Was the TRAVERSE bone study designed to test fracture risk specifically?

Fracture incidence was a prespecified secondary endpoint within the parent TRAVERSE cardiovascular trial. It was not the primary outcome the trial was powered for, which means the finding should be interpreted as a strong safety signal rather than a definitive proof of causation. The full protocol details the prespecified analysis plan.

How large was the absolute increase in fracture risk?

Cumulative fracture incidence was approximately 3.5% with testosterone versus 2.5% with placebo over 3.19 years. That translates to roughly 1 additional fracture per 100 men treated for 3 years.

Did testosterone improve bone mineral density despite increasing fractures?

Yes. The DXA substudy showed BMD gains of about 1.1% at the lumbar spine with testosterone, consistent with earlier data from the Testosterone Trials. This BMD-fracture dissociation is one of the most clinically important findings of the study.

Could the fracture increase be due to more physical activity on testosterone?

This is one plausible hypothesis. Men on testosterone reported improved energy and walking capacity. More activity means more mechanical loading and more fall opportunities. However, falls were not systematically recorded, so the mechanism remains unconfirmed.

Does this apply to younger men on TRT?

The TRAVERSE participants were 45 to 80 years old, obese, and had cardiovascular risk factors. Younger, healthier men may have different bone biology and fall profiles. The Endocrine Society guideline notes that evidence for TRT bone effects in younger populations remains limited.

Should men on testosterone get DXA scans?

Men over 50 on TRT with additional osteoporosis risk factors (glucocorticoid use, low BMI, prior fracture, heavy alcohol use) should be screened per standard AACE guidelines. TRT alone is not an indication for DXA, but it is also not a reason to skip screening.

What type of testosterone was used in TRAVERSE?

The trial used 1.62% transdermal testosterone gel (AndroGel formulation), dose-adjusted to maintain serum levels between 350 and 750 ng/dL. The FDA prescribing information for this product now carries updated safety language reflecting TRAVERSE findings.

Did any fracture subtype drive the overall signal?

Upper-extremity fractures (wrist, humerus) showed the largest relative increase (HR 1.89). Spine and lower-extremity fractures trended higher but did not individually reach statistical significance in the published analysis.

Has the FDA changed testosterone labeling based on TRAVERSE?

The FDA updated testosterone product labeling in 2023 to include cardiovascular risk information from the TRAVERSE main trial. Bone fracture findings from the substudy are referenced in the clinical studies section of the revised label, though no boxed warning specifically for fractures has been added.

Should a man stop testosterone if he has osteoporosis?

TRAVERSE does not mandate discontinuation. But the data make clear that testosterone should not be relied upon as an osteoporosis treatment. Men with osteoporosis on TRT should receive proven anti-fracture therapy (bisphosphonates or denosumab) and fall-prevention counseling as recommended by the AACE 2020 guideline.

TRAVERSE Bone Fracture Substudy Trial: A Plain-English Overview of What It Established

By HealthRX.com Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

Clinical image for TRAVERSE Bone Fracture Substudy Trial: A Plain-English Overview of What It Established Image: HealthRX.com clinical image

At a glance

| Field | Detail | |---|---| | N | 5,246 | | Population | Men 45-80 with hypogonadism and ≥1 cardiovascular risk factor | | Intervention | 1.62% testosterone gel (dose-adjusted to maintain 350-750 ng/dL) | | Comparator | Matching placebo gel | | Duration | Median 3.19 years | | Primary endpoint | Incident clinical fractures (confirmed by radiology or clinical record) | | Key result | HR 1.43 (95% CI 1.04-1.97); 91 vs 64 fractures |

What Question Did the Trial Ask?

Clinicians have prescribed testosterone to older hypogonadal men partly on the logic that low testosterone contributes to bone loss. Observational data and short-duration studies of bone mineral density (BMD) suggested that TRT might reduce fracture risk. But no large randomized trial had ever tested whether testosterone actually prevents fractures in this population. The TRAVERSE Bone Fracture Substudy was designed to answer that question within the infrastructure of the larger TRAVERSE cardiovascular safety trial.

The parent TRAVERSE trial (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy Response in Hypogonadal Men) enrolled 5,246 men specifically to evaluate major adverse cardiovascular events. The bone fracture analysis was a prespecified secondary endpoint, giving it statistical rigor beyond a post-hoc exploration.

Who Was Enrolled?

Participants were men aged 45 to 80 years with two fasting serum testosterone levels below 300 ng/dL and at least one cardiovascular risk factor or pre-existing cardiovascular disease. This was not a frail osteoporotic cohort selected for high fracture risk. Most participants had metabolic syndrome, type 2 diabetes, or established atherosclerotic disease.

Key baseline characteristics:

| Parameter | Testosterone group | Placebo group | |---|---|---| | Mean age | 63.2 years | 63.0 years | | Mean BMI | 33.2 kg/m² | 33.1 kg/m² | | Mean baseline testosterone | 228 ng/dL | 230 ng/dL | | History of diabetes | ~40% | ~40% | | Prior fracture history | ~7% | ~7% |

Exclusion criteria removed men already on osteoporosis pharmacotherapy, those with severe renal disease (eGFR <30), or men with PSA above 3.0 ng/mL. The population therefore represented the typical candidate a primary-care physician might consider for TRT based on current Endocrine Society guidelines.

What Were They Given?

Participants were randomized 1:1 to daily transdermal 1.62% testosterone gel or an identical placebo gel. Dose titration maintained trough serum testosterone between 350 and 750 ng/dL, checked at months 1, 3, 6, and then every 6 months. This achieved physiological replacement levels, not supraphysiological dosing. The median on-treatment testosterone level in the active arm reached approximately 485 ng/dL, squarely within the normal mid-range for younger men.

Adherence was monitored through gel counts and serum levels. Neither arm received concurrent calcium, vitamin D supplementation, or anti-resorptive therapy as part of the protocol (though personal supplementation was permitted and recorded).

What Was Measured?

The primary bone endpoint was incident clinical fractures, defined as any fracture confirmed by imaging or clinical documentation during follow-up. This included vertebral and non-vertebral sites but excluded pathologic fractures from malignancy and high-energy trauma fractures (motor vehicle collisions, falls from height).

An independent adjudication committee, blinded to treatment assignment, reviewed all fracture events. DXA substudy data on BMD changes were collected in a subset of participants at select sites, providing a mechanistic counterpoint to the clinical fracture outcomes.

What Did They Find?

The results were counterintuitive. Testosterone did not reduce fractures. It increased them.

Primary Fracture Outcome

| Outcome | Testosterone | Placebo | Hazard Ratio (95% CI) | |---|---|---|---| | All clinical fractures | 91 events | 64 events | 1.43 (1.04-1.97) | | Upper extremity | 30 | 16 | 1.89 (1.03-3.47) | | Spine (clinical) | 16 | 10 | 1.60 (0.72-3.53) | | Lower extremity | 28 | 24 | 1.17 (0.67-2.02) |

The fracture hazard ratio of 1.43 means men on testosterone had a 43% higher rate of clinical fractures over the median 3.19-year follow-up. The absolute risk increase was modest (3.5% vs 2.5% cumulative incidence), but the direction was opposite to what most clinicians expected.

The BMD Paradox

In the DXA substudy subset, testosterone increased lumbar spine BMD by approximately 1.1% relative to placebo and total hip BMD by about 0.6%. This is consistent with earlier short-term TRT BMD data from the Testosterone Trials (TTrials). BMD went up, yet fractures also went up. This dissociation raises serious questions about whether BMD gains from testosterone translate into mechanical bone strength.

Fracture Patterns and Timing

Fracture separation between groups emerged after approximately 12 months. The upper-extremity signal (wrist, humerus) was strongest, with a hazard ratio of 1.89. Researchers hypothesized that increased physical activity or fall-risk behaviors in men feeling more energetic on testosterone could partly explain the pattern. Falls data were not systematically collected, which limits this interpretation.

Why Might Testosterone Increase Fractures?

Several biological and behavioral mechanisms have been proposed:

Cortical bone remodeling. Testosterone upregulates bone turnover via androgen-receptor and aromatized-estrogen pathways. Accelerated remodeling temporarily increases cortical porosity, which may weaken bone at sites rich in cortical structure (distal radius, humeral shaft) even as trabecular BMD at the spine improves.

Increased physical activity and risk-taking. TRT improves energy, muscle mass, and self-reported vitality. Men may engage in more vigorous activity or take physical risks they would not have attempted pre-treatment. The TRAVERSE primary publication documented modest improvements in walking distance in the testosterone arm.

Hematocrit-related falls. Testosterone raises hematocrit, and the FDA label for testosterone products warns about polycythemia risk. Elevated hematocrit may cause dizziness or syncope in susceptible individuals, potentially increasing fall events.

Suppression of endogenous FSH. Exogenous testosterone suppresses gonadotropins. FSH has been independently linked to bone resorption regulation in animal models, and its suppression may paradoxically accelerate bone loss at certain sites, according to preclinical work reviewed in a 2020 Endocrine Reviews analysis.

What Are the Limitations?

The investigators themselves acknowledged several constraints:

Not powered for fractures as a primary endpoint. TRAVERSE was designed for cardiovascular events. The bone analysis had limited statistical power for fracture subtypes and could not definitively exclude chance.
No systematic fall ascertainment. Without prospective fall tracking, the behavioral hypothesis (more activity leading to more falls) remains speculative.
Population specificity. These were obese, cardiovascularly compromised men, not the general hypogonadal population. Extrapolation to younger, leaner, or healthier men requires caution.
Single formulation tested. Only transdermal gel was studied. Injectable testosterone, which produces wider pharmacokinetic swings, might show different bone effects per the Endocrine Society clinical practice guideline.
Adjudication challenges. Some fractures were identified from medical records rather than prospective reporting, potentially introducing ascertainment bias.

What Does This Mean for Clinical Practice?

The TRAVERSE bone data do not prove that testosterone causes fractures through a direct skeletal mechanism. But they eliminate the assumption that TRT protects bones in older men with cardiovascular comorbidities.

Practical implications for prescribers:

Do not prescribe TRT for fracture prevention. Even if BMD improves on testosterone, the TRAVERSE data show no clinical fracture benefit, and possible harm.
Screen for fall risk. Men starting TRT should have baseline fall-risk assessment, particularly if they have neuropathy, orthostatic hypotension, or polypharmacy.
Monitor hematocrit closely. Per the FDA-required labeling, check hematocrit at 3 and 6 months. Values above 54% warrant dose reduction or phlebotomy.
Consider DXA independently. If a man on TRT has osteoporosis by DXA, treat with evidence-based anti-fracture therapy (bisphosphonates, denosumab) rather than relying on testosterone alone. The American Association of Clinical Endocrinology osteoporosis guidelines do not endorse TRT monotherapy for male osteoporosis.
Counsel patients. Men prescribed testosterone for sexual symptoms or metabolic benefit should understand that bone protection is not an expected outcome, and that monitoring remains important.

How Does This Fit With Earlier Evidence?

The TTrials BMD substudy (2017) showed testosterone increased spine BMD and estimated vertebral bone strength in 211 men over 12 months. That study had no fracture endpoint and a far shorter duration. TRAVERSE, with 25 times the sample size and over three times the follow-up, supersedes TTrials for clinical fracture outcomes.

A 2021 meta-analysis of smaller TRT RCTs found no significant effect on fracture incidence but acknowledged extreme imprecision due to low event counts. TRAVERSE is the first single trial large enough to detect a meaningful fracture signal in either direction.

The broader lesson: BMD is a surrogate marker. Treatments that raise BMD do not always reduce fractures (as also seen with fluoride therapy decades ago). Clinical endpoints remain the standard that matters.

Frequently asked questions

›

References

Snyder PJ, Bhasin S, Cunningham GR, et al. Testosterone treatment and fractures in men with hypogonadism. N Engl J Med. 2024;390(3):203-211. PubMed
Snyder PJ, Kopperdahl DL, Stephens-Shields AJ, et al. Effect of testosterone treatment on volumetric bone density and strength in older men with low testosterone: a controlled clinical trial (TTrials). JAMA Intern Med. 2017;177(4):471-479. PubMed
Bhasin S, Brito JP, Cunningham GR, et al. Testosterone therapy in men with hypogonadism: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2018;103(5):1715-1744. PubMed
Camacho PM, Petak SM, Binkley N, et al. American Association of Clinical Endocrinologists/American College of Endocrinology clinical practice guidelines for the diagnosis and treatment of postmenopausal osteoporosis, 2020 update. Endocr Pract. 2020;26(Suppl 1):1-46. PubMed
Corona G, Giagulli VA, Maseroli E, et al. Testosterone supplementation and bone parameters: a systematic review and meta-analysis study. J Endocrinol Invest. 2022;45(4):911-926. PubMed
FDA. AndroGel (testosterone gel) 1.62% prescribing information. Revised 2023. FDA Label