Epidemiology and Infection Pneumococcal serotypes in children, clinical cambridge.org/hyg presentation and antimicrobial susceptibility in the PCV13 era C. Izquierdo1 , P. Ciruela1,2, S. Hernández1, J. J. García-García2,3,4, C. Esteva2,3,4, Original Paper F. Moraga-Llop5, A. Díaz-Conradi6, J. Martínez-Osorio3, A. Solé-Ribalta3, M. F. de Cite this article: Izquierdo C et al (2020). Pneumococcal serotypes in children, clinical Sevilla2,3,4, S. González-Peris5, G. Codina5, A. M. Planes5, S. Uriona5, M. Campins5, presentation and antimicrobial susceptibility in the PCV13 era. Epidemiology and Infection C. Muñoz-Almagro2,3,4,7, L. Salleras2,8 and A. Domínguez2,8 148, e279, 1–12. https://doi.org/10.1017/ 1 S0950268820002708 Agència de Salut Pública de Catalunya, Generalitat de Catalunya, Barcelona, Spain; 2CIBER de Epidemiología y Salud Pública (CIBERESP), Madrid, Spain; 3Hospital Sant Joan de Déu, Universitat de Barcelona, Esplugues de Received: 29 May 2020 Llobregat, Barcelona, Spain; 4Malalties prevenibles amb vacunes, Institut de recerca Sant Joan de Déu, Esplugues Revised: 4 October 2020 de Llobregat, Barcelona, Spain; 5Hospital Vall d’ Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain; Accepted: 15 October 2020 6 Hospital HM Nens, HM Hospitales, Barcelona, Spain; 7Departament de Medicina. Universitat Internacional de Key words: Catalunya, Barcelona, Spain and 8Departament de Medicina, Universitat de Barcelona, Barcelona, Spain Antimicrobial susceptibility; invasive pneumococcal disease; PCV13 vaccine; Abstract penicillin resistance; serotype distribution; Streptococcus pneumoniae The aim was to analyse invasive pneumococcal disease (IPD) serotypes in children aged ⩽17 years according to clinical presentation and antimicrobial susceptibility. We conducted a pro- Author for correspondence: spective study (January 2012–June 2016). IPD cases were diagnosed by culture and/or real- C. Izquierdo, E-mail: conchita.izquierdo@gencat.cat time polymerase chain reaction (PCR). Demographic, microbiological and clinical data were analysed. Associations were assessed using the odds ratio (OR) and 95% confidence intervals (CI). Of the 253 cases, 34.4% were aged <2 years, 38.7% 2–4 years and 26.9% 5– 17 years. Over 64% were 13-valent pneumococcal conjugate vaccine (PCV13) serotypes. 48% of the cases were diagnosed only by real-time PCR. Serotypes 3 and 1 were associated with complicated pneumonia (P < 0.05) and non-PCV13 serotypes with meningitis (OR 7.32, 95% CI 2.33–22.99) and occult bacteraemia (OR 3.6, 95% CI 1.56–8.76). Serotype 19A was more frequent in children aged <2 years and serotypes 3 and 1 in children aged 2–4 years and 5–17 years, respectively. 36.1% of cases were not susceptible to penicillin and 16.4% were also non-susceptible to cefotaxime. Serotypes 14, 24F and 23B were associated with non-susceptibility to penicillin (P < 0.05) and serotypes 11, 14 and 19A to cefotaxime (P < 0.05). Serotype 19A showed resistance to penicillin (P = 0.002). In conclusion, PCV13 ser- otypes were most frequent in children aged ⩽17 years, mainly serotypes 3, 1 and 19A. Non- PCV13 serotypes were associated with meningitis and occult bacteraemia and PCV13 sero- types with pneumonia. Non-susceptibility to antibiotics of non-PCV13 serotypes should be monitored. Introduction Streptococcus pneumoniae infections are associated with substantial morbidity and mortality worldwide. The World Health Organization estimated that approximately 335 000 (240 000– 460 000) deaths were due to S. pneumoniae in children aged <5 years in 2015 [1]. The epidemiology of the invasive pneumococcal disease (IPD) has changed after the intro- duction of pneumococcal conjugate vaccines [2]. In countries where the 7-valent pneumococ- cal conjugate vaccine (PCV7) containing serotypes 4, 6B, 9V, 14, 18C, 19F and 23F was introduced, the global IPD incidence decreased but non-PCV7 serotypes emerged [2–5]. Later, the 10-valent pneumococcal conjugate vaccine (PCV10), which included the additional serotypes 1, 5 and 7F, and 13-valent pneumococcal conjugate vaccine (PCV13) which includes © The Author(s), 2020. Published by the additional serotypes 3, 6A and 19A [6] were introduced, and a significant decline in some Cambridge University Press. This is an Open additional serotypes was observed [7]. Access article, distributed under the terms of In Catalonia, Spain, PCV7 was available in 2001, but only children with risk factors were vac- the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), cinated for free [8, 9]. Estimated vaccine coverage of around 50% was reached [9]. Between 2005 which permits unrestricted re-use, and 2007, a change in IPD serotypes was found in children aged <2 years compared with the pre- distribution, and reproduction in any medium, vaccine period, with a decrease in PCV7 serotypes and an increase in non-PCV7 serotypes [10]. provided the original work is properly cited. PCV13 replaced PCV7 in 2010 and, in July 2016, was included in the official paediatric vaccin- ation calendar [11]. PCV13 covered up to 70% of IPD cases in our reference area. In 2013, the estimated coverage of PCV13 in children aged <2 years was 64% (92.5% fully vaccinated) [12]. Studies worldwide have found that the changing distribution of pneumococcal serotypes has led to changes in the rates of S. pneumoniae resistance to antibiotics [13, 14]. 2 C. Izquierdo et al. The aim of this study was to analyse the distribution of S. 33C), 11A/11D, 12F/(12A/44/46), 13, 16F, 17F, 18/(18A/18B/ pneumoniae serotypes according to the clinical presentation of 18C/18F), 19A, 19F, 20(20A/20B), 21, 22F/22A), 23A, 23B, 24/ IPD and antimicrobial susceptibility in children aged ⩽17 years (24A/24B/24F), 31, 34, 35A/(35C/42), 35B, 35F/47F, 38/25F and in a community with intermediate PCV13 coverage before its 39] was used [15]. introduction in the official paediatric vaccination calendar. Serotypes identified by real-time PCR as group level (6A/6C, 7F/7A, 9V/9A and 19F/19B/19C) were defined as PCV13 sero- types 6A, 7F, 9V and 19F, respectively. Methods PCR-positive samples that were negative for the serotypes Study design included in the sequential multiplex PCR (including all vaccine serotypes) were classified as other non-vaccine serotypes (ONV). We conducted a prospective observational study between January Susceptibility to penicillin and cefotaxime was determined by 2012 and June 2016 in children aged ⩽17 years with IPD attended the microdilution method according to the Clinical and in three paediatric hospitals: Hospital Vall d’Hebron of Barcelona, Laboratory Standards Institute (CLSI) criteria. Current EUCAST Hospital Sant Joan de Déu of Esplugues, Barcelona and Hospital breakpoints were used to interpret susceptibility [21]. A min- de Nens of Barcelona. The estimated reference population aged imum inhibitory concentration (MIC) study was carried out at ⩽17 years of the hospitals was 442 761, representing 31.9% of the National Microbiology Center. The penicillin breakpoints this age group in Catalonia. used were: MIC >0.06 mg/l, non-susceptible and MIC >2 mg/l, IPD was diagnosed through the isolation of S. pneumoniae by resistant. For cefotaxime, the breakpoint was: MIC >0.5 mg/l, culture or detection of bacterial DNA by real-time polymerase non-susceptible. chain reaction (PCR) in any normally sterile site [15], such as blood, cerebrospinal fluid, pleural fluid or articular fluid, among others. We established the presence of S. pneumoniae DNA by amp- Statistical analysis lification of the autolysin (lytA) gene and the wzg (cpsA) gene, Proportions were compared using the χ 2 test or Fisher’s exact test, according to published assays [16–18]. We included only samples as appropriate. All statistical tests were two-tailed and statistical that were positive for the lytA and wzg genes in real-time PCR. significance was established as P < 0.05. If the number of cases All clinical samples were analysed by culture and real-time was zero, this was corrected by adding 0.5, and P values were cal- PCR. culated according to Sheskin’s method [22] The clinical evaluation was made by the paediatricians partici- Associations between variables were assessed using the odds pating in the study according to established diagnostic criteria [19]. ratio (OR) and 95% confidence intervals (CI). The analyses were conducted using the Statistical Package for Data collected Social Sciences (SPSS 19.0 for Windows) and EPIDAT (program for the epidemiological analysis of tabulated data; version 3.1). Epidemiological and clinical characteristics, including age, gender, underlying medical condition, clinical presentation, intensive care unit (ICU) admission, clinical outcome and PCV13 vaccination Data confidentiality and ethical aspects status were collected. The method of diagnosis, the serotype Informed consent was obtained from all individual participants and antimicrobial susceptibility were analysed. included in the study. The study complies with the principles of A subject was considered vaccinated if they had received the the Declaration of Helsinki and the legal structure with respect last dose of PCV13 ⩾15 days before symptom onset. to international human rights and biomedicine and protection of personal data laws. The Ethics Committee of Hospital Sant Serotype identification and antimicrobial susceptibility Joan de Déu approved the study. Strains isolated by culture were serotyped using the Quellung reaction or dot blot at the National Centre of Microbiology Results (Majadahonda, Spain) [20]. When the diagnosis was made only Baseline characteristics by real-time PCR, serotypes were identified at the Molecular Microbiology Department, Hospital Sant Joan de Déu, in accord- During the study period, 263 cases of IPD were diagnosed in chil- ance with previously-validated methods [15]. dren aged ⩽17 years. Serotyping was not possible in ten cases Capsular typing of all culture-negative and real-time (3.8%), which were excluded. Of the 253 cases included, 151 PCR-positive samples was made using two methods, depending (59.7%) were male patients; 87 (34.4%) were children aged <2 on the amount of S. pneumoniae DNA available. If the amount years, 98 (38. 7%) in those aged 2–4 years, and 68 (26.9%) in was low (detection of lytA gene DNA and an additional capsular those aged 5–17 years (Table 1). gene of S. pneumoniae by real-time PCR with the cycle threshold There was a seasonal variation, with 72% of cases detected dur- (CT) of >30 cycles), a previously described, real-time multiplex ing the cool months (October to March; P < 0.001). PCR technique that detects all pneumococcal capsular types and With respect to PCV13 vaccination, 170 children (67.2%) were differentiates serotypes 1, 3, 4, 5, 6A/C, 6B/D, 7F/A, 8, 9V/A/N/ not immunised, 43 (17%) were fully immunised and 39 (15.4%) L, 14, 15B/15C, 18C/18B, 19A, 19F/19B/19C, 23A and 23F was partly immunised; 129 children (51.0%) had received no dose used [16]. If the amount of S. pneumoniae DNA was high of any conjugate vaccine. In one case, the vaccination status (PCR-positive samples with CT of ⩽30 cycles), sequential multiplex could not be determined. PCR combined with fragment analysis and automated fluorescent Fifteen cases (6%) had a chronic disease, including congenital capillary electrophoresis to differentiate serotypes [1, 2, 3, 4, 5, immunodeficiency (2 cases), congenital heart disease (3 cases), 6A/6B, 6C, 6,7C/(7B/40), 7F/7A, 9N/9L, 9V/9A, 10A, 10F/(10C/ chronic pulmonary disease (1 case), chronic kidney disease (1 Epidemiology and Infection 3 Table 1. Demographic characteristics and clinical presentation in children aged case), cerebrospinal fluid leak (1 case), or had received immuno- ⩽17 years with IPD suppressive therapy in the previous 6 months (7 cases). N (%) The most frequent clinical presentation was pneumonia (187 cases; 73.9%), of which 150 cases (80. 2%) were complicated (253 cases) pneumonia. This was followed by occult bacteraemia (25 cases; Age group 9.9%) and meningitis (18 cases; 7.1%). The presentation in the remaining 23 cases (9.1%) was septic shock (7 cases; 2.8%), mas- <2 years 87 (34.4) toiditis (7 cases; 2.8%), osteoarticular infection (6 cases; 2.3%), 2–4 years 98 (38.7) orbital cellulitis (2 cases; 0.8%) and pancreatitis (1 case; 0.4%). 5–17 years 68 (26.9) Figure 1 shows the distribution of clinical manifestations by the age group. Complicated pneumonia was the most frequent clinical Sex presentation in all age groups. Male 151 (59.7) Forty–nine cases (19.4%) were admitted to the ICU: 7 out of 7 Seasonality (100%) with septic shock, 15 out of 18 (83.3%) with meningitis, 23 out of 150 (15.3%) with complicated pneumonia, 1 out of 7 October–March 182 (72.0) (14.3%) with mastoiditis, 2 out of 37 (5.4%) with non-complicated Diagnosis pneumonia and 1 out of 25 (4%) with occult bacteraemia (ICU admis- sion was caused by convulsions). Of the 49 ICU cases, 55.1% were Culture only 62 (24.5) caused by PCV13 serotypes, and 14.3% had an underlying disease. PCR only 122 (48.2) Two patients died: an 18-month-old child with serotype-35F Culture + PCR 69 (27.3) meningitis who had congenital immunodeficiency, and a 23-month-old child with septic shock caused by a non-vaccine Clinical presentation serotype which was not identified. Overall pneumonia 187 (73.9) Complicated pneumonia 150 (80.2) Serotypes and clinical presentation Occult bacteraemia 25 (9.9) One hundred and sixty-three (64.4%) cases were due to PCV13 Meningitis 18 (7.1) serotypes (Fig. 2); 41.5% patients had received ⩾1 dose of Other formsa 23 (9.1) PCV13 (34/82 cases) and 75.2% patients had not received any conjugated vaccine (97/129; P < 0.001). ICU admission The most frequent serotypes were: 3 (20.9%), 1 (19.0%), 19A Yes 49 (19.4) (6.7%) and 14 (5.9%), all PCV13 serotypes. Underlying medical condition Overall pneumonia and complicated pneumonia, respectively, were mainly caused by PCV13 serotypes (143/187 cases, 76.4% Yes 15 (6.0) and 123/150 cases, 82.0%, respectively): 87.7% of IPD cases caused Case fatality rate 2 (0.8) by PCV13 serotypes had pneumonia, compared with only 48.9% of Yes IPD cases caused by non-PCV13 serotypes (OR 7.47, 95% CI 4.0– 13.96). Similarly, 75.5% of cases caused by PCV13 serotypes led to Serotype groups complicated pneumonia compared with only 30% of IPD cases PCV13 163 (64.4) caused by non-PCV13 serotypes (OR 7.2, 95% CI 4.04–12.75) Non-PCV13 90 (35.6) (Table 2). PCV13 serotypes were more likely to cause overall pneu- monia and complicated pneumonia than non-PCV13 serotypes. Pneumococcal vaccinationb A total of 92.5% of IPD cases caused by serotype 3 were overall PCV13 pneumonia and 83% of IPD cases caused by serotype 3 were com- ⩾1 dose 82 (32.4) plicated pneumonia. Serotype 3 caused overall pneumonia and complicated pneumonia more frequently than the other clinical Fully vaccinated 43 (17.0) forms (OR 5.5, 95% CI 1.9–15.92 and OR 4.3, 95% CI 2.01– Partly vaccinated 39 (15.3) 9.35, respectively). No dose 170 (67.2) Serotype 1 was also associated with overall pneumonia (OR 21.82, 95% CI 2.95–161.63) and complicated pneumonia (OR Non-vaccinated (any PCV) 129 (51.2) 5.16, 95% CI 2.21–12.04). Serotype 1 was more likely to cause Unknown 1 (3.9) overall pneumonia and complicated pneumonia than other clin- Antimicrobial susceptibility (n = 122 strains) ical forms (OR 21.82, 95% CI 2.95–161.63 and OR 5.16, 95% CI 2.21–12.04, respectively). Penicillin non-susceptiblec 44 (36.1) Meningitis was mainly caused by non-PCV13 serotypes (14/ Cefotaxime non-susceptible 20 (16.4) 18, 77.8%). Although there was no predominant serotype, IPD, invasive pneumococcal disease; CFR, case fatality rate; ICU, intensive care unit; PCV13, 15.6% of IPD cases caused by non-PCV13 serotypes presented 13-valent pneumococcal conjugate vaccine. as meningitis, compared with only 2.5% of cases caused by a Other forms: septic shock (7 cases), mastoiditis (7 cases), osteoarticular infection (6 cases), PCV13 serotypes (OR 7.32, 95% CI 2.33–22.99). orbital cellulitis (2 cases), pancreatitis (1 case). b 1 missing case. Occult bacteraemia was also caused mainly by non-PCV13 sero- c 4 cases were penicillin-resistant. types (16/25, 64%), and 17.8% of IPD cases caused by non-PCV13 serotypes presented as occult bacteraemia compared with only 4 C. Izquierdo et al. Fig. 1. Distribution of IPD clinical presentation in children aged ⩽17 years by age group. IPD: invasive pneumococcal disease. Fig. 2. Distribution of Streptococcus pneumoniae serotypes causing IPD in children aged ⩽17 years by age groups. IPD: invasive pneumococcal disease. ONV: other non-vaccine serotypes. #1: P < 0.001; #2: P = 0.001; #3: P = 0.028. 5.5% of cases caused by PCV13 serotypes (OR 3.6, 95% CI 1.56– 2.26–1043.02), but because there are only two cases with this sero- 8.76). Some non-PCV13 serotypes were associated with occult bac- type the finding cannot be evaluated. teraemia, namely 12F (OR 19.73, 95% CI 1.72–226.12) and 10A Other clinical presentations were caused mainly by (OR 10.22, 95% CI 1.94–53.74). Serotype 18C, a PCV13 serotype non-PCV13 serotypes (16/23; 69.6%), with 17.8% of IPD cases was also associated with occult bacteraemia (OR 48.61, 95% CI caused by non-PCV13 serotypes presenting as other clinical Epidemiology and Infection Table 2. Distribution of Streptococcus pneumoniae serotypes causing IPD in children aged ⩽17 years by clinical presentation Total Meningitis Overall pneumonia Complicated pneumonia Occult bacteraemia Others Serotype n % n % P value* n % P value* n % P value* n % P value* n % P value* PCV13 1 48 19.0 0 0.0 – 47 97.9 <0.001 41 85.4 <0.001 0 0.0 – 1 2.1 ns 3 53 20.9 1 1.9 ns 49 92.5 0.001 44 83.0 <0.001 0 0.0 – 3 5.7 ns 4 2 0.8 0 0.0 – 1 50.0 ns 0 0.0 – 1 50.0 ns 0 0.0 – 6A a 3 1.2 0 0.0 – 3 100.0 ns 3 100.0 ns 0 0.0 – 0 0.0 – 6B 2 0.8 0 0.0 – 2 100.0 ns 1 50.0 ns 0 0.0 – 0 0.0 – 7F a 9 3.6 0 0.0 – 9 100.0 ns 9 100.0 ns 0 0.0 – 0 0.0 – 9V a 6 2.4 0 0.0 – 4 66.7 ns 4 66.7 ns 2 33.3 ns 0 0.0 – 14 15 5.9 1 6.7 ns 12 80.0 ns 9 60.0 ns 1 6.7 ns 1 6.7 ns 18C 2 0.8 0 0.0 – 0 0.0 – 0 0.0 – 2 100.0 0.009 0 0.0 – 19A 17 6.7 1 5.9 ns 15 88.2 ns 12 70.6 ns 0 0.0 – 1 5.9 ns 19F a 5 2.0 1 20.0 ns 1 20.0 ns 0 0.0 – 2 40.0 ns 1 20.0 ns 23F 1 0.4 0 0.0 – 0 0.0 – 0 0.0 – 1 100.0 ns 0 0.0 – Non-PCV13 6C 1 0.4 0 0.0 – 1 100.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 8 2 0.8 0 0.0 – 1 50.0 1 50.0 ns 0 0.0 – 1 50.0 ns 10A 6 2.4 1 16.7 ns 1 16.7 ns 0 0.0 – 3 50.0 0.014 1 16.7 ns 11A 5 2.0 0 0.0 – 3 60.0 ns 1 20.0 ns 0 0.0 – 2 40.0 ns 12F 3 1.2 0 0.0 – 1 33.3 ns 1 33.3 ns 2 66.7 0.026 0 0.0 – 13 1 0.4 1 100.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 15A 3 1.2 1 33.3 ns 0 0.0 ns 0 0.0 – 1 33.3 ns 1 33.3 ns 15B 2 0.8 1 50.0 ns 0 0.0 – 0 0.0 – 1 50.0 ns 0 0.0 – 15C 1 0.4 0 0.0 – 0 0.0 – 0 0.0 – 1 100.0 ns 0 0.0 – 16F 2 0.8 1 50.0 ns 0 0.0 – 0 0.0 – 1 50.0 ns 0 0.0 – 22F 2 0.8 1 50.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 1 50.0 ns 23A 1 0.4 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 1 100.0 ns 23B 3 1.2 0 0.0 – 1 33.3 ns 1 33.3 ns 1 33.3 ns 1 33.3 ns 24 2 0.8 0 0.0 – 1 50.0 ns 0 0.0 – 1 50.0 ns 0 0.0 – 24A 2 0.8 1 50.0 ns 1 50.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 24F 6 2.4 1 16.7 ns 4 66.7 ns 2 33.3 ns 1 16.7 ns 0 0.0 (Continued ) 5 6 C. Izquierdo et al. presentations compared with only 4.3% of IPD cases caused by 7F: 4 out of 9 cases were identified as group level (7F/7A); 9V: 2 out of 6 cases were identified as group level (9V/9A); 19F: 1 out of 5 case was identified as group level (19F/19B/19C); 6A: 2 out of 3 cases were identified as group level (6A/6C). P value* <0.001 PCV13 serotypes (OR 4.81, 95% CI 1.90–12.22). ns ns ns ns – – – – – Serotypes and clinical presentation according to age group Others 50.0 50.0 50.0 50.0 10.8 17.8 0.0 0.0 0.0 4.3 9.1 PCV13 serotypes were those most frequently detected in 5– % 17-year-olds, representing 77.9% of all serotypes in this age group (OR 2.41, 95% CI 1.27–4.59) and non-PCV13 serotypes were the most frequently detected serotypes (55.2%) in the <2 0 1 1 1 0 0 1 4 7 16 23 n years age group (OR 3.63, 95% CI 2.10–6.29). In the 5–17 years age group, IPD was more likely to be caused by PCV13 serotypes. P value* 0.002 Serotype 19A was more frequent in the <2 years age group, sero- Occult bacteraemia type 3 in the 2–4 years age group (58.5%; OR 2.79, 95% CI 1.50– – – – – – ns ns ns ns 5.20) and serotype 1 in the 5–17 years age group (60.4%; OR 6.50, 95% CI 3.31–12.77) (Fig. 2). Serotype 19A more frequently 0.0 0.0 0.0 0.0 5.5 9.9 50.0 50.0 17.8 100.0 100.0 caused IPD in the <2 years age group (58.8%; OR 2.95, 95% CI % 1.08–8.05), as did serotype 3 in the 2–4 years age group (58.5%; OR 2.79, 95% CI 1.50–5.20) and serotype 1 in the 5–17-years age group (60.4%; OR 6.50, 95% CI 3.31–12.77) (Fig. 2). 16 25 1 1 1 0 1 0 0 0 9 n Table 3 and Figure 3 show the serotype distribution by age groups and the association to clinical presentation. P value* <0.001 In children aged <2 years, PCV13 serotypes were associated with Complicated pneumonia – – – – – – – – ns *The P value indicates differences in the relationship between a serotype and a clinical form with respect to the other serotypes. ‘ns’ indicates P value⩾0.05. overall pneumonia (OR 5.29, 95% CI 2.08–13.41) and complicated pneumonia (OR 4.84, 95% CI 1.91–12.22). Serotype 3 caused overall pneumonia in 91.7% of cases (OR 12.57, 95% CI 1.54–102.33) and 50.0 54.1 75.5 30.0 59.3 0.0 0.0 0.0 0.0 0.0 0.0 complicated pneumonia in 83.3% (OR 10.62, 95% CI 2.16–52.3). % Serotype 19A was associated with complicated pneumonia (OR 4.32, 95% CI 1.03–18.07) in 70% of cases. 0 0 0 0 0 0 1 20 27 123 150 In the 2–4 years age group, PCV13 serotypes were associated n with overall pneumonia (OR 5.42, 95% CI 1.70–17.23) and com- plicated pneumonia (OR 6.92, 95% CI 2.61–18.36), while P value* non-PCV13 serotypes were associated with meningitis (OR <0.001 27.38, 95% CI 1.42–527.7). Serotype 3 was associated with com- ns ns – – – – – – – Overall pneumonia plicated pneumonia (OR 6.7, 95% CI 1.85–24.24) and serotype 9V with occult bacteraemia (OR 102.7, 95% CI 4.27–2472.3). 0.0 0.0 0.0 0.0 0.0 50.0 50.0 75.7 87.7 48.9 73.9 In the 5–17 years age group, PCV13 serotypes were associated % with overall pneumonia (OR 8.17, 95% CI 1.91–34.86) and com- plicated pneumonia (OR 6.05, 95% CI 1.81–20.18), whereas non-PCV13 serotypes were associated with other clinical forms 0 0 0 1 0 0 1 28 44 143 187 n (OR 9.3, 95% CI1.51–57.12). Serotype 1 was associated with com- plicated pneumonia (OR 10.43, 95% CI 2.17–50.12). P value* <0.001 Serotypes and method of diagnosis ns – – – – – – – – Meningitis IPD, invasive pneumococcal disease; ONV, other non-vaccine serotypes. The diagnosis was made by culture in 62 cases (24.5%), culture and real-time PCR in 69 (27.3%) and real-time PCR in 122 (48.2%). 0.0 0.0 0.0 0.0 0.0 0.0 2.5 7.1 13.5 15.6 100.0 % PCV13 and non-PCV13 serotypes showed no significant dif- ference in the percentage of cases diagnosed only by PCR (52.1% vs. 41.1%). 14 18 0 0 0 0 0 1 0 5 4 n There were significant differences between serotypes in the method of diagnosis: 86.8% of cases due to serotype 3 were diag- 100.0 0.4 0.8 0.8 0.8 0.4 0.4 0.8 14.6 64.4 35.6 nosed only by real-time PCR vs. 38% of cases due to other sero- % types (P < 0.001). By contrast, other serotypes, such as serotype 14 Total and 24F, were diagnosed less frequently by real-time PCR (13.3%, 2/15 and 0%, 0/6) than the other serotypes (P = 0.005 and P = 1 2 2 2 1 1 2 37 90 163 253 n 0.030, respectively). Table 2. (Continued.) Non-PCV13 Non-susceptibility to antibiotics by serotypes PCV13 Serotype Total Antimicrobial susceptibility was studied in 122 (93.1%) strains ONV 35B 25F 33F 35F 27 31 38 isolated. Forty-four cases (36.1%) were non-susceptible to penicil- lin (MIC > 0.06), and four cases (3.3%) were penicillin-resistant. a Epidemiology and Infection Table 3. Distribution of Streptococcus pneumoniae serotype groups causing IPD in children aged ⩽17 years by clinical presentation and age group Clinical presentation Total Meningitis Overall pneumonia Complicated pneumonia Occult bacteraemia Other forms Age Serotype P P P P P groups group n % n % value* n % value* n % value* n % value* n % value* 0–17 years PCV13 163 64.4 4 2.5 – 143 87.7 <0.001 123 75.5 <0.001 9 5.5 – 7 4.3 – Non-PCV13 90 35.6 14 15.6 <0.001 44 48.9 – 27 30.0 – 16 17.8 0.002 16 17.8 <0.001 Total 253 100.0 18 7.1 187 73.9 150 59.3 25 9.9 23 9.1 <2 years PCV13 39 44.8 3 7.7 – 29 74.4 <0.001 23 59.0 0.001 4 10.3 – 3 7.7 – Non-PCV13 48 55.2 10 20.8 ns 17 35.4 – 11 22.9 – 12 25 ns 9 18.8 ns Total 87 100.0 13 14.9 46 52.9 34 39.1 16 18.4 12 13.8 2–4 years PCV13 71 72.4 0 0 – 65 91.5 0.004 57 80.3 <0.001 4 5.6 – 2 2.8 – Non-PCV13 27 27.6 4 14.8 0.005 18 66.7 – 10 37.0 – 2 7.4 ns 3 11.1 ns Total 98 100.0 4 4.1 83 84.7 67 68.4 6 6.1 5 5.1 5–17 years PCV13 53 77.9 1 1.9 ns 49 92.5 0.006 43 81.1 0.002 1 1.9 – 2 3.8 – Non-PCV13 15 22.1 0 0 – 9 60.0 – 6 40.0 – 2 13.3 ns 4 26.7 0.018 Total 68 100.0 1 1.5 58 85.3 49 72.1 3 4.4 6 8.8 *The P value indicates differences in the relationship between a serotype group and a clinical form with respect to the other serotypes. ‘ns’ indicates P value⩾0.05. 7 8 C. Izquierdo et al. Fig. 3. Distribution of Streptococcus pneumoniae sero- types in children aged ⩽17 years by clinical presentation. • <2 age group. #1: P < 0.001; #2: P = 0.004 (non- complicated and complicated pneumonia); #3:P = 0.033. • 2–4 age group. #4: P = 0.002; #5:P = 0.003 • 5–17 age group. #6:P < 0.001; #7: P = 0.004; #8: P = 0.004 The P values indicate the differences in the relationship between a serotype and a clinical form with respect to the other serotypes. IPD: invasive pneumococcal disease. ONV: other non-vaccine serotypes. Table 4. Distribution of Streptococcus pneumoniae serotypes causing IPD in children aged ⩽17 years by non-susceptibility to antibiotics Epidemiology and Infection Penicillin >0.06 + a Cases analysed Penicillin >0.06 Penicillin>2b Cefotaxime>0.5c Cefotaxime>0.5d Total cases Serotype n n % n % P value* n % P value* n % P value* n % P value * PCV13 1 48 27 56.3 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 3 53 6 11.3 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 4 2 2 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 6A e 3 1 33.3 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 6B 2 1 50.0 1 100 ns 0 0.0 – 0 0.0 – 0 0.0 – 7Fe 9 1 11.1 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 9V e 6 5 83.3 1 20.0 ns 0 0.0 – 1 20.0 ns 1 20.0 ns 14 15 11 73.3 11 100.0 <0.001 0 0.0 – 9 81.8 <0.001 9 81.8 <0.001 18C 2 2 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 19A 17 11 64.7 10 90.9 <0.001 3 27.3 0.002 6 54.5 0.003 6 54.5 0.003 e 19F 5 4 80.0 3 75.0 ns 1 25.0 ns 1 25.0 ns 1 25.0 ns 23F 1 1 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – Non-PCV13 6C 1 1 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 8 2 2 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 10A 6 5 83.3 0 0.0 – 0 0.0 – 0 0.0 0 0.0 24F 6 6 100.0 5 83.3 0.023 0 0.0 – 0 0.0 – 0 0.0 – 11A 5 5 100.0 3 60.0 ns 0 0.0 – 3 60.0 0.031 3 60.0 0.031 12F 3 3 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 13 1 1 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 15A 3 3 100.0 1 33.3 ns 0 0.0 – 0 0.0 – 0 0.0 – 15B 2 2 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 15C 1 1 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 16F 2 2 100.0 1 50.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 22F 2 2 100.0 0 0.0 – 0 0.0 – 0 0.0 – 0 0.0 – 23B 3 3 100.0 3 100.0 0.045 0 0.0 – 0 0.0 – 0 0.0 – 24 2 2 100.0 2 100.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 24A 2 2 100.0 2 100.0 ns 0 0.0 – 0 0.0 – 0 0.0 – 27 2 2 100.0 1 50.0 ns 0 0.0 – 0 0.0 – 0 0.0 – (Continued ) 9 10 C. Izquierdo et al. Twenty cases (16.4%) were also non-susceptible to cefotaxime 7F: 4 out of 9 cases were identified as group level (7F/7A); 9V: 2 out of 6 cases were identified as group level (9V/9A); 19F: 1 out of 5 cases was identified as group level (19F/19B/19C); 6A: 2 out of 3 cases were identified as group level (6A/6C). P value * (Table 4). 0.010 With respect to the clinical form, 5 out of 44 cases with MIC > – – – – – – Penicillin >0.06 + Cefotaxime>0.5d 0.06 were considered resistant due to the meningeal location (ser- otypes 14, 19A, 19F, 24A and 24F). One case (serotype 14) was also non-susceptible to cefotaxime. 23.6 16.4 0.0 0.0 0.0 0.0 0.0 6.0 % There were no significant differences in the percentage of non- susceptible penicillin strains between PCV13 (36.1%) and non-PCV13 serotypes (36.0%). PCV13 serotypes were associated 0 0 0 0 0 3 17 20 with strains not susceptible to cefotaxime (P = 0.010) and with n isolates not susceptible to either penicillin or cefotaxime (P = 0.010). P value* Serotype 19A showed a penicillin MIC of >0.06 mg/l in 90.9% 0.010 of cases studied (P < 0.001) and MIC >2 mg/l in 27.3% (P = – – – – – – 0.002). In addition, 54.5% of serotype 19A strains were non- Cefotaxime>0.5c susceptible to cefotaxime (P = 0.003) and non-susceptible to peni- cillin (P = 0.003). 23.6 16.4 0.0 0.0 0.0 0.0 0.0 6.0 % All serotype 14 strains (100%) showed a penicillin MIC of *The P value indicates differences in the relationship between a serotype and non-susceptibility to an antibiotic with respect to the other serotypes. ‘ns’ indicates P value⩾0.05. >0.06 mg/l (P < 0.001) and 81.8% of cases were non-susceptible to both cefotaxime (P < 0.001) and penicillin (P < 0.001). Serotype 11A showed non-susceptibility to cefotaxime in 60% 0 0 0 0 0 3 17 20 n of strains compared with 14.5% for other serotypes (P = 0.031). These strains were also non-susceptible to penicillin (P = 0.031). P value* Other serotypes showed significant differences in antimicrobial susceptibility, namely 24F, which showed a penicillin MIC of – – – – – – ns >0.06 mg/l in 83.3% (P = 0.023), and 23B in 100% (P = 0.045). Penicillin>2b Discussion 0.0 0.0 0.0 0.0 0.0 5.6 0.0 3.3 % PCV13 serotypes caused 64.4% of IPD cases in children aged ⩽17 years during the study period, and the most frequent serotypes were 3, 1, 19A and 14. Other studies in Catalonia have shown a n 0 0 0 0 0 4 0 4 higher frequency of PCV13 serotypes, despite the decreased inci- dence in all age groups in this period [23]. However, other authors P value* have found a higher percentage of non-PCV13 serotypes. Makwana et al. observed that 77.5% of cases of IPD in children – – – – – – ns a aged <5 years in England and Wales between 2010 and 2016 Penicillin >0.06 were caused by non-PCV13 serotypes although there was a high immunisation rate with 97.3% having received ⩾1PCV13 dose 0.0 0.0 0.0 0.0 0.0 36.1 36.0 36.1 % [24]. A study conducted in Madrid (Spain) between 2012 and Number and percentage of isolates of a given serotype, with penicillin MIC >0.06 mg/l. 2015, with vaccination coverages between 67% and 95%, found that non-PCV13 serotypes accounted for 68.2% of cases in chil- dren aged <15 years [25], somewhat higher than the 55.2% 0 0 0 0 0 26 18 44 n described by Janoir et al. [14] in France in 2011–2012 in the IPD, invasive pneumococcal disease; ONV, other non-vaccine serotypes. same age group, with a vaccination coverage of at least 90%. The differences between these results might be explained, at 44.2 55.6 48.2 100.0 100.0 100.0 100.0 100.0 Cases analysed % least in part, by a lower PCV13 coverage (about 64% in children aged <2 years) in Catalonia than in other areas because during the study period, PCV13 was not included in the Catalan immunisa- 72 50 122 2 2 1 1 2 Not susceptible to either penicillin or cefotaxime. tion schedule [12]. In fact, a systematic review by Balsells et al. in n children in the post-PCV era found large differences between countries. Non-PCV13 serotypes accounted for 42.2% of child- hood IPD cases, but marked regional differences were observed, Total cases from 9.2% in one Eastern Mediterranean country to 71.9% in 163 253 2 2 1 1 2 90 n With cefotaxime MIC >0.5 mg/l. Europe [26]. With penicillin MIC >2.0 mg/l. With respect to the distribution of the main serotypes, Ceyhan Table 4. (Continued.) et al. [27] found that, in Turkish children aged <18 years with 25.5% of IPD cases having received any PCV dose, serotype Non-PCV13 19F was the most frequent, followed by 14, 3 and 6B (in equal Serotype amounts), while Camilli et al. [28], in Italian children aged <5 PCV13 35B 33F 35F Total 31 38 years in 2012–2014 with PCV coverage at 24 months of age of about 80% found that serotypes 1, 19A and 14 were among the b d a e c Epidemiology and Infection 11 four most frequent serotypes, together with 24F. In addition to the There were some limitations. First, the population studied corre- differences in PCV coverage, the diagnostic techniques used in sponds to three paediatric hospitals, although these hospitals serve these studies might help to explain the differences observed 32% of the population aged ⩽17 years requiring hospitalisation in [29]. In our study, serotype 3 was the most frequent and also Catalonia. In the subgroup analyses, the small number of cases the most frequently diagnosed only by real-time PCR (87% of observed in some situations may explain why the confidence inter- cases due to serotype 3 were diagnosed only by real-time PCR). vals of the measure of association were very wide. Finally, antibiotic When real-time PCR is not available for the diagnosis, another sensitivity could only be studied in 48% of cases serotyped, since a distribution of serotypes in the same age groups may be found, large part of the cases were diagnosed by real-time PCR. because some serotypes are more antimicrobial-resistant than others and real-time PCR makes it possible to diagnose the more sensitive serotypes. Obando et al. [30] found evidence of Conclusions pneumococcal infection in 84% of culture-negative pleural fluid PCV13 serotypes were the most frequently found IPD serotypes samples on the basis of ply or wzg gene detection by PCR; in children aged ⩽17 years, especially serotype 19A in children these cases were significantly more likely to have received anti- aged <2 years, serotype 3 in the 2–4 years age group and serotype microbial drug therapy before sampling. In our study 48% of 1 in the 5–17 years age group. IPD cases were diagnosed only by real-time PCR, which served Non-PCV13 serotypes were the main cause of meningitis, to increase the etiologic diagnosis and serotype identification. occult bacteraemia and other clinical presentations, while Serotype 19A was associated with age <2 years, serotype 3 with PCV13 serotypes were mainly responsible for pneumonia. the 2–4 years age group and serotype 1 with the 5–17 years age PCV13 and non-PCV13 cases presented a high frequency of group. Other studies have found similar results, with 100% of non-susceptibility to penicillin. Non-susceptibility to both peni- cases due to serotype 19A in children aged <2 years, 63.6% of cillin and cefotaxime was associated with serotypes 19A and 14, cases due to serotype 3 in the 2–4 years age group and 80% of and serotype 19A was associated with resistance to penicillin. cases due to serotype 1 in children aged >4 years [21]. The non-susceptibility to antibiotics of non-PCV13 serotypes Compared with the pre-PCV13 era, [23] in our region the is concerning and should be monitored to apply the appropriate most frequent serotype was 19A (24.6%) in children aged <2 disease prevention strategies. years and serotype 1 (46.8%) in children aged 2–4 years, while serotype 3, now the most frequent, represented only 2.8% and Financial support. This work was supported by the Plan Nacional I + D + I, Instituto de Salud Carlos III -Subdirección General de Evaluación y Fomento de 4.2% of cases in these age groups, respectively. This may be la Investigación Sanitaria (Projects: PI11/02081 and PI11/02345) and explained by the replacement of serotypes. cofunded by the Fondo Europeo de Desarrollo Regional (FEDER) (funding PCV13 serotypes were associated with overall pneumonia and for the serotyping study and for participating in congresses) and the complicated pneumonia in children aged ⩽17 years, mainly sero- Agència de Gestió d’ Ajuts Universitaris i Recerca (AGAUR: grant number type 19A in children aged <2 years, serotype 3 in those aged <5 2017 SGR 1342) (funding of the English manuscript revision). years and serotype 1 in the 5–17 years age group. A Polish study conducted between 2011 and 2013 in all age groups, Conflict of interest. JJ Garcia-Garcia has received honoraria for speaking at found that PCV13 serotypes caused 78.4% of cases of pneumonia, symposia (Pfizer and GSK), and financial support for attending symposia and serotype 3 and serotype 1 were associated with pneumonia, as (Pfizer). None were concerned with the present study. C Muñoz-Almagro reports travel grants from Pfizer, research grants from in our study [31]. BioMerieux, Stat DX and Instituto de Salud Carlos III, personal fees from GSK In our study, non-PCV13 serotypes were associated with as consultor to an advisory board and honoraria for speaking at symposia from occult bacteraemia, especially 10A and 12F, and with meningitis Roche and Biomerieux. None were concerned with the present study. in the 2–4 years age group. Similarly, other authors have reported For the remaining authors, no competing interests were declared. that 90% of meningitis cases and 64% of occult bacteraemia cases during 2011–2012 were produced by non-PCV13 serotypes [32]. Ethical standards. Informed consent was obtained from all individual par- The percentage of strains not susceptible to penicillin (36.1%) ticipants included in the study. Written informed consent was obtained from a or to cefotaxime (16.4%) was higher than that described by other parent or guardian for participants under 16 years of age. The study complies with the principles of the Declaration of Helsinki and the legal structure with authors [23, 25]. Serotypes 19A and 14 were associated with respect to international human rights and biomedicine and protection of per- strains not susceptible to penicillin or cefotaxime, whereas sero- sonal data laws. The Ethics Committee of Hospital Sant Joan de Déu approved type 23B and 24F were not susceptible to penicillin. Other authors the study. have also described non-susceptibility to penicillin in serotypes 19A and 14 [28, 31], serotype 24F [14, 28] and serotype 23B Data. No additional data are available. [33]. Skoczyńska et al. [31] found that non-susceptibility to peni- cillin was also associated with vaccine-serotypes 19F, 6B and 9V, References which in our study could not be demonstrated, perhaps due to the small number of strains studied. They found an association 1. Wahl B et al. (2018) Global burden of Streptococcus pneumoniae in chil- between non-susceptibility to cefotaxime in serotype 19A and dren younger than 5 years in the era of pneumococcal conjugate vaccines 14, as in our study. The non-susceptibility to antibiotics of (PCV): 2000–2015. The Lancet Global Health 6, e744–e757. non-PCV13 serotypes is concerning and should be monitored. 2. Whitney CG et al. (2003) Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine. The New Among the strengths of the study is the high percentage of England Journal of Medicine 348, 1737–1746. cases serotyped, since serotyping was not possible in only 3.8% 3. Rückinger S et al. (2009) Reduction in the incidence of invasive pneumo- of cases. The same diagnostic techniques (culture and real-time coccal disease after general vaccination with 7-valent pneumococcal con- PCR) were used uniformly in all cases, improving the diagnostic jugate vaccine in Germany. Vaccine 27, 4136–4141. sensitivity and the range of serotypes identified, which permitted 4. Harboe ZB et al. (2010) Early effectiveness of heptavalent conjugate a more precise estimate of the burden of IPD. pneumococcal vaccination on invasive pneumococcal disease after the 12 C. Izquierdo et al. introduction in the Danish childhood immunization programme. Vaccine in clinical specimens. European Journal of Clinical Microbiology & 28, 2642–2647. Infectious Diseases 34, 705–711. 5. Roche PW et al. (2008) Invasive pneumococcal disease in Australia, 2006. 19. Janoff EN and Musher DM (2020) Streptococcus pneumoniae. In Bennet Communicable Diseases Intelligence 32, 18–30. JE, Dolin R and Blaser MJ (eds), Mandell, Douglas, and Bennett’s 6. Centers for Disease Control and Prevention (2010) Licensure of a Principles and Practice of Infectious Diseases, 9th Edn. Philadelphia: 13-valent pneumococcal conjugate vaccine (PCV13) and recommenda- Elsevier, pp. 2473–2491. tions for use among children – advisory committee on immunization 20. Fenoll A et al. (1997) Dot blot assay for the serotyping of pneumococci. practices (ACIP). Morbidity and Mortality Weekly Report 59, 258–261. Journal of Clinical Microbiology 35, 764–766. 7. Tin Tin Htar M, Christopoulou D and Schmitt HJ (2015) Pneumococcal 21. The European Committee on Antimicrobial Susceptibility Testing serotype evolution in Western Europe. BMC Infectious Diseases 15, 419. (2018) Breakpoint tables for interpretation of MICs and zone diameters. 8. Bernaola E et al. (2009) Vaccination schedule of the Spanish association Version 8.1, . Available at http://www.eucast.org/clinical_breakpoints/ of pediatrics: recommendations 2009. Anales de Pediatría (Barcelona) 70, Accessed July 2019. 72–82. 22. Sheskin DJ (2004) Handbook of Parametric and Nonparametric Statistical 9. Ciruela P et al. (2013) Risk factors for invasive pneumococcal disease in a Procedures, 3rd edn. Boca Raton: Chapman & Hall /CRC. community with a high proportion of non-vaccine serotypes. Vaccine 31, 23. Ciruela P et al. (2018) The changing epidemiology of invasive pneumo- 960–966. coccal disease after PCV13 vaccination in a country with intermediate vac- 10. Salleras L et al. (2009) Changes in serotypes causing invasive pneumococ- cination coverage. Vaccine 36, 7744–7752. cal disease (2005–2007 vs. 1997–1999) in children under 2 years of age in a 24. Makwana A et al. (2018) Characteristics of children with invasive population with intermediate coverage of the 7-valent pneumococcal con- pneumococcal disease after the introduction of the 13–valent pneumococ- jugated vaccine. Clinical Microbiology and Infection 15, 997–1001. cal conjugate vaccine in England and Wales, 2010–2016. Pediatric 11. Generalitat de Catalunya, Departament de Salut New vaccine schedule Infectious Disease Journal 37, 697–703. (2016) Available at http://canalsalut.gencat.cat/ca/salut-a-z/v/vacunacions/ 25. Picazo J et al. (2017) Effect of the different 13-valent pneumococcal con- calendari-de-vacunacions/. Accesed July 2019. jugate vaccination uptakes on the invasive pneumococcal disease in chil- 12. Savulescu C et al. (2017) Effect of high valency pneumococcal conjugate dren: analysis of a hospital-based and population-based surveillance vaccines on invasive pneumococcal disease in children in SpIDnet coun- study in Madrid, Spain, 2007–2015. PLoS One 12, e0172222. tries: an observational multicentre study. The Lancet Respiratory 26. Balsells E et al. (2017) Serotype distribution of Streptococcus pneumoniae Medicine 5, 648–656. causing invasive disease in children in the post-PCV era: a systematic 13. Tomczyk S et al. (2016) Prevention of antibiotic-nonsusceptible invasive review and meta-analysis. PLoS One 12, e0177113. pneumococcal disease with the 13-valent pneumococcal conjugate vac- 27. Ceyhan M et al. (2016) Serotype distribution of Streptococcus pneumoniae cine. Clinical Infectious Diseases 62, 1119–1125. in children with invasive diseases in Turkey: 2008–2014. Human Vaccines 14. Janoir C et al. (2016) Insight into resistance phenotypes of emergent non & Immunotherapeutics 12, 308–313. 13-valent pneumococcal conjugate vaccine type pneumococci isolated 28. Camilli R et al. (2017) Impact of pneumococcal conjugate vaccine (PCV7 from invasive disease after 13-valent pneumococcal conjugate vaccine and PCV13) on pneumococcal invasive diseases in Italian children and implementation in France. Open Forum Infectious Diseases 3, ofw020. insight into evolution of pneumococcal population structure. Vaccine 15. Selva L et al. (2012) Rapid and easy identification of capsular serotypes of 35, 4587–4593. Streptococcus pneumoniae by use of fragment analysis by automated 29. Selva L et al. (2012) Serotype 3 is a common serotype causing invasive fluorescence-based capillary electrophoresis. Journal of Clinical pneumococcal disease in children less than 5 years old, as identified by Microbiology 50, 3451–3457. real-time PCR. European Journal of Clinical Microbiology & Infectious 16. Tarragó D et al. (2008) Identification of pneumococcal serotypes from Diseases 31, 1487–1495. culture-negative clinical specimens by novel real-time PCR. Clinical 30. Obando I et al. (2008) Paediatric parapneumonic empyema, Spain. Microbiology and Infection 14, 828–834. Emerging Infectious Diseases 14, 1390–1397. 17. Centers for Disease Control and Prevention (CDC) (2011) PCR For 31. Skoczyńska A et al. (2015) Recent trends in epidemiology of invasive detection and characterization of bacterial meningitis pathogens: pneumococcal disease in Poland. European Journal of Clinical Neisseria meningitidis, Haemophilus influenzae, and Streptococcus pneu- Microbiology & Infectious Diseases 34, 779–787. moniae. In WHO & CDC. Laboratory Methods for the Diagnosis of 32. Picazo J et al. (2013) Expansion of serotype coverage in the universal Meningitis Caused by Neisseria meningitidis, Streptococcus pneumo- pediatric vaccination calendar: short-term effects on age- and serotype- niae, and Haemophilus influenzae. 2nd edn. Geneve & Atlanta: WHO dependent incidence of invasive pneumococcal clinical presentations in & CDC. pp. 105–156 [cited 2020 Oct 4]. Available at https://www.cdc. Madrid, Spain. Clinical and Vaccine Immunology 20, 1524–1530. gov/meningitis/lab-manual/index.html. 33. Van der Linden M, Perniciaro S and Imöhl M (2015) Increase of sero- 18. del Amo E et al. (2015) Estimation of the invasive disease potential of types 15A and 23B in IPD in Germany in the PCV13 vaccination era. Streptococcus pneumoniae in children by the use of direct capsular typing BMC Infectious Diseases 15, 207.