Growth Hormone Stimulation Test: What It Actually Measures

Why Random GH Levels Are Clinically Useless

A single blood draw for GH tells you almost nothing. GH is secreted in pulsatile bursts, primarily during deep sleep, exercise, and fasting. Between those bursts, circulating GH can drop to undetectable levels even in completely healthy individuals. A random GH of 0.1 ng/mL at 2 p.m. Might look alarming on paper, but it is physiologically normal.

The Pulsatile Secretion Problem

The anterior pituitary releases GH in 6 to 12 discrete pulses per 24 hours, with the largest bursts occurring during slow-wave sleep [1]. Peak amplitude varies by age, sex, and body composition. In a healthy young male, nocturnal GH peaks can reach 20 to 30 ng/mL, while daytime troughs sit below 0.2 ng/mL [2]. This 100-fold variation within a single day makes any isolated measurement unreliable.

Why Stimulation Solves the Problem

The GH stimulation test bypasses pulsatility by pharmacologically forcing the pituitary to release its GH stores. If the gland can respond with a strong peak, the hypothalamic-pituitary-somatotroph axis is intact. If it cannot, deficiency is confirmed. The Endocrine Society's 2011 clinical practice guideline calls provocative testing "mandatory" for confirming adult GH deficiency (AGHD) in most clinical scenarios [3].

What the Test Actually Measures

The GH stimulation test quantifies the peak serum GH concentration achievable when the pituitary is maximally provoked. It is a measure of somatotroph reserve capacity, not of daily GH production or tissue-level GH action.

Somatotroph Reserve vs. Daily Output

Think of it like a cardiac stress test for the pituitary. Resting cardiac output does not predict maximal cardiac output under dobutamine. GH stimulation works the same way. The pituitary may produce adequate baseline GH under normal daily conditions but fail under pharmacological stress, revealing early or partial deficiency. Conversely, a strong stimulated peak confirms that the machinery for GH synthesis, storage, and secretion is functional.

What It Does Not Measure

The test does not measure GH bioactivity, receptor sensitivity, or downstream IGF-1 generation. A patient could pass a stim test with a peak of 12 ng/mL yet still have GH insensitivity (Laron syndrome) or hepatic IGF-1 generation failure. These are rare but clinically distinct entities. The stim test answers one question: can the pituitary release GH when asked?

The Provocative Agents: How Each One Works

Four main stimulation agents are used clinically. Each triggers GH release through a different mechanism, and the diagnostic cutoff values differ by agent.

Insulin Tolerance Test (ITT)

The ITT remains the reference standard per the Endocrine Society and AACE guidelines [3][4]. Regular insulin (0.1 to 0.15 U/kg IV) induces hypoglycemia (blood glucose must fall below 40 mg/dL), which triggers a counter-regulatory GH surge. Blood is drawn at 0, 30, 45, 60, and 90 minutes.

A peak GH below 5 ng/mL confirms severe AGHD. Between 5 and 10 ng/mL is a gray zone requiring clinical correlation. The ITT simultaneously tests the entire hypothalamic-pituitary axis, including cortisol response, making it efficient. The drawback is real: symptomatic hypoglycemia with sweating, tremor, and rare seizure risk. It is contraindicated in patients with coronary artery disease, seizure disorders, or age over 65 [3].

Glucagon Stimulation Test (GST)

Glucagon (1 mg IM for patients under 90 kg, 1.5 mg for those above) triggers GH release through a mechanism that is still not fully characterized but likely involves secondary hypoglycemia and direct hypothalamic stimulation. Sampling occurs at 0, 60, 90, 120, 150, 180, and 210 minutes.

The Endocrine Society guideline endorses GST as the primary alternative to ITT, using a peak GH cutoff of 3 ng/mL [3]. A 2015 meta-analysis (N=314 across 8 studies) reported 97% sensitivity and 88% specificity at that threshold [5]. Nausea occurs in roughly 30% of patients, but serious adverse events are rare.

Macimorelin Oral Test

Macimorelin (Macrilen) earned FDA approval in December 2017 as the first oral GH stimulation agent [6]. It is a synthetic ghrelin-receptor agonist dosed at 0.5 mg/kg dissolved in water. Blood draws occur at 30, 45, 60, and 90 minutes after ingestion.

The key trial (N=157 AGHD patients, N=99 matched controls) showed 92% sensitivity and 96% specificity using a peak GH cutoff of 2.8 ng/mL [7]. Its advantages are significant: no IV access required, no hypoglycemia, and completion in 90 minutes versus 2 to 4 hours for ITT or GST. The Endocrine Society's 2019 update added macimorelin as an acceptable alternative to ITT [8].

GHRH-Arginine Test

This combination test pairs growth hormone-releasing hormone (GHRH, 1 mcg/kg IV) with arginine (0.5 g/kg IV over 30 min). It was widely considered the most reliable and reproducible test until GHRH (Geref) was discontinued in the U.S. Market. The test is still used in Europe and select research centers. BMI-adjusted cutoffs apply: peak GH below 11.5 ng/mL for BMI <25, below 8.0 ng/mL for BMI 25 to 30, and below 4.2 ng/mL for BMI above 30 [9].

Normal Ranges and Diagnostic Cutoffs

"Normal" on a GH stimulation test means the peak GH exceeds the agent-specific cutoff. There is no single universal number. The following cutoffs represent current consensus from the Endocrine Society and AACE guidelines [3][4].

Adult Cutoffs by Agent

| Stimulation Agent | Diagnostic Cutoff (Severe GHD) | Assay Note | |---|---|---| | Insulin tolerance test | Peak GH <5 ng/mL | Requires glucose <40 mg/dL | | Glucagon stimulation test | Peak GH <3 ng/mL | BMI <30 preferred | | Macimorelin | Peak GH <2.8 ng/mL | FDA-approved cutoff | | GHRH-arginine | BMI-adjusted (4.2 to 11.5 ng/mL) | Agent availability limited |

Pediatric Cutoffs

For children, the Pediatric Endocrine Society and most pediatric guidelines define GHD as a peak GH below 10 ng/mL on two separate provocative tests using different agents [10]. A single failed test is insufficient for diagnosis due to the 20 to 25% false-positive rate inherent to individual provocative tests in children. Priming with sex steroids (estradiol or testosterone) before testing is recommended for prepubertal children aged 9 and older to reduce false positives.

What a Low Result Means

A peak GH below the diagnostic cutoff confirms growth hormone deficiency. The clinical significance depends heavily on the patient's age, symptom burden, and number of other pituitary hormone deficiencies.

In Adults

Adult GH deficiency affects an estimated 1 to 3 per 10,000 adults, though it is significantly more common in patients with known pituitary disease [3]. Causes include pituitary tumors (most common), pituitary surgery, cranial radiation, traumatic brain injury, and idiopathic GHD carried over from childhood. Symptoms include increased visceral adiposity, reduced lean mass, fatigue, impaired quality of life, and adverse lipid profiles.

The 2011 Endocrine Society guideline states: "GH replacement therapy should be considered in adults with clinical features suggesting GHD who have biochemically confirmed deficiency" [3]. Replacement doses typically start at 0.1 to 0.2 mg/day of recombinant human GH, titrated to normalize IGF-1 levels. Women on oral estrogen require higher doses due to first-pass hepatic antagonism of GH signaling.

In Children

Pediatric GHD presents with short stature, growth velocity below the 25th percentile, delayed bone age, and in severe cases, neonatal hypoglycemia or micropenis in boys. The FDA has approved recombinant GH for pediatric GHD since 1985, and treatment typically continues until near-final height or epiphyseal closure [10].

What a High Result Means

A peak GH above the cutoff is a normal, expected result. It rules out GH deficiency. A very high stimulated peak (above 20 to 30 ng/mL) is not pathological in the context of provocative testing; it simply reflects strong pituitary reserve.

When High Baseline GH Is the Concern

If the clinical question is excess GH production (suspected acromegaly), the stimulation test is the wrong test entirely. Acromegaly is diagnosed with an oral glucose tolerance test (OGTT), where failure of GH to suppress below 1 ng/mL (or 0.4 ng/mL with ultrasensitive assays) after a 75 g glucose load confirms autonomous GH secretion [11]. The Endocrine Society's acromegaly guideline explicitly recommends the OGTT as the confirmatory test, not a stimulation test [11].

Factors That Affect Test Accuracy

Several physiological and pharmacological variables can produce misleading results. Awareness of these confounders is critical for accurate interpretation.

Body Mass Index

Obesity is the single largest confounder. Adiposity blunts GH secretion across all provocative agents. A 2008 study (N=224) demonstrated that obese subjects (BMI above 30) had mean peak GH values 60 to 70% lower than lean controls on ITT [12]. This means a BMI of 35 could push a peak GH from 7 ng/mL (passing) down to 3 ng/mL (failing), producing a false-positive diagnosis of GHD. The GHRH-arginine test addressed this with BMI-stratified cutoffs, but other agents lack validated obese-specific thresholds.

Medications

Glucocorticoids suppress GH secretion dose-dependently. Patients on prednisone equivalents above 5 mg/day may produce falsely low peaks. Estrogen (particularly oral formulations) lowers IGF-1 without affecting pituitary GH secretion, complicating the screening-to-confirmation pathway. Hypothyroidism should be corrected before testing, as low T4 impairs GH secretion [3].

Age and Sex

GH secretion declines with age at approximately 14% per decade after age 30, a phenomenon called somatopause [2]. The diagnostic cutoffs in current guidelines do not adjust for age, meaning a 70-year-old tested with an ITT may "fail" the 5 ng/mL cutoff due to physiological decline rather than pathological deficiency. Sex steroid status also matters: testosterone priming in prepubertal boys and estradiol priming in prepubertal girls reduces the rate of false-positive GHD diagnoses from roughly 25% to under 5% [10].

Assay Variability

GH immunoassay results can vary by 20 to 40% between different commercial platforms due to antibody specificity, calibration standards, and recognition of GH isoforms [13]. The WHO International Standard 98/574 for recombinant 22-kDa GH has improved but not eliminated this variability. A peak of 4.8 ng/mL on one assay might read as 5.5 ng/mL on another. Clinicians should interpret borderline results in the context of the specific assay platform used.

How to Prepare for the Test

Preparation is straightforward but strict. Noncompliance with fasting or medication instructions is a common cause of indeterminate results.

Pre-Test Checklist

Patients must fast for 8 to 12 hours before the test. Only water is permitted. Exercise should be avoided the morning of testing, as even moderate physical activity can raise baseline GH and confuse interpretation. Medications that affect GH secretion (glucocorticoids, GH therapy itself, dopamine agonists) should be held per the ordering endocrinologist's instructions. Hypothyroidism must be treated to euthyroid status before testing [3].

During the Test

An IV catheter is placed, and baseline blood is drawn. The provocative agent is administered, and serial blood samples are collected over 90 to 210 minutes depending on the agent. Patients undergoing ITT must be monitored continuously with glucose checks, and 50% dextrose must be available at bedside. A physician must be present during ITT. Glucagon and macimorelin tests are lower risk and can be supervised by trained nursing staff in most endocrine testing centers [4].

The Screening Step Before Stimulation Testing

Not every patient with fatigue or poor body composition needs a GH stim test. The Endocrine Society guideline recommends measuring serum IGF-1 as an initial screen [3]. IGF-1 reflects integrated 24-hour GH secretion because the liver produces IGF-1 in proportion to GH exposure, and IGF-1 has a half-life of 12 to 16 hours (versus 15 to 20 minutes for GH itself).

When IGF-1 Alone Confirms GHD

In patients with three or more other pituitary hormone deficiencies plus a low IGF-1, the probability of GHD exceeds 95%, and some guidelines accept this combination without provocative testing [4]. For patients with isolated suspected GHD or only one to two other deficiencies, stimulation testing remains required.

IGF-1 Pitfalls

IGF-1 can be normal in up to 35% of adults with confirmed GHD on stimulation testing [3]. Malnutrition, liver disease, uncontrolled diabetes, and hypothyroidism all lower IGF-1 independent of GH status. A normal IGF-1 does not rule out GHD. A low IGF-1 in the right clinical context (prior pituitary surgery, radiation, TBI) is strongly suggestive and warrants confirmatory stimulation testing.

How Results Guide Treatment Decisions

A confirmed GH deficiency on stimulation testing opens the door to recombinant human GH (rhGH) replacement, but the decision to treat involves more than a single lab number.

Adult Treatment Initiation

The Endocrine Society recommends starting rhGH at 0.1 to 0.2 mg/day in men and 0.3 mg/day in women (higher due to oral estrogen effects), titrating every 4 to 8 weeks based on IGF-1 levels, clinical response, and side effects [3]. The target is an IGF-1 in the upper half of the age-adjusted reference range. A 2-year open-label extension of the HypoCCS database (N=1,988) showed improvements in body composition, lipid profiles, and quality of life scores with rhGH replacement [14].

Retesting and Transition

Children diagnosed with GHD require retesting after attaining final height. The transition period (ages 15 to 25) uses adult diagnostic criteria. Approximately 40% of children diagnosed with idiopathic isolated GHD will pass adult retesting, indicating their childhood GHD was transient [10]. Those who continue to fail require adult-dose GH replacement, typically at lower doses than pediatric treatment.

Patients with structural pituitary disease (tumors, surgery, radiation) and two or more additional pituitary deficiencies generally do not require retesting, as the pretest probability of persistent GHD exceeds 95% [3].