Over a period of years there have been numerous reports from many developing countries including Pakistan showing a rising treatment failure rate with first line drugs (amoxicillin and cotrimoxazole) in non-severe pneumonia in children. The reasons for this rise in treatment failure rates are not entirely clear but one of the reasons is thought to be increasing antimicrobial resistance of H. influenzae and Strep. pneumoniae to first line drugs. There is microbiological data which suggests that this resistance can be overcome by increasing the dose of amoxicillin. The investigators propose to treat non-severe pneumonia with double dose amoxicillin in an attempt to demonstrate a decrease in treatment failure rates. This will be a multicentre, double blind randomized controlled trial in children 2-59 months of age with non-severe pneumonia comparing the clinical outcome between the standard and double dose groups.

It has been felt that the WHO criteria for treatment failure in children with pneumonia are too stringent and have never been systematically evaluated in the community. The investigators also propose to modify WHO criteria and field test them in this trial in an attempt to demonstrate that less stringent treatment failure criteria would have an impact on the overall treatment failure rates in pneumonia.

Hypothesis:

Therapy outcome with double dose of oral amoxicillin is not different than the standard dose of amoxicillin, when used for three days for the treatment of non-severe pneumonia in 2-59 months old children.

INTRODUCTION:

World Health Organization's (WHO) standard case management strategy for reducing ARI related mortality recommends oral cotrimoxazole and oral amoxicillin as first line drugs for the treatment of pneumonia. In 1989, the Pakistan Ministry of Health (MOH) adopted this strategy to control ARI mortality and recommended cotrimoxazole for treatment of outpatients pneumonia due to lower costs. A number of studies subsequently performed have shown significant in vivo and in vitro resistance of H. influenzae and S. pneumoniae, the commonest bacteria causing childhood pneumonia, to cotrimoxazole.

This increase in clinical failure with oral cotrimoxazole has put a lot of pressure both at national and international forums on the National ARI Control Programme of Pakistan to change its recommendations from cotrimoxazole to amoxicillin as the first line of therapy. Besides being efficacious for non-severe pneumonia amoxicillin has shown good efficacy in children with severe pneumonia and bacteremic children.

although on a case by case basis for pneumonia, in vitro resistance does not correlate very well with in vivo failures. One can hypothesize that if bacterial pneumonia is a certain proportion of all pneumonia cases, the rise in clinical failures may be related to increasing antimicrobial resistance of S. pneumoniae and H. influenzae to amoxicillin.

It is felt that WHO criteria for treatment failure are too stringent. Improvement is defined as the return of respiratory rate to below the cut off for that particular age at 48 hours. Some of the children with viral pneumonia will continue to have respiratory rate above the cut off on the first follow-up. These children are otherwise well and show no signs of deterioration of the disease but according to WHO criteria they would be classified as treatment failure. The treatment failure rates hence are influenced by the definitions used.

The existing data shows that if we increase the dose of amoxicillin to achieve higher MICs then the eradication of causative organism tend to be much more complete. As yet, no clinical trial has been conducted to address this important question in the treatment of childhood pneumonia.

We propose a multicentre, randomized, controlled double blind trial in which we will compare standard versus double dose oral amoxicillin for three days for the treatment of non-severe pneumonia in children less than five years of age.

Primary Objectives:

1. To compare the proportion of children 2 - 59 months of age presenting with non-severe pneumonia, who achieve clinical resolution on day 5 with standard (15 mg/kg/8hrly) versus double dose (30 mg/kg/8hrly) of oral amoxicillin therapy given for 3 days.

Secondary Objective:

To follow the clinical course of non-severe pneumonia with the alternative criteria of treatment failure (signs of deterioration i.e. lower chest indrawing and appearance of danger signs) on or before day 3 and compare them with other children who have persistent fast breathing (respiratory rate above the cut off for age) on day 3.

Null Hypothesis:

Therapy outcome with double dose of oral amoxicillin is not different than the standard dose of amoxicillin, when used for three days for the treatment of non-severe pneumonia in 2-59 months old children.

STUDY POPULATION AND SITE:

Patients:

They will be recruited from those coming to outpatient departments (OPD) in large tertiary care hospitals in Pakistan.

CONSENT:

The purpose of the study will be explained to parents and oral informed consent to participate obtained. A proposed consent form is attached (Annex. 1).

Study Design:

This will be a randomized, double blind, multi-centre trial.

Definitions:

ARI is defined according to the WHO ARI Programme guidelines 2 for children who have either a cough or difficult breathing.

A clinical resolution is defined as return of respiratory rate to normal (no pneumonia) according to WHO ARI standard case management classification 2.

Clinical Failure is defined as development of chest indrawing; or any other danger sign; or persistence of fast breathing at day 5 follow-up or thereafter leading to therapy change.

Improved will be defined as slower respiratory rate (either back to normal range for age, or lower by more than 5 compared to previous evaluation), feeding better according to mother and lower body temperature than previous assessment.

Same will be defined as still breathing fast (respiratory rate ± 5 breaths/minute compared to previous evaluation or higher than that), with no chest indrawing and or danger sign.

Worse will be defined as development of lower chest indrawing or any danger sign.

A relapse is defined as development of recurrence signs of pneumonia till day 14 after fast breathing has disappeared initially.

Follow-up:

All children will be followed-up on day 3, 5 and 14.

Sample size:

Keeping in mind the overall treatment failure rate of 20.9% with amoxicillin from the MASCOT study (1999 - 2001), it is expected that the proportion of patients who will fail to respond with standard dose amoxicillin will be around 21.0%.7 In order for double dose to be more efficacious there should be a difference of more than 5% in the treatment success rate as compared to the standard dose. We assume that the failure rate will not be more than 13% with double dose amoxicillin. By using the power of 80% and the confidence interval of 95%, the estimated sample size comes to 369 for each group. Assuming the loss to follow up to be 15% (as it is an out patient study and having day 14 follow-up) the estimated sample size will be 425 patients in each group, a total of 850 patients. Thus 900 patients will be enrolled in one and a half ARI season.

DATA ENTRY AND ANALYSIS:

Data forms will all be filled on auto copy forms and will be reviewed by the site co-ordinator.

Data will be entered using Epi Info 6 and will be entered twice by two separate operators and both data bases will be validated using Epi Info 6.

Outcome measures:

A clinical resolution is defined as return of respiratory rate to normal (no pneumonia) on day 5, according to WHO ARI standard case management classification.

Clinical/treatment Failure is defined as development of chest indrawing; or any other danger sign; or persistence of fast breathing at day 5 follow-up or thereafter leading to therapy change.

Data analysis:

Analysis will be carried out on intention to treat basis. For primary outcome proportion of therapy failures and their 95% CI will be calculated for both the two regimens. Then we will compare proportion of therapy failure using alternative criteria and WHO defined conventional therapy failure criteria.

Inclusion Criteria:

1. They are aged 2-59 months.
2. Diagnosed with WHO defined non-severe pneumonia (Table 1).

Exclusion Criteria:

• Children with signs of WHO defined severe or very severe disease.
• Known penicillin allergy, with a history of accelerated rash, urticaria, or anaphylactic symptoms. Complicating acute non-pulmonary or chronic illness .
• The children living outside the municipal limits of the city who cannot be followed up .
• Children who have taken the appropriate doses of WHO-recommended dose of anti microbial drug for 48 hours prior to presentation .
• Children who have prior history of wheezing or bronchial asthma and are wheezing now .
• Children whose parents or guardians refuse to give consent.
• Previously enrolled patients in the present study.
• Hospitalization in the past two weeks.