Estratégias de transfusão de hemácias em pacientes internados em UTI: revisão sistemática e metanálise em rede e tradicional

Abstract

Background: Hemoglobin threshold on which critically ill patients should receive transfusion of red blood cells (RBC) is still controversial. Traditional meta-analysis showed that there is no difference between a restrictive versus a more liberal transfusional strategy. Moreover various studies used different levels of hemoglobin in each strategy. Objective: A network meta-analysis was performed to compare the effects of different thresholds for RBC transfusion on early mortality and other clinical outcomes in critically ill patients. Methods: Electronic databases were searched for randomized clinical trials comparing 2 levels of hemoglobin as threshold for red blood transfusion in adults admitted to the intensive care unit (ICU). After analyzing titles and abstracts of all the studies found in the search, the relevant ones were selected for full-text review and data extraction. The primary outcome was early mortality (28 or 30-day mortality) and secondary outcomes were late mortality (60, 90, 120 or 180-day), ICU and hospital length of stay, number of patients who received transfusions, ischemic events and adverse events of any kind. The network meta-analysis was performed by a frequentist based approach with multivariate random effect meta-analysis. In addition, a traditional meta-analysis was performed, comparing a restrictive (7 g/dl of hemoglobin) to a liberal transfusion strategy (9 or 10 g/dl of hemoglobin). Results: Seven randomized clinical trials were included (2848 patients). There were no significant differences among the thresholds screened at the network metaanalysis (7, 9 or 10 g/dl) regarding early mortality or the secondary outcomes. The threshold of 7 g/dl was associated with higher probability of not being transfused. Traditional meta-analysis showed no differences among the outcomes, except for the lower relative risk of being transfused in the restrictive strategy. Conclusions: A lower threshold may reduce the risk of being transfused. Since this decrease is not followed by negative impact regarding mortality, length of stay or occurrence of adverse events, it can be used in critically ill patients.