AndroGel Variable Absorption: Alternatives Without This Side Effect

What varies with a gel, and what the label lists

AndroGel is applied to the skin. The Teva testosterone gel 1% label says topical testosterone products may have different doses, strengths or application instructions that result in different systemic exposure, and that its own dosage and administration differ from testosterone gel 1.62% [5]. That label is not AndroGel's, so its figures are not AndroGel 1.62% figures.

The label lists blood pressure increases, increased red blood cell mass, venous thromboembolism, prostate risks, effects on spermatogenesis, edema and sleep apnea among its warnings [5].

How much of a gel dose reaches the blood

The clearest published numbers come from a 2000 study in the Journal of Clinical Endocrinology and Metabolism. Swerdloff and colleagues followed 227 hypogonadal men applying 1% testosterone gel at 50 mg or 100 mg daily, or wearing a testosterone patch, with blood sampling on days 1, 30, 90 and 180. About 9% to 14% of the testosterone applied in the gel was bioavailable, and an earlier study had shown steady state levels 48 to 72 hours after the first application [1].

Two details matter for anyone whose levels move around.

First, within a dose, mean exposure held steady. Pharmacokinetic parameters for serum total and free testosterone were very similar on days 30, 90 and 180 in all treatment groups, and mean serum steady state levels remained stable over the 180 days of gel application [1]. The study reports group averages, not each man's levels.

Second, dose changes did not scale in a straight line. Raising men from 50 mg to 75 mg did not significantly increase their average 24-hour testosterone, while lowering men from 100 mg to 75 mg reduced levels to the normal range for most of them [1].

Intramuscular testosterone cypionate and enanthate

Intramuscular testosterone skips the skin entirely. The ester is injected in an oil vehicle, sits in a depot, and is released and split to free testosterone slowly rather than crossing the outer skin layer.

Injections remove the skin-absorption step, but they still share the blood pressure increase identified across testosterone products [2]. In a network meta-analysis of 29 placebo-controlled randomized trials covering 3,393 men, mean hematocrit rose 4.0 percentage points with intramuscular enanthate or cypionate (95% CI 2.9 to 5.1) and 3.0 points with gel (1.8 to 4.3) compared with placebo, while the patch rose 1.4 points (0.2 to 2.6) [3]. The intramuscular esters raised hematocrit significantly more than the patch; the other comparisons between formulations were not statistically significant [3].

A retrospective single-center review of 178 hypogonadal men followed for three years found erythrocytosis, defined as hematocrit above 50%, in 66.7% of men on injectable testosterone, 12.8% on gels and 35.1% on pellets [6]. That comparison has an important caveat: the injection group was younger (mean 42.5 years versus about 54) and started with a higher baseline hematocrit, and the study was not randomized [6].

Cardiovascular outcome data are route-specific as well. TRAVERSE, the 5,246-man trial behind FDA's class-wide label changes, tested a 1.62% testosterone gel against placebo: a major adverse cardiovascular event occurred in 7.0% of men on testosterone and 7.3% on placebo [7]. No outcome trial of that size exists for injectable testosterone [7].

Weekly subcutaneous testosterone enanthate

Xyosted is testosterone enanthate injected under the skin once a week, so it does not depend on getting testosterone through the skin barrier the way a gel does. It does not avoid the blood pressure effect: FDA's ambulatory monitoring studies confirmed a blood pressure increase with all testosterone products, class-wide [2].

What about the testosterone patch?

The patch is one of the FDA-approved formulations, along with gel, the buccal system and injection [4]. It is transdermal too, so it does not skip skin, but its estimated hematocrit rise was smaller than gel's in the 29-trial analysis (1.4 points versus 3.0), though that difference was not statistically significant, while the patch versus intramuscular ester comparison was [3].

What stays the same whichever route is used

Blood pressure. FDA's February 2025 class-wide labeling action followed ambulatory blood pressure studies that, in the agency's words, confirmed an increase in blood pressure with use of all testosterone products [2]. The testosterone gel 1% label reports a mean 24-hour rise of 1.9/1.3 mm Hg at week 16 (169 men with usable recordings), and 3.0/2.2 mm Hg in men already treated for hypertension; it also says to measure blood pressure periodically and that the product is not recommended in men with uncontrolled hypertension [5].

Hematocrit. The same label says to check hematocrit before starting treatment and to re-evaluate it 3 to 6 months later and then annually [5].

Clots. Venous thromboembolism has been reported with testosterone products; the label reports 1.7% of men on testosterone versus 1.2% on placebo in TRAVERSE [5].

Fertility. The label says exogenous androgens may lead to azoospermia [5], and the Endocrine Society guideline recommends against starting testosterone therapy in men planning fertility in the near term, with monitoring of symptoms, serum testosterone, hematocrit and prostate cancer risk during the first year [8]. Recovery after stopping is variable: in a retrospective study of 66 men with infertility after testosterone use, 70% reached a total motile count above 5 million within 12 months of stopping testosterone and starting high-dose hCG plus a selective estrogen receptor modulator [9].

Household exposure. Gels carry a boxed warning about secondary exposure, because virilization has been reported in children exposed to testosterone gel [5]. That warning belongs to the topical products.

The trade-off in one paragraph

A gel gives daily dosing without a needle, and in the 2000 study about 9% to 14% of the applied dose was bioavailable, with dose changes that did not move average levels in a straight line [1]. Injections skip the skin and stretch dosing out, but they do not change testosterone's class effects, and pooled trial data estimate a hematocrit rise of 4.0 points on intramuscular esters versus 3.0 on gel, a difference that was not statistically significant [3].