Preventing MothertoChild Transmission of HIV: The U'S' Experience

1

• Mortality Among HIV-1-Infected Women Individual
  Patient Data Meta-Analyses
• Description of, and Risk Factors for,
• Mortality Among Breastfeeding Women
• Mortality According to Infant Feeding Modality
• (Formula Feeding vs. Breastfeeding)
• (J.S. Read, for the Breastfeeding and HIV
  International Transmission Study)

2
Background

• Recent analyses regarding breastfeeding and
  maternal mortality among HIV-1-infected women
  conflicting results
• Nduati et al Lancet 2001 357 1651-1655
• Coutsoudis et al AIDS 2001 15 653-655

3
Background

• Nduati et al analysis of data from randomized
  clinical trial of breastfeeding vs. formula
  feeding
• Analysis by intention to treat
• Mortality of women over 2 year period after
  delivery higher in BF group (18/197) vs. FF
  (6/200) (p 0.009)
• Cumulative probability of maternal death at 24
  months BF (10.5) vs. FF (3.8) (p 0.02)

4
Background

• Relative risk of death for mothers assigned to BF
  vs. FF 3.2 (95CI 1.3, 8.1) (p 0.01)
• Attributable risk of maternal death due to
  breastfeeding 69
• Significant associations between
• CD4 and maternal death
• Viral load and maternal death

5
Background

• Issues raised in accompanying editorial (Newell
  Lancet 2001 357 1634-1635), including
• Women randomized to BF had lower median CD4
  counts and higher median peripheral blood viral
  load concentrations at baseline than women
  randomized to FF
• CD4 399 vs. 415
• Viral load 48,000 vs. 37,000
• Information regarding timing of maternal deaths
  was not presented
• Loss to follow-up occurred earlier in FF group
  compared to BF group (therefore, deaths could
  have been underreported in FF group)

6
Background

• Coutsoudis et al Analysis of data from
  randomized clinical trial of vitamin A
  supplementation in South Africa
• Mothers chose infant feeding modality (NOT
  randomized)
• N 566 (410 BF vs. 156 never BF)

7
Background

• No differences in maternal mortality rates
  according to infant feeding modality were
  observed
• 0.49 (2/410) ever BF
• 1.92 (3/156) never BF
• Morbidity similar among those who BF gt 3 mo
  (34/239 14) vs.BF lt 3 mo (18/171 10.5) (p gt
  0.10)

8
Objectives BHITS Mortality Analyses

• Mortality among BF women
• 1. To more reliably estimate the mortality risk
  (and the timing of mortality in relation to
  duration of breastfeeding) among HIV-infected
  breastfeeding women over a 12-18 month period
  after delivery
• 2. To identify factors associated with mortality
  among HIV-infected, breastfeeding women
• Mortality among BF and FF women
• 1. Compare mortality among BF and FF women
  (enrolled at sites with relatively large
  proportions of women who never breastfed)

9
Methods, Results and Conclusions

• Analyses ongoing
• Preliminary results to be discussed
• During BHITS investigators meeting on December
  11 and
• (To the extent possible) at workshop on December
  12/13

10
Writing Committee

• Anna Coutsoudis
• Francois Dabis
• Wafaie Fawzi
• Philippe Gaillard
• Geert Haverkamp
• D. Robert Harris
• J. Brooks Jackson

• Valeriane Leroy
• Nicolas Meda
• Philippe Msellati
• Marie-Louise Newell
• Ruth Nduati
• Jennifer S. Read
• Stefan Wiktor

11
Acknowledgements

• A collaboration between NICHD and the Ghent
  Working Group on Mother-to-Child Transmission
• Supported by N01 HD 43208 (NICHD contract)
• Data management center Westat, Rockville,
  Maryland, USA