Introduction: Increasing the hematocrit is considered to increase oxygen delivery to the patient, especially when hypoxic conditions exist and the patient may become more stable. The aim of this study was to evaluate the relationship between hematocrit and hospital mortality via subgroup analyses of trauma and non-trauma patients. Methods: The hospital length of stay (LOS) and LOS in the intensive care unit (ICU) and hospital after extracorporeal life support (ECLS) treatment of 81 patients were analyzed and compared. In-hospital survival until extracorporeal membrane oxygen (ECMO) weaning and hospital discharge were defined as the clinical outcome. Results: Significantly increased mortality, with a relative risk of 1.73 with a 95% confidence interval of 1.134 to 2.639, was identified in the group with an hematocrit greater than 31%. However, no significant differences in relative risk (95% confidence interval) of death for each group were found among groups with an hematocrit less than or equal to 25%, 26-28% and 29-31%. Additionally, no significant relationship between survival and median hematocrit level was observed at a significance level of 0.413 and an Exp (B) of 1.089 at a 95% confidence interval of 0.878 to 1.373 in binary logistic regression analysis; a model was established with a -2 log likelihood of 40.687 for the entire group of patients. Moreover, a significant increase in mortality was observed as the average number of transfusions per day in the hospital increased (significance level 0.024, Exp (B) 4.378, 95% confidence interval for Exp (B) 1.212 to 15.810). Conclusion: Because a variety of factors influence therapy, the indication for transfusion should be re-evaluated and adapted repeatedly on a case-by-case basis. Further studies are needed to demonstrate whether an acceptable outcome from ECLS device therapy can also be achieved with a low hematocrit and a restrictive indication for transfusion. [ABSTRACT FROM AUTHOR]