What is the difference between subq and IM testosterone?

Dose and interval by route
TrialDose and schedulePeople
McFarlandCypionate under the skin, 50 to 100 mg weekly, median 75 mgEleven transgender men9
SprattCypionate or enanthate under the skin weekly, starting at 50 mg, adjusted to 50 to 150 mg (median 75/80 mg)63 transgender men11
KaminetskyEnanthate autoinjector under the skin, 75 mg weekly to start, then 50, 75 or 100 mg from week 7150 men with hypogonadism12

The Depo-Testosterone (cypionate) label gives 50 to 400 mg every two to four weeks for replacement in the hypogonadal male, deep in the gluteal muscle, and the Xyosted (enanthate) label a starting 75 mg under the skin of the abdomen once a week.12

Which route the labels name

The Depo-Testosterone (cypionate) label says it is for intramuscular use only, deep in the gluteal muscle.1 Xyosted's instructions say to inject under the skin of the abdomen, not within 2 inches of the navel, and nowhere else.2 The Azmiro (cypionate) and Hikma (enanthate) labels also name deep gluteal intramuscular injection.34

Of the people who logged testosterone cypionate, 84% logged the muscle and 15% under the skin.15 Of the people who logged testosterone enanthate, 91% logged the muscle and 9% under the skin.15

Blood levels by route

A review in the Journal of Clinical Endocrinology and Metabolism says the evidence is limited but suggests similar doses under the skin and in the muscle give comparable testosterone levels.5 In a crossover pilot of fourteen transgender men already on intramuscular testosterone (Wilson), 3 weeks in the muscle then 8 weeks under the skin, total testosterone exposure was comparable (mean dose-normalized AUC 1.7 ± 0.6 under the skin against 1.9 ± 0.6 nmol·days/L/mg in the muscle), with wide variability.6 In a phase 2 study of thirty-nine men with hypogonadism (Kaminetsky), randomised to 50 or 100 mg of enanthate under the skin weekly for 6 weeks, two 100 mg doses (week 5 and week 6 combined) gave a similar total exposure to 200 mg of enanthate in the muscle in a reference group.7

In a comparison of 234 hypogonadal men (Choi), not randomised by route, trough total testosterone rose from 313.6 to 536.4 ng/dL after 12 weeks of intramuscular cypionate and from 246.6 to 552.8 ng/dL with subcutaneous enanthate, both at 100 mg weekly.8

Needle and injection site

In McFarland, eleven transgender men already on weekly subcutaneous cypionate injected it themselves into abdominal fat with a 25-gauge 5/8” needle on a 1-mL luer lock syringe.9 The Australian Immunisation Handbook, writing about vaccines, gives 25 to 27 gauge and 16 mm for under the skin, and 22 to 25 gauge and 25 mm for the muscle.10

Of the people who logged testosterone enanthate, 39% logged the shoulder and 25% the glute as the site.15

Pain, reactions and preference

In the Spratt retrospective cohort of sixty-three transgender men on weekly subcutaneous cypionate or enanthate, minor and transient local reactions were reported in 9, and all 22 who had switched from intramuscular to subcutaneous preferred subcutaneous and none preferred intramuscular.11

In the 52-week Kaminetsky study of 150 men with hypogonadism who started on 75 mg of enanthate weekly from a subcutaneous autoinjector, more than 95% of patients reported no injection-related pain.12

Hematocrit, blood pressure and monitoring

The Depo-Testosterone label says to check hemoglobin and hematocrit periodically in patients on long-term androgen administration, and to monitor blood pressure periodically, especially in men with hypertension.1 In the 26-week Gittelman study of 133 men aged 18 to 75 on a subcutaneous enanthate autoinjector, 75 mg once weekly adjusted when indicated to 50 or 100 mg, increased hematocrit of 52% or more was the most frequent adverse drug reaction, in 10 patients (7.5%).13

None of the studies read randomised the same ester and dose to the two routes and reported hematocrit; the Wilson crossover measured hemoglobin, 160 ± 9 g/L at the first visit and 153 ± 9 g/L at the last, a difference with p > 0.05.6

Dose and interval by route

The table is set at the top of the page

In the studies read, the subcutaneous trials in men with hypogonadism used the enanthate autoinjector, none of cypionate under the skin turned up, and Choi's comparison differs in ester and device as well as route.

From mg to mL to syringe units

An FDA summary for an insulin syringe says U-100 means there are 100 units of insulin in every mL.14 Of the people who logged testosterone enanthate, 51% set up a 250 mg/mL vial.15 Worked example: 125 mg of testosterone enanthate from a 250 mg/mL vial is 0.5 mL, or 50 units on a U-100 syringe; 8% of the people who logged testosterone enanthate logged 125 mg at 50 units.15

Frequently asked questions

Does a different needle draw up the oil than the one used to inject?

The Australian Immunisation Handbook, written for vaccines, says to draw up with a 19 or 21 gauge needle if required and to change the needle before giving the injection, though using the same needle may be more practicable in mass vaccination clinics.10

Why can testosterone oil look cloudy?

The Hikma enanthate label says a wet needle or wet syringe may make the oil cloudy without changing its potency.4

How the logged figures are counted

Milligram app dose logs, aggregated and anonymous, the 90 days to 17 November 2026. Each figure is a share of people who set up or logged the item, and each share is of at least 50 people.

References

  1. DailyMed (Pfizer label). Depo-Testosterone (testosterone cypionate) injection, label. 2025. dailymed.nlm.nih.gov
  2. DailyMed (Antares Pharma label). Xyosted (testosterone enanthate) injection, label. 2025. dailymed.nlm.nih.gov
  3. DailyMed (Azurity label). Azmiro (testosterone cypionate) injection, label. 2025. dailymed.nlm.nih.gov
  4. DailyMed (Hikma label). Testosterone enanthate injection (Hikma), label. 2025. dailymed.nlm.nih.gov
  5. Journal of Clinical Endocrinology and Metabolism (Figueiredo et al.), PubMed abstract. Testosterone Therapy With Subcutaneous Injections: A Safe, Practical, and Reasonable Option. 2022. eutils.ncbi.nlm.nih.gov
  6. American Journal of Health-System Pharmacy (Wilson et al.), PubMed abstract. Pharmacokinetics, safety, and patient acceptability of subcutaneous versus intramuscular testosterone injection for gender-affirming therapy: A pilot study. 2018. eutils.ncbi.nlm.nih.gov
  7. Sexual Medicine (Kaminetsky, Jaffe, Swerdloff). Pharmacokinetic Profile of Subcutaneous Testosterone Enanthate Delivered via a Novel, Prefilled Single-Use Autoinjector: A Phase II Study. 2015. ebi.ac.uk
  8. Journal of Urology (Choi et al.), PubMed abstract. Comparison of Outcomes for Hypogonadal Men Treated with Intramuscular Testosterone Cypionate versus Subcutaneous Testosterone Enanthate. 2022. eutils.ncbi.nlm.nih.gov
  9. Journal of the Endocrine Society (McFarland et al.). Serum Testosterone Concentrations Remain Stable Between Injections in Patients Receiving Subcutaneous Testosterone. 2017. ebi.ac.uk
  10. Australian Government Department of Health, Disability and Ageing. Administration of vaccines, The Australian Immunisation Handbook. immunisationhandbook.health.gov.au
  11. Journal of Clinical Endocrinology and Metabolism (Spratt et al.), PubMed abstract. Subcutaneous Injection of Testosterone Is an Effective and Preferred Alternative to Intramuscular Injection: Demonstration in Female-to-Male Transgender Patients. 2017. eutils.ncbi.nlm.nih.gov
  12. Journal of Urology (Kaminetsky et al.), PubMed abstract. A 52-Week Study of Dose Adjusted Subcutaneous Testosterone Enanthate in Oil Self-Administered via Disposable Auto-Injector. 2019. eutils.ncbi.nlm.nih.gov
  13. Journal of Sexual Medicine (Gittelman et al.), PubMed abstract. Safety of a New Subcutaneous Testosterone Enanthate Auto-Injector: Results of a 26-Week Study. 2019. eutils.ncbi.nlm.nih.gov
  14. US Food and Drug Administration (device clearance summary). FDA 510(k) summary K233794, insulin syringe. 2023. accessdata.fda.gov
  15. Milligram app dose logs, aggregated and anonymous, the 90 days to 17 November 2026. Each figure is a share of people who set up or logged the item, and each share is of at least 50 people. How these figures are counted