
Peter Attia on Treating Low Testosterone (and why TRT isn't always the right answer)
What this covers
This ten-minute clip comes from a longer episode in which Peter Attia describes how he approaches low testosterone in men. It starts with the background: male testosterone peaks in the 20s and then declines slowly, rather than dropping sharply as hormones do at menopause. It then explains why a lab number is a weak guide. Hormones act by binding androgen receptors, and the resulting complex has to reach the nucleus to influence transcription factors. Reported levels, including free testosterone, are calculated from total testosterone, SHBG, and albumin, and Attia compares this to the Friedewald estimate of LDL-C. No direct measure of androgen receptor saturation exists, so the response to treatment has to carry much of the diagnostic weight.
The practical middle of the clip concerns how to treat. Attia prefers intramuscular injections to gels, because gel absorption varies with sweat, exercise, skin, and application site. He prefers 40-50 mg twice weekly to the historical every-two-weeks schedule, which was set for convenience and leaves supraphysiologic levels for several days followed by a trough. He says to look for a "hell yes" response rather than treating the first trial as a placebo-controlled one. He also says to stop and reassess if the response is equivocal, except where the aim is insulin resistance or muscle building.
The risk discussion is where positions get more contested. Attia says TRT does not raise prostate cancer risk and that low testosterone is linked to higher-grade disease, though it may worsen benign prostatic enlargement. He allows a probable slight, short-term rise in cardiovascular events, possibly through blood pressure, and argues for managing other health problems first. He also covers acne and androgenetic alopecia, mentions 5-alpha reductase inhibitors as an option, and says a patient who puts hair ahead of treatment benefits should be respected.
Low testosterone should be diagnosed and treated based on a combination of symptoms and biochemical markers rather than testosterone levels alone, because androgen receptor saturation varies between individuals and cannot be directly measured, making symptom response the most reliable indicator of therapeutic benefit.
- Testosterone levels are estimates (measured indirectly via SHBG and albumin calculations) and do not reveal actual free testosterone or androgen receptor saturation status
- Individuals have variable androgen receptor density and function, so the same testosterone level produces different clinical effects
- Clinical response to treatment (symptoms improvement) is more reliable than lab normalization, requiring adequate dosing and monitoring for symptom resolution
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Not all men have the same density of androgen receptors, and not all androgen receptors function in exactly the same way, which means two men with identical testosterone levels can experience different clinical effects.
“we know that not all men have the same density of anderen receptors and we know that not all Androgen receptors function in the exact same way”
Measured testosterone levels (including free and bioavailable testosterone) are estimates based on calculations using total testosterone, sex hormone binding globulin, and albumin rather than direct measurements, meaning they do not actually reveal the free testosterone concentration or androgen receptor saturation status.
“we sit here and we measure testosterone levels in men and maybe we measure bioavailable or free testosterone but those are just estimates they aren't actually telling you free testosterone level um you're measuring total testosterone you're measuring sex hormone binding globulin you're measuring albumin you use those to estimate the free amount of testosterone but that's still an estimate”
Male testosterone peaks in the 20s and then undergoes a slow, steady decline throughout adulthood, unlike women who experience a sharp drop during menopause.
“male testosterone probably peaks in the 20s and it's just a slow steady decline it's not um it's not like in the case of women where they you know they go through puberty they have these hormones that are cyclical and then fall off a cliff with men it sort of you go through puberty you kind of peak and then you're on a slow uh decline down”
Testosterone and other hormones exert their effects by binding to androgen receptors, which then form a complex that must enter the nucleus to impact transcription factors, making the density and functional capacity of androgen receptors the critical determinant of hormonal effect rather than circulating testosterone levels alone.
“these hormones work by binding to Androgen receptors and the testosterone Androgen receptor complex has to make its way into the nucleus where it impacts transcription factors”
Testosterone replacement therapy does not increase the risk of prostate cancer; in fact, lower testosterone is associated with higher risk of high-grade prostate cancer, contrary to popular belief.
“we know know that the lower the testosterone the higher the risk of high-grade prostate cancer um so again contrary to popular belief testosterone replacement therapy does not increase the risk of prostate cancer”
Testosterone administered as a gel has variable absorption that differs both between individuals and within the same individual across different times of day, depending on factors such as recent exercise, sweating, skin exfoliation, hair density, and the anatomical location of application, making injections more reliable.
“you have variable absorption and it doesn't just vary by individual it varies by time a day so you know for example like if you're if you just finished a workout and you're sweating and even if you go and have a shower you're still kind of in a less absorptive State than maybe if you're cold um you know what part of your body do you put it on do you have to exfoliate the skin for first you have hair on the skin you know you want to put it on an area that doesn't have hair there's just more issues with it”
The historical standard of testosterone replacement every two weeks was implemented to minimize patient inconvenience (fewer doctor visits) rather than for pharmacological reasons, but this creates problematic pharmacokinetics with supraphysiologic levels for 4-5 days followed by subphysiologic levels before the next dose.
“the problem with doing it every 2 weeks which was usually done in the days when people would go to their doctor to get the injection and you wanted to minimize the inconvenience of that um you're just super physiologic for you know four or five days then you're kind of physiologic and then you're actually backed down to being very subphysiologic before the dose”
A large study found that high-risk men given testosterone showed a slight increase in major adverse cardiac events (MACE) at one year post-initiation of testosterone replacement therapy, but this increased risk vanished by two and three years, suggesting that the highest-risk men (those closest to having a cardiovascular event) were pushed over the edge by testosterone initiation.
“there was um a very large study that looked at kind of high-risk men and they were given testosterone and at one year post initiation of trt there was a slight increase in the risk of major adverse cardiac events in the testosterone group compared to the placebo group that vanished at two and three years uh almost suggesting that the highest risk men probably those that were closest to having an event were actually pushed over the edge a little bit”
Clinicians should not initiate testosterone replacement therapy until they have addressed and managed all other relevant health conditions and factors (cardiovascular risk, metabolic health, etc.), indicating a comprehensive approach to patient readiness before testosterone therapy.
“we're not keen to put guys on testosterone until we have the house in order with respect to everything else”
The analogy between how testosterone levels are measured (calculated from indirect markers rather than directly) and how LDL-C is calculated by the Friedewald formula (an estimate rather than direct measurement) illustrates why both markers should be treated with caution.
“kind of like ldlc calculated by the fredwell formula is an estimate”
Testosterone replacement therapy should be administered via intramuscular injection twice weekly at doses of 40-50 mg rather than the standard historical protocol of every two weeks, because twice-weekly dosing produces much more consistent and steady testosterone levels throughout the week.
“we recommend um an injection we also recommend instead of doing it every two weeks which is standard doing it twice a week at obviously a much lower dose so typical dose would be somewhere between 80 and 100 milligrams of testosterone a week so it would be 50 40 to 50 milligrams twice a week and that produces just a much more steady uh level because you're you're really trying to get the steadiest level possible”
Low testosterone should be diagnosed based on a combination of both laboratory levels and clinical symptoms, not laboratory values alone, because symptoms are more clinically relevant than isolated lab measurements.
“low tea is really a combination of levels and symptoms and it's really important to remember that symptoms matter because levels are really well how can I put it delicately I mean just not as helpful as we'd like to believe they are”
Testosterone replacement therapy can cause hair loss in individuals genetically susceptible to androgenetic alopecia through androgen pathway mechanisms, and can increase acne in susceptible individuals, but these side effects are typically more common at supraphysiologic rather than physiologic replacement doses, occurring in clinical practice approximately once per year or less frequently.
“it also there are there are side effects of testosterone right it will drive hair loss in an individual who's susceptible to hair loss through um uh the the sort of Androgen Pathways there um it can increase acne in a susceptible individual again these things are typically more the type of side effects that people talk about when testosterone is being used in super physiologic levels so I'm just trying to think the last time we saw a patient who had acne um I'll probably see it once a year”
For individuals concerned about testosterone-induced hair loss, 5-alpha reductase inhibitors can be used to block conversion of testosterone to dihydrotestosterone, which is the more potent androgen driving androgenetic alopecia, as an alternative to discontinuing testosterone therapy.
“you can take a 5 Alpha reductase inhibitor uh so those are drugs that block the conversion of testosterone to dihydrotestosterone which is a more potent Androgen and that's the that's the Androgen that's driving Androgen specific hair loss”
Some patients may prioritize hair preservation over testosterone therapy benefits, and this is a valid clinical decision that should be respected rather than overridden by the clinician.
“or they might say you like I've had patients say oh you know what like my My Hair Matters more to me than my testosterone I don't want to take testosterone so so so those are the things that we just kind of want to point out”
Testosterone replacement therapy may increase the risk of benign prostatic hypertrophy (BPH, prostate enlargement), and patients should be monitored for this side effect.
“what it does do is potentially increase uh BPH benign prostatic hypertrophy so it does increase the size of the prostate potentially um so so one you know needs to be mindful of that”
In susceptible individuals receiving testosterone replacement therapy, there is probably a slight short-term increase in the risk of major adverse cardiovascular events (heart attack, stroke), likely mediated through increased blood pressure rather than through other mechanisms.
“I do believe that in a susceptible individual in the short run there's there's probably a slight increase in the risk of cardiovascular events with testosterone and it's probably born through an increase in blood pressure”
When administering testosterone replacement therapy, clinicians assume that androgen receptors are not saturated, meaning additional testosterone will increase receptor saturation and nuclear transcription, but there is no direct way to measure androgen receptor saturation status in individual patients.
“if you're giving a guy testosterone in the presence of mild to low te you're assuming his Androgen receptors are not saturated and therefore giving him more testosterone will lead to an increased saturation of the AR and will lead to more nuclear transcription but we have no way of measuring that”
Testosterone replacement therapy should be initiated based on a combination of clinical symptoms (reduced libido, reduced energy, mood changes) and signs (insulin resistance, difficulty building muscle, difficulty recovering from exercise) paired with biochemical labs showing total testosterone below the 30th-40th percentile or lower, with free testosterone commensurate with total testosterone levels.
“I got to see a certain set of symptoms in combination with a biochemical set of labs that makes sense and then we have to test it out”
When testing testosterone replacement therapy, the initial response should not be evaluated as a double-blind placebo-controlled trial because placebo effects will be present, so instead clinicians should look for a "hell yes" response, and if the response is equivocal, the treatment should be discontinued and symptoms reassessed to determine if the deficit reappears.
“it's not going to be a placebo test so we're going to have a placebo effect and then if the if if the response we see isn't a hell yes I think we should pull pull it all off and see if we notice a response in the deficit”
Some individuals may already have androgen receptor saturation at baseline low testosterone levels, meaning that increasing testosterone will not produce clinical benefit, while others may be severely under-saturated and benefit dramatically from increases, which explains the wide variability in clinical response to testosterone replacement therapy.
“this guy might have been already saturated at at 8 nanograms per decler where he started and all that additional testosterone may have done him no good whereas somebody else might have been woefully undersaturated and when you you know increased him by 150% you actually got benefit from it”
In individuals where testosterone replacement therapy is being considered primarily for insulin resistance and muscle-building outcomes rather than symptom relief, treatment should be continued even if the patient reports no subjective symptom improvement, because the relevant outcomes are objective metabolic and physical changes.
“unless we were only treating this for insulin resistance and muscle mass those were the only things in which case I would say we still say the course and see if those things get better”