Assessment of mother-to-child HIV-1 and HIV-2 transmission: an AIDS reference laboratory collaborative study

A prospective study was carried out to assess HIV-1 and HIV-2 mother-to-child transmission (MTCT) rates in Portugal between 1999 and 2005 by analysing the proportion of diagnosed infected children born to HIV-positive mothers. Serial blood samples were collected from 1315 children at risk of HIV-1 infection, 131 children at risk of HIV-2 infection and six children at risk of both HIV-1 and HIV-2 infections attending 25 Health Institutions. HIV proviral DNA was detected by nested polymerase chain reaction (PCR) and statistical analysis was performed using spss. DNA PCR using HIV-1 and HIV-2 long terminal repeat (LTR) primers amplified 92.5% and 75% of maternal HIV infections, respectively. Overall, MTCT occurred in 3.4% [95% confidence interval (CI) 2.5-4.6%] of HIV-1 and 1.5% (95% CI 0.2-5.4%) of HIV-2 mother-child pairs. A significant decrease in HIV-1 MTCT was observed with time, from 7.0% (95% CI 2.6-14.6%) in 1999 to 0.5% (95% CI 0.0-2.5%) in 2005. HIV MTCT was associated with an absence of antiretroviral therapy in infected pregnant women (P < 0.0001). Of the 48 infected children (46 with HIV-1 and two with HIV-2), the schedule of blood sample collection was followed for only 26 children. In 14 (53.8%) of those 26 children the infections were diagnosed in the first sample collected before they were 48 h old, suggesting in utero transmission. Despite the national recommendations for antenatal HIV testing, a high overall proportion (22.2% for HIV-1 and 44.3% for HIV-2) of mothers did not access any MTCT prevention measures, mostly because of late diagnosis in pregnancy. A small but significant proportion of HIV-2 infection was found in mothers with no identifiable link with West Africa. HIV-2 transmission rates are low (1.5% in this study), and this may have led to a lower uptake of interventions, but in the absence of interventions transmission does occur. HIV-1 transmission was also associated with a lack of intervention, mostly as a result of late presentation. Use of primers restricted to a single sequence led to false-negative maternal results in a significant proportion of cases. In part this may have been attributable to very low HIV DNA loads as well as primer template mismatches. HIV infection was not documented in children born to mothers with negative HIV DNA PCR results.