FAQ  /  Hormone Optimization

I Heard From My PCP That Testosterone Can Make My Blood Too Thick and Cause Blood Clots. Is That True?

What's True: Testosterone Does Raise Hematocrit

---

Testosterone stimulates red blood cell production, partly by increasing erythropoietin and partly by suppressing hepcidin, the hormone that regulates iron availability for red cell production. The result is a real, expected, dose-related rise in hematocrit for most men on therapy, particularly with injectable formulations. This isn't a side effect that occasionally shows up in a few patients, it's a well-established, predictable response, which is exactly why hematocrit is checked at baseline and monitored throughout treatment. None of what follows on this page is an argument against monitoring hematocrit. It's the actual point of monitoring it.

What the Guidelines Actually Say About When It Matters

---

The American Urological Association's 2018 guideline recommends withholding testosterone therapy if hematocrit exceeds 50% before starting treatment, specifically to investigate the cause, and recommends dose reduction or temporary discontinuation once hematocrit reaches 54% or higher. The Endocrine Society's 2018 guideline uses hematocrit above 50% as a relative contraindication to initiating therapy and hematocrit above 54% as the threshold to stop therapy. Other professional societies use thresholds ranging from 52% to 55%.

Notice the gap between those two numbers. The "above 50%" language is about investigating a new, unexplained finding before someone starts therapy. The number that actually triggers a treatment change, dose reduction or discontinuation, sits at 52% to 54%. A hematocrit in the high 40s or right around 50 in someone already on stable, monitored therapy isn't in the zone any guideline treats as an action point, even though it's common for that range to get talked about as if it already is.

It also helps to know that "normal" hematocrit itself isn't a single fixed number. Different labs use different upper limits, some set it at 50%, others at 51%, others at 54%, and some run higher still. A national reference lab commonly used in the U.S. lists a normal range up to 60%. Those ranges were built from populations without secondary erythrocytosis, not from men on testosterone therapy or from people adapted to high altitude, where normal runs meaningfully higher. Men living in parts of the Andes have a standard hematocrit range of 45% to 61%, and that population isn't experiencing an epidemic of blood clots. A number that reads as high on one lab's printout can sit inside another lab's normal range entirely, which is worth keeping in mind before treating any single hematocrit value as inherently alarming.

Where the Blood Clot Claim Actually Comes From

---

A large retrospective, propensity-matched claims database found an association between elevated hematocrit and clotting events, but only at a cutoff of 52% or higher. It's worth being clear about what that kind of study can and can't show. Retrospective, observational data like this can only demonstrate association, and association is vulnerable to confounding, meaning some other factor, not the one being studied, could be driving the outcome. Men whose hematocrit climbs higher on testosterone therapy might also differ from men whose hematocrit doesn't in ways that have nothing to do with hematocrit itself, and a retrospective design like this one can't fully separate that out. That doesn't mean the finding is meaningless. It means the correct next step is to test it directly, which is what a randomized trial does. Separately, prospective cohort data drawn from the Atherosclerosis Risk in Communities Study and the Cardiovascular Health Study found that an elevated hematocrit did not correlate with increased venous thromboembolism risk at all, though that study's own upper hematocrit threshold was fairly low.

That's exactly the sequence that played out here. The retrospective finding flagged hematocrit as something worth testing directly. TRAVERSE tested it.

What the TRAVERSE Trial Shows About Hematocrit Specifically

---

TRAVERSE is the largest randomized trial to date on testosterone's cardiovascular safety, and a follow-up analysis of that trial tested specifically whether the rise in hematocrit during treatment was associated with venous thromboembolism or cardiovascular events. It wasn't. The result was a hazard ratio of 0.94, with a 95% confidence interval of 0.84 to 1.05. A hazard ratio compares how often an event happens in one group versus another, a value of 1.0 means no difference between groups, and this result sits right at that line, with the confidence interval crossing 1.0 as well, meaning the data doesn't support any meaningful association in either direction. This is the strongest evidence available on the specific question this page is about, and it points the same direction as the physiology and the guideline structure above: hematocrit elevation at the levels testosterone therapy typically produces has not been shown to raise clotting risk.

What TRAVERSE Showed About Cardiovascular Risk Overall

---

TRAVERSE was actually built to answer a bigger question than hematocrit alone. In 2014, the FDA required a boxed warning on all testosterone products about a possible increased risk of heart attack, stroke, and cardiovascular death, based on limited earlier data, and asked for a definitive trial to settle it. TRAVERSE was that trial: over 5,200 men, ages 45 to 80, with hypogonadism and either documented cardiovascular disease or several major cardiovascular risk factors, randomized to testosterone or placebo. The primary result was unambiguous: a serious cardiovascular event occurred in 7% of men on testosterone versus 7.3% on placebo, no meaningful difference between groups. That result was strong enough that in July 2025, the FDA removed the cardiovascular boxed warning that had been on every testosterone product for over a decade. On the central safety question TRAVERSE was designed to answer, the result came back clean, and it's now reflected in the product labeling itself, not just the trial data.

TRAVERSE also tracked several secondary safety measures, including venous thromboembolism, atrial fibrillation, and acute kidney injury, and found somewhat higher rates of each in the testosterone group. These weren't part of the primary, adjudicated cardiovascular outcome the boxed warning addressed, and none of them changed the trial's central conclusion or the FDA's decision. They also come from a trial population enrolled specifically for elevated cardiovascular risk, documented disease or multiple major risk factors, which matters for how far these numbers generalize to a lower-risk patient starting therapy. As with any single trial's secondary findings, the honest read is that this is a signal worth watching and confirming in further research, not a settled answer, which is exactly why hematocrit and cardiovascular risk factors get monitored throughout therapy rather than checked once at baseline and set aside.

Why This Belief Persists Anyway

---

A few things keep this idea circulating even though the leap from "thicker blood" to "clot risk" hasn't actually been demonstrated at TRT-typical hematocrit levels. Viscosity really does rise with hematocrit, that part is basic, uncontested physiology, and higher viscosity is a genuine mechanism for increased clotting risk in general. The gap is in the next step: assuming that mechanism is actually operating at the hematocrit elevations testosterone therapy produces. As covered above, TRAVERSE tested that assumption directly and didn't confirm it. "Your blood is getting thicker" is accurate. "Therefore you're at risk for a clot" is the part the evidence doesn't support at this range, and framing the two together tends to land as an alarm rather than as the more precise, and more accurate, description of what's actually going on.

Caution that's genuinely appropriate at the higher end, hematocrit climbing toward 54% and beyond, tends to get applied reflexively to any elevation, including levels well inside the normal range, rather than reserved for where the guidelines actually place it.

There's also a specific reason the bench-level viscosity data oversells the real-world picture. The studies most often cited to support "higher hematocrit means thicker, more dangerous blood" measure viscosity in a rigid glass viscometer, and in that setting, viscosity does rise sharply with hematocrit. But blood doesn't flow through rigid glass tubes in the body, it flows through small, distensible vessels, and it behaves differently there. As vessels narrow toward capillary size, red cells migrate toward the center of the flow, leaving a thin, low-viscosity layer of plasma along the vessel wall, a well-described phenomenon called the Fåhræus-Lindqvist effect. Blood is also a shear-thinning fluid, meaning its viscosity itself drops as flow speeds up, which is exactly what happens in smaller vessels. The net effect is that apparent viscosity in vivo is meaningfully lower than what a benchtop viscometer reading on the same blood would suggest. The physics behind "thicker blood" isn't wrong, it's just measured in a setting that doesn't fully represent how that blood actually moves through a real vessel.

There's a second piece of physiology worth understanding here too. Secondary erythrocytosis doesn't happen in isolation, it happens alongside an increase in total blood volume. A larger blood volume expands the vascular bed, lowers peripheral resistance, and increases cardiac output, and the net effect is that optimal oxygen delivery in that expanded-volume state actually occurs at a somewhat higher hematocrit than it would in someone with a normal, untreated blood volume. In other words, the body isn't just tolerating a higher hematocrit on testosterone therapy, in a meaningful sense it's adapted to need one. A moderate rise, within the ranges guidelines already treat as expected and monitored, isn't automatically something to correct back down to an untreated baseline.

None of that makes the concern malicious. It usually reflects genuine caution built around a threshold that applies at a meaningfully higher hematocrit than what's actually showing up on the labs in question.

Testosterone Also Has Some Real Vascular Benefits

---

Focusing only on hematocrit leaves out the rest of the picture. Testosterone has several documented effects on blood vessels that work in the opposite direction of a clotting concern. It promotes nitric oxide production in the endothelium, the lining of blood vessels, which supports vasodilation and healthy blood flow, and this effect has been shown consistently across mechanistic and clinical studies. Testosterone therapy has also been shown, in a randomized, placebo-controlled trial, to lower fibrinogen and plasminogen activator inhibitor type 1, two substances that promote clotting when elevated, rather than raise them. None of this cancels out the importance of monitoring hematocrit, but it's a reminder that testosterone's overall effect on clotting risk isn't a one-way street toward danger. The full picture includes mechanisms that lower clot risk alongside the hematocrit rise that gets most of the attention.

A Familiar Pattern

---

What to Actually Do

---

• Keep monitoring hematocrit as usual, and don't change a stable, working dose based on this finding alone. A hematocrit in the high 40s to around 50 in an established patient isn't an action point by any major guideline's own standard. If it trends up toward the mid-50s over time, that's the point to revisit dosing, not before.
• If hematocrit does run higher than you'd like, the first question is what is driving it, and dose is the lever every guideline reaches for before phlebotomy. Testosterone dose and delivery both matter, and so do causes that have nothing to do with testosterone, including untreated sleep apnea, smoking and simple dehydration. Routine blood donation is not a standard requirement of testosterone therapy, and repeated phlebotomy has its own downsides including iron depletion. That whole question, including what the evidence does and does not support about splitting doses more frequently, is covered in will I have to donate blood or get phlebotomy on testosterone.
• Bring the guideline thresholds to the conversation if you want to. The AUA and Endocrine Society numbers above are publicly available guidance, not this practice's opinion, and they're worth having on hand if your PCP is open to discussing the specific data.
• If you or a close family member has a personal history of blood clots, make sure that's on record. Across published case series of men who developed clots while on testosterone therapy, a large majority turned out to have a previously undiagnosed clotting disorder, such as Factor V Leiden, that predated and likely drove the event rather than the testosterone alone. That's a reason to flag personal or family clotting history clearly, not a reason for general concern if that history isn't present.
• Bring any new PCP concern to your next visit here regardless. If your hematocrit or testosterone level looks different than usual, or if new symptoms come with it, that's always worth a conversation with your care team.

— Dr. Darrell Wilcox · @wellnessdoc_4everyoung on Instagram

This content is for educational purposes only and does not constitute medical advice. It is not a substitute for professional medical evaluation, diagnosis, or treatment, and it is not a recommendation to disregard concern raised by any treating physician. The guideline thresholds cited here (American Urological Association, Endocrine Society) apply to hematocrit management in testosterone therapy specifically and do not extend to other hormones or other safety markers. Individual risk factors, including a personal or family history of clotting disorders, smoking status, and other cardiovascular risk factors, can change how a given hematocrit level should be interpreted, and this content does not establish what is appropriate for any individual patient. Always discuss changes in your lab results, symptoms, or treatment plan with your care team before making any changes.