Background: The 2013 WHO guidelines incorporated simplified and more effective
antiretroviral regimens for the purposes of preventing mother-to-child transmission of HIV. With ideal
implementation of these recommendations, perinatal HIV transmission could be reduced to less than
2%. However, loss to follow-up (LTFU) has the potential to erode the success of programs and a
number of studies report high rates of LTFU within the prevention of mother-to-child transmission
(PMTCT) care cascade. We evaluated the timing and magnitude of LTFU in a large programmatic
PMTCT cohort in Nigeria in order to focus future efforts to reduce loss in this high burden setting.
Methods: From 2004-2014, the APIN/Harvard PEPFAR program supported antenatal HIV screening for nearly one
million pregnant women and provided PMTCT care to over 30,000 women. The care cascade for women enrolling in the
PMTCT program includes antenatal, delivery, and infant follow-up services through 12-18 months of life. In this
retrospective cohort analysis, we examined data collected between 2004-2014 from 31 clinical sites in Nigeria and
assessed the numbers of mothers and infants enrolled and LTFU at various points along the care cascade.
Results: Among 31,504 women (median age 30, IQR: 27-34) entering PMTCT care during the antenatal period, 20,679
(66%) completed the entire cascade of services including antenatal, delivery, and at least one infant follow-up visit. The
median gestational age at presentation for antenatal care services was 23 weeks (IQR: 17-29). The median infant age at
last follow-up visit was 12 months (IQR: 5-18). The greatest loss in the PMTCT care cascade occurred prior to delivery
care (21%), with a further 16% lost prior to first infant visit. Of the 38,223 women who entered at any point along the
PMTCT cascade, an HIV DNA PCR was available for 20,202 (53%) of their infants. Among infants for whom DNA PCR
results were available, the rate of HIV transmission for infants whose mothers received any antenatal and/or delivery care
was 2.8% versus 20.0% if their mother received none.
Conclusion: In this large cohort analysis, the proportion of women LTFU in the PMTCT care cascade was lower than that
reported in previous cohort analyses. Nevertheless, this proportion remains unacceptably high and inhibits the program
from maximally achieving the goals of PMTCT care. We also provide the largest analysis to date on rates of perinatal
HIV transmission, with low rates among women receiving NNRTI- or PI-based regimens, approaching that reported in
clinical trials. However, among mothers who received any antenatal care, infant outcomes were unknown for 48%, and
women presented later in pregnancy than that recommended by current guidelines. Implementation research to evaluate
ways to improve integration of services, particularly transitions from antenatal to delivery and pediatric care, are critically
needed to reduce LTFU within PMTCT programs and achieve the ultimate goal of eliminating pediatric HIV infection.