Concurrent Bacteremia Due to Non-vaccine Serotype 24F Pneumococcus in Twins: A Rapid Increase in Serotype 24F-invasive Pneumococcal Disease and its High Invasive Potential

Abstract

Although concurrent bacteremia in siblings is rare, serotype 24F Streptococcus pneumoniae was isolated from the blood of twin 1-year-old girls within a 3-day interval, supporting the high invasive potential of this serotype. As the prevalence of childhood serotype 24F-invasive pneumococcal diseases increases in Europe and the Western Pacific Region, investigation and surveillance of this serotype are necessary.

The number of patients with invasive pneumococcal disease (IPD) has decreased after the implementation of the pneumococcal conjugate vaccine (PCV); however, the incidence of IPD due to non-vaccine serotypes (NVT) has increased worldwide due to serotype replacement.^1,2 PCV7 was licensed in Japan in February 2010 and was used on a voluntary basis until April 2013. In April 2013, PCV7 was approved for routine vaccination in Japan, and the vaccine was switched to 13-valent PCV (PCV13) in October 2013. IPD due to NVT such as 24F and 15A emerged in 2013 and increased rapidly thereafter. As differences in invasiveness among serotypes have been reported, studies on the virulence of NVT have become vital in the PCV13 era.

Concurrent development of bacteremia is rare among invasive bacterial infections, with the exception of meningococcal infections and an outbreak of nosocomial catheter-associated bloodstream infections. Here, we report the first episode of concurrent bacteremia caused by Streptococcus pneumoniae in twins. The concurrent onset of IPD in siblings, including twins, could provide clues to the invasiveness of specific serotypes.

CASE REPORT

A 22-month-old girl, one of a pair of identical twins, presented to our emergency department due to a febrile seizure in March 2016. She had rhinorrhea that had persisted for 2 weeks. Her past history was unremarkable. Her axillary body temperature was 39.0°C, and tonic-clonic seizures ceased spontaneously within 5 minutes. On physical examination, her consciousness levels appeared normal, and no other abnormal findings were noted. Laboratory tests indicated bacterial infection with an increased white blood cell count of 32,220/μL, and a C-reactive protein level of 4.4 mg/dL. Immunologic tests were normal as below: IgG 752 mg/dL, IgA 40 mg/dL, IgM 173 mg/dL, C3 103 mg/dL, C4 11 mg/dL, CH50 21.3 U/mL, CD4 57.9%, CD8 24.0% and CD4/8 2.41. Cerebrospinal fluid (CSF) findings were normal. There was no abnormal finding on the chest radiograph, and the spleen appeared normal on abdominal ultrasonography. After taking cultures from the blood, urine, pharynx and CSF, ceftriaxone (CTRX) at a dose of 120 mg/kg per day was started. S. pneumoniae was isolated from blood and pharyngeal specimens, while CSF and urine cultures were negative for this bacterial species.

Three days after onset of the first case, her identical twin also presented with fever and rhinorrhea. Her physical examination findings were unremarkable. Laboratory tests showed similar white blood cell counts and C-reactive protein levels—22,600/μL and 3.2 mg/dL, respectively—to her sister. Immunologic tests were normal as below: IgG 780 mg/dL, IgA 42 mg/dL, IgM 161 mg/dL, C3 117 mg/dL, C4 13 mg/dL, CH50 36.6 U/mL, CD4 48.0%, CD8 23.5% and CD4/8 2.04. Because her sister had received a diagnosis of IPD, a CTRX dose of 120 mg/kg per day was started immediately after taking cultures. S. pneumoniae was also identified in the blood culture but not in pharyngeal cultures.

S. pneumoniae strains isolated from blood samples from both patients were penicillin-susceptible, serotype 24F and sequence type 5496. Serotype was determined using the Quellung reaction with pneumococcal antisera (Statens Serum Institut, Copenhagen, Denmark). Multilocus sequence typing was performed and sequence type was identified using the multilocus sequence typing online database (https://pubmlst.org/spneumoniae/). The results of the antimicrobial susceptibility tests were identical. Minimum inhibitory concentrations of the antimicrobials tested were as follows: penicillin G ≤0.015 µg/mL; ampicillin ≤0.03 µg/mL; cefotaxime ≤0.03 µg/mL; meropenem ≤0.008 µg/mL; erythromycin ≥8 µg/mL; clindamycin ≥8 µg/mL; and vancomycin 0.25 µg/mL.

Both twins had been vaccinated 4 times with PCV13, with the last vaccinations given 6 months prior to admission. Both patients improved after a 7-day chemotherapy regimen with CTRX. Neither of them showed any signs of primary immunodeficiency disease and had family history of it. They did not develop any other invasive bacterial disease for 3 years after discharge. We obtained consent from the parents of the patients to report this case.

INCIDENCE OF INVASIVE PNEUMOCOCCAL DISEASE

We have tested children for IPD since 2008 in Kagoshima Prefecture, Japan. This surveillance was performed in part by the Research Program on Emerging and Re-emerging Infectious Diseases from Japan Agency for Medical Research and Development.² Annual incidence of IPD among children <5 years of age during January 2008 to June 2019 in Kagoshima Prefecture is shown in Figure 1A. IPD due to serotype 24F emerged in 2013, when PCV13 was approved for routine vaccination, and increased rapidly thereafter. As shown in Figure 1B, among a total of 50 NVT isolates from January 2008 to June 2019, the most frequent causative serotype was 24F (17 isolates; 34%), followed by 15A (10 isolates; 20%).

FIGURE 1.: A: Annual incidence of IPD among children <5 years of age from January 2008–June 2019 in Kagoshima Prefecture, Japan. B: Non-PCV13 serotype distribution of Streptococcus pneumoniae isolates from children <5 years of age with IPD from January 2008–June 2019 in Kagoshima Prefecture, Japan.

DISCUSSION

Although the risk of IPD within the same household is higher within 1 year of the primary IPD diagnosis, concurrent onset of IPD in siblings, including twins, has not been reported thus far.³ Although IPD cases in siblings with IRAK-4 (interleukin-1 receptor-associated kinase 4) deficiency have been reported, the serotypes of the causative pneumococci, serotypes 15C and 15A, were different.⁴

Serotype 24F pneumococcus is one of the most prevalent NVT causing IPD, but it is rarely isolated from healthy children and patients with otitis media or pneumonia in Japan, which suggests increased invasiveness of this serotype.⁵ A French study, which assessed the invasive potential of different serotypes by comparing childhood IPD serotypes with nasopharyngeal-colonizing serotypes, showed that only serotype 24F was associated with significant invasive disease potential after PCV13 implementation.⁶ A recent meta-analysis of the relative invasive potential of S. pneumoniae among children immunized with PCV also demonstrated that serotype 24F (as well as serotypes 8, 12F, and 33F) is at the upper end of the invasiveness spectrum based on epidemiologic data.⁷ In addition, a time series analysis by a French national survey demonstrated a sharp increase in pediatric pneumococcal meningitis from 2015 to 2016, which was primarily related to serotype 24F.⁸

In the present case, the source of infection was unknown. The patients did not attend a daycare center or have other siblings, and an outbreak of IPD was not noted in the neighborhood at the time. Nasopharyngeal cultures were not obtained from the parents. The first patient in this case study carried pneumococcus in the pharynx, although the serotype was unknown since the isolated strain was not stocked. Two probable hypotheses on the infection route in the second patient are acquisition from the same source as the first patient or transmission from the first patient. The incubation period of IPD is unknown, but less invasive strains likely colonize the host nasopharynx and therefore need longer periods to cause invasive disease. Considering that the twins developed bacteremia due to the same strain around the same time, it can be assumed that the incubation period from acquisition of pneumococcus to onset of bacteremia was relatively short in these cases and that the strain was highly invasive. Thus, we believe that this case supports the high invasive potential of serotype 24F pneumococcus.

Serotype 24F is one of the most prevalent NVT in Europe and the Western Pacific Region, but not in North America.⁹ In Japan, nationwide surveillance of IPD in children demonstrated that IPD caused by serotype 24F dramatically increased after the introduction of PCV13.^5,10 For these reasons, surveillance and in vitro analysis of the invasiveness of serotype 24F is essential in the PCV13-era. In addition, serotype 24F is the leading serotype in Europe 8 years after PCV13 implementation and none of the vaccines available (PCV7, PCV10 and PCV13) or in development (PCV15 and PCV20) contain this very invasive serotype. It is a big problem for these countries.

In conclusion, this case report of concurrent pneumococcal bacteremia in twins supports previous reports of the high invasive potential of serotype 24F pneumococcus. In the PCV13 era, serotype 24F is one of the most prevalent NVT causing IPD in many areas of the world, thus warranting an investigation of its virulence as well as its surveillance.