Vaccination and Pneumococcal Disease in South Africa

Published - Written by Brian Honeyman

Since the beginning of medical school I, like most of us, have been bombarded by medical questions from friends, family, and even brand-new acquaintances.  If I had a dollar for every rash that I’ve been asked to “take a quick look at” my loans would be a bit less daunting.  One question often posed by friends with young children: how important are vaccines really?  In a time when the public is inundated with information of questionable scientific merit, it becomes critical for all physicians-in-training to understand this commonly discussed issue of vaccines and their effectiveness in diverse populations.  In this week’s NEJM, von Gottberg and colleagues offer an opportunity to examine one nation’s experience with a pneumococcal vaccine program.

The largest number of deaths due to pneumococcal disease among children <5 years of age occur in Africa.  Additionally, HIV-infected children aged <1 have a 20-fold greater rate of invasive pneumococcal disease-associated (IPD) hospitalizations than HIV-uninfected children. In 2009, South Africa adopted a 7-valent pneumococcal conjugate vaccine (PCV), and later, in 2011, replaced it with a 13-valent PCV.  The country reached 82% completion of the 3 PCV doses (doses at 6 and 14 weeks, with a booster at 9 months of age) in all children under 12 months old.  While these vaccines have a demonstrated effect in resource-rich countries, this study provides a multi-year examination of the effectiveness of PCV in a predominantly low-resource population with higher rates of pneumococcal disease and carriage, and high HIV prevalence.  Over the study period, HIV incidence rates in infants <2 months old decreased from 9.8% in 2008 to 2.8% in 2011.

This study compared annual IPD incidence pre- and post-vaccine rollout, comparing 2005-2008 to 2011-2012, in all ages and further stratified the results by HIV status.  Years 2009-2010 were excluded to account for the gradual rollout of the PCV vaccination program.  They deployed nationwide laboratory surveillance for IPD gathered from 2005 through 2012 in 459 hospitals in South Africa.  35,192 cases of IPD defined as Streptococcus pneumonia cultured from normally sterile-site specimens were found, serotyped, and the antimicrobial minimum inhibitory concentrations (MIC) were determined.  

The investigators found that in all ages, the rates of IPD decreased by 40% between 2008 and 2012, with a non-significant increase of 6% in non-vaccine serotype IPD.  In the high risk age group of children <2 years the incidence of all-serotype IPD decreased 69% over the study period. Similar reductions in disease were seen when stratified by HIV-infection.  Specifically, in HIV-uninfected children aged <2 years incidence of vaccine-serotype IPD cases decreased by 85% while non-vaccine serotype disease increased by 33%; results were comparable in HIV-infected children. Significant reductions of IPD with vaccine-related serotypes were also observed in the 25-44 age group regardless of HIV-infection, driven mostly by reductions in PCV-7-serotype disease rates; -57% in HIV-infection and -59% in HIV-uninfected.  Other age groups showed reductions in IPD but failed to reach statistical significance.

Across all age groups the rates of antimicrobial-resistant pneumococcal disease decreased. The rates (per 100,000) of penicillin nonsusceptible IPD decreased by 57% (from 4.3 to 1.9), ceftriaxone nonsusceptible IPD decreased from by 58% (from 0.8 to 0.3), and multidrug-resistant IPD rates decreased by 52% (from 2 to 1).

NEJM Deputy Editor Lindsey Baden, MD, notes: “Pneumococcus remains a tremendous cause of severe illness especially in children and immunocompromised patients. Improving the pneumococcal vaccine and its deployment can significantly decrease disease from pneumococcus and it can also apply immunologic selective pressure on this bacillus, as seen in the decrease in the circulation of antibiotic resistant strains. ”

These data demonstrate effectiveness of the recent vaccination program in South Africa in prevention of IPD.  Further surveillance will be needed to determine the long-term impact of non-vaccine serotype disease.


Brian Honeyman is a 4th year medical student at Boston University, and recently completed an elective course at the editorial offices of the New England Journal of Medicine.