The aim of this systematic review was to evaluate the longitudinal evolution of midterm to long-term results of reverse total shoulder arthroplasty (RTSA) for patients with massive irreparable rotator cuff tears (miRCT). Databases were scanned for studies of RTSA for miRCT. Studies with a minimum Level IV of evidence were considered eligible. Studies were included if they reported a minimum of 5 years of follow-up and excluded if they reported RTSA combined with tendon transfers or as revision arthroplasty. Data were grouped based on results after 5 to 7 years, 7 to 10 years, and 10 to 20 years of follow-up. Eight studies with a total of 365 shoulders were included. After a mean follow-up of 9.5 years (range, 5-20 years), the preoperative absolute and relative Constant scores were significantly improved from 24 to 59 points (P = .004) and from 33% to 74% (P = .009). The preoperative Subjective Shoulder Value improved from 23% to 72% (P = .049). Active anterior elevation and abduction also improved significantly (P = .004 and P = .014, respectively), but active external rotation remained unchanged (P = .855). None of the clinical scores or active ranges of motion significantly deteriorated up to 20 years after the operation (P > .05). After 10 years, 42% of the RTSAs showed grade III or IV inferior scapular notching. Pooled long-term results of RTSA for miRCT show significant improvement of overhead function and of objective and subjective outcome scores up to 20 years after surgery. Shoulder function and outcome scores also showed no significant deterioration between 5 and 20 years of follow-up. Longer follow-up will be needed to determine ultimate longevity.