The Diagnostic Bar — and Why It Exists

Testosterone deficiency is not diagnosed by a number alone. The American Urological Association is explicit on this: the clinical diagnosis requires both a consistently low testosterone level and symptoms or signs that are plausibly related to that deficiency.[1] A low number without symptoms is not sufficient to treat. Symptoms without a confirmed low number are not sufficient either.

The threshold the AUA uses is a total testosterone below 300 ng/dL — confirmed on at least two separate early-morning measurements, ideally at the same laboratory.[1] The morning timing matters because testosterone follows a daily rhythm, peaking in the early hours and falling through the afternoon. A single low-normal result drawn at 3pm is not a diagnosis.

📊 The Evidence — Which Symptoms Actually Point to Low T

The European Male Aging Study established that after accounting for age, only three symptoms had a reliable relationship with low testosterone: poor morning erections, decreased libido, and erectile dysfunction.[2] These are the symptoms that most specifically suggest testosterone deficiency — not fatigue, not low energy, not brain fog. The AUA also recommends testing in certain high-risk situations even without symptoms: unexplained anemia, significant bone loss, diabetes, HIV, chronic opioid use, pituitary dysfunction, prior chemotherapy or testicular radiation, and male infertility.[1]

⚠ Look for the Underlying Cause First Before treating low testosterone, it's worth understanding why it's low. Obesity, type 2 diabetes, sleep apnea, and opioid medications are common causes of functional hypogonadism — low testosterone driven by something else entirely. Treating the underlying cause often normalizes testosterone without any replacement therapy. Prescribing testosterone without this step risks treating a symptom of the real problem while leaving the real problem unaddressed.[3]

What About Exercise?

Patients frequently ask whether exercise can raise testosterone levels — and the honest answer is: acutely yes, chronically not much, with some important exceptions.

📊 The Evidence — Exercise and Testosterone

Exercise does produce a short-term spike in testosterone — a meta-analysis of 48 studies found a meaningful acute increase after both moderate and vigorous activity.[8] But this spike returns to baseline within about 30 minutes, and resistance training's boost lasts only slightly longer. High-intensity exercise can actually suppress testosterone for up to 72 hours afterward due to cortisol's opposing effects.[9]

At the level that matters most — resting baseline testosterone — a systematic review of 11 randomized trials found that exercise training has a negligible effect in men who already have normal levels.[10] The important exception is overweight and obese men: vigorous aerobic exercise can meaningfully raise total, free, and bioavailable testosterone in this group, likely because weight loss itself restores normal hormonal signaling.[11] High-intensity interval training (HIIT) has also shown roughly 28–30% increases in total and free testosterone in physically inactive middle-aged adults in one trial.[12]

Particularly striking: in a randomized trial of men aged 50–70 with low-normal testosterone, exercise training was more effective than testosterone treatment for improving aerobic capacity, strength, and reducing visceral fat.[13] Testosterone did increase lean mass, but when combined with exercise it added little beyond what exercise alone achieved. For men with low-normal testosterone and symptoms that could plausibly be explained by deconditioning or obesity, a structured exercise program is worth trying before reaching for a prescription.

What Treatment Actually Improves — and By How Much

The largest and most rigorous clinical trial program on testosterone in men is the Testosterone Trials (TTrials), a coordinated set of seven placebo-controlled trials in older men with documented low testosterone. Combined with the more recent TRAVERSE trial, they give us the clearest picture available of what testosterone therapy actually does.

Outcome Effect Magnitude
Libido / sexual desire Improved ~25% increase; small effect size (SMD 0.17–0.25)
Sexual activity Improved ~40% increase (~0.58 acts/week); small effect size (SMD 0.23)
Erectile function Modest ~35% improvement on scoring scale; less effective than sildenafil (Viagra)
Anemia correction Clinically meaningful Hemoglobin rose ≥1.0 g/dL in 33–50% of men with baseline anemia
Bone mineral density Improved Increased volumetric bone density and bone strength
Physical function Modest 6–7 meter improvement in 6-minute walk distance; did not improve in men who walked slowly at baseline
Fatigue / energy Inconsistent 4–5% improvement in one trial; no improvement on FACIT Fatigue scale in TTrials
Mood / depression Slight 3–4% improvement overall; no benefit in men with diagnosed major depression
Cognition No improvement No benefit found across trials
📊 The Evidence — Effect Sizes in Context

The improvement in erectile function with testosterone — about 2.6 points on a standard scoring scale — is less than half the improvement seen with sildenafil (Viagra), which produces about 5.7 points.[4] This means testosterone is unlikely to be the right treatment when erectile dysfunction is the primary complaint, unless testosterone is genuinely and substantially low.

Effect sizes across sexual outcomes are consistently small by statistical standards — a standardized mean difference of about 0.2, where 0.5 is considered medium and 0.8 large.[5][6] These are real improvements, but they're meaningful primarily in men with unequivocally low testosterone. The lower the starting testosterone level — particularly below 200 ng/dL — the greater the likelihood of a noticeable benefit.[4]

Formulations: How Testosterone Is Delivered

Testosterone replacement comes in several forms. The right choice depends on patient preference, lifestyle, skin sensitivity, and whether fertility is a concern. Each has tradeoffs.

Topical gels and creams

Applied daily to skin (shoulders, upper arms, or abdomen). Convenient but require care to avoid transfer to partners or children through skin contact. Most widely prescribed formulation.

Injections Most cost-effective

Testosterone cypionate is my preferred formulation for most patients — it's inexpensive, widely available as a generic, and can be self-administered at home with proper instruction. Given every 1–2 weeks. The main tradeoff is that levels peak shortly after injection and trough before the next dose, which some patients notice as energy or mood variation. Weekly smaller doses can smooth this out.

Patches

Applied nightly to skin. Steady levels. Significant skin irritation in many patients limits use.

Pellets

Implanted under the skin every 3–6 months. Cannot be easily adjusted or removed if side effects occur or the dose needs to change. Associated with higher rates of supraphysiologic levels. I don't recommend them for this reason.

Nasal gel

Applied inside the nose three times daily. Produces rapid peaks and troughs. Minimal transfer risk. Inconvenient dosing schedule limits adherence.

Oral (newer formulations)

Testosterone undecanoate taken with food. Avoids the liver toxicity of older oral androgens. Requires twice-daily dosing and has more variable absorption than other routes.

Monitoring on Testosterone Therapy

Starting testosterone is not a set-it-and-forget-it decision. Regular monitoring matters both to confirm the treatment is working and to catch the most important risks before they become problems.

What to Check and When

After starting or adjusting therapy, testosterone levels and hematocrit (the proportion of red blood cells in the blood) are checked at 3–6 months, then annually once stable. The target range for total testosterone is the mid-normal range — roughly 400–700 ng/dL — not the upper end of normal or above it. PSA (prostate-specific antigen) is checked before starting in men over 40 and monitored annually thereafter.

Erythrocytosis — The Most Important Risk to Monitor

Testosterone stimulates red blood cell production, which is why it can help men with anemia — but it can also push hematocrit too high, a condition called erythrocytosis (too many red blood cells). When blood becomes too thick, the risk of clots, stroke, and cardiovascular events rises. Hematocrit above 54% generally requires dose reduction, temporary discontinuation, or therapeutic phlebotomy (a controlled blood draw to reduce red cell concentration).[1]

⚠ A Simple Solution: Donate Blood For patients whose hematocrit is trending upward on testosterone therapy, regular blood donation is an underused and underappreciated option. It reduces red cell concentration just as effectively as therapeutic phlebotomy, it's free, it's available at most community blood centers, and — unlike a clinic procedure — it directly benefits other people. I encourage patients with rising hematocrit to donate whole blood every 8 weeks if their levels warrant it. It's one of the few situations in medicine where managing a side effect and helping your community are exactly the same action.
⚠ Other Risks Worth Knowing
  • Sleep apnea: Testosterone can worsen existing sleep apnea or unmask it in predisposed men. Screen before starting and monitor symptoms.
  • Prostate: Testosterone does not cause prostate cancer, but it can stimulate growth of existing prostate cancer. PSA monitoring is standard. Men with active or high-risk prostate cancer should not receive testosterone.
  • Skin: Acne and oily skin are common, particularly early in treatment.
  • Cardiovascular: The picture is complex and covered in detail in Post 2.5. Short version: TRAVERSE found no increase in major cardiovascular events in men with pre-existing cardiovascular disease, but erythrocytosis risk requires management.

Fertility: The Conversation That Can't Wait

This is the part of the testosterone conversation I consider non-negotiable to have before starting treatment, regardless of a patient's age or stated plans. Testosterone replacement suppresses the pituitary signals (LH and FSH) that drive sperm production. Most men on testosterone therapy produce little to no sperm while on treatment. This effect is often reversible after stopping — but recovery takes months and is not guaranteed, particularly after long-term use.

📊 The Evidence — Fertility Impact Testosterone therapy reliably suppresses sperm production, often to the point of azoospermia (no detectable sperm), in the majority of men within 3–4 months of starting.[1] Recovery of sperm production after stopping testosterone typically takes 6–18 months, and in a minority of men — particularly after years of use — it may not recover fully.[3] For men who want to preserve fertility options, sperm banking before starting therapy is straightforward and worth discussing. Alternatively, human chorionic gonadotropin (hCG), which mimics LH, can be used alongside testosterone to maintain some testicular function during treatment, though it adds cost and complexity.
My Synthesis The fertility conversation is one I have with every man before starting testosterone, including men in their 50s and 60s who say they're done having children. Circumstances change, relationships change, and the consequences of not having this conversation are irreversible in a way that's hard to undo. I'd rather spend ten minutes on a conversation that turns out to be unnecessary than skip it and have a patient regret it later.

On treatment itself: the outcomes table above is worth sitting with. The benefits are real but modest for most men. Sexual function improves — that's the most consistent finding and the most clinically meaningful in men with documented deficiency. Fatigue and energy are unlikely to improve unless anemia is part of the picture. I tell patients upfront: if fatigue is your primary complaint and your sexual function is fine, testosterone is probably not going to solve it.