Oral Glucose Suppression Test for Growth Hormone

The oral glucose suppression test is the confirmatory test for acromegaly and pituitary gigantism.

In healthy people a glucose load suppresses growth hormone (GH) secretion to very low levels, because hyperglycaemia inhibits GH release from the pituitary through hypothalamic somatostatin. In acromegaly, autonomous secretion from a somatotroph adenoma is not suppressed, and GH may even rise paradoxically.

The test is done when serum IGF-1 is raised or equivocal, or when a random GH is not conclusive, because GH is pulsatile and a single random value cannot diagnose or exclude acromegaly.

Acromegaly has a prevalence of 40 to 125 in every million people and an incidence of 3 to 4 in every million each year, with a mean delay of 5 to 10 years between onset and diagnosis. Untreated disease doubles mortality through cardiovascular, respiratory and neoplastic complications. Over 95% of cases arise from a pituitary adenoma. The test also assesses disease control after surgery or medical therapy, where GH suppression to under 1 microg/L with a normal IGF-1 defines remission.

Indications. Clinical features of acromegaly (enlarging hands and feet, coarsening of features, jaw prognathism, sweating, carpal tunnel syndrome, snoring, new diabetes or hypertension) with raised or borderline IGF-1. Pituitary mass with raised IGF-1. Assessment of cure 12 weeks or more after pituitary surgery. Not needed when IGF-1 is clearly raised with typical features, and not suitable as the first test.

Contraindications and cautions. Diabetes mellitus with random glucose over 11 mmol/L (the glucose load is unsafe and GH is not suppressed normally in uncontrolled diabetes). Recent gastric surgery with dumping. Severe illness. Pregnancy interpretation is altered by placental GH.

Patient preparation. Fasting from midnight (8 to 10 hours), water allowed. The patient rests for 30 minutes before the baseline sample. No exercise, stress or smoking. Drugs that affect GH are reviewed (oestrogens, levodopa, dopamine agonists, somatostatin analogues and pegvisomant interfere with interpretation, and dose timing is planned with the endocrine team).

Procedure.

• Insert a cannula and take a baseline sample for GH and glucose at 0 minutes.
• Give 75 g anhydrous glucose in 250 to 300 mL water (or 1.75 g/kg to a maximum of 75 g in children), drunk within 5 minutes.
• Take samples for GH and glucose at 30, 60, 90 and 120 minutes.
• The samples are separated and frozen if not analysed on the day. GH is measured by a sensitive two site immunoassay (functional sensitivity under 0.05 microg/L), and assays are calibrated to the recombinant 22 kDa reference preparation.

Normal response. GH falls to under 1 microg/L (under 0.4 microg/L with sensitive assays) at one or more time points, most often at 60 to 120 minutes. A normal result with a normal IGF-1 excludes acromegaly.

Acromegaly.

• GH nadir over 1 microg/L (over 0.4 microg/L by newer, more sensitive assays) after glucose, with raised IGF-1: confirms acromegaly.
• Paradoxical rise of GH in 20 to 30% of patients.
• Lack of suppression occurs also in 10 to 20% of patients with diabetes, adolescents (during puberty), liver or renal failure, anorexia nervosa, malnutrition, heroin use and pregnancy, so the result is read with IGF-1.

Post-treatment assessment.

• Remission: GH nadir under 1 microg/L (or under 0.4 microg/L with sensitive assays) and age- and sex matched normal IGF-1, measured at least 12 weeks after surgery.
• A GH under the cut off with persistently raised IGF-1 suggests discordance, and the test is repeated.
• Discordance between GH and IGF-1 occurs in up to 30% of treated patients, and the clinical picture and tumour imaging decide.

Other uses. Pituitary gigantism in children with accelerated growth. GH excess from ectopic growth hormone releasing hormone production (rare) is separated by measuring serum growth hormone releasing hormone and imaging.

Critical values and findings needing immediate action.

• Confirmed acromegaly with visual field loss, headaches or cranial nerve palsy: urgent pituitary MRI and ophthalmology review.
• Pituitary apoplexy (sudden headache, visual loss, ophthalmoplegia): emergency, with hydrocortisone and neurosurgical review.
• Glucose over 14 mmol/L during the test in a patient with unknown diabetes: stop the test and treat hyperglycaemia.
• Symptomatic hypoglycaemia after the load (late dumping): oral carbohydrate and end of test.

Interferences and pitfalls.

• Assay differences: more sensitive assays give lower nadirs, and the older 1 microg/L cut off misclassifies some active disease. The cut off for the assay in use is applied.
• IGF-1 is the first line test and the glucose suppression test is confirmatory. IGF-1 needs age- and sex specific reference ranges, and is low in malnutrition, liver disease, hypothyroidism and uncontrolled diabetes, and raised in pregnancy.
• Non-suppression with normal IGF-1 is of uncertain meaning and the test is repeated.
• Delayed glucose absorption or vomiting invalidates the test.
• Bromocriptine, cabergoline, octreotide, lanreotide, pasireotide and pegvisomant affect GH and IGF-1 and are stopped for an appropriate washout before testing for cure, if clinically safe.
• Pegvisomant raises GH and lowers IGF-1, so only IGF-1 is used to monitor it.
• Diabetes mellitus: the test is not done if glucose is uncontrolled, and the interpretation uses IGF-1.

Special groups.

• Children and adolescents: GH is higher in puberty, and the test is done by a paediatric endocrinologist.
• Pregnancy: placental GH variant raises IGF-1 and alters results, and the test is avoided.
• Diabetes: glucose tolerance may be impaired, and GH may be paradoxically high even in the absence of acromegaly.
• Regions: late diagnosis is common where access to MRI is limited.

Follow up tests. IGF-1, prolactin, TSH and free T4, morning cortisol, LH, FSH and testosterone or oestradiol, pituitary MRI, visual fields, glucose and HbA1c, lipid profile, echocardiography, sleep study for sleep apnoea, colonoscopy for polyps, and bone densitometry.