The efficacy of hyperbaric oxygen therapy in stroke patients: a real-world study

Background Hyperbaric oxygen therapy (HBOT) is widely used as an adjunctive intervention in stroke rehabilitation; however, its therapeutic benefit remains uncertain. This study investigated the clinical effectiveness of HBOT in improving recovery among patients with subacute-to-chronic stroke. Methods A total of 564 patients (14 days to 6 months after stroke onset) treated at Shanghai Yangzhi Rehabilitation Hospital between January 2017 and December 2024 were enrolled. Patients were assigned to either the HBOT group ( n = 250), which received 15–40 HBOT sessions at 1.8 ATA in addition to routine rehabilitation, or the non-HBOT group ( n = 314), which received routine rehabilitation alone. Propensity score matching (PSM) and inverse probability weighting (IPW) were used to balance baseline characteristics between groups. The primary outcome was the change in National Institutes of Health Stroke Scale (NIHSS) score from baseline (ΔNIHSS). Secondary outcomes included changes in Functional Test for the Hemiplegic Upper Extremity–Hong Kong (FTHUE-HK), Berg Balance Scale (BBS), and Modified Rankin Scale (mRS) scores, expressed as ΔFTHUE-HK, ΔBBS, and ΔmRS, respectively. Results After 1:1 PSM, 83 matched pairs were included. No significant between-group differences were observed in ΔNIHSS, ΔFTHUE-HK, ΔBBS, or ΔmRS. Sensitivity analyses confirmed these findings. IPW applied to the original cohort demonstrated a modest but statistically significant reduction in NIHSS score ( β = −0.3428, p = 0.040); however, the observed improvement was well below the established 2-point minimal clinically important difference. Post hoc subgroup analyses suggested that patients who completed 15–24 HBOT sessions experienced greater neurological improvement (interaction p = 0.004). Disease duration was not associated with treatment response. Conclusions HBOT was not associated with clinically meaningful improvements in overall functional recovery among patients with subacute-to-chronic stroke. Although weighted analyses identified a small statistically significant reduction in NIHSS score, the effect was not clinically meaningful. Exploratory subgroup analyses suggested a possible dose-related benefit that warrants confirmation in prospective studies.