TRT and Heart Health: The Definitive Guide

What do we know about testosterone therapy and the heart?

In the largest trial so far, testosterone therapy (TRT) did not raise the risk of heart attack, stroke or death from heart disease. That trial, TRAVERSE, followed about 5,200 men at high heart risk who used a daily testosterone gel. It did find more of three other problems: an irregular heart rhythm called atrial fibrillation, sudden kidney injury, and blood clots in the lungs. Separate studies show that every testosterone product can raise blood pressure a little. [1], [2], [3]

The worry started between 2010 and 2015, when a small trial and several records studies suggested more heart problems in men on testosterone. The US Food and Drug Administration (FDA) then asked testosterone makers to run a large safety trial, and TRAVERSE was that trial. In February 2025 the FDA told testosterone makers to add its results to their labels, recommended removing boxed-warning language about heart outcomes, and required a blood-pressure warning on every product. Some labels have not been updated yet. [3]

"Heart health" covers separate questions with different evidence: heart attacks and strokes (MACE), blood pressure, heart rhythm, blood clots, heart failure, cholesterol and the kidneys. This guide takes them one at a time.

TRT and heart health: the key numbers

This table shows what each major clinical guideline and product label says about the heart and blood clots before and during testosterone treatment. The rules are what each document says, checked in September 2026.

[6], [7], [11], [8], [5], [12], [13], [14], [15], [16], [2], [17], [18], [19]

No guideline sets a blood-pressure, heart-rate or heart-tracing (ECG) number for starting or stopping testosterone. What most of them share: wait 3 to 6 months after a heart attack or stroke (4 months in the VA's rules), avoid uncontrolled heart failure, ask about personal and family clot history, and keep hematocrit at or below 54%. US labels add one blood-pressure rule: do not use testosterone with uncontrolled high blood pressure. [6], [7], [8], [5], [2]

Why do guidelines differ?

Mostly on timing and severity. The waiting time after a heart attack or stroke ranges from 3 to 6 months, and the EAA says only "recent". Heart failure rules range from any uncontrolled heart failure (Endocrine Society, EAU) to only the most severe class (BSSM). On clots, the AUA said in 2018 that there was no definitive evidence linking testosterone to them, while the Endocrine Society and the VA rule out thrombophilia. The AUA and Endocrine Society guidelines predate TRAVERSE; the EAU 2026 edition, the VA's 2026 rules and the Endocrine Society's 2026 statement came after it. [13], [12], [6], [7], [8], [5], [11]

How do the units work?

Blood pressure is written as two numbers in millimeters of mercury (mm Hg): the top number, systolic, is the pressure when the heart beats, and the bottom number, diastolic, is the pressure between beats. A rise of "3.9/1.5 mm Hg" means systolic up 3.9 and diastolic up 1.5.

Testosterone levels appear in ng/dL (US) or nmol/L (most other countries). TRAVERSE enrolled men with two fasting levels under 300 ng/dL (10.4 nmol/L) and adjusted the gel to keep testosterone between 350 and 750 ng/dL (about 12 to 26 nmol/L). The testosterone unit converter switches between the two. [1]

How can testosterone affect the heart and blood vessels?

Through several separate routes, and they do not all point the same way. Some are measured in people on treatment; others are links seen in records or genetic studies that cannot prove cause. [3], [2], [22]

A high hematocrit may explain part of the risk, but no trial has shown that it causes heart events; the hematocrit on TRT guide covers it in full. Genetic (Mendelian randomization) studies of lifelong natural testosterone disagree: one found no direct effect on heart artery disease, and a 2026 study found about 17% higher odds, apparently through blood pressure. Neither tests treatment. [22], [30], [31]

How did the worry about TRT and the heart start?

With one small trial stopped early and a run of records studies between 2010 and 2015. Larger trials and pooled analyses since then have not confirmed a rise in heart attacks or strokes, but the early studies explain why labels, guidelines and many doctors are cautious. [32], [33], [34], [1]

• 2010: the TOM trial. In 209 frail men aged 65 or older using testosterone gel for 6 months, the safety board stopped the trial early after 23 men on testosterone and 5 on placebo had heart-related adverse events. These were reported events, not a planned heart outcome, and the authors said the small size and unusual group "prevent broader inferences". [32]
• 2013–2014: pooled trials and records studies. A meta-analysis of 27 short trials found more heart-related events on testosterone (odds ratio 1.54), though not in industry-funded trials. A VA records study reported deaths, heart attacks or strokes in 25.7% of men on testosterone vs 19.9% by 3 years; it was later corrected, and the likely range for the difference includes zero. An insurance study found 36% more heart attacks in the 90 days after a first prescription (rate ratio 1.36; the abstract gives no absolute counts). [35], [33], [34]
• 2014: regulators respond. The FDA began looking into reports of strokes, heart attacks and deaths, added a warning about blood clots in the veins to every US testosterone label, and held an advisory committee that led to the call for a large industry trial. A European review found no consistent evidence of higher heart risk. [3], [36], [37]
• 2015–2018: more caution, more data. A March 2015 FDA safety communication required label changes about a possible heart risk and urged caution for low levels caused by aging alone. In 2018 the FDA required 24-hour blood-pressure studies for every product. [3]
• 2023: TRAVERSE published. See the next section. [1]
• 2025: relabeling. In February the FDA told testosterone makers to add the TRAVERSE results to their labels, recommended removing boxed-warning language about heart outcomes and required blood-pressure warnings for every product. In July 2025 the boxed blood-pressure warnings on Xyosted, Jatenzo, Tlando and Kyzatrex were removed, and blood pressure became a standard warning. [3], [38]
• 2026: more changes requested. In June 2026 HHS announced that the FDA is requesting further label changes (dropping the limit on use for aging-related low testosterone, narrowing the prostate cancer warning to metastatic disease, and revising the enlarged-prostate warning). These are requests, not yet in any label. Checked September 2026. [39], [40]

Doserly

How the heart question developed, 2010 to 2026

1. 2010Study

   TOM trial stopped early: 23 vs 5 men with heart-related adverse events (209 frail men aged 65+)

2. 2013Study

   Meta-analysis of 27 small trials: odds ratio 1.54 for heart-related events

3. 2013Study

   VA records study: 25.7% vs 19.9% at 3 years (later corrected)

4. 2014Study

   Insurance study: 36% more heart attacks in the 90 days after a first prescription

5. 2014Regulator

   FDA: vein-clot warning on all US labels; advisory committee calls for a large trial

6. Nov 2014Regulator

   EU review: no consistent evidence of higher heart risk

7. Mar 2015Regulator

   FDA safety communication and label changes

8. 2018Regulator

   FDA requires 24-hour blood-pressure studies for every product

9. 2023Study

   TRAVERSE: 7.0% vs 7.3% (hazard ratio 0.96)

10. Feb 2025Regulator

    FDA tells makers to add TRAVERSE, drop boxed heart language and warn on blood pressure

11. Jul 2025Regulator

    Boxed blood-pressure warnings on Xyosted, Jatenzo, Tlando and Kyzatrex removed

12. Jun 2026Requested, pending

    HHS: FDA requests further label changes (requested, pending; checked September 2026)

Each marker names one study or regulatory action; the June 2026 changes are requested, not yet in any label.

What did the TRAVERSE trial find?

TRAVERSE found that testosterone gel did not raise the combined risk of heart attack, stroke or death from heart disease compared with a placebo gel, in middle-aged and older men at high heart risk. It also found more atrial fibrillation, acute kidney injury and clots in the lungs, and more fractures. [1], [2], [41]

Who was in the trial, and what did they take?

• Men: 5,204 men aged 45 to 80 with two fasting testosterone levels under 300 ng/dL (10.4 nmol/L), symptoms of low testosterone, and either existing heart disease or several heart risk factors. [1], [42]
• Their health: average age 63.3; about 55% already had heart disease; 69% had diabetes, 84% high cholesterol and 93% high blood pressure. [2]
• Treatment: a daily 1.62% testosterone gel or a placebo gel, with the dose adjusted to keep testosterone between 350 and 750 ng/dL. Average testosterone on the gel rose from 220 ng/dL at the start to about 440 ng/dL at 12 months. [1], [2]
• The question: whether testosterone was "noninferior" for major heart events. The trial ran until at least 256 heart events had happened, and counted testosterone as safe enough if the upper end of the likely range for its hazard ratio stayed below 1.5 (noninferiority). [20]

What were the main results?

[1], [21], [2], [41]

The last column shows absolute differences. Percentages are out of the 2,596 men on testosterone gel and 2,602 men on placebo in the trial's safety analysis. The main result had a hazard ratio of 0.96, with a likely range of 0.78 to 1.17, well inside the 1.5 limit. Heart attack, stroke and heart death each "appeared to be similar" in the two groups. The trial's registry record gives a hazard ratio of 1.02 for the wider count that adds stents and bypass, 0.98 for death from any cause and 1.11 for heart failure (likely range 0.76 to 1.62). [1], [21]

Not every rise was statistically clear. Venous clots overall had a hazard ratio of 1.46, but its likely range (0.92 to 2.32) includes no difference; the rise in clots reaching the lungs is the clearer signal, which the Endocrine Society describes as about a 50% relative increase. Blood pressure barely moved: systolic pressure rose 1.0 mm Hg on testosterone and fell 0.5 mm Hg on placebo by 36 months. [21], [11], [2]

Doserly

TRAVERSE: absolute results

Testosterone gel (2,596 men)Placebo gel (2,602 men)

Heart attack, stroke or heart death

Testosterone 7.0% vs placebo 7.3% (182 vs 190 men; hazard ratio 0.96, 95% CI 0.78–1.17; met the 1.5 safety margin)

Nonfatal rhythm problems needing treatment

Testosterone 5.2% vs placebo 3.3%

Atrial fibrillation

Testosterone 3.5% vs placebo 2.4%

Clinical fracture

Testosterone 3.5% vs placebo 2.5% (91 of 2,601 vs 64 of 2,603 in the full analysis set)

Acute kidney injury

Testosterone 2.3% vs placebo 1.5%

Venous clots (all)

Testosterone 1.7% vs placebo 1.2%

Pulmonary embolism

Testosterone 0.9% vs placebo 0.5%

About 5,200 men aged 45–80 at high heart risk; daily gel; mean treatment 21.7 months; about 61% stopped their study gel. Not a trial of injections.

What did the TRAVERSE substudies find?

• Fractures: 3.50% vs 2.46% (hazard ratio 1.43); testosterone did not prevent fractures, and the reason for the rise is unknown (side effects). [41]
• Diabetes: no significant reduction in progression from prediabetes (7.8% vs 10.7% at 12 months). [43]
• Sex life: in men with low libido, sexual activity and desire rose; erections did not improve (sexual health). [44]
• Mood: small gains in mood and energy; no benefit for persistent low-grade depression (mental health). [45]
• Prostate: high-grade prostate cancer in 5 vs 3 men, not a significant difference, in screened men (prostate). [42]
• Anemia: corrected more often on testosterone (45.0% vs 33.9% at 12 months). [46]

What are TRAVERSE's limits?

• Only a gel. It tested a daily 1.62% gel, not injections, pellets or oral capsules, which produce different peaks. In a pooled analysis of trials, hematocrit rose by about 4.0 points more than placebo on cypionate or enanthate injections, 4.3 on oral undecanoate and 3.0 on gel; only the difference between injections and patches was statistically significant, so the gap between gel and injections was not clear. [1], [47]
• About two years of use. Men used the gel for 21.7 months on average and were followed for 33 months. The EAU says the evidence is reliable for about 3 years of treatment, after which no study can exclude long-term heart events. [1], [5]
• Many stopped early. About 61% of men stopped their study gel (testosterone or placebo) before the end, which makes real differences harder to detect. [2]
• Only high-risk men aged 45 to 80. It did not enroll younger men or men without heart risk factors, so its absolute numbers do not transfer to them. [1]
• Moderate levels. Its target range was 350 to 750 ng/dL, and average testosterone on the gel was about 440 ng/dL, so it says little about higher levels. [1], [2]

How do experts read TRAVERSE?

Differently, mainly in how far they go:

• Endocrine Society (July 2026 statement): no meaningful rise in heart attack or stroke over 1 to 4 years, but roughly a 50% relative increase in lung clots and more fractures; long-term safety "remains unestablished". [11]
• European Expert Panel for Testosterone Research (2026): testosterone is safe from a heart point of view in carefully selected, monitored men. [48]
• Androgen Society (2024), a testosterone-focused professional group: it is "conclusively determined" that TRT does not raise heart attack, stroke or heart death. [49]
• A reviewer writing on the American College of Cardiology's website (2023, not an ACC guideline): reassuring overall, but it "may be prudent to avoid" testosterone in men with past clots, and perhaps with intermittent atrial fibrillation or earlier kidney problems. [50]

Labels have not all caught up. Newer-format brand labels, such as Kyzatrex and AndroGel 1.62%, carry the TRAVERSE results. The US cypionate and enanthate vial labels and the Testopel pellet label still say long-term heart safety trials "have not been conducted" (checked September 2026). [2], [18], [17], [51], [52]

What do other studies show about heart attacks and strokes?

Pooled randomized trials published after TRAVERSE agree with it: no clear rise in heart attacks, strokes or deaths, but more heart-rhythm problems. Records studies point in both directions, because they compare men who chose or were given treatment with men who were not. [53], [54], [55], [56], [57]

What do pooled trials show?

The analyses overlap: several include TRAVERSE, so their totals should not be added together.

Do records studies show more heart attacks?

Some do and some show fewer. In a 2025 study of a large international health-records network (TriNetX), 117,908 treated men followed for 5 years had more atrial fibrillation (hazard ratio 1.27) and clots (1.26) than matched untreated men, slightly fewer heart attacks (0.94) and no difference in strokes or deaths. Earlier US studies found more heart events after starting injections than gels (1.26), fewer heart events in men given any testosterone (16.9 vs 23.9 per 1,000 person-years, hazard ratio 0.67), and fewer heart attacks and strokes in VA men whose levels reached normal. Men who get treatment, and who reach normal levels, differ from other men in ways records cannot fully adjust for. [60], [56], [57], [61]

What about long-term use?

This is the least settled question. In a Scottish records study, 440 men aged 51 or older with at least 2 years of prescriptions had more heart events than untreated men (adjusted hazard ratio 1.55); its heart-event count included heart failure, and the study cannot show cause. TRAVERSE followed men for 33 months on average, and the EAU says the trial evidence covers about 3 years of treatment; the EAU and the Endocrine Society both say long-term safety is not established. [62], [5], [11]

Is low testosterone itself bad for the heart?

Very low natural testosterone goes with higher death rates, but that does not prove treatment helps. Pooled data from large population studies linked only quite low natural testosterone, under about 213 ng/dL (7.4 nmol/L), with higher death from any cause, and under about 153 ng/dL (5.3 nmol/L) with more heart deaths. The AUA tells men that low testosterone is a heart risk factor but that it cannot be said for sure whether treatment raises or lowers heart risk. [63], [6]

Does testosterone raise blood pressure?

Yes, a little, with every product tested. In 24-hour monitoring studies required by the FDA, average systolic pressure rose by about 2 to 5 mm Hg after 3 to 6 months. US labels say to measure blood pressure regularly, especially in men who already have high blood pressure, and not to use testosterone when it is uncontrolled. [3], [2]

Ambulatory blood-pressure monitoring means wearing a cuff that measures automatically through a full day and night, which is more reliable than a single clinic reading. The results by product:

[64], [65], [66], [67], [18], [68], [2]

Doserly

How much each product raised 24-hour blood pressure

Average rise in 24-hour systolic pressure (the top number), mm Hg, in each product's US label study.

Bars run from 0 to 6 mm Hg.

Jatenzo (oral undecanoate capsules)

+4.9 systolic / +2.5 diastolic mm Hg · after 4 months · not yet leveling off

Tlando (oral undecanoate capsules)

+4.3 systolic / +1.4 diastolic mm Hg · after 4 months

Xyosted (enanthate autoinjector, under the skin)

+3.9 systolic / +1.5 diastolic mm Hg · after 12 weeks

Aveed (undecanoate injection into muscle)

+3.1 systolic / +2.0 diastolic mm Hg · after 16 weeks

AndroGel 1.62% (gel on the skin)

+1.9 systolic / +1.3 diastolic mm Hg · after 16 weeks

Kyzatrex (oral undecanoate capsules)

+1.7 systolic / +0.6 diastolic mm Hg · after 4 months

One randomized comparison (16 weeks, 673 men, no placebo group)

Fortesta gel

+2.78 systolic mm Hg

Aveed injection

+2.77 systolic mm Hg

Testim gel

+2.15 systolic mm Hg

Separate studies: not a head-to-head comparison. None used cypionate or standard enanthate vials.

Do injections raise blood pressure more than gels?

Not in the one randomized comparison. In a 16-week study of 673 men randomly assigned to one of three products (with no placebo group), 24-hour systolic pressure rose by 2.77 mm Hg on Aveed injections, 2.78 on Fortesta gel and 2.15 on Testim gel. The label numbers above come from separate studies and should not be compared as if they were one trial. None of these studies used cypionate or standard enanthate vials. [4]

How big is a 2 to 5 mm Hg rise?

Small on one day, but it adds up across years, which is why labels warn that it "can increase cardiovascular (CV) risk over time". Some men rise more: in a published 4-month study of Tlando (138 men, average 24-hour systolic rise 3.8 mm Hg), the quarter of men whose hematocrit rose most (6 to 14 points) averaged 8.3 mm Hg, and in pooled heart-failure trials systolic pressure rose 5.68 mm Hg. [2], [23], [69]

The evidence conflicts on one point. A long-running registry of undecanoate injections reported falling, not rising, blood pressure (a median systolic drop of 12.5 mm Hg in men not on blood-pressure medicines). It had no randomized comparison, came from one practice and its authors report industry grants. The controlled 24-hour studies outrank it. [70]

The EAU recommends checking blood pressure before starting, at 3 months and 12 months, then yearly. The TRT side effects guide covers swelling and fluid retention. [5]

Can TRT cause atrial fibrillation or other heart-rhythm problems?

It can make them more likely. In TRAVERSE, atrial fibrillation occurred in 3.5% of men on testosterone and 2.4% on placebo, and nonfatal heart-rhythm problems needing treatment in 5.2% and 3.3%. A 2025 meta-analysis of long trials found about 50% more rhythm problems (risk ratio 1.53; the abstract gives no event counts). [2], [53]

Atrial fibrillation (AFib) is an irregular, often fast rhythm in the heart's upper chambers. It can cause palpitations, breathlessness or tiredness, or no symptoms at all, and it raises the risk of stroke because blood can pool and clot in the heart.

Why do studies disagree about AFib?

Because randomized trials and records studies ask different questions, and natural testosterone seems to have a U-shaped link with AFib. [71]

• Randomized evidence points up. TRAVERSE found more AFib, and it drives the signal in pooled trials: without it, a 2024 meta-analysis found no significant rise. [2], [58]
• Records studies point both ways. Higher AFib in 117,908 treated men in a large records network (hazard ratio 1.27). Lower AFib in VA men whose levels reached normal (0.79), in a large US records study (3.6% vs 4.0%) and in men with diabetes (0.91). Another US records study found a rise that was not statistically clear (risk ratio 1.48). [60], [72], [73], [74], [75]
• Natural testosterone: both low and high-normal. In the Framingham study, men aged 55 to 69 with lower natural testosterone developed AFib more often, and a Finnish study found a weak link. In 4,570 healthy older men, those in the top two-fifths of natural testosterone had about twice the AFib risk of men in the middle. [27], [76], [28]
• Genetics: no link with other rhythm measures. A genetic study found no link between lifelong testosterone level and heart block, fast rhythms, resting heart rate or the QT interval in men. It did not report a testosterone estimate for AFib itself. [77]

The randomized result is the most reliable, and a 2026 review suggests the risk may rise at both low and high levels, which fits all three kinds of evidence. No TRAVERSE paper has yet reported who developed AFib, when, or at what testosterone or hematocrit level. [71]

Does TRT change the heart tracing (ECG)?

Slightly. In a trial of testosterone gel, the QT interval lengthened less with age than on placebo (a 6.3 ms difference); a second trial found a smaller, non-significant change. The meaning is unclear. [78]

Can you take TRT if you have AFib?

None of the guidelines in the key numbers lists AFib as a reason not to use testosterone, and no trial we found reported results separately for men who already have AFib. The reviewer on the American College of Cardiology's website suggested that it may be prudent to avoid testosterone in men with intermittent AFib. Men who take warfarin for AFib need more frequent INR checks when testosterone starts or stops, because testosterone changes how strongly warfarin thins the blood. [50], [2], [19]

Does TRT cause blood clots?

It can raise the risk of clots in the veins, especially in the first months and in men with a clotting disorder. Every US testosterone label has warned about deep vein thrombosis (DVT) and pulmonary embolism since 2014. In TRAVERSE, clots reaching the lungs occurred in 0.9% of men on testosterone and 0.5% on placebo. [18], [36], [2]

When is the clot risk highest?

In the first months, according to records studies. [24], [25]

• First 6 months. In a large UK study, men in their first 6 months of testosterone had 63% more clots (rate ratio 1.63), about 10 extra clots per 10,000 men per year above a base rate of 15.8. After 6 months, the rate was the same as in untreated men (1.00). [24]
• Around month 3. In one US clinic's series, clots peaked about 3 months after starting and fell by 10 months; 65% of clots came within the first 8 months. [25], [79]
• Other records studies disagree. A US study found men were about twice as likely to have been using testosterone just before a clot (odds ratio 2.32), while another found no link (0.90) and VA records found clots in about 4 to 5 per 1,000 men whether or not they were treated, in a group that excluded men with clotting disorders. [80], [81], [82]
• Trials are too small to settle it. Before TRAVERSE, pooled trials showed a wide range of possible effects (odds ratio 1.42, range 0.22 to 9.03). [83]

Who is most at risk of a clot?

Men with thrombophilia, an inherited or acquired tendency to clot, such as factor V Leiden. In a small US referral series of 17 people who had a clot or bone-death event (osteonecrosis) after starting testosterone or hCG, 76% had at least one clotting disorder, against 19% of healthy controls, and two men clotted again while still taking testosterone even on blood thinners. The EAU notes a study in which 40 of 42 men with clots on testosterone had thrombophilia. UK labels say that continuing testosterone after a first clot in a man with thrombophilia needs careful evaluation. [84], [79], [5], [19]

The Endocrine Society and the VA list thrombophilia as a reason not to start, the VA also lists a past clot without a clear trigger, and the EAA and EAU ask about personal and family clot history before treatment. The Endocrine Society notes case reports of clots in men with thrombophilia even without a high hematocrit, especially in the first 6 months. [7], [8], [13], [5]

What are the warning signs of a clot?

Labels list pain, swelling, warmth and redness in one leg for a deep vein clot, and sudden shortness of breath for a clot in the lungs. They say to stop testosterone and get checked if a clot is suspected. Sudden breathlessness or chest pain needs emergency care. [18]

What about heart failure, cholesterol and the kidneys?

Is TRT safe with heart failure?

Uncontrolled or severe heart failure is a reason not to start under most guidelines, because testosterone can cause salt and water retention. Labels warn that swelling, with or without heart failure, can be a serious problem in men with existing heart, kidney or liver disease; UK labels say to stop at once if it happens. [7], [5], [2], [19]

The trials in stable heart failure are small, and they disagree. A 2012 meta-analysis of 4 trials (198 patients) found walking distance improved by 54 meters with no extra heart events. A larger 2020 analysis of 8 trials (332 patients) found no clear benefit for exercise, heart function, quality of life or hospital stays, and systolic pressure 5.68 mm Hg higher. In TRAVERSE, heart failure events were similar (55 vs 50 men). [85], [69], [21]

Does TRT affect cholesterol?

Usually only a little, and the direction varies by product. US labels ask for cholesterol checks regularly, especially after starting. In Kyzatrex trials, total, LDL and HDL cholesterol and triglycerides all fell; the Depo-Testosterone label says cholesterol "may increase". [2]

Pooled trials found no harmful change overall. The UK TestES review found cholesterol and triglycerides slightly lower on testosterone, and in pooled trials of men with type 2 diabetes, total cholesterol and triglycerides fell slightly, with mixed results for HDL. [59], [86]

Adding an aromatase inhibitor is a separate question. A very small year-long trial found no cholesterol change with anastrozole, but the VA does not recommend aromatase inhibitors on TRT and notes they increased heart events in men with existing heart disease. See the anastrozole guide. [87], [8]

Can TRT affect the kidneys?

TRAVERSE found more acute kidney injury: 2.3% of men on testosterone and 1.5% on placebo, about 8 extra cases per 1,000 men. A US records study found a similar rise (risk ratio 1.53), and the VA states that testosterone increases the risk of acute kidney problems. A records study of men with diabetes found slightly less kidney injury (hazard ratio 0.93). No study has explained the TRAVERSE finding. [2], [75], [8], [74]

Is TRT safe for your heart?

For most men with confirmed low testosterone who are checked as guidelines describe, the best evidence shows no rise in heart attacks or strokes over about 2 to 3 years, with small rises in blood pressure, atrial fibrillation, kidney injury and lung clots. Men with certain heart and clotting conditions need more caution, and long-term safety is not established. [1], [2], [5], [11]

Who should be especially cautious?

• Men with a recent heart attack, stroke, stent or bypass (see the waiting times), uncontrolled high blood pressure or uncontrolled or severe heart failure. [6], [2], [5], [17]
• Men with a past clot, a clotting disorder or a family history of clots. See who is most at risk. [7], [8], [5]
• Men with atrial fibrillation or kidney disease. No trial we found reported results separately for men who already have AFib, and the VA lists kidney failure as a reason not to start; see the AFib question and the kidneys. [50], [8]
• Men with a high hematocrit, untreated sleep apnea or who smoke. These raise red-cell counts; see the hematocrit guide.
• Drug-tested athletes. Testosterone is prohibited in sport at all times. [10]
• Under-18s. Testosterone for teenagers belongs only in specialist care.
• Women. This guide is written for men. Testosterone for women uses different doses and products; see testosterone therapy for women.

Which medicines matter for the heart on TRT?

• Warfarin and similar blood thinners. Testosterone changes how strongly they work, so labels advise more frequent INR checks, especially when testosterone starts or stops. [2], [19]
• Steroid medicines (corticosteroids). Taking them with testosterone can increase fluid retention. [88]
• Aromatase inhibitors such as anastrozole. The VA notes more heart events with them in men with existing heart disease. [8]
• SGLT2 inhibitors (diabetes and heart medicines). They raise hematocrit on their own; see the hematocrit guide.
• Other anabolic steroids. Non-prescribed steroids carry much larger heart risks; see the next question.

Bring a complete list of what you take, including supplements, to a pharmacist or prescriber. Checked September 2026.

Is prescribed testosterone the same as steroid misuse for the heart?

No. The alarming heart findings come from non-prescribed steroids at doses far above replacement. Long-term users had weaker heart pumping (ejection fraction 52% vs 63% in non-users) and more artery plaque, and in Danish men caught using steroids, heart attacks were about 3 times as common over 11 years (adjusted hazard ratio 3.00 in 1,189 steroid users against 59,450 matched men; the abstract gives no event counts). A single cycle thickened the heart muscle, which recovered after about 8 months off. These studies do not describe replacement doses; see TRT vs steroids. [89], [90], [91]

Is testosterone legal, and is it allowed in sport?

Checked September 2026. Rules differ by country and change often.

• United States. Testosterone and its esters are Schedule III controlled substances, prescription only. [9]
• United Kingdom. Testosterone is a Class C controlled drug. [92]
• Canada. Testosterone is a Schedule IV controlled substance. [93]
• Australia. Testosterone is a Schedule 4 prescription-only medicine. [94]
• Sport. The World Anti-Doping Agency (WADA) prohibits testosterone at all times, and its 2027 list, published in September 2026, keeps the ban. [10], [95]

What do guidelines say to do about heart risk?

Check heart risk before starting, wait after a recent heart event, rule out uncontrolled heart failure and clotting disorders, then monitor blood pressure, hematocrit and cholesterol on treatment. This section describes what clinicians weigh; it is not a personal plan. [5], [12], [7]

• Before starting: assess blood pressure, cholesterol, diabetes and smoking (EAU and BSSM strong recommendations; AUA), and take a personal and family clot history (EAA, EAU). [5], [12], [6], [13]
• With stable heart disease: still a candidate for a supervised trial of treatment (CUA); treat with caution and check closely (EAU). [14], [5]
• On treatment: blood pressure at 3 and 12 months, then yearly, and cholesterol and blood sugar at 12 months, then yearly (EAU); hematocrit on each body's schedule. [5], [7]
• New symptoms: report chest pain, shortness of breath, dizziness or fainting (AUA); stop and get checked if a clot is suspected (labels). [6], [18]

What if you have had a heart attack or a stent?

Guidelines do not rule testosterone out for good; they say to wait: 3 to 6 months (AUA), 6 months (Endocrine Society) or 4 months after a heart attack, stroke, stent or bypass (VA). In TRAVERSE, about 55% of men already had heart disease, and heart events overall were not more common on testosterone; the published results reviewed here do not break this down by past stents or bypass. The decision belongs to you, your cardiologist and your prescriber together. [6], [7], [8], [14], [2]

Should testosterone be stopped before surgery?

No guideline says so. In records of 7,906 hip replacements in adults, people on testosterone had more leg clots at 1 year (4.3% vs 3.0%), heart events (2.9% vs 1.7%) and sudden damage to the kidneys (7.6% vs 5.6%), but a records study cannot show that testosterone caused this. Tell your surgical team that you take testosterone so they can plan clot prevention. [96]

Practical questions about TRT and the heart

Does TRAVERSE apply to injections?

Not directly. TRAVERSE tested only a daily gel, and no randomized trial has measured heart attacks or strokes with injections. What is known: the Endocrine Society says high hematocrit is more common with injections than with gels, though a pooled analysis of trials could not show a clear gap (about 4.0 vs 3.0 points above placebo), an insurance study found more heart events after starting injections than gels (hazard ratio 1.26, which cannot prove cause), and in the one randomized blood-pressure comparison an injection and two gels raised pressure by about the same amount. [1], [7], [47], [56], [4]

Does TRAVERSE apply to younger men?

Only partly. It enrolled men aged 45 to 80 with heart disease or several risk factors. Younger, healthier men start with a much lower risk of heart attack or stroke, so the same relative effect would mean fewer events; but the trial could not test that, and the rhythm and clot findings have not been studied in younger men. [1]

Does the dose or testosterone level matter for the heart?

Probably, though no trial has compared doses for heart outcomes. Higher doses and levels raise hematocrit more, men with the largest hematocrit rises had the largest blood-pressure rises, and high-normal natural testosterone went with more AFib in older men. In VA records, men whose treated levels reached normal did better than men whose levels stayed low. TRAVERSE kept levels between 350 and 750 ng/dL. [23], [28], [61], [1] The TRT dosage guide covers how doses are set, and the testosterone level visualizer shows how injection frequency changes peaks and troughs.

Can an injection cause sudden coughing or chest tightness?

Rarely, yes. Oil-based injections can cause pulmonary oil microembolism (POME), where tiny oil droplets reach the lungs. The Aveed label's boxed warning lists an urge to cough, breathlessness, throat tightening, chest pain, dizziness and fainting during or right after the injection, and a European safety review in 2025 recommended adding POME to the information for oil-based testosterone injections. The symptoms overlap with a heart attack or a lung clot, so they need urgent assessment. [67], [97]

What do people on TRT report about their heart?

Public forums are full of men watching their blood pressure, feeling their heart pound at night, or deciding with a cardiologist whether to continue. The threads below are recent public Reddit discussions, read with their replies; they show what people experience, not how often it happens.

What do people report about blood pressure?

Rises, mostly, often alongside other changes. Three weeks into testosterone cream, one man who usually measured 110/70 recorded up to 140/97 during the day and about 125/75 at night, and planned a lower dose with his provider. Another, two years into cream, moved from an average of 116/70 to 135–140 over 75–80, which settled on a blood-pressure medicine. A man already on a blood-pressure medicine saw systolic readings of 160–170 a week after raising his dose; they returned to normal once his medicine was doubled. [98], [99], [100]

Other stories show how many things move the number. One man's readings fell from about 130/90 to about 110/70 three months into cypionate, while he drank more water and had stopped exercising. Another's rise from 120/80 to 150/90 settled within two weeks of stopping daily electrolyte tablets. Men with high blood pressure before TRT describe the hardest time: one reading of 201/105 led to a hospital stroke assessment, with hematocrit at 54.7%. [101], [102], [103]

What about palpitations and AFib?

Palpitations are common, and many settle. A 50-year-old felt a forceful heartbeat a few days after starting, with unchanged heart rate and blood pressure, and it calmed within days. Another man's palpitations stopped after his prescriber lowered his weekly dose from 160 to 105 mg. One man's resting heart rate rose from about 62 to 78 beats a minute at a level near 800 ng/dL. [104], [105], [106]

A few threads describe atrial fibrillation. A 33-year-old went into AFib about three months into weekly cypionate; he stopped, was treated with a blood thinner and heart-rate medicine, and felt well six weeks later. Another man, whose heart monitor showed irregular beats 22% of the time, was advised by his cardiologist to stop and retest. Causes are rarely clear-cut: one man's first episode followed a very stressful week, four weeks after starting anastrozole, and his cardiologist blamed the stress. [107], [108], [109]

What do people report about clots and heart events?

Few threads, but serious ones. A 53-year-old with uncontrolled blood pressure, frequent donations for a hematocrit up to 54% and a small pulmonary embolism stopped on his doctors' advice; four weeks later his blood pressure was much better, but his libido and erections were gone. Another poster clotted two years into treatment, restarted at a low dose after six months of blood thinners, clotted again and was advised to stay off. [110], [111]

Cardiologists vary after a heart event. Two months into TRT, one man had a heart attack from long-standing blockages and needed bypass surgery; his cardiologist read TRAVERSE but wanted more data, noting that it used a gel, and planned to revisit the question in a year. A 64-year-old who had raised his dose had chest pain first treated as a heart attack, which turned out to be inflammation of the heart's outer lining. [112], [113]

A few men describe heart-structure findings. A man on scrotal cream with very high levels developed a thickened heart wall and, two years later, weaker pumping, and wrote that the cause was not established. After a steroid cycle at doses far above replacement, another poster's heart walls measured thicker on ultrasound. [114], [115]

How can you judge a heart story?

Ask: Was blood pressure measured the same way each time, seated and rested, at home and in the clinic? What was it before TRT? What else changed, such as caffeine, salt, electrolytes, alcohol, weight, stress, sleep or other medicines? What were the testosterone level and hematocrit, not only the dose? Was a rhythm problem confirmed on an ECG or monitor? Were other hormones or steroids involved? And what did the next reading show?

These selected discussions show what people experience and the questions research has not answered. They are not a survey of all men on TRT, a success rate or a substitute for the studies above.

What should you track?

Blood pressure, hematocrit and cholesterol, plus any new chest pain, breathlessness, palpitations or leg swelling. Use this summary to prepare questions for your prescriber; it is not a personal testing plan. [5], [2], [6]

[5], [2], [7], [6], [18]

One blood-pressure reading is one point on a line; home readings taken the same way (seated, rested, at similar times) show a trend. The TRT blood work guide covers when to draw each test.

Common questions about TRT and heart health

Does TRT cause heart attacks?

Not in the largest trial. In TRAVERSE, heart attack, stroke or heart death occurred in 7.0% of men on testosterone gel and 7.3% on placebo, and pooled trials agree on those outcomes. TRAVERSE did find more atrial fibrillation (3.5% vs 2.4%), acute kidney injury (2.3% vs 1.5%) and lung clots (0.9% vs 0.5%). The studies that first raised the worry were small or based on records. TRAVERSE tested a gel for about two years, so injections and longer use are less certain. [1], [2], [53] See the TRAVERSE results.

Did the FDA remove the heart warning from testosterone?

Partly. There was never a class-wide boxed warning about heart attacks. In February 2025 the FDA told makers to add TRAVERSE to their labels and recommended removing boxed-warning language about heart outcomes, and in July 2025 four products' boxed blood-pressure warnings became standard warnings. Labels still warn about blood pressure and clots, and the US cypionate and enanthate vial labels still carry older heart wording. Checked September 2026. [3], [38], [17] See the history.

Does TRAVERSE prove TRT is safe?

It shows that a daily gel did not raise heart attacks or strokes over about two years in high-risk men aged 45 to 80. It also found more AFib, kidney injury, lung clots and fractures, and it did not test injections, higher levels or longer use. The Endocrine Society says long-term safety "remains unestablished". [2], [11] See its limits.

Can you take TRT if you have AFib?

No guideline rules it out, but no trial we found reported results separately for men who already have AFib, and TRAVERSE found more new AFib on testosterone (3.5% vs 2.4%). If you take warfarin, INR needs closer checks when testosterone starts or stops. [2] See AFib and heart rhythm.

Can you take TRT with factor V Leiden or after a blood clot?

Guidelines are cautious. The Endocrine Society and the VA list thrombophilia, which includes factor V Leiden, as a reason not to start, and the VA also lists a past clot without a clear trigger. UK labels say continuing after a first clot in a man with thrombophilia needs careful evaluation. [7], [8], [19] See blood clots.

Does TRT raise cholesterol?

Usually not much. Pooled trials found no harmful change, and a large UK review found cholesterol and triglycerides slightly lower on testosterone. The direction varies by product, so labels ask for regular checks. [59], [2] See cholesterol.

Glossary

Plain explanations of the medical, heart and research terms used in this guide. Underlined terms in the text link here.

Ambulatory blood-pressure monitoring (ABPM)

Wearing a cuff that measures blood pressure automatically, again and again, through a full day and night. It gives a more reliable average than a single clinic reading.

Absolute risk

The share of people who have an event, such as 7.0 out of every 100 men. It is easier to judge than a relative risk, which only says how much higher one group's risk is than another's.

Aromatase inhibitor

A medicine, such as anastrozole, that blocks the enzyme that turns testosterone into estradiol. It is sometimes added to TRT, off-label.

Atrial fibrillation (AFib)

An irregular, often fast rhythm in the heart's upper chambers. It can cause palpitations, breathlessness or tiredness, or no symptoms, and it raises the risk of stroke.

Boxed warning

The most prominent warning on a US medicine label, printed in a box at the top, for serious risks.

Cholesterol (LDL and HDL)

Fats carried in the blood. LDL is the kind linked with artery plaque; HDL is often called good cholesterol. Triglycerides are another blood fat measured in the same test.

Cohort study (records study)

A study that follows a group of people, or looks back at their records, without assigning treatments. It can show links but not prove that one thing caused another.

Controlled trial

A study that compares people who receive a treatment with a similar group who do not, often receiving a placebo instead. Randomly assigning people to each group makes the comparison fairer.

Edema

Swelling caused by fluid building up in the body's tissues, often in the ankles and legs.

Ejection fraction

The share of blood the heart's main pumping chamber pushes out with each beat, measured on a heart scan. About 55% or more is usually normal.

Clinical guideline

Recommendations written by a medical society or health system, based on its review of the evidence and expert judgment. Different guidelines can reach different conclusions.

Hazard ratio

A measure comparing how often an event happens over time in two groups. A hazard ratio of 1 means no difference; 0.96 means about 4% lower in the first group.

Hematocrit

The share of the blood made up of red blood cells. Testosterone therapy can raise it, which is why it is checked.

INR (international normalized ratio)

A blood test that shows how long blood takes to clot. People taking warfarin have it checked to keep their dose in range.

Intramuscular (IM)

Injected into a muscle. Most testosterone cypionate and enanthate is given this way.

Product label (prescribing information)

The official document approved by a medicines regulator for a product, describing its approved uses, doses, warnings and monitoring. In the UK it is called the summary of product characteristics.

MACE (major adverse cardiovascular events)

A combined count of serious heart and circulation events used in trials. In TRAVERSE it meant death from heart disease, a nonfatal heart attack or a nonfatal stroke.

Mendelian randomization

A genetic study method that uses gene variants affecting a hormone level to estimate the effect of that level over a lifetime. It describes natural levels, not treatment.

Meta-analysis and systematic review

A systematic review gathers all studies on a question in a planned way. A meta-analysis combines their results statistically.

Noninferiority trial

A trial designed to show that a treatment is not worse than a comparison by more than a set margin. TRAVERSE's margin was a hazard ratio of 1.5 for heart events.

NYHA class

The New York Heart Association scale for heart failure, from class I (no limits on activity) to class IV (symptoms even at rest).

Odds ratio

A measure comparing the odds of an outcome in two groups. An odds ratio of 2 means roughly twice the odds; 1 means no difference.

Placebo

A dummy treatment with no active ingredient, used as a comparison in studies. In TRAVERSE it was a gel with no testosterone.

QT interval

A measurement on a heart tracing (ECG) of how long the heart's lower chambers take to reset between beats. A long QT can raise the risk of dangerous rhythms.

Subcutaneous (SC)

Injected into the fatty layer just under the skin. Xyosted is given this way, and some prescribers use this route for other testosterone injections.

Systolic and diastolic pressure

The two numbers in a blood-pressure reading, in mm Hg. Systolic, the top number, is the pressure as the heart beats; diastolic, the bottom number, is the pressure between beats.

Thrombophilia

An inherited or acquired tendency to form blood clots more easily than normal. Factor V Leiden is a common inherited type.

Transdermal

Absorbed through the skin, as with testosterone gels and patches.

TRT (testosterone replacement therapy)

Prescribed testosterone, by injection, gel or other forms, for men with low testosterone.

Venous thromboembolism (VTE)

A blood clot in a vein, usually in the leg (deep vein thrombosis, DVT), which can travel to the lungs (pulmonary embolism).

WADA

The World Anti-Doping Agency, which publishes the list of substances banned in sport. Testosterone is prohibited at all times.

How this guide was researched

This guide is built from a thorough review of the sources cited throughout it: clinical guidelines from 8 medical bodies, product labels from the US and UK, published clinical trials and studies, regulatory documents and trial records. We also reviewed public online forums where men on TRT describe their own heart, blood-pressure and clot experiences.

The guide cites 115 sources, including 43 original studies in people, 10 guideline documents and statements, and 18 public community discussions. Each type of source answers a different question. Guidelines and labels show what clinicians are told to do. Trials and studies show what was measured. Personal accounts show what individual people experienced. Every numbered citation links to its entry below, labeled by source type.

How this guide was made

Research and drafting were AI-assisted. Every cited source was checked against the original, and the guide was reviewed and edited by Doserly before publication. It has not had an independent clinical review, and Doserly does not currently have medical reviewers. Doserly makes a medication and health-tracking app and runs Doserly Academy, both of which are promoted in this guide. Read our editorial policy for how guides are researched, updated and corrected.

This guide is for educational purposes. It summarizes what the reviewed sources report so the research is easier to understand; it is not medical advice. For a deeper dive, or to check any point for yourself, go straight to the cited sources.

Explore the sources