Key takeaways
- HOT-LoCO is a Swedish phase II double-blind sham-controlled RCT of 80 long COVID patients, published in 2025.
- HBOT arm received 10 sessions at 2.4 ATA over 6 weeks; sham arm received 10 sessions at 1.03 ATA.
- No significant difference between arms at 13 weeks on primary endpoints (physical functioning, role limitations).
- Safety profile was favorable — adverse events were comparable between arms with no serious events attributable to HBOT.
- The negative result does not refute the 40-session Efrati Lab protocol, but it does strongly suggest that 10 sessions is below the therapeutic threshold for long COVID.
The HOT-LoCO trial
The Hyperbaric Oxygen Treatment of Long COVID (HOT-LoCO) trial was a Swedish-led phase II double-blind randomised controlled trial published in 2025. Eighty adults with long COVID (post-COVID condition persisting at least three months after infection) were randomised to either active HBOT at 2.4 ATA for 90-minute sessions, or a sham arm at 1.03 ATA, for a total of 10 sessions over 6 weeks.
The primary endpoints at 13 weeks were physical functioning and role limitations, measured on the Short Form-36 (SF-36) questionnaire.
The result: no significant difference
At 13 weeks, HOT-LoCO found no statistically significant difference between the active HBOT and sham arms on either primary endpoint. Both arms improved modestly from baseline, consistent with placebo and natural recovery effects. The safety profile was favorable with no serious adverse events attributable to HBOT.
On the face of it, this is a negative trial. But its interpretation requires close attention to one specific design choice: the 10-session dose.
Why the 10-session dose matters
The Efrati Lab RCTs that established HBOT as a research-stage treatment for post-COVID cognitive symptoms — most notably Zilberman-Itskovich et al. 2022 (Scientific Reports) — used 40 sessions at 2.0 ATA over approximately 60 days. That trial demonstrated significant cognitive, physical, and perfusion-MRI benefits vs. sham.
HOT-LoCO's 10 sessions at 2.4 ATA represents one-quarter the exposure of the Efrati protocol. The HOT-LoCO investigators acknowledged this design trade-off up front: they were explicitly testing whether a lower-cost, lower-commitment course could produce clinical benefit. If it could, HBOT would become dramatically more accessible for long COVID.
The negative result tells us this: it probably cannot. Ten sessions does not appear to be enough.
How this fits the broader evidence base
HBOT evidence for long COVID in 2024-2025 is actively evolving:
- Zilberman-Itskovich 2022 (Sci Rep) — RCT, 40 sessions at 2.0 ATA, significant benefit across cognitive and physical domains.
- Zilberman-Itskovich 2024 follow-up (Sci Rep) — one-year durability of gains, including sleep, psychiatric, and pain measures.
- Prospective registries 2024-2025 — multiple cohorts reporting 56–63% clinically relevant improvement at 3 months after full-course HBOT.
- HOT-LoCO 2025 (negative) — 10 sessions insufficient.
The pattern is consistent: full-course HBOT (40-60 sessions at 2.0 ATA) produces clinically meaningful benefit in a majority of post-COVID patients; abbreviated protocols do not. This is a dose-response question, not an on/off question.
What it means for clinicians and patients
For patients: if considering HBOT for long COVID, choose a facility that offers the full Efrati Lab protocol — 40 sessions at 2.0 ATA, hard-shell chamber, with air breaks. Short "trial" courses may fail to reach therapeutic effect and could leave patients with the misleading impression that HBOT does not work.
For clinicians and researchers: HOT-LoCO is a useful boundary condition. It establishes that 10 sessions is sub-therapeutic. Future trials should test the Efrati protocol dose (40+ sessions) in different populations and compare long-term outcomes against standard of care — not test abbreviated protocols against sham.
For payers: long COVID HBOT remains off-label and cash-pay (approximately $6,000–$18,000 for a full course). Insurance coverage will require the UHMS Oxygen Therapy Committee to formally review the growing evidence base — a review widely anticipated but not yet announced.
The bigger lesson
HOT-LoCO is a good reminder that trial design — especially dose — is where HBOT research either succeeds or fails. The DoD HOPPS and BRIMS TBI trials ran into similar interpretation problems a decade earlier. The modern HBOT evidence base increasingly points toward adequate dose being the single most important variable. Future trials that mimic the Efrati 2.0 ATA / 40-session protocol are the ones most likely to produce interpretable results.
For an evidence-aligned view of what HBOT can and cannot do, cross-reference the full indication evidence on our long COVID evidence page and the studies database.
Frequently Asked Questions
Why only 10 sessions?
HOT-LoCO was deliberately testing a shorter protocol — asking whether a lower-cost, lower-commitment HBOT course could help long COVID. The pre-specified hypothesis was that even a brief course might show benefit if the mechanism is primarily acute anti-inflammatory.
Does this contradict Zilberman-Itskovich 2022?
No — the Zilberman-Itskovich RCT used 40 sessions at 2.0 ATA and showed significant cognitive, physical, and imaging benefits. HOT-LoCO is best interpreted as evidence that 10 sessions is sub-therapeutic, not that HBOT is ineffective for long COVID.
What does this mean for patients?
If pursuing HBOT for long COVID, choose a facility that uses the full Efrati-protocol course (40 sessions at 2.0 ATA). Short "trial" courses may not reach therapeutic effect and could falsely suggest HBOT does not work.
Is HBOT for long COVID now disproven?
No. HOT-LoCO reinforces that dose matters. The current clinical consensus from the broader evidence base (Zilberman-Itskovich 2022, 2023, and growing prospective registries) supports 40-60 sessions at 2.0 ATA. The question is dose, not whether HBOT works.
