Background Skeletal muscle dysfunction is a major contributor to morbidity in chronic cardiac and respiratory diseases. Iron deficiency is deleterious in these conditions but the underlying mechanisms are poorly understood. Animal studies suggest a direct effect of iron deficiency on skeletal muscle but human studies are inconsistent. We hypothesised that iron deficiency would disturb human skeletal muscle metabolism, and that intravenous iron supplementation would have beneficial physiological effects. Methods We performed a prospective, double-blind, randomised, controlled, clinical physiology study of the effects of iron status on human skeletal muscle metabolism. Otherwise healthy individuals with absolute iron deficiency (ID), and an iron-replete (IR) control group, underwent 31P magnetic resonance spectroscopy (31P-MRS) of exercising calf muscle followed by cardiopulmonary exercise testing (CPET). CPET included both cycle ergometer exercise to volitional fatigue and a subsequent submaximal period of exercise at 65% VO2max. Participants were randomised to receive intravenous iron (ferric carboxymaltose, 15 mg/kg, maximum 1g) or saline at the end of the first study morning. Identical assessments were performed around a week later. Mixed-effects modelling was used to examine the physiological effects of iron status and iron supplementation. Results Thirteen ID individuals completed the study and were matched with controls. Baseline group characteristics were similar. VO2max did not differ significantly between groups (ID 38.5 v. IR 40.5 ml/kg/min; 95% CI for difference -0.9 to 4.5; P = NS). Despite similar venous blood lactate concentrations at volitional fatigue, net lactate clearance during submaximal exercise was impaired in the ID group (ID 4.8 v. IR 7.5 mmol/L/h; 95% CI for difference 0.7-4.7; P < 0.05). A significant interaction between baseline iron status and the effect of intravenous iron on lactate clearance was detected (P < 0.05). Additionally, intravenous iron significantly raised the lactate threshold irrespective of baseline iron status (P < 0.05). 31P-MRS revealed more marked intracellular acidosis during small muscle mass exercise in the ID group, despite similar phosphocreatine kinetics. Conclusion This study provides clear evidence of an effect of iron bioavailability on human metabolism. Iron deficiency appears to promote anaerobic glycolysis, as evidenced by abnormal lactate kinetics during whole body exercise, and myocyte acidosis during light small muscle mass exercise. Administration of intravenous iron corrects this to an extent, and appears to have an effect even in iron-replete individuals. These observations have implications for the pathophysiology of skeletal muscle dysfunction in chronic cardiopulmonary diseases, as well as for human athletic performance.