Invasive Pneumococcal Disease in High-risk Children: A 10-Year Retrospective Study

Streptococcus pneumoniae is one of the most common bacterial pathogens of childhood.1 Health Canada reported an average incidence rate of 10.0 invasive pneumococcal disease (IPD) cases per 100,000 individuals of all ages annually between 2009 and 2019.2 The risk and severity of IPD are higher in persons with underlying immune deficiency.3–8 A key determinant for IPD and disease severity is its serotype, defined by its capsular polysaccharide. S. pneumoniae has >90 serotypes differing in incidence, virulence and antimicrobial resistance patterns.9 Ongoing population serotype-specific surveillance is important to inform optimal immunization approaches.

Multivalent pneumococcal vaccines are included in routine Canadian child immunization schedules. Pneumococcal polysaccharide vaccine 23 (PPSV23) was authorized in Canada in 1983, received public funding in 1995 and has since been recommended for adults >65 years of age and patients 2–64 years old with high-risk conditions that predispose to IPD.10 Pneumococcal conjugate vaccine 7 (PCV7) was authorized in June 2001, followed by PCV10 in November 2009 and PCV13 in November 2010. For children with high-risk conditions, the US Advisory Committee on Immunization Practices and Canadian National Advisory Committee on Immunization (NACI) currently recommend 4 doses of PCV13 in infancy (at 2, 4, 6 and 12–15 months of age) and 1 dose of PPSV23 ≥2 years of age (with an additional booster at 5 years of age for children with immunocompromising conditions or asplenia).11,12 Healthy, non-high-risk children are recommended to receive 3 doses of PCV13 and no PPSV23 vaccination.9 The primary objective of this study was to describe the IPD case numbers and clinical outcomes in pediatric patients with underlying high-risk conditions. Secondary objectives included determining risk factors for IPD, isolating serotype patterns and vaccination trends in high-risk pediatric patients.

METHODS

A retrospective chart review identified all high-risk pediatric patients <18 years of age admitted to the Hospital for Sick Children (SickKids) for IPD from January 1, 2009, to December 31, 2018, inclusive. IPD was defined as the detection of S. pneumoniae in culture or by molecular test from a normally sterile body site [eg, blood, cerebral spinal fluid (CSF), peritoneal fluid, synovial fluid and bone]. Pneumonia was defined by the detection of S. pneumoniae in bronchoalveolar lavage or blood culture with evidence of infiltrates on chest radiograph. IPD diagnoses were identified using both a search of microbiology records by a microbiologist for culture/molecular confirmation, and International Classification of Disease-10 discharge diagnostic codes related to IPD to ensure complete capture (Table, Supplemental Digital Content 1, https://links.lww.com/INF/E867).

High-Risk Patient Definition

All identified IPD cases were assessed for eligibility as a high-risk pediatric patients through a medical chart review. High-risk conditions included: transplant (solid organ and hematopoietic stem cell), malignancy, primary immunodeficiency, asplenia, sickle cell disease (SCD), cochlear implant, nephrotic syndrome, chronic lung disease, CSF leak, HIV and use of immunosuppressive therapy (chemotherapy, biologic drugs or immune-suppressing medications) in the 90 days preceding IPD. Although SCD patients have limited spleen function, we described SCD and asplenia from other causes separately.

Patient chart review was used to collect data on each patient’s underlying high-risk condition, demographic information, pneumococcal vaccination status, disease severity, clinical course and patient outcomes. If a patient experienced multiple IPD episodes, the most recent episode was analyzed in this study.

Serotype Analysis, Antimicrobial Susceptibility and Vaccine Status

Since January 1, 1995, pneumococcal isolates from sterile sites from patients at the Hospital for Sick Children have been routinely submitted to the Toronto Invasive Bacterial Disease Network (TIBDN). TIBDN performs active, population-based surveillance for IPD in the greater Toronto area. Isolates received are confirmed as S. pneumoniae and serotyped using latex agglutination and Quellung reaction.13–15 Antimicrobial susceptibility of pneumococcal isolates was performed using broth microdilution (Trek Sensititre STP6F) and interpreted to Clinical and Laboratory Standards Institute standards.16 Only isolates collected from sterile sites and with complete antibiotic susceptibility data were included. Nonmeningitis resistance breakpoints for penicillin and ceftriaxone were used unless the isolate was from a CSF sample and then meningitis breakpoints were used. As part of a well-established and long-running network, TIBDN collects vaccination data by contacting the patient’s parent/guardian and relevant primary care physicians. Patient chart data was used to identify vaccine status, vaccine type, and date of vaccine receipt. Children who received mixed conjugate vaccine programs (ie, PCV7, then PCV13) were considered fully vaccinated based on the total number of conjugate vaccines received.

Data Management and Statistical Analysis

Baseline and demographic characteristics were summarized using descriptive statistics. A two-tailed Fisher exact test was used to test the significance of the decrease in the proportion of IPD hospitalizations caused by conjugate vaccine-contained serotypes, comparing the pre to post-PCV13 periods. Vaccine effectiveness analysis was used to assess the proportion of children with IPD due to PCV-contained serotypes in vaccinated and unvaccinated children.17 Significance was defined as P = 0.05. Ethics approval was obtained from The Hospital for Sick Children (REB: 1000063186).

RESULTS

Over 10 years, 94 high-risk pediatric patients were hospitalized for IPD. The median age at diagnosis was 5.5 years (IQR: 3.6–7.9) and 55% of patients were female (Table 1). The number of IPD cases decreased over time. Six children had a previously documented hospitalization for IPD, of which 4 (67%) had active leukemia, the others received a transplant or were on immunosuppressive therapy. The first episode of IPD requiring hospitalization occurred a median of 224 days before the second episode (IQR: 70–239). Children with active malignancy accounted for the most cases (n = 33, 35%), followed by transplant recipients (n = 17, 18%) and SCD (n = 14, 15%). Of the children with active malignancy, the mean time after a cancer diagnosis to IPD hospitalization was 2.6 years. Chemotherapy was received within 3 months before IPD among 23 children, 6 received hematopoietic stem cell transplant, 4 were on other immune suppressive medications and none were on biologic treatments. Among the children with acute lymphoblastic leukemia (n = 17), 1 was receiving induction treatment, 13 were receiving maintenance and 3 had completed treatment. Of the children who received transplants, the mean time from transplant to IPD hospitalization was 3.3 years. One child experienced graft rejection, and all 17 received an immunosuppressant within 3 months of the IPD event (Table, Supplemental Digital Content 1, https://links.lww.com/INF/E867). Of the children with SCD, 9 (64%) were older than 5 years at the time of IPD. The median age of IPD among patients receiving antibiotic prophylaxis was 4.0 years, and 9.5 years among those who were off antibiotic prophylaxis.

TABLE 1. - Cohort Demographics and Clinical Course Demographic Information n (%) Sex  Female 51 (54)  Male 43 (46) Age at onset of first episode of IPD (years)  Mean 6.3  Median (IQR) 5.4 (3.6–7.7) High-risk group n (%)* Malignancy 33 (35)  Hematologic 23 (25)  Solid organ 10 (11) Solid organ transplant 17 (18)  Liver 9 (10)  Heart 4 (4)  Multi-visceral 3 (3)  Kidney 1 (1) Sickle cell disease 14 (15) Asplenia 9 (10) Hematopoietic stem cell transplant 8 (9) Nephrotic syndrome 8 (9) Chronic lung disease 7 (7) Use of immune suppressing medications 32 (34)  Chemotherapy 23 (25)  Tacrolimus 14 (15)  Corticosteroids 12 (13)  Mycophenolate mofetil 3 (3)  Sirolimus 3 (3) Primary immunodeficiency 4 (4) HIV 2 (2) Cochlear implant 1 (1) CSF leak 1 (1) Preventative measures n (%) Pneumococcal conjugate vaccination  Full conjugate vaccine pneumococcal immunization† 34 (36)  High-risk child recommendation: 4 doses 20 (21)  General healthy child recommendation: 3 doses 14 (15)  Partial conjugate vaccine pneumococcal immunization 5 (5)  2 doses 4 (4)  1 dose 1 (1)  Unknown number of pneumococcal conjugate vaccine dDoses 12 (13) PPSV23 vaccination (age eligible, N = 82) 3 (4) No pneumococcal vaccination‡ 15 (16) Unknown vaccine status 28 (30) Vaccine received§  PCV7 22 (23)  PCV10 2 (2)  PCV13 15 (16)  PPSV23 & PCV7 1 (1)  PPSV23 & PCV13 2 (2)  Unknown vaccine type 9 (10) Prophylactic pneumococcal antibiotic use 11 (12)  Penicillin 6 (6)  Amoxicillin 5 (5) No prophylactic pneumococcal antibiotic use documented 83 (88) Immunoglobulin use (within the past 30 days) 2 (2) Clinical course n (%) Site of first positive test  Blood 81 (86)  Bronchiolar fluid 4 (4)  CSF 4 (4)  Joint fluid 1 (1)  Middle ear 0 (0)  Pleural fluid 0 (0)  Peritoneal fluid 1 (1)  Other 2 (2) Duration of IPD admission (days)  Mean 11  Median (IQR) 5 (3.3, 10.8) ICU admission 15 (16) Duration of ICU admission (days)  Mean 7.5  Median (IQR) 2 (1,9.5) Assisted ventilation 10 (11) Duration of assisted ventilation (days)  Mean 9.8  Median (IQR) 5 (2, 11) ECMO use 0 (0) Vasopressor use 1 (1) Hemodynamic instability 5 (5) Acute respiratory distress syndrome 3 (3) Surgical procedure 7 (7) End organ dysfunction 10 (11)  Neurologic sequelae 5 (5)  Renal failure 4 (4)  Myocarditis 1 (1) Permanent disabilities 1 (1) Death 0 (0) Full recovery 93 (99)

CSF indicates cerebral spinal fluid; ECMO, Extracorporeal membrane oxygenation; ICU, Intensive Care Unit; IPD, invasive pneumococcal disease; PCV, pneumococcal conjugate vaccine; PPSV, pneumococcal polysaccharide vaccine.

*Patients with multiple risk factors were included in each respective subcategory, thus the total number exceeds the total study population.

†Two of these children were <15 months of age, and thus vaccinated appropriate to age.

‡Serotypes in unvaccinated individuals: 15B (n = 4), 7F (n = 2), 19A (n = 2), 22F (n = 2), 23B (n = 2), 35B (n = 2) and 21 (n = 1).

§Recorded based on highest level of vaccine coverage (ie, if received PCV7 and then PCV13, they were classified as PCV13 recipients).


IPD Prevention and Vaccination

Of the 66 (70%) patients with known pneumococcal vaccination status at the time of IPD, 34 (36%) received age-appropriate pneumococcal conjugate vaccinations as per NACI guidelines for high-risk children (Table 1).12 Five patients were partially immunized (1–2 doses) based on their age and general pediatric recommendations and 15 (23%) were unvaccinated. Of the unvaccinated children, many had an active malignancy at the time of IPD (n = 7, 47%), or were taking an immunosuppressant medication (n = 5, 33%). However, all unvaccinated children with cancer were diagnosed with a malignancy after 1 year of age. The median age of IPD in the unvaccinated children was 9.0 years (IQR: 4.7–9.7), older than the vaccinated cohort where the median age was 5.8 years (IQR: 4.0–9.0). A higher proportion of unvaccinated children were born between 1995 and 2004 (83%). PCV7 was the most common pneumococcal vaccine received (n = 22, 43%), followed by PCV13 (n = 15, 29%). Only 3 of the 84 age-eligible children (4%) received both a conjugate vaccine and PPSV23 and all were born after 2007 (Table, Supplemental Digital Content 2, https://links.lww.com/INF/E868). No children received only PPSV23 vaccination.

In unvaccinated/partially vaccinated children and fully vaccinated children, 14/20 (70%) and 2/34 (6%) cases were due to PCV13-contained serotypes, respectively (Table, Supplemental Digital Content 3, https://links.lww.com/INF/E869). Therefore, an unadjusted estimate of vaccine effectiveness for PCV13 against PCV13 serotypes is 97.3%.17

Prophylactic antibiotics were used by 12 patients (13%). Patients with SCD (n = 8) and transplant patients with asplenia (n = 3) composed 92% of patients using prophylactic antibiotics (Table 1).

Clinical Course and Patient Outcomes

The most common presentation was bacteremia (n = 81, 86%), followed by pneumonia (n = 23, 25%) (Table 2). Of the 23 patients with pneumonia, 19 (83%) were also bacteremic. In total 9 (10%) patients presented with meningitis, 8 of whom had SCD or asplenia. Of the 14 SCD patients and 9 asplenic patients with IPD, 43% (n = 6) and 22% (n = 2), respectively presented with pneumococcal meningitis. Three of the meningitis cases were receiving prophylactic antibiotics (penicillin n = 1 and amoxicillin n = 2).

TABLE 2. - Type of invasive pneumococcal disease Episode by High-Risk Group Bacteremia n (%) Pneumonia n (%)* Meningitis n (%) Joint n (%) Endocarditis n (%) Solid organ transplant (n = 17) 15 (88) 6 (35) 1 (7) 0 (0) 0 (0)  Heart (n = 4) 3 (75) 3 (75) 0 (0) 0 (0) 0 (0)  Kidney (n = 2) 1 (50) 0 (0) 0 (0) 0 (0) 0 (0)  Liver (n = 12) 11 (92) 3 (25) 1 (8) 0 (0) 0 (0)  Multi-visceral (n = 1) 1 (100) 0 (0) 0 (0) 0 (0) 0 (0) Malignancy (n = 33) 30 (91) 4 (12) 1 (3) 1 (3) 1 (3)  Hematologic (n = 23) 21 (91) 4 (17) 1 (4) 1 (4) 1 (4)  Solid tumour (n = 10) 9 (90) 0 (0) 0 (0) 0 (0) 0 (0) Hematopoietic stem cell transplant (n = 8) 7 (88) 0 (0) 1 (13) 1 (13) 1 (13) Use of other immune suppressing medications (n = 32) 29 (91) 9 (28) 1 (3) 0 (0) 0 (0) Primary immunodeficiency (n = 4) 4 (100) 2 (50) 0 (0) 0 (0) 0 (0) Asplenia (n = 9) 8 (89) 0 (0) 2 (22) 1 (11) 1 (11) Sickle cell disease (n = 14) 9 (64) 2 (14) 6 (43) 0 (0) 0 (0) Cochlear implant (n = 1) 1 (100) 1 (100) 0 (0) 0 (0) 0 (0) Nephrotic syndrome (n = 8) 8 (100) 0 (0) 0 (0) 0 (0) 0 (0) Chronic lung disease (n = 7) 5 (71) 6 (86) 0 (0) 0 (0) 0 (0) CSF leak (n = 1) 1 (100) 0 (0) 0 (0) 0 (0) 0 (0) HIV (n = 2) 2 (100) 1 (50) 0 (0) 0 (0) 0 (0) Total† 81 (86) 23 (25) 9 (10) 2 (2) 1 (1)

CSF indicates cerebral spinal fluid.

*Pneumonia was defined by detection of S. pneumoniae in bronchoalveolar lavage or blood culture with evidence of infiltrates on chest radiograph. Most cases (n = 22) had a secondary bacteremia, 1 case had non-bacteremic pneumonia.

†Patients with multiple risk factors were included in each respective subcategory, thus the total number exceeds the total study population.

The duration of hospitalization was less than 7 days in 59% of cases, with a median of 5 days (IQR: 3.3–10.8) (Table 1). The hospitalization course was complicated by end-organ dysfunction (n = 10, 11%), Intensive Care Unit admission (n = 15, 16%) or the need for surgery (n = 7, 7%). Of the 10 patients with end-organ dysfunction, half (n = 5) were children with SCD who had a vaso-occlusive crisis. Of the 15 patients requiring Intensive Care Unit-level care, a large proportion had underlying chronic lung disease (n = 6, 40%) or SCD (n = 4, 27%). Surgical procedures included cardiac valve replacement and chest tube placement. One patient with meningitis experienced permanent neurologic sequelae. No patients died. A full recovery was made by 99% of patients (n = 93).

Isolate Serotypes

IPD isolates serotype data was retrieved for 66 patients. The 6 most common serotypes were 15B, 22F, 19A, 23A, 23B and 9N (Fig. 1). From 2016 to 2018, the most recent 3 years of the study, there was no IPD caused by PCV13-contained serotypes, despite PCV13-contained serotypes composing 27% (n = 7) in the first 3 years (2009–2011). IPD caused by PPSV23 persisted over the study period and composed 50% (n = 7) of the 14 cases of IPD from 2016 to2018 (Fig. 2). Of the 6 children with IPD caused by PPSV23-contained serotypes between 2016 and 2018, none received PPSV23 vaccination, and 1 was completely unvaccinated against S. pneumoniae. However, before 2016, 1 child who received the PPSV23 vaccine had IPD due to a serotype contained within PPSV23. The proportion of hospitalizations caused by PCV13-contained serotypes was significantly lower after PCV13 introduction when comparing pre-PCV13 (2009/2010; n = 8 of 22 total IPD cases) to post-PCV13 (2011–2018; n = 6 of 43 total IPD cases) time periods (P = 0.04).

F1FIGURE 1.:

Yearly IPD serotype data from 2009 to 2018, grouped by vaccine coverage. IPD, invasive pneumococcal disease.

F2FIGURE 2.:

Proportion of IPD serotype vaccine-coverage groups in pre-PCV13, early PCV13 and late PCV13 time periods, stratified by vaccination status. (Vaccine Coverage Groups: PCV7 represents IPD serotypes contained within PCV7, PCV13 represents serotypes uniquely covered by PCV13 and not PCV7, PPSV23 represents serotypes unique to PPSV23 and not contained within PCV13, and Non-vaccine type represents serotypes not included in current pneumococcal vaccines). IPD indicates invasive pneumococcal disease; PCV, pneumococcal conjugate vaccine; PPSV, pneumococcal polysaccharide vaccine.

Of the 20 children with no vaccine or partial immunization (≤2 doses), 14 IPD episodes were due to serotypes included in vaccines (PPSV23-contained n = 10, PCV13-contained n = 3 and PCV-7-contained n = 1). IPD caused by a vaccine-contained serotype occurred in 2 immunized patients. Both had bacteremia without a focus of infection, 1 due to serotype 3 in a fully vaccinated transplant recipient, and 1 due to serotype 22F in a child using immune-suppressive medications who received 2 doses of PCV13.

Antimicrobial Susceptibility

Complete antimicrobial susceptibility results were available for 64 isolates (68%) (Table 3). Erythromycin had the highest proportion of resistant (36% of isolates resistant), followed by trimethoprim-sulfamethoxazole (11%), cefuroxime (11%), meropenem (8%) and penicillin (5%) based on nonmeningitis resistance breakpoints. Of the 6 patients on penicillin prophylaxis, all IPD isolates were sensitive to penicillin. All isolates were susceptible to vancomycin and levofloxacin. Serotype 19A had the highest proportion of resistant with 100% of isolates resistant to erythromycin, 67% resistant to cefuroxime and meropenem, 50% resistant to penicillin, 33% resistant to trimethoprim-sulfamethoxazole and 17% resistant to ceftriaxone. High proportions of resistance were also seen in serotypes 15B and 19F (Table 3). Over the study period, meropenem and penicillin proportions of resistance decreased from 12% to 8%, respectively in 2009–2011 and to 0% in 2016–2018.

TABLE 3. - Antibiotic Sensitivities of Invasive Pneumococcal Disease Isolates by Serotype and Time Period, Using Non-Meningitis Resistance Breakpoints Serotype n = % Resistant TMP-SMX Penicillin * Erythromycin Cefuroxime Ceftriaxone † Meropenem Vancomycin Levofloxacin ≥1 Antibiotic ≥2 Antibiotics 15B 10 10 0 20 10 10 0 0 0 20 10 22F 9 0 0 11 0 0 0 0 0 11 0 19A 6 33 33 100 67 0 67 0 0 100 67 23A 5 20 0 40 0 0 0 0 0 60 0 23B 4 0 0 25 0 0 0 0 0 25 0 9N 3 0 0 0 0 0 0 0 0 0 0 7F 2 0 0 50 0 0 0 0 0 50 0 35F 2 0 0 0 0 0 0 0 0 0 0 19F 2 50 0 50 50 50 50 0 0 50 50 16F 2 0 0 0 0 0 0 0 0 0 0 15C 2 50 0 50 0 0 0 0 0 100 0 21 2 0 0 50 0 0 0 0 0 50 0 Others 15 7 0 47 7 0 0 0 0 47 13 Total 64 11 5 36 11 2 8 0 0 39 12 Time Period n = N Cases per Year TMP-SMX Penicillin* Erythromycin Cefuroxime Ceftriaxone

留言 (0)

沒有登入
gif