Natural and bio-engineered stem cell-derived extracellular vesicles for spinal cord injury repair: A meta-analysis with trial sequential analysis.

Background: Stem-cell derived extracellular vesicles (EVs) have shown promise in preclinical spinal cord injury (SCI) models but lack a comprehensive literature review for clinical translation guidance.

Methods: This meta-analysis with trial sequential analysis systematically search PubMed, Web of Science, Embase, and Cochrane Library databases. Prespecified inclusion criteria were studies reporting on measurable outcomes relevant to SCI repair. Risk of bias and quality of reporting were assessed. Random-effects meta-analyses and subgroup analyses comparing natural and bio-engineered EVs were performed. The study was registered with PROSPERO (CRD42024512122).

Findings: The search identified 3935 records, of which 39 studies were included, totaling 1801 animals. Administration of EVs significantly improved locomotor function as measured by Basso-Beattie-Bresnahan or Basso-Mouse-Scale scores at 1 week (natural EVs: SMD 1.50, 95 % CI 1.06-1.95; bio-engineered EVs: SMD 1.93, 95 % CI 1.34-2.52) and 3 weeks (natural EVs: SMD 2.57, 95 % CI 1.96-3.17; bio-engineered EVs: SMD 3.16, 95 % CI 2.29-4.02) post-injury. Subgroup analyses indicated surface modification approaches were most effective among bio-engineered EV strategies. EVs also promoted nerve growth (SMD 2.95, 95 % CI 2.12-3.78), enhanced neuron conductivity (MD 0.75, 95 %CI 0.59-0.90), alleviated inflammation (SMD -3.12, 95 % CI -4.15--2.10), and reduced lesion size (SMD -2.90, 95 % CI -3.87--1.93).

Conclusions: Both natural and bio-engineered EVs improve functional and pathological outcomes in animal models of SCI. The enhanced benefits observed with bio-engineered EVs, particularly those utilizing surface modification approaches, highlight the importance of continued exploration into bio-engineering techniques to optimize EVs' therapeutic efficacy for SCI repair. Protocol Registration CRD42024512122.