Secondary Hypogonadism and Effects of Testosterone Replacement Therapy on Cardiovascular Events

In 408 men with pituitary tumors followed 8 years, TRT showed no increase in heart attack or stroke risk — but untreated low T was tied to higher mortality

Journal: Journal of Clinical Endocrinology and Metabolism | Published: 2026-04-22 | Type: Retrospective cohort study | PMID: 41252267 Authors: Munro V, Law J, Matheson K, Imran SA (Division of Endocrinology, Dalhousie University, Halifax, Canada) Funding/COI: Not listed for either

Summary

Munro et al., 2026 followed 408 men from the Halifax Neuropituitary Registry with nonfunctioning pituitary adenomas or prolactinomas over a median 8.1 years, comparing major adverse cardiovascular events (MACE) across three groups: no secondary hypogonadism (SHG), TRT-treated SHG, and untreated SHG. TRT-treated men showed no elevated MACE risk relative to eugonadal controls. The untreated group had higher all-cause mortality, though they were also older and carried more baseline disease burden.

Claims

• No significant difference in MACE (stroke, MI, cardiovascular death) between TRT-treated SHG (n=214), no-SHG (n=150), and untreated SHG (n=44) by multivariable logistic regression
• Untreated SHG was associated with increased mortality compared to both TRT-treated SHG and no-SHG groups
• Untreated SHG patients had significantly lower median testosterone: 3.2 nmol/L (IQR 0.6–5.1) vs 14.8 nmol/L in both treated and eugonadal groups
• MACE outcomes were not affected by adenoma size, presence of additional hormonal deficits, or achieved testosterone level
• TRT-treated group split: 55.6% intramuscular depot, 44.4% transdermal; all titrated to mid-normal range
• Untreated group was meaningfully older (mean 61.7 vs 53.7 years in no-SHG) and had higher baseline CVD (27.3% vs 15%)

Study Quality

This is a single-center retrospective cohort study drawing from a prospective registry (Halifax Neuropituitary Registry, 2006–2023, Nova Scotia province ~1M population). Outcomes were cross-referenced against provincial cardiac and stroke registries, which reduces the ascertainment bias typical of chart-only reviews. The cohort was followed systematically — patients seen at 3, 6, and 12 months post-diagnosis, then annually — and testosterone adequacy was verified via serial measurements and prescription pickup records. The authors pre-specified a sample size of 398 to detect a >5% difference in event rates at α=0.05, and they met it.

That said, the retrospective design cannot establish causation, and the untreated group (n=44) is small. Multivariable adjustment was used to account for the older age and higher comorbidity burden in untreated patients, but residual confounding is likely: the decision not to treat was itself driven by clinical severity, making the untreated arm a selected high-risk group rather than a clean control. The mortality signal in untreated SHG needs to be read with that caveat prominently in mind.

Red Flags

• Funding and conflicts of interest both unlisted — for a TRT safety paper, that omission is a concern
• Untreated SHG group (n=44) is underpowered for mortality subgroup analysis; small cells inflate apparent risk
• Untreated group was significantly older (mean 61.7 vs 53.7 years) and had 1.8× the baseline CVD rate (27.3% vs 15%) — residual confounding likely survives multivariable adjustment
• Single-center, single-province study; Nova Scotia's demographics and healthcare access may not generalize
• "Untreated" partly reflects refusal, prostate cancer concerns, or side effects — not a random allocation; these are sicker or more cautious patients
• Median follow-up 8.1 years sounds long, but IQR of 3.3–14.1 indicates wide variance; shorter-followed patients contribute less event data

Strengths

• First study specifically assessing CV safety of TRT in the SHG-with-pituitary-disease population — a genuine gap in the literature that hypopituitarism guidelines have flagged as unresolved
• Long median follow-up (8.1 years) relative to prior TRT trials; the landmark TRAVERSE trial averaged only 33 months
• Provincial registry linkage for cardiac and stroke outcomes reduces underreporting
• Well-defined SHG diagnosis criteria (two testosterone draws ≥8 weeks apart, morning fasting, confirmed low LH/FSH) rather than relying on a single value
• TRT adequacy verified prospectively with standardized titration to mid-normal range — addresses a key methodological weakness of prior database studies

Verdict

This paper fills a specific and legitimate gap: prior TRT safety data largely excluded men with structural pituitary disease, and the one trial most cited (TRAVERSE) excluded men with testosterone below 3.5 nmol/L — exactly the population this study treats. The null finding on MACE is credible for what it is: a reasonably powered retrospective cohort with outcome registry linkage and long follow-up. The mortality signal in untreated men is hypothesis-generating at best given the confounding, and the authors acknowledge this. The missing funding and COI disclosures are an editorial failure by the journal, not necessarily a scientific one, but readers should note it. For a first-in-population study, this is solid groundwork — not definitive, but a meaningful addition to the safety literature for a population that has historically been studied by exclusion.