By Philip R. Fischer, MD, DTM&H

Professor of Pediatrics, Division of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, MN

Dr. Fischer reports no financial relationships relevant to this field of study.

SYNOPSIS: Approximately 200,000 preschool-aged children in communities in Malawi, Niger, and Tanzania were treated twice yearly with either azithromycin or placebo. Communities in which azithromycin was provided had 13.5% less all-cause mortality than did placebo-treated communities. In children 1 to 5 months of age, the mortality was 25% lower with azithromycin than with placebo.

SOURCE: Keenan JD, Bailey RL, West SK, et al. Azithromycin to reduce childhood mortality in sub-Saharan Africa. N Engl J Med 2018;378:1583-1592.

Azithromycin has been effective in reducing ocular morbidity due to trachoma, and large trachoma studies have suggested that widespread use of azithromycin also might be effective in preventing pneumonia, diarrhea, and malaria. Thus, the investigators sought to rigorously determine if preventive azithromycin might be effective in reducing mortality in African children.

Researchers undertook a prospective study in three countries of sub-Saharan Africa — Niger in West Africa, Tanzania in East Africa, and Malawi in South Africa. Realizing that reducing transmissible microbial colonization in one person might affect neighbors, the investigators randomized the study at a community level. Communities were assigned to either azithromycin (a single 20 mg/kg/dose every six months over three years) or placebo treatment of children 1 to 59 months of age. Overall, 1,533 communities were included, with 190,238 children involved at the beginning of the study (with others added or removed later as they grew in or out of the age range or moved in or out of the community).

Adverse effects of treatment were reported rarely and could not be tied causally to the use of azithromycin. Shifts in antibiotic resistance patterns were not evaluated.

The rate of childhood death varied markedly by country at baseline (9.1 per 1,000 person-years in Malawi, 22.5 in Niger, and 5.4 in Tanzania) and after the study. Overall, mortality was 13.5% lower in communities that received azithromycin than in placebo-treated communities. In each of the three countries, mortality was lower with azithromycin treatment — 5.7% lower in Malawi, 18.2% lower in Niger, and 3.4% lower in Tanzania. Children 1 to 5 months of age were at the greatest risk of dying but also had the greatest reduction in mortality with azithromycin treatment (24.9%).

Verbal autopsies done after a randomized 250 of the deaths suggested that malaria accounted for 41% of deaths, diarrhea for 18%, and pneumonia for 12%. Malaria was a more common cause of death in Niger, and pneumonia was relatively more common in Tanzania.

The authors wisely pointed out the risk of hypertrophic pyloric stenosis when macrolides are used in young infants, and they reasonably cautioned about the development of antibiotic resistance when antibiotics are used widely in mass distribution campaigns. Nonetheless, they concluded that azithromycin was most effective in children who were 1 to 5 months of age and prevented one of four deaths that were expected among children of this age.


Millions of preschool-aged children die from infections in sub-Saharan Africa each year. Malaria caused by Plasmodium falciparum, pneumonia caused by Streptococcus pneumoniae, and diarrhea caused by enterotoxigenic Escherichia coli are common in Africa and potentially are preventable with the use of azithromycin.

Keenan and colleagues have performed a great service in showing that a community-wide treatment of 1- to 5-month-old children with a single dose of azithromycin every six months is associated with a significant reduction in childhood death.

The authors of this new paper carefully cautioned against generalizing these results to areas of varying mortality rates and varying causes of childhood mortality. They also noted that identifying and treating children in communities only once every six months means that some newborns would not be identified and could outgrow the window of greatest opportunity (1-5 months of age) before ever being identified and treated. However, individual treatment programs might miss some of the decreased contagiousness resulting from mass treatment programs.

It could be that we all will be advocating for prophylactic azithromycin for all infants in resource-limited areas of sub-Saharan Africa within a few years. However, there are lessons from recent history that remind us to be cautious about widespread implementation of new preventive techniques. We can learn from experience with a previous rotavirus vaccine, a new malaria vaccine, and deworming programs.

Rotavirus is a common cause of childhood diarrhea around the world, and rotavirus vaccines can reduce the morbidity and mortality of rotavirus diarrhea. After licensure, a previous rotavirus vaccine was found to be associated with an increased risk of intussusception in the United States.1 Use of that vaccine was halted, and new vaccines were developed. Fortunately, new rotavirus vaccines are not associated with an increased risk of intussusception in Africa.1 Whenever a new product or treatment is used widely, ongoing surveillance for potential complications is warranted.

Early results with a malaria vaccine showed 36% protection during four years of follow-up.2 Although not protecting a majority of children, this vaccine still could have prevented thousands of childhood deaths in Africa. However, follow-up to seven years showed that there was a rebound with more malaria subsequently in vaccine recipients.2 Even widespread interventions that start well might have longer term negative effects that unfavorably counter-balance the early positive effects.

Intestinal parasite infections are related to undernutrition and poor school performance in resource-limited areas. The World Health Organization has advocated mass deworming in areas endemic for soil-transmitted helminths, and this intervention has been considered a cost-effective approach to improve attendance at school in countries where helminths are endemic.3 Even though it is inexpensive for individuals (less than 50 cents per child), the overall cost around the world to implement deworming is approximately $276 million per year.3

A recent systematic review of studies involving more than 1 million subjects now has shown that mass deworming is associated with no improvement in weight gain, no improvement in cognition, and no improvement in school attendance.3 Nonetheless, it is possible that deworming still might be useful for certain individuals in some population subgroups.

Keenan and colleagues have taken global child health forward significantly with their data showing decreased mortality in communities where asymptomatic young children were treated with azithromycin. As further studies are conducted and as other potential side effects are evaluated, it could be that azithromycin will indeed prevent a significant number of deaths in sub-Saharan Africa without dangerous adverse effects. However, caution and subsequent surveillance are important when widespread interventions are considered and implemented.


  1. Tate JE, Mwenda JM, Armah G, et al. Evaluation of intussusception after monovalent rotavirus vaccination in Africa. N Engl J Med 2018;378:1521-1528.
  2. Olotu A, Fegan G, Wambua J, et al. Seven-year efficacy of RTS,S/AS01 malaria vaccine among young African children. N Engl J Med 2016;374:2519-2529.
  3. Welch VA, Ghogomu E, Hossain A, et al. Mass deworming to improve developmental health and wellbeing of children in low-income and middle-income countries: A systematic review and network meta-analysis. Lancet Glob Health 2017;5:e40-e50.