Objective: To analyze the clinical effect of extracorporeal membrane oxygenation (ECMO) in the treatment of burn patients with acute respiratory distress syndrome (ARDS). Methods: The retrospective observational study and the systematic review were applied. From March 2014 to July 2020, five burn patients with ARDS received ECMO treatment in the First Affiliated Hospital of Army Medical University (the Third Military Medical University). All the five patients were male, aged from 40 to 62 years. The average total burn surface area was 58.8% total body surface area (TBSA) and four cases had severe inhalation injury. Patient's ECMO starting time, duration and mode, and whether successfully weaned or the cause of death, and others. were recorded. Furthermore, the changes of oxygenation and infection before, during, and after utilizing ECMO were analyzed. PubMed and Web of Science from the establishment of each database to August 2021 were searched using "Extracorporeal Membrane Oxygenation", "ECMO", "burn", "inhalation" as the search terms and "Title/Abstract" as the field to retrieve the clinical articles that meet the selection criteria . Basic information were extracted from the articles, including sample size, gender, age, total burn area, inhalation injury, the indication of ECMO, the start and lasting time of ECMO, ECMO mode, rate of successful weaning, complications of ECMO, mortality, the combined application of continuous renal replacement therapy (CRRT). Results: Five patients started venovenous ECMO on an average of 10.2 days after injury and lasted an average of 180.4 hours. Three out of 5 patients were weaned successfully with one patient survived. Four patients died of multiple organ dysfunction syndrome (MODS) and septic shock. Compared with those before ECMO treatment, the arterial oxygen partial pressure (PaO2) and oxygen saturation in arterial blood (SaO2) of three successfully weaned patients obviously increased during and after ECMO treatment. The fraction of inspired oxygen (FiO2) decreased below 50% and PaO2/FiO2 ratio increased above 200 mmHg (1 mmHg=0.133 kPa) during and after ECMO. Furthermore, lactic acid and respiratory rate decreased, basically. Compared with those before ECMO, PaO2 and SaO2 in the other two patients during ECMO, who failed to be weaned, continuously decreased while lactic acid increased. Before and during ECMO, the PaO2/FiO2 ratios of unsuccessfullg weaned cases were less than 200 mmHg, and partial pressure of carbon dioxide in arterial blood (PaCO2) were more than 40 mmHg. Compared with those before ECMO, there were no significant changes in body temperature during and after ECMO, which were less than 38 ℃. Compared with those before ECMO, the leucocyte number (the index without this in unsuccessfully weaned cases was omitted, the same as below) in four patients showed a significant decrease during ECMO, but rose after removal of ECMO. The proportion of neutrophils in three patients were slightly higher during ECMO than before ECMO, and did not change significantly after removal of ECMO. Compared with those before ECMO, platelet counts in three patients were significantly reduced during ECMO, and all five patients during ECMO were below normal levels. Compared with those before ECMO, the procalcitonin levels in four deaths were significantly increased during ECMO. Catheter culture of microorganism was performed in three successfully weaned patients, all of which were negative. A total of 13 literature were included, ranging from 1990 to 2019. The sample size in 6 studies was less than 10, and the sample size in 4 studies was between 10 and 20, and only 2 literatures had a sample size larger than 50. ECMO was applied in 295 burn patients with overall mortality of 48.8% (144/295), including 157 adults and 138 children. The most common indication of ECMO was severe ARDS. Among 157 adult burn patients (95 males and 65 females), 36 cases had inhalation injury. The average burn area was 27%-37%TBSA in 5 reported studies and was more than 50%TBSA in 2 reported studies. The most common mode was venovenous ECMO. ECMO treatment began 26.5 hours to 7.4 days after injury and lasted from 90 hours to 18 days, and the rate of successful weaning ranged from 50% to 100%. The most common complications were bleeding and infection. The mortality was 52.9% (83/157). MODS and sepsis were the leading causes of death. Among 138 pediatric burn patients (77 boys and 61 girls), 29 patients had inhalation injury. The average burn area was 17%-50.2%TBSA in 3 studies. ECMO treatment lasted from 165.2 hours to 324.4 hours. Bleeding was the most common complication. The mortality was 44.2% (61/138). Conclusions: ECMO is an effective strategy for the salvage treatment of burns complicated with ARDS. Furthermore, the prevention and treatment of bleeding, infection and organ dysfunction should be emphasized during the use of ECMO. More importantly, evidence-based guidelines for burns are urgently needed to further improve the clinical effect of ECMO.