Objective: To explore the fluid resuscitation strategy in shock stage in severely burned children with different burn areas in different age groups, and to evaluate the curative effect. Methods: A retrospective cohort study was conducted. From January 2015 to June 2020, 235 children with severe and above burns who met the inclusion criteria were hospitalized in the First Affiliated Hospital of Nanchang University, including 150 males and 85 females, aged 3 months to 12 years. After admission, it was planned to rehydrate the children with electrolyte, colloid, and water according to the domestic rehydration formula for pediatric burn shock, and the rehydration volume and speed were adjusted according to the children's mental state, peripheral circulation, heart rate, blood pressure, and urine output, etc. The actual input volume and planned input volume of electrolyte, colloid, water, and total fluid of all the children were recorded during the 8 hours since fluid replacement and the first and second 24 hours after injury. According to urine output during the 8 hours since fluid replacement, all the children were divided into satisfactory urine output maintenance group (119 cases) with urine output ≥1 mL·kg -1 ·h -1 and unsatisfactory urine output maintenance group (116 cases) with urine output <1 mL·kg -1 ·h -1 , and the electrolyte coefficient, colloid coefficient, and water coefficient of the children were calculated during the 8 hours since fluid replacement. According to the total burn area, children aged <3 years (155 cases) and 3-12 years (80 cases) were divided into 15%-25% total body surface area (TBSA) group and >25%TBSA group, respectively. The electrolyte coefficient, colloid coefficient, water coefficient, and urine output of the children were calculated or counted during the first and second 24 hours after injury, and the non-invasive monitoring indicators of body temperature, heart rate, respiratory rate, and percutaneous arterial oxygen saturation and efficacy indicators of hematocrit, platelet count, hemoglobin, albumin, creatinine, and alanine aminotransferase (ALT) of the children were recorded 48 hours after injury. The prognosis and outcome indicators of all the children during the treatment were counted, including complications, cure, improvement and discharge, automatic discharge, and death. Data were statistically analyzed with independent sample or paired sample t test, Mann-Whitney U test, chi-square test, and Fisher's exact probability test. Results: During the 8 hours since fluid replacement, the actual input volume of electrolyte of all the children was significantly more than the planned input volume, and the actual input volumes of colloid, water, and total fluid were significantly less than the planned input volumes ( Z =13.094, 5.096, 13.256, 7.742, P <0.01). During the first and second 24 hours after injury, the actual input volumes of electrolyte of all the children were significantly more than the planned input volumes, and the actual input volumes of water and total fluid were significantly less than the planned input volumes ( Z =13.288, -13.252, 3.867, 13.183, -13.191, 10.091, P <0.01), while the actual input volumes of colloid were close to the planned input volumes ( P >0.05). During the 8 hours since fluid replacement, compared with those in unsatisfactory urine output maintenance group, there was no significant change in electrolyte coefficient or colloid coefficient of children in satisfactory urine output maintenance group ( P >0.05), while the water coefficient was significantly increased ( Z =2.574, P <0.05). Among children <3 years old, compared with those in >25%TBSA group, the electrolyte coefficient and water coefficient of children were significantly increased and the urine output of children was significantly decreased in 15%-25%TBSA group during the first and second 24 hours after injury ( Z =-3.867, -6.993, -3.417, -5.396, -5.062, 1.503, P <0.05 or P <0.01), while the colloid coefficient did not change significantly ( P >0.05); the levels of efficacy indicators of hematocrit, platelet count, and hemoglobin at 48 h after injury were significantly increased, while ALT level was significantly decreased ( Z =-2.720, -3.099, -2.063, -2.481, P <0.05 or P <0.01); the levels of the rest of the efficacy indicators and non-invasive monitoring indicators at 48 h after injury did not change significantly ( P >0.05). Among children aged 3-12 years, compared with those in >25%TBSA group, the electrolyte coefficient and water coefficient of children in 15%-25%TBSA group were significantly increased during the first and second 24 hours after injury, the colloid coefficient during the second 24 h was significantly decreased ( Z =-2.042, -4.884, -2.297, -3.448, -2.480, P <0.05 or P <0.01), while the colloid coefficient during the first 24 hours after injury, urine output during the first and second 24 hours after injury, and the non-invasive monitoring indicators and efficacy indicators at 48 hours after injury did not change significantly ( P >0.05). Complications occurred in 17 children during the treatment. Among the 235 children, 211 cases were cured, accounting for 89.79%, 5 cases were improved and discharged, accounting for 2.13%, 16 cases were discharged automatically, accounting for 6.81%, and 3 cases died, accounting for 1.28%. Conclusions: The electrolyte volume in early fluid resuscitation in severely burned children exceeding the volume calculated by the formula can obtain a good therapeutic effect. Among children <3 years old, the volume of fluid resuscitation should be appropriately increased in children with extremely severe burns compared with children with severe burns during fluid resuscitation; among children aged 3-12 years, the colloid volume should be appropriately increased in children with extremely severe burns compared with children with severe burns during fluid resuscitation; non-invasive monitoring indicators can be used to monitor hemodynamics and guide fluid resuscitation in severely burned children.