Purpose This systematic review aimed to answer the PICO question: do adhesive protocols used for non-carious cervical lesions (NCCLs) using a universal adhesive system influence marginal degradation, marginal staining, and retention of these restorations? The self-etching adhesive strategy and selective enamel etching were compared with the etch-and-rinse strategy as a control. Materials and Methods The study searched various databases, including PubMed, Web of Science, Cochrane Central Register of Controlled Trials, Scopus, Embase, and grey literature, to find randomized clinical trials (RCTs) comparing self-etching (SE) or selective enamel etching (SEE) to the etch and rinse (ER) strategy. The risk of methodological bias was assessed using the Cochrane RoB 2 tool. Data were dichotomized and analyzed using RevMan v 5.3, adopting the Mantel-Haenszel method. The quality of evidence was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE). Results Twenty RCTs were included in the meta-analysis. Results showed that using universal adhesives with the SE strategy resulted in clinical signs of marginal degradation at 12 months, 24 months, and 36 months of follow-up, and marginal staining at 24 months. The adhesive strategy did not interfere with the retention of restorative material used for NCCLs over 36 months, as assessed based on both the United States Public Health Service (USPHS) and World Dental Federation (FDI) criteria. Conclusion With moderate certainty of evidence, after 24 months of follow-up, the SE strategy results in the detection of clinical signs of marginal degradation and staining. The adhesive strategy adopted did not influence the retention rate of the restorations over 36 months of follow-up.