Transcript Antibiotics as Part of the Management of Severe Acute

Antibiotics as Part of the Management of
Severe Acute Malnutrition
Indi Trehan MD MPH DTM&H
Assistant Professor, Washington University in St. Louis Department of Pediatrics
Visiting Lecturer, University of Malawi Department of Paediatrics and Child Health
17 October 2013
Community-Based Management of Severe Acute Malnutrition
Adjuncts to RUTF?
WHO, 2003
2003; WHO/WFP/UNSCN/UNICEF, 2007
Antibiotics for Severe Acute Malnutrition
A need for prospective data
Bulletin
of the
World
Health Organization 2011; 89: 593
PLoS
ONE
2013;
8: e53184
Antibiotics for Severe Acute Malnutrition
RCT design
Population :: Children 6-59 months old with severe acute malnutrition (SAM) who qualify
for community-based therapy (i.e., have an appetite and a reliable caretaker)
• Kwashiorkor (edematous malnutrition) and/or Marasmus (weight-for-height Z-score < -3)
• Within walking distance of one of 18 rural clinic sites in southern Malawi
• Exclusions :: obvious chronic debilitating illness (excluding HIV & TB); recently enrolled in a
therapeutic feeding program for acute malnutrition (inpatient or outpatient)
Intervention :: Empiric oral antibiotics in addition to RUTF
• Amoxicillin 80-90 mg/kg/d div BID x7d
• Cefdinir ~14 mg/kg/d div BID x7d
Comparison :: Placebo in addition to RUTF
Outcomes
• Primary :: adverse effects; nutritional recovery; mortality
• Secondary :: time to recovery; growth parameters (height, weight, MUAC)
Antibiotics for Severe Acute Malnutrition
Which antibiotics to use?
Retrospective review of all admission blood cultures obtained from the 4322 children admitted
to the Nutritional Rehabilitation Unit at Queen Elizabeth Central Hospital in Blantyre
August 2005 – March 2008
33 (3.4%)
Grampositives
3351
no blood
culture
Malawi Medical Journal 2009; 21: 29
971
blood
cultures
808 (83%)
no growth or
contaminants
73 (7.5%)
nonTyphoidal
Salmonella
57 (5.9%)
other
Gramnegatives
Antibiotics for Severe Acute Malnutrition
Which antibiotics to use?
Penicillin
or
Ampicillin
33 (3.4%)
Gram-positives
Erythromycin
Chloramphenicol
Cotrimoxazole
Gentamicin
Tetracycline
Ceftriaxone
Ciprofloxaxin
Penicillin
100
%
73 (7.5%)
non-Typhoidal
Salmonella
Ampicillin
100
%
Ampicillin
57 (5.9%) other
Gram-negatives
Malawi Medical Journal 2009; 21: 29
100
%
100
%
100
%
Baseline Characeristics
New England Journal of Medicine 2013; 368: 425
Adverse Events
Variable
Amoxicillin
Cefdinir
Placebo
Number of children who took all 7 days of
intervention
865/879 (98)
887/897 (99)
865/872 (99)
Fever since enrollment
309/876 (35)
339/889 (38)
337/870 (39)
Cough since enrollment
239/874 (27)A
280/889 (31)
301/871 (35)A
Diarrhea since enrollment
322/878 (37)B
282/889 (32)B,C
352/871 (40)C
Vomiting since enrollment
114/877 (13)
124/890 (14)
137/872 (16)
43/865 (5)
31/872 (4)
37/857 (4)
865/879 (98)
883/893 (99)
855/871 (98)
Rash since enrollment
Reported to have a good appetite since
enrollment
Values are presented as no./total no. (%).
All pairwise comparisons with P > 0.05 except for the following:
AP = 0.001 for cough since enrollment for amoxicillin vs. placebo
BP = 0.03 for diarrhea since enrollment for amoxicillin vs. cefdinir
CP < 0.001 for diarrhea since enrollment for cefdinir vs. placebo
New England Journal of Medicine 2013; 368: 425
24.4% (4.1%-40.4%) reduction in failure rate with amoxicillin
38.9% (21.1%-52.7%) reduction in failure rate with cefdinir
35.5% (6.9%-55.4%) reduction in mortality rate with amoxicillin
44.3% (18.0%-62.2%) reduction in mortality rate with cefdinir
Secondary Outcomes
New England Journal of Medicine 2013; 368: 425
Survival Analyses
New England Journal of Medicine 2013; 368: 425
Recovered
2 p=0.0005
88.7%
90.9%
p=0.0001
RR 0.94
(0.91 - 0.97)
5.8% difference
(2.8% - 8.8%)
2 p=0.0045
7.4%
4.8%
4.1%
p=0.0025
RR 1.80 (1.22 - 2.64)
3.3% diff (1.2% - 5.4%)
p=0.021
RR 0.96
(0.93 - 0.99)
3.6% difference
(0.6% - 6.7%)
Amoxicillin Cefdinir
Died
85.1%
p=0.018
RR 1.55 (1.07-2.24)
2.6% diff (0.5% - 4.8%)
Placebo
Amoxicillin Cefdinir
Placebo
• Amoxicillin decreased failure rate by 24% and death rate by 36%.
• Cefdinir decreased failure rate by 39% and death rate by 44%.
• Need to treat only 31 children with this $5 intervention to save 1 life.
• Antibiotic therapy needs to be vigorously incorporated into CMAM programs and its
role emphasized to funding agencies and to those involved in frontline implementation.
New England Journal of Medicine 2013; 368: 425
Antibiotics for Severe Acute Malnutrition
Bacteremia and gut translocation
American
International
Pediatric
Journal
Journal
ofPaediatrics
Disease
Diseases
of Infectious
Journal
in2006;
Childhood
Diseases
2000;
19:2002;
1984;
312 6:
138:
187551
Annals ofInfectious
Tropical
26:
1992;
12: 319
433
Antibiotics for Severe Acute Malnutrition
Bacteremia and gut translocation
Archives
American
Trends
PLoS
ONE
inof
Microbiology
Journal
2011;
Disease
6:ofe18580
inClinical
Childhood
2010; Nutrition
18: 487
1997; 1987;
76: 236
45: 1433
Support
• Mark Manary
• St. Louis Nutrition Project
• Project Peanut Butter
• Eunice Kennedy Shriver
National Institute for Child
Health and Development
• T32-HD049338
• L40-HD066655
• Hickey Family Foundation
• USAID
• Agency for Educational Development
• Office of Health, Infectious Diseases, and
Nutrition
• Office of Food for Peace
Understanding the Gut Microbiome…in Kwashiorkor
Disruption in normal development
Science 2013; 339: 548
Understanding the Gut Microbiome…in Kwashiorkor
Disruption in normal development
• “Identical” twin pairs discordant for kwashiorkor
• Overall gene contents of the fecal microbiota in children with kwashiorkor fails to develop
with increasing age
• Microbiome of children kwashiorkor less “mature” than their healthy co-twin
Understanding the Gut Microbiome…in Kwashiorkor
Gnotobiotic mouse model
• Fecal microbial communities from discordant twins at the time of diagnosis were
transferred into germ-free gnotobiotic mice
• Mice then fed diet based on typical Malawian foods
• Mice who received kwashiorkor microbiome lost weight
• Mice who received healthy microbiome sustained their weight
• Kwashiorkor mice fed RUTF gained weight
• Unfortunately, weight gain not that well sustained after RUTF ended
Understanding the Gut Microbiome…in Kwashiorkor
Gnotobiotic mouse model
Nature Reviews Gastroenterology & Hepatology 2013; 10: 261
Understanding the Gut Microbiome…in Kwashiorkor
Gnotobiotic mouse model
• Meaningful changes in the fecal taxonomic, genetic,
and metabolic content accompanied these
transplantations and dietary shifts in the recipient
mice
• PCR for 22 common bacterial, protozoal, and
viral enteric pathogens revealed no evidence
that this was due to any of these microbes
• Significant differences in 37 species-level taxa
• Bilophila wadsworthia, a sulphite-reducing
organism previously linked to IBD flares
• Clostridia innocuum, a symbiont that is an
opportunistic pathogen in
immunocompromised hosts
• 3 species of Bifidobacteria, 2 species of Lactobacilli,
2 Ruminococcus species recovered after receiving
RUTF, as did amounts of essential and nonessential
amino acids
Understanding the Gut Microbiome…in Kwashiorkor
Gnotobiotic mouse model
• Analysis of urinary metabolites revealed an inhibition of the Krebs (TCA) cycle in
kwashiorkor mice fed a Malawian diet -- indicative of impaired cellular metabolism and
energy production
• Taurine, cysteine, methionine concentrations also disrupted, suggestive of disordered sulfur
metabolism -- consistent with prior data linking sulfur amino acid metabolism to
kwashiorkor