Interventions to reduce morbidity and mortality in formula-fed HIV exposed uninfected infants

1
Interventions to reduce morbidity and mortality
in formula-fed HIV exposed uninfected infants

• Chhagan, M. K.
• Kauchali, S.
• Coutsoudis, A.
• Coovadia, H

2
Overall Goal

• Identify feasible and effective public health
  interventions that would preserve the health and
  well-being of HIV-exposed uninfected infants born
  to HIV-infected mothers who choose to formula feed

3
What is the reality at PMTCT clinics in urban
centres offering VCT and Infant feeding
counseling?
Kassier 2003. Health Systems Trust Report
4
What is the evidence that infants who escape HIV
infection in first few months of life
(HIV-free survival) are still at risk of
morbidity and mortality?
HIV-free is not synonymous to morbidity-free
5
The MASHI trial severe diarrhea in first 6
months by feeding practice (n1200 HIV women)
Lockman et al., MASHI study
team. Abstract IAS, 2006
6
The MASHI trial mortality in first 6 months by
feeding practice(n1200 HIV women)
7
Infectious morbidity increased among cohort of
HIV exposed uninfected infants in Latin America

• 45 of 338 formula-fed infants experienced at
  least one episode of infection (respiratory
  infectionsgtdiarrhea) in first 6 months of life
• Mussi-Pinhata et al. Abstract CROI 2005.

8
Excess mortality among infants in a era of
HIV/AIDS
HIV status
Brahmbatt et al. JAIDS, 2006 Rakai, Uganda
(median duration of breastfeeding20months)
9
Mortality Zimbabwe

• 2 yr follow-up of cohort of infants born to HIV
  (n345) and HIV-(n351) mothers.
• Cumulative mortality at 2yrs
• HIV uninfected non-exposed 1.9
• HIV uninfected exposed 2.6
• Nathoo, K. Cent Afr J Med 2004

RR1.4
10
Pooled analysis mortality in SSA

• Mortality in 3468 infants of HIV mothers
• By age 2 yrs 52.5 of HIV-infected infants died
  and 7.6 of uninfected exposed infants died
• (No M-C- cohort for comparison of excess
  mortality)
• Newell, M et al. Lancet 2004

11
Conceptual framework
HIV mothers who choose formula
HIV- mothers
Increased pathogen burden and transmission
Common environmental risk factors for childhood
infections
12
Guidelines for cotrimoxazole prophylaxis

• All HIV-exposed infants born to HIV-infected
  mothers, commence at 4-6 weeks age
• UNTIL HIV infection definitely excluded AND no
  longer breastfed.
• Joint WHO/UNAIDS/UNICEF statement on use of
  cotrimoxazole prophylaxis in HIV exposed and HIV
  infected children. 2004.
• South Africa PMTCT program performs HIV PCR at 6
  weeks age followed by early decision to
  discontinue prophylaxis

13
Proposed interventions
Intervention A Standard counseling on formula
preparation PLUS placebo syrup Intervention B
Standard counseling on formula preparation PLUS
continued CTX prophylaxis up to 12 mo
age Intervention C Standard counseling on
formula preparation PLUS placebo syrup PLUS home
hygiene package (infection control practices
hand washing and soap for entire household
sterilization of utensils for formula feeding
disinfectants for sterilization safe food
preparation and storage) NOTE All arms will
receive standard care and support for maternal
health and wellbeing

FFPlacebo
FFCTX
FFPlaceboHygiene
Infections
Malnutrition
14
Rationale for hygiene interventions
Excess morbidity and mortality in formula fed
infants who escape HIV infection Samples of
commercial infant milk samples from 94 mothers at
PMTCT clinics in SA E. coli isolated in 64 and
enterococci in 26 despite relatively good
education levels and counselling on formula
feeding. Bergstrom 2003. Health Systems
Trust.
FFPlacebo
FFCTX
FFPlaceboHygiene
Infections
Malnutrition
15
Hygiene interventions
Hand-washing and disinfection reduced diarrhea
incidence by 53 in children. Pakistan- 25
neighborhoods, 300 households, 1523 children.
Lube. JAMA 2004. Hygiene and safe water
storage in households reduced diarrhea episodes
by 25 in HIV-infected persons. Lule. Am J Trop
Med Hyg 2005. NOTE Large effect sizes
associated with hygiene interventions
FFPlacebo
FFCTX
FFPlaceboHygiene
Infections
Malnutrition
16
Rationale for cotrimoxazole (CTX) prophylaxis
CTX prophylaxis effective in HIV-infected
children CHAP trial Chintu. Lancet 2004 High
frequency of sub-clinical PCP carriage in
HIV-infected persons. PCP infections identified
in HIV-uninfected infants. Morris. Emerging Inf
Dis. 2004. 2771 HIV-negative household members
living with HIV-infected cases in Uganda 35
with positive bacterial cultures for stool
pathogens, incl Shigella, Salmonella,
Campylobacter, E coli. Mermin AIDS 2005.

FFPlacebo
FFCTX
FFPlaceboHygiene
Infections
Malnutrition
17
Cotrimoxazole to protect HIV-uninfected persons
living in a high-risk environment

• CTX prophylaxis in HIV-infected adults reduces
  morbidity in uninfected household members. In
  HIV-negative children lt10 yrs of age, reductions
  in mortality (HR 0.37), malaria (HR 0.65),
  diarrhea (HR 0.58) and hospitalizations (HR 0.53)
  (Mermin AIDS 2005)
• - CTX prophylaxis associated with fewer lower
  respiratory tract infections (OR0.44) among HIV
  exposed uninfected infants (Coutsoudis, SAMJ
  2005)

FFPlacebo
FFCTX
FFPlaceboHygiene
Infections
Malnutrition
18
Specific aims and hypotheses

• Primary aims
• To compare the following outcomes between
  intervention arms among formula-fed HIV-exposed
  uninfected infants between 6 wks to 24 months age
  
• incidence rates of diarrheal and respiratory
  illness
• growth trajectories (weight, height, MUAC)
• cumulative mortality

19
Specific aims and hypotheses

• Secondary aims
• Compare incidence rates of severe diarrheal and
  respiratory illness across intervention groups
• Compare cumulative incidence of all-cause
  household morbidity across groups
• Measure of pathogen burden and transmission
  within households
• Compare incidence rates of diarrheal and
  respiratory illness between HIV-exposed and
  unexposed infants

20
Design Cluster randomized trial

• Study subjects
• MC- pairs attending PMTCT follow-up
• (consider enrolling antenatally)
• Include additional random sample of M-C- pairs
  attending 6 week immunization visit, to mask and
  provide reference morbidity and mortality
  incidence rates
• Clusters 12 PMTCT Clinics per arm
• Sample size 12 clusters per arm, 100 subjects
  per cluster
• Clusters blocks Formal urban (3), informal urban
  (3), peri-urban(3), rural (3)
• Total subject pairs for interventions 3600
• Total M-C- reference/mask subjects 600
• TOTAL SAMPLE SIZE 4200
• Enrollment 12 months
• Follow-up 24 months
• Preparation of sites 12 months
• Study duration 48 months

FFPlacebo
FFCTX
FFPlaceboHygiene
Infections
Malnutrition
21
Cluster RCT

• Cluster RCT chosen over individual because
  interventions (hygiene practice) susceptible to
  contamination
• e.g. household hygiene and infection control
  practices in Pakistan used Cluster RCT design
• Sample size estimated after accounting for
  clustering

22
Public health impact