• Post-measles pneumonia
  
• Admission to hospital
  

• Weight gain or loss during the first month of infection
  
• Diarrhoea
  
• Severe fever
  
• Oral thrush
  
• Stomatitis
  
• Conjunctivitis
  
• Otitis media
  

Background The case-fatality rate of measles in developing countries is still high, particularly in infants. It is estimated that measles is responsible for more than one million deaths per year, and that most of these deaths are due to complications of the disease.

Most of the severe complications of measles in developing countries are due to secondary bacterial and viral infections causing pneumonia and diarrhea.

A study from the fifties showed no benefit from treating measles cases prophylactically with antibiotics, and this together with the fear for developing antibiotic resistance has given rise to the dogma that it is harmful to give prophylactic antibiotics in measles infection.

A more recent study from Niakhar, a rural area of Senegal, has shown that children treated with prophylactic antibiotics had a lower frequency of respiratory complications. In 1987 it was decided that all children younger than 3 years of age seen within the first 2 weeks of the onset of measles symptoms should be treated with the antibiotic trimethoprim-sulfamethoxazole for 7 days irrespective of whether they had signs of bacterial infection at the time of clinical examinations. Children younger than 3 years of age who had received prophylactic antibiotics were less likely to have respiratory symptoms on days 8 to 15 of illness than children of the same age group who had not received antibiotics because they were seen for the first time on days 8 to 15 (relative risk, 0.37 (0.15 to 0.94)). Further, the case-fatality rates adjusted for age declined 2-fold between 1983-1986 and between 1987-1991 (mortality ratio, 0.41 (0.21 to 0.81)).

As this study was not an unbiased evaluation, it would be desirable to do a randomized doubleblind placebocontrolled trial of prophylactic use of antibiotics in measles infection. This could potentially prevent a large number of measles-related complications and deaths.

Participation and randomization Measles cases included in the study will receive treatment with either co-trimoxazole or the identical looking placebo. The co-trimoxazole and the placebo will be packed in identical looking sacks marked with a randomization number. The code will be broken only after the end of the study period.

Informed consent will be obtained from the parents or guardians. It will be explained that the study will examine whether antibiotics can prevent later complications, it is not known whether this is indeed the case. Therefore there will be one group receiving active treatment, and another group receiving placebo, and we do not know to which group the measles case belongs. It will also be explained if they do not want to participate, the management of the measles case will be as otherwise done in the study area.

Measles cases not included in the study will receive standard treatment.