Perinatal zidovudine prophylaxis in HIV type-1-infected pregnant women with thalassaemia carriage in Thailand

Abstract

Background To investigate a possible interaction between α-thalassaemia, β-thalassaemia and haemoglobin-E trait and the haematological parameters of HIV type-1 (HIV-1)-infected pregnant women receiving zidovudine prophylaxis for the prevention of mother-to-child HIV-1 transmission in Thailand. Methods The study sample was composed of HIV-1-infected pregnant women receiving zidovudine (300 mg twice daily) from 28 weeks of gestational age to delivery as part of the Perinatal HIV Prevention Trial (PHPT-1), a large trial investigating zidovudine use in pregnancy. These women were randomly selected and screened for haemoglobin abnormalities. Haemoglobin levels, haematocrit and erythrocyte, leukocyte, absolute neutrophil and absolute lymphocyte counts were measured at 26, 32 and 35 weeks of gestation and at delivery. PCR genotyping techniques were used to screen for haemoglobin abnormalities, which included α-thalassaemia-1 Southeast Asian type deletion, β-thalassaemia mutation (codons 41/42 [-TCTT], codon 17 [A→T], intervening sequence-I nucleotide 1 [G→T], codons 71/72 [+A]) and haemoglobin-E trait. The evolution of haematological parameters between 26 weeks and delivery was compared according to thalassaemia carriage using linear mixed models adjusted for baseline sociodemographic characteristics, HIV clinical stage, CD4+ T-cell count and viral load. Results At baseline, women with thalassaemia or haemoglobin-E trait had significantly lower haemoglobin level and red blood cell counts than women with no haemoglobin abnormalities, whereas absolute neutrophil and leukocyte counts were significantly higher. Exposure to zidovudine until delivery did not increase this difference. Conclusions Zidovudine exposure did not appear to have increased haematological toxicity in HIV-1-infected pregnant women with thalassaemia.