HBOT for Sudden Hearing Loss: What the Evidence Shows

Two disclosures frame everything below, and they belong at the top. First, UHMS-listed is not FDA-cleared: hyperbaric chambers are cleared for about thirteen indications (carbon monoxide poisoning, decompression sickness, non-healing diabetic wounds, radiation injury, and others), and hearing loss is not among them (FDA, archived; UHMS indications), so every US hearing-loss use is off-label. Second, Medicare does not cover it (NCD 20.29), which makes the studied 10-20 session course a cash expense for most American patients. Neither fact cancels the evidence; both belong in the decision. New to the therapy itself? Start with how hyperbaric oxygen therapy works, and for the shorter primer on this exact topic see our hearing loss evidence guide.

Severity and timing decide who benefits

The headline averages hide the subgroup story, and the subgroups are where the decision lives. Two factors predict a larger effect across the literature: how much hearing was lost and how fast treatment starts. The UHMS literature, summarized in the StatPearls chapter, cites average gains of about 19.3 dB in moderate loss and 37.7 dB in severe loss, enough in the severe group to move an average patient from hearing-aid territory toward near-normal hearing (NBK459160). The Rhee meta-analysis found the benefit concentrated in severe-to-profound baseline loss, in salvage use, and in courses totaling at least 1,200 minutes (Rhee 2018). And the trials consistently favor starts within the first week or two. The uncomfortable corollary: mild loss, treated late, is the profile least likely to benefit, and the one most likely to recover on its own anyway.

Why an "option" grade is still meaningful

The AAO-HNS guideline grades HBOT as an option, the weakest actionable tier, below "recommendation." Skeptics quote the tier; marketers skip it. Both miss what the tier means here. In a guideline that also issued strong recommendations against antivirals, thrombolytics, and vasodilators for the same condition, HBOT is one of the only adjuncts named as worth offering at all, and the wording was deliberately tightened in 2019: combined with steroids, inside a defined window (Chandrasekhar 2019). Add the UHMS listing, which moved real referral practice when it arrived (Sherlock 2021), and the fair reading is: a genuine, bounded, time-sensitive option, graded conservatively because the trials are small. That is a strong position by HBOT standards, and this page still refuses to round it up to a promise.

UHMS-approved: yes. The Undersea and Hyperbaric Medical Society board approved idiopathic sudden sensorineural hearing loss as an HBOT indication on October 8, 2011 (UHMS approval notice), and it has been carried in the Hyperbaric Oxygen Therapy Indications Manual since the 2014 edition; the current 15th edition's chapter is publicly reprinted (LeGros 2024). This reflects a recognized body of evidence, and it is real.

FDA-cleared: no. The FDA clears hyperbaric chambers for a specific list of about thirteen conditions, and hearing loss is not on it (FDA, archived; the full list is mapped in our FDA-cleared indications guide). Note also what clearance even is: chambers are Class II devices reviewed through 510(k) equivalence, so a clearance is about the equipment, not proof the therapy works. Using a chamber for SSNHL in the US is off-label use of a cleared device, legal for physicians to prescribe, and accurately described only with all of those words attached.

Insurance-covered: usually no. SSNHL is not on Medicare's covered list (NCD 20.29), and commercial insurers largely follow, so American courses are typically self-pay. Health Canada, by contrast, does approve HBOT for SSNHL, a reminder that regulatory status stops at the border and should never be imported across it. How coverage works for the indications that are covered in the US is mapped in our insurance coverage guide. For order of magnitude on the cash side: the StatPearls chapter cites about $2,000-5,000 for a 10-session outpatient course (a 2017-era estimate), and the studied course runs 10-20 sessions (NBK459160).

The trials behind the AAO-HNS option and the meta-analyses above all used clinical hard chambers at 2.0-2.5 ATA breathing 100% oxygen. Soft portable chambers operate around 1.3 ATA, typically concentrator-fed, and their FDA clearance covers acute mountain sickness only: they cannot reach the pressure class of a single study on this page. Our own soft-shell S1 (1.3-1.5 ATA, concentrator-fed) sits in that class, and we do not present it as hearing-loss equipment, because it is not the equipment the evidence used. The regulatory and engineering details live in our medical-grade chambers explainer, the ATA pressure guide, and the soft-shell chamber guide.

For an operator whose community includes ENT referrals, the equipment-class takeaway is narrow and factual: the SSNHL protocol sits in hard-shell, medically engineered territory, certified pressure vessels, documented oxygen delivery, trained oversight, and physician direction. The single-person rung of that class in our lineup is the Superhuman L1 (2.0 ATA hard-shell, from $49,900), with the two- and four-person models shown below. Our chambers are built to that engineering class, which is the only claim we make here. It is a claim about hardware. The clinical claim belongs to the trials and the guideline, and the treatment decision belongs to the patient and their ENT, inside a window that equipment shopping does not pause.

The trials are small and imperfect. Cochrane pooled 392 patients across seven trials and called them small and generally poor quality, adding that the clinical significance of the hearing gain "remains unclear" (Bennett 2012). The strictest endpoint, a 50% return of hearing, did not reach significance. The field argues about the pooling in public: published critiques flagged heterogeneity in the 2018 meta-analysis (Lammers 2019) and weighting in the 2022 one. An average gain of 10 dB is real, but whether every patient notices 10 dB in daily life is a fair question the trials do not fully answer.

Spontaneous recovery muddies everything. Depending on the estimate, roughly a quarter to two thirds of patients recover useful hearing with no treatment at all, and placebo arms in the controlled trials run 35-39% (NBK459160; LeGros 2024). That is precisely why the randomized trials matter and why the anecdote of a friend who "got better after ten sessions" cannot settle anything. It is also why the benefit is expressed as added recovery on top of steroids and natural history, not as a cure rate.

The protocol is not settled, and the scope is narrow. The optimal pressure, session count, and start time are still debated; the guideline covers adults only, and pediatric data are absent; HBOT monotherapy is unestablished; and nothing here extends to noise-induced loss, age-related loss, acoustic trauma, or tinnitus, each of which is a different, mostly empty evidence base. The fair verdict: a genuine, guideline-listed, time-sensitive adjunct with a modest average benefit, off-label and usually self-pay in the US, useless outside its window and its indication. For the sharpest contrasts in this series, see the wound healing pillar, where the indication is fully cleared and covered but the flagship double-blind trial was negative; the stroke pillar, where an uncontrolled literature met its sham-controlled test and lost; and the Lyme pillar, where no controlled trial exists at all. Sudden hearing loss is the page where the evidence genuinely supports action, as long as the action starts this week.