Nonunions represent one of the major indications for clinical settings with stem cell-based therapies. The objective of this research was to systematically assess the current evidence for the efficacy of bone marrow-derived cell-based approaches associated or not with bone scaffolds for the treatment of nonunions. We searched MEDLINE (PubMed) and CENTRAL up to July 2019 for clinical studies focused on the use of cell-based therapies and bone marrow derivatives to treat bone nonunions. Three investigators independently extracted the data and appraised the risk of bias. We analysed 27 studies including a total number of 347 participants exposed to four interventions: bone marrow concentrate (BMAC), BMAC combined with scaffold (BMAC/Scaffold), bone marrow-derived mesenchymal stromal cells (BMSCs), and BMSC combined with scaffold (BMSC/Scaffold). Two controlled studies showed a positive trend in bone healing in favour of BMAC/Scaffold or BMSC/Scaffold treatment against bone autograft, although the difference was not statistically significant (RR 0.11, 95% CI -0.05; 0.28). Among single cohort studies, the highest mean pooled proportion of healing rate was reported for BMAC (77%; 95% CI 63%-89%; 107 cases, n = 8) and BMAC/Scaffold treatments with (71%; 95% CI 50%-89%; 117 cases, n = 8) at 6 months of follow-up. At 12 months of follow-up, an increasing proportion of bone healing was observed in all the treatment groups, ranging from 81% to 100%. These results indicate that BMAC or BMAC/Scaffold might be considered as the primary choice to treat nonunions with a successful healing rate at a midterm follow-up. Moreover, this meta-analysis highlighted that the presence of a scaffold positively influences the healing rate at a long-term follow-up. More case-control studies are still needed to support the clinical improvement of cell-based therapies against autografts, up to now considered as the gold standard for the treatment of nonunions.