Background Haemolytic disease of the newborn (HDN) is a persistent public health problem in Brazil due to the lack of efficient prophylactic measures. The standard therapeutic interventions for the newborn are phototherapy and exchange transfusion (ET). Intravenous immunoglobulin has been used for several years in some neonatal centres to reduce the need for ET among newborns with jaundice, but the scientific evidence on the efficacy of this treatment is still not definitive. Objective To assess the evidence of efficacy of non-specific immunoglobulin in decreasing the need for ET in newborns with haemolytic disease of the newborn. Methods All randomized controlled trials on the use of intravenous immunoglobulin (IVIG) to reduce the need of ET in the treatment of Rhesus haemolytic disease were included by searching MEDLINE, Cochrane Library, EMBASE, Scopus, Lilacs, Scielo, www.clinicaltrial.gov and abstracts from SPR. Results Six trials were found including 281 neonates: Rubo et al. (1992) (32); Alpay et al. (1999) (23); Dagoglu et al. (1995) (41); Nasseri et al. (2006) (13); Santos et al. (2010) (92); Smits-Wintjens et al. (2011) (80). Most of the trials used the immunoglobulin in a prophylactic approach and only one used it when the jaundice was already established. Only two studies used placebo in the control group and described a method of blinding the intervention after allocation. The phototherapy protocols were not the same neither the efficacy of the units used. There was difference on the doses of the IVIG and on the number of infusions. Conclusions Although current guidelines support the therapeutic effect of the intravenous immunoglobulin effect on decreasing the need for ET in newborns with Rhesus haemolytic disease it shall not yet be routinely recommended.