AndroGel Cardiovascular Impact Long-Term: What the Evidence Actually Shows

Why Long-Term Cardiovascular Data for AndroGel Took So Long to Arrive

For years, the cardiovascular profile of testosterone gel was inferred from short trials and observational registries, not from a prospective randomized trial powered for hard outcomes. The FDA issued a safety communication in 2014 requiring a new boxed warning about possible increased cardiovascular risk, based largely on two retrospective analyses published in JAMA and PLOS ONE that suggested excess myocardial infarction rates in testosterone users. [1][2]

Those studies had major methodological problems, including lack of active comparator arms and failure to confirm hypogonadism at baseline. Their findings were contested, but they shifted prescribing patterns and prompted the FDA to mandate a large, prospective outcomes trial.

The Gap Between Short-Term Physiology and Long-Term Outcomes

The T-Trials, published in the New England Journal of Medicine in 2016 (N=790, mean age 72), were a coordinated set of seven placebo-controlled trials examining testosterone gel over 12 months. [3] They showed that AndroGel raised serum testosterone from a mean of 234 ng/dL to the mid-normal range and improved sexual function, walking distance, and bone mineral density. The cardiovascular sub-trial of the T-Trials found a statistically significant increase in non-calcified coronary plaque volume in the testosterone arm (mean increase 41 mm3 vs. 11 mm3 placebo, P<0.001). [4] That finding, in older men with pre-existing cardiovascular disease, reinforced regulatory concern.

Twelve months of data, however, cannot answer what happens over three to five years of daily gel application. That question required TRAVERSE.

TRAVERSE Trial: The Definitive Long-Term Cardiovascular Dataset

The Testosterone Replacement therapy for Assessment of long-term Vascular Events and efficacy ResponSE (TRAVERSE) trial enrolled 5,204 men aged 45 to 80 with confirmed hypogonadism (two morning testosterone readings <300 ng/dL) and either pre-existing cardiovascular disease or high cardiovascular risk. [5] Participants were randomized to testosterone gel 1.62% (target serum testosterone 350 to 750 ng/dL) or matching placebo gel. Median follow-up was 33 months.

Primary Endpoint: MACE Was Non-Inferior

The primary endpoint was time to first major adverse cardiovascular event (MACE), defined as non-fatal myocardial infarction, non-fatal stroke, or cardiovascular death. The testosterone arm reached a MACE rate of 7.0% versus 7.3% in the placebo arm (hazard ratio 0.96, 95% CI 0.78 to 1.17). [5] The pre-specified non-inferiority margin was 1.5, so the trial met its primary endpoint. Testosterone gel did not increase the risk of the composite MACE outcome in this high-risk population over a median of nearly three years.

Secondary Endpoints: Where the Signal Emerged

Non-inferiority on MACE does not mean cardiovascular neutrality. Three secondary outcomes showed statistically higher rates in the testosterone arm:

• Atrial fibrillation: 3.5% testosterone vs. 2.4% placebo
• Pulmonary embolism: 0.9% testosterone vs. 0.5% placebo
• Acute kidney injury: 2.3% testosterone vs. 1.5% placebo

The atrial fibrillation finding is consistent with prior observational data. A 2018 meta-analysis of 10 trials (N=2,971) published in the European Heart Journal found that testosterone therapy roughly doubled the odds of atrial fibrillation (OR 2.26, 95% CI 1.33 to 3.86). [6] The pulmonary embolism signal connects mechanistically to the polycythemia risk discussed in the next section.

Polycythemia, Hematocrit, and Thromboembolic Risk

Testosterone is a potent stimulator of erythropoiesis. Transdermal formulations produce less erythropoiesis than intramuscular testosterone esters because they avoid the supraphysiologic testosterone peaks seen after an injection, but the effect is not zero. [7]

What the Numbers Look Like

In TRAVERSE, erythrocytosis (hematocrit above 54%) occurred in 5.8% of testosterone-treated men versus 1.0% of placebo men. [5] That fivefold difference in erythrocytosis rates is the most likely mechanistic driver behind the pulmonary embolism gap.

The current FDA-approved label for AndroGel 1.62% states that therapy should be interrupted if hematocrit exceeds 54%, with dose reduction or discontinuation based on whether hematocrit normalizes on re-check. [8] Clinicians should obtain a baseline complete blood count before initiation and recheck at 3 to 6 months, then annually thereafter.

Venous Thromboembolism Beyond Pulmonary Embolism

A population-based cohort study published in BMJ in 2012 (N=44,115 men initiating testosterone therapy) found a twofold increase in venous thromboembolism risk in the first six months of therapy (rate ratio 2.0, 95% CI 1.5 to 2.7). [9] Gel formulations were not analyzed separately in that study, but the biologic mechanism, increased red cell mass and viscosity, applies across routes of administration.

Men with a personal or strong family history of deep vein thrombosis or pulmonary embolism, and men with known thrombophilic conditions such as Factor V Leiden, should be counseled explicitly about this risk before starting AndroGel.

Blood Pressure and Lipid Effects of Testosterone Gel

Systolic Blood Pressure

Data on testosterone and blood pressure are mixed. Short-term trials generally show either a neutral or mildly favorable effect, with some vasodilatory activity attributed to androgen receptor-mediated endothelial nitric oxide signaling. [10] In TRAVERSE, baseline-to-month-12 systolic blood pressure changes did not differ significantly between arms. However, men with poorly controlled hypertension (systolic above 160 mmHg) were excluded from TRAVERSE, so the dataset does not represent the highest-risk hypertensive patients seen in everyday practice.

LDL, HDL, and Triglycerides

A 2022 meta-analysis of 30 randomized controlled trials published in Annals of Internal Medicine found that testosterone therapy lowered LDL by a mean of 5.3 mg/dL and triglycerides by 22 mg/dL, while also lowering HDL by 3.2 mg/dL. [11] The clinical significance of these modest shifts on a background of statin therapy, which is nearly universal in men with the cardiovascular risk profile of the TRAVERSE population, is likely small. Lipid panels should still be checked at 3 to 6 months and annually.

Cardiac Morphology: Coronary Plaque and Left Ventricular Hypertrophy

Coronary Plaque Progression

The cardiovascular sub-study of the T-Trials used coronary CT angiography to measure plaque volume at baseline and 12 months in 138 men. Non-calcified plaque volume increased by a mean of 41 mm3 in the testosterone group versus 11 mm3 in the placebo group (P<0.001). [4] Non-calcified plaque is the type most associated with acute coronary syndromes, which gives this finding clinical weight despite the small sample size.

The TRAVERSE investigators did not collect coronary CT angiography data at scale, so it is unknown whether the plaque progression signal seen at 12 months in the T-Trials would continue, plateau, or reverse over three or more years of therapy.

Left Ventricular Effects

Supraphysiologic testosterone levels, as seen with anabolic steroid abuse, cause left ventricular hypertrophy and myocardial fibrosis. Replacement dosing is a different situation. A 52-week randomized trial in 423 men with hypogonadism found no significant change in left ventricular mass index with testosterone gel versus placebo. [12] Keeping serum testosterone within the normal physiologic range, generally 350 to 750 ng/dL with AndroGel 1.62% per the product label, appears to avoid the adverse cardiac remodeling seen at supraphysiologic levels.

Atrial Fibrillation: Mechanisms and Clinical Management

The atrial fibrillation finding from TRAVERSE (3.5% vs. 2.4%, absolute risk difference 1.1 percentage points over 33 months) deserves its own section because it represents the clearest actionable safety signal from the trial.

Proposed Mechanisms

Androgens may promote atrial fibrillation through several pathways: activation of cardiac ion channels that shorten atrial refractory period, promotion of atrial fibrosis, and indirect effects mediated by increases in red cell mass that raise right atrial pressure. [13] In men with pre-existing structural heart disease or enlarged left atria, these effects may reach a threshold that triggers arrhythmia.

Pre-Treatment Screening and Monitoring

The American Heart Association's 2023 scientific statement on sex hormones and cardiovascular disease recommends that clinicians obtain a baseline electrocardiogram in men over 65 or those with known structural heart disease before initiating testosterone therapy. [14] Any new palpitations, exertional dyspnea, or irregular pulse during therapy should prompt an ECG and, if AF is confirmed, cardiology referral before continuing testosterone.

Practical Decision Points for AF-Risk Patients

Men who are already anticoagulated for a prior AF episode represent a distinct group. Testosterone-related erythrocytosis raises hematocrit, which can affect time-in-therapeutic-range for warfarin. In these patients, direct oral anticoagulants are preferred over warfarin during testosterone therapy, and hematocrit monitoring every three months rather than every six is reasonable based on the TRAVERSE safety profile.

FDA Label Changes and Regulatory History

2014 Boxed Warning Addition

The FDA added a new warning to all testosterone products in March 2015 noting that "cardiovascular risk has been reported with use of testosterone products." [8] This language was added after the agency reviewed the observational data from Finkle et al. (PLOS ONE, 2014) and Vigen et al. (JAMA, 2013), two studies that showed increased myocardial infarction rates in testosterone users despite their methodological limitations. [1][2]

Post-TRAVERSE Label Field

TRAVERSE was completed after the boxed warning was added. The trial's non-inferiority finding on MACE did not lead the FDA to remove the cardiovascular risk language from the label, partly because of the atrial fibrillation and pulmonary embolism signals in secondary endpoints. As of the 2024 prescribing information update, AndroGel's label still carries warnings about cardiovascular risk, thromboembolism, and erythrocytosis, while acknowledging that MACE equivalence was demonstrated in TRAVERSE for men without a history of prior MI or stroke who had high-risk features only. [8]

Who Should Not Use AndroGel: Absolute and Relative Contraindications

Absolute Contraindications

Men with known or suspected prostate carcinoma or breast carcinoma should not receive AndroGel. The FDA label also lists untreated severe sleep apnea as a contraindication because testosterone worsens upper airway obstruction. [8]

Relative Contraindications Based on Cardiovascular Profile

Based on the TRAVERSE and T-Trials data combined, the following groups warrant especially careful individualized risk-benefit discussions before starting gel therapy:

• Men with recent acute coronary syndrome (within 90 days): TRAVERSE excluded men within 90 days of a cardiac event, so there is no safety data for this window.
• Men with known atrial fibrillation who are not anticoagulated: the TRAVERSE AF signal applies most directly to this group.
• Men with hematocrit above 50% at baseline: close to the threshold where treatment interruption would be required.
• Men with Factor V Leiden, prothrombin gene mutation, or other inherited thrombophilia: the pulmonary embolism signal makes VTE prophylaxis discussion necessary.

The Endocrine Society's 2018 clinical practice guideline on male hypogonadism recommends against testosterone therapy in men who had a myocardial infarction or stroke within the previous six months. [15]

Monitoring Protocol: Practical Clinical Guidance

Baseline Workup Before First Prescription

Before writing a first prescription for AndroGel, the following should be documented in the chart:

• Two morning serum testosterone levels (obtained before 10 AM, at least one week apart) confirming values <300 ng/dL
• LH and FSH to classify primary vs. Secondary hypogonadism
• Complete blood count including hematocrit
• Fasting lipid panel
• PSA and digital rectal examination (or structured risk discussion for patients declining DRE)
• Blood pressure with mean of two readings
• 12-lead ECG in men over 65 or those with structural heart disease

On-Treatment Monitoring Schedule

The Endocrine Society guideline specifies serum testosterone measurement at 3 to 6 months after dose initiation, targeting a mid-normal range of 400 to 700 ng/dL for gel formulations. [15] Hematocrit should be rechecked at the same visit.

After the first year, annual CBC, PSA, lipid panel, and testosterone level constitute the minimum surveillance set. If hematocrit climbs above 54% at any point, gel should be held until hematocrit falls below 50%, then restarted at a lower dose or with a less frequent application schedule.

Target serum testosterone for AndroGel 1.62% is 350 to 750 ng/dL per label; the Endocrine Society prefers the 400 to 700 ng/dL window to minimize the erythrocytosis and AF risk visible at the higher end of the therapeutic range.

Benefit-Risk Summary Across Cardiovascular Domains

The evidence base for AndroGel's cardiovascular profile is now substantially richer than it was a decade ago. The table below summarizes the primary outcomes data:

| Cardiovascular Domain | Direction of Effect | Key Source | |---|---|---| | MACE (MI, stroke, CV death) | Non-inferior to placebo (HR 0.96) | TRAVERSE [5] | | Atrial fibrillation | Increased (3.5% vs. 2.4%) | TRAVERSE [5] | | Pulmonary embolism | Increased (0.9% vs. 0.5%) | TRAVERSE [5] | | Erythrocytosis (Hct above 54%) | Fivefold increase (5.8% vs. 1.0%) | TRAVERSE [5] | | Coronary plaque (non-calcified) | Increased at 12 months | T-Trials CV sub-study [4] | | LDL cholesterol | Modestly reduced (~5 mg/dL) | Meta-analysis [11] | | Left ventricular mass | No significant change at replacement doses | RCT data [12] | | Systolic blood pressure | Neutral in controlled trials | TRAVERSE [5] |

For a man with symptomatic hypogonadism and no prior atrial fibrillation, no thrombophilia, hematocrit below 50%, and no recent acute coronary event, the TRAVERSE data support that the benefit of symptom improvement and the absence of excess MACE risk outweigh the modestly elevated AF and PE rates. For a man with two or more of those risk factors, a conservative strategy such as watchful waiting, lifestyle optimization, and repeat testosterone confirmation before prescribing is appropriate.