Abstract
Background
Non-pharmaceutical interventions to mitigate the spread of SARS-CoV-2 were implemented across the United States in 2020. These public health measures might influence pneumococcal colonization in younger children and their subsequent risk of invasive pneumococcal disease. Our objective was to evaluate pneumococcal colonization rates in children ≤ 5 years of age with and without respiratory symptoms during the first 12 months of SARS-CoV-2 pandemic (April 2020 – March 2021).
Study design
This is a retrospective cross-sectional study evaluating pneumococcal colonization rates, density and serotype distribution across three study periods (April-July [Period 1], August-November [Period 2] and December-March [Period 3]) after implementation of non-pharmaceutical interventions in Kansas City, Missouri. Children aged ≤5 years with and without respiratory symptoms seeking care at Children’s Mercy Kansas City hospital system with a residual standard of care nasal mid-turbinate sample were included. The odds of pneumococcal colonization across study periods were calculated using multivariable logistic regression.
Results
A total of 311 children met inclusion criteria (126 symptomatic and 185 asymptomatic). The overall pneumococcal colonization (23% vs. 13%, p = 0.03) and SARS-CoV-2 positivity (34.1% vs. 17.8%, p = 0.001) rates were higher in symptomatic children than in asymptomatic children. The odds of pneumococcal colonization of symptomatic (Period 2: OR 1.09; 95% CI 0.33–3.64, and Period 3: OR 0.46; 95% CI 0.13–1.59) and asymptomatic (Period 2: OR 0.55; 95% CI 0.18–1.7, and Period 3: OR 0.37; 95% CI 0.11–1.2) children did not statistically differ across study periods. Pneumococcal colonization density was also not different across study periods among study groups. Overall, non-PCV15, non-PCV20 serotypes were the most frequently identified serotypes (56.8%).
Conclusions
Pneumococcal colonization rates and density did not significantly change across study periods as adherence to non-pharmaceutical interventions gradually relaxed during the first year of the SARS-CoV-2 pandemic.
Introduction
Nasopharyngeal colonization with Streptococcus pneumoniae plays a pivotal role in the development of mucosal and invasive pneumococcal diseases (IPD). Children ≤5 years of age have the highest rates of pneumococcal colonization and contribute to horizontal dissemination of pneumococcus within communities. Daycare and school attendance, and other crowding conditions are considered risk factors for pneumococcal colonization [1]. Presence of a respiratory virus even in the absence of symptoms has been suggested to increase pneumococcal colonization rates and facilitate pneumococcal transmission and disease [2,3].
The surge of SARS-CoV-2 cases observed in March 2020 in the United States prompted the implementation of non-pharmaceutical interventions (NPI) (social distancing, frequent handwashing, wearing masks in public, and shelter-in-place orders) to reduce the spread of SARS-CoV-2 [4]. These interventions not only reduced the transmission of SARS-CoV-2 and other respiratory viruses, but also reduced the incidence of IPD and other bacterial infections [5]. In 2020, among 26 countries in six continents, the incidence of IPD decreased by 82% at 8 weeks following NPI implementation compared with corresponding data from 2018 and 2019 [5]. The causes of the altered dynamics of respiratory infections during the first years of the SARS-CoV-2 pandemic are incompletely understood. However, the changes in human behavior resulting from the implementation of NPI appeared to be a substantial factor that in turn lead to alterations in nasopharyngeal microbiota, changes in nasopharyngeal colonization rates and density, resurgence of endemic viruses outside their typical seasons, waning immunity, increase severity of rebounded viral and bacterial infections, etc. It has been hypothesized that the decrease in IPD incidence was secondary to an interruption in the transmission of S. pneumoniae leading to decrease rates of pneumococcal colonization. Studies from France, Belgium and Israel reported no differences in pneumococcal colonization in children before and during the SARS-CoV-2 pandemic, while a study from Serbia reported increased pneumococcal colonization during the pandemic [6–9]. However, there is little data on the effect of NPI on pneumococcal colonization in children in the United States. We hypothesized that pneumococcal colonization rates in children ≤ 5 years of age would have decreased with the implementation of NPI and then would gradually increase due to two factors: relaxation of NPI plus onset of the winter season in the Kansas City region. We evaluated serotype-specific pneumococcal colonization dynamics in children with and without respiratory symptoms during the first 12 months of SARS-CoV-2 pandemic.
Materials and methods
Study design
This is a retrospective cross-sectional study evaluating variations in pneumococcal colonization prevalence and serotype distribution among children aged ≤5 years with and without respiratory symptoms, and seeking care at Children’s Mercy - Kansas City (CMKC) hospital system during the first 12 months of SARS-CoV-2 pandemic (April 1, 2020 to March 31, 2021). The CMKC population coverage area and demographics are detailed in S1 Appendix in S1 File. The study period was divided in three 4-month periods: April/May/June/July (Period 1, spring-summer), August/September/October/November (Period 2, summer-fall) and December/January/February/March (Period 3, winter-spring). Period 1 represents the time of strict recommendations for NPI, e.g., social distancing, masking and shelter-in-place orders to reduce the spread of SARS-CoV-2. Period 2 represents the time when NPI were at least partially lifted in our region, with cities reopening to activities and children being back in daycare and school. Period 3 represents winter season with maintenance of partial NPI. NPI implementation and relaxation dates are listed in S2 Appendix in S1 File. This retrospective study was approved by the Institutional Review Board at CMKC with a waiver of consent and HIPAA authorization (#STUDY00001775) because research involved no more than minimal risk to the subjects.
Study population
Children aged ≤5 years who had a residual sample remaining from a nasal mid-turbinate specimen obtained for viral testing as part of routine care were included in this study. The asymptomatic group consisted of children without respiratory symptoms, but requiring SARS-CoV-2 testing. Two subgroups made up the asymptomatic group: (1) children requiring SARS-CoV-2 polymerase chain reaction (PCR) screening 72 hours prior to a scheduled surgery or a sedated procedure (pre-procedure) and exhibiting no respiratory symptoms (rhinorrhea, nasal congestion, cough, sore throat, earache, fever, respiratory distress/shortness of breath), and (2) children exposed to SARS-CoV-2 but exhibiting no respiratory symptoms as determined by the ordering provider. The symptomatic group consisted of children with respiratory symptoms that sought care at CMKC and required SARS-CoV-2 PCR and/or other viral PCR testing (BioFire® FilmArray® Respiratory Panel or Cepheid Xpert® Xpress Flu/RSV) as recommended per treating provider. SARS-CoV-2 testing required the ordering provider to select an indication for testing; the main categories included: asymptomatic exposed, asymptomatic pre-procedure, and symptomatic. Eligible children were identified by performing an electronic health record (EHR) query. Convenience sampling was used and stratified within the asymptomatic and symptomatic groups for each study period. Samples were obtained from ten different CMKC facilities across Kansas City (S1 Appendix in S1 File). This approach helped diversify the study population, reducing the risk that the convenience sample would represent only a narrow subset of the population. If subjects had multiple encounters generating a residual specimen during the study period, only the first encounter was considered for inclusion. Subject identification and data collection started on 01/11/2021.
Data collection
Demographic and clinical data such as age, sex, smoking exposure, pneumococcal immunization, complex chronic conditions, and viral testing result were obtained from EHR. The CMKC Integrated Care Solutions Data Repository (network that coordinates the medical care of patients enrolled in commercial insurance and Medicaid plans) was queried to obtain pneumococcal immunization and antibiotic use (within 4 weeks prior to sample collection), which could impact colonization results. Additional pneumococcal immunization information was also obtained from the Kansas and Missouri immunization registries. As reference, the coverage of ≥ 3 PCV13 doses by age 35 months in Missouri and Kansas was 88.6% (95% CI: 82.9–93.1%) and 92.7% (95% CI: 88–96.1%), respectively, in 2019 [10].
Pneumococcal colonization and molecular serotyping
Samples were collected using flocked swabs. Each swab was placed into a 3 ml tube of Universal Transport Medium and underwent viral testing; the sample residual was stored at –80⁰C. De-identified samples were used for the following research procedures. Deoxyribonucleic acid (DNA) was extracted using QIAGEN DNeasy (Germantown, Maryland, USA) automated extraction system and human-specific glyceraldehyde 3-phosphate dehydrogenase (GAPDH) real-time PCR was used to evaluate quality of samples. Real-time PCR assay targeting lytA gene was used for S. pneumoniae detection with a cycle threshold value ≤35 considered positive [11]. Positive samples for S. pneumoniae underwent sequential multiplex serotype-specific PCR testing which detects 40 pneumococcal serotype/serogroups (1, 2, 3, 4, 5, 6A/6B/6C/6D, 6C/6D, 7C/7B/40, 7F/7A, 8, 9N/9L, 9V/9A, 10A, 10F/10C/33C, 11A/11D, 12F/12A/12B/44/46, 13, 14, 15A/15F, 15B/15C, 16F, 17F, 18C/18F/18B/18A, 19A, 19F, 20, 21, 22F/22A, 23A, 23B, 23F, 24F/24A/24B, 31, 33F/33A/37, 34, 35A/35C/42, 35B, 35F/47F, 38/25F/25A, 39) and each reaction includes the internal control gene cpsA [12]. Amplified PCR products were analyzed by gel electrophoresis. To calculate the colonization density, a standard curve of lytA was generated with known DNA concentrations from control strains, plotted against the cycle threshold values to calculate CFU/ml and then transform to log10.
Sample size
The pneumococcal colonization rate in children <5 years in the United States is ~ 30% [13,14]. Higher colonization rates (40−50%) have been reported in children with respiratory infections [3,15]. We hypothesized pneumococcal colonization rates during Period 1 would be lower than these previously reported rates due to interventions implemented to reduce the spread of SARS-CoV-2, followed by a gradual increase of 10% for asymptomatic group and 15% for symptomatic group in colonization rates from period to period. We used Cochran-Armitage trend test to calculate the sample size assuming the above trend of pneumococcal colonization rates, 80% power, and equal sample size per period. The total calculated sample size for asymptomatic group was 186 children (62 per period) and for symptomatic group was 126 children (42 per period).
Statistical analysis
Pearson’s chi-square and Fisher’s exact test were used for when examining bivariate associations between categorical variables. The overall proportion of children with pneumococcal colonization and vaccine serotypes were compared across study periods using the Cochran-Armitage test for trend. Pneumococcal colonization density was compared across study periods using the Kruskal-Wallis test and Wilcoxon Rank-sum test, where appropriate. Nearly all variables included in this analytic framework were categorical with the exception of colonization density. The colonization density variable exhibited a slight positive skewness, though we did not consider it substantial enough to necessitate any data transformation or outlier exclusion. Given the relatively small number of positive specimens, we presented non-parametric comparisons.
We performed multivariable logistic regression modeling to examine the relationship between pneumococcal colonization and study periods. Variables included in the logistic regression models were identified a priori and included the following potential confounders: sex, exposure to tobacco smoke, complex chronic conditions, pneumococcal immunization, prior antibiotic use, and SARS-CoV-2 positive results. This list was further refined based on the strength of associations observed during the bivariate comparisons. The regression models were built incrementally, which allowed for an assessment for any change in coefficient directionality, significant changes in adjusted coefficients, and significant changes in standard errors. More robust multicollinearity assessments were not completed. Analyses were stratified into asymptomatic and symptomatic groups. No imputation methods were performed as missing data were limited to two variables: SARS-CoV-2 positivity (<1%) and PCV-13 vaccination status (<8%). Missing SARS-CoV-2 status applied only to 3 symptomatic participants. Missing PCV-13 vaccination records were handled by categorizing immunization into three groups: ≥ 3 doses, 0–2 doses, and no vaccine record. We further conducted an analysis excluding the participants without vaccine records, and the results remained consistent, indicating that missingness did not influence our results. Analyses were completed using Stata (StataCorp. Stata Statistical Software: Release 17. College Station, TX).
Results
Of the 312 samples, one (from the asymptomatic group) was GADPH negative and excluded from the analysis. Selected characteristics of the overall asymptomatic and symptomatic groups are presented in Table 1. All 185 asymptomatic children had SARS-CoV-2 testing performed; 152 (82.2%) prior to a sedated procedure (categories of procedure listed in S1 Table in S1 File) and 33 (17.8%) because they were SARS-CoV-2 exposed children exhibiting no symptoms. A significant proportion of asymptomatic children requiring a sedated procedure were younger than 2 years of age, male and had a complex chronic condition when compared to asymptomatic children exposed to SARS-CoV-2 (S2 Table in S1 File). SARS-CoV-2 testing was performed in 123/126 (97.6%) symptomatic children. Three symptomatic children were not tested for SARS-CoV-2, but they were included because they had a nasal mid-turbinate specimen for respiratory panel testing. Complex chronic conditions [16] (S3 Table in S1 File) were more common in the asymptomatic group than the symptomatic group (40.5% vs 23.8%, p = 0.002). For the symptomatic group, the most common symptoms documented were fever (65.9%), cough (61.9%), nasal congestion (53.2%), rhinorrhea (28.6%), vomiting (17.5%), shortness of breath (16.7%), diarrhea (13.5%), otalgia (7.9%), sore throat (6.3%), and headache (2.4%). Pneumococcal immunization data were not available for 24 children (7.7%). Among children with confirmed pneumococcal immunization data, 205/287 (71.4%) received ≥ 3 doses of 13-valent pneumococcal conjugate vaccine (PCV13) (120/168 asymptomatic and 85/119 symptomatic children). The overall SARS-CoV-2 positivity rates for asymptomatic and symptomatic children were 17.8% and 34.1% (p = 0.001), respectively. Within the asymptomatic group, the SARS-CoV-2 positivity rate among children exposed to SARS-CoV-2 was higher than among children requiring a sedated procedure (39.4% vs. 13.4%, p = 0.0004). However, pneumococcal colonization rates were similar in the two groups (S2 Table in S1 File). The overall pneumococcal colonization rates for asymptomatic and symptomatic children were 13.5% and 23% (p = 0.03), respectively. SARS-CoV-2 positivity rate was similar between colonized and non-colonized children (22.2% vs. 25.2%, p = 0.65). Of the 54 samples positive for pneumococcal colonization, 12 (22.2%) were also positive for SARS-CoV-2, 2 (3.7%) for rhinovirus/enterovirus, and 1 (1.8%) for coronavirus NL63.
Table 1. Selected characteristics of asymptomatic and symptomatic groups.
| Asymptomatic (N = 185) | Symptomatic (N = 126) | p value | ||
|---|---|---|---|---|
| Age | 0.043 | |||
| <2 years | 93 (50.3%) | 78 (61.9%) | ||
| 2-5 years | 92 (49.7%) | 48 (38.1%) | ||
| Sex | 0.441 | |||
| Female | 77 (41.6%) | 58 (46.0%) | ||
| Male | 108 (58.4%) | 68 (54.0%) | ||
| Race/ethnicity | <0.001 | |||
| Hispanic | 23 (12.4%) | 32 (25.4%) | ||
| Black (non-Hispanic) | 19 (10.3%) | 32 (25.4%) | ||
| White (non-Hispanic) | 126 (68.1%) | 54 (42.9%) | ||
| Other/unknown | 17 (9.2%) | 8 (6.3%) | ||
| Complex chronic condition | 0.002 | |||
| No | 110 (59.5%) | 96 (76.2%) | ||
| Yes | 75 (40.5%) | 30 (23.8%) | ||
| Smoking exposure | 0.407 | |||
| Yes | 24 (13.0%) | 12 (9.5%) | ||
| No | 154 (83.2%) | 106 (84.1%) | ||
| Unknown | 7 (3.8%) | 8 (6.3%) | ||
| Number of PCV13 doses | 0.505 | |||
| 0 | 17 (9.2%) | 14 (11.1%) | ||
| 1 | 22 (11.9%) | 10 (7.9%) | ||
| 2 | 9 (4.9%) | 10 (7.9%) | ||
| 3 | 32 (17.3%) | 26 (20.6%) | ||
| 4 | 88 (47.6%) | 59 (46.8%) | ||
| Unknown | 17 (9.2%) | 7 (5.6%) | ||
| Number of PPSV23 doses | 0.335 | |||
| 0 | 167 (90.3%) | 117 (92.8%) | ||
| 1 | 1 (0.5%) | 2 (1.6%) | ||
| Unknown | 17 (9.2%) | 7 (5.6%) | ||
| Prior antibiotic use | 0.307 | |||
| No | 184 (99.5%) | 123 (97.6%) | ||
| Yes | 1 (0.5%) | 3 (2.4%) | ||
| Location of sample collection | 0.471 | |||
| Outpatient setting | 136 (73.5%) | 98 (77.8%) | ||
| Inpatient setting | 49 (26.5%) | 28 (22.2%) | ||
| Respiratory panel testing | <0.001 | |||
| Not tested | 173 (93.5%) | 100 (79.4%) | ||
| Negative | 11 (5.9%) | 18 (14.3%) | ||
| Positivea | 1 (0.5%) | 8 (6.3%) | ||
| SARS-CoV-2 PCR testb | 0.001 | |||
| Negative | 152 (82.2%) | 80 (65.0%) | ||
| Positive | 33 (17.8%) | 43 (35.0%) | ||
| Pneumococcal colonization | 0.03 | |||
| Negative | 160 (86.5%) | 97 (77.0%) | ||
| Positive | 25 (13.5%) | 29 (23.0%) | ||
aRhinovirus/enterovirus (n = 5), coronavirus NL63 (n = 2), respiratory syncytial virus (n = 1), and adenovirus (n = 1).
bThree patients in the symptomatic group did not have a SARS-CoV-2 test done; thus the denominator used was 123. One of these three patients was positive for rhinovirus/enterovirus. PPSV23, 23-valent pneumococcal polysaccharide vaccine
The pneumococcal colonization rate did not change across study periods (Fig 1). The odds of colonization of asymptomatic children were similar during Period 2 (odds ratio [OR] 0.55; 95% confidence interval [CI] 0.18–1.7) and Period 3 (OR 0.37; 95% CI 0.11–1.2), using Period 1 as reference and after adjusting for age, sex, SARS-CoV-2 results, presence of a complex chronic condition, and ≥ 3 doses of PCV13. When compared with Period 1, the odds of colonization of symptomatic children were also similar for Period 2 (OR 1.09; 95% CI 0.33–3.64) and Period 3 (OR 0.46; 95% CI 0.13–1.59).
Fig 1. Pneumococcal colonization rates in asymptomatic and symptomatic patients by study periods.
Pneumococcal serotype was confirmed in 44/54 (81.5%) positive samples. The most common serotypes were 15A/15F (n = 8), 11A/11D (n = 6), 35B (n = 5), 15B/15C (n = 4), 19F (n = 3), 23A (n = 3), and 23B (n = 3). Seven of 44 (15.9%) serotypes were PCV13 serotypes (19F [n = 3], 19A [n = 2] and 3 [n = 2]). Six of seven (85.7%) patients colonized with a PCV13 serotype had received ≥ 3 doses of PCV13. Two isolates (4.5%) were among the serotypes unique to PCV15 and 10 isolates (22.7%) among the serotypes unique to PCV20. Twenty-five isolates (56.8%) were serotypes that would not be covered by PCV13, PCV15 or PCV20.
Pneumococcal density was not different across study periods for the asymptomatic and symptomatic groups (Fig 2). Colonization density was also not impacted by SARS-CoV-2 positivity (Fig 3). The highest colonization density was observed among unique PCV20 serotypes (Fig 4).
Fig 2. Pneumococcal colonization density across the three study periods for the asymptomatic and symptomatic groups.
Fig 3. Pneumococcal colonization density by SARS-CoV-2 results.
Fig 4. Pneumococcal colonization density by vaccine serotype categories.
PCV15 category includes serotypes unique to PCV15 (non-PCV13), PCV20 category includes serotypes unique to PCV20 (non-PCV13 and non-PCV15), non-PCV category includes serotypes not covered by PCV13, PCV15 or PCV20, and unknown represents the samples without a confirmed serotype. A post-hoc analysis was completed following the Kruskal-Wallis test, with p-values adjusted for multiple comparisons. A significant difference in colonization density was observed between PCV20 and non-PCV categories (p = 0.017).
Discussion
In our study, pneumococcal colonization rates and density did not significantly change across study periods despite NPI gradually relaxing and winter season starting over the first year of the SARS-CoV-2 pandemic. However, our results may have been impacted by our relatively small sample size and the absence of a pre-pandemic comparison group. Pneumococcal colonization (23% vs. 13%, p = 0.03) and SARS-CoV-2 positivity (34.1% vs. 17.8%, p = 0.001) rates were higher for children in the symptomatic group than in children exhibiting no respiratory symptoms, which is consistent with previous studies [3,15]. As might be expected, the most common detected serotypes were serotypes not covered by PCV13, PCV15 or PCV20 (56.8%). Only 7 children were colonized with PCV13 serotypes, but 6 had received ≥ 3 doses of PCV13. Interestingly, colonization density for PCV13 serotypes was lower compared with higher-valency vaccine serotypes. This was also not unexpected because PCV15 and PCV20 were not in use prior to or during the study period.
Our study did not include a pre-pandemic control group, and pneumococcal colonization rates for Kansas City children under 5 years of age before 2020 are unavailable for comparison. The most comparable pre-pandemic data comes from a study conducted in St. Louis, Missouri, involving young mostly asymptomatic children who required sedation for procedures or minor surgeries from 2013 to 2016 [17]. This study found a 30.8% and 28.9% pneumococcal colonization prevalence among children 0–2 years old and >2–6 years old, respectively. Using these data as a pre-pandemic comparison, the pneumococcal colonization rates observed in Period 1 of our study were substantially lower for the asymptomatic group (16.1%) than the symptomatic group (23.8%); lower rates in the asymptomatic group were present in each study period. It is noteworthy that detected colonization rates in our study were lower despite pneumococcal identification using PCR-based methods, which are more sensitive that the culture-based methods used in the St. Louis study. However, without local rates from the months immediately preceding the implementation of the shelter-at-home order, it is challenging to contextualize our observed rates from Period 1.
In 2020, IPD rates were historically low in the United States and worldwide [5,18–20]. The greatest reduction in IPD cases occurred during the second and third quarters of 2020 coinciding with total lockdown [19,21]. This was thought to be secondary to a decrease in pneumococcal colonization due to stringent NPI. However, studies from France, Israel, Belgium and South Africa reported no change in pneumococcal colonization rates in young children before and during the pandemic, proposing that the temporary decline in IPD was associated with interruption of the transmission of respiratory viruses [6–8,22].
To our knowledge, the only United States study evaluating pneumococcal colonization in children during the pandemic included adult data in the analysis. This was a community-based household surveillance study evaluating the association of respiratory viruses and pneumococcal semiquantitative colonization density from November 2019 to June 2021 in Seattle, Washington [23]. Their overall (children plus adult) pneumococcal colonization rate from November 2019 to March 2020 was 12%; and declined to 6% and 5% in the March-December 2020 and January-June 2021 periods, respectively [23]. Their lower rates may have been due to the fact that colonization rates vary and tend to decrease with age.
Our study evaluated pneumococcal colonization exclusively in children ≤ 5 years of age in three periods after NPI implementation. The odds of pneumococcal colonization remained relatively stable across study periods for symptomatic and asymptomatic children. A slightly lower pneumococcal colonization rate in Period 3 (December 2020 – March 2021) was not statistically lower compared to the other two periods for either study groups. We anticipated a higher colonization rate in Period 3 compared to Period 1 based on the previous knowledge that pneumococcal colonization rates and IPD increase during winter when seasonal respiratory viruses circulate [24,25]. That said, influenza, respiratory syncytial virus, and human metapneumovirus circulation was markedly low during the winter of 2020–2021 [26], which may have contributed to an overall unchanged colonization rate and the absence of the usual seasonal-virus-induced uptick in pneumococcal colonization in the third period of our study. Other studies also showed a somewhat lower colonization rate towards the end of 2020; however, direct comparisons with our study are challenging due to differences in pneumococcal identification techniques, age groups, overall timeframes, and specific timing of NPI [7,23]. In Israel, pneumococcal colonization rates from June 2020 to February 2021 were not significantly different from previous years, except for lower rates in the October-December 2020 period [7]. In contrast, Belgian data showed high colonization rates (67.4%) during the winter 2020–2021 (November to March) in children <3 years of age, a rate comparable to previous winters; however, these investigators did not assess colonization rates during early pandemic months [8].
In Serbia, a significant increase in pneumococcal colonization was noted from February-March 2020 to September-November 2020 (24.9% vs. 32.1% p = 0.0085) and to April-June 2021 (24.9% vs. 37.7%, p < 0.0001) [9]. However, this study design differed from our study, included only 2–5 year old children and combined data from symptomatic and asymptomatic children. Further, the largest Serbian increase was not detected until April-June 2021, after the COVID-19 rates in children had plunged. One could speculate that our study timeframe stopped short of a potential increase in colonization as was seen in Serbia. Children attending daycare were also over represented among the colonized group, about 84.6% of their participants [9]. We did not have access to daycare attendance data, which is a known risk factor for transmission of upper respiratory tract pathogens. However, during the initial months of NPI, daycare facilities were closed with few exceptions. As a result, daycare attendance during Period 1 was likely lower than in subsequent periods when daycare centers were allowed to reopen. That said, we were unable to evaluate whether the unchanged colonization rate across study periods was related to limited daycare attendance, despite the relaxation of NPIs in Periods 2 and 3.
Another confounding factor may have been the fact that single sample snapshot surveillance colonization rates necessarily detect colonization that has been present for varying lengths of time. Colonization has been reported to last a median of 60 days, with some persisting for 12 months [27]. So, colonization detected later during period 1 (three months into onset of the pandemic) may have persisted into the second period, so that rates did not decrease as much as we anticipated. It is also possible that strict NPI may have not been in place long enough to see a significant impact on pneumococcal colonization rates. However, some families may have remained more cautious about the level of exposures in preschool age children despite gradual relaxation of NPI, so that recolonization was less likely.
The few studies reporting pneumococcal serotype distributions revealed no difference from previous years [6,7,9]. We reported a predominance of non-PCV15, non-PCV20 serotypes with 15A and 35B among the most common serotypes; our data are consistent with a United States colonization study conducted from 2021 to 2023 [28]. Some of these non-PCV15, non-PCV20 serotypes (serotypes 35B, 15A, 23A, 23B) are also among the most common serotypes causing acute otitis media and IPD in children in the United States [29]. Although these relevant serotypes are not included in current pediatric PCVs, they are included in a new 21-valent PCV recently approved for use in the adult population and under evaluation for pediatric use.
Regarding overall pneumococcal colonization density, we did not observe a difference across the three study periods in either study group. This is consistent with findings in Seattle and Israel where colonization density did not change after NPI implementation [7,23]. Yet, studies from South Africa and Vietnam reported a reduction in pneumococcal colonization density even when colonization rates remain unchanged in those areas; suggesting that the IPD decline may have been secondary to a decline in S. pneumoniae density after NPI implementation [22,30].
We observed a trend toward higher density among serotypes in the higher valency vaccines. Indeed, pneumococcal colonization density in our study among PCV13 serotypes was lower than that among serotypes unique to PCV15 and PCV20. This lower density may be related to a protective effect of PCV13 vaccination as most children (71.4%) included in our study had received ≥ 3 doses of PCV13. However, studies have reported inconsistent results regarding the effect of PCVs on pneumococcal nasopharyngeal density in children [31]. The slightly higher colonization density observed in the asymptomatic group in Period 3 seems to be driven by serotypes unique to PCV20, but our numbers are too small in the subcategories to draw any conclusions.
Pneumococcal colonization density can be influenced by viral co-detection, which our study was not designed to thoroughly evaluate. Almost all children were tested for SARS-CoV-2 (308/311), but a considerable proportion were not evaluated for other viruses, potentially impacting density results. We found a 3.9% (12/308) co-detection of S. pneumoniae and SARS-CoV-2, and no difference in colonization density based on SARS-CoV-2 positive results. This finding may suggest independent pathogenic dynamics between S. pneumoniae and SARS-CoV-2, a distinct host-pathogen interaction, or methodological factors such as the timing of sample collection during the illness. Studies from Seattle and Belgium did not identify any S. pneumoniae and SARS-CoV-2 co-detection; however, co-detection with other viruses was reported in Seattle [8,23]. Our somewhat higher virus co-detection may indicate that children included in our study had more social interactions or healthcare related exposures due to our patients’ relatively high rate of complex chronic conditions which likely required more frequent medical attention, particularly in the asymptomatic group. Or it may indicate a less than complete adherence to NPI.
The strengths of our study include the parallel evaluation of serotype-specific pneumococcal colonization in two groups of children based on the presence or absence of respiratory symptoms during the SARS-CoV-2 pandemic, the availability of pneumococcal immunization data for most children (92.3%) and the use of quantitative methods to determine S. pneumoniae density. Nonetheless, the findings in this study are subject to some limitations. First, the number of subjects included in each study period may not have been large enough to detect a significant difference. Second, we do not have data on adherence to NPI or social behavior practices of the children or their household contacts due to the retrospective design of the study. It is possible that some families were not able to strictly adhere to NPI at the beginning of the pandemic (e.g., essential workers); while other families decided to continue limiting their social interactions even when schools and daycares reopened. Variation in NPI adherence and care-seeking behaviors could have introduced potential confounders. Third, we compared pneumococcal colonization rates during discrete time periods across the first year of the SARS-CoV-2 pandemic after NPI implementation; however, the absence of pre-pandemic baseline pneumococcal colonization rate data limits the trend analysis of our results. Fourth, a large percentage of children in this study were not evaluated for respiratory viruses other than SARS-CoV-2. Thus, undetected respiratory virus co-infection could have impacted pneumococcal colonization density results, particularly for the asymptomatic group. In addition, our study timeframe did not continue to the time when other viral infection rates increased. Fifth, this was a cross-sectional study; therefore, repeat samples from each child over time were not obtained to evaluate whether NPI impacted the acquisition of pneumococcus or new pneumococcal serotypes. Lastly, this is a single-center study and we used a convenience sample, both of which limit the external validity of our results, as they cannot be readily generalized to other areas where NPI and PCV coverage may differ. Convenience sampling may have also contributed to over-representation of children with complex chronic conditions compared to other studies.
In conclusion, we did not detect significant decrease in pneumococcal colonization after NPI was in place for several months, nor did we detect an increase when NPI was relaxed and winter season effects could have been expected. The anticipated changes in overall pneumococcal colonization rates and density among children with and without respiratory symptoms were not observed despite shifting strictness of recommended NPI during the first year of the SARS-CoV-2 pandemic. Even though our results may have been impacted by a relatively small sample size in each study period, our results reflect those of studies using national surveillance systems which also did not report significant changes in pneumococcal colonization. Integrating more completely documented human behavioral monitoring in future prospective multicenter studies is of utmost importance for thoroughly understanding the impact of NPI on the complex pneumococcal colonization dynamics, as well as the viral-bacterial interactions in pneumococcal transmission and disease pathogenesis.
Supporting information
S2 Appendix. Dates of community mitigation measures implemented in the Kansas City Metro Area. S1 Table. Procedure categories for which asymptomatic group required SARS-CoV-2 testing, by pneumococcal colonization status. S2 Table. Characteristics of asymptomatic participants. S3 Table. Complex chronic condition categories identified among patients in the asymptomatic and symptomatic groups. S1 Data. Minimal anonymized dataset.
(ZIP)
Acknowledgments
We thank Ana Contreras, CMKC clinical research coordinator, for her assistance with reviewing Missouri and Kansas immunization registries; and Rachel Moran, CMKC Integrated Care Solutions data analytics manager, for retrieving pneumococcal immunization and antibiotic use data.
Data Availability
All relevant data are within the manuscript and its Supporting Information files.
Funding Statement
This work was supported by Merck Investigator Studies Program [MISP 60517 to L.O.] (https://misp-investigator-studies.com). The study sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; writing of the manuscript or decision to submit the manuscript for publication.
References
- 1.Bogaert D, Engelen MN, Timmers-Reker AJ, Elzenaar KP, Peerbooms PG, Coutinho RA, et al. Pneumococcal carriage in children in The Netherlands: a molecular epidemiological study. J Clin Microbiol. 2001;39(9):3316–20. doi: 10.1128/JCM.39.9.3316-3320.2001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Fan RR, Howard LM, Griffin MR, Edwards KM, Zhu Y, Williams JV, et al. Nasopharyngeal pneumococcal density and evolution of acute respiratory illnesses in young children, Peru, 2009-2011. Emerg Infect Dis. 2016;22(11):1996–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.DeMuri GP, Gern JE, Eickhoff JC, Lynch SV, Wald ER. Dynamics of bacterial colonization With Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis during symptomatic and asymptomatic viral upper respiratory tract infection. Clin Infect Dis. 2018;66(7):1045–53. doi: 10.1093/cid/cix941 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.CDC COVID-19 Response Team. Coronavirus disease 2019 in children - United States, February 12-April 2, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(14):422–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brueggemann AB, Jansen van Rensburg MJ, Shaw D, McCarthy ND, Jolley KA, Maiden MCJ, et al. Changes in the incidence of invasive disease due to Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis during the COVID-19 pandemic in 26 countries and territories in the Invasive Respiratory Infection Surveillance Initiative: a prospective analysis of surveillance data. Lancet Digit Health. 2021;3(6):e360–70. doi: 10.1016/S2589-7500(21)00077-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rybak A, Levy C, Angoulvant F, Auvrignon A, Gembara P, Danis K, et al. Association of nonpharmaceutical interventions during the COVID-19 pandemic with invasive pneumococcal disease, pneumococcal carriage, and respiratory viral infections among children in France. JAMA Netw Open. 2022;5(6):e2218959. doi: 10.1001/jamanetworkopen.2022.18959 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Danino D, Ben-Shimol S, van der Beek BA, Givon-Lavi N, Avni YS, Greenberg D, et al. Decline in pneumococcal disease in young children during the coronavirus disease 2019 (COVID-19) pandemic in Israel associated with suppression of seasonal respiratory viruses, despite persistent pneumococcal carriage: a prospective cohort study. Clin Infect Dis. 2022;75(1):e1154–64. doi: 10.1093/cid/ciab1014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Willen L, Ekinci E, Cuypers L, Theeten H, Desmet S. Infant pneumococcal carriage in Belgium not affected by COVID-19 containment measures. Front Cell Infect Microbiol. 2022;11:825427. doi: 10.3389/fcimb.2021.825427 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Petrović V, Milosavljević B, Djilas M, Marković M, Vuković V, Andrijević I, et al. Pneumococcal nasopharyngeal carriage in children under 5 years of age at an outpatient healthcare facility in Novi Sad, Serbia during the COVID-19 pandemic. IJID Reg. 2022;4:88–96. doi: 10.1016/j.ijregi.2022.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Centers for Disease Control and Prevention. ChildVaxView. Vaccination coverage among young children (0 – 35 Months). Accessed 2025 February 24.
- 11.Carvalho MaG, Tondella ML, McCaustland K, Weidlich L, McGee L, Mayer LW, et al. Evaluation and improvement of real-time PCR assays targeting lytA, ply, and psaA genes for detection of pneumococcal DNA. J Clin Microbiol. 2007;45(8):2460–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.da Gloria Carvalho M, Pimenta FC, Jackson D, Roundtree A, Ahmad Y, Millar EV, et al. Revisiting pneumococcal carriage by use of broth enrichment and PCR techniques for enhanced detection of carriage and serotypes. J Clin Microbiol. 2010;48(5):1611–8. doi: 10.1128/JCM.02243-09 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Reisman J, Rudolph K, Bruden D, Hurlburt D, Bruce MG, Hennessy T. Risk factors for pneumococcal colonization of the nasopharynx in Alaska native adults and children. J Pediatric Infect Dis Soc. 2014;3(2):104–11. doi: 10.1093/jpids/pit069 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Desai AP, Sharma D, Crispell EK, Baughman W, Thomas S, Tunali A, et al. Decline in Pneumococcal nasopharyngeal carriage of vaccine serotypes after the introduction of the 13-valent pneumococcal conjugate vaccine in children in Atlanta, Georgia. Pediatr Infect Dis J. 2015;34(11):1168–74. doi: 10.1097/INF.0000000000000849 [DOI] [PubMed] [Google Scholar]
- 15.Xu Q, Wischmeyer J, Gonzalez E, Pichichero ME. Nasopharyngeal polymicrobial colonization during health, viral upper respiratory infection and upper respiratory bacterial infection. J Infect. 2017;75(1):26–34. doi: 10.1016/j.jinf.2017.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Feudtner C, Feinstein JA, Zhong W, Hall M, Dai D. Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation. BMC Pediatr. 2014;14:199. doi: 10.1186/1471-2431-14-199 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McFarland M, Szasz TP, Zhou JY, Motley K, Sivapalan JS, Isaacson-Schmid M, et al. Colonization with 19F and other pneumococcal conjugate vaccine serotypes in children in St. Louis, Missouri, USA. Vaccine. 2017;35(34):4389–95. doi: 10.1016/j.vaccine.2017.06.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.McNeil JC, Flores AR, Kaplan SL, Hulten KG. The indirect impact of the SARS-CoV-2 pandemic on invasive group A streptococcus, Streptococcus pneumoniae and Staphylococcus aureus infections in Houston area children. Pediatr Infect Dis J. 2021;40(8):e313–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ricketson LJ, Kellner JD. Changes in the incidence of invasive pneumococcal disease in Calgary, Canada, during the SARS-CoV-2 pandemic 2020-2022. Microorganisms. 2023;11(5):1333. doi: 10.3390/microorganisms11051333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Teng JLL, Fok KMN, Lin KPK, Chan E, Ma Y, Lau SKP, et al. Substantial decline in invasive pneumococcal disease during coronavirus disease 2019 pandemic in Hong Kong. Clin Infect Dis. 2022;74(2):335–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ciruela P, Soldevila N, Garcia-Garcia JJ, Gonzalez-Peris S, Diaz-Conradi A, Redin A. Effect of COVID-19 pandemic on invasive pneumococcal disease in children, Catalonia, Spain. Emerg Infect Dis. 2022;28(11):2321–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Olwagen CP, Downs SL, Izu A, Tharasimbi L, Van Der Merwe L, Nunes MC, et al. Bacterial nasopharyngeal colonisation in children in South Africa before and during the COVID-19 pandemic: an observational study. Lancet Microbe. 2024;5(1):e34–42. doi: 10.1016/S2666-5247(23)00260-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bennett JC, Emanuels A, Heimonen J, O’Hanlon J, Hughes JP, Han PD, et al. Streptococcus pneumoniae nasal carriage patterns with and without common respiratory virus detections in households in Seattle, WA, USA before and during the COVID-19 pandemic. Front Pediatr. 2023;11:1198278. doi: 10.3389/fped.2023.1198278 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Talbot TR, Poehling KA, Hartert TV, Arbogast PG, Halasa NB, Edwards KM. Seasonality of invasive pneumococcal disease: temporal relation to documented influenza and respiratory syncytial viral circulation. Am J Med. 2005;118(3):285–91. [DOI] [PubMed] [Google Scholar]
- 25.Ampofo K, Bender J, Sheng X, Korgenski K, Daly J, Pavia AT, et al. Seasonal invasive pneumococcal disease in children: role of preceding respiratory viral infection. Pediatrics. 2008;122(2):229–37. doi: 10.1542/peds.2007-3192 [DOI] [PubMed] [Google Scholar]
- 26.Olsen SJ, Winn AK, Budd AP, Prill MM, Steel J, Midgley CM, et al. Changes in influenza and other respiratory virus activity during the COVID-19 pandemic - United States, 2020-2021. MMWR Morb Mortal Wkly Rep. 2021;70(29):1013–9. doi: 10.15585/mmwr.mm7029a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Nelson KN, Grijalva CG, Chochua S, Hawkins PA, Gil AI, Lanata CF, et al. Dynamics of colonization of Streptococcus pneumoniae strains in healthy peruvian children. Open Forum Infect Dis. 2018;5(3):ofy039. doi: 10.1093/ofid/ofy039 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kaur R, Schulz S, Sherman A, Andrejko K, Kobayashi M, Pichichero M. Anticipated effects of higher-valency pneumococcal conjugate vaccines on colonization and acute otitis media. Pediatr Infect Dis J. 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kaur R, Gierke R, McGee L, Gonzalez E, Kobayashi M, Pichichero M, et al. Pneumococci isolated from children in community-based practice differ from isolates identified by population- and laboratory-based invasive disease surveillance. J Infect Dis. 2024;230(5):1243–52. doi: 10.1093/infdis/jiae184 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Nation ML, Manna S, Tran HP, Nguyen CD, Vy LTT, Uyen DY, et al. Impact of COVID-19 nonpharmaceutical interventions on pneumococcal carriage prevalence and density in Vietnam. Microbiol Spectr. 2023;11(1):e0361522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Jagne I, von Mollendorf C, Wee-Hee A, Ortika B, Satzke C, Russell FM. A systematic review of pneumococcal conjugate vaccine impact on pneumococcal nasopharyngeal colonisation density in children under 5 years of age. Vaccine. 2023;41(19):3028–37. doi: 10.1016/j.vaccine.2023.03.063 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
S2 Appendix. Dates of community mitigation measures implemented in the Kansas City Metro Area. S1 Table. Procedure categories for which asymptomatic group required SARS-CoV-2 testing, by pneumococcal colonization status. S2 Table. Characteristics of asymptomatic participants. S3 Table. Complex chronic condition categories identified among patients in the asymptomatic and symptomatic groups. S1 Data. Minimal anonymized dataset.
(ZIP)
Data Availability Statement
All relevant data are within the manuscript and its Supporting Information files.