Abstract. We hypothesized that myocardial stunning would be reversed through increased cyclic GMP caused by nitroprusside, and that this would be accomplished through a decreased proportion of regional work during diastole. Hearts were instrumented to measure left ventricular pressure, and regional myocardial mechanics were recorded using a miniature force transducer and ultrasonic dimension crystals in eight open‐chest anesthetized dogs. Following baseline (CON), the left anterior descending coronary artery (LAD) was occluded for 15 min, followed by a 30‐min recovery (STUN). Then intracoronary LAD infusion of sodium nitroprusside (NP) (4 μg/kg/min) was begun. The time delay (msec) to regional shortening increased significantly from 18 ± 13 to 73 ± 13 following stunning, but was reduced to 49 ± 18 by NP. Total regional work (g*mm/min) at baseline (1368 ± 401 CON) was unchanged with stunning (1320 ± 333 STUN), but reduced (961 ± 240) following NP. Time to peak force development (msec) increased significantly with stunning from 284 ± 13 (CON) to 333 ± 11 (STUN), but was reduced to 269 ± 12 following NP. The percentage work during systole was reduced from 96%± 2% (CON) to 77%± 7% (STUN), but returned to 98%± 1% with NP. Regional O2 consumption was unaffected by either treatment. Cyclic GMP was unchanged by stunning (2.9 ± 0.3–2.9 ± 0.4 pmol/g) but increased significantly with NP (4.6 ± 0.6). These data indicated that regional myocardial stunning could be attenuated by nitroprusside, which increased cyclic GMP, decreased contractile delay, increased the proportion of work done during systole, and reduced time of shortening.