Cortisol and Testosterone: How Chronic Stress Is Quietly Dismantling Your Hormones

Here is what nobody told me when I was at my worst.

My testosterone panel came back low-normal. My cortisol came back "within range." My doctor looked at the numbers and said I was fine. I was not fine. I was sleeping nine hours and waking up exhausted. I had no libido. My training wasn't doing anything except making me feel worse. And every single time I sat in that office and described my symptoms, the labs got cited at me like evidence that I was imagining it.

The Mechanism Your Doctor Missed

What nobody ordered — what almost no one gets ordered — was a diurnal cortisol assessment. A measure of how cortisol is behaving across the day, not just at 8am when the reference range was designed to catch Addison's disease and Cushing's syndrome. Because a single morning cortisol draw will not catch functional hypercortisolism. It will read as completely normal while your hypothalamic-pituitary-gonadal axis is being suppressed around the clock.

The relationship between cortisol and testosterone is not a soft correlation. It is a direct physiological suppression mechanism operating at multiple levels simultaneously. If you are a man in your late thirties or forties with fatigue, low libido, poor gym recovery, and labs that keep coming back "normal," this is the mechanism your doctor missed. I'm going to walk through it completely — the physiology, the lab gap, the downstream effects on sleep and training, and what you can actually do about it before you consider any hormonal intervention.

The Seesaw Most Men Never Hear About

When your cortisol goes up, your testosterone goes down. Not as a metaphor. Not as a general wellness concept. As a direct, documented, acute endocrine event.

In 1983, Cumming and colleagues administered ACTH — the pituitary signal that drives cortisol production — to healthy men and measured what happened downstream. Within four hours, both LH (the pituitary signal to the testes) and testosterone were suppressed (Cumming et al., J Clin Endocrinol Metab, 1983). The mechanism is that fast and that upstream. It does not require years of chronic stress to manifest. A single cortisol spike is enough to put the brakes on your testosterone production within hours.

Now scale that to modern life. You are not dealing with an acute cortisol spike that resolves in a few hours. You are dealing with a low-grade chronic cortisol elevation that does not switch off — from work pressure, poor sleep, financial anxiety, inflammatory diet, overtraining, or some combination of all of them running simultaneously. The suppression is not acute. It is continuous. And it runs in the background while your doctor looks at your testosterone number and tells you it is within the normal range.

The seesaw is real. Understanding why it does not shut off under chronic conditions is the next part.

How the HPA Axis Became a Chronic Problem

Your hypothalamic-pituitary-adrenal axis — the HPA axis — is one of the most elegantly designed systems in human biology. Under acute stress, it fires, dumps cortisol, gives you the energy and focus to deal with the threat, and then shuts down via negative feedback. Cortisol itself signals the hypothalamus and pituitary to stop producing CRH and ACTH. The loop closes. You recover.

That system was designed for a world of acute, resolvable threats. It was not designed for a world where the threat never fully resolves.

Bruce McEwen's work on allostatic load — the cumulative physiological cost of chronic stress — established that repeated or sustained activation of the stress response changes how the system regulates itself (McEwen, NEJM 338:171, 1998). The negative feedback that shuts down acute cortisol release becomes dysregulated. Instead of a sharp spike and recovery, you get a chronically elevated or diurnally flattened cortisol pattern that keeps the HPG axis suppressed without ever spiking high enough to look like pathology on a single blood draw.

This is not a disease state in the traditional sense. There is no tumor. No gland is failing. What is happening is that the normal regulatory architecture of the stress axis has been chronically overloaded, and the downstream hormonal systems — including the one that produces testosterone — are paying the cost.

Three Ways Cortisol Kills Testosterone

The suppression does not happen at one point in the chain. It happens at three.

First: GnRH and LH suppression. Glucocorticoids directly inhibit the hypothalamic release of gonadotropin-releasing hormone (GnRH). Less GnRH means the pituitary sends less luteinizing hormone (LH) to the testes. Less LH means less testosterone is produced. This is not a theory — it is the primary mechanism documented in Whirledge and Cidlowski's review of glucocorticoid suppression of the HPG axis, which tracked the suppression at multiple levels of the axis simultaneously (Whirledge & Cidlowski, Nat Rev Endocrinol 6:446, 2010).

Second: Leydig cell blunting. Even when LH does reach the testes, the Leydig cells — which are the actual testosterone-producing cells — carry glucocorticoid receptors. Sustained cortisol exposure reduces their sensitivity and output directly. So even if the pituitary is still sending the LH signal at normal levels, the target cells are less capable of responding to it. This is a second, independent point of suppression downstream of the first.

A note on a third proposed mechanism: you will see the "pregnenolone steal" hypothesis discussed widely in men's health content, including probably on this site. The idea is that cholesterol — the shared upstream precursor for both cortisol and testosterone — gets preferentially routed toward cortisol synthesis under chronic demand, leaving less available for testosterone production. It is a plausible-sounding mechanistic story. It is also contested. Many endocrinologists reject the shared-precursor competition model as an oversimplification; the regulatory enzymes that govern steroidogenesis do not simply "steal" substrate from one pathway to feed another in the way the metaphor implies. The two mechanisms described above — HPG suppression and Leydig cell blunting — are well-supported by the literature. The "steal" framing is not. I am flagging it rather than repeating it.

Two confirmed suppression mechanisms running simultaneously, with a third that is proposed but not settled. Each documented point reduces testosterone. Together, they can push a man well below his functional optimum while his testosterone number stays inside the reference range that was built to catch hypogonadism, not functional suppression.

What "Normal" Cortisol Labs Miss

This is the part that kept me dismissed for years.

The standard cortisol test is a single serum draw, usually in the morning, usually around 8am. That draw was calibrated to catch primary adrenal failure (Addison's disease) and hypercortisol disease (Cushing's syndrome). Those are pathological states with dramatic cortisol deviations. The test is very good at finding them. It is useless for finding what most chronically stressed men actually have.

What most chronically stressed men have is not Cushing's. It is a flattened diurnal cortisol curve. Healthy cortisol should spike sharply in the morning — the cortisol awakening response — drop through the day, and reach its nadir in the evening before sleep. Under chronic stress, that curve flattens. Morning cortisol is blunted. Evening cortisol stays elevated when it should be near zero. The total area under the curve may not look dramatically different from normal on a snapshot. But the timing is wrong, and timing is everything for HPG axis function.

Fries and colleagues documented the flattened diurnal cortisol slope as a robust biomarker of chronic stress exposure — and noted explicitly that it is invisible to single-timepoint testing (Fries et al., Psychoneuroendocrinology 34:1, 2009). Your doctor orders one morning draw, it falls in range, you get sent home. The mechanism that is suppressing your testosterone is still running. It just did not show up on the test that was never designed to find it.

The Sleep Loop

Testosterone is not manufactured evenly throughout the day. The majority of daily testosterone synthesis happens during sleep, specifically during REM and slow-wave stages. Disrupt the sleep, and you disrupt the production line.

Elevated evening cortisol — the hallmark of the flattened diurnal curve — directly degrades sleep architecture. It delays sleep onset, fragments sleep, and suppresses the slow-wave stages where testosterone synthesis peaks. You sleep for eight hours and the quality is not there. You wake tired. Your testosterone, which should have been replenished overnight, was not.

Leproult and Van Cauter ran one of the cleanest human experiments on this. They restricted healthy young men — not clinically ill, not stressed, just sleep-restricted — to five hours per night for one week. Daytime testosterone levels dropped 10 to 15 percent (Leproult & Van Cauter, JAMA 305:2173, 2011). One week. Healthy young men. Ten to fifteen percent.

Now consider the chronically stressed man who is not hitting five hours of clean sleep, but eight hours of fragmented, cortisol-disrupted sleep. The arithmetic is not better just because the hours are higher. The quality of the sleep is what drives the testosterone synthesis. And his cortisol, running too high into the evening, is sabotaging the quality of every night.

The loop looks like this: chronic stress elevates cortisol, elevated evening cortisol degrades sleep, degraded sleep suppresses testosterone synthesis, low testosterone reduces stress resilience, which makes everything feel harder, which feeds back into elevated cortisol. Once you are in it, every component makes every other component worse.

Our guide on sleep and testosterone covers the LH pulse architecture that sleep depends on and a two-week protocol to test the fix on yourself.

Why Training Harder Under Chronic Stress Makes It Worse

The instinct makes sense. You feel terrible. You feel soft. You feel weak. So you train harder, because training fixes things. Except under chronic cortisol elevation, it does not fix things. It extends the problem.

Cortisol is catabolic. That is its job — to mobilize energy, break down tissue, suppress inflammation acutely so you can deal with the immediate demand. In the context of recovery from exercise, you need a cortisol spike followed by a rapid return to baseline, which allows the anabolic window to open: testosterone rises, growth hormone pulses, muscle protein synthesis runs. The cortisol spike is not the problem. The failure to return to baseline is the problem.

Under chronic stress, baseline cortisol is already elevated. Your post-exercise cortisol spike starts from a higher floor, stays elevated longer, and the anabolic rebound is blunted. Duclos and colleagues studied overtrained athletes and found that their testosterone-to-cortisol ratio at baseline was chronically suppressed compared to well-trained controls — not just during exercise, but at rest (Duclos et al., Eur J Endocrinol 148:421, 2003). They were training just as hard. Their hormonal environment could not support the adaptation.

Training five or six days a week under sustained cortisol elevation is not building you. It is providing additional cortisol stimulus to a system that is already overloaded. You feel worse after training, not better. You are slower to recover. Your body composition does not improve. The mechanism is not mysterious — you are pushing a system that cannot absorb the load.

The answer is not to stop training. It is to structure training so it is not another stressor your system has to survive. That means mandatory deload weeks. It means treating sleep as the primary recovery tool. And it means addressing the cortisol load before assuming the training volume is the main variable.

Reducing the Cortisol Load Before Chasing Testosterone

The most important thing I can tell you is this: you cannot TRT your way out of a chronically elevated cortisol environment.

If the HPA axis is still running hot, exogenous testosterone will suppress endogenous production, which is expected, but it will not fix the underlying suppression mechanism. You will still have elevated evening cortisol degrading your sleep. You will still have glucocorticoids suppressing GnRH signalling and blunting your Leydig cell response. You will feel better initially because you are supplementing above the suppression. But you have not fixed anything upstream. You have added a treatment on top of an unresolved cause.

Before the conversation reaches hormone replacement, these are the evidence-ranked cortisol levers, in order of impact:

Morning light exposure. Ten to fifteen minutes of direct outdoor light within thirty minutes of waking entrains the circadian rhythm and sharpens the morning cortisol spike, which is the metabolically correct version of the pattern. A sharp morning spike with a steep decline is healthy. A flat all-day curve is not.

Mandatory deload weeks. Every fourth week, training volume drops 40 to 50 percent. This is not optional. It is the mechanism by which the system absorbs the load from the previous three weeks. Without it, the cumulative cortisol from training stacks on top of the cortisol from life stress. The deload is the reset.

Ashwagandha. This has the strongest human randomized-controlled trial data in the adaptogen category. Chandrasekhar and colleagues randomized chronically stressed adults to 300mg standardized ashwagandha extract twice daily or placebo for sixty days. Serum cortisol dropped 27.9 percent in the treatment group (Chandrasekhar et al., Indian Journal of Psychological Medicine 3:357, 2012). That is a meaningful reduction from a well-run trial, not a supplement company claim.

Phosphatidylserine. 400 to 800mg before training blunts the cortisol spike from exercise. It does not eliminate the cortisol response — you still get the training stimulus — but it reduces the overshoot that extends recovery time.

Start with the first three before adding any supplement. Sleep, light, deload. If those are not in place, nothing downstream works as intended.

When to Bring It to a Clinician

There are thresholds where lifestyle intervention is not enough and a clinician conversation is warranted. Here is what they look like in practice.

Evening cortisol above mid-range on a four-point diurnal test is a clinical signal. Mid-range for most labs on a saliva collection is roughly 1.5 to 2 ng/mL. Above that at 10pm is a meaningful finding, not noise.

Testosterone below 400 ng/dL with concurrent high cortisol warrants a conversation about whether the suppression is the primary driver. If you address the cortisol and the testosterone corrects, you have your answer. If it does not fully correct, you have a clearer picture of what needs to be treated and why.

Central fat accumulation plus sleep disruption plus mood dysregulation as a triad is a phenotypic presentation that tracks strongly with functional hypercortisolism even when a standard serum draw reads normal. The cluster matters.

When you see a clinician, here is what to ask for specifically: a four-point diurnal cortisol — either salivary or dried urine — not single serum. DHEA-S alongside the cortisol, because it serves as a counter-regulatory marker and tracks adrenal reserve. And the testosterone panel should be free and total testosterone with SHBG, not just total, because SHBG elevation — which cortisol can also drive — will suppress your free testosterone independent of your production rate.

Tsigos and Chrousos's foundational work on HPA neuroendocrinology and the clinical implications of chronic stress laid out how far upstream this mechanism runs (Tsigos & Chrousos, J Psychosom Res 53:865, 2002). The point is not that everyone needs a clinician for this. The point is that when the levers have been addressed and the pattern persists, you deserve a clinician who orders the right tests, not one who looks at a single morning cortisol draw and sends you home.

Our guide on high cortisol symptoms covers the validated screening tests worth requesting and the exact language to use with your prescriber.

Fix the Cortisol Environment First

If you are tired, have low libido, are training hard and not recovering, and your labs keep coming back normal — the problem might not be that nothing is wrong. The problem might be that the wrong things are being measured.

Cortisol and testosterone are not two separate concerns. They are one system. The HPG axis that produces your testosterone is downstream of the HPA axis that manages your stress response. When the HPA axis is chronically activated, the HPG axis pays. That is not a metaphor. It is documented physiology with a forty-year research trail.

Fix the cortisol environment first. Sleep, consistent timing, deload, targeted adaptogens. Get the diurnal curve tested properly. Bring those results to a clinician who knows what they are looking at.

You are not imagining it. And you deserve an answer that addresses what is actually happening.

Key Takeaways

• Cortisol suppression of testosterone is a direct, documented physiological mechanism, not a soft correlation — a single ACTH-driven cortisol spike suppresses LH and testosterone within four hours.
• Chronic stress keeps the HPA axis activated continuously instead of allowing the normal spike-and-recovery pattern, which keeps testosterone production suppressed around the clock.
• Two well-supported pathways drive the suppression — GnRH/LH suppression at the hypothalamus and pituitary, and direct Leydig cell blunting in the testes. The popular "pregnenolone steal" idea is contested and should not be treated as established fact.
• A single morning cortisol draw is calibrated to catch Addison's disease and Cushing's syndrome, not the flattened diurnal curve of chronic stress — a four-point diurnal saliva or dried urine test is what actually captures it.
• Elevated evening cortisol degrades sleep architecture, which independently suppresses testosterone synthesis, creating a self-reinforcing loop between stress, sleep, and hormone production.
• Training harder under chronic cortisol elevation extends the problem rather than fixing it — sleep, consistent wake timing, morning light, and mandatory deload weeks come before chasing testosterone with medication.
• Exogenous testosterone alone does not resolve a chronically elevated cortisol environment — it treats the output while the upstream suppression mechanisms keep running.

References

1. Cumming DC, Quigley ME, Yen SS. Acute suppression of circulating testosterone levels by cortisol in men. J Clin Endocrinol Metab. 1983;57(3):671-673. Link
2. McEwen BS. Protective and damaging effects of stress mediators. N Engl J Med. 1998;338(3):171-179. Link
3. Whirledge S, Cidlowski JA. Glucocorticoids, stress, and fertility. Minerva Endocrinol. 2010;35(2):109-125. Link
4. Fries E, Dettenborn L, Kirschbaum C. The cortisol awakening response (CAR): facts and future directions. Int J Psychophysiol. 2009;72(1):67-73. Link
5. Leproult R, Van Cauter E. Effect of 1 week of sleep restriction on testosterone levels in young healthy men. JAMA. 2011;305(21):2173-2174. Link
6. Duclos M, et al. Testosterone and cortisol responses in overtrained athletes. Eur J Endocrinol. 2003;148:421.
7. Chandrasekhar K, Kapoor J, Anishetty S. A prospective, randomized double-blind, placebo-controlled study of safety and efficacy of a high-concentration full-spectrum extract of ashwagandha root in reducing stress and anxiety in adults. Indian J Psychol Med. 2012;34(3):255-262. Link
8. Tsigos C, Chrousos GP. Hypothalamic-pituitary-adrenal axis, neuroendocrine factors and stress. J Psychosom Res. 2002;53(4):865-871. Link