The use of hyperbaric oxygen (HBO2) in the field of traumatic brain injury (TBI) is becoming more widespread and increasing yearly, however there are few prognostic reports on long-term functional efficacy. The aim of this study was to assess the functional prognosis of patients with moderate-to-severe TBI 5–8 years following HBO2 treatments and to explore the optimal HBO2 regimen associated with prognosis, using a retrospective study. Clinical data were retrospectively collected as a baseline for patients with moderate-to-severe TBI treated with HBO2 during inpatient rehabilitation from January 2014 to December 2017. The primary outcome measure was the Disability Rating Scale (DRS) and the secondary outcome measure was the Glasgow Outcome Scale. A total of 133 patients enrolled, with 9 (6.8%) dying, 41 (30.8%) remaining moderately disabled or worse (DRS scores 4–29), 83 (62.4%) remaining partially/mildly disabled or no disability (DRS scores 0–3). Logistic regression analysis revealed that age at injury (odds ratio (OR), 0.96; 95% confidence interval (CI), 0.92–0.99), length of intensive care unit stay (OR, 0.94; 95% CI, 0.88–0.99), and HBO2 sessions (OR, 0.97; 95% CI, 0.95–0.99) were variables that independently influenced long-term prognosis. Cubic fitting models revealed that 14 and 21.6 sessions of HBO2 could be effective for moderate and severe TBI, respectively. This study highlighted that HBO2 in moderate-to-severe TBI may contribute to minimize death and reduce overall disability in the long-term. However, clinicians should be cautious of the potential risk of adverse long-term prognosis from excessive HBO2 exposure when tailoring individualized HBO2 regimens for patients with moderate-to-severe TBI. The study was registered on ClinicalTrials.gov (NCT05387018) on March 31, 2022.