Long-term Follow-up (LTFU) After Hyperbaric Oxygen or Sham Chamber Sessions in Military Post-concussive Syndrome
Completed United States Department of Defense clinical trials of hyperbaric oxygen (HBO2) for persistent post-concussive symptoms (PCS) lack LTFU. We report challenges encountered during LTFU data collection in a cross-section of this study population and the limited results.
Methods:In this observational cohort study, investigators contacted participants from two randomized, controlled trials of HBO2 for PCS: the Defense Advanced Research Projects Agency/Virginia Commonwealth University HBO2(DARPA/VCU) study and US Army HBO2 for Persistent PCS After Mild Traumatic Brain Injury (HOPPS) study. Participants completed an electronic survey assessing PCS (Rivermead Post-Concussive Symptom Questionnaire), post-traumatic stress disorder (PTSD), anxiety, depression, quality of life, current therapies, and interim trauma. Data are presented as the mean ±1 standard deviation.
Results:Of 132 randomized DARPA/VCU (n=60) and HOPPS (n=72) participants, 40 (30%) completed the survey (42 could not be contacted; 50 declined/were lost to follow-up). Survey participants were male, age 28.1±6.6 years. Most (88%) sustained multiple baseline head injuries. Time from original study randomization to LTFU was 39.2±6.1 months (range 28-51). At LTFU, participants reported continued symptoms of PTSD, depression, anxiety, and reduced quality of life. Among DARPA/VCU participants, total PCS scores worsened from baseline to LTFU in the 1.5 atmospheres absolute (ATA) equivalent HBO2 group (mean change 7.4±15.8) and improved in the sham (-8.0±7.7) and 2.0 ATA equivalent HBO2 group (-3.3±7.4). Individual changes varied widely, range -23 to +28 points. In HOPPS participants, total PCS scores worsened in all groups: local care (mean change 10.5±8.7), sham (7.9±11.9), and 1.5 ATA HBO2 (1.0±19.4).
Conclusions:In this limited, cross-sectional sample, PCS and PTSD symptoms did not appear to improve over time according to descriptive analyses. Low participation rates and potential response bias limit our ability to perform statistical hypothesis testing and to draw conclusions from these data. Future studies should consider prospective planning of longitudinal follow-up and regular engagement with participants to minimize attrition.