Impact of COVID-19 Nonpharmaceutical Interventions on Pneumococcal Carriage Prevalence and Density in Vietnam

Nonpharmaceutical interventions (NPIs) implemented to contain SARS-CoV-2 have decreased invasive pneumococcal disease. Previous studies have proposed the decline is due to reduced pneumococcal transmission or suppression of respiratory viruses, but the mechanism remains unclear. We undertook a secondary analysis of data collected from a clinical trial to evaluate the impact of NPIs on pneumococcal carriage and density, drivers of transmission and disease, during the COVID-19 pandemic in Ho Chi Minh City, Vietnam. Nasopharyngeal samples from children aged 24 months were assessed in three periods - one pre-COVID-19 period (n = 1,537) and two periods where NPIs were implemented with increasing stringency (NPI period 1 [NPI-1, n = 307], and NPI period 2 [NPI-2, n = 262]). Pneumococci were quantified using lytA quantitative PCR and serotyped by DNA microarray. Overall, capsular, and nonencapsulated pneumococcal carriage and density were assessed in each NPI period compared with the pre-COVID-19 period using unadjusted log-binomial and linear regression. Pneumococcal carriage was generally stable after the implementation of NPIs. In contrast, overall pneumococcal carriage density decreased by 0.44 log10 genome equivalents/mL (95% confidence interval [CI]: 0.19 to 0.69) in NPI-1 and by 0.84 log10 genome equivalents/mL (95% CI: 0.55 to 1.13) in NPI-2 compared with the pre-COVID-19 period. Reductions in overall pneumococcal density were driven by reductions in capsular pneumococci, with no corresponding reduction in nonencapsulated density. As higher pneumococcal density is a risk factor for disease, the decline in density provides a plausible explanation for the reductions in invasive pneumococcal disease that have been observed in many countries in the absence of a substantive reduction in pneumococcal carriage. IMPORTANCE The pneumococcus is a major cause of mortality globally. Implementation of NPIs during the COVID-19 pandemic led to reductions in invasive pneumococcal disease in many countries. However, no studies have conducted a fully quantitative assessment on the impact of NPIs on pneumococcal carriage density, which could explain this reduction. We evaluated the impact of COVID-19 NPIs on pneumococcal carriage prevalence and density in 2,106 children aged 24 months in Vietnam and found pneumococcal carriage density decreased up to 91.5% after NPI introduction compared with the pre-COVID-19 period, which was mainly attributed to capsular pneumococci. Only a minor effect on carriage prevalence was observed. As respiratory viruses are known to increase pneumococcal carriage density, transmission, and disease, this work suggests that interventions targeting respiratory viruses may have the added benefit of reducing invasive pneumococcal disease and explain the reductions observed following NPI implementation.

Effect of different schedules of ten-valent pneumococcal conjugate vaccine on pneumococcal carriage in Vietnamese infants: results from a randomised controlled trial

Background

WHO recommends a three-dose infant pneumococcal conjugate vaccine (PCV) schedule administered as a two-dose primary series with booster (2 + 1) or a three-dose primary series (3 + 0). Data on carriage impacts of these and further reduced PCV schedules are needed to inform PCV strategies. Here we evaluate the efficacy against carriage of four different PCV10 schedules.

Methods

Participants within an open-label, randomised controlled trial in Ho Chi Minh City, Vietnam, were allocated to receive PCV10 in a 3 + 1 (2,3,4,9 months, n = 152), 3 + 0 (2,3,4 months, n = 149), 2 + 1 (2,4,9.5 months, n = 250) or novel two-dose (2,6 months, n = 202) schedule, or no infant doses of PCV (two control groups, n = 197 and n = 199). Nasopharyngeal swabs collected between 2 and 24 months were analysed (blinded) for pneumococcal carriage and serotypes. Trial registration: ClinicalTrials.gov NCT01953510.

Findings

Pneumococcal carriage prevalence was low (10.6-14.1% for vaccine-type (VT) at 12-24 months in unvaccinated controls). All four PCV10 schedules reduced VT carriage compared with controls (the 2 + 1 schedule at 12, 18, and 24 months; the 3 + 1 and two-dose schedules at 18 months; and the 3 + 0 schedule at 24 months), with maximum reductions of 40.1%-64.5%. There were no differences in VT carriage prevalence at 6 or 9 months comparing three-dose and two-dose primary series, and no differences at 12, 18, or 24 months when comparing schedules with and without a booster dose.

Interpretation

In Vietnamese children with a relatively low pneumococcal carriage prevalence, 3 + 1, 2 + 1, 3 + 0 and two-dose PCV10 schedules were effective in reducing VT carriage. There were no discernible differences in the effect on carriage of the WHO-recommended 2 + 1 and 3 + 0 schedules during the first two years of life. Together with the previously reported immunogenicity data, this trial suggests that a range of PCV schedules are likely to generate significant direct and indirect protection.

Funding

NHMRC, BMGF.

Effect of pneumococcal conjugate vaccine on prevalence of otitis media with effusion among children in Vietnam

PURPOSE

Otitis media with effusion (OME) is common in young children and is associated with Streptococcus pneumoniae infection. We aimed to determine the impact of pneumococcal conjugate vaccine (PCV) introduction on the prevalence of OME and OME associated with vaccine-type (VT) or non-VT.

METHODS

Population-based cross-sectional surveys were conducted in pre- (2016) and post-PCV periods (2017, 2018, and 2019) at selected communes in Nha Trang, Vietnam. For each survey, we randomly selected 60 children aged 4-11 months and 60 aged 14-23 months from each commune. Nasopharyngeal sample collection and tympanic membrane examination by digital otoscope were performed. S. pneumoniae was detected and serotyped by lytA qPCR and microarray. Odds ratios (OR) and 95% confidence intervals (CIs) were calculated using Firth's logistic regression, stratified by age group.

RESULTS

Over the four surveys, 2089 children had a bilateral ear examination. Compared to pre-PCV, the prevalence of OME reduced in 2018 (OR 0.51, 95 %CI 0.28-0.93) and in 2019 (OR 0.53, 95 %CI 0.29-0.97) among the <12-month-olds, but no significant reduction among the 12-23-month-olds. The prevalence of OME associated with VT pneumococcus decreased in 2018 and 2019 (2018: OR 0.14, 95 %CI 0.03-0.55; 2019: OR 0.20, 95 %CI 0.05-0.69 in the <12-months-olds, 2018: OR 0.05, 95 %CI 0.00-0.44, 2019: OR 0.41, 95 %CI 0.10-1.61 in the 12-23-months-olds). The prevalence of OME associated with non-VT pneumococcus increased in the 12-23-month-olds in 2017 (OR 3.09, 95 %CI 1.47-7.45) and returned to the pre-PCV level of prevalence in 2018 and 2019 (OR 0.94, 95 %CI 0.40-2.43 and 1.40, 95 %CI 0.63-3.49).

CONCLUSION

PCV10 introduction was associated with a reduction of OME prevalence in infants but not in older children.

Effect of a 2+1 schedule of ten-valent versus 13-valent pneumococcal conjugate vaccine on pneumococcal carriage: Results from a randomised controlled trial in Vietnam

BACKGROUND

Pneumococcal conjugate vaccines (PCVs) generate herd protection by reducing nasopharyngeal (NP) carriage. Two PCVs, PCV10 and PCV13, have been in use for over a decade, yet there are few data comparing their impact on carriage. Here we report their effect on carriage in a 2+1 schedule, compared with each other and with unvaccinated controls.

METHODS

Data from four groups within a parallel, open-label randomised controlled trial in Ho Chi Minh City contribute to this article. Three groups were randomised to receive a 2+1 schedule of PCV10 (n = 250), a 2+1 schedule of PCV13 (n = 251), or two doses of PCV10 at 18 and 24 months (controls, n = 197). An additional group (n = 199) was recruited at 18 months to serve as controls from 18 to 24 months. NP swabs collected at 2, 6, 9, 12, 18, and 24 months were analysed (blinded) for pneumococcal carriage. This study aimed to determine if PCV10 and PCV13 have a differential effect on pneumococcal carriage, a secondary outcome of the trial. We also describe the serotype distribution among unvaccinated participants.

TRIAL REGISTRATION

ClinicalTrials.gov NCT01953510.

FINDINGS

Compared with unvaccinated controls, a 2+1 schedule of PCV10 reduced PCV10-type carriage by 45-62% from pre-booster through to 24 months of age, and a 2+1 schedule of PCV13 reduced PCV13-type carriage by 36-49% at 12 and 18 months of age. Compared directly with each other, there were few differences between the vaccines in their impact on carriage. Vaccine serotypes accounted for the majority of carriage in unvaccinated participants.

INTERPRETATION

Both PCV10 and PCV13 reduce the carriage of pneumococcal vaccine serotypes. The introduction of either vaccine would have the potential to generate significant herd protection in this population.

FUNDING

National Health and Medical Research Council of Australia, Bill & Melinda Gates Foundation.