported to be safe and to offer patients with heart failure moderate benefits in survival, left ventricular function, and quality of life [31, 32]. Cell-based therapy for chronic ischemic and nonischemic disease involves a range of cellular products and delivery routes including autologous or allogenic bone marrow mononuclear cells and MSC administered by intramyocardial injections, percutaneous intracoronary infusion, and exceptionally peripheral intravenous infusion [31, 33, 34].

Several mechanisms may explain the clinical benefit of MSCs treatment for patients with heart failure, including reductions in myocardial cell apoptosis, modulation of inflammation, myocardial fibrosis, neovascularization, and increased cell differentiation [35]. Incorporation of MSCs into tissues involves multiple processes consisting of cell recruitment, migration, and adhesion [36]. Umbilical cord MSCs have a high migration capacity and have shown a good response to serum in heart failure patients, thus this cell type might sense biological cues mediating the therapeutic effect by systemic delivery [20]. Our meta-analysis suggests that compared with control interventions for heart failure, MSC treatment is associated with significantly improved LVEF and reduced rehospitalization rates, but with no significant influence on cardiovascular death.

Clinical trials have validated the safety of MSC-based therapies, with no increase observed in acute infusion toxicity, organ system complications, infection, death, or malignancy in treated patients [37, 38]. A phase 2 study confirmed the safety of intravenous administration of allogeneic MSCs (up to 5 × 106 cells/kg) in patients with acute myocardial infarction, as evidenced by no increase in adverse event rate and a decrease in hospitalization rate and arrhythmic events at the 6‑month follow-up [38]. In addition, the safety of the intravenous administration of ischemia-tolerant allogeneic MSCs has been evaluated in patients with nonischemic cardiomyopathy, and the results revealed no increase in death, hospitalizations, and serious adverse events at the 90-day follow-up [39]. There is no statistically significant difference in the incidence of myocardial infarction, the recurrence of heart failure, and total death between MSC treatment and control intervention for heart failure patients based on the results of this meta-analysis.

Limitations

This meta-analysis has several potential limitations that should be taken into account. First, our analysis is based on only six RCTs and five of them have a modest sample size (n < 100). Overestimation of the treatment effect is more likely in smaller trials compared with larger samples. Next, the trials involved allogeneic and autologous MSCs as well as umbilical cord MSCs administered by intravenous infusion, intracoronary transplantation, and intramyocardial injection in included RCTs, and these different cellular products and delivery routes may have some impact on the pooled results. Finally, the optimal cell sources and doses as well as the delivery routes remain undefined and future studies should focus on these issues.

Practical conclusion

MSC treatment can provide some benefits for patients with heart failure.