TREATMENT OF SEVERE PNEUMONIA WITH ORAL ANTIBIOTICS

1
TREATMENT OF SEVERE PNEUMONIA WITH ORAL
ANTIBIOTICS
Lozano JM, on behalf of the APPIS Trial Group.
Department of Pediatrics and Clinical
Epidemiology Unit, School of Medicine, Pontificia
Universidad Javeriana, Bogotá, Colombia
Abstract TREATMENT OF SEVERE PNEUMONIA WITH ORAL
ANTIBIOTICS Lozano JM, on behalf of the APPIS
Trial Group. Javeriana University, Bogotá,
Colombia Problem Statement Injectable penicillin
is recommended treatment for WHO-defined severe
pneumonia. Oral amoxicillin, if found equally
effective, could reduce referral, hospitalisation
and treatment costs. Objectives To determine
whether an equivalent proportion of children aged
3-59 months with severe pneumonia improves after
48 hours of treatment with oral amoxicillin or
injectable penicillin. Design Open equivalence
randomised trial. Setting Nine tertiary care
hospitals in eight countries. Study Population
1,702 hospitalised children with clinically
severe pneumonia (cough and/or difficult
breathing plus chest indrawing). Intervention(s)
IV benzyl penicillin (200.000 U/kg/day) or oral
amoxicillin (45 mg/kg/day). Outcome Measure(s)
Main the proportion of children who failed to
improve or worsened at 48 hours of treatment.
Secondary the proportion of failures or relapses
at 5 and 14 days and of side effects. Results
845 subjects were allocated to penicillin and 857
to amoxicillin. There was no significant
difference in the proportion of failures at 48
hours (penicillin 161 19, amoxicillin 167
19 rate difference -0.4, 95 CI -4.2 to
3.3). Cumulative treatment failures at 5 and 14
days were also equally distributed in the two
groups. A total of 11 and 4 deaths were observed
within 30 days in the penicillin and in the
amoxicillin groups, respectively (difference
0.8, 95 CI 0.1 to 2). Infancy (OR 2.72, 95
CI 1.95 to 3.79), very fast breathing (1.94, 1.42
to 2.65) and hypoxia (1.95, 1.34 to 2.82) at
baseline predicted treatment failure by
multivariate analysis. Conclusions Oral
amoxicillin and parenteral penicillin are equally
effective in the treatment of hospitalised
children with severe pneumonia in developing
countries. Funding Source(s) Department of Child
and Adolescent Health and Development, World
Health Organization Applied Research in Child
Health (ARCH) Project USAID.
Results

• Introduction
• Standard WHO guidelines recommend that children
  with severe pneumonia (acute respiratory
  infection (ARI) and lower chest wall indrawing
  (LCI)) be hospitalized and treated with
  parenteral antibiotics. However, hospitalization
  required for the administration of injectable
  therapy has several drawbacks
• It is associated with an increased risk of
  morbidity
• in some settings injection needles and
  administration equipment are in short supply or
  periodically unavailable
• hospitalization can substantially increase the
  cost of health care
• referred children may not be brought or able to
  get to the hospital
• Research Questions
• Primary. Is treatment failure at 48 h among
  children aged 3-59 months with WHO-defined severe
  pneumonia treated with oral amoxicillin
  equivalent to that with injectable penicillin?
• Secondary. Do the two treatments remain
  clinically equivalent through follow-up on days 5
  and 14?

• Discussion
• Oral amoxicillin and injectable penicillin have
  equivalent clinical efficacy at 48 hours and
  beyond.
• The relatively high failure rate for severe
  pneumonia can be partially explained by the
  rather stringent clinical criteria used to
  establish treatment failure at this endpoint. The
  low case fatality rate supports this observation.
  
• RSV was identified in one quarter of the
  children, which is consistent with the median age
  of subjects in our study (8 months) . This is
  similar to previously reported data from
  developing countries.
• An increased failure rate among HIV-infected
  children was observed in Durban and Ndola (data
  not shown). This suggests that empiric treatment
  of severe pneumonia with oral amoxicillin or
  parenteral penicillin alone may be insufficient
  in areas of high HIV prevalence.
• Infancy, severe tachypnea and hypoxemia were
  predictive of treatment failure at 48 hours in a
  multivariable model (data not shown). These have
  also been reported elsewhere as predictors of
  therapy failure for pneumonia.

Methods Design. Randomized non-blinded
equivalency trial. Setting. Tertiary care
facilities at 9 international sites Colombia,
Ghana, India, Mexico, Pakistan, South Africa (2
sites), Vietnam and Zambia. Study population.
Children aged 3 to 59 months with WHO-defined
severe pneumonia (lower chest indrawing in
children with cough and/or difficult breathing)
were eligible. Children with a recent history of
very severe infectious/non infectious disease,
chronic or congenital illness, asthma, clinically
evident HIV infection, persistent vomiting, known
penicillin allergy or more than 48 hours of
antibiotic therapy for the present illness were
excluded. Written informed consent was obtained
from legal guardians. Randomization and
Interventions. Randomization lists, stratified by
site, were prepared in advance and implemented
using sealed, opaque and numbered envelopes.
Children were assigned to either oral amoxicillin
syrup (45mg/kg/day divided in three doses) or
parenteral penicillin G (200,000 IU/kg/day
divided in four doses). Outcomes. All subjects
were hospitalized for at least 48 hours. Children
whose clinical status improved were discharged at
48 hours with a course of oral amoxicillin and
asked to return for reassessments at 5 days and
14 days. Treatment failure up to or at 48 hours
of randomization was defined as the appearance of
danger signs, low oxygen saturation, persistence
of LCI at 48 hours, life-threatening or serious
adverse drug reaction, receiving another
antibiotic, newly diagnosed co-morbid condition,
parents/guardian withdrew consent, child left
against medical advice or death

• Conclusions
• This study demonstrates clinical equivalency
  between injectable penicillin and oral
  amoxicillin in the treatment of severe pneumonia.
  These finding have several important beneficial
  implications
• reducing the risk of needle-associated
  complications and needle-borne infections
• decrease treatment administration costs
• the full public health benefit of these findings
  will depend on additional research which
  determines the role of oral amoxicillin in the
  treatment of severe pneumonia at the community or
  household level.