Itraconazole

Limits on what this shows

• Null and lower-risk findings: Label capsule trials: hypokalemia only 2% of 602 [1]. Itraconazole produced less hypokalemia than amphotericin B in three comparisons [15][16][17] and Le 2017 [18] reports hypokalemia significantly more common with amphotericin B than itraconazole in 440 patients with talaromycosis (counts not in the abstract read). The meta-analysis of 3,597 prophylaxis patients (Glasmacher 2003) notes hypokalemia in three studies without counts [26]. In the label toenail trial (200 mg once daily for 12 weeks, N = 112) the adverse event tables list no hypokalemia [1].
• At-risk subgroups: Higher exposure (doses of 400 to 600 mg/day; the label states exposure is greater with the oral solution than with capsules at the same dose), prolonged courses, high plasma concentrations (Beck suggests drug monitoring may help [7]), people already hypertensive or on potassium-wasting drugs (loop or thiazide diuretics, amphotericin B), neutropenic and oncology patients with gastrointestinal losses, and people on other QT-prolonging drugs; the Brandi report shows 100 mg/day can be enough in individuals [21].

Direction

Loss / disturbance. Serum potassium falls because of a drug-induced mineralocorticoid-excess state. This is not a dietary shortfall and not an instruction to raise potassium intake: the cause is the drug, the remedy is a prescriber decision (dose or exposure change, switching azole, blood pressure and potassium management). Hypertension and edema are features of the syndrome, not nutrient findings (Hard Rule 2). Low potassium also raises the drug's own QT and torsade risk (nutrient state modifying drug toxicity, Hard Rule 4).

Regulatory status

Tier A. Sporanox capsule label, warnings: pseudoaldosteronism "manifested by the onset of hypertension or worsening of hypertension" with abnormal findings "(hypokalemia, low serum renin and aldosterone, and elevated 11-deoxycortisol)"; "Monitor blood pressure and potassium levels and manage as necessary." Postmarketing table lists pseudoaldosteronism under endocrine disorders. Adverse reactions in systemic fungal infection trials with capsules (N = 602): edema 4%, hypertension 3%, hypokalemia 2%. The Tolsura label repeats the warning (section 5.4) and the same N = 602 table; the oral solution label states "Edema, hypokalemia and menstrual disorders have been reported in clinical trials with itraconazole capsules" and carries the same monitoring instruction [1][3][4]. The toenail trial (N = 112, 200 mg once daily for 12 weeks) lists hypertension 2% as a discontinuation reason and no hypokalemia [1].

Population evidence (this drug in every cohort)

Records below were found in PubMed listings [52] and read at abstract level. 1. Lee 2019 [12]: prospective phase 2, itraconazole in biochemically recurrent prostate cancer (no infection, no chemotherapy, no amphotericin), 21 enrolled and 19 evaluable for the primary endpoint; edema 52%, fatigue 38%, hypertension 24%, hypokalemia 24% (denominator not stated in the abstract; 24% corresponds to 5 of 21). Single arm, abstract level, read through a summarizing fetch tool. 2. Antonarakis 2013 [11]: noncomparative randomized phase II, 46 men with castration-resistant prostate cancer, itraconazole 200 mg/day (closed early, n = 17) or 600 mg/day (n = 29); the abstract names "a syndrome of hypokalemia, hypertension, and edema" among the common toxicities without per-arm counts (full text blocked by a bot check, so no counts were retrieved). 3. Glasmacher 2005 [13]: open-label randomized trial, itraconazole oral solution 2.5 mg/kg twice daily (n = 248) against fluconazole 400 mg daily (n = 246) in 494 neutropenic patients, 8 weeks; "more hypokalaemia occurred in the itraconazole group", total adverse events similar; counts and P value are in the paywalled full text and were not retrieved. 4. Wang 2006 [14]: open-label noncomparative phase IV, intravenous itraconazole, 156 evaluable patients with invasive fungal infection; hypokalemia was the most common adverse event, 13.5%. 5. Wu 2013 [15]: retrospective, 76 patients with hematologic disease and invasive fungal infection; hypokalemia 14.0% with itraconazole against 42.4% with amphotericin B (P = 0.005). 6. Park 2006 [16]: matched case-control trial of intravenous itraconazole against amphotericin B deoxycholate in persistent neutropenic fever, 96 patients (48 per arm); hypokalemia 24.6% against 66.7% (the denominators behind 24.6% were not given in the abstract). 7. Yoshida 2020 [17]: randomized, 102 patients, intravenous itraconazole (n = 50) against liposomal amphotericin B (n = 52); grade 3 to 4 hypokalemia-related events significantly fewer with itraconazole (P <.01). Entries 5 to 7 are active-comparator designs against a potassium-wasting drug and show that itraconazole hypokalemia is lower than amphotericin B, not that it is absent. Case series and reports [21]: Brandi 2021 (2 patients, hypertension and hypokalaemia at 100 mg/day), Denolle 2014 (itraconazole 200 mg/day, home BP 157/78, recurrent hypokalemia and leg edema, resolved 3 months after stopping), Sasaki 1999 (400 mg/day, edema at one month, potassium 2.5 mEq/L, recovered after stopping), Nelson 1993 (ventricular fibrillation attributed to itraconazole-induced hypokalaemia, letter, no abstract), Kondo 2021 (potassium 2.1 mEq/L with QT prolongation, itraconazole plus an herbal drug). Supporting records, not used to set the grade: pharmacovigilance and listing-level items. Nakano 2026 [22] (FAERS, posaconazole-focused) found itraconazole with a pseudoaldosteronism signal; Chai 2022 [23] (FAERS, 22,566 triazole records) flagged hypokalemia as a strong signal without a hypokalemia-specific ratio in the abstract; Tucker 1990 [19] (189 patients on chronic therapy) reported that hypokalemia occurred less often than gastrointestinal reactions, without a count; Zhang 2011 [20] (71 enrolled, 60 analyzed, intravenous then oral solution) lists hypokalemia among the most common drug-related events without a count; Benitez 2019 [24] (review) places hypokalemia among long-term azole effects reversible on dose reduction or stopping. Tier A is the label and cohorts 1 to 7; case reports are Tier B; pharmacovigilance is Tier C for causal strength.

Mechanism

Named and demonstrated. Hoffmann 2018 [6]: a patient with apparent mineralocorticoid excess (hypertension, hypokalemia, metabolic alkalosis, low plasma renin activity) on itraconazole; "inhibition of 11-beta-hydroxysteroid dehydrogenase 2 was demonstrated", effects resolved on withholding itraconazole and did not recur with voriconazole. Beck 2020 [7] (abstract-level, via a summarizing fetch tool): itraconazole mainly causes pseudohyperaldosteronism via 11-beta-HSD2 inhibition, about threefold more potent than posaconazole there, whereas posaconazole acts about sevenfold more strongly on CYP11B1; voriconazole, isavuconazole and fluconazole did not affect these targets. Inderbinen 2021 [8]: human 11-beta-HSD2 inhibited by itraconazole and its hydroxy metabolite at nanomolar IC50 values (rat 1.5 to 6 times weaker, mouse and zebrafish above 7 micromolar). The label's low renin and aldosterone with raised 11-deoxycortisol fits this picture.

Counter-evidence held on record

Counter-evidence and confounding No study compared potassium on itraconazole with placebo or no treatment, and none reports a mean serum potassium change from baseline, so the size of the average effect is not known; what exists is the frequency of reported hypokalemia events (2% to 24%). The high figures come from small or non-comparative cohorts (Lee n = 21; Wang n = 156, intravenous formulation) in sick patients with other causes of potassium loss, and the 2% label figure comes from capsule trials. Confounding by indication was weighed (Hard Rule 6) and is argued down by the prostate cancer cohorts without infection or amphotericin B, by the randomized fluconazole comparison, by positive dechallenge in multiple case reports, by no recurrence on voriconazole, and by the demonstrated enzyme mechanism. Counts for Antonarakis and Glasmacher were not retrieved, so the pooled magnitude rests on Lee, Wang, Wu, Park and the label. Framing caution Monitoring prompt, not a deficiency claim and not a reason to take potassium. This is a drug-induced loss; hypertension and edema are part of the syndrome. Potassium, blood pressure and drug-level questions belong with the prescriber. Supplementing potassium without clinical direction is not advised.

Limits on what this shows

• Null findings: Label toenail trial (N = 112, 200 mg once daily, 12 weeks): hypertension 2% as a discontinuation reason, no edema listed [1]. Label systemic trials: heart failure not tabulated with a rate in the text read.
• At-risk subgroups: Label: ventricular dysfunction or a history of heart failure. Higher doses and longer courses in the oncology cohorts. Older people and those on other negative inotropes or QT-prolonging drugs (label interaction tables).

Direction

Retention-type clinical syndrome plus cardiac toxicity. Sodium and water retention is the expected physiological consequence of mineralocorticoid excess, but no abstract read reports serum or urinary sodium for itraconazole, so the nutrient itself is unmeasured (Hard Rule 2). Edema, hypertension and congestive heart failure are outcomes. Dietary sodium advice is a clinical decision and not a depletion or supplementation claim.

Regulatory status

Tier A. Boxed warning: itraconazole "should not be administered for the treatment of onychomycosis in patients with evidence of ventricular dysfunction"; negative inotropic effects "were seen" when given intravenously to dogs and healthy volunteers; discontinue if signs or symptoms of heart failure appear. Postmarketing: congestive heart failure, peripheral edema, pulmonary edema, dyspnea, bradycardia. Clinical trials with the solution and intravenous forms also list cardiac failure and left ventricular failure. Systemic fungal infection trials with capsules (N = 602): edema 4%, hypertension 3% [1][3].

Population evidence

Lee 2019 [12]: edema 52% and hypertension 24% in 21 men on itraconazole for prostate cancer (abstract level). Antonarakis 2013 [11]: hypokalemia, hypertension and edema syndrome among the common toxicities in 46 men at 200 or 600 mg/day (no counts in the abstract). Case reports [21]: facial and pretibial edema after one month at 400 mg/day (Sasaki), leg edema with hypertension at 200 mg/day (Denolle). No source read gives an ejection fraction or a heart failure rate for itraconazole; the negative inotropy evidence in this review is the label statement only.

Mechanism

Edema and hypertension: mineralocorticoid excess through 11-beta-HSD2 inhibition (see the potassium entry [6][7][8]). Heart failure: direct negative inotropy (label); the label does not tie it to the steroid pathway. The two mechanisms are different and should not be merged.

Counter-evidence held on record

Counter-evidence and confounding Oncology and infection cohorts carry their own causes of edema and hypertension (androgen deprivation, corticosteroids, hypoalbuminemia, fluids). The heart failure statement rests on the label and on animal and healthy-volunteer intravenous data cited there; the primary reports were not opened. Framing caution Clinical outcome entry. Not a sodium deficiency or excess claim and not a reason to change sodium intake without a prescriber. Report swelling, sudden weight gain or breathlessness promptly.

Limits on what this shows

• Null findings: No adrenal insufficiency with itraconazole alone in the 12-patient arm of Skov [43].
• At-risk subgroups: People using inhaled, oral or injected corticosteroids metabolized by CYP3A4 (budesonide, fluticasone, others).

Direction

Hormone suppression through CYP3A4 inhibition of steroid clearance (interaction), plus steroid-pathway disturbance in the mineralocorticoid-excess syndrome. Not a deficiency of any nutrient (Hard Rule 2).

Regulatory status

Tier A. Label: adrenal insufficiency listed among events infrequently reported in systemic fungal trials; "elevated 11-deoxycortisol" in the pseudoaldosteronism warning [1].

Population evidence

Skov 2002 [43]: controlled clinical trial, 25 patients on itraconazole plus budesonide, 12 on itraconazole alone and 30 controls; "Eleven of the 25 patients treated with both itraconazole and budesonide had adrenal insufficiency"; none on itraconazole alone or in controls; mineralocorticoid and gonadal insufficiency were not seen (abstract level, via a summarizing fetch tool). Bishnoi 2024 [44]: one patient with exogenous Cushing syndrome who developed adrenal insufficiency on itraconazole (case letter, no abstract numbers read). In vitro: Jager 2023 and 2024 [25] describe azole effects on adrenal steroidogenesis in H295R cells without numbers in the abstracts.

Mechanism

Interaction: itraconazole inhibits CYP3A4, raising budesonide exposure (label boxed-warning mechanism for CYP3A4 substrates [1]; the Skov abstract read does not itself state the mechanism). Steroidogenesis: enzyme inhibition in vitro [25].

Counter-evidence held on record

Counter-evidence and confounding Small arms; the combination, not itraconazole alone, produced the finding. Histoplasmosis itself causes adrenal insufficiency, so adrenal findings in treated fungal disease are confounded by indication. Framing caution Hormone and drug-interaction effect for the prescriber; not a nutrient entry and not a reason to supplement anything.

Limits on what this shows

• Null findings: The label text read does not mention grapefruit; MedlinePlus raises it without data [9]. Wikipedia (Tier D) says capsules may be taken with orange juice or cola because acid improves absorption [49]; the labels support only cola, in fasting patients with achlorhydria, and say nothing about orange juice in the text read.
• At-risk subgroups: People on acid suppression, achlorhydria or after gastric surgery; people switching between capsule, solution and Tolsura (not interchangeable).

Direction

Nutrient or diet to drug. Filed as timing interactions under Hard Rule 4. Nothing here says itraconazole lowers a nutrient.

Regulatory status

Tier A. Capsules: absolute bioavailability about 55%, maximal after a full meal; "take with a full meal"; absorption reduced with reduced gastric acidity (H2-receptor antagonists, proton pump inhibitors); give acid-neutralizing medicines at least 2 hours before or 2 hours after; 8 oz non-diet cola increased absorption in fasting AIDS patients with relative or absolute achlorhydria; capsule exposure is lower than solution exposure [1][2]. Oral solution: bioavailability 30% higher fasting, "take without food, if possible" [4]. Tolsura: "must be administered with food"; acid-reducing drugs including omeprazole increase its exposure and a dose reduction may be needed [3]. S. boulardii: "Not recommended during and 2 weeks after" treatment [1][2]. Consumer pages (Tier D, cross-check only): MedlinePlus advises capsules with a full meal, solution on an empty stomach, antacids 2 hours apart, and to discuss grapefruit with a doctor; Drugs.com lists only the food effect [9][10].

Population evidence

The labels summarize healthy-volunteer and patient pharmacokinetic studies whose primary reports were not opened. No percent change for acid suppressors or cola is stated in the label text read. An attempt to retrieve the primary grapefruit, cola and omeprazole studies from PubMed was refused repeatedly (HTTP 429), so no primary effect size is cited here.

Mechanism

Capsule dissolution needs acid and fat; the cyclodextrin solution and the Tolsura formulation differ in how food and acid affect them (label). Probiotic yeast: the label gives the instruction without a stated mechanism in the text read.

Counter-evidence held on record

Counter-evidence and confounding Formulation-specific: capsule and solution give opposite food advice, so a single "take with food" statement is wrong for the solution. Framing caution Timing guidance for the prescriber and pharmacist; separation interval for acid-neutralizing medicines is 2 hours before or after. Not a depletion claim.

Limits on what this shows

• Null findings: Hyperkalemia is listed in the same label sentence, so the list does not show a consistent electrolyte direction.
• At-risk subgroups: Not identifiable from the sources read; patients on diuretics, amphotericin B or with diarrhea have independent magnesium loss.

Direction

Possible loss or disturbance through the drug; no mechanism is named. Not an indication to supplement.

Regulatory status

Tier A, weak. Sporanox label, other clinical trials, metabolism and nutrition disorders: "hyperglycemia, hyperkalemia, hypomagnesemia", no frequencies [1].

Population evidence

No cohort, trial or case report measuring magnesium on itraconazole was retrieved. The Benitez 2019 review of long-term azole adverse effects names hypokalemia and hyponatremia, not magnesium [24]. In the Le 2017 talaromycosis trial (itraconazole 221 against amphotericin B 219) the abstract states hypomagnesemia was significantly more frequent with amphotericin B than with itraconazole, and in Pasqualotto 2023 (118 patients, all given itraconazole after induction) hypomagnesemia frequency was similar across the three amphotericin arms; neither abstract gives counts and neither has a no-itraconazole control [18][51].

Mechanism

None named. If mineralocorticoid excess lowers magnesium it is by analogy with potassium wasting and was not shown for itraconazole in anything read.

Counter-evidence held on record

Counter-evidence and confounding A bare list entry without rates cannot distinguish a drug effect from disease, co-medication (amphotericin B, diuretics) or intravenous fluids in the treated population. Framing caution Context flag only. No magnesium supplement is implied.

Limits on what this shows

• Null findings: The two rat studies above are direct nulls for itraconazole.
• At-risk subgroups: People prescribed oral tretinoin (ATRA) or other retinoids. No evidence for people not on a retinoid.

Direction

Possible retention of an exogenous retinoid drug (excess direction, drug interaction). Not a vitamin A deficiency; adding vitamin A would be the hazard (Hard Rules 4 and 5).

Regulatory status

No vitamin A statement in the label text read [1][2].

Population evidence

Cordoba 2008 (Eur J Clin Pharmacol, PMID 18584165): one patient with acute promyelocytic leukemia, hypercalcemia attributed to an ATRA and itraconazole interaction, treated with zoledronic acid (letter; only title and indexing seen). Ueno 1995 (Am J Hematol, PMID 7485119): letter titled "Itraconazole and retinoid resistance" (no abstract seen). Both from a PubMed listing read through a summarizing tool [45]. Vanden Bossche 1988 [46]: rat skin microsomes; itraconazole had no effect on retinoic acid metabolism up to 10^-5 mol/L, whereas ketoconazole inhibited with 50% inhibition at 6.5 x 10^-7 mol/L. Van Wauwe 1990 [47]: rats given intravenous retinoic acid; itraconazole had no sparing effect on elimination (plasma below 1 ng/mL as in controls) while ketoconazole raised the half-life from 27 to 43 minutes (abstract; itraconazole dose not given).

Mechanism

Retinoic acid hydroxylation (CYP26) inhibition is the usual azole route; for itraconazole the two animal studies were negative, so the mechanism is not shown.

Counter-evidence held on record

Counter-evidence and confounding Case letters with no numbers read, one patient each, and the animal nulls argue against a pharmacokinetic interaction; hypercalcemia is a known ATRA effect on its own. Framing caution Drug-interaction check for prescribers if itraconazole and a retinoid drug are combined; not a vitamin A deficiency claim and no vitamin A supplement advice.

Limits on what this shows

• Null findings: None found, because no study measured vitamin D metabolites in itraconazole users.
• At-risk subgroups: None identified.

Direction

Possible reduction of the active hormone through enzyme inhibition (hormone-level, not dietary). Not an indication to supplement; Hard Rule 2 applies.

Regulatory status

The label text read contains no vitamin D or calcium statement [1][2].

Population evidence

Zhang 2013 [39]: 16-HBE bronchial epithelial cells; 500 nM 25(OH)D3 and 50 nM 1,25(OH)2D3 induced TSLP; itraconazole at 10^-6 M, used as a 1-alpha-hydroxylase inhibitor, reduced 25(OH)D3-induced TSLP but not 1,25(OH)2D3-induced TSLP (abstract level, summarizing tool). Cusato 2022 [40]: observational pharmacogenetic study, 636 plasma samples across azoles (itraconazole share not stated); CYP27B1, VDR and CYP24A1 variants related to drug exposure, no serum vitamin D values; the study is about vitamin D genes affecting azole levels, the opposite direction. Brodkin 2025 [41]: one dog with pythiosis and hypercalcemia treated with itraconazole; animal case, no numbers in the abstract.

Mechanism

Proposed: azole inhibition of CYP27B1; shown only as a pharmacological tool effect in the cell study. No human enzyme or hormone data.

Counter-evidence held on record

Counter-evidence and confounding Cell work at a single concentration, no human exposure data, and the human cohort evidence runs the other way (vitamin D genes affect azole exposure). Framing caution Mechanistic only. Not a vitamin D deficiency claim and no supplement recommendation.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C source supports a yellow grade (no itraconazole record measures zinc status in a mammalian system) and no circulating claim was found to rebut. A disclosed evidence gap, not a finding of safety, with part of it tooling-limited. Zinc, Copper and Iron are separate entries; no source is shared.

What was searched

PubMed listing "itraconazole zinc" (42 results; page 1, 10 records, read through the format=pubmed listing and a summarizing tool) [48]: reviews of tinea versicolor and leishmaniasis, zinc-itraconazole coordination complexes tested on parasites and Sporothrix in vitro (Azevedo-Franca 2020), zinc pyrithione as a separate topical antifungal, a fungal zinc cluster transcription factor, and cancer papers where "zinc" appears only in the GLI1 "zinc finger" gene name. No record measured serum zinc, zinc absorption or a zinc supplement interaction with itraconazole. The label text read has no zinc statement [1][2]; MedlinePlus and Drugs.com name none [9][10].

Channel outcomes

Pages 2 to 5 of the 42 results (32 records) were not read; no query with "serum zinc", "zinc deficiency" or "taste" was run. Open-web search was unavailable (shared budget exhausted), so the absence of a circulating zinc claim is tooling-limited; the absence of a primary study in the 10 records read is a literature gap as far as read.