RESUMO
OBJECTIVES: To evaluate the efficacy of blood transfusion compared to no intervention in obstetric patients. MATERIAL AND METHODS: A systematic review was performed with Cochrane Database of Clinical Trials, PubMed, EMBASE and LILACS databases searched as of September, 2016. Two authors independently selected relevant clinical trials, assessed their methodological quality and extracted data, using the GRADE approach. RESULTS: Five studies within a total of 6,297 met the inclusion criteria, with women generally aged 20-40 years. Three included studies allocated women to receive blood transfusion or no intervention. Two other studies allocated women with either restricted or full blood supplies. The major issue regarding risk of bias was the extent of concealment of randomization and blinding. There was no statistically significant difference between blood transfusion versus no transfusion or restricted blood supply on mortality (relative risk 0.82 [95% confidential interval 0.32 to 2.09], p = 0.68; two studies; I2 = not applicable). CONCLUSIONS: Very low-quality evidence suggests no significant difference between blood transfusion and no intervention in obstetric patients, underlining the need for more robust clinical trials evaluating this area.
