AUTHORS
Emma Carpenter, N/A, Clio González-Zacarías, PhD2, Silvie Suriany3, Alisha Memon4, Sharon O'Neil, PhD, MHA, ABPP-CN5, Dawn Ward, MD6, Alyssa Ziman, MD7, Ellen Klapper, MD8, Lefan Zhuang, MD9, Christopher Clement Denton, MD, MS10, John Wood I, MD, PhD11
AFFILIATIONS
1 Children's Hospital Los Angeles, Los Angeles, California, United States
2 Children's Hospital Los Angeles, Los Angeles, California, United States
3 Children's Hospital Los Angeles, Los Angeles, California, United States
4 Children's Hospital Los Angeles, Los Angeles, California, United States
5 Children's Hospital Los Angeles, Los Angeles, California, United States
6 David Geffen School of Medicine at UCLA, Los Angeles, California, United States
7 David Geffen School of Medicine at UCLA, Los Angeles, California, United States
8 Cedars-Sinai Medical Center, Los Angeles, California, United States
9 City of Hope, Los Angeles, California, United States
10 Children's Hospital Los Angeles, Los Angeles, California, United States
11 Children's Hospital Los Angeles, Los Angeles, California, United States
ABSTRACT
Introduction
Anemia lowers arterial oxygen content (CaO2), and cerebral blood flow (CBF) is expected to rise to defend brain oxygen delivery. Whether this compensation is complete in iron deficiency anemia (IDA), the most common anemia worldwide, is unknown, as is whether it reverses with iron therapy and on what timescale.
Methods
Design: NCT05929729 is a randomized, open-label, parallel Phase 4 trial at Children’s Hospital Los Angeles comparing intravenous versus standard-of-care oral iron in adult women with IDA, of which we report a interim analysis of 51 women imaged of a planned 120.
Participants: 34 women with IDA and 17 healthy controls; female only, aged 16–60 years by eligibility. IDA 36.7 ± 11.1 y, Hb 9.69 ± 1.29 g/dL; controls 31.9 ± 9.6 y, Hb 13.41 ± 0.57 g/dL.
Measurements: CBF was measured by phase-contrast MRI and normalized to brain volume (nCBF).¹?² OEF and venous saturation were measured by T2-relaxation-under-spin-tagging (TRUST).³ Cerebral oxygen delivery was derived once as nCBF × CaO2, with CaO2 = 1.36 × Hb × SaO2. Fatigue was scored with the 13-item FACIT-Fatigue scale (0–52, higher = less fatigue; MCID 3).4
Statistical analysis: Group comparisons were adjusted for age with false-discovery correction across 8 imaging measures. Longitudinal change was tested by Wilcoxon signed-rank on within-subject pairs.
Follow-up: Imaging was obtained at 2 weeks (IV arm only, n = 12) and at 3 months (IV + oral pooled, n = 19); pooling was justified by equivalent Hb repletion by route (+2.03 vs +2.05 g/dL) and addresses the effect of iron repletion rather than of assigned route.
Results
Compensatory hyperemia was present but graded. nCBF was 64.4 ± 11.3 in IDA versus 56.7 ± 14.7 mL/100g/min in controls (adjusted β +7.8, p = 0.095), scaled negatively with Hb across subjects (−2.52 mL/100g/min per g/dL, p = 0.022, n = 37) and more tightly within-person over treatment (−5.2 per g/dL, r = −0.72, p = 0.002, n = 16). Despite this rise, oxygen delivery remained ≈18% below controls (8.34 ± 1.54 vs 10.15 ± 2.38 mL O2/100g/min; adjusted β −1.72, 95% CI −3.09 to −0.36, p = 0.015): the CBF response could in principle have compensated fully but did not. OEF did not participate: 0.449 ± 0.054 versus 0.434 ± 0.067 (adjusted β +0.012, p = 0.58), with no dependence on Hb (p = 0.95), indicating that compensation operates through CBF alone, consistent with mathematical modeling of reduced regulatory capacity in anemia.5 Fatigue tracked hemoglobin (FACIT +3.43 points per g/dL, 95% CI 2.08–4.78, p = 5.5×10?6, n = 51; >1 MCID per g/dL), scaling with red cell mass rather than with the cerebral flow response (nCBF adjusted for Hb: p = 0.63).
At two weeks (IV arm), biochemical iron stores were fully replete (ferritin 5.6 → 371 ng/mL, TSAT 7.3 → 53%, both p = 0.002, n = 9), yet nCBF was unchanged (64.9 → 61.7, p = 0.43) and OEF was unchanged (p = 0.44, n = 6). Hb had barely moved (+0.56 g/dL, p = 0.052), and oxygen delivery therefore did not change (+0.09, p = 0.82). By three months, Hb had risen by 2.04 g/dL (p = 0.001, n = 19); the hyperemia resolved in proportion to each participant’s own Hb increment along the same −5.2 mL/100g/min per g/dL slope; the group-mean nCBF change did not reach significance (−2.7, p = 0.46); and oxygen delivery moved from 8.60 into the control interquartile range (9.73, +1.13, p = 0.003, n = 16). An exploratory OEF fall (0.469 → 0.433, p = 0.021, n = 12) was observed but should be read cautiously given no baseline group difference and n = 12.
Conclusion
In adult IDA, cerebral compensation relies on CBF alone: the brain raises flow but does not recruit oxygen extraction, and delivery therefore remains subnormal. At two weeks, once iron stores are already replete, hemodynamics do not respond, hemoglobin has not yet risen and thus delivery is necessarily unchanged. By three months, as Hb repletes, hyperemia resolves proportionally and delivery returns to the control range. Cerebral hemodynamics in IDA track red cell mass, not iron availability as such, with implications for counseling on when patients should expect neurologic benefit from iron therapy.
DISCLOSURES
Emma Carpenter, N/A: No relevant conflicts of interest to declare.