Oral Glucose Tolerance Test (OGTT): Drugs That Distort This Test

What the OGTT Measures and Why Drug Interference Matters

The OGTT tracks how quickly your body clears a standardized glucose load from the bloodstream over one to three hours. A result that crosses a diagnostic threshold carries real clinical weight: it can label a patient with gestational diabetes mellitus (GDM), prediabetes, or type 2 diabetes, triggering insulin therapy, dietary restriction, or lifelong monitoring.

Drug interference is not rare. A 2019 analysis published in Diabetes Care noted that medication-induced glucose dysregulation accounts for a meaningful proportion of secondary hyperglycemia diagnoses in hospitalized patients, with corticosteroids implicated in the majority of drug-induced cases [1]. When a drug is on board at the time of testing, the result may reflect the drug's pharmacology more than the patient's true metabolic status.

How Drugs Shift Glucose During an OGTT

Drugs alter OGTT values through several mechanisms:

• Reduced insulin secretion. Thiazide diuretics and diazoxide suppress beta-cell output directly, raising post-load glucose [2].
• Increased hepatic glucose output. Corticosteroids stimulate gluconeogenesis and blunt peripheral insulin signaling [3].
• Enhanced insulin sensitivity. Metformin and thiazolidinediones lower post-load values, potentially masking impaired glucose tolerance.
• Altered gastrointestinal absorption. GLP-1 receptor agonists slow gastric emptying, which shifts the timing of the glucose peak rather than simply lowering the 2-hour value [4].
• Sympathomimetic stimulation. Epinephrine and related agents promote glycogenolysis, acutely raising plasma glucose [5].

The Diagnostic Stakes

The American Diabetes Association's 2024 Standards of Medical Care define a 2-hour plasma glucose of 140–199 mg/dL as impaired glucose tolerance (prediabetes) and ≥200 mg/dL as diagnostic of diabetes [6]. A single drug pushing a patient's result from 135 to 145 mg/dL places them in a fundamentally different clinical category with different treatment and surveillance expectations. That 10 mg/dL difference from a medication effect is clinically actionable information that the ordering clinician must account for.

Corticosteroids: The Highest-Impact Offender

Corticosteroids are the single drug class most consistently documented to raise OGTT values above diagnostic thresholds. Prednisone, dexamethasone, methylprednisolone, and inhaled fluticasone at high doses all drive glucose up, though oral systemic agents cause the most pronounced distortion [3].

Mechanism and Magnitude

Cortisol and synthetic glucocorticoids increase hepatic gluconeogenesis, reduce GLUT-4 translocation in skeletal muscle, and promote adipose lipolysis. The net effect is post-prandial and post-load hyperglycemia that can exceed fasting hyperglycemia. A patient on prednisone 20 mg/day may have a normal fasting glucose of 95 mg/dL but a 2-hour OGTT value above 200 mg/dL due to the afternoon glucocorticoid surge [3].

A retrospective cohort study of 2,231 patients on systemic glucocorticoids found that steroid-induced hyperglycemia occurred in roughly 32% of non-diabetic patients receiving short courses of prednisone ≥20 mg/day [7]. That prevalence makes pre-test disclosure of steroid use essential.

Clinical Recommendation

If a patient is on a systemic corticosteroid for an acute indication (e.g., a 5-day prednisone burst for asthma), the OGTT should be postponed until at least two weeks after the course ends. Patients on chronic low-dose steroids (≤5 mg prednisone/day) may still be tested, but the result must be interpreted alongside the steroid dose and duration. Inhaled corticosteroids at standard doses (e.g., fluticasone 250 mcg/day) produce minimal OGTT distortion and do not typically require a wash-out [8].

Thiazide Diuretics and Loop Diuretics

Hydrochlorothiazide (HCTZ) and chlorthalidone raise plasma glucose by inhibiting pancreatic beta-cell potassium channels via hypokalemia, which reduces calcium-dependent insulin secretion [2]. The effect is dose-dependent.

Evidence From Clinical Trials

The ALLHAT trial (N=33,357) found that patients randomized to chlorthalidone had a statistically higher rate of new-onset diabetes at 4 years compared with those on amlodipine or lisinopril (11.6% vs. 9.8% and 8.1%, respectively) [9]. The glucose elevation in ALLHAT was modest on average but pushed borderline patients across the ADA diagnostic threshold. Loop diuretics such as furosemide cause less glucose distortion than thiazides but can still contribute when hypokalemia is severe (serum K <3.0 mEq/L).

Testing Guidance

A patient on HCTZ 25 mg/day should have their serum potassium corrected to ≥3.5 mEq/L before an OGTT is performed. The 2023 AACE/ACE Diabetes Consensus Statement recommends noting diuretic use and potassium status in the lab order to allow correct interpretation [10].

Atypical Antipsychotics

Second-generation antipsychotics (SGAs), including olanzapine, clozapine, quetiapine, and risperidone, produce weight gain and insulin resistance that distorts OGTT results even after short-term use [11].

Mechanism and Ranking by Risk

Olanzapine and clozapine carry the highest metabolic burden, with studies showing that olanzapine can raise fasting glucose by 8–15 mg/dL within 8 weeks of initiation [11]. Risperidone and quetiapine carry intermediate risk; aripiprazole and ziprasidone carry the lowest risk in this class.

A systematic review and meta-analysis of 48 randomized controlled trials (N=17,498) published in JAMA Psychiatry found that olanzapine increased fasting glucose by a mean of 8.2 mg/dL compared with placebo (P<0.001) [12]. In a patient whose baseline 2-hour OGTT value sits near 130 mg/dL, that shift could produce a false-positive prediabetes diagnosis.

What to Do

The FDA label for olanzapine includes a specific warning about hyperglycemia and diabetes mellitus. Clinicians ordering an OGTT for a patient on an SGA should document the agent, dose, and duration in the lab requisition. Switching to a lower-metabolic-risk agent before testing is sometimes possible after psychiatric consultation, but should never be done without the prescribing psychiatrist's involvement.

Beta-Blockers: The Hidden Masking Problem

Beta-blockers do not reliably raise post-load glucose, but they alter OGTT interpretation in a different and dangerous way: they mask the adrenergic warning symptoms of hypoglycemia (tachycardia, tremor, anxiety) that would otherwise alert a patient or technician to a low glucose reading during the fasting period [5].

Which Agents Cause Glucose Distortion

Non-selective beta-blockers (propranolol, nadolol) also inhibit glycogenolysis and gluconeogenesis, which can lower glucose during the fasting window and produce a falsely low fasting value. Cardioselective agents (metoprolol, atenolol) cause less metabolic interference but are not entirely neutral [5].

In a crossover study of 12 healthy volunteers, propranolol 80 mg twice daily reduced the mean 2-hour OGTT glucose by 18 mg/dL compared with placebo, an effect large enough to reclassify one participant from prediabetes to normal [13].

Immunosuppressants: Tacrolimus and Cyclosporine

Post-transplant diabetes mellitus (PTDM) is defined by an OGTT performed at least 45 days after transplant, per International Consensus Guidelines [14]. Both tacrolimus and cyclosporine directly suppress insulin secretion by inhibiting calcineurin in pancreatic beta cells. Tacrolimus carries a higher diabetogenic risk than cyclosporine at equivalent immunosuppressive doses.

Sirolimus (rapamycin) impairs insulin signaling at the mTOR pathway level and can raise 2-hour OGTT values substantially. An OGTT performed on a patient within 30 days of tacrolimus initiation may overestimate diabetic risk because the drug effect has not yet reached steady state [14].

Medications That Lower OGTT Values (False-Negative Risk)

Some drugs push results in the opposite direction, producing false-normal or falsely low readings that miss true prediabetes or GDM.

Metformin and Thiazolidinediones

Metformin reduces hepatic glucose output and improves insulin sensitivity. A patient taking metformin 1,000 mg twice daily for polycystic ovary syndrome (PCOS) will have a meaningfully lower 2-hour OGTT value than their unmedicated baseline. The DPP (Diabetes Prevention Program, N=3,234) showed that metformin reduced 2-hour glucose by approximately 20 mg/dL in participants with impaired glucose tolerance [15]. Testing a patient on metformin without disclosing the medication could lead to a false-negative for GDM or prediabetes.

GLP-1 Receptor Agonists

Semaglutide, liraglutide, and tirzepatide all slow gastric emptying significantly. During an OGTT, delayed gastric emptying shifts the glucose absorption curve, blunting the 2-hour peak and potentially pushing a result below the 140 mg/dL threshold even in a patient with genuine impaired glucose tolerance [4]. The ADA 2024 Standards explicitly note that interpretation of OGTT results requires consideration of GLP-1 RA use [6].

High-Dose Salicylates

Aspirin at doses above 3 g/day (used historically for rheumatic conditions, less common now) stimulates insulin secretion and enhances glucose disposal, lowering OGTT values by 10–30 mg/dL [16]. Standard low-dose aspirin (81 mg/day) does not produce clinically meaningful OGTT distortion.

Insulin Secretagogues

Sulfonylureas (glipizide, glyburide, glimepiride) and meglitinides (repaglinide, nateglinide) stimulate insulin release independent of glucose concentration. If taken before an OGTT, they can drive a reactive hypoglycemia pattern that lowers the 2-hour reading below the patient's true metabolic state.

Oral Contraceptives, Progestins, and Hormone Therapy

Estrogen-progestin combinations modestly raise post-load glucose through progestin-mediated insulin resistance. The effect varies substantially by progestin type: levonorgestrel and norethindrone show more glucose-raising activity than drospirenone or desogestrel [17].

The table below summarizes the direction and approximate magnitude of effect for commonly used hormonal agents:

| Agent | OGTT Direction | Approximate 2-hr Shift | Evidence Level | |---|---|---|---| | Prednisone ≥20 mg/day | Raises | +30 to +80 mg/dL | High (RCTs, cohorts) | | HCTZ 25 mg/day | Raises | +10 to +20 mg/dL | High (ALLHAT) | | Olanzapine | Raises | +8 to +25 mg/dL | High (meta-analysis) | | Tacrolimus | Raises | +15 to +40 mg/dL | Moderate | | Propranolol | Lowers fasting | -10 to -20 mg/dL | Low-moderate | | Metformin 2 g/day | Lowers | -15 to -25 mg/dL | High (DPP) | | Semaglutide | Lowers peak | variable | Moderate | | OCP (levonorgestrel) | Raises | +5 to +15 mg/dL | Moderate | | High-dose ASA (≥3 g/day) | Lowers | -10 to -30 mg/dL | Low |

A 2021 systematic review in Obstetrics and Gynecology concluded that combined oral contraceptives containing third-generation progestins produce significantly less glucose intolerance than older formulations, though the OGTT is still mildly affected in susceptible individuals [17].

Stimulants, Vasopressors, and Acute-Care Agents

Epinephrine (adrenaline), norepinephrine, and high-dose dopamine stimulate glycogenolysis through beta-2 and alpha-adrenergic receptors, producing acute hyperglycemia. An OGTT performed within 48 hours of vasopressor use in an ICU patient is essentially uninterpretable for outpatient diagnostic purposes [5].

Pseudoephedrine and other over-the-counter sympathomimetics can also raise glucose modestly, though the clinical magnitude is generally smaller than prescription sympathomimetics.

How to Prepare for a Valid OGTT

Standard Preparation Protocol

The ADA 2024 Standards and ACOG Practice Bulletin No. 190 specify the following requirements for a valid OGTT [6][18]:

1. Eat at least 150 g of carbohydrate daily for three days before the test (to avoid carbohydrate restriction-induced glucose intolerance).
2. Fast for 8–14 hours the night before (water is allowed).
3. Avoid smoking, vigorous physical activity, and acute illness on the test day.
4. Sit quietly during the test period.
5. Report all medications, supplements, and herbal products to the ordering clinician at least one week before scheduling.

Timing the Test Around Drug Cycles

For drugs with short half-lives (e.g., a 5-day steroid burst, a single dose of a stimulant), a two-week delay is generally adequate. For drugs with long half-lives or prolonged receptor effects (e.g., depot medroxyprogesterone acetate, long-acting antipsychotics), the ordering clinician must decide whether to delay testing, switch agents, or interpret results with explicit caveats in the report.

The Endocrine Society's 2021 Clinical Practice Guideline on Diabetes in Pregnancy states: "Clinicians should document all pharmacological agents with known glycemic effects before interpreting an OGTT and should repeat testing after any suspected drug-induced glucose elevation resolves" [19].

Normal OGTT Range and Diagnostic Thresholds

Non-Pregnant Adults (75 g, 2-Hour Test)

• Normal: 2-hour plasma glucose <140 mg/dL
• Prediabetes (impaired glucose tolerance): 140–199 mg/dL
• Diabetes: ≥200 mg/dL (requires confirmation on a separate day unless symptoms are present) [6]

Gestational Diabetes (Carpenter-Coustan Criteria, 100 g, 3-Hour Test)

GDM is diagnosed when two or more values meet or exceed thresholds [18]:

• Fasting: ≥95 mg/dL
• 1-hour: ≥180 mg/dL
• 2-hour: ≥155 mg/dL
• 3-hour: ≥140 mg/dL

IADPSG Criteria (75 g, 1-Step GDM Screening)

The International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria, which the ADA endorses as an alternative, diagnose GDM when any single value meets or exceeds [6]:

• Fasting: ≥92 mg/dL
• 1-hour: ≥180 mg/dL
• 2-hour: ≥153 mg/dL

What a High OGTT Means

A 2-hour value of ≥200 mg/dL in a non-pregnant adult indicates diabetes mellitus if confirmed on repeat testing or corroborated by HbA1c ≥6.5% or fasting plasma glucose ≥126 mg/dL [6]. Values of 140–199 mg/dL identify impaired glucose tolerance, which carries a roughly 5–10% annual conversion rate to type 2 diabetes without intervention, as documented in the DPP (N=3,234) [15].

A high value during a GDM screen signals increased risk of macrosomia, neonatal hypoglycemia, preeclampsia, and long-term maternal type 2 diabetes. ACOG Practice Bulletin No. 190 states that "women diagnosed with GDM should receive nutritional counseling from a registered dietitian and begin glucose self-monitoring within one week of diagnosis" [18].

What a Low OGTT Means

A 2-hour OGTT value well below 140 mg/dL in a symptomatic patient, or one who dropped sharply from a high 1-hour value to a low 2-hour value, may suggest reactive hypoglycemia rather than normal glucose homeostasis. Reactive hypoglycemia (post-challenge glucose <70 mg/dL at 2 or 3 hours) can indicate early insulin hypersecretion, which itself is a marker of beta-cell stress [20].

Drug-induced low values (from metformin, sulfonylureas, or GLP-1 RAs taken before the test) must be excluded before attributing reactive hypoglycemia to an intrinsic metabolic problem.