AIMS Heart failure (HF) patients breathe with a rapid shallow pattern during exercise. This study examined the relationship between cardiac size and tachypnoeic breathing in HF patients during exercise. METHODS AND RESULTS Thirty-seven HF patients [age = 55 ± 13 years, ejection fraction (EF) = 27 ± 10%, New York Heart Association (NYHA) class = 2.3 ± 1.2] and 42 controls (CTL) (age = 56 ± 14 years, EF = 63 ± 8%) were recruited. Participants underwent maximal exercise testing, pulmonary function testing, and chest radiography for calculation of total thoracic cavity volume (TTCV), diaphragm, heart, and lung volumes. Heart failure patients were divided into two groups: Group A = cardiac volume < median (n = 18) and Group B = cardiac volume ≥ median of the HF patients (n = 19). There was no difference between groups for TTCV (CTL = 8203 ± 1489 vs. Group A = 8694 ± 1249 vs. Group B = 8195 ± 1823 cm(3)). Cardiac volume was different between groups for both absolute (CTL = 630 ± 181 vs. Group A = 894 ± 186 vs. Group B = 1401 ± 382 cm(3), P< 0.001 for all comparisons) and %TTCV (CTL = 8 ± 2 vs. Group A = 10 ± 1 vs. Group A = 18 ± 5%, P< 0.001 for all comparisons). Similarly, total lung volume as a %TTCV was significantly different among the groups (CTL = 70 ± 4 vs. Group A = 65 ± 5 vs. Group A = 58 ± 7%, P< 0.01 for all comparisons). In HF patients, there was a trend (P = 0.10) towards an independent association between cardiac size and tidal volume (V(T)) at 75% of VO(2) peak whereas this relationship was statistically significant at VO(2) peak (P = 0.02) as patients with larger cardiac size had reduced V(T). CONCLUSION This study demonstrates the close relationship between cardiac size and breathing pattern during exercise in HF patients. These results suggest cardiac size may pose a significant constraint on the lungs during exercise and may contribute to tachypnoeic breathing.