REF-1413 · Burden et al., 2015 · British Journal of Sports Medicine · Paper · PubMed

Does treating low iron without anaemia boost endurance?

This meta-analysis of 17 studies in endurance athletes says VO2max rose (g 0.61); a later one of oral-iron RCTs found a non-significant trend (REF-1409). Possible, not proven.

Main caveat: Abstract only. Designs, number of participants, routes, doses, deficiency definitions, heterogeneity and risk of bias are unknown from what was read.
fairly strong evidenceMeasured on trained to elite athletes

Burden and colleagues asked whether iron treatment improves iron status and aerobic capacity in endurance athletes who are iron-deficient but not anaemic.

Pooling 17 studies, they found large effects of iron on serum ferritin (Hedges' g 1.088), serum iron (1.004) and transferrin saturation (0.741), and moderate effects on haemoglobin concentration (0.695) and VO2max (0.610, 95% CI 0.399 to 0.821), each with p < 0.001.

Regression showed that the effect on ferritin depended on treatment duration, with treatments beyond 80 days having the least effect. The authors conclude that iron treatment improves the iron status and aerobic capacity of these athletes. Our reading: this is the "yes" side of the performance question the base has carried in PHYS-36 without an entry.

The later meta-analysis restricted to oral-iron RCTs (REF-1409: 13 trials, mostly young women of any iron status) found ferritin up but VO2max only a non-significant trend (SMD 0.49), and REF-1412 (iron-deficient adults, any route) found no VO2max gain (SMD 0.11).

The disagreement is about which studies were pooled and how depleted the athletes were, more than about iron's effect on stores.

Protocol

Design: meta-analysis of studies that measured the effect of iron treatment on serum ferritin, serum iron, transferrin saturation, haemoglobin concentration and VO2max in iron-deficient non-anaemic endurance athletes; 17 eligible studies from online databases. Search dates, inclusion criteria, designs, routes, doses and total participants are not given in the abstract.
Analysis: pooled Hedges' g with 95% CIs; regression of the ferritin effect on treatment duration.
Population: iron-deficient non-anaemic endurance athletes; n = not stated in the abstract.
Funding and interests: PubMed lists a UK Department of Health grant (10/104/06); a conflict-of-interest statement is not in the abstract record.

Key Results

OutcomeHedges' g95% CIpAuthors' label
Serum ferritin1.0880.914 to 1.263<0.001large
Serum iron1.0040.828 to 1.181<0.001large
Transferrin saturation0.7410.564 to 0.919<0.001large
Haemoglobin concentration0.6950.533 to 0.836<0.001moderate
VO2max0.6100.399 to 0.821<0.001moderate
Duration: a significant interaction between the ferritin effect and treatment duration; treatments lasting beyond 80 days appeared to have the least effect on ferritin.
Not in the abstract: heterogeneity, sensitivity analyses, risk of bias, effects by route, sex or baseline ferritin, and any performance outcome other than VO2max.

Mechanisms

Measured (pooled): iron raised stores and transport markers and, moderately, haemoglobin concentration and VO2max in athletes classed as non-anaemic.
Authors' interpretation: iron treatment improves iron status and aerobic capacity in this group.
Inference: a moderate haemoglobin effect in athletes defined as non-anaemic suggests some were near the anaemia line, which would favour a VO2max gain; REF-1409 adds that pooling intramuscular with oral iron inflates the ferritin effect; neither can be checked from this abstract.
Inference: that courses beyond 80 days had the least ferritin effect is more likely a difference between the studies with long courses than a sign that iron stops working; the abstract does not say.

Practical Application

Does treating low iron without anaemia boost endurance? This meta-analysis of 17 studies in endurance athletes says VO2max rose (g 0.61); a later one of oral-iron RCTs found a non-significant trend (REF-1409). Possible, not proven.
State the disagreement, not a verdict: to the athlete asking whether fixing their iron will make them faster, the base now holds this "yes" (A-, from the abstract), REF-1409's "not proven" (A-, oral RCTs only), REF-1412's fatigue-yes, capacity-no (A) and REF-1420's within-group gains in female athletes (B+). A gain is most plausible when stores are clearly depleted; the base can say so with these sources.
What it does not license: iron for athletes with normal ferritin (only iron-deficient athletes were pooled) or a promise of a faster race. Not from this paper; the base's own practice: iron only after a blood test, prescribed by a physician (SYNTH-06).

Limitations

Abstract only. Designs, number of participants, routes, doses, deficiency definitions, heterogeneity and risk of bias are unknown from what was read; the full text was not open.
Mixed routes: per REF-1409, it combined oral iron with intramuscular injections, which had the largest effects on ferritin.
Surrogate outcome. VO2max is the only capacity measure in the abstract; no race or time-trial result is reported.
Unexplained duration finding (least ferritin effect beyond 80 days).
Funding and interests: a UK Department of Health grant per PubMed; the authors' interests are not in the abstract record. The record prints no competing-interest statement, so none is reported here; interests printed in later papers by the same authors are not this paper's.
How this entry was checked and changed
Why this level

(from the abstract) — a meta-analysis (rubric A) held at A- because the abstract gives neither the designs of the 17 studies, their sizes nor their heterogeneity, and REF-1409 reports that it combined oral iron with intramuscular injections, the latter giving the largest ferritin effects