Objective: To investigate the effects of dexmedetomidine on myocardium of rats at early stage after severe burn. Methods: Twenty specific pathogen free male SD rats were immersed in 90 ℃ hot water for 20 s, causing 30% total body surface area (TBSA) full-thickness scald (hereafter referred to as burn) on the back. And then they were divided into burn resuscitation group (BR) and burn resuscitation+ dexmedetomidine group (BRD) according to the random number table, with 10 rats in each group. Sodium lactate Ringer's solution (2 mL·kg(-1)·%TBSA(-1)) were intraperitoneally injected into rats of both groups after burn. Dexmedetomidine with dose of 1 μg/kg was intraperitoneally injected into rats of group BRD at the same time point. Another 5 rats in sham injury group (SI) were immersed in 37 ℃ water bath causing sham injury, and fluid resuscitation of rats in group SI was the same as that in group BR. Five rats of group BR and BRD were respectively selected at post burn hour (PBH) 6 and 24. And then left ventricular end-systolic internal diameter (LVIDs), left ventricular end-diastolic internal diameter (LVIDd), ejection fraction (EF), and cardiac output (CO) were determined with small animal ultrasonic imaging system. Plasma levels of cardiac troponin (cTn) I and cTnT were detected by enzyme-linked immunosorbent assay, and morphological changes of myocardium were observed under optical microscope and transmission electron microscope (observed only at PBH 24). In rats of group SI, morphological change of myocardium was observed at PBH 24, and the other indexes were detected as above. Data were processed with one-way analysis of variance and SNK test. Results: At PBH 6, EF value of rats in group BR [(98.0±2.8) %] was obviously higher than that in group SI [(91.0±0.4)%, P<0.05]. The other 3 cardiac ultrasound indexes of rats in group BR were close to those in group SI (with P values above 0.05). Each cardiac ultrasound index of rats between groups BRD and SI was close at PBH 6 (with P values above 0.05). At PBH 24, LVIDs levels of rats in group BR [(0.66±0.59) mm] and group BRD[(0.69±0.27) mm] were obviously lower than LVIDs level of rats in group SI [(1.65±0.33) mm, with P values below 0.05]. LVIDd, EF, and CO levels of rats were close among 3 groups at PBH 24 (with P values above 0.05). At PBH 6, the plasma levels of cTnI [(17.40±1.59) ng/mL] and cTnT [(1 488±229) pg/mL] of rats in group BR were significantly higher than those in group SI [(1.84±0.92) ng/mL and (169±12) pg/mL, with P values below 0.01]. At PBH 6 in group BRD, the plasma level of cTnI of rats [(2.58±0.60) ng/mL] was close to that in group SI (P>0.05), and the plasma level of cTnT [(649±190) pg/mL] was higher than that in group SI (P<0.01). At PBH 24, the plasma levels of cTnI and cTnT of rats in group SI were close to those in groups BR and BRD (with P values above 0.05). At PBH 24, the plasma level of cTnI of rats in group BRD was obviously lower than that in group BR (P<0.01). At PBH 6, the myocardial structures of rats in group BR and group BRD were normal, which were close to myocardial structure of rats in group SI at PBH 24. At PBH 24, obviously damaged myocardial tissue, disorderly arrangement of myofilament, and seriously damaged mitochondria were observed in rats of group BR, which were significantly ameliorated in rats of group BRD. Conclusions: Dexmedetomidine can protect the myocardium of rats with severe burn at early stage.