What are the normal reference ranges for dehydroepiandrosterone sulfate (DHEA‑S) and total testosterone in an adult woman being evaluated for hirsutism?

Normal DHEA-S and Testosterone Reference Ranges in Female Hirsutism Evaluation

When evaluating hirsutism in women, normal DHEA-S and testosterone levels do not exclude hyperandrogenism, and the diagnostic approach must prioritize calculated free testosterone or free androgen index over total testosterone alone, using liquid chromatography-tandem mass spectrometry (LC-MS/MS) when available.

First-Line Androgen Testing

Measure total testosterone (TT) and calculated free testosterone (cFT) or free androgen index (FAI) as your initial biochemical assessment, not DHEA-S. 1

• Total testosterone by LC-MS/MS demonstrates 74% sensitivity and 86% specificity for detecting hyperandrogenism, with higher specificity (92%) than direct immunoassays (78%). 1
• Calculated free testosterone shows the highest diagnostic accuracy at 89% sensitivity and 83% specificity when calculated using the Vermeulen equation from high-quality TT and SHBG measurements. 1
• Free androgen index (FAI = total testosterone/SHBG ratio) provides 78% sensitivity and 85% specificity, serving as an acceptable alternative when LC-MS/MS is unavailable. 1
• Obtain samples in the morning (8–10 AM) while fasting to maximize diagnostic accuracy and allow concurrent metabolic screening (fasting glucose, 2-hour oral glucose tolerance test, lipid panel). 1

Second-Line Testing: DHEA-S and Androstenedione

Only measure DHEA-S and androstenedione if first-line testosterone measurements are normal but clinical suspicion remains high, as these markers have substantially poorer specificity. 1

DHEA-S Characteristics

• Sensitivity 75%, specificity only 67% for detecting hyperandrogenism—the lowest specificity among all androgen markers. 2
• Only 8–33% of PCOS patients have elevated DHEA-S, with higher rates in certain phenotypes and ethnic groups (approximately 20% in White patients, 33% in Black patients when age-adjusted). 1
• Age-adjusted reference ranges are mandatory because DHEA-S peaks between ages 20–30 years and declines steadily thereafter; failure to use age-specific cutoffs leads to overdiagnosis. 1
• DHEA-S >600 μg/dL (>16.3 μmol/L) raises concern for an adrenal source, particularly adrenocortical carcinoma, and warrants adrenal imaging. 1

Androstenedione Characteristics

• Sensitivity 75%, specificity 71% for detecting hyperandrogenism—better than DHEA-S but still inferior to testosterone measurements. 2
• Particularly useful when SHBG is low (<30 nmol/L), as it is less affected by SHBG fluctuations than total testosterone. 1

Critical Diagnostic Pitfalls

Normal Androgens Do Not Exclude Hyperandrogenism

Up to 30% of women with clinical hyperandrogenism (hirsutism, acne, androgenic alopecia) have normal total testosterone but elevated free testosterone or FAI. 3

• SHBG fluctuations due to age, weight, oral contraceptives, or metabolic factors can mask hyperandrogenism by lowering free testosterone despite normal total testosterone. 1
• Direct immunoassay methods for free testosterone must be avoided in women due to poor accuracy at low serum concentrations typical of the female range. 1
• Relying solely on DHEA-S misses the majority of cases: 67–92% of PCOS patients have normal DHEA-S levels. 1

Timing and Hormonal Contraception

• Hormonal contraception (including progestin-only implants) suppresses the hypothalamic-pituitary-ovarian axis, making all androgen measurements unreliable; remove or allow expiration before testing. 4
• Oral contraceptives reduce testosterone by 72%, androstenedione by 68.5%, and DHEA-S by 41%, normalizing DHEA-S in all patients with pretreatment elevation. 5

When to Suspect Androgen-Secreting Tumors

Rapid-onset virilization (clitoromegaly, voice deepening) or markedly elevated androgens warrant urgent imaging, but routine screening with testosterone and DHEA-S has poor positive predictive value. 6

• Total testosterone >8.7 nmol/L (>250 ng/dL) has 100% sensitivity but only 9% positive predictive value for ovarian tumors due to the rarity of these neoplasms (2.3% of hyperandrogenic women). 6
• DHEA-S >16.3 μmol/L (>6000 ng/mL) suggests adrenal source, but in one population study of 478 hyperandrogenic women, no adrenocortical tumors were identified despite 10 patients exceeding this threshold. 6
• Clinical evaluation (rapid progression, severe virilization) is more cost-effective than biochemical screening for identifying the rare patient with an androgen-secreting tumor. 6

Comprehensive Metabolic and Hormonal Workup

All women evaluated for hirsutism require exclusion of alternative diagnoses and metabolic screening, regardless of androgen levels. 1, 4

• TSH to exclude thyroid disease. 1, 4
• Morning prolactin to rule out hyperprolactinemia (3.15-fold increased risk in PCOS). 4
• 2-hour oral glucose tolerance test (75-gram load) to screen for diabetes and insulin resistance. 1, 4
• Fasting lipid panel to assess cardiovascular risk. 1, 4
• Consider 17-hydroxyprogesterone if non-classic congenital adrenal hyperplasia is suspected, though ACTH stimulation testing is more sensitive than basal steroid levels. 7, 8

Practical Algorithm

1. Obtain morning fasting sample for total testosterone (LC-MS/MS), SHBG, TSH, prolactin, fasting glucose, 2-hour OGTT, and lipid panel. 1
2. Calculate free testosterone or FAI from TT and SHBG. 1
3. If TT and cFT/FAI are elevated: hyperandrogenism confirmed; proceed to determine etiology (PCOS accounts for 95% of cases). 1
4. If TT and cFT/FAI are normal but clinical features are strong: measure androstenedione and DHEA-S (age-adjusted). 1
5. If rapid virilization or TT >8.7 nmol/L or DHEA-S >16.3 μmol/L: obtain pelvic ultrasound and/or adrenal CT to exclude tumor. 1, 6