This is a clinical trial to determine if oral amoxicillin is equivalent in efficacy to injectable penicillin (the standard treatment) in the treatment of WHO-defined severe pneumonia in children between the ages of 2 and 59 months.

BACKGROUND Injectable penicillin is recommended treatment for WHO-defined severe pneumonia (lower chest indrawing [LCI]). Oral amoxicillin, if found equally effective, could reduce referral, hospitalization, and treatment costs. We aimed to determine whether oral amoxicillin and parenteral penicillin were equivalent in the treatment of severe pneumonia in children aged 3-59 months.

METHODS This multicentre randomized, open label equivalency study was conducted at tertiary care centres in 8 countries. Children aged 3-59 months with severe pneumonia were hospitalized for 48 hours and, if improved, discharged with a 5-day course of oral amoxicillin. Follow-up evaluation occurred at 5 and 14 days after enrolment. The primary outcome was treatment failure (persistence of LCI or new danger signs) at 48 hours. Analyses were by intention to treat and per protocol.

RESULTS 1702 children were randomized to receive either oral amoxicillin (857) or parenteral penicillin (845) for 48 hours. Treatment failure was 19% in each group (risk difference -0.4; 95% CI -4.2 to 3.3) at 48 hours, 22% (cumulative) in each group at 5 days, and 25% in each group at 14 days. Relapse at day 5 and 14 occurred in 45/1375 (3.4%) and 65/1330 (4.8%) of the children, respectively. Infancy (OR 2.72, 95% CI 1.95-3.79), very fast breathing (1.94, 1.42-2.65) and hypoxia (1.95, 1.34-2.82) at baseline predicted treatment failure by multivariate analysis.

CONCLUSIONS Injectable penicillin and oral amoxicillin are equivalent in the treatment of severe pneumonia in developing world areas in a controlled environment. Additional research is needed in a public health setting in rural and urban areas before these finding can be recommended for widespread implementation. Potential benefits of oral therapy for pneumonia include a decrease in the 1) the risk of needle-borne infections; 2) need for referral or hospitalization; 3) administration costs and 4) costs to the family. (word count 294).

Inclusion Criteria:

• Children aged three to 59 months admitted with severe pneumonia. Severe pneumonia is defined as lower chest indrawing in children with cough and/or difficult breathing, who are able to drink and do not have central cyanosis, regardless of the respiratory rate. Known HIV infected patients in clinical category N or A (CDC) will be included.

Exclusion Criteria:

• Presence of any danger sign during current illness (convulsions, abnormally sleepy or difficult to awake, stridor in a calm child).
• Severe malnutrition defined as weight for age equal or less than -3 SD or kwashiorkor).
• Hospitalization in the last two weeks. This will exclude cases with possible nosocomial pneumonia that could require second line antibiotics.
• Known prior episodes of asthma. These patients may have respiratory distress without infection and most often do not need antibiotics. Children having had 3 or more episodes of wheezing in the past will also be excluded. Wheezing that improves after receiving bronchodilator therapy, as explained in the footnote . These children may have asthma.
• Measles in the last month. These patients may have immune suppression.
• Previous history of HIV infection in clinical category B or C (CDC). These patients have immune suppression and infections that may require other diagnostic or treatment measures (see annex 1, page 39).
• Known or clinically recognizable chronic conditions (congenital cardiac or respiratory anomalies, chronic lung disease including bronchopulmonary dysplasia, neurological impairment that affects respiratory function, renal diseases, malignant or hematological diseases).
• Other diseases requiring antibiotic therapy on presenting, such as meningitis, evident tuberculosis, dysentery, osteomyelitis, septic arthritis, etc. Use of other antibiotics needed for treatment.
• Children Lower Chest Indrawing due to non-infective cause such cardiac failure, active rickets, or severe anemia, or with signs of severe dehydration according to WHO criteria (see annex 2, page 42).
• Children with SaO2 <80% in room air at sea level, or SaO2 < 75% in Bogota and Mexico (for measurement of oxygen saturation see Study Manual).
• Known prior anaphylactic reaction to penicillin or amoxycillin.
• Known antibiotic therapy for 48 hours or more prior to admission. Clinically, these children would be considered for a change of treatment to second line antibiotics on admission. Evidence of antibiotic use include any of the following: a) parental report that an antibiotics has been given; b) parent can provide evidence for prescription for antibiotics or c) has the container with antibiotic or d) recognizes medication when shown appropriate containers for locally available antibiotics. Children that have received antibiotics for less than 48 hours prior to admission will be accepted in the trial.
• Inability to receive oral medications (three or more episodes of vomiting per hour). These children should not be given oral antibiotics.
• Previous inclusion in the trial or children already included in another trial.
• Living outside the catchment area of the hospital. These patients may be difficult to follow after discharge.
• Parental or caretaker refusal to participate in the trial.