Patient-reported outcomes following craniofacial regenerative stem cell therapy: a secondary analysis and machine learning approach

Patient-reported outcomes ( PROs ) are important clinical outcomes, however frequently overlooked. This study assessed PROs and early wound healing following craniofacial bone defect reconstructions [extraction sockets, EX maxillary sinus floor augmentation, SA craniofacial trauma, CT and alveolar cleft reconstruction, AC] in which stem cell therapy was evaluated. Three randomised studies were included in the analysis. A total of 64 participants who received either: (1) experimental stem cell therapy (bone marrow aspirate from the posterior iliac crest) combined with a scaffold ( n = 35), or (2) standard-of-care therapy using a guided bone regeneration or autogenous bone block grafting ( n = 29). Early wound healing indicators (wound healing index, WHI ) and PROs- including self-assessed discomfort ( DF ), adverse events ( AE ), and quality of life ( QoL )- were evaluated at 1, 2 and 4 weeks post-surgery. Mixed models were used to compare outcomes between groups, whilst machine learning (ML) models were trained on EX and SA/CT datasets, then tested on SA data. DF increased (worsened) across all groups ( p < 0.001) over time, and the total QoL score rose (worsened) with age ( p = 0.02). Hispanic ethnicity recorded the highest overall QoL scores ( p = 0.01). AE increased with age and among females ( p < 0.001). No differences were found between groups. ML performed well with nonlinear approaches predicting for WHI, QoL, and DF whilst all models predicted the binary outcome AE. This study provides evidence of equivalent impacts of stem-cell regenerative therapy vs. standard-of-care therapies. Further research should focus on PROs when performing oral and craniofacial reconstructive procedures. Trial registration: ClinicalTrials.gov identifiers NCT00755911 and NCT00980278.