TRT is safe when it's properly monitored and risky when it isn't. The monitoring schedule exists to catch the handful of real problems early — and to avoid the common mistake of crushing estrogen along the way.
Check at baseline, then at 3–6 months, then annually: testosterone (timed to the formulation, aiming mid-normal), haematocrit, and — in appropriate-age men — PSA with a prostate exam. Stop or reduce the dose if haematocrit exceeds ~54%; get a urology opinion if PSA rises more than 1.4 ng/mL in the first year or crosses 4.0. Evidence ●●●●●
The most common is erythrocytosis — testosterone raises red-blood-cell production, thickening the blood (hence the haematocrit checks). Others: acne and oily skin, fluid retention, a small HDL dip, worsening of existing sleep apnoea, accelerated hair loss in genetically predisposed men (via DHT), and gynaecomastia (via conversion to estrogen). Mood generally improves in hypogonadal men — the 'roid rage' idea doesn't hold at physiologic replacement doses.
The large TRAVERSE trial (~5,200 hypogonadal men at high cardiovascular risk) found TRT did not increase major adverse cardiac events (heart attack, stroke, cardiovascular death) versus placebo. But it did show higher rates of atrial fibrillation, acute kidney injury, and pulmonary embolism / blood clots — reassuring on the headline, but worth counselling and monitoring, not a free pass. Evidence ●●●●○
Some testosterone converts to estradiol, and men genuinely need estradiol for bone density, libido and mood. Aromatase inhibitors (like anastrozole) are widely over-used to 'control estrogen' on TRT — but crushing estradiol harms bone and libido and the evidence doesn't support routine use. An AI is for genuinely symptomatic high estradiol or gynaecomastia, not a default add-on. Evidence ●●●●○ (against routine AI use)
When not to start
TRT is generally not started with active prostate or breast cancer, an unevaluated PSA above ~4, an already-high haematocrit, untreated severe sleep apnoea, a heart attack or stroke in the last ~6 months, a clotting disorder, or a desire for children in the near term. These are clinician judgements — and reasons the baseline work-up matters.
The Testosterone Guide