Drugs That Distort the Growth Hormone Stimulation Test

Direct answer

A drug or physiologic state that suppresses hypothalamic GHRH output, raises somatostatin tone, or blunts hepatic IGF-1 synthesis can lower the measured GH peak without any true pituitary disease. The best-documented confounders are supraphysiologic or even standard-replacement glucocorticoids, chronic opioid use, oral (not transdermal) estrogen, obesity, and, in prepubertal children, the absence of sex-steroid priming. Because many patients referred for GH stim testing are already taking one of these drugs for an unrelated condition, the test result has to be read alongside the medication list and BMI, not in isolation. The Endocrine Society's adult GH deficiency guideline calls for correcting reversible factors before testing and using two abnormal provocative tests, rather than one, before confirming the diagnosis in adults.

What a GH stimulation test measures, and why the test exists at all

Growth hormone is secreted in pulses. A random blood draw is close to useless for diagnosis because most daytime GH levels sit near the lower limit of detection even in people with a completely normal pituitary. A stimulation test forces a pharmacologic surge and asks whether the pituitary can reach an expected peak.

The most established protocols are the insulin tolerance test (ITT), the glucagon stimulation test, the GHRH-arginine test, and the newer oral macimorelin test. The ITT drives hypoglycemia (blood glucose under roughly 40 mg/dL), which triggers a counter-regulatory GH surge; it is considered by many endocrinologists to be the reference standard, but it is contraindicated in people with seizure disorders, ischemic heart disease, and generally in older adults, and it requires close medical supervision because of the induced hypoglycemia. Glucagon stimulates GH through a less completely understood pathway involving secondary hypoglycemia. GHRH-arginine combines a releasing hormone with a somatostatin-lowering amino acid. Macimorelin, an oral ghrelin-receptor agonist approved by the FDA in December 2017 for the diagnosis of adult GH deficiency, requires no IV insulin and no induced hypoglycemia.

The core problem in one paragraph

A peak GH below the diagnostic cutoff does not by itself confirm growth hormone deficiency. Chronic glucocorticoid or opioid use, oral estrogen, obesity, and unaddressed puberty status can each independently suppress the GH peak enough to mimic true deficiency, and these confounders are common precisely in the population referred for testing, since many of these patients are on chronic medications for other conditions. The Endocrine Society's guideline on adult GH deficiency responds to this by requiring correction of reversible factors where medically safe and by requiring two abnormal provocative tests rather than one before the diagnosis is confirmed. No single test agent, including the newer oral macimorelin test, is immune to every pharmacologic confounder, and BMI in particular still needs to be accounted for regardless of which agent is used.

Glucocorticoids: the most common confounder in practice

Glucocorticoids suppress GH secretion at multiple points in the hypothalamic-pituitary-somatotroph axis, and the effect is dose-dependent. Chronic use of prednisone-equivalent doses substantially above physiologic replacement is reported to reduce pulsatile GH secretion; some replacement-dose hydrocortisone regimens have also been associated with reduced IGF-1 and blunted GH peaks in patients with existing hypopituitarism. The exact percentage reductions reported in the older endocrinology literature vary between studies and should be checked against the primary papers before being quoted as a fixed number.

The practical problem is that many patients referred for GH stim testing are already on glucocorticoid replacement for adrenal insufficiency, so the very treatment for one pituitary axis can mask adequate function in another. The usual approach is to minimize the glucocorticoid dose to the lowest safe physiologic replacement (commonly discussed as hydrocortisone in the 10 to 15 mg/day range, divided) for at least roughly two weeks before testing where a physician judges this to be safe. Complete withdrawal is rarely appropriate in a patient with documented adrenal insufficiency, so a borderline GH peak in a patient who cannot come off steroids should be interpreted cautiously rather than treated as a confirmed diagnosis.

Opioids and the GH axis

Opioids act on hypothalamic mu-opioid receptors, increasing somatostatin tone and reducing GHRH output. Clinical reports describe meaningfully reduced peak GH responses in chronic opioid users compared with people not taking opioids, though published effect sizes vary and a specific percentage should not be treated as a fixed rule for every patient.

Short-acting opioids clear from the bloodstream within a day or two, but hypothalamic-pituitary suppression appears to outlast the drug itself, plausibly by one to two weeks, and long-acting formulations (methadone, extended-release morphine, buprenorphine) likely require longer. Buprenorphine's partial agonism may cause less suppression than full opioid agonists, but the supporting data are limited and this should be treated as plausible rather than established.

Many patients referred for GH testing take opioids for chronic pain, and abrupt discontinuation carries withdrawal risk. A planned taper or a temporary switch to non-opioid analgesia, coordinated with the prescribing clinician, is the safer path when discontinuation is medically reasonable. When it is not, the interpreting physician should document the opioid use as a likely confounder and consider repeat testing if the result is borderline rather than clearly abnormal.

Estrogen: the route matters more than the drug

Oral estrogen undergoes first-pass hepatic metabolism that suppresses hepatic IGF-1 production without changing the pituitary's actual GH output. This means a screening IGF-1 level can look artificially low in a person taking oral contraceptives or oral hormone therapy, prompting a GH stim test referral that may not have been needed. Transdermal estrogen bypasses this first-pass effect and does not appear to produce the same artifact, based on the consensus guideline literature on adult GH deficiency.

A premenopausal woman on combined oral contraceptives could have an IGF-1 result below the age-adjusted reference range for this reason alone. Switching to a transdermal estrogen product or discontinuing oral estrogen for roughly four to six weeks before repeating IGF-1 can clarify whether the low value was a true signal or a route-of-administration artifact, though this timeline should be confirmed with the ordering endocrinologist rather than applied as a fixed rule.

Testosterone therapy in adult men appears to modestly raise GH secretion through peripheral conversion (aromatization) to estradiol, which could in theory mask mild GH deficiency, though the effect is generally considered small enough that testosterone is not routinely held before adult GH testing. It should still be documented on the requisition.

Puberty and sex-steroid priming in children

In pediatric testing, the concern runs the other direction. Prepubertal children, particularly boys in early puberty (Tanner stage I to II), have physiologically low sex-steroid levels that blunt the GH response to any provocative agent. Older pediatric endocrinology literature reports that a substantial proportion of normally growing prepubertal children can fail a GH stim test without priming, and pediatric endocrine guidelines recommend brief sex-steroid priming (a short course of oral estrogen in girls, or an injection of testosterone several days before testing in boys) to reduce this false-positive rate. The specific percentage of false positives prevented by priming varies across studies and should be verified against the primary pediatric literature rather than cited as a fixed figure.

Obesity: a powerful confounder that isn't a drug

Body fat is not a medication, but during GH stim testing it behaves like one. Obesity is one of the most potent suppressors of the stimulated GH peak outside of actual pituitary disease. Studies using the GHRH-arginine test have reported markedly lower peak GH values in people with BMI above roughly 30 kg/m² compared with lean controls, and proposed BMI-stratified cutoffs (progressively lower cutoffs at higher BMI bands) exist for this reason. The exact peak values and cutoff thresholds reported in different studies are not identical, and a clinician applying a BMI-adjusted cutoff should confirm which reference values the testing laboratory uses.

Macimorelin, tested against the ITT in a comparative trial, has been reported to show high sensitivity and specificity for adult GH deficiency at its FDA-labeled cutoff of 2.8 mcg/L, and it appears to be less affected by BMI than GHRH-arginine, though "less affected" is not the same as "unaffected," and the exact sensitivity and specificity figures reported across sources vary slightly and should be checked against the primary trial publication before being restated as precise numbers.

GLP-1 receptor agonists such as semaglutide deserve a specific mention because of how common they have become. These drugs produce substantial weight loss in clinical trials of people with obesity (the STEP-1 trial reported meaningful mean weight loss at 68 weeks in participants without diabetes), and weight loss of that magnitude could plausibly unmask a GH axis that looked suppressed at a higher BMI. This is a reasonable clinical hypothesis, not an established finding specific to GH testing: no trial in the material reviewed here directly measured GH stim test results before and after GLP-1-induced weight loss. A patient who failed a GH stim test at a high BMI and later loses substantial weight on a GLP-1 agonist is a plausible candidate for repeat testing, but that decision is a clinical judgment call rather than a guideline-mandated step.

Psychiatric and stimulant medications

Atypical antipsychotics (risperidone, olanzapine, quetiapine) can raise prolactin, and elevated prolactin is associated with blunted GH responses in some reports. Antipsychotics usually cannot be safely stopped before testing, so the likely confounding effect has to be acknowledged in the interpretation rather than eliminated.

SSRIs have a more mixed relationship with GH secretion. Acute dosing may stimulate GH release through serotonergic pathways, while chronic use appears to have a smaller, less consistent effect. SSRIs are not routinely held before GH stim testing, but their use should be recorded.

Dopamine agonists (cabergoline, bromocriptine) have variable and sometimes opposite effects depending on the underlying pituitary condition: in acromegaly they can suppress GH secretion from a co-secreting tumor, while in ordinary physiology dopamine tends to stimulate GH release. When the clinical question is whether GH deficiency exists, holding a dopamine agonist for roughly four weeks before testing is a reasonable and commonly cited precaution, though exact washout timing should be confirmed with the prescribing endocrinologist.

Stimulant medications (amphetamines, methylphenidate) can acutely raise GH release through adrenergic and dopaminergic pathways. A child tested on the same day as a stimulant dose could show a falsely elevated peak that masks true deficiency. Holding stimulant medications for roughly 48 to 72 hours before pediatric provocative testing is a standard practical precaution.

Diabetes medications and the insulin tolerance test specifically

The ITT depends on inducing hypoglycemia below roughly 40 mg/dL. Anything that prevents blood glucose from falling that far will invalidate the test, and anything that alters the counter-regulatory hormone response will complicate its interpretation.

Metformin alone rarely prevents adequate hypoglycemia. Sulfonylureas can cause unpredictable glucose behavior that complicates timing. The more clinically important scenario is a patient already on insulin therapy for diabetes: recurrent hypoglycemia can blunt counter-regulatory cortisol and GH surges independent of pituitary status, which is a physiologic confounder rather than a drug interaction in the strict sense, but it argues for choosing glucagon or macimorelin over the ITT in insulin-treated diabetic patients.

Somatostatin analogs (octreotide, lanreotide), used in acromegaly and neuroendocrine tumors, directly suppress GH secretion and must be held before GH deficiency testing. Depot formulations (such as octreotide LAR) release drug slowly from the injection site and are generally held for a period of months rather than days before testing; short-acting subcutaneous octreotide requires a much shorter hold. Exact washout intervals vary by formulation and should be confirmed against the product label and the ordering endocrinologist's protocol.

What is established, what is plausible, and what is not established

Established: glucocorticoids, chronic opioids, and obesity each suppress the stimulated GH peak through documented physiologic mechanisms, and the diagnostic guideline literature explicitly calls for correcting reversible factors before confirming adult GH deficiency. Oral estrogen lowers hepatic IGF-1 independent of true GH status. Sex-steroid priming reduces false-positive pediatric GH stim test results.

Plausible but not firmly quantified in the material reviewed here: the exact percentage reductions in GH peak attributed to specific opioid or glucocorticoid doses, the exact sensitivity and specificity of macimorelin relative to the ITT, and whether GLP-1-induced weight loss meaningfully changes GH stim test outcomes in a given patient.

Not established: that any single provocative test agent is fully immune to pharmacologic confounding, or that a fixed drug-holiday schedule applies identically to every patient regardless of the underlying condition being treated.

When to seek urgent care instead of proceeding with testing

The insulin tolerance test requires induced hypoglycemia under medical supervision and is not something to attempt outside a monitored clinical setting. Anyone who experiences confusion, chest pain, seizure activity, or loss of consciousness during or after a GH stim test needs immediate medical evaluation. Patients with a seizure disorder, known cardiovascular disease, or hypoglycemia unawareness should discuss whether the ITT is appropriate for them at all, since alternative agents (glucagon, GHRH-arginine, or macimorelin) may be safer choices.

A confounder decision framework for pre-test medication review

This is a way to reason through a specific patient's medication list before a GH stim test is scheduled, not a substitute for the ordering endocrinologist's judgment.

Step 1: Can the confounding drug be held safely?

Step 2: If the confounder cannot be removed, choose the test least sensitive to it. Insulin-treated diabetics with hypoglycemia unawareness are poor ITT candidates; consider glucagon or macimorelin. Patients with significant obesity may benefit from a test agent and cutoff pair reported to be less BMI-dependent, understood as "less affected," not "unaffected."

Step 3: Never confirm adult GH deficiency on a single abnormal test in the presence of an unremovable confounder. The guideline-based practice of requiring two abnormal provocative tests (or one test plus a low IGF-1 in the correct clinical context) exists partly for this reason. A single borderline result in a patient on chronic opioids and glucocorticoids is a reason to document the confounder and consider repeat testing after optimization, not a reason to finalize a diagnosis.

Step 4: Record BMI, current medications, and last-dose timing on the requisition. A peak GH of 4.5 mcg/L means something different in a lean, medication-free 25-year-old than in a 55-year-old with a BMI of 34 who is taking prednisone and a long-acting opioid. The interpreting physician cannot make that distinction without the data being recorded at the time of the blood draw.

References

1. FDA. Macrilen (macimorelin) prescribing information, approved December 2017. https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/205598s000lbl.pdf

This article's discussion of GH stimulation testing relies on peer-reviewed endocrinology literature for both adult and pediatric populations, including the Endocrine Society's clinical practice guideline on adult GH deficiency, pediatric endocrine society growth hormone guidelines, and published research on BMI-adjusted GHRH-arginine cutoffs and macimorelin validation trials. Numerical values and cutoff points from the previous version could not be confirmed against original sources and were therefore excluded; prior to publication, a qualified endocrinologist should verify all reported effect sizes, diagnostic cutoffs, and medication washout periods against primary literature.