Testosterone deficiency affects 4.0 to 13.8 million men in the United States and is defined as a combination of low serum testosterone levels plus signs and symptoms of hypogonadism.^1,2 Traditionally thought of as a disease afflicting elderly men, more young men are now presenting with concerns related to testosterone deficiency. However, the diagnosis of testosterone deficiency in young men remains challenging due to uncertainty about what is considered a “normal” testosterone level in younger men.

Testosterone levels decline with age,³ yet urologists and endocrinologists have used an age-independent cutoff to evaluate all men with concerns for testosterone deficiency. The AUA guidelines suggest using a cutoff of 300 ng/dL to define low testosterone.⁴ However, this 300 ng/dL cutoff was derived from testosterone replacement trials that primarily studied men older than 45 years of age. This makes it challenging to evaluate and treat men <45 years old who present with symptoms of hypogonadism but have testosterone levels greater than 300 ng/dL. Evaluation of testosterone deficiency is further complicated by the fact that there are conflicting cutoffs for low testosterone reported in literature and by major professional societies (Table 1). As a result, clinicians have struggled to evaluate, counsel, and treat younger men presenting with concerns about testosterone deficiency.

Table 1. Total Testosterone Cutoffs for Testosterone Deficiency Recommended by Different Professional Organizations.

Given this clinical challenge, we performed the first study evaluating population-based testosterone levels for younger men in the United States. Specifically, we used the 2011 to 2016 National Health and Nutrition Examination Survey (NHANES) to evaluate testosterone levels in men 20 to 44 years old. The NHANES dataset includes cross-sectional, population-based surveys run by the CDC. We analyzed testosterone levels for men 20 to 44 years old who had morning (6:00 AM to 10:00 AM) total testosterone levels measured with isotope dilution high-performance liquid chromatography/tandem mass spectrometry, the gold standard for evaluating testosterone levels. Since NHANES is a weighted sample, the distribution of race and ethnicity is representative of the US population, with 60% identifying as non-Hispanic White, 9.7% as non-Hispanic Black, 21.4% as Hispanic, and 8.9% as Other. We excluded men on hormonal medications (eg, testosterone injections), men with a history of testicular cancer, men with history of orchiectomy, and men with afternoon or evening blood draws. Finally, we used the 2018 AUA Guideline for Testosterone Deficiency definition of “normal testosterone” as the middle tertile of the population to identify age-specific cutoffs for low testosterone levels in men between 20 to 44 years old.

We found that the mean total testosterone of men 20 to 44 years old was 466 ng/dL. Middle tertile, “normal,” testosterone levels were 409 to 558 ng/dL (age 20-24 years), 413 to 575 ng/dL (25-29 years), 359 to 498 ng/dL (30-34 years), 352 to 478 ng/dL (35-39 years), and 350 to 473 ng/dL (40-44 years). Age-specific cutoffs for low testosterone levels were 409, 413, 359, 352, and 350 ng/dL, respectively (Table 2, Figure).

Table 2. Total Morning Testosterone Levels (ng/dL), by Age Group.

This kind of population-based analysis has several limitations. Our age-specific cutoffs for testosterone were modeled off the AUA guideline definition that a normal testosterone level lies within the middle tertile. However, no randomized control trials have been performed to evaluate whether the middle tertile is correlated with hypogonadal symptoms or the likelihood of patients benefiting from treatment of testosterone deficiency. Furthermore, we do not know if our population suffered from hypogonadal symptoms, as the NHANES does not query men for hypogonadal signs and symptoms. Finally, the NHANES only obtains 1 serum testosterone value from each subject.

These limitations notwithstanding, our findings have important implications for the management of young men presenting with concerns regarding testosterone deficiency. This analysis suggests that men 20 to 44 years of age have different testosterone reference ranges than older men. Clinicians should be cognizant of these age-specific differences in testosterone reference ranges and accordingly tailor their approach of young men presenting with testosterone deficiency. For example, consideration should be given to treating young men with hypogonadal symptoms who have testosterone levels below our described age-specific testosterone cutoffs. These men previously may have been disqualified from testosterone treatment based off the age-independent historical 300 ng/dL cutoff. Additionally, these testosterone reference ranges can provide patients with a more realistic and accurate expectation for typical testosterone levels in their age range. This is especially important given the rise of direct-to-consumer online platforms that often do not provide guideline-concordant care in regards to testosterone treatment.⁵ For instance, 1 study by Dubin et al found that 50% of direct-to-consumer platforms recommended that an optimal testosterone level for a 34-year-old man was >1000 ng/dL.⁶

Importantly, we are not suggesting that these cutoffs for testosterone lead to treatment of all men with testosterone levels below the described reference ranges. What remains key to discussion with patients is that these population-based cutoffs must be considered in the context of hypogonadal symptoms and fertility goals. Urologists are uniquely positioned to address concerns regarding testosterone deficiency while ensuring younger patients’ fertility goals are considered when treatment is indicated.