AndroGel (Testosterone Topical) and Erythrocytosis: Alternatives Without This Side Effect

Does AndroGel raise hematocrit?

Testosterone gels do. In the pooled analysis, gels raised mean hematocrit by 3.0% (95% CI 1.8 to 4.3) compared with placebo; that figure covers gels as a group, not AndroGel specifically [1]. The testosterone gel 1% label, a generic label and not the AndroGel 1.62% label, carries a polycythemia warning: increases in hematocrit, reflecting a rise in red blood cell mass, 'may require lowering or discontinuation of testosterone', and an increase in red cell mass may raise the risk of thromboembolic events [2].

In that label's trial tables, hematocrit appears only inside a combined "lab test abnormal" category, which also covers lipid, glucose, potassium, creatinine and bilirubin abnormalities. That category was reported by 6% of men on 50 mg of gel, 5% on 75 mg and 3% on 100 mg over 180 days, and by 9.3% of 162 men in the 3-year extension study, so it gives no hematocrit-specific rate [2].

Observational numbers are higher. In a retrospective single-center study of 247 men on testosterone replacement (194 on testosterone undecanoate, 18 on enanthate and 35 on gel), 57% of patients reached a hematocrit above 0.46, 23% above 0.50 and 5% above 0.54 [4]. Those figures cover the whole cohort, not gel alone.

How does the gel compare with injections?

In the 2022 network meta-analysis of 29 placebo-controlled randomized trials covering 3,393 men, every formulation raised mean hematocrit compared with placebo: gel by 3.0% (95% CI 1.8 to 4.3), intramuscular testosterone enanthate or cypionate by 4.0% (2.9 to 5.1), oral testosterone undecanoate by 4.3%, intramuscular testosterone undecanoate by 1.6% and patch by 1.4% [1]. The only comparison between formulations that reached statistical significance was intramuscular enanthate or cypionate versus patch; gel versus injection did not [1].

Retrospective data show a larger gap. In a single-center review of 178 hypogonadal men followed for 3 years, erythrocytosis (hematocrit above 50%) occurred in 66.7% of men on injectable testosterone, 12.8% on gels and 35.1% on pellets. The injection group was younger (mean 42.5 versus about 54 years), started with higher hematocrit, and the study was not randomized [5]. Reviews rank short-acting injectables highest for erythrocytosis [6], and say all formulations raise hemoglobin and hematocrit, injectables the most [7].

In a nationwide cohort of 53,971 US veterans with type 2 diabetes, the higher odds of erythrocytosis with combined SGLT-2 inhibitor and testosterone therapy were not modified by whether the testosterone was topical or injectable [8].

Which testosterone forms raise it least?

• Patch. Mean hematocrit rose 1.4% (95% CI 0.2 to 2.6) in the network meta-analysis, the smallest increase of the formulations pooled; intramuscular testosterone undecanoate was close behind at 1.6% [1].
• Pellets. In the 178-man review, 35.1% of men on pellets developed a hematocrit above 50% [5]. In a separate retrospective series of 273 men treated with subcutaneous pellets, one patient developed erythrocytosis [9].
• Injections. Highest incidence in both reviews [6][7] and 66.7% above 50% in the retrospective review [5], though the pooled mean change was not significantly different from gel [1].

Non-testosterone options: clomiphene, enclomiphene, hCG

Clomiphene citrate is a selective estrogen receptor modulator that acts centrally to raise luteinizing hormone and follicle-stimulating hormone, increasing testosterone production without suppressing the hypothalamic-pituitary-gonadal axis [10]. In an age-matched comparison of 31 men per group, median testosterone rose from 247 to 504 ng/dL on clomiphene, from 230 to 412 ng/dL on gels and from 224 to 1,104 ng/dL on injections, with similar satisfaction scores across all three [11].

Enclomiphene was compared with AndroGel 1.62% in two parallel randomized, double-blind, placebo-controlled phase III trials in overweight men aged 18 to 60 with secondary hypogonadism. Total testosterone rose in all treatment groups over 16 weeks; sperm concentration stayed in the normal range on enclomiphene while the gel group had a marked reduction in spermatogenesis [12].

Human chorionic gonadotropin-based regimens have been studied for restoring sperm production after testosterone exposure: in a series of 49 men with azoospermia or severe oligospermia on exogenous testosterone, spermatogenesis returned or improved in 47 (95.9%) on hCG plus clomiphene, tamoxifen, anastrozole or recombinant FSH [13].

Is there an alternative that avoids erythrocytosis?

Not among the testosterone formulations compared in the pooled trials: gel, patch, oral testosterone undecanoate and the studied injections all raised mean hematocrit versus placebo [1]. A separate retrospective study reported erythrocytosis in men using pellets [5]. Clomiphene, enclomiphene and hCG are non-testosterone approaches [10][12][13], but the papers cited for them report testosterone, sperm and satisfaction outcomes, not erythrocytosis rates.

What the label says about a rising hematocrit

The AndroGel 1.62% label says to check hematocrit before starting, to re-evaluate it 3 to 6 months after starting, and then annually; it says that if hematocrit becomes elevated, therapy should be stopped until hematocrit decreases to an acceptable concentration [19]. The generic testosterone gel 1% label has the same wording [2].

Published thresholds differ. A 2025 narrative review of testosterone adverse-effect management says therapeutic phlebotomy is indicated when hematocrit exceeds 52% [3], and the 2010 Endocrine Society guideline recommends against starting testosterone therapy in a man with a hematocrit above 50% [15].