In the United States, approximately 100,000 people have sickle cell disease (SCD), a condition that increases the risk of invasive bacterial infections caused by encapsulated bacteria such as Streptococcus pneumoniae (pneumococcus) due to functional asplenia [1]. Invasive pneumococcal disease (IPD), such as pneumococcal bacteremia and meningitis, causes significant morbidity and mortality in children with SCD compared with children without SCD. To prevent IPD, the 2014 Evidence-Based Management of Sickle Cell Disease Expert Panel Report recommends that all children with SCD receive the recommended number of 13-valent pneumococcal conjugate vaccine (PCV13) doses followed by two doses of the 23-valent pneumococcal polysaccharide vaccine (PPSV23) spaced 3 years apart [2]. In addition, oral penicillin prophylaxis is recommended through age 5 years unless a child had a splenectomy or IPD [2]. Understanding that the recommendations by the Advisory Committee on Immunization Practices (ACIP) may change over time, the panel recommended that children with SCD should follow the most up-to-date recommendation by the ACIP. With the approval of the new, 20-valent pneumococcal conjugate vaccine (PCV20) for children, in 2023, the ACIP recommended PCV20 as an option to complete the recommended pneumococcal vaccine series for children with certain underlying conditions who were previously recommended to receive PPSV23, including children with SCD [3].

Incidence of IPD among children with SCD in the United States dramatically declined after introduction of the 7-valent PCV (PCV7) in 2000, but remained higher than the incidence in the general pediatric population [4–6]. In the years prior to introduction of PCV13 in 2010, nearly a quarter of all IPD among children with SCD was caused by a serotype not covered in PCV7 but covered in PCV13; however, children with SCD had a smaller proportion of cases of IPD caused by serotypes covered in PCV13 compared with healthy children [4]. The assessment of PCV13 introduction among children with SCD in the United States is limited [7]. Most of those affected by SCD are of African ancestry or self-identify as Black [2], and in 2018–2019, Black children had approximately two times higher IPD rates compared with White children [8]. Therefore, we sought to characterize IPD cases and estimate incidence of IPD in Black or African American (Black) children with and without SCD before and after the introduction of PCV13. In addition, we estimated the proportion of IPD cases that were caused by serotypes contained in the new 15-valent and 20-valent PCVs (PCV15, PCV20) that were approved for use among US children in 2022 and 2023, respectively [3, 9].

A case of IPD was defined as S. pneumoniae isolation or detection from a normally sterile site (e.g., blood or cerebrospinal fluid) among Black children less than age 18 years residing in Active Bacterial Core surveillance (ABCs) catchment areas during 2005–2019. ABCs is an active population- and laboratory-based surveillance system for invasive bacterial disease in the United States. ABCs catchment area expanded over time, and as of 2019, there were approximately 1.6 million Black children aged less than 18 years residing in the surveillance area [10]. Surveillance methods have been previously described [4, 11]. Briefly, ABCs collects demographic, hospitalization, outcome information, and relevant underlying conditions (including SCD) for all cases via medical chart review. All available pneumococcal isolates are tested for antibiotic susceptibility by the Clinical Laboratory and Standards Institute broth microdilution procedure (all isolates sent to the Minnesota Department of Health or isolates sent to CDC before 2015) or predicted by whole genome sequencing (2015–2019 for isolates sent to CDC). Serotyping is performed using Quellung reaction, polymerase chain reaction, or whole genome sequencing.

A total of 1725 cases of IPD occurred in Black children aged less than 18 years in the catchment area during 2005–2019 (Table 1). Of these, 119 (6.9%) had SCD identified during their medical chart review. An additional 355 (20.6%) had one or more IPD risk factor(s) without SCD. In all time periods, children with SCD and those with other risk factors were significantly older at the time of culture than children without IPD risk factors.

Our study showed that overall IPD incidence in Black children aged less than 18 years with SCD has dropped by 50% after PCV13 introduction, from 2005–2009 to 2015–2019. Despite the reduction, the IPD incidence continues to be dramatically higher than the incidence among Black children without SCD. A higher proportion of non-PCV13 serotypes among children with SCD compared to children without SCD could have contributed to these disparities. In the late-PCV13 period, greater than 95% of IPD cases among children with SCD were due to non-PCV13 serotypes; 41% were due to serotypes in PCV20, none were due to PPSV23-unique serotypes, and slightly more than half were due to serotypes that are not contained in any of the existing pneumococcal vaccines.

The 2014 Evidence-Based Management of Sickle Cell Disease recommends that children with SCD receive timely administration of both PCV and PPSV23 and oral penicillin prophylaxis to prevent IPD [2]. PCVs induce T-cell dependent immune response, which allows a more robust immune response compared with PPSV23. In children, PPSV23 is only recommended for children aged ≥2 years with underlying conditions that increase the risk of IPD, as T-cell independent antibody response to PPSV23 is poor in younger children [16]. Therefore, children who are recommended to receive PPSV23 are recommended to receive both PCV and PPSV23, with PPSV23 providing additional serotype coverage not contained in PCV. In our study, additional serotypes included in PCV15 and PCV20 (but not in PCV13) caused 19% and 41% of IPD, respectively, among children with SCD in the late-PCV13 period. On the other hand, there was no IPD case caused by PPSV23-unique serotypes among children with SCD during the entire study period. This indicates that use of new PCVs that were recommended in recent years could further reduce IPD incidence in this population. However, still over 50% of IPD in children with SCD was caused by a non-vaccine serotype (NVT) in the late-PCV13 time period, pointing to the need for additional prevention measures such as oral penicillin prophylaxis and the development of vaccines with broader coverage. Importantly, the proportion of IPD cases nonsusceptible to penicillin did not increase over time. Continued IPD surveillance is important to monitor the impact of new vaccines and efforts to increase penicillin prophylaxis utilization.

Oral penicillin prophylaxis is known to be an effective way in protecting children with SCD from IPD and has been recommended for high-risk children for nearly four decades [17, 18]. Children with sickle cell anemia, the most severe form of SCD, are recommended to receive oral penicillin prophylaxis until age 5 years unless a child had splenectomy or had IPD [2]. While we were unable to assess adherence to penicillin prophylaxis among children with IPD in our study, studies have reported that adherence to antibiotic prophylaxis among children with SCD has been suboptimal [19, 20]. This suggests the importance of continued education and emphasis on oral penicillin prophylaxis for children with indications, along with receiving recommended pneumococcal vaccine doses. Additionally, consistent with previous reports [4, 7], children with SCD had an older age of onset of IPD compared with children without SCD, and 25% of children with SCD and IPD in our study were of age ≥5 years in all time periods. In general, IPD incidence decreases with increasing age in children; however, the older age of onset seen in children with SCD suggests that the reduction in IPD incidence in children with SCD is smaller compared with children without SCD, supported by the higher IRR among children aged 5–17 years compared with children aged less than 5 years. It is possible the older age of IPD onset among children with SCD is related to cessation of penicillin prophylaxis after age 5 years. A survey of pediatric hematologists indicated 76% routinely recommended cessation of penicillin prophylaxis after 5 years of age, with 27% of pediatric hematologists who recommend cessation of penicillin prophylaxis after 5 years of age citing concerns related to emergence of antibiotic resistance as a reason for their recommendation [21]. In our study, there were no statistically significant differences in the proportion of IPD isolates nonsusceptible to parental or oral penicillin in children with SCD compared with those from other groups in any time period. However, the possibility of unintended consequences, such as development of resistance against non-pneumococcal pathogens or disruption of microbiome, cannot be ruled out [22]. Furthermore, for children who have not received PCV20, ACIP currently [3] recommends a second dose of PPSV23 after 5 years of the first dose, instead of 3 years as recommended by the Expert Panel Report [2]. Further research may be needed to identify the most appropriate age to cease penicillin prophylaxis and vaccination schedules for children with SCD.

Our study is subject to a few limitations. First, because the number of children living with SCD was not known, we estimated rates of SCD in the ABCs catchment area to define the SCD population. It is possible these estimates either over- or underestimated the total number of children with SCD in the surveillance catchment area. However, sensitivity analyses showed even in the most extreme assumptions of over- or underestimation of the number of children with SCD in the ABCs catchment area population, children with SCD continued to have a significantly higher incidence rate compared with children without SCD, with higher IRR among children aged 5–17 years than children aged less than 5 years. Second, we were unable to characterize the risk profile of IPD cases among children with SCD by SCD type. Third, we were unable to capture receipt of preventive measures (e.g., pneumococcal vaccination, penicillin prophylaxis, hydroxyurea utilization), limiting our ability to attribute noted differences between groups to differences in receipt of preventive measures or potential waning effectiveness of these preventive measures over time. Fourth, differences in case ascertainment efforts among children with and without SCD could have potentially overestimated the IRR between the two groups. In children with no IPD risk factors, the proportion of hospitalized children was lower compared with children with SCD during the study period, although the proportion increased over time. It has been reported that obtaining blood cultures from children with no IPD risk factors has become less common since PCV introduction, especially in the outpatient setting, as detection of S. pneumoniae from cultures obtained from otherwise healthy children has become rare [23]. On the other hand, national guidelines recommend febrile patients with SCD should be hospitalized and closely monitored whenever infection is suspected [2], making it potentially more likely for IPD cases to be detected and hospitalized in children with SCD compared with children without SCD. Fifth, as HIV status is not reportable in ABCs New York site, we could have underestimated the number of IPD cases among children with other IPD risk factors; however, IPD cases among children with HIV have been less common compared with non-HIV risk factors, so the degree of underestimation is expected to be small. Lastly, given the small number of IPD cases among children with SCD in the late-PCV13 period, certain estimates (e.g., serotype distribution) may have been unreliable.

This report demonstrates that introduction of PCV13 reduced the incidence of IPD in children with and without SCD, primarily due to reduction in PCV13-type IPD cases. However, our analyses show children with SCD are still at much greater risk of IPD infection than healthy children, pointing to the need for better prevention strategies to reduce the overall morbidity due to IPD in children with SCD. Use of new PCVs may help prevent additional disease burden in this vulnerable population.