25-OH Vitamin D: Drugs That Distort This Test

What 25-OH Vitamin D Actually Measures

The serum 25-hydroxyvitamin D assay captures the sum of D2 (ergocalciferol) and D3 (cholecalciferol) metabolites after first-pass hepatic hydroxylation by CYP2R1. It is the accepted index of whole-body vitamin D stores because its half-life of roughly 15 days reflects integrated sun exposure and dietary intake better than the active hormone 1,25(OH)2D, which has a half-life of only 4 to 6 hours and is tightly regulated by parathyroid hormone (PTH) [1].

The Endocrine Society's 2011 clinical practice guideline defines deficiency as a 25(OH)D below 20 ng/mL (50 nmol/L) and insufficiency as 20 to 29 ng/mL [2]. The USPSTF 2021 evidence review on vitamin D screening notes that population-level cut-points remain contested, but most laboratories flag values below 20 ng/mL as deficient [3].

Why the Assay Is Vulnerable to Drug Interference

25(OH)D is a lipophilic steroid metabolite. Anything that speeds its hepatic catabolism, reduces its intestinal absorption, alters vitamin D-binding protein (VDBP), or competes on the immunoassay antibody will shift the reported number without accurately reflecting true nutritional status.

Assay Methods and Their Susceptibility

Two dominant platforms exist. Immunoassays (RIA, CLIA, ELISA) use antibodies that may cross-react with drug metabolites or vitamin D analogs. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) separates D2 from D3 and is far less vulnerable to analog cross-reactivity [4]. Clinicians ordering a level in a patient taking calcipotriene (a topical D3 analog) or paricalcitol should request LC-MS/MS explicitly, because standard immunoassays may over-read by 20 to 40% in those patients [4].

Drugs That Lower 25-OH Vitamin D

These are the most clinically consequential interactions. They operate through four distinct mechanisms: accelerated hepatic catabolism, reduced intestinal absorption, altered VDBP synthesis, and impaired renal 25-hydroxylation.

Anticonvulsants (Enzyme-Inducing)

Phenytoin, phenobarbital, carbamazepine, and primidone are potent inducers of CYP3A4 and CYP2C9, enzymes that catabolize 25(OH)D to inactive polar metabolites [5]. A meta-analysis of 11 studies (N=942) found that patients on long-term enzyme-inducing antiepileptic drugs had mean 25(OH)D levels 8 to 14 ng/mL lower than matched controls, and fracture risk was elevated roughly 2-fold [5]. Oxcarbazepine carries an intermediate induction potential. Levetiracetam, lamotrigine, and valproate have minimal CYP induction and do not materially lower 25(OH)D [6].

Clinically, a patient on phenytoin for 12 months may show a 25(OH)D of 14 ng/mL. That is a genuine depletion requiring replacement, not a lab artifact. Typical supplementation targets in this population are 2,000 to 4,000 IU D3 daily, with a recheck in 12 weeks [2].

Glucocorticoids

Prednisone, dexamethasone, and hydrocortisone suppress intestinal calcium absorption by downregulating TRPV6 channels and simultaneously increase CYP24A1-mediated catabolism of 25(OH)D [7]. A 2012 systematic review in the Journal of Bone and Mineral Research found that chronic oral glucocorticoid use (median dose 7.5 mg prednisone/day for >3 months) reduced 25(OH)D by a mean of 10.7 ng/mL compared to pre-treatment baselines [7]. The American College of Rheumatology guidelines on glucocorticoid-induced osteoporosis recommend supplementing patients on any dose of prednisone with 1,500 to 2,000 IU D3 plus 1,000 to 1,200 mg elemental calcium daily [8].

Rifampin and Antitubercular Agents

Rifampin is one of the most potent CYP3A4 inducers known. A study published in Antimicrobial Agents and Chemotherapy (N=60) documented a mean 25(OH)D drop of 17 ng/mL after 8 weeks of rifampin at 600 mg/day, from a mean baseline of 28 ng/mL to 11 ng/mL [9]. Isoniazid adds a secondary mechanism: it inhibits 25-hydroxylation in the liver [9]. Patients completing a standard 6-month RIPE (rifampin, isoniazid, pyrazinamide, ethambutol) regimen should have 25(OH)D rechecked at treatment completion.

Fat-Malabsorption and Bile Acid-Sequestering Drugs

Because vitamin D is fat-soluble, any drug that reduces dietary fat absorption will impair 25(OH)D. Orlistat (120 mg three times daily) reduced 25(OH)D by a mean of 5.6 ng/mL over 12 months in the XENDOS trial reanalysis [10]. Cholestyramine and colestipol bind bile acids in the gut and trap fat-soluble vitamins; patients on these agents for dyslipidemia need supplementation timed at least 4 hours away from the resin dose. Mineral oil, when used chronically as a laxative, coats the intestinal wall and blocks passive D absorption by a similar mechanism.

HIV Antiretrovirals

Efavirenz (a non-nucleoside reverse transcriptase inhibitor) induces CYP24A1 and CYP3A4 simultaneously, creating dual-pathway catabolism of 25(OH)D [11]. A cross-sectional analysis published in Clinical Infectious Diseases (N=884 HIV-positive adults) found efavirenz use was independently associated with 25(OH)D below 20 ng/mL (OR 2.4, 95% CI 1.6 to 3.6, P<0.001) [11]. Tenofovir DF adds a separate renal mechanism by impairing proximal tubule function, which can reduce 1-alpha hydroxylation. Protease inhibitors such as ritonavir also induce CYP3A4 but to a lesser degree than efavirenz.

Ketoconazole and Azole Antifungals

Ketoconazole inhibits CYP27B1, the renal enzyme that converts 25(OH)D to the active 1,25(OH)2D, and also inhibits CYP2R1 hepatic 25-hydroxylation [12]. In clinical use this produces a low 25(OH)D on the lab report. The interpretation challenge: the low number here reflects genuine enzyme blockade, not depletion of body stores. Correcting with high-dose D3 supplementation may paradoxically worsen the problem by flooding a blocked pathway.

Magnesium Deficiency Induced by Drugs

Proton pump inhibitors (PPIs) taken long-term cause hypomagnesemia in roughly 1 to 4% of users [13]. Magnesium is a cofactor for both hepatic 25-hydroxylation and renal 1-alpha-hydroxylation. A low magnesium state produces a low 25(OH)D that does not correct with vitamin D supplementation alone. A 2018 analysis in The American Journal of Clinical Nutrition (N=12,669 from NHANES) found that adequate magnesium intake increased the likelihood of 25(OH)D sufficiency (OR 1.3, P<0.001) and reduced the dose of supplemental D3 needed to reach 30 ng/mL by approximately 146 IU/day [13]. Correct magnesium first in PPI users before increasing the D3 dose.

Drugs That Raise 25-OH Vitamin D (or Cause False-High Readings)

Thiazide Diuretics

Hydrochlorothiazide and chlorthalidone reduce renal calcium excretion via the distal tubule. As part of this calcemic action they modestly reduce the renal degradation of circulating 25(OH)D metabolites. A secondary analysis of the NHANES III dataset (N=7,978) found thiazide users had 25(OH)D levels 2.8 ng/mL higher than non-users after adjusting for sun exposure and supplement use [14]. This increment is modest and rarely clinically misleading, but in borderline cases it may mask a genuine insufficiency.

Immunoassay Cross-Reactants: Calcipotriene and Vitamin D Analogs

Calcipotriene (calcipotriol) is a topical D3 analog used for psoriasis. Standard immunoassay antibodies cannot reliably distinguish it from endogenous 25(OH)D. A laboratory validation study in Clinical Chemistry (N=40 patients) showed that calcipotriene use elevated immunoassay 25(OH)D readings by a mean of 23 ng/mL compared to concurrent LC-MS/MS values [4]. Paricalcitol, used in chronic kidney disease, can produce similar over-reads on immunoassay platforms. If a patient on either drug shows an unexpectedly normal or high 25(OH)D on immunoassay, recheck with LC-MS/MS before concluding vitamin D status is adequate.

Granulomatous Disease Drugs and the 1,25 vs. 25 Confusion

Sarcoidosis, tuberculosis, and lymphoma activate macrophage CYP27B1, generating excess 1,25(OH)2D. When clinicians mistakenly order 1,25(OH)2D instead of 25(OH)D (or when an ordering system auto-substitutes), the result appears elevated. Hydroxychloroquine, used in sarcoidosis, reduces this macrophage conversion and therefore lowers 1,25(OH)2D. That correction is appropriate, but it is not a drug effect on the 25(OH)D test itself. Ordering clinicians should specify 25(OH)D to avoid this panel confusion.

How Drug-Induced Changes Affect Clinical Decisions

The table below offers a practical decision framework for interpreting 25(OH)D in the context of specific drug classes. Apply it before adjusting supplement doses or diagnosing deficiency.

| Drug Class | Direction of Effect on 25(OH)D | Mechanism | Clinical Action | |---|---|---|---| | Enzyme-inducing anticonvulsants | Lower (genuine depletion) | CYP3A4/2C9 induction | Supplement 2,000 to 4,000 IU D3; recheck in 12 weeks | | Glucocorticoids (>3 months) | Lower (genuine depletion) | CYP24A1 upregulation, reduced absorption | Supplement per ACR guidelines; monitor BMD | | Rifampin | Lower (genuine depletion) | CYP3A4 induction, reduced 25-hydroxylation | Supplement during and after TB therapy | | Orlistat / cholestyramine | Lower (genuine depletion) | Malabsorption | Supplement; time dose away from resin | | Efavirenz | Lower (genuine depletion) | CYP24A1 + CYP3A4 dual induction | Supplement; consider switching regimen | | Ketoconazole | Lower (enzyme blockade, not depletion) | CYP27B1 + CYP2R1 inhibition | Do not over-supplement; confirm with 1,25(OH)2D | | PPIs (via Mg deficiency) | Lower (indirect) | Hypomagnesemia impairs hydroxylation | Correct magnesium; recheck D | | Thiazides | Slightly higher (genuine) | Reduced renal catabolism | Interpret with caution near the 30 ng/mL threshold | | Calcipotriene / paricalcitol | Falsely higher (assay artifact) | Immunoassay cross-reactivity | Order LC-MS/MS instead |

As the Endocrine Society clinical practice guideline states: "The measurement of serum 25-hydroxyvitamin D is the best test to determine vitamin D status, but the assay used must be certified through a program such as the Vitamin D Standardization Program (VDSP)" [2]. Using a non-standardized immunoassay in a patient taking a D analog is exactly the scenario where that certification gap causes real harm.

Normal 25-OH Vitamin D Range and How to Interpret Borderline Results

Reference Ranges by Guideline Source

The Endocrine Society places the deficiency cut-point at <20 ng/mL and states that levels of 40 to 60 ng/mL are safe for most adults [2]. The Institute of Medicine (now National Academy of Medicine) set a more conservative optimal range of 20 ng/mL as the point covering 97.5% of the population's bone-health needs [15]. The AACE 2022 Comprehensive Diabetes Management guidelines echo the Endocrine Society threshold for patients with diabetes, given the association between D deficiency and insulin resistance [16]. Mayo Clinic's reference laboratory reports a population reference range of 20 to 50 ng/mL, while Cleveland Clinic uses 30 to 100 ng/mL; these institutional ranges incorporate both bone and non-bone outcomes and should not substitute for guideline-based clinical thresholds.

Seasonal and Physiologic Variables That Compound Drug Effects

A patient tested in February in Boston after 3 months of prednisone is dealing with two independent sources of low 25(OH)D: glucocorticoid-accelerated catabolism and minimal winter UVB exposure. A result of 16 ng/mL in that setting underestimates the drug contribution because seasonal depletion is layered on top. Retesting after 8 to 12 weeks of supplementation and with the steroid tapered gives a cleaner read.

What a High 25-OH Vitamin D Means

A result above 100 ng/mL warrants concern for vitamin D toxicity, though symptomatic hypercalcemia generally appears above 150 ng/mL [2]. True toxicity almost always reflects supplementation above 10,000 IU/day for months, not sunlight exposure. Check serum calcium, phosphorus, and PTH concurrently. If the result is unexpectedly high in a patient on no supplements, consider immunoassay cross-reactivity (calcipotriene, paricalcitol) before diagnosing toxicity.

How to Raise 25-OH Vitamin D Safely

Correcting drug-induced depletion requires matching the supplementation strategy to the mechanism.

Supplementation Dosing Tiers

For patients with drug-induced depletion (anticonvulsants, glucocorticoids, rifampin), standard maintenance doses of 600 to 800 IU/day are inadequate. The Endocrine Society recommends 1,500 to 2,000 IU/day for adults at risk and states that pharmacologic doses of 6,000 to 10,000 IU/day may be required to correct deficiency in patients with malabsorption or drug interactions, followed by a maintenance dose of 1,500 to 2,000 IU/day [2]. These higher doses should be supervised.

D3 vs. D2 in Drug-Interaction Contexts

Cholecalciferol (D3) raises 25(OH)D roughly 87% more efficiently per unit dose than ergocalciferol (D2) in head-to-head trials [17]. In patients on enzyme inducers where every milligram of substrate matters, D3 is the preferred form. A 2012 Cochrane review of 50 randomized trials confirmed that D3 supplementation reduced all-cause mortality (RR 0.94, 95% CI 0.91 to 0.98) while D2 showed no mortality signal, though the authors cautioned that this difference could relate to dosing frequency differences [17].

Timing for Absorption-Blocking Drugs

Patients on cholestyramine, colestipol, or orlistat should take vitamin D supplements at least 4 hours before or after the drug dose. Taking D3 with the largest-fat meal of the day increases absorption by approximately 50% compared to a fasting dose, per a pharmacokinetic study in The Journal of Bone and Mineral Research [18].

How to Lower 25-OH Vitamin D

Deliberate lowering is rarely needed outside toxicity scenarios. When a patient presents with 25(OH)D above 150 ng/mL and symptomatic hypercalcemia, the first step is stopping all supplemental vitamin D and calcium. Hydration (2 to 3 liters IV normal saline) corrects hypercalcemia faster than any drug in most cases. Glucocorticoids (prednisone 20 to 40 mg/day) reduce intestinal calcium absorption and macrophage-driven 1,25(OH)2D synthesis in granulomatous cases and can lower hypercalcemia within 5 to 7 days [2]. Bisphosphonates are not first-line for D toxicity specifically, but they may be used if bone resorption is contributing. A serum calcium check 48 to 72 hours after stopping supplementation is the appropriate monitoring interval.

Who Should Be Screened and When to Retest After Drug Changes

The USPSTF found insufficient evidence to recommend universal vitamin D screening in asymptomatic adults [3]. Targeted testing is appropriate in patients starting any of the following: enzyme-inducing anticonvulsants, long-term glucocorticoids (>3 months at any dose), rifampin-based TB therapy, HIV antiretrovirals (especially efavirenz), and chronic orlistat or bile acid sequestrant therapy.

Retest timing depends on the mechanism. For depletion drugs (anticonvulsants, glucocorticoids), recheck 25(OH)D 12 weeks after starting replacement therapy, because 25(OH)D has a 15-day half-life and takes 8 to 12 weeks to reach a new steady state at a fixed supplementation dose [1]. For assay-artifact scenarios (calcipotriene users), switch to LC-MS/MS at the next draw rather than rechecking on the same immunoassay platform.