Motivation: Right heart failure, contractile dysfunction and arrhythmia are associated with pulmonary hypertension (PH) induced in rats by monocrotaline (1-3). Voluntary wheel running can improve cardiac function and delay the onset of right heart failure in these animals (3). Although the mechanisms associated with this benefit are unknown, amelioration of cardiac electrophysiological and structural remodelling may play a part (1-3). Understanding the mechanisms by which exercise acts on PH may potentially reveal new therapeutic pathways for treatment. Methods: Male Wistar rats (180-190 g) were injected with 60 mg/kg monocrotaline to induce PH, or an equivalent volume of saline (control; CON). Animals were randomly assigned free access to a running wheel (exercise; EX) or not (sedentary; SED). PH-SED (n = 12) were sacrificed upon external signs of heart failure; CON-SED (n = 6), CON-EX (n = 11) and PH-EX ( n = 6) animals were sacrificed on matched days, unless heart failure developed earlier in PH-EX animals. Isolated hearts were Langendorff perfused, then action potentials and calcium transients were optically mapped with RH237 and Rhod2 respectively. Hearts were then fixed in 5% paraformaldehyde, and imaged using diffusion tensor MRI in a Bruker 9.4 T scanner at 200 µm isotropic resolution to quantify cardiac geometry and myocyte orientations. Animal-specific computational models based on these experimental data were used to asses vulnerability to arrhythmia (1,2,4). Results: Mean daily voluntary running distance was not significantly different between CON-EX and PH-EX (mean ± SEM: 3.1 ± 0.6 vs. 4.4 ± 1.1 km, p = 0.29, t-test). Voluntary exercise prolongs survival in PH: no PH-EX animals developed heart failure signs on the day of such signs in PH-SED animals (21 ± 0.26 days). PH significantly increased right ventricular action potential duration (APD90) at 5 Hz pacing (CON 63.3 ± 4.3 ms, PH 92.3 ± 4.6 ms, p = 0.0003, two-way ANOVA), but exercise had no effect (SED 77.1 ± 7.7 ms, EX 78.4 ± 5.6 ms, p = 0.52). There were no PH or exercise effects on calcium transient duration. Myocyte disarray, quantified by R2 from a linear fit to transmural myocyte elevation angles, was increased by PH in the right ventricular wall (CON 0.80 ± 0.05, PH 0.65 ± 0.02, p = 0.01) but was not significantly affected by exercise (SED 0.75 ± 0.04, EX 0.73 ± 0.05, p = 0.74). Exercise improved myocyte disarray in the left ventricular wall (SED 0.85 ± 0.02, EX 0.89 ± 0.05, p = 0.04). Computational simulations show that structural (myocyte orientation) remodelling underpins instability of re-entrant arrhythmias to a significantly greater extent than electrophysiological (APD90) remodelling in PH (p = 0.004, one-way ANOVA). Conclusion: These data suggest that exercise may be a suitable therapeutic intervention to prevent right heart failure and arrhythmia in PH.