Table 1.
Characteristics of included systematic reviews.
| Study ID | Population | Databases and date of search | Year range of included studies | Number of studies included with study design | Type of stem cells | Type of control | Outcomes | Risk of bias of included studies | Risk of bias tool used | Grade | Publication bias | Key findings |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bernardi 201918 | Individuals with refractory CD | PubMed, ScienceDirect, Jan 2008–Dec 2018 | 2008–2018 | Thirteen RCTs (1 on luminal CD, 12 on perianal fistulizing CD) | Adipose-derived MSCs | Standard treatments, various other treatments | Healing of fistulas, symptom improvement | Not assessed | Not assessed | Not assessed | Not assessed | Promising results with MSCs in healing perianal fistulas; need for more studies on luminal CD; variability in dosage and administration methods |
| Cao 201719 | Patients diagnosed with CD | PubMed, Web of Science, up to Sep 30, 2016 | 2009–2016 | Fourteen RCTs | ASCs, BM-MSCs | Placebo or standard of care | Overall healing rate, clinical response, AEs | Low to moderate | NOS | Not assessed | Not assessed | MSCs are effective for Crohn’s fistula; CDAI baseline is a candidate for evaluating effectiveness |
| Cao 202120 | Patients with Crohn’s fistula | PubMed, Cochrane Library, EMBASE, Jun 2005–Aug 2020 | 2005–2020 | Twenty-nine studies (RCTs and cohort studies) | Adipose-derived and BM-MSCs | Placebo, fibrin glue | Healing rate, AEs, CDAI, PDAI, IBDQ, CRP | Moderate to high | NOS | Not assessed | Not assessed | MSCs show a higher healing rate (61.75%) vs. placebo (40.46%) a lower incidence of AEs; the optimal dose is identified as 3 × 10^7 cells/ml |
| Cheng 201921 | Patients with perianal CD | PubMed, Cochrane Library, EMBASE, CNKI, up to Oct 2018 | 2005–2018 | Thirteen studies (five RCTs, eight nonrandomized experimental) | Autologous and allogeneic MSCs from adipose tissue and bone marrow | Placebo, fibrin glue, standard care | Fistula healing, clinical response, AEs | Moderate to high | Cochrane risk-of-bias tool | Not assessed | Funnel plots indicate no publication bias | Local MSC therapy is safe and effective; higher healing rates with autologous MSCs and size-based dosing |
| Cheng 20207 | Patients with complex perianal fistulas (either of cryptoglandular origin or associated with CD) | PubMed and EMBASE, up to Mar 2020 | 2009–2020 | Seven RCTs | Autologous and allogeneic MSCs | Fibrin glue, saline solution | Healing rate, AEs, re-epithelialization | Moderate to high | Cochrane risk-of-bias tool | Not assessed | Funnel plot indicates no publication bias | Local MSC therapy is safe and efficacious for complex perianal fistulas; significant long-term efficacy; no significant difference in AEs |
| Cheng 202322 | Patients with perianal CD | PubMed, EMBASE, Cochrane Library, up to Mar 2022 | 2009–2022 | Six RCTs | Autologous and allogeneic MSCs | Saline solution, fibrin glue | Healing rate, AEs | Moderate to high | Cochrane risk-of-bias tool | Not assessed | Funnel plot indicates no publication bias | Local MSC injection is safe and efficacious for perianal fistulas in CD; significant long-term efficacy; no significant difference in AEs |
| Choi 201923 | Patients with complex perianal fistulas (CD and non-CD) | PubMed, EMBASE, Cochrane Library, up to Aug 2017 | 2005–2017 | Sixteen studies (3 RCTs, 13 non-RCTs) | Autologous and allogeneic MSCs | Conventional surgical methods | Healing rate, AEs, re-epithelialization | Moderate to high | MINORS | Not assessed | Funnel plot and Orwin’s fail-safe N indicate possible publication bias | Stem cell therapy is effective for complex perianal fistulas; higher healing rates with autologous MSCs; further large-scale RCTs needed |
| Ciccocioppo 201924 | Patients with CD or cryptoglandular fistulas | MEDLINE, EMBASE, Web of Science, Cochrane, CINAHL, ClinicalTrials.gov, May 2017 | 2003–2017 | Twenty-three studies (4 RCTs, 10 one-arm trials, 7 observational) | Autologous and allogeneic MSCs from adipose tissue and bone marrow | Placebo, fibrin glue, standard care | Fistula closure, radiological healing, AEs | Moderate to high | Cochrane risk-of-bias assessment instrument | Not assessed | Funnel plots indicate no publication bias | 80% fistula closure in MSC-treated patients; 64% in MSC vs. 37% in control in RCTs; low incidence of treatment-related AEs |
| Dave 201525 | Patients with IBD, including CD and UC | PubMed (since inception to Mar 2015), EMBASE (since inception to Nov 2014) | 2009–2013 | Twelve studies (RCTs and observational) | MSCs from various sources including BM, adipose tissue, and UC | Placebo, standard of care, fibrin glue | Healing of perianal fistulas, clinical remission, AEs | High | Cochrane risk of bias tool | Not assessed | Possible publication bias indicated by funnel plot | MSCs show promise in treating IBD with healing of perianal fistulas and induction of clinical remission; challenges include cost and characterization |
| El-Nakeep 202226 | Patients with medically refractory CD | MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), ClinicalTrials.gov, WHO ICTRP; up to Mar 19, 2021 | 2009–2020 | Seven RCTs | HSCs and MSCs | Standard of care, placebo | Clinical remission, CDAI <150 at 24 weeks, fistula closure short-term, fistula closure long-term, total AEs, SAEs, withdrawal due to AEs | Moderate to high | Cochrane risk-of-bias tool | Low to very low certainty | Possible publication bias indicated by funnel plot | SCT shows uncertain effects on clinical remission and CDAI <150 at 24 weeks; beneficial for fistula closure short and long-term; likely increases SAEs |
| Ko 202127 | Patients with IBD, including CD and UC | PubMed, from inception to Oct 29, 2020 | 2016–2022 | Thirty-two studies | MSCs | Placebo, standard of care | Healing of perianal fistulas, clinical remission, AEs | Moderate to high | Not assessed | Not assessed | Not assessed | Local MSC injections for PFCD support long-term efficacy and safety; mixed evidence for systemic MSC infusion in luminal IBD due to methodological heterogeneity |
| Lee 201728 | Patients with Crohn’s anal fistula | MEDLINE, EMBASE, Cochrane Library, Jan 1995–Mar 2016 | 1995–2016 | Thirty-nine retrospective, 16 prospective cohorts, 5 open-label, 3 RCTs | MSC, ASC | Various surgical interventions including setons, advancement flaps, fistula plugs | Fistula healing rate | Overall high risk of bias | Cochrane ROBINS-I and ROB tool | Not assessed | Not assessed | Surgical interventions for Crohn’s anal fistula are heterogeneous with high bias. Standardization needed for better understanding of treatment options |
| Lei Ye 201629 | CD patients, age ≥18, refractory to or unsuitable for current therapies | Cochrane Library, PubMed, Medline, EMBASE, ISI Web of Knowledge, ClinicalTrials.gov, up to Sep 2015 | 2007–2015 | Eighteen articles (six clinical trials with HSCs, 12 with MSCs) | Autologous/Allogeneic MSCs, HSCs | Self-control or placebo controls using fibrin glue or routine therapies | Clinical remission, endoscopic response, perianal fistulas healing/closure, SAEs | Moderate to high | Cochrane ROB-1, NOS | Not assessed | Not assessed | MSCs reduce CDAI and alleviate CD symptoms; low incidence of SAEs |
| Li 202330 | Patients with perianal fistulizing CD | PubMed, Cochrane Library, EMBASE; Mar 2022 | 2016–2022 | Five RCTs | MSCs | Placebo | Efficacy (remission), safety (TEAEs, perianal abscess, proctalgia) | High | Cochrane risk of bias tool | Not assessed | Low possibility of publication bias indicated by symmetrical funnel plot | MSCs treatment leads to definite remission (OR 2.06, P<.0001); no significant increase in TEAEs, perianal abscess, or proctalgia |
| Lightner 201831 | Patients with perianal CD | PubMed, Cochrane Library Central Register of Controlled Trials, EMBASE; Jan 1, 2003 - Oct 31, 2017 | 2003–2017 | Eleven studies (phase I, II, III trials) | MSCs (autologous and allogeneic) | Placebo, fibrin glue, no treatment | Safety and efficacy of MSCs, AEs, SAEs, fistula healing rates | Moderate | Cochrane Collaboration’s risk of bias tool for RCTs, NOS | Not assessed | Not assessed | MSCs improve healing rates for perianal CD; no significant increase in AEs or SAEs; higher healing rates with MSCs vs. conventional treatments |
| Narang 201632 | Adults with cryptoglandular fistula in ano | MEDLINE (PubMed and Ovid), EMBASE (Ovid), Cochrane Library, 2007–2014 | 2007–2014 | Twenty-one articles (two RCTs, rest observational) | Not specified | Placebo or no treatment | Fistula closure rate, complications | Moderate to high | MINORS | Not assessed | Not assessed | New techniques are in early stages, with difficult-to-reproduce results and lacking long-term data. No clear evidence currently favors any specific technique |
| Qiu 201733 | Patients with active CD | PubMed, Cochrane Library CENTRAL, EMBASE; initial search Feb 5, 2015; updated Oct 15, 2016 | 2002–2016 | Twenty-one studies (RCTs and observational) | HSCs and MSCs (both autologous and allogeneic) | Various, including placebo and standard of care | Clinical response, clinical remission, fistula healing, endoscopic remission, SAEs, recurrence | Varied, mostly moderate to high | Cochrane risk of bias tool for RCTs, NOS | Not assessed | Egger test indicates publication bias exists for clinical response but not for fistula healing | Stem cell therapy potentially effective for refractory CD; high efficacy in inducing fistula healing; toxicity is a significant barrier |
| Qiu 202434 | Adult patients with medically refractory CD or CD-related fistula | PubMed, CENTER (Cochrane Library), EMBASE (Ovid); up to 5 Sep 2023 | 2009–2023 | Twelve RCTs | ADSCs, BM-MSCs, HSCs, placenta-derived cells, UC-MSCs | Placebo, no treatment | Clinical remission, SAEs | Varied, mostly moderate to high | Cochrane risk of bias tool (ROB 2.0) | Moderate certainty | Minimal risk of publication bias detected | SCT significantly increases likelihood of CR vs. placebo/no treatment; not associated with higher likelihood of SAEs |
| Wang 202335 | Patients with complex perianal fistulas of cryptoglandular or CD origin | PubMed, EMBASE, Cochrane Library database, US ClinicalTrials.gov; up to May 15, 2022 | 2009–2020 | Six clinical trials, 10 publications | ASCs, BSCs | Placebo, fibrin glue, saline solution, surgery | Healing rates (HR), safety, efficacy, AEs, SAEs, recurrence, re-epithelialization | Low to high | Cochrane risk of bias tool | Not assessed | No publication bias detected | MSCs therapy superior to conventional treatment in short, long, and over-long-term follow-up; no statistical difference in medium-term efficacy; both autologous and allogeneic MSCs effective |
AEs, adverse events; ASC, adipose-derived stem cell; BM-MSC, bone marrow-derived mesenchymal stem cell; CD, Crohn’s disease; CDAI, Crohn’s Disease Activity Index; CRP, C-reactive protein; HSC, hematopoietic stem cell; IBDQ, Inflammatory Bowel Disease Questionnaire; ICTRP, International Clinical Trials Registry Platform; MSC, mesenchymal stem cell; NOS, Newcastle–Ottawa Scale; PDAI, Perianal Disease Activity Index; PFCD, Perianal Fistulizing Crohn’s Disease; RCT, randomized controlled trial; ROB, risk of bias; SAEs, serious adverse events; TEAEs, treatment-emergent adverse events; UC-MSC, umbilical cord mesenchymal stem cell.