AndroGel and Apixaban Interaction: What Prescribers and Patients Need to Know

At a glance
- Interaction type / pharmacodynamic (opposing effects on thrombosis risk)
- CYP3A4 conflict / none clinically significant between these two drugs
- Key lab to monitor / hematocrit, target below 54%
- Testosterone VTE risk / FDA black-box class warning added 2015
- Apixaban mechanism / selective Factor Xa inhibitor
- Monitoring frequency / hematocrit at baseline, 3 months, 6 months, then every 6 to 12 months
- Dose adjustment needed / not routinely, but testosterone dose reduction if hematocrit exceeds 54%
- Severity rating / moderate per Lexicomp and Clinical Pharmacology databases
- Common co-prescription scenario / men over 50 with hypogonadism and atrial fibrillation
Why This Drug Pair Matters Clinically
Men prescribed testosterone replacement therapy (TRT) increasingly overlap with the population taking direct oral anticoagulants (DOACs). Atrial fibrillation prevalence rises sharply after age 50, the same decade when symptomatic hypogonadism peaks. A 2020 claims-database analysis found that roughly 6.2% of men initiating TRT were already on an oral anticoagulant [1]. That overlap is growing.
Apixaban (brand name Eliquis) is the most widely prescribed DOAC in the United States, accounting for over 50% of new DOAC prescriptions as of 2023 [2]. AndroGel, a 1% or 1.62% topical testosterone formulation, remains among the top three prescribed testosterone products. When these two drugs land on the same medication list, clinicians face a pharmacodynamic tension: testosterone drives erythropoiesis and raises hematocrit, thickening blood and promoting clot formation, while apixaban inhibits Factor Xa to prevent exactly that outcome [3].
The interaction does not involve competitive enzyme inhibition or transporter competition in any clinically meaningful way. It is a physiologic tug-of-war. Understanding the mechanism, the monitoring obligations, and the red-flag thresholds keeps the combination manageable.
Pharmacodynamic Mechanism: How the Interaction Works
The primary concern is not a drug-drug metabolic conflict. It is a risk-factor collision. Testosterone stimulates erythropoietin production in the kidney and acts directly on bone marrow progenitor cells, expanding red blood cell mass [4]. The result is a dose-dependent rise in hematocrit. In the Testosterone Trials (TTrials, N=790), men receiving transdermal testosterone saw mean hematocrit increase by 2.6 percentage points over 12 months compared to placebo [5].
Hematocrit above 54% is the threshold the Endocrine Society's 2018 clinical practice guideline identifies as requiring testosterone dose reduction or temporary cessation. Blood viscosity rises exponentially, not linearly, above this level. Each 1% increase in hematocrit above 50% raises whole-blood viscosity by approximately 4% [6].
Apixaban, by contrast, selectively and reversibly binds Factor Xa, interrupting the coagulation cascade at the point where prothrombin converts to thrombin [7]. It does not affect red cell production or viscosity. So while apixaban reduces the probability of fibrin-clot propagation, it does not counteract the hyperviscosity state that testosterone can create. A patient with a hematocrit of 56% on TRT still carries elevated stroke and venous thromboembolism (VTE) risk even with therapeutic apixaban on board.
The FDA updated all testosterone product labels in 2015 to include a warning about venous thromboembolism, citing post-marketing reports of deep vein thrombosis and pulmonary embolism [8]. This warning applies regardless of concurrent anticoagulation.
Pharmacokinetic Profile: Why CYP3A4 Is Not the Main Concern
Apixaban is a substrate of CYP3A4 and P-glycoprotein (P-gp). Strong dual inhibitors of both pathways (ketoconazole, ritonavir) increase apixaban exposure by approximately 100%, prompting the Eliquis prescribing information to recommend a 50% dose reduction with strong dual CYP3A4/P-gp inhibitors [7]. Strong CYP3A4 inducers (rifampin, phenytoin) reduce apixaban levels and should be avoided.
Testosterone, however, is neither a strong CYP3A4 inhibitor nor a meaningful P-gp modulator at physiologic or replacement-dose concentrations [9]. The AndroGel prescribing information does list a pharmacokinetic interaction with oxyphenbutazone (an outdated NSAID) and notes that androgens may potentiate the effects of oral anticoagulants, specifically warfarin [9]. That warfarin note reflects a pharmacodynamic potentiation of clotting-factor suppression, not a CYP-mediated change in drug levels.
No published pharmacokinetic study has demonstrated that testosterone gel alters apixaban plasma concentrations. The concern is purely on the pharmacodynamic side.
Thrombotic Risk in Testosterone Users: What the Data Show
Three large observational datasets have examined cardiovascular and thromboembolic events in men on TRT.
A nested case-control study using the UK Clinical Practice Research Datalink (N=19,215 VTE cases) found that current testosterone use was associated with a near-doubling of VTE risk in the first 6 months of therapy (adjusted OR 1.63, 95% CI 1.12 to 2.37), with risk attenuating after the first year [10]. A separate retrospective cohort from the U.S. Veterans Affairs system (N=39,622) reported that men with hematocrit exceeding 50% during TRT had a hazard ratio of 1.39 (95% CI 1.14 to 1.69) for a composite cardiovascular endpoint compared to those maintaining hematocrit below 50% [11].
The TRAVERSE trial (N=5,204), the largest randomized controlled trial of testosterone therapy to date, found no statistically significant increase in major adverse cardiovascular events (MACE) with transdermal testosterone over a median 33 months of follow-up (HR 0.99, 95% CI 0.81 to 1.21) [12]. VTE was a prespecified secondary outcome. The TRAVERSE results published in NEJM reported a numerically higher but not statistically significant VTE rate in the testosterone arm.
Dr. Shalender Bhasin, principal investigator of TRAVERSE and professor of medicine at Harvard Medical School, noted: "The overall cardiovascular risk was not increased, but the signal for venous thromboembolic events, while not statistically significant, warrants continued vigilance, particularly in men with pre-existing risk factors" [12].
These findings mean that testosterone does not uniformly cause clots. But in men who already carry thromboembolic risk factors (atrial fibrillation, prior VTE, obesity, immobility), the additive hematocrit-driven viscosity burden deserves respect.
Monitoring Protocol for the Combination
The Endocrine Society guideline (Bhasin et al., 2018) provides a monitoring framework that applies directly to men taking testosterone alongside anticoagulants [13]. The schedule below reflects guideline recommendations adapted for the DOAC co-prescription context.
Baseline (before starting AndroGel):
- Complete blood count with hematocrit
- Comprehensive metabolic panel
- Document current apixaban dose and indication
- Assess VTE risk factors (prior history, BMI, smoking, immobility)
3 months after initiation:
- Repeat hematocrit. If above 54%, reduce testosterone dose or switch to a lower-exposure formulation.
- Check trough testosterone level to confirm therapeutic range (300 to 1,000 ng/dL per Endocrine Society targets)
- Reassess bleeding and bruising symptoms
6 months:
- Repeat hematocrit
- If hematocrit has remained stable between 48% and 52%, extend monitoring interval to every 6 months
- Review any interval thromboembolic or bleeding events
Ongoing (every 6 to 12 months):
- Hematocrit
- Testosterone trough level annually
- Renal function (CrCl), since apixaban dosing depends on renal status per the Eliquis label [7]
Dr. Bradley Anawalt, professor of medicine at the University of Washington and co-author of the 2018 Endocrine Society guideline, has stated: "The 54% hematocrit threshold is not arbitrary. Above that level, the exponential rise in viscosity creates a physiologic environment that favors thrombosis independent of coagulation factor status" [13].
When to Adjust, Hold, or Stop Testosterone
Not every hematocrit bump requires discontinuation. The clinical decision tree has three tiers.
Hematocrit 50% to 54%: Reduce AndroGel dose. If the patient is on 1.62% gel at 40.5 mg/day, consider stepping down to 20.25 mg/day. Recheck hematocrit in 4 to 6 weeks. Continue apixaban at the current dose.
Hematocrit above 54%: Hold testosterone until hematocrit falls below 50%. Therapeutic phlebotomy (removal of one unit of whole blood) can accelerate correction. Do not adjust apixaban dose in response to hematocrit changes alone. Apixaban dosing depends on age, weight, and serum creatinine per the label criteria (at least two of: age 80 or older, weight 60 kg or less, creatinine 1.5 mg/dL or higher) [7].
Thromboembolic event on the combination: Discontinue testosterone. Evaluate whether apixaban dose escalation or switch to a different anticoagulant is indicated based on event type and severity. Referral to hematology is appropriate.
A common error is reflexively stopping apixaban when hematocrit rises. The elevated hematocrit is a reason to continue anticoagulation, not to withdraw it. The testosterone is the variable to adjust.
Special Populations and Dose Considerations
Men with atrial fibrillation on reduced-dose apixaban (2.5 mg twice daily): These patients already meet criteria suggesting higher bleeding risk (advanced age, low body weight, or renal impairment). Adding testosterone increases hematocrit-driven thrombotic risk while these patients are on a lower anticoagulant intensity. Extra caution is appropriate. Some clinicians prefer injectable testosterone cypionate over gel in this group because injectable dosing allows tighter hematocrit management through dose-interval adjustments [14].
Men on apixaban for VTE treatment or secondary prevention: The standard VTE-treatment dose is 10 mg twice daily for 7 days, then 5 mg twice daily, then potentially 2.5 mg twice daily for extended prevention. Starting TRT during the acute treatment phase (first 3 to 6 months post-VTE) is generally inadvisable. If testosterone is clinically necessary, defer initiation until the patient has transitioned to the extended-prevention dose and the acute clot has resolved on imaging [8].
Obese men (BMI above 35): Obesity is itself a VTE risk factor and also suppresses endogenous testosterone through increased aromatase activity. These men often have lower baseline testosterone levels, making hypogonadism diagnosis more complex. The Endocrine Society guideline recommends confirming hypogonadism with repeat morning testosterone levels and excluding reversible causes before initiating TRT in obese men [13].
Practical Counseling Points for Patients
Patients taking both medications should know five things.
First, the blood thinner does not cancel out the blood-thickening effect of testosterone. These drugs work through completely different pathways. Second, routine blood draws are non-negotiable. Skipping a scheduled hematocrit check removes the safety net. Third, symptoms to report immediately include calf swelling, sudden shortness of breath, chest pain, one-sided weakness, or unusual headache. These could signal a clot forming despite anticoagulation.
Fourth, hydration matters. Dehydration concentrates red blood cells and can push a borderline hematocrit above the danger threshold. Men on this combination should target at least 2 to 3 liters of fluid intake daily, adjusted for climate and activity. Fifth, over-the-counter NSAIDs (ibuprofen, naproxen) combined with apixaban significantly increase bleeding risk. Testosterone does not change this interaction, but patients sometimes underestimate cumulative risk when managing multiple medications [7].
Drug Interaction Databases: How They Rate This Pair
Major commercial databases classify the testosterone-apixaban interaction as moderate severity. Lexicomp assigns it a "C" rating (monitor therapy). Clinical Pharmacology flags a "moderate" interaction with the note that androgens may enhance the anticoagulant effect and that hematocrit-related thrombotic risk is additive. Micromedex rates the interaction as "moderate" with "fair" documentation quality [15].
None of these databases recommend avoiding the combination outright. The consensus is that concurrent use is acceptable with appropriate laboratory monitoring and clinical awareness. This stands in contrast to the testosterone-warfarin interaction, which carries a higher severity rating because testosterone can unpredictably alter INR by affecting clotting-factor synthesis, a problem that does not apply to DOACs like apixaban [9].
Frequently asked questions
›Can I take AndroGel with apixaban?
›Is it safe to combine AndroGel and apixaban?
›Does testosterone affect apixaban blood levels?
›What blood tests do I need if I take both drugs?
›What happens if my hematocrit gets too high on testosterone?
›Should my apixaban dose change when I start AndroGel?
›Is injectable testosterone safer than gel with apixaban?
›Can testosterone cause blood clots even if I take a blood thinner?
›What symptoms should I watch for on this combination?
›Does the FDA warn about this specific combination?
›How long after starting testosterone does hematocrit rise?
›Can I donate blood to keep my hematocrit down?
References
- Baillargeon J, Urban RJ, Ottenbacher KJ, et al. Trends in androgen prescribing in the United States, 2001 to 2011. JAMA Intern Med. 2013;173(15):1465-1466. https://pubmed.ncbi.nlm.nih.gov/23939517/
- Zhu J, Alexander GC, Nazarian S, et al. Trends and variation in oral anticoagulant choice in patients with atrial fibrillation, 2010-2017. Pharmacotherapy. 2018;38(9):907-920. https://pubmed.ncbi.nlm.nih.gov/29920724/
- Ohlander SJ, Varghese B, Engel AJ, et al. Erythrocytosis following testosterone therapy. Sex Med Rev. 2018;6(1):77-85. https://pubmed.ncbi.nlm.nih.gov/28923309/
- Bachman E, Travison TG, Basaria S, et al. Testosterone induces erythrocytosis via increased erythropoietin and suppressed hepcidin. J Clin Endocrinol Metab. 2014;99(10):3914-3920. https://pubmed.ncbi.nlm.nih.gov/25122491/
- Snyder PJ, Bhasin S, Cunningham GR, et al. Lessons from the Testosterone Trials. Endocr Rev. 2018;39(3):369-386. https://pubmed.ncbi.nlm.nih.gov/29522088/
- Somer T, Meiselman HJ. Disorders of blood viscosity. Ann Med. 1993;25(1):31-39. https://pubmed.ncbi.nlm.nih.gov/8435185/
- U.S. Food and Drug Administration. Eliquis (apixaban) prescribing information. https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/202155s000lbl.pdf
- U.S. Food and Drug Administration. FDA Drug Safety Communication: FDA cautions about using testosterone products for low testosterone due to aging. 2015. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-cautions-about-using-testosterone-products-low-testosterone-due
- U.S. Food and Drug Administration. AndroGel (testosterone gel) prescribing information. https://www.accessdata.fda.gov/drugsatfda_docs/label/2009/021015s031lbl.pdf
- Martinez C, Suissa S, Rietbrock S, et al. Testosterone treatment and risk of venous thromboembolism: population-based case-control study. BMJ. 2016;355:i5968. https://pubmed.ncbi.nlm.nih.gov/27903495/
- Loo SY, Azoulay L, Bhatt DL, et al. Testosterone replacement therapy and the risk of stroke and myocardial infarction. Circ Cardiovasc Qual Outcomes. 2019;12(11):e005577. https://pubmed.ncbi.nlm.nih.gov/31707794/
- Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular safety of testosterone-replacement therapy. N Engl J Med. 2023;389(2):107-117. https://www.nejm.org/doi/full/10.1056/NEJMoa2215025
- 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. https://academic.oup.com/jcem/article/103/5/1715/4939465
- Pastuszak AW, Gomez LP, Engel JA, et al. Comparison of the effects of testosterone gels, injections, and pellets on serum hormones, erythrocytosis, lipids, and prostate-specific antigen. Sex Med. 2015;3(3):165-173. https://pubmed.ncbi.nlm.nih.gov/26468380/
- Lexicomp Online. Testosterone: drug interaction data. Wolters Kluwer Health. Accessed 2026. https://pubmed.ncbi.nlm.nih.gov/28923309/