BackgroundIron deficiency (ID) is frequent and associated with diminished exercise capacity in heart failure (HF), but its contribution to unexplained dyspnea without a HF diagnosis at rest remains unclear. Methods and ResultsConsecutive patients with unexplained dyspnea and normal echocardiography and pulmonary function tests at rest underwent prospective standardized cardiopulmonary exercise testing with echocardiography in a tertiary care dyspnea clinic. ID was defined as ferritin of <300 µg/L and a transferrin saturation of <20% and its impact on peak oxygen uptake (peakVO2), biventricular response to exercise, and peripheral oxygen extraction was assessed. Of 272 patients who underwent cardiopulmonary exercise testing with echocardiography, 63 (23%) had ID. For a similar respiratory exchange ratio, patients with ID had lower peakVO2 (14.6 ± 7.6 mL/kg/minvs 17.8 ± 8.8 mL/kg/min; P = .009) and maximal workload (89 ± 50 watt vs 108 ± 56 watt P = .047), even after adjustment for the presence of anemia. At rest, patients with ID had a similar left ventricular and right ventricular (RV) contractile function. During exercise, patients with ID had lower cardiac output reserve (P < .05) and depressed RV function by tricuspid s' (P = .004), tricuspid annular plane systolic excursion (P = .034), and RV end-systolic pressure-area ratio (P = .038), with more RV–pulmonary artery uncoupling measured by tricuspid annular plane systolic excursion/systolic pulmonary arterial pressure ratio (P = .023). RV end-systolic pressure-area ratio change from rest to peak exercise, as a load-insensitive metric of RV contractility, was lower in patients with ID (2.09 ± 0.72 mm Hg/cm2 vs 2.58 ± 1.14 mm Hg/cm2; P < .001). ID was associated with impaired peripheral oxygen extraction (peakVO2/peak cardiac output; P = .036). Cardiopulmonary exercise testing with echocardiography resulted in a diagnosis of HF with preserved ejection fraction in 71 patients (26%) based on an exercise E/e' ratio of >14, with equal distribution in patients with (28.6%) or without ID (25.4%, P = .611). None of these findings were influenced in a sensitivity analysis adjusted for a final diagnosis of HFpEF as etiology for the unexplained dyspnea. ConclusionsIn patients with unexplained dyspnea without clear HF at rest, ID is common and associated with decreased exercise capacity, diminished biventricular contractile reserve, and decreased peripheral oxygen extraction.