The Bone Disease That Waits: Dysbaric Osteonecrosis in Professional Divers

A professional diver comes in for a routine fitness-to-dive examination. He has no pain, no stiffness and no recollection of a bends hit that frightened him. He passes the physical, the lung function, the neurological screen. Then somebody orders a magnetic resonance scan of his shoulders and finds a wedge of dead bone sitting just beneath the joint surface of the humeral head — a lesion that has probably been there for years, that plain X-rays would very likely have called normal, and that will eventually decide whether he keeps diving and whether his shoulder keeps working.

That is the scenario a group of Turkish diving-medicine physicians have now put numbers behind. In a retrospective review of 120 professional divers who came through routine fitness-to-dive evaluations at Çanakkale Onsekiz Mart University between June 2018 and March 2024, Dr Şefika Körpinar and colleagues identified dysbaric osteonecrosis in ten of them — 8.3 percent of a working population that is, on the whole, young and healthy. The paper appears in the third-quarter 2026 issue of Undersea and Hyperbaric Medicine.

Dysbaric osteonecrosis deserves a moment of explanation, because it is the quiet member of the decompression family. It is not the bends. Decompression sickness announces itself — joint pain, skin mottling, neurological signs — and sends a diver to a recompression chamber. Dysbaric osteonecrosis is a chronic, structural disease of bone that develops in people who have spent years absorbing decompression stress without ever having an acute event that anybody recorded. The classical lesions are juxta-articular, sitting next to the shoulder and hip joints, where the damage eventually produces collapse of the joint surface and secondary arthritis.

What the Refined Data Showed

The Çanakkale cohort is small and single-centre, and the authors are clear that it is retrospective. Within those limits, though, the pattern is unusually informative, because the study looked at both imaging modalities in the same divers. Dysbaric osteonecrosis was found most often in the right humerus. MRI identified nine of the ten affected divers; conventional radiography identified one, a difference that was statistically significant (p < 0.0001). That is the paper’s central practical claim, since early lesions are exactly the ones X-rays struggle to see.

The factors associated with lesions read like a description of a long, heavy career rather than of a single bad dive. Cumulative diving duration beyond 900 hours, more than 15 hours of diving per week, working depths beyond 35 metres and more than 9.5 years of experience all carried significant associations — as did two things that had already happened to the diver: a previous history of decompression sickness, and a previous diagnosis of dysbaric osteonecrosis. Those last two are also the least surprising, since a lesion of this kind tends to be permanent.

The authors’ proposal is not universal screening. It is a narrower one: that MRI should be justifiable as part of periodic fitness-to-dive assessment for divers with prolonged or intense exposure, or with a history of decompression-related complications, especially when nothing is bothering them.

Why the Bubble Debate Still Matters

What causes the lesion is less settled than the lesion itself. The same journal issue carries a pointed exchange on the underlying mechanism. Dr Ran Arieli of the Israel Naval Medical Institute reviews the two competing schools — the older view that decompression bubbles are the primary cause of decompression illness, and the newer argument, associated with Dr Stephen Thom, that circulating microparticles shed from endothelium do most of the initial injury. Arieli’s verdict is that bubbles remain the main initial cause and that microparticles amplify harm rather than initiate it.

A second paper approaches the physics from a different direction. Dr Oswaldo Monteiro Del Cima and colleagues at the Federal University of Viçosa in Brazil, working with the decompression modeller Bruce Wienke, reproduced Ostwald ripening — the transfer of gas from small bubbles to large ones — in a viscous fluid. Small bubbles disappeared, larger ones grew, and the population thinned over time. They then ran the effect into a reduced gradient bubble model schedule for a closed-circuit rebreather dive to 128 metres, where it lengthened the first decompression stop. The authors call the clinical extension a conjecture and the study involved no divers, but the direction of travel is uncomfortable: in a supersaturated tissue, the bubbles that persist are the big ones.

Bone is an attractive place for that to happen. Its blood supply is comparatively poor, which is why it off-gasses slowly and why a lesion can form without producing anything a diver would notice.

Fitness-to-Dive Is Not One Test

If screening for a silent disease depends on getting divers into the clinic, it is worth knowing how solid the clinic’s instruments are. Two papers in the same issue of Diving and Hyperbaric Medicine suggest that the answer is: less solid than the paperwork implies.

The first is a validation study of the FitMáx questionnaire, a self-report tool for estimating aerobic capacity. Dr Pieter Dijk and colleagues at the Royal Netherlands Navy Diving and Submarine Medical Center put it to 157 Navy divers and recompression personnel — a deliberately fit population — and compared its estimates with directly measured peak oxygen uptake on a cycle ergometer. The correlation was moderate (r = 0.631), the questionnaire systematically underestimated fitness by 1.8 mL·kg⁻¹·min⁻¹, the limits of agreement spanned roughly ±10.9 mL·kg⁻¹·min⁻¹, and the false-positive and false-negative rates were 8.9 percent and 10.8 percent. The centre’s conclusion was blunt: it has not adopted FitMáx for fitness-to-dive evaluations, and direct exercise testing remains the more reliable route to individual risk stratification.

The second paper looks at the psychological side of the same decision. Dr Calina Ouliaris and colleagues, including Dr Thijs Wingelaar of the Dutch military diving service, synthesised 29 studies on mental health and neurodevelopmental factors in diving assessment. Diagnosis alone cannot determine fitness to dive, they conclude. What the evidence supports instead is a set of functional questions — how a diver behaves under stress, whether psychiatric stability holds, how medication interacts with the diving environment, how much cognitive reserve is left when cognitive performance is already degraded by depth, cold and breathing resistance. Panic and maladaptive behaviour are named as key pathways to bad outcomes, and risk as probabilistic rather than binary.

Read together, the two papers describe an assessment process that works best when it is individualised, resource-intensive and honest about uncertainty — an awkward fit with an industry that needs bodies in the water.

The Workforce Behind the Scan

How awkward becomes clearer from a qualitative study by Dr Holly Blake and colleagues at the University of Nottingham, based on interviews and focus groups with 20 men drawn from commercial diving, supervision and industry representation. Three themes emerged: entry into the occupation shaped by financial incentive, risk acceptance and informal routes; psychological strain marked by traumatic exposure, a normalisation of distress, stigma around disclosure and support that arrived after the fact; and training described as generic, low-fidelity and poorly aligned with offshore and saturation work, with few chances to practise emergency skills.

The study measures nothing, and says so. What it describes is the culture in which a silent occupational disease has to be found. Surveillance for dysbaric osteonecrosis depends on divers reporting histories and turning up to periodic examinations; if distress is normalised and help-seeking is stigmatised, the workforce that accumulates the most decompression dose is the one least likely to be captured by a screening pathway.

Where the Chamber Stands — and Where It Does Not

Hyperbaric oxygen therapy is a genuinely decisive treatment in acute decompression illness, and the same journal issue contains a reminder of how decisive. Clinicians at the Naval Hospital of the Eastern Theater Command in China describe a 33-year-old diver who surfaced emergently from 56 metres after an underwater explosion and arrived comatose with bleeding from the nose and ears. An initial 60-metre recompression protocol helped only partly, with symptom recurrence during decompression; a 15-metre air saturation decompression with extended station times followed, after which MRI four days later showed at least a 50 percent reduction in lesion volume and the diver was discharged neurologically intact on day 11, back on land duties at three months. That is a single case report, not a protocol, and the authors treat it as a demonstration of principle — ultra-high-pressure recompression with prolonged desaturation.

Dysbaric osteonecrosis sits outside that picture, and the regulatory documents say so plainly. Decompression illness and gas embolism are named both among the FDA-cleared indications for hyperbaric oxygen devices — Class II devices cleared through the 510(k) pathway under Product Code CBF — and in Medicare’s national coverage determination 20.29, whose covered list runs from acute carbon monoxide intoxication through decompression illness, osteoradionecrosis and chronic refractory osteomyelitis. Dysbaric osteonecrosis appears on neither list. Nothing in the new cohort study tests hyperbaric oxygen as a treatment for the lesions; its subject is detection, and the tools the authors discuss are imaging and work-pattern surveillance, not pressure.

The wider evidence base in hyperbaric medicine is unevenly distributed, which is worth remembering whenever a chamber is proposed for a new target. The issue reprints an inventory from the fifteenth edition of the UHMS Hyperbaric Oxygen Therapy Indications manual, in which Dr Michael Bennett of the Cochrane Collaboration catalogues 416 citations covering roughly 390 randomised trials. That is a large number of trials for a specialty still accused of practising on thin evidence — and, as Bennett notes, of variable quality, concentrated in a handful of indications rather than spread evenly.

One caution belongs here too. The low-pressure devices sold for wellness use — soft-sided fabric chambers and some hard-sided “mild” units, typically designed for operational pressures of three to five pounds per square inch — cannot deliver decompression treatment, and a critical review in the same issue by Dr Ryan Murray of Northern Westchester Hospital argues that they frequently fail to meet FDA clearance, NFPA 99 and NFPA 101 codes and ASME-PVHO standards, with reported harms including otic and pulmonary barotrauma, suffocation, fire, structural failure and death. Device safety is not theoretical in the clinical setting either: the FDA’s August 25, 2025 letter to health care providers states that the agency is aware of fires involving hyperbaric oxygen therapy devices that resulted in serious injuries and deaths, and asks providers to follow manufacturer instructions on training, continuous patient monitoring, hyperbaric-compatible clothing such as cotton, and the exclusion of prohibited electrical or static-generating items.

What Would Actually Change the Numbers

The honest summary of the Çanakkale paper is that it is a prevalence estimate from one centre, with a clear imaging message and no outcome data. It cannot say whether finding an asymptomatic lesion early changes what happens to that shoulder, because nobody has followed a screened cohort long enough to find out. It cannot say what screening interval is worth the cost, or whether decompression practice that reduces acute hits would also reduce bone lesions — although that is where preventive research is pointing. One proposal in the current literature, from Sergio Angelini and Alessandro Piacente, uses tissue-specific gradient factors inside a ZH-L16C decompression framework to attack “undeserved hits”, the episodes that occur inside accepted limits; it draws on a DAN Europe analysis of more than 39,000 dives containing 295 undeserved hits, and notes decompression sickness appearing at gradient-factor settings as low as 70, mostly in intermediate tissues. The paper is simulation work, a proposal rather than a validated algorithm, and one of its authors has commercial links to dive-equipment interests — all of which should be stated when it is cited.

For now, the practical content of this cluster is narrower than a chamber protocol and more durable than a marketing claim. Long, deep careers carry a measurable bone cost. Plain films under-detect it; MRI is more sensitive but not free. Fitness-to-dive decisions are better made with direct testing and individualised judgement than with a questionnaire, and they are made, uncomfortably often, in a workforce that has learned not to complain. The intervention that matters most is still the unglamorous one: fewer, shallower, better-controlled decompressions in the first place.

Sources

← Back to blog