Frequently Asked Questions

Yes. In this randomized, sham-controlled trial, 68% of veterans in the hyperbaric oxygen therapy (HBOT) group met the predefined treatment response threshold of at least a 30% reduction in CAPS-5 score at the end of treatment, compared with 4% in the sham group (P < .001). At 3-month follow-up, response rates were 61% with HBOT versus 4% with sham (P < .001), and the study also found a significant group-by-time interaction for CAPS-5 scores (F = 27.33, P < .001).

Remission was more common with HBOT than with sham and continued to increase after treatment ended. By the end of treatment, 25% of the HBOT group achieved remission, defined as a CAPS score lower than 20, compared with 3.6% of the sham group (P < .01). At 3-month follow-up, remission rates were 39.2% in the HBOT group and 0% in the sham group (P < .01).

The protocol tested was 60 daily sessions delivered 5 days per week in a multiplace chamber. Each HBOT session consisted of 90 minutes of 100% oxygen at 2 atmospheres absolute (ATA) with 5-minute air breaks every 20 minutes. The sham protocol used 21% oxygen by mask at 1.02 ATA for 90 minutes, with brief initial compression to 1.2 ATA followed by decompression to support blinding.

The study enrolled male veterans aged 25 to 60 years with combat-associated PTSD lasting at least 5 years. Participants had persistent debilitating symptoms despite at least 1 course of trauma-focused psychotherapy and pharmacotherapy and still met CAPS-5 diagnostic criteria for PTSD with a score above 20. Patients with a history of traumatic brain injury or other known brain pathology were excluded.

Yes, the trial found significant improvements in depressive symptoms with HBOT compared with sham. On the Beck Depression Inventory-II, the between-group effect was significant (F = 4.2, P = .043), and the depression domain of the DASS-21 also improved significantly (F = 4.55, P = .034). The anxiety and stress domains of the DASS-21 did not reach statistical significance.

HBOT was associated with increased resting-state functional connectivity compared with sham across several brain networks implicated in PTSD. The study found significant group-by-time increases involving the frontoparietal network, default mode network, salience network, and thalamic connectivity, with no significant group-by-time connectivity decreases. The hippocampi and amygdala did not show significant group-by-time interactions in the seed-to-voxel analysis.

HBOT was generally well tolerated in this trial, and adverse events were reported as mild and self-limited. There were 23 documented mild, self-limited adverse events in the HBOT group and 15 in the sham group. Eight self-remitting barotrauma events were documented overall, including 6 in the HBOT group and 2 in the sham group.

Some HBOT-treated participants reported surfacing of new memories during treatment. Seven subjects in the HBOT group reported new memories emerging during the treatment course, accompanied by distress that lasted several days and then resolved; no patients in the sham group reported this phenomenon. The article notes that, because symptom worsening and memory surfacing can occur, treatment should be provided only by centers with professionals experienced in PTSD care.

Yes, the sham condition appeared credible based on the trial's blinding assessment. After the first session, 64% of participants in both groups believed they were receiving HBOT and 36% believed they were receiving sham treatment, with no difference between groups (P = 1.0). Patients, physicians, therapists, and assessors were blinded to treatment assignment.

No. The article states that the findings should be applied only to the evidence-based protocol used in the trial and should not be used to encourage treatment in non-medical grade facilities, monoplaces that vary from the tested protocol, home-use chambers, or clinics without a professional multidisciplinary medical team. The authors also emphasize the importance of staff trained to manage PTSD-specific risks, including symptom worsening and memory surfacing during treatment.

The main limitations were the relatively small sample size and the short follow-up period. The clinical analysis included 28 patients in the HBOT group and 28 in the sham group, and follow-up lasted 3 months after treatment. The authors also note that generalizability may be limited because the demands of the trial may have selected for veterans with substantial symptom burden.