Chemotherapy Takes Sperm. Can an Oxygen Chamber Give Some Back?

Stylised cross-section of a seminiferous tubule: concentric rings of teal germ cells stepping from dark at the outer wall to pale at a clear central lumen, fine sperm tails streaming inward, amber oxygen specks diffusing in from capillaries at the periphery, on a deep navy field

Cyclophosphamide is one of the drugs that made childhood cancer survivable. It is also an alkylating agent, and at the doses used for lymphoma, leukaemia, testicular cancer, sarcoma or before a stem-cell transplant, it is reliably toxic to the germ cells of the testis. Sperm banking before the first dose is standard practice for a reason: once the drug has been through, there is often very little to bank.

Which is why a mouse paper published on 19 October in Toxicology Mechanisms and Methods is worth reading carefully — and why it should not be taken as licence to skip the bank.

What the chemotherapy model actually showed

Ma L and colleagues, working across the andrology and hyperbaric departments at hospitals in Zaozhuang, Shandong, gave male BALB/c mice five consecutive days of intraperitoneal cyclophosphamide, then treated them with hyperbaric oxygen at 2.5 ATA for 60 minutes a day, five days a week, for two weeks. Testicles were then examined for sperm parameters, serum sex hormones, histopathology, oxidative stress markers and a panel of RhoA/ROCK proteins.

Against the cyclophosphamide-only group, the numbers were not subtle. Sperm count rose 80.95 per cent and progressive motility 160.96 per cent, both at p < 0.0001. The abnormal-sperm rate fell 38.88 per cent. Testicular malondialdehyde, a marker of lipid peroxidation, dropped 40.90 per cent while superoxide dismutase activity rose 69.53 per cent. RhoA and ROCK1 protein expression fell; Nrf2 and haem oxygenase-1 rose.

The part that lifts this above the usual descriptive animal study is the knockout. In mice engineered without RhoA, essentially all of the benefit disappeared — hormone levels, sperm parameters, testicular indices, oxidative stress markers, the Nrf2/HO-1 response. That conversion of a correlation into a pathway claim is what a mechanistic paper is for. It is also, on its own, a rodent result about a rodent dose: no mating, no litters, no fertility endpoint at all, and a pressure at or above what most clinical wound protocols use. Whether it means anything for a 22-year-old about to start chemo is a separate question, and a much harder one.

The testis lives close to its own oxygen limit

That the organ responds to hyperbaric oxygen at all is not surprising, because the testis is unusually exposed to oxygen injury twice over. It runs a high metabolic rate, its blood supply is a countercurrent arrangement built to keep the organ a couple of degrees below core temperature, and its germinal epithelium is a fast-dividing tissue sitting at low physiological oxygen tension. Interrupt the blood supply and you get the classic ischaemia-reperfusion injury of testicular torsion: germ-cell apoptosis, sloughing of the seminiferous epithelium, atrophy.

Zhang and colleagues at the PLA Navy General Hospital in Beijing modelled exactly that in Sprague-Dawley rats in 2013. Hyperbaric oxygen preserved the seminiferous tubules, suppressed germ-cell apoptosis and prevented testicular atrophy after torsion. The mechanism they traced will look familiar to anyone who has read this literature: malondialdehyde down, superoxide dismutase and haem oxygenase-1 up, neutrophil infiltration down, reduced mRNA for TNF-α, IL-1β and CD44 with suppressed NF-κB, p38 and JNK, and less nitric oxide through inducible nitric oxide synthase.

The same wall of antioxidant and anti-inflammatory effects shows up when the injury is heat rather than ischaemia. Hsieh and colleagues at Chi-Mei Medical Center in Tainan preconditioned adult male rats at 2.0 ATA for two hours daily for fourteen days, then ran them to exertional heat stroke. The preconditioned animals finished with a core temperature of 40 °C against 43 °C, scrotal temperature 34 °C against 36 °C, plasma testosterone 6.8 against 3.5 ng/mL, follicle-stimulating hormone 196.3 against 513.8 mIU/mL, luteinising hormone 131 against 189 IU/L, better erectile function, and lower histopathology scores across epididymis, testis, seminal vesicle, prostate and bladder.

All of it is injury prevention or injury repair in animals. None of it is enhancement. The distinction matters, because it is the difference between a chamber used to salvage a torsed testis — there is a single paediatric case report in Diving and Hyperbaric Medicine from 2014, and it is exactly that, a case report — and a chamber sold to a healthy man who wants a higher sperm count. The animal literature supports the first framing. It says nothing about the second.

Nine trials, one large number, a very thin file

The human literature has been pooled once, by Liu B and colleagues in Changsha, and the resulting paper is both the best argument for the therapy and the best argument for caution. Published in Medical Gas Research in 2025, registered on INPLASY as 2024120101 and reported to PRISMA 2020, it screened 291 records and kept nine randomised trials.

The pooled effects are large. Sperm viability, six studies, weighted mean difference 22.99 (95% CI 14.91–31.07). Sperm density, eight studies, 14.90 (11.92–17.88). Motility, five studies, 23.91 (18.53–29.29). Normal sperm rate, three studies, 15.34 (9.70–20.97). Abnormal-sperm rate, two studies, −16.79 (−18.31 to −15.28). And the one that would matter most to a couple: clinical pregnancy rate across six studies, risk ratio 3.45 (95% CI 1.38–8.73). Leave-one-out sensitivity analysis did not shift any of these, which is a genuinely reassuring test.

Then you read the table of included studies, and the picture changes. Jadad scores are 2 or 3 out of 5. Sample sizes run from 18 to 180, and most are under 100. Every single trial is hyperbaric oxygen added to something else — antibiotics for chronic bacterial prostatitis, varicocelectomy, pentoxifylline plus vitamin E, IVF, or vaguely specified “usual care”. And the nine trials cover four different diseases: varicocele, asthenozoospermia, chronic bacterial prostatitis and idiopathic infertility. Heterogeneity is high for every outcome except the abnormality rate, which had two studies and was pooled with a fixed-effects model.

The authors do not hide any of this. Their own limitations paragraph cites small samples, likely publication bias from unpublished negative results, and the fact that “the majority of the studies included in this analysis focused on Chinese populations, which may limit generalizability due to regional and ethnic factors.” They call for multicentre randomised trials.

Two of the included studies are worth putting side by side, because they show what the pooled numbers are actually made of. Metelev and colleagues in Moscow enrolled 90 men with idiopathic infertility, gave 60 of them hyperbaric oxygen before IVF and left 30 as controls, and reported sperm DNA fragmentation falling from 33.2 ± 7.5 to 11.9 ± 5.9 per cent by TUNEL assay, seminal reactive oxygen species from 0.89 to 0.39 mV/s, and a pregnancy rate of 63.3 per cent against 36.7 per cent. Those are striking figures from a study the abstract describes only as having a control group. Worth noting as a measure of the field’s record-keeping: the review’s own table transcribes the two group sizes the other way round.

The second is smaller still. Özgök Kangal and Özgök in Ankara went back through the files of 15 men who had received hyperbaric oxygen between 2015 and 2019 for any reason, sorted them into DNA-fragmentation, globozoospermia and azoospermia groups, and followed what happened in assisted reproduction. Thirty-one embryos, 19 transferred, three healthy pregnancies — 30 per cent. Mean sperm count rose from 8.4 ± 11.1 to 15.7 ± 15.0 million per mL. Testicular sperm extraction showed positive change in four of five azoospermic men. DNA fragmentation improved in half of the first group, at p = 0.500. Fifteen patients, three pregnancies, and a p-value that means nothing.

The same pressure lowers testosterone in healthy men

The cluster has one human dataset pointing the other way, and it comes from diving medicine rather than fertility clinics. Verratti and colleagues, with Enrico Camporesi, studied six healthy, well-trained recreational male divers across four conditions combining exercise with different hyperbaric and hyperoxic exposures, measuring prolactin, FSH, LH, cortisol, oestradiol and total testosterone immediately afterwards.

Exercise raised prolactin, and hyperbaric oxygen amplified that rise. Hyperbaria reduced oestradiol, an effect hyperoxia partly blocked. And hyperbaria — not hyperoxia — reduced total testosterone, while FSH, LH and cortisol did not move. The authors’ own interpretation of that last finding is blunt: “TT reduction may be interpreted as an acute and transient fertility impairment.”

Six men, acute measurements, immediately after exercise, in a study the authors call a pilot. It is not strong evidence of harm. But it is the only human gonadal-hormone data in this whole file, it was collected in healthy men rather than injured ones, and it runs in the opposite direction to the rodent work. Any account of hyperbaric oxygen as a fertility treatment that omits it is incomplete.

Two further corrections belong here, because both circulate in clinic marketing. The 2018 World Neurosurgery paper titled “Improving Sperm Viability After Spinal Cord Injury Using Hyperbaric Therapy” reads like a clinical study; it is 18 rats with a contused spinal cord, in which sperm viability on day 28 was 86.8 ± 5.6 per cent against 48.8 ± 21.8 per cent, with no difference in sperm concentration at all. And the fertility work is not exclusively male: Ma Y and colleagues in Changzhou reported in 2023 that ten days of hyperbaric oxygen improved serum AMH, reduced follicular apoptosis and improved oocyte maturation, fertilisation and blastocyst formation in 40-week-old female mice, with PI3K-Akt the most prominent pathway among 27 differentially expressed genes. Forty-week-old mice are not women in their forties, and mouse oocytes are not a pregnancy.

What no regulator has said

Fertility is not a cleared indication for hyperbaric oxygen, and this is not a technicality.

The FDA’s letter to health care providers of 25 August 2025 — still the agency’s current statement on the modality, and verified live for this article — describes HBOT devices as Class II devices cleared through the 510(k) pathway, identifiable by Product Code CBF, and it is a safety communication, not an endorsement: the agency says it “is aware of recent reports of fires that occurred with HBOT devices that resulted in serious injuries and deaths,” that the root cause is not known, and it asks providers to follow instructions for use, maintain fire prevention measures, train and maintain staff, monitor patients for the entire treatment, keep up cleaning and maintenance intervals, avoid prohibited electrical and static devices, and dress patients in hyperbaric-compatible material such as cotton.

The Undersea and Hyperbaric Medical Society’s fifteenth-edition indication list runs to fifteen numbered entries, from air or gas embolism through carbon monoxide poisoning, gas gangrene, crush injury, decompression sickness, problem wounds, severe anaemia, intracranial abscess, necrotising soft tissue infection, refractory osteomyelitis, delayed radiation injury, compromised grafts and flaps, thermal burns, idiopathic sudden sensorineural hearing loss and avascular necrosis. Fertility, sperm quality and gonadal protection appear nowhere on it.

So every human study described above is an off-label use, in a field where the trials are small, unblinded, add-on in design, and concentrated in one country — and where the marketing has already run ahead of the file. There is also a dose mismatch in that market: the mouse work used 2.5 ATA and the clinical reproductive series used medical chambers, while much of the consumer offering is 1.3 to 1.5 ATA in soft-sided vessels — a different device at a different pressure, with a different risk profile.

What would actually settle it

The trial nobody has run is straightforward to describe and hard to fund: randomised and sham-controlled, in men scheduled to receive gonadotoxic chemotherapy, with sperm cryopreservation as the comparator standard rather than an afterthought, with DNA fragmentation and live birth as endpoints, and with serial gonadotrophins and testosterone built in so that the Verratti signal can be confirmed or dismissed.

Until someone runs it, the honest reading of this literature is narrow. In animals, hyperbaric oxygen reliably reduces oxidative and inflammatory injury in testicular tissue, and the newest paper adds a specific molecular pathway to that story. In men, nine small trials suggest sperm parameters and pregnancy rates may improve when the chamber is added to another treatment — a signal strong enough to justify a proper trial and far too weak to justify a purchase. And for the young man facing cyclophosphamide this month, the intervention with the evidence behind it remains the one that happens before the first infusion, in a cryopreservation laboratory, not in a pressure vessel.

Sources

← Back to blog