Evidence question

PubMed sourcedAbstract-based analysisUpdated Sep 2026Human reviewedUpdated Aug 2026

Evidence / Men's Hormone Health

CoverageMen's Hormone Health

Do transdermal and injectable testosterone differ in safety or efficacy?

Reviewed by Neal Rouzier, MD, Preventive Medicine · Aug 2026

Updated Aug 2026·14 studies analyzed·2 core studies·Applies to: hypogonadal men

Bottom line

All FDA-approved formulations can restore testosterone when dosed appropriately. Transdermal routes may have less erythrocytosis than injectable esters in some comparisons, but transfer, adherence, and peak-trough patterns differ by formulation.[1–6]

Evidence confidence

Evidence Confidence

Moderate
LimitedModerateHigh

Relevant evidence exists, but study designs, populations, or conclusions differ enough to keep certainty moderate.

Key uncertainty. Head-to-head cardiovascular outcome trials by formulation are lacking.

Ask RAGMD

Get the live evidence synthesis

10 complimentary queries. Full clinical intelligence — PubMed citations, conflict analysis, Evidence Confidence — with 7-day free trial.

Also asked

TRT gel vs injection▼

All FDA-approved formulations can restore testosterone when dosed appropriately. Transdermal routes may have less erythrocytosis than injectable esters in some comparisons, but transfer, adherence, and peak-trough patterns differ by formulation.

Testosterone formulation comparison▼

All FDA-approved formulations can restore testosterone when dosed appropriately. Transdermal routes may have less erythrocytosis than injectable esters in some comparisons, but transfer, adherence, and peak-trough patterns differ by formulation.

Body of evidence

Clinical Evidence Brief

RAGMD analyzed 13 relevant studies.

Overall finding

2 studies address sleep-related cognitive symptoms, with weighted findings showing a positive association. 3 studies address bone mineral density, with weighted findings showing no meaningful effect. 2 studies address venous thromboembolism, with weighted findings showing no meaningful effect. 1 study addresses weight / body composition, with weighted findings showing unclear findings. 6 randomized or systematic-review reports and 7 observational studies address treatment. Choose formulation based on patient preference, adherence, transfer risk, peak-trough tolerance, and monitoring response rather than a one-size-fits rule.

Evidence by outcome

Direction: ↑ Favors treatment / reassuring · ↓ Harm signal · ↔ No clear effect · ± Mixed · ↗ Observational link · ? Unclear

Positive link

Sleep-related cognitive symptoms

high · 2 studies

2 studies address sleep-related cognitive symptoms, with weighted findings showing a positive association.

Mixed measurement

No clear effect

Bone mineral density

moderate · 3 studies

3 studies address bone mineral density, with weighted findings showing no meaningful effect. This is an objectively measured outcome. This is a surrogate endpoint and does not by itself prove a hard clinical outcome.

Objectively measured · Surrogate endpoint

No clear effect

Venous thromboembolism

moderate · 2 studies

2 studies address venous thromboembolism, with weighted findings showing no meaningful effect. This is an objectively measured outcome. This is a hard clinical endpoint.

Objectively measured · Hard clinical endpoint

Unclear

Weight / body composition

moderate · 1 study

1 study addresses weight / body composition, with weighted findings showing unclear findings. This is an objectively measured outcome. This is a surrogate endpoint and does not by itself prove a hard clinical outcome.

Objectively measured · Surrogate endpoint

Unclear

Quality of life

moderate · 1 study

1 study addresses quality of life, with weighted findings showing unclear findings. This is a patient-reported / subjective outcome.

Patient-reported

No clear effect

Fractures

moderate · 1 study

1 study addresses fractures, with weighted findings showing no meaningful effect. This is an objectively measured outcome. This is a hard clinical endpoint.

Objectively measured · Hard clinical endpoint

Magnitude / clinical importance

Validated effect estimates in the analyzed set include or 10. Statistical significance is not the same as clinical importance.

Interpret magnitude in absolute terms for the patient in front of you, not from relative estimates alone.

Agreement across studies

Most studies agree on quality of life (unclear findings), while bone mineral density findings are no meaningful effect and more consistent across domains.

Supports main conclusion: 1 · Neutral / mixed: 12 · Credible conflicting: 0

Study counts describe the landscape; RAGMD weights studies according to design, quality, and directness.

Conflicting findings

  • The results indicated that OSA is significantly correlated with the decrease in serum testosterone levels in men.
  • Estrogens primarily regulate bone homeostasis in adult men, and testosterone and estradiol levels must decline substantially to impact the skeleton.
  • It has generally been held that estrogen and testosterone are the major sex steroids regulating bone metabolism in women and men, respectively.
  • In men with hypogonadism and preexisting or a high risk of cardiovascular disease, testosterone-replacement therapy was noninferior to placebo with respect to the incidence of major adverse cardiac e…

Risks & harms

4 studies report safety-relevant findings. Design mix: Randomized controlled trial, Study design not clearly classified, Narrative review. Rare-event inference should not rest on underpowered trials alone.

Who this applies to

Most studies included hypogonadal men.

What remains uncertain

  • Head-to-head cardiovascular outcome trials by formulation are lacking.
  • Head-to-head cardiovascular outcome trials by formulation are lacking.
  • Surrogate improvement does not prove a hard clinical outcome.

Clinical takeaway

Choose formulation based on patient preference, adherence, transfer risk, peak-trough tolerance, and monitoring response rather than a one-size-fits rule. Injectable routes may raise hematocrit more in some patients; transdermal routes require consistent application and transfer precautions.

Positive link

Sleep-related cognitive symptoms?

A positive association

high[34536053 · 35904664]

No clear effect

Bone mineral density?

No meaningful effect

moderate[19820017 · 26901812 · 11730247]

No clear effect

Venous thromboembolism?

No meaningful effect

moderate[37326322 · 26205547]

Unclear

Weight / body composition?

Unclear findings

moderate[35904664]

Unclear

Quality of life?

Unclear findings

moderate[35904664]

No clear effect

Fractures?

No meaningful effect

moderate[19820017]

Unclear

Hematocrit / erythrocytosis?

Unclear findings

limited[35467476]

Unclear

Fertility / spermatogenesis?

Unclear findings

limited[27855957]

6 randomized or systematic-review reports and 7 observational studies address treatment. Weighted treatment findings show mixed findings.

6 RCT / systematic review · 7 observational

Large randomized trials strongly influence the conclusion

2 landmark studies with randomized or systematic-review designs strongly inform this conclusion.

Supporting studies provide additional context

12 supporting studies add context beyond the core evidence base.

Regimen & duration

Observational and supporting studies provide context and do not outrank higher-appropriateness designs.

Long-term uncertainty

Study quality or consistency across available evidence limits certainty.

14 analyzed studies

Strongest evidence

The aging male : the official journal of the International Society for the Study of the Aging Male2022

Landmark studyPMID 35467476

The HEAT-Registry (HEmatopoietic Affection by Testosterone): comparison of a transdermal gel vs long-acting intramuscular testosterone undecanoate in hypogonadal men.

T substitution with intramuscular TU or T gel increase T concentrations effectively. Long-acting TU leads to a higher rate of hematocrit levels >50%, whilst at the same time it seems to be more efficient to ameliorate anemia in the subgroup of respectively affected hypogonadal patients .

Study
Evidence
Relevance
The New England journal of medicine2023

Landmark studyPMID 37326322

Cardiovascular Safety of Testosterone-Replacement Therapy.

In men with hypogonadism and preexisting or a high risk of cardiovascular disease, testosterone-replacement therapy was noninferior to placebo with respect to the incidence of major adverse cardiac events. (Funded by AbbVie and others; TRAVERSE ClinicalTrials.gov number, NCT03518034.).

RCT
Evidence
Relevance

Supporting literature

Association between obstructive sleep apnea and male serum testosterone: A systematic review and meta-analysis.

Obstructive sleep apnea and serum total testosterone: a system review and meta-analysis.

Effects of aromatase inhibition on bone mineral density and bone turnover in older men with low testosterone levels.

+9 more

More in this focus area

Related evidence analyses

Other structured clinical questions in Men's Hormone Health.

Go deeper on Men's Hormone Health

Full evidence synthesis, study comparisons, conflicting evidence, and expert commentary are available with RAGMD. Ask follow-up questions, explore related topics, and see how the evidence has evolved.