The impact of the introduction of ten- or thirteen-valent pneumococcal conjugate vaccines on antimicrobial-resistant pneumococcal disease and carriage: A systematic literature review

Abstract

Background: A systematic review in 2019 found reductions in antimicrobial resistance (AMR) of pneumococcal vaccine serotypes following pneumococcal conjugate vaccine (PCV) introduction. However, few low- or middle-income countries were included as not many had introduced higher valent PCVs (PCV10 or PCV13). The aim of our review is to describe AMR rates in these samples following the introduction of PCV10 or PCV13.

Methods: We conducted a systematic literature review of published papers that compared AMR for invasive pneumococcal disease (IPD), otitis media (OM) and nasopharyngeal carriage (NPC) samples following introduction of PCV10 or PCV13 to the pre-PCV period. Included studies published from July 2017 to August 2020 had a post-licensure observational study design and reported on our defined outcomes: IPD, OM, NPC and other (sputum or mixed invasive and non-invasive pneumococcal) isolates from people of all ages. Rates of AMR in the pre- and post-period were extracted.

Results: Data were extracted from 31 studies. Among IPD isolates, penicillin AMR rates following PCV10 or PCV13 introduction declined in 32% (n = 9/29) of included studies, increased in 34% (n = 10/29) and showed no change in 34% (n = 10/29). Cephalosporins AMR declined in 32% (n = 6/19) of studies, increased in 21% (n = 4/19) and showed no change in 47% (n = 9/19). Macrolides AMR declined in 33% (n = 4/12) of studies, increased in 50% (n = 6/12), and showed no change in 17% (n = 2/12). AMR to other antibiotics (including multidrug resistance) declined in 23% (n = 9/39) of studies, increased in 41% (n = 16/39) and showed no change in AMR in 36% (n = 14/39). There were no obvious differences between AMR; in setting which used PCV10 vs PCV13, according to time since PCV introduction or by World Bank income status of the respective country. The only study including OM isolates found no change in penicillin resistance. There were few studies on AMR in NPC (four studies), OM (one study) or other isolates (five studies). The results followed similar patterns to IPD isolates.

Conclusions: We observed considerable heterogeneity in the findings between and within studies, e.g. no evidence of reduction in amoxicillin AMR with an increase in macrolides AMR. Reasons for such diverse findings include the period covered by different studies and variation in other pressures towards AMR.

Figure 1

Flow diagram of the study selection process.