Abstract: Objective To retrospectively analyze the risk factors and clinical manifestations of myocardial damage of patients with severe burn in order to provide evidence for its prevention and treatment. Methods Two hundred and fifty-two patients with severe burn admitted to 5 burn centers from January 2010 to June 2015, conforming to the study criteria, were treated in accordance with the fluid resuscitation formula of the Third Military Medical University. According to the creatine kinase isoenzyme-MB (CK-MB) level before treatment on admission, patients were divided into non-myocardial damage group (n=118, CK-MB level less than 75 U/mL) and myocardial damage group (n=134, CK-MB level higher than or equal to 75 U/mL). Data of patients in two groups were collected and evaluated such as gender, age, body mass, number of patients with chemical burn, admission time after injury, total burn area, full-thickness burn area, number of patients with inhalation injury, levels of haemoglobin, hematocrit, and blood lactate on admission and at post injury hour (PIH) 24 and 48, volumes of urine output and fluid input at PIH 24 and 48, levels of creatinine, urea nitrogen, total bile acid, diamine oxidase on admission and at PIH 24 and 48, and mortality. Furthermore, patients were divided into three groups, i. e. less than 50% total body surface area (TBSA) group (n=110), larger than or equal to 50% TBSA and less than 80% TBSA group (n=83), and larger than or equal to 80% TBSA group (n=59) according to the total burn area, and the incidence rates of myocardial damage in patients of three groups were recorded. Data were processed with chi-square test, t test, Wilcoxon test, analysis of variance for repeated measurement, and the values of P were adjusted by Bonferroni. Basic data of 252 patients were processed with binary logistic regression analysis. Receiver operating characteristic curve of total burn area of 252 patients was drawn to predict myocardial damage. Results (1) There were no statistically significant differences in age, body mass, number of patients with chemical burn, number of patients with inhalation injury, and full-thickness burn area between two groups (with t values respectively 0.20 and 0.31, χ² values respectively 0.49 and 4.10, Z=1.42, P values above 0.05). There were statistically significant differences in gender, admission time after injury, and total burn area of patients between two groups (χ²=5.00, with t values respectively 2.44 and 3.13, P<0.05 or P<0.01). (2) Gender, admission time after injury, and total burn area were independent risk factors related to myocardial damage in the patients (with odds ratios respectively 2.608, 3.620, and 1.030; 95% confidence intervals respectively 1.315-5.175, 1.916-6.839, and 1.011-1.049; P values below 0.01). (3) The incidence rates of myocardial damage of patients in less than 50% TBSA group, larger than or equal to 50% TBSA and less than 80% TBSA group, and larger than or equal to 80% TBSA group were 38.2% (42/110), 54.2% (45/83), and 61.0% (36/59) respectively, and there was statistically significant difference among them (χ²=9.46, P<0.05). (4) The total area under receiver operating characteristic curve of total burn area to predict myocardial damage of 252 patients was 0.706 (with 95% confidence interval 0.641-0.772, P<0.01), and 51.5% TBSA was chosen as the optimal threshold value, with sensitivity of 62.6% and specificity of 65.3%. (5) Compared with those in non-myocardial damage group, except the levels of haemoglobin and hematocrit at PIH 48 (with t values respectively -0.76 and -0.61, P values above 0.05), the levels of haemoglobin, hematocrit, and blood lactate of patients in myocardial damage group were significantly increased at each time point (with t values from -2.80 to -2.06, P<0.05 or P<0.01). Compared with those in non-myocardial damage group, the volume of urine output of patients was significantly declined (with t values respectively 2.05 and 3.68, P<0.05 or P<0.01), while the volume of fluid input of patients was not obviously changed in myocardial damage group at PIH 24 and 48 (with t values respectively 1.01 and 1.08, P values above 0.05). (6) Compared with those in non-myocardial damage group, the level of creatinine of patients was significantly increased on admission and at PIH 24 and 48 (with Z values from -2.91 to -1.99, P<0.05 or P<0.01), the level of urea nitrogen of patients was only significantly increased at PIH 24 and 48 (with t values respectively -4.75 and -5.24, P values below 0.01), the level of total bile acid of patients was not obviously changed on admission and at PIH 24 and 48 (with t values from -0.81 to -0.20, P values above 0.05), and the level of diamine oxidase of patients was only significantly increased on admission and PIH 24 in myocardial damage group (with t values respectively -3.97 and -2.02, P<0.05 or P<0.01). (7) Compared with that in myocardial damage group, the mortality of patients in non-myocardial damage group was significantly declined (χ²=5.81, P<0.05). Conclusions Patients with severe burn have high incidence of myocardial damage, which may be predicted by total burn area. Severely burned patients with myocardial damage are more likely to suffer from decline of effective circulating volume, tissue oxygenation disorders, and damage in other organs in shock stage.