Strokes treated by fast-acting Vitamin D (rodents thus far)

Acute, Fast-Acting Vitamin D in Ischemic Stroke: A Mechanistic, Preclinical, and Clinical Research Report

There is a biologically credible but clinically unproven case for administering a fast-acting vitamin D metabolite (calcifediol or calcitriol) within the acute window of ischemic stroke. Mechanistically, calcitriol/VDR signaling targets nearly every arm of the ischemia-reperfusion cascade — NF-κB, NLRP3, TLR2/4, MMP-9, NMDAR excitotoxicity, oxidative stress (Nrf2/HO-1), apoptosis (Bcl-2/cytochrome-c/caspase-3), mitochondrial bioenergetics (AMPK/AKT/GSK-3β), and BBB integrity — and rodent MCAO studies giving calcitriol 1 µg/kg IP within 30 min of reperfusion consistently reduce infarct volume by 25–50% and improve neurological scores.

However, the only published acute human RCTs use slow-acting cholecalciferol (single IM 300,000–600,000 IU within 24 h of stroke), and the only clinical study using calcitriol acutely (Naderian 2023; n=38; 10 µg single oral dose) measured only molecular biomarkers (TLR4/NF-κB/IL-1β/IL-6 down-modulation at 24 h) — not infarct volume, NIHSS, or mRS. No RCT of acute calcifediol in ischemic stroke has been published or is registered. The strongest defensible protocol — extrapolating from the Castillo/Nogués calcifediol-COVID paradigm and the Hesami stroke RCT — would be: in any acute ischemic stroke patient with measured or presumptive 25(OH)D <30 ng/mL, give oral calcifediol 0.5–1.0 mg (500–1000 µg, ≈ 20–40,000 IU equivalent) at hospital admission, repeat 0.266 mg on day 3 and day 7, while monitoring serum calcium. Adjunctive low-dose calcitriol (0.25–0.5 µg PO BID for 3 days) could be considered for the most rapid genomic and non-genomic VDR activation, with calcium monitoring. This protocol is currently a clinical hypothesis, not a standard-of-care intervention, and should be confirmed in a properly powered RCT.

TL;DR (3 bullets)

Strong preclinical case, weak human evidence: Acute post-stroke calcitriol (1 µg/kg IP within 30 min) reproducibly reduces infarct volume and improves neurological scores in MCAO rats via VDR-mediated suppression of NF-κB/NLRP3/TLR4, Nrf2/HO-1 induction, BBB stabilization, and reduction of MMP-9 — but the few small acute human RCTs used cholecalciferol (intrinsically too slow to act in the acute window) and no RCT of acute calcifediol or calcitriol in ischemic stroke exists.
Pharmacokinetics decisively favor calcifediol or calcitriol over cholecalciferol for an acute-window intervention. Calcitriol gives near-immediate VDR activation (Tmax 3–6 h, t½ 5–8 h). Calcifediol bypasses hepatic 25-hydroxylation and raises serum 25(OH)D within hours (substantial elevation by 4 h after a 0.5–1 mg oral dose; predictable, linear dose-response). Standard oral D3 takes 7–30+ days to peak and is irrelevant for the first 72 h after stroke onset.
The Castillo/Nogués calcifediol-COVID trials provide the strongest paradigm for acute high-dose calcifediol in critical illness (≈80% relative reduction in ICU admission, OR 0.13–0.21 for ICU/mortality after adjustment), demonstrating the feasibility, safety, and rapid bioactivity of a 0.5 mg oral loading dose. Translating this paradigm into ischemic stroke is plausible but requires a dedicated phase II RCT with infarct-volume and 90-day mRS endpoints.

Key Findings

VDR signaling is operational throughout the ischemic brain. Neurons, astrocytes, microglia, oligodendrocytes, and cerebrovascular endothelial cells all express VDR, CYP27B1, and CYP24A1. Both 25(OH)D₃ and 1,25(OH)₂D₃ cross the blood-brain barrier and the blood-CSF barrier; CSF concentrations correlate with serum. Local intracerebral 1α-hydroxylation makes vitamin D effectively a neurosteroid.
Calcitriol (1 µg/kg IP) given immediately post-MCAO reduces infarct volume in nearly every rodent study that has tested early (≤30 min) administration. Mechanisms include: TLR4/MyD88/NF-κB suppression (Ramshini 2025; Vahidinia/Tameh group), NLRP3 inflammasome inhibition via ROS/TXNIP modulation (2025/2026 study with calcitriol at 30 min, 24 h, 48 h post-MCAO), Nrf2/HO-1 induction (Vahidinia 2022), NMDAR/CYP46A1-mediated antiexcitotoxicity (Khassafi 2022), AMPK/AKT/GSK-3β-mediated mitochondrial protection (Yuan/Li 2022), and TGF-β/Smad2/3-mediated peri-infarct angiogenesis with VEGF upregulation. VDR antagonist pyridoxal-5-phosphate partially abolishes neuroprotection, confirming the pathway is genuinely VDR-dependent.
One important negative preclinical finding: Balden et al. (Endocrinology 2012) reported that vitamin D3 at 10 µg/kg given IP starting 4 h post-MCAO did not reduce infarct volume, although vitamin D deficiency (chronic dietary) clearly worsened infarct at 5 days. This implies the therapeutic window may be narrow (likely <2–3 h) and/or D3 is too slow-acting (it must first undergo hepatic 25-hydroxylation), reinforcing the rationale for calcitriol or calcifediol over D3 in the acute window.
BBB protection and tPA synergy. Calcitriol post-ischemia attenuates vasogenic edema, reduces BBB permeability, and downregulates MMP-9 via VDR-IKKβ-NF-κB blockade (Panahpour 2019; Won 2015). Because tPA-induced hemorrhagic transformation is largely MMP-9–mediated, this is a mechanistically attractive (though completely untested in humans) rationale for adjunctive calcitriol/calcifediol with thrombolysis.
Observational human data are highly consistent. Across >20 cohorts: low admission 25(OH)D predicts higher NIHSS, larger DWI infarct volume (Turetsky 2015; Park 2015), worse 90-day mRS, more early neurological deterioration, and higher mortality (Wajda 2019: severe deficiency <10 ng/mL → IRR 2.52 for all-cause mortality at 44.9 months). In a Chinese non-diabetic stroke cohort (n=266), VDD increased poor functional outcome OR 3.2 and mortality OR 3.9. None of these prove causality, but they establish a strong dose-response gradient and define the population most likely to benefit.
The only published acute clinical RCTs use cholecalciferol (slow PK):
- Hesami 2022 (Iran, n=41, double-blind RCT): single IM 600,000 IU cholecalciferol within 24 h of moderate ischemic stroke (NIHSS 5–15, baseline 25(OH)D ≤30 ng/mL). NIHSS at 48 h and Barthel Index at 3 months improved significantly vs. control; NSE biomarker was unchanged.
- Iranian 300,000 IU IM RCT (n=60): improved IL-6, mRS, MMSE; TNF-α and NIHSS not significantly improved.
- Open-label Indian trials (~2017–2018): IM 600,000 IU improved Scandinavian Stroke Scale and 24-week survival.
- Momosaki 2019 (Japan, n=100, RCT): 2000 IU/day x 8 weeks after acute stroke; no benefit on Barthel Index gain — but this is a sub-acute rehabilitation dose, not an acute intervention.
The only acute human study with a fast-acting metabolite is Naderian 2023 (Kashan; IRCT2017012532174N1; n=38). A single 10 µg oral calcitriol dose given on admission reduced TLR2, TLR4, NF-κB, IL-1β, and IL-6 mRNA expression at 24 h vs. control. Clinical outcomes (NIHSS, mRS, infarct volume) were not reported. This is proof-of-concept that acute calcitriol modulates the predicted molecular targets in human ischemic stroke patients.
The Castillo/Nogués calcifediol-COVID trials are the most informative paradigm for acute high-dose calcifediol in inflammatory critical illness. Castillo 2020 (Córdoba, RCT, n=76): oral calcifediol 0.532 mg on admission + 0.266 mg on days 3, 7, then weekly. ICU admission 2% (1/50) vs. 50% (13/26); adjusted OR 0.03 (95% CI 0.003–0.25). Nogués 2021 (Barcelona, n=838): calcifediol-treated 4.5% ICU vs. 21% untreated; adjusted OR 0.13 (0.07–0.23); mortality OR 0.21 (0.10–0.43). ALBACOVIDIOL (n=230) showed in-hospital mortality 12.6% vs. 23.4% (OR 0.47, 0.23–0.95). These trials demonstrate that 0.5 mg oral calcifediol is feasible, well-tolerated, and pharmacodynamically active within hours.
VITdAL-ICU (Amrein 2014) showed no benefit of 540,000 IU enteral cholecalciferol on hospital length of stay overall (n=480), but in the prespecified severe-deficiency subgroup (25(OH)D ≤12 ng/mL): 28-day mortality 36.3% placebo vs. 20.4% vitamin D (HR 0.52, NNT=6). VIOLET (NEJM 2019) showed no benefit of pre-ICU high-dose D3 in critically ill patients without specific deficiency targeting. Together these support: (a) the deficient subset is the most likely to respond, (b) cholecalciferol PK is suboptimal for acute intervention, (c) maintenance dosing matters.
No registered RCT specifically testing acute calcifediol or calcitriol in ischemic stroke is currently active on ClinicalTrials.gov. The 2023 Fleet et al. systematic review (10 trials, 691 patients) confirmed all available stroke-vitamin D RCTs are post-acute/rehabilitation focused. The major active stroke-related vitamin D study (NCT04070833, Brigham, ancillary to VITAL) examines pre-event D3 and post-stroke functional outcome at 6/12 months — not acute intervention.

A. Mechanistic Rationale

Genomic VDR signaling targets virtually every step of the ischemic cascade:

Excitotoxicity / NMDAR / Calcium homeostasis. Calcitriol upregulates the NR3A NMDAR subunit (which dampens NMDA-mediated Ca²⁺ influx) via the MEK/ERK-CREB pathway in hippocampal neurons; PD98059 abolishes this effect (Fu 2013, Mol Med Rep). Calcitriol pretreatment in MCAO rats normalizes NMDAR and CYP46A1 expression (Khassafi 2022). VDR ligands also acutely modulate voltage-gated Cl⁻ channels (non-genomic, second timescale).
Inflammation / TLR/NF-κB/NLRP3. Calcitriol antagonizes TLR4 (molecular docking confirmed), suppresses MyD88, NF-κB, IL-1β, IL-6, TNF-α (Ramshini 2025; Atif 2013; Naderian 2023 in humans). VDR directly binds IKKβ to block NF-κB nuclear translocation. NLRP3 inflammasome assembly is inhibited via the ROS/TXNIP axis (2026 ScienceDirect study). Microglial polarization shifts from M1 to M2; Th1/Th17 → Th2/Treg.
Oxidative stress. Calcitriol activates Nrf2/HO-1 (Vahidinia 2022), upregulates SOD, glutathione peroxidase, glutathione, and γ-GCL; downregulates NADPH oxidase (NOX2) subunits and MMP-9 (Mongolian gerbil study, Kalueff group). MDA and superoxide anion fall significantly.
Apoptosis and mitochondrial bioenergetics. Calcitriol downregulates p53, cytochrome-c, caspase-3 and upregulates Bcl-2, BDNF, p-AMPK, p-AKT, p-GSK-3β, ATP, and succinate dehydrogenase (Yuan/Li 2022). VDR antagonist P5P partially blocks all of these effects, confirming the pathway is VDR-mediated.
BBB and endothelial integrity. Calcitriol prevents hypoxia/reoxygenation-induced BBB disruption in cerebral endothelial cell culture via VDR-mediated NF-κB inhibition (Won 2015, PLoS ONE-published). In MCAO rats, post-ischemic calcitriol reduces vasogenic edema, Evans blue extravasation, and improves neurological scores 24 h after 1-h MCAO (Panahpour 2019). The same NF-κB blockade that reduces edema also suppresses MMP-9, the principal driver of tPA-related hemorrhagic transformation — a tantalizing but untested adjunctive rationale.
Angiogenesis/repair. 1,25-D3 upregulates TGF-β/Smad2/3/VEGF, increasing peri-infarct microvessel density and CD31⁺ vasculature (PMC8966232) — relevant for sub-acute recovery rather than the first 24–72 h.
Combinatorial neurosteroid synergy. Atif/Stein group showed progesterone + vitamin D hormone (VDH) outperform either alone in tMCAO rats, with stronger BDNF/TrkB/Erk1/2 induction and TLR4/NF-κB suppression. The most efficacious combination uses low (not high) VDH dose with progesterone, suggesting non-monotonic dose-response.

Genomic vs. non-genomic timescale. Genomic effects (gene transcription via VDR-RXR-VDRE) take hours and require the pathway calcifediol → calcitriol → VDR. Non-genomic effects (membrane VDR, ion channel modulation, MAPK activation) occur within seconds to minutes once calcitriol is at the target. This dual timescale is critical: the first hours of stroke involve excitotoxic Ca²⁺ overload (non-genomic targets), while the 6–72 h window involves transcriptional reprogramming (genomic targets). A fast-acting metabolite captures both windows; slow-acting D3 captures neither in time.

B. Preclinical Evidence — Quantitative Summary

Study	Model	Drug & Dose	Timing	Key Result
Ramshini 2025 (IBRO Neurosci Rep)	Rat tMCAO, 1 h occlusion	Calcitriol 1 µg/kg IP	30 min, 24 h, 48 h post	Infarct volume ↓ (P<0.001), neurological score ↑ (P<0.05); TLR4, MyD88, NF-κB, FGFR2 mRNA all reduced
Anonymous 2026 (Brain Res Bull)	Rat tMCAO	Calcitriol 1 µg/kg IP	30 min, 24 h, 48 h post	Infarct ↓, neurological deficit ↓; ROS/TXNIP/NLRP3 axis suppressed
Vahidinia 2022	Rat tMCAO	Calcitriol pretreatment	Before injury	Nrf2/HO-1 axis activated, infarct ↓
Khassafi 2022 (JNEN)	Rat MCAO 1 h + 23 h reperfusion	Calcitriol pretreatment	Before injury	NMDAR and CYP46A1 normalization, neuroprotection
Yuan/Li 2022 (PMC9618954)	Rat MCAO	1,25-D3 (calcitriol)	Post-injury	Infarct ↓; AMPK/AKT/GSK-3β/VDR up; P53/CytC/caspase-3 down; reversed by VDR antagonist P5P
Panahpour 2019	Rat 1 h MCAO + 23 h reperfusion	Calcitriol IP	Post-ischemia	BBB permeability ↓, vasogenic edema ↓, BDNF ↑, antioxidant enzymes ↑, infarct ↓
Fu 2013 (Mol Med Rep)	Rat MCAO, 7 d outcome	Calcitriol IP	Post-injury	Infarct ↓ at 7 d, NR3A and p-CREB ↑; PD98059 abolishes effect
Atif 2013 (Neuropharmacology)	Rat tMCAO + OGD neurons	Progesterone + VDH	5 min pre-reperfusion + 7 d	Combination > monotherapy at day 7; better infarct reduction & functional outcomes
Atif 2020	Rat tMCAO + LPS	P + VDH	5 min pre-reperfusion + d1–3	Reduced post-stroke systemic inflammation
Balden 2012 (Endocrinology)	Rat ET-1 MCAO	D3 (cholecalciferol) 10 µg/kg IP	Starting 4 h post	No infarct reduction — important negative; though chronic VDD worsened infarct
Evans 2018 (JCBFM)	Mouse 1 h MCAO	Diet-induced VDD	Pre-injury	No effect on 24 h infarct in young male mice
PMC5932460	Mongolian gerbil global ischemia	D3 pretreatment	Pre-injury	NOX2, MMP-9, MDA all ↓; VDR ↑
PMC8966232	Rat MCAO	1,25-D3	Post-injury	TGF-β/Smad2/3/VEGF ↑, peri-infarct angiogenesis ↑
Kim 2020 (Int J Emerg Med)	Rat global ischemia (4-vessel)	Paricalcitol 1 µg/kg IP	5 min, 1, 2, 3 d post	Improved neurological function, motor function, hippocampal neuron survival

Summary of preclinical pattern: - The most efficacious regimen is calcitriol 1 µg/kg IP within 30 minutes of reperfusion plus continued daily dosing. - Calcitriol given very early (≤30 min) reproducibly works; cholecalciferol given at 4 h does not (Balden) — consistent with the slow PK of D3. - Effect sizes are clinically meaningful: typical 25–50% reduction in TTC-stained infarct volume. - Mechanism is unambiguously VDR-dependent (P5P abolishes effect). - Most studies use Iranian, Korean, or Chinese groups; replication by independent labs (Atif/Stein at Emory) supports robustness.

C. Human Clinical Evidence

Acute administration RCTs (within 24 h of onset):

Naderian 2023 (J Neuroimmunol; IRCT2017012532174N1): n=38 ischemic stroke patients on admission. Single 10 µg oral calcitriol given immediately. At 24 h, TLR4, TLR2, NF-κB, IL-1β, IL-6 mRNA all significantly reduced vs. control. Clinical (NIHSS, mRS) and infarct outcomes were NOT reported — this is purely a molecular proof-of-concept that acute calcitriol activates the predicted human pathway.
Hesami 2022 (Stroke Res Treatment, n=41 RCT): Single IM 600,000 IU cholecalciferol within 24 h (moderate ischemic stroke, NIHSS 5–15, deficient). Significant improvement in NIHSS at 48 h and BI at 3 months; NSE biomarker not changed. Limitations: small n; cholecalciferol PK means 25(OH)D rise was not measurable until well beyond the 48-h NIHSS endpoint, suggesting either non-genomic effects or that the IM depot acts as a slow-release reservoir of D3.
Iranian 300,000 IU IM RCT (Torkaman 2021, n=60): IL-6 ↓, mRS ↓, MMSE ↑; NIHSS and TNF-α not significantly changed.
Indian open-label trials (~2017–2018): IM 600,000 IU cholecalciferol → Scandinavian Stroke Scale improved; one trial reported improved survival at 24 weeks.

Subacute / rehabilitation RCTs (mostly negative or null):

Momosaki 2019 (Japan, n=100): 2000 IU/day x 8 wk after acute stroke — no Barthel Index gain difference (vehicle 19.5±13.1 vs. D3 19.0±14.8; p=0.88).
Wroclaw single-blind RCT (Ali 2024, n=160 in rehab phase): mixed signal.
Canneva 2025 (Italy, n=85 retrospective): calcifediol vs. cholecalciferol in subacute rehab — calcifediol raised 25(OH)D significantly faster (P<0.001 at 4 months) but no functional outcome difference.

Observational data on baseline 25(OH)D and outcomes:

Wajda 2019 (n=240, 44.9-mo follow-up): 25(OH)D <10 ng/mL → IRR 2.52 (95% CI 1.44–4.68) for all-cause mortality, independent of age and mRS. Severe deficiency in 62.9% of patients.
Park 2015 (n=235): 25(OH)D in lowest quartile → 3.0× odds of NIHSS ≥6 and worse 90-day mRS.
Daubail 2013 (Eur J Neurol, n=382): 25(OH)D predicts severity and 100-day functional outcome.
Turetsky 2015 (JSCD, n=96): low vitamin D independently associated with larger DWI lesion volume.
Liu 2018 (Chinese non-diabetics, n=266): VDD → adjusted OR 3.2 for poor functional outcome and 3.9 for mortality at 90 days.
Wang 2019 (post-thrombolysis, n=478): low 25(OH)D → adjusted OR 2.10 for poor 90-day mRS after IV alteplase.
Multiple recent prospective studies confirm Pearson r ≈ −0.4 between 25(OH)D and admission NIHSS.

The Castillo/Nogués calcifediol-COVID paradigm (most informative for acute high-dose calcifediol PK and safety):

Castillo 2020 (JSBMB, NCT04366908): 76 hospitalized COVID-19 patients, randomized 2:1 to oral calcifediol 0.532 mg on admission then 0.266 mg on days 3, 7, then weekly. ICU admission 1/50 (2%) vs. 13/26 (50%); adjusted OR 0.03 (95% CI 0.003–0.25); 0/50 deaths vs. 2/26.
Nogués 2021 (n=838): adjusted OR 0.13 (0.07–0.23) for ICU; OR 0.21 (0.10–0.43) for in-hospital mortality.
ALBACOVIDIOL 2024 (n=230): OR 0.47 (0.23–0.95) for in-hospital mortality.
Implications for stroke: these studies prove that 0.5 mg oral calcifediol given acutely is (a) safe, (b) raises serum 25(OH)D rapidly, and (c) can deliver clinically meaningful outcome benefit in acute inflammatory critical illness. Whether the effect translates to ischemic stroke is the key unanswered question.

Critical illness paradigm:

VITdAL-ICU (Amrein 2014, JAMA, n=480): 540,000 IU enteral cholecalciferol → primary endpoint (length of stay) negative; subgroup with 25(OH)D ≤12 ng/mL → 28-day mortality 36.3% → 20.4%, HR 0.52 (0.30–0.89), NNT=6.
VIOLET (Ginde NEJM 2019): early 540,000 IU D3 in ICU patients without specific deficiency targeting — null. Both reinforce: deficient subset most likely to respond; cholecalciferol PK suboptimal.
VITDALIZE (NCT03188796): ongoing phase III.

D. Pharmacokinetics and Dosing

Parameter	Calcitriol (1,25-OH₂D₃)	Calcifediol (25-OH-D₃)	Cholecalciferol (D₃)
Mechanism	Direct VDR ligand	Bypasses hepatic 25-hydroxylation	Requires hepatic 25-OH + renal/extra-renal 1α-OH
Tmax (oral)	3–6 h	~4 h (small dose); ~7 d for 100,000 IU bolus	7–14 days for 25(OH)D rise
Half-life	5–8 h (pediatric/HD up to 27 h)	~15 days (some report 21–199 days at high cumulative load)	Cholecalciferol itself ~24 h; 25(OH)D pool ~2–3 weeks
Time to detectable VDR effect	Minutes (non-genomic); hours (genomic)	Hours after first dose	Days after first dose
Linearity of dose-response	Highly potent, narrow range	Linear, predictable	Variable, baseline-dependent
Hypercalcemia risk at therapeutic doses	Highest; tight monitoring needed	Low at typical doses; possible at >0.5 mg/d sustained	Very low
Brain penetration	Crosses BBB; locally synthesized	Crosses BBB; substrate for local CYP27B1	Crosses BBB but inactive until hydroxylated
Acute use feasibility	Yes (preferred for fastest VDR effect)	Yes (preferred for sustained 25(OH)D + safer Ca profile)	No (too slow for the acute window)

Calcifediol oral PK quantitative data: - Single 0.532 mg loading dose (Castillo COVID protocol) raises serum 25(OH)D substantially within 24 h. - Brandi 2021 (Italy, n=50, 20 vs 30 µg/d): peak 25(OH)D 59.3 ng/mL by day 90 (20 µg) and 72.3 ng/mL by day 60 (30 µg). - Vaes 2017 (older adults): 10–15 µg/d calcifediol > 20 µg/d D3 — calcifediol reached >75 nmol/L by 4 weeks. - Calcifediol is ≈3.2-fold more potent on a weight basis than cholecalciferol; 100-µg weekly normalizes deficiency reliably.

Calcitriol oral PK (FDA Rocaltrol label): - Peak serum 60.0±4.4 pg/mL at 2 h after 0.5 µg PO (baseline 40 pg/mL). - Half-life 5–8 h, prolonged in CKD/HD. - 99.9% protein-bound (vitamin D binding protein/DBP). - Oral or IV available.

Liposomal vitamin D3: - Compared to oily formulation, liposomal D3 produces ≈4–5× higher AUC₁day for 25(OH)D rise (Lewinski 2022, Nanomedicine) in deficient subjects — but this is still cholecalciferol, requiring hepatic 25-OH; "rapid" here means hours-faster D3 absorption, not hours-faster 25(OH)D ceiling. Less suitable than calcifediol for the acute stroke window unless data emerge directly demonstrating same-day 25(OH)D normalization. - Sublingual / oral spray D3: marketing claims of "rapid" absorption are not supported by rigorous PK comparing 25(OH)D Tmax to oral D3; mechanistic biology (hepatic conversion) constrains the kinetics regardless of absorption route.

Brain bioavailability: - Both 25(OH)D and 1,25(OH)₂D are present in human CSF; concentrations correlate with serum. - Neurons, astrocytes, microglia, oligodendrocytes express CYP27B1 → local 1α-hydroxylation. Therefore raising serum 25(OH)D substantially (the calcifediol approach) provides substrate for local activation; raising serum calcitriol provides direct VDR ligand. Both strategies should be biologically active.

Safety considerations for acute use: - Calcitriol: hypercalcemia risk significant; monitor calcium daily for 3–7 days; contraindicated in advanced CKD without nephrology supervision; dose 0.25–0.5 µg PO BID is well established. - Calcifediol: tolerated up to single oral 1 mg in healthy adults and COVID-19 patients (Castillo, Nogués) without hypercalcemia events of clinical concern. Half-life is long enough that effect persists 1–2 weeks. - Loading IM cholecalciferol 600,000 IU was used safely in stroke RCTs but PK is too slow for acute neuroprotection and irrelevant outside a subacute/recovery rationale.

E. Current Trials

Active or planned trials directly testing acute fast-acting vitamin D in ischemic stroke: essentially none.

NCT04070833 (Brigham & Women's, ancillary to VITAL): cohort follow-up of stroke survivors who were already randomized to D3 2000 IU/d in VITAL — examines functional outcome at 6/12 months. Pre-stroke randomization, not acute intervention.
NCT06817512 (ISNIS): Vitamin K2 (not D), 1-year intervention. Not relevant.
VITDALIZE (NCT03188796): ICU population; cholecalciferol 540,000 IU + 4000 IU/d. Mixed critical illness, will likely include some stroke.
No registered RCT of calcifediol or calcitriol acute administration in ischemic stroke as of the May 2026 search. The Fleet 2023 systematic review (Sage; 10 trials, 691 patients) confirmed all stroke vitamin D RCTs to date are post-acute. This is a clear, fundable trial gap.

F. Gaps, Critiques, and Why This Has Not Been Adopted

Translational gap. Rodent calcitriol 1 µg/kg IP given within 30 min of reperfusion does not map cleanly to human dosing or timing. The human-equivalent dose (HED) of 1 µg/kg in a 250-g rat is approximately 0.16 µg/kg in humans (≈11 µg for a 70-kg adult by allometric scaling); the typical 0.25–0.5 µg PO calcitriol dose used in renal disease is therefore in the right order of magnitude, but no dose-finding work has been done in stroke.
Timing window. Balden 2012 negative result (D3 at 4 h post-stroke) suggests that any clinically useful trial must deliver fast-acting metabolite within the first 1–2 hours — i.e., in the same window as IV thrombolysis. For calcifediol that is feasible orally; for calcitriol the IV route may be required.
Cholecalciferol's PK makes it a suboptimal test article for acute studies, yet it dominates the published acute RCTs. This has produced a literature where positive signals (Hesami) are mechanistically implausible to be from rapid 25(OH)D rise, and may reflect non-genomic effects of D3, IM depot release, or chance. The Naderian calcitriol trial is the only one with appropriate PK but did not measure clinical outcomes.
Heterogeneity of ischemic stroke (lacunar vs. large-vessel; reperfused vs. not; cardioembolic vs. atherothrombotic) means a benefit may exist in a subset (e.g., large-vessel post-thrombectomy, where reperfusion injury and BBB disruption are maximal — the population most analogous to the rodent tMCAO model). No subgroup analysis exists.
Confounding by deficiency state. Most observational signal comes from severe deficiency (<10 ng/mL). The Wajda IRR of 2.5 may reflect frailty, comorbidity, or sarcopenia rather than the stroke-specific pathway. Only an RCT that randomizes within a narrow band (e.g., 10–20 ng/mL) and stratifies by 25(OH)D can disentangle.
Hypercalcemia and CYP24A1 induction. Acute calcitriol risks transient hypercalcemia; sustained calcifediol can also induce CYP24A1, which catabolizes both calcifediol and calcitriol — relevant for repeated dosing.
Commercial and regulatory inertia. Calcifediol (Hidroferol) is widely available in Spain, Italy, and Latin America but not FDA-approved as a generic drug in the US (Rayaldee, an extended-release formulation, is approved only for secondary hyperparathyroidism in CKD). This regulatory geography partly explains why the strongest acute trials (Castillo, Nogués) come from Spain, and why a US acute stroke RCT would have IND complexity.
No biomarker validation. Optimal serum 25(OH)D target in acute stroke is unknown. The c19early.org/Cordoba inference of "≥50 ng/mL" comes from COVID-19 cytokine storm logic, not stroke-specific data.
No human data on calcitriol adjunctive to tPA. The MMP-9 / hemorrhagic transformation rationale is mechanistically attractive but completely untested in humans.

Recommendations

Stage 1 — Immediate, low-risk clinical practice (where the evidence already supports action): - Measure 25(OH)D at admission in any acute ischemic stroke patient. Severe deficiency (<10 ng/mL) is independently associated with 2.5× mortality and is itself a clinical priority. - For severe deficiency, initiate calcifediol 0.266–0.532 mg PO on admission (or IM cholecalciferol 600,000 IU if calcifediol is not available locally), followed by structured outpatient repletion. This is supported by the Hesami RCT (acute), the broad observational literature, and the calcifediol-COVID paradigm. Threshold to change: a properly powered RCT showing no benefit, or evidence of harm, would supersede this.

Stage 2 — Hypothesis-driven adjunctive use in selected acute patients (research/expanded-access framework): - The strongest theoretical protocol — for testing, not for routine use — would be: - Calcifediol 0.5–1.0 mg PO at hospital admission (rapid 25(OH)D rise, 4-h pharmacodynamic onset, validated safe dose from COVID-19 trials). - Plus calcitriol 0.25 µg PO BID for 3 days (fastest VDR engagement, captures the genomic anti-inflammatory window between 6–72 h), with daily serum calcium and creatinine monitoring; hold if Ca >10.5 mg/dL. - Maintenance: calcifediol 0.266 mg on days 3 and 7, then weekly; transition to oral D3 ≥4000 IU/d once 25(OH)D is documented ≥40 ng/mL. - Patient selection: prioritize large-vessel occlusion patients (the LVO/MCAO clinical analog) undergoing thrombectomy or thrombolysis — the population with maximum reperfusion injury and BBB disruption, where mechanism most strongly predicts benefit. - This protocol should be considered hypothesis-generating and offered only in IRB-approved trial settings or as compassionate use after risk-benefit discussion. Threshold to expand: completion of a phase II RCT (n ≈ 200–300, primary endpoint 90-day mRS shift or DWI lesion-growth from admission to 24 h) with a positive effect size and no safety signal.

Stage 3 — Trial agenda (the field's actual unmet need): - A phase II randomized, double-blind, placebo-controlled trial of calcifediol 0.5 mg PO on admission + 0.266 mg on days 3 and 7 in acute ischemic stroke patients within 12 h of onset, baseline 25(OH)D ≤30 ng/mL, NIHSS 6–25. Primary endpoint: 90-day mRS ordinal shift. Secondary: DWI lesion growth from admission to 24/72 h, hemorrhagic transformation rate, NIHSS at 7 d, mortality. Stratify by reperfusion therapy (tPA, thrombectomy, neither) and baseline 25(OH)D quartile. Sample size ~400 for a 0.10 mRS shift detection at α=0.05, 80% power. Threshold to advance to phase III: mRS shift OR ≥1.30 with 95% CI excluding 1.0, or DWI lesion-growth reduction ≥30%, with no excess hemorrhage. - A separate mechanistic substudy comparing acute calcitriol vs. calcifediol vs. placebo on serum MMP-9, IL-6, TNF-α, and sICAM-1 trajectories at 0, 6, 24, 72 h would resolve which metabolite is preferable.

Stage 4 — Don't do this: - Do not rely on standard oral cholecalciferol (any dose, any route) for acute neuroprotection in the first 72 h. Its pharmacokinetics are incompatible with the acute window. Use it only for sub-acute repletion / long-term maintenance. - Do not use calcitriol monotherapy without calcium monitoring in any patient with CKD stage 3+ or pre-existing hypercalcemia.

Caveats

The clinical evidence base for acute fast-acting vitamin D in ischemic stroke is preclinical-dominant. Translating rodent calcitriol-1-µg/kg-at-30-min into a human protocol involves dosing, timing, and pharmacokinetic assumptions that have not been validated in stroke patients. The single human acute calcitriol study (Naderian 2023, n=38) is molecular-only.
The Hesami 2022 acute D3 RCT result is plausibly real but mechanistically puzzling. A 48-h NIHSS improvement after IM cholecalciferol pre-dates measurable 25(OH)D rise and may reflect (a) non-genomic D3 effects, (b) regression to the mean / small-sample chance, or (c) trial conduct artifacts. It should not be over-interpreted.
The Castillo/Nogués calcifediol-COVID effect sizes (OR 0.03–0.21) are exceptionally large for a hormone intervention and have attracted methodological criticism (non-blinded, allocation by ward, post-hoc adjustments). They are best read as a strong signal that calcifediol is biologically active in acute inflammatory critical illness — not as a quantitative prediction for stroke. The COVIDIOL multicenter RCT (NCT04366908) was registered to confirm but has not, to my knowledge, fully resolved the question.
Vitamin D and stroke prevention RCTs (VITAL, VIDA, D-Health) are uniformly null on stroke incidence — but these test cholecalciferol in non-deficient populations for prevention, not fast-acting metabolites in the acute window. They do not refute the acute hypothesis but underscore that vitamin D is not a pan-cardiovascular protectant in unselected populations.
Hypercalcemia, CYP24A1 induction, and drug interactions (especially in CKD and with thiazides) are real risks that scale with calcitriol potency. Any acute protocol must include calcium monitoring.
Geographic and assay variability in 25(OH)D measurement is substantial. The thresholds quoted here (<10, 10–20, 20–30, ≥30 ng/mL) assume LC-MS/MS or comparable; immunoassays vary by 10–20%.
The AHA/ASA acute stroke guidelines do not currently recommend any vitamin D intervention, and any protocol described above is research-grade or hypothesis-driven, not standard of care.
Source-quality note: Some interpretive material on calcifediol acute use comes from advocacy sites (e.g., vitamindstopscovid.info, c19early.org) which contain useful PK summaries but should not be treated as primary evidence; primary citations should be Castillo 2020 (JSBMB), Nogués 2021 (JCEM), Quesada-Gómez/Bouillon reviews, the Iranian acute stroke RCTs, the Atif/Stein and Tameh/Vahidinia preclinical series, and FDA/EMA labels for Rocaltrol and Hidroferol.

Related in VitaminDWiki

Forms of Vitamin D