Advertisement

Swiss Evaluation Registry for Pediatric Infective Endocarditis (SERPIE) - Risk factors for complications in children and adolescents with infective endocarditis

  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Stefanie Katharina Schuler
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Pediatric Cardiology, Pediatric Heart Center, University Children's Hospital Zurich, University of Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland

    Children's Research Center, University Children's Hospital Zurich, University of Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Pierre Alex Crisinel
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Pediatric Infectious Diseases, Service of Pediatrics, Department Woman-Mother-Child, Lausanne University Hospital and University of Lausanne, Route du Bugnon 46, CH-1011 Lausanne, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Raphael Joye
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Pediatric Cardiology Unit, Children's University Hospital, Rue Willy-Donzé 6, CH-1205 Geneva, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Marie Rohr
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Pediatric Infectious Diseases Unit, Department of Pediatrics, University Hospital Geneva, University of Geneva, Rue Willy-Donzé 6, CH-1205 Geneva, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Sabrina Bressieux-Degueldre
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Pediatric Cardiology Unit, Service of Pediatrics, Department Woman-Mother-Child, Lausanne University Hospital and University of Lausanne, S; Route du Bugnon 46, CH-1011 Lausanne, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Martin Glöckler
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Department of Congenital Heart Diseases, Inselspital, Bern University Hospital, University of Bern, Freiburgstrasse 15, CH-3010 Bern, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Paolo Paioni
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Division of Infectious Diseases and Hospital Epidemiology, University Children's Hospital Zurich, University of Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Philipp K.A. Agyeman
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Department of Pediatrics, Inselspital, Bern University Hospital, University of Bern, Freiburgstrasse 15, CH-3010 Bern, Switzerland
    Search for articles by this author
  • Walter Knirsch
    Correspondence
    Corresponding author at: University Children's Hospital Zurich, Pediatric Heart Center, Pediatric Cardiology, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland.
    Affiliations
    Pediatric Cardiology, Pediatric Heart Center, University Children's Hospital Zurich, University of Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland

    Children's Research Center, University Children's Hospital Zurich, University of Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
Open AccessPublished:November 01, 2022DOI:https://doi.org/10.1016/j.ijcard.2022.10.173

      Highlights

      • SERPIE is the first nation-wide registry for pediatric infective endocarditis in Switzerland
      • Most patients with pediatric infective endocarditis suffered from a congenital heart disease
      • The right ventricular to pulmonary artery outflow tract conduit was mostly infected
      • The most common pathogen for infective endocarditis is Staphylococcus aureus
      • Staphylococcus aureus is significantly associated with severe complications

      Abstract

      Background

      Infective endocarditis (IE) in pediatric patients is a severe cardiac disease and its actual epidemiology and clinical outcome in Switzerland is scarcely studied.

      Methods

      Retrospective nationwide multicenter data analysis of pediatric IE in children (<18 years) between 2011 and 2020.

      Results

      69 patients were treated for definite (40/69;58%) or possible IE (29/69;42%). 61% (42/69) were male. Diagnosis was made at median 6.4 years (IQR 0.8–12.6) of age with 19 patients (28%) during the first year of life. 84% (58/69) had congenital heart defects. IE was located on pulmonary (25/69;35%), mitral (10/69;14%), tricuspid (8/69;12%) and aortic valve (6/69;9%), and rarely on ventricular septal defect (VSD;4/69;6%) and atrial septal defect (ASD;1/69;1%). In 22% (16/69) localization was unknown. 70% (48/69) had postoperative IE, with prosthetic material involved in 60% (29/48; right ventricular to pulmonary artery conduit (24), VSD (4), ASD (1)). Causative organisms were mostly Staphylococci spp. (25;36%) including Staphylococcus aureus (19;28%), and Streptococci spp. (13;19%). 51% (35/69) suffered from severe complications including congestive heart failure (16;23%), sepsis (17;25%) and embolism (19;28%). Staphylococcus aureus was found as a predictor of severe complications in univariate and multivariate analysis (p = 0.02 and p = 0.033). In 46% (32/69) cardiac surgery was performed. 7% (5/69) died.

      Conclusions

      IE in childhood remains a severe cardiac disease with relevant mortality. The high morbidity and high rate of complications is associated with Staphylococcus aureus infections. Congenital heart defects act as a risk factor for IE, in particular the high number of cases associated with prosthetic pulmonary valve needs further evaluation and therapeutic alternatives.

      Keywords

      1. Introduction

      Infective endocarditis (IE) in childhood is associated with high morbidity and mortality [
      • Dixon G.
      • Christov G.
      Infective endocarditis in children: an update.
      ,
      • Knirsch W.
      • Nadal D.
      Infective endocarditis in congenital heart disease.
      ,
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ]. The incidence of pediatric IE is reported with 0.43–0.84/100′000 children per year [
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Jortveit J.
      • Eskedal L.
      • Klcovansky J.
      • Døhlen G.
      • Holmstrøm H.
      Prevalence of infective endocarditis in children.
      ,
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ], with increasing trend in recent years [
      • Sakai Bizmark R.
      • Chang R.K.R.
      • Tsugawa Y.
      • Zangwill K.M.
      • Kawachi I.
      Impact of AHA’s 2007 guideline change on incidence of infective endocarditis in infants and children.
      ,
      • Baltimore R.S.
      • et al.
      Infective endocarditis in childhood: 2015 update - a scientific statement from the American Heart Association.
      ,
      • Rushani D.
      • et al.
      Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors.
      ,
      • Rosenthal L.B.
      • et al.
      The changing epidemiology of pediatric endocarditis at a children’s hospital over seven decades.
      ].
      In developed countries, congenital heart disease (CHD) plays the dominant role as predisposing factor for pediatric IE [
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ,
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ]. This includes children with native and prosthetic valves affected [
      • Dixon G.
      • Christov G.
      Infective endocarditis in children: an update.
      ,
      • Rushani D.
      • et al.
      Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors.
      ,
      • Day M.D.
      • Gauvreau K.
      • Shulman S.
      • Newburger J.W.
      Characteristics of children hospitalized with infective endocarditis.
      ]. Nowadays, prosthetic valve IE in children with CHD are associated with severe clinical course leading to complications [
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Day M.D.
      • Gauvreau K.
      • Shulman S.
      • Newburger J.W.
      Characteristics of children hospitalized with infective endocarditis.
      ]. From the clinical point of view, early determination of risk factors contributing to complications is essential.
      Due to the limited cases of pediatric IE, multicenter surveys such as nation-wide registries are needed. In Switzerland, two single-center studies on pediatric infective endocarditis have been published [
      • Weber R.
      • et al.
      Interventions using foreign material to treat congenital heart disease in children increase the risk for infective endocarditis.
      ,
      • Knirsch W.
      • Schuler S.K.
      • Martin C.
      • Roland W.
      Time-trend population analysis of the clinical and epidemiologic effect on pediatric infective endocarditis after change of antibiotic prophylaxis guidelines.
      ]. So far, nation-wide data are missing. Therefore, the Swiss Evaluation Registry for Pediatric Infective Endocarditis (SERPIE) was founded in 2020 with the purpose to build up a prospective data collection of the microbiological spectrum, diagnostics, predisposing risk factors, clinical course, complications, therapy, and outcome of pediatric IE in Switzerland. To optimally prepare for the launch of the SERPIE, we conducted the analysis based on national retrospective data on pediatric IE presented here.
      As it seems that especially children with CHD experience more severe courses of IE [
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Day M.D.
      • Gauvreau K.
      • Shulman S.
      • Newburger J.W.
      Characteristics of children hospitalized with infective endocarditis.
      ], and in the last two studies conducted in Switzerland, mainly children with CHD were found to be at risk for IE, especially when foreign material was used for treatment of the CHD [
      • Weber R.
      • et al.
      Interventions using foreign material to treat congenital heart disease in children increase the risk for infective endocarditis.
      ,
      • Knirsch W.
      • Schuler S.K.
      • Martin C.
      • Roland W.
      Time-trend population analysis of the clinical and epidemiologic effect on pediatric infective endocarditis after change of antibiotic prophylaxis guidelines.
      ], the SERPIE study also aimed to analyze factors associated with complicated clinical courses.

      2. Methods

      2.1 Study design

      In 2020, a network of pediatric cardiologist and pediatric infectious disease specialists was built up on behalf of the “Verein Kinderherzforschung Schweiz” (Association of Children's Heart Research Switzerland) and with the consent of the participating pediatric cardiac surgeons and the Swiss Society of Pediatric Cardiology. In Switzerland, the pediatric cardiac surgery program concentrates on four main tertiary pediatric heart centers performing invasive pediatric cardiac procedures including cardiac surgery as well as catheter-based cardiac interventions. This includes the tertiary Pediatric University Children's Hospital in Berne, Geneva, Lausanne, and Zurich.
      This study was designed as a retrospective nationwide multicenter analysis including cases of pediatric infective endocarditis in children under 18 years of age treated in Switzerland between 2011 and 2020. After ethical committee approval was achieved (BASEC Number 2021–00862), patients were searched in the local medical data bases in each of the main four pediatric cardiac center, since our preliminary discussion revealed that all cases treated for IE have been doubtfully treated within these four main centers or in direct cooperation with them.

      2.2 Patients and variables

      Patients' consent was given by primary (general) consent for retrospective data analysis, or a secondary consent was obtained in written form for study participation. Only patients fulfilling the modified Duke criteria for definite or possible IE were included [
      • Li J.S.
      • et al.
      Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis.
      ]. Patients' data were recorded in an encoded way into REDCap®. We collected data analysis regarding diagnostic, microbiological, clinical and therapeutic variables, and outcomes. To calculate the incidence of IE in Switzerland, national data from the Federal Statistical Office on children under 18 years of age living in Switzerland were used (https://www.bfs.admin.ch/bfs/de/home/stastiken/bevoelkerung/stand-entwicklung/bevoelkerung.assetdetail.18404679.html).
      Based on the AHA guideline for the antibiotic endocarditis prophylaxis from 2007 [
      • Wilson W.
      • et al.
      Prevention of infective endocarditis: guidelines from the American Heart Association.
      ], which were adopted in 2008 for Switzerland [
      • Flückiger U.
      • Jaussi A.
      Revidierte schweizerische Richtlinien für die Endokarditis-Prophylaxe.
      ], all patients were classified into a “predisposed” or “non-predisposed” group, i.e. antibiotic prophylaxis was recommended or not during high-risk interventions. This includes patients with prosthetic valve, patients with previous IE, patients with unrepaired cyanotic CHD including palliative shunts and conduits or patients with any type of CHD repaired with a prosthetic material within 6 months after procedure or lifelong if residual shunt remains [
      • Wilson W.
      • et al.
      Prevention of infective endocarditis: guidelines from the American Heart Association.
      ]. The first day of intravenous antibiotic therapy was defined as the day of diagnosis of IE. Postoperative IE was diagnosed if the patient underwent previous cardiac surgery or catheter intervention. Severe complications were defined as sepsis, embolism or severe congestive heart failure [
      • Habib G.
      • et al.
      2015 ESC guidelines for the management of infective endocarditis.
      ]. Severe congestive heart failure was defined as impaired ventricular myocardial function determined by echocardiography, hemodynamic instability, and clinical signs of congestive heart failure such as hepatomegaly, pulmonary oedema, or cardiac-related renal failure. Patients were classified into “acute” versus “subacute” IE accordingly to their clinical course. IE was defined as “acute” if severe complications occurred and/or early surgical intervention during antibiotic treatment was performed. Predisposing factors for IE were known CHD, prior history of IE, comorbidities like immunodeficiency, genetic or oncologic comorbidities, and temporary or permanent inserted central venous catheter line before diagnosis of IE.

      2.3 Statistics

      Statistical analysis was performed using STATA Version 16 (StataCorp, Texas, United States of America). Continuous variables are described as medians and interquartile ranges (IQRs), categorial variables as counts and frequencies. Yearly and median incidence was calculated using the yearly respective median numbers of patients treated for IE and children under 18 years of age living in Switzerland. Intergroup comparison was done using Kruskal-Wallis test for continuous variables and chi-square test or Fisher's exact test for categorial variables. A multivariate logistic regression analysis with adjustment for age, co-morbidities and gender was performed for each potential predictor identified in the univariate analyses (with P-value ≤0.05) and adjusted odds ratio (OR) and 95% confidence interval (CI) were calculated. All tests were 2-tailed, and a P-value ≤0.05 was considered statistically significant.

      3. Results

      3.1 Patients

      The medical data base research revealed a total number of 95 patients fulfilling inclusion criteria. 24 patients were treated as part of humanitarian project and lost to follow up. For 48 patients a primary consent was available. Of the remaining 23 patients, 2 refused consent so that finally a total of 69 patients were included in the study.
      Patients fulfilled the modified Duke criteria for definite IE and for possible IE in 58% (40/69) and 42% (29/69) of cases, respectively. IE was diagnosed at a median age of 6.4 years (IQR 0.8–12.6) with 28% (19/69) during the first year of life. 61% (42/69) were male patients. The median incidence of IE was 0.33 (IQR 0.27–0.54) patients per year and 100′000 children under 18 years of age living in Switzerland.

      3.2 Predisposing factors for IE

      Fifty-eight of the 69 (84%) patients had a documented congenital heart disease (Table 1) as a predisposing factor, of those 55% were cyanotic CHD (32/58). 17% were unrepaired CHD (10/58), while 83% (48/58) patients had postoperative IE. At least 60% (29/48) of the patients with postoperative IE were proven to be associated with prosthetic material. Postoperative IE was diagnosed at a median of 1.5 years (IQR 0.4–8.8) after cardiac intervention. Ten of 48 (21%) patients with postoperative IE had an early postoperative IE within 30 days after cardiac procedure. Two of the 69 (3%) patients already suffered from IE before. According to the guidelines for antibiotic prophylaxis in Switzerland, 61% (42/69) of the patients were at high risk for IE, where antibiotic prophylaxis is recommended [
      • Flückiger U.
      • Jaussi A.
      Revidierte schweizerische Richtlinien für die Endokarditis-Prophylaxe.
      ]. In only 2 cases prior dental treatment was documented, these children received antibiotic prophylaxis.
      Table 1Predisposing factors for IE in children.
      Congenital heart disease
      some patients presented with more than one predisposing factor.
      RV-PA conduit
      Cyanotic CHDn = 32n = 26
       Tetralogy of Fallot (TOF)n = 9n = 9
       Pulmonary atresia (PA)n = 7n = 7
       Double outlet right ventricle (DORV)n = 5n = 3
       Truncus arteriosus communisn = 4n = 4
       Transposition of the great arteries (TGA)n = 3n = 2
       Hypoplastic left heart syndrome (HLHS)n = 2n = 1
       Double outlet right ventricle with tricuspid atresia (DORV +TA)n = 1n = 0
       Unbalanced AVSDn = 1n = 0
      Non-cyanotic CHDn = 26n = 6
       Ventricular septal defect (VSD)n = 8n = 0
       Bicuspid aortic valve (AV)n = 6n = 2
       Atrioventricular septal defect (AVSD)n = 4n = 0
       Aortic valve (AV) stenosisn = 3n = 2
       Atrial septal defect (ASD)n = 2n = 0
       Absent pulmonary valve (PV)n = 1n = 1
       Shone-complexn = 1n = 1
       Complete atrioventricular (AV)-blockn = 1n = 0
      Other predisposing factors
      some patients presented with more than one predisposing factor.
       Central venous linen = 9 (one permanent)
       Genetic syndromesn = 5
       DiGeorge syndromen = 3 (one with trisomy 9p)
       Trisomy 21n = 1
       Right isomerism with asplenian = 1
      Immunodeficiencyn = 4
       Hypogammaglobulinemian = 2
       Autoimmune neutropenian = 1
       Chronic granulomatous diseasen = 1
      Oncologic comorbiditiesn = 3
       Common ALLn = 2
       Kidney clear cell sarcoman = 1
      Pretermn = 2
      Previous IEn = 2
      No known predisposing factorsn = 8
      low asterisk some patients presented with more than one predisposing factor.
      In the group of patients with CHD, the following additional risk factors were found: central venous catheter (n = 6), genetic syndromes (n = 5), oncologic comorbidities (n = 3), primary immunodeficiency (n = 2) and previous IE (n = 2). Two patients with CHD were born preterm with persistent ductus arteriosus (PDA) and IE occurring early after birth.
      In 29 of 69 (42%) a possible source for transient bacteremia was identified, mostly skin or soft tissue infections (n = 10), or central venous catheters (n = 9; see Table A in appendix).

      3.3 Localization of IE

      IE affected valves in pulmonary (25/69; 35%), mitral (10/69; 14%), tricuspid (8/69; 12%) and aortic position (6/69; 9%), VSD (4/69; 6%), ASD II (1/69; 1%) and PDA (2/69; 3%). Multiple infections occurred in 2 of 69 (3%) patients, with one patient with IE affecting tricuspid valve and a VSD-patch, and one patient with an unrepaired AVSD with infected mitral and tricuspid valve and infected PDA. Localization was undetermined in 16 of 69 (22%). In total, isolated right sided IE occurred in 31 of 69 (45%), isolated left sided IE in 15 of 69 (22%).
      Prosthetic valve associated IE affected only the right ventricular to pulmonary artery (RV-PA) conduit in 24 of 48 (41%) postoperative IE. The type of RV-PA conduit affected included in 18 of 24 (75%) predominately the Contegra valve conduit (Medtronic Co., Münchenbuchsee. Switzerland), in 4 of 24 (17%) the Melody valve (Medtronic Co., Münchenbuchsee, Switzerland), in 1 of 24 (4%) a homograft, and in 1 of 24 (4%) a Shelhigh-Conduit (Shelhigh Inc., New Jersey, USA). Of note, all IE localized in a VSD or ASD were associated with foreign material such as VSD patches (n = 3) including a loose patch VSD, or Amplatzer (Abbott Medical Co., Baar, Switzerland) septal occluder for VSD (n = 1) or ASD (n = 1) closure. All 16 of 69 (22%) patients with undetermined localization of IE had CHD, the majority (13/16; 81%) had postoperative IE with implanted foreign material in 12 of 13 (92%) patients. Implanted foreign material in these patients included VSD patch (n = 10), RV-PA (n = 7, Contegra valve (n = 5), Melody valve (n = 2)), prosthetic mitral valve, ASD patch or other (each n = 1).

      3.4 Imaging findings

      In all patients (69/69; 100%) transthoracic echo (TTE) was performed for diagnosis, 29% (20/69) additionally underwent transesophageal echo (TEE). Positive echocardiographic findings according to the Duke criteria were found in 42 of 69 (61%) patients including vegetation (38/42; 90%), new valve regurgitation (8/42; 19%), new dehiscence of prosthesis (2/42; 5%), and paravalvular abscess (2/42; 5%; multiple findings included). In 10 of 69 patients (14%) only pathologic echo findings not fulfilling the Duke criteria were found, in 17 of 69 patients (25%) echo was negative. Therefore, the positive rate according to Duke was 55% (38/69) in TTE and 60% (12/20) in TEE. With all pathologic findings included the positive rate was 71% (49/69) in TTE and 70% (14/20) in TEE.
      One patient with negative echo presented the picture of IE on the pulmonary valve in PET-CT that was performed to ensure diagnosis.

      3.5 Microbiological findings

      Pathogens determined in blood culture are listed in Table 2. Blood culture negative IE was found in 8 of 69 (12%) patients. In 3 of 8 (38%) patients the causative pathogen could be determined by positive serology for Coxiella burnetii (n = 1), positive culture for coagulase negative Staphylococcus on an explanted valve (n = 1) and positive culture for Nocardia on tracheal secretion in a patient with chronic granulomatous disease (n = 1). No pathogen could be determined by any technique in 5 of 69 (7%) patients, all of them meeting the Duke criteria for possible IE.
      Table 2Blood culture findings.
      All patients

      n = 69 (100%)
      Foreign material associated IE

      n = 33 (48%)
      Non foreign material IE

      n = 20 (29%)
      Unknown localization

      n = 16 (23%)
      Staphylococcus spp.n = 25 (36%)n = 10 (14%)n = 11 (16%)n = 4 (6%)
      Staphylococcus aureusn = 19 (28%)n = 7 (10%)n = 9 (13%)n = 3 (4%)
      Coagulase negative Staphylococcusn = 6 (9%)n = 3 (4%)n = 2 (3%)n = 1 (1%)
      Streptococcus spp.n = 13 (19%)n = 6 (9%)n = 3 (4%)n = 4 (6%)
      Viridans-Streptococcin = 10 (14%)n = 6 (9%)n = 0 (0%)n = 4 (6%)
      Non-Viridans-Streptococcin = 3 (4%)n = 0 (0%)n = 3 (4%)n = 0 (0%)
      Enterococci spp.n = 8 (12%)n = 2 (3%)n = 2 (3%)n = 4 (6%)
      HACEKn = 6 (9%)n = 2 (3%)n = 2 (3%)n = 2 (3%)
      Other pathogensn = 9 (13%)n = 4 (6%)n = 3 (4%)n = 2 (3%)
      Hemophilus influenza type Bn = 2n = 0n = 1n = 1
      Granulicatella adiacensn = 2n = 1n = 0n = 1
      Abiotrophia defectivan = 1n = 1n =0n = 0
      E. colin = 1n = 1n = 0n = 0
      Klebsiella oxytocan = 1n = 0n = 1n = 0
      Serratia marescensn = 1n = 0n = 1n = 0
      Candida albicansn = 1n = 1n = 0n = 0
      Blood-culture negative IEn = 8 (12%)n = 5 (7%)n = 3 (4%)n = 0 (0%)
      Abbreviations: HACEK = Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, Kingella.
      Associated bacterial findings with the same pathogen determined in blood culture were found in 18% (11/61) with positive cultures of cerebrospinal fluid (n = 3), explanted valve material (n = 7), wound swab of necrotizing fasciitis (n = 1) and urine (n = 1).

      3.6 Clinical findings and complications

      Clinical findings are shown in Table 3. The diagnosis was made at median 6.5 days (IQR 3–14) after first symptom, which most often was fever (59/69; 86%). 58% (40/69) of the patients were diagnosed with acute IE.
      Table 3Clinics, complications, therapy and outcome.
      Clinical findings
      Fever and/or shiveringn = 66 (96%)
      Decreased general health conditionn = 36 (52%)
      New / altered valve function
      multiple findings possible.
      n = 29 (42%)
       Aortic valve regurgitationn = 5
       Mitral valve regurgitationn = 10
       Mitral valve stenosisn = 1
       Pulmonary valve regurgitationn = 4
       Pulmonary valve stenosisn = 10
       Acute obstruction of melody valven = 1
       Tricuspid valve regurgitationn = 3
       Tricuspid valve stenosisn = 1
      Gastrointestinal symptoms
      multiple findings possible.
      n = 24 (35%)
       Nausea and/or vomitingn = 17
       Inappetencen = 10
       Diarrhean = 13
      Abdominal painn = 8
      Respiratory tract infection
      multiple findings possible.
      /
      without impaired cardiac function.
      n = 8 (12%)
       Coughn = 4
       Dyspnean = 2
       Rhinitisn = 2
       Sore throatn = 2
      IE specific skin affection
      multiple findings possible.
      n = 2 (3%)
       Osler nodesn = 2
       Splinter hemorrhagesn = 2
       Janeway lesionsn = 1
      Unspecific skin affection
      multiple findings possible.
      n = 7 (10%)
       Maculopapular exantheman = 3
       Petechian = 3
       Suffusion on left handn = 1
      Other symptoms
      multiple findings possible.
      n = 19 (28%)
       Headachen = 11
       Weight lossn = 6
       Myalgia / arthralgian = 5
      No symptomsn = 1 (1%)
      Severe complications
      Congestive heart failuren = 16 (23%)
      Sepsisn = 17 (25%)
      Embolismn = 19 (28%)
       Lungn = 11
       Brainn = 9
       Spleenn = 3
       Kidneyn = 2
       Extremitiesn = 2
       Retinan = 1
      Deathn = 5 (7%)
      Surgical intervention
      Surgery donen = 32 (46%)
       Early surgical intervention needed24/32 (75%)
      Interval surgery after diagnosis IE, median days (IQR)21 (7–41)
       For isolated left heart infections7 (3−20)
       For isolated right heart infections31 (11–64)
      Indication for surgery
      some patients fulfilled more than one indication for surgery.
       Valve regurgitation13/32 (41%)
       Valve stenosis12/32 (38%)
       Embolism or large valvular vegetations10/32 (31%)
       Pseudoaneurysm due to rupture of ventriculo-prostetic suture1/32 (3%)
       Change of pacemaker leads1/32 (3%)
       AVSD-direct closure1/32 (3%)
       Loose VSD-patch1/32 (3%)
      Type of surgery
      some patients received more than one type of surgery.
       Valve replacement15/32 (47%)
       Valve reconstruction12/32 (38%)
       Large vegetation removal9/32 (28%)
       Thrombus removal in right ventricle2/32 (6%)
       VSD-Closure2/32 (6%)
       ASD-Closure1/32 (3%)
       Change of pacemaker leads1/32 (3%)
      Hospitalization
       Duration of hospitalization, median days (IQR)42 (23–46)
       Intensive care neededn = 40 (58%)
      low asterisk multiple findings possible.
      low asterisklow asterisk without impaired cardiac function.
      low asterisklow asterisklow asterisk some patients fulfilled more than one indication for surgery.
      low asterisklow asterisklow asterisklow asterisk some patients received more than one type of surgery.
      One of the 69 patients (1%) did have no symptoms. This was a 3-month-old boy with complete congenital AV-blockage after implantation of a pacemaker with ventricular pacing leads. At echocardiographic follow-up 30 days after intervention, he presented with new tricuspid valve stenosis and pacemaker leads suspicious of infection. Coagulase negative Staphylococcus was identified on the replaced tricuspid valve.
      Infectious diseases associated with the manifestation of IE were found in 20% (14/69) of patients, most often pneumonia (11/69; 16%) and meningitis (3/69; 4%; see Table B in appendix). Many children already presented with multiple infectious foci at the start of hospitalization, so it is not possible to say with certainty which one appeared first. 3 of the 69 patients (4%) suffered severe invasive infections: one immunocompetent patient had invasive Candida infection with pneumonia, a patient with hypogammaglobulinemia had invasive Hemophilus influenzae type B infection with epiglottitis, and one patient with chronic granulomatous disease suffered from systemic nocardiosis with septic shock and acute respiratory distress syndrome (ARDS).

      3.7 Clinical course, therapy and outcome

      Information on severe complications, therapy and outcome are listed in Table 3 (more detailed information in Table C in appendix). Intravenous specific antibiotic therapy (69/69; 100%) was performed for at least 4 weeks according to international guidelines [
      • Knirsch W.
      • Mackenzie C.R.
      • Schäfers H.-J.
      • Baumgartner H.
      • Kramer H.-H.
      Infektiöse Endokarditis und Enokarditisprophylaxe - Leitlinien der deutschen Gesellschaft für Pädiatrische Kardiologie.
      ]. Pathogens found in patients who required surgical intervention (Table 3) were Staphylococcus spp. (15/32; 47%), Streptococcus spp. (5/32; 16%), HACEK (3/32; 9%), Enterococcus (3/32; 9%) and other (4/32; 13%). Severe complications occurred in 52% (Table 3), 7% of patients (5/69) died.
      A 2-month-old preterm infant (gestational age 33 6/7) died within 10 days after valve replacement due to severe congenital mitral valve regurgitation. He suffered IE of unknown localization and died of coagulase-negative staphylococcal sepsis with severe heart failure and cardiogenic shock.
      A 17-year-old patient with DiGeorge syndrome and low native T cells, postoperative pulmonary atresia with implantation of a Melody valve died of pulmonary valve IE positive for Staphylococcus aureus. He suffered fulminant sepsis with multi-organ failure and DIC.
      Another 4-month-old with HLSH died three months after cardiac surgery due to a coagulase-negative Staphylococcus IE on the tricuspid valve. Although he underwent surgical intervention with valve repair because of severe tricuspid regurgitation, he died of sepsis with heart failure and severe hypoxemia.
      A 4-year-old child died after aortic valve IE and systemic nocardiosis with septic shock and multiple organ failure. After IE diagnosis, he was diagnosed with previously unknown chronic granulomatous disease.
      A 6-month-old patient with AVSD repair three months prior to diagnosis of IE at unknown location died of enterococcal sepsis following necrotizing enterocolitis with ARDS and renal failure.
      In our study period one patient had a relapse of IE.

      3.8 Predictors for IE with complications

      In univariate analysis, Staphylococcus aureus-infections was found as a predictor of severe complications, whereas age, CHD, prosthetic material and CRP levels at admission were not associated with complicated clinical courses (Table 4). In the multivariate analysis adjusted for age, gender and risk factors, Staphylococcus aureus-infection was also found as a predictor of severe complications (p = 0.033, OR 4.079, 95%-CI 1.117–14.89).
      Table 4Univariate analysis - predictors of severe complications in children with IE.
      All IEBy Severe Complications


      n = 69
      Present

      n = 35 (51%)
      Not Present

      n = 34 (49%)
      p-value
      Age, median years (IQR)6.39 (0.81–12.60)6.90 (0.6–14.3)5.35 (0.8–11.4)0.50
      Male, n total (%)42/69 (61%)25/42 (60%)17/42 (40%)0.07
      Diagnosis after first symptome, median days (IQR)6.5 (3–14)6 (3–14)7 (3−13)0.90
      Risk factors, n total (%)59/69 (86%)27/59 (46%)32/59 (54%)0.08
       Congenital heart disease58/69 (84%)27/58 (46%)31/58 (54%)0.19
       Cyanotic heart disease32/69 (46%)13/32 (41%)19/32 (59%)0.12
       CHD recommended for antibiotic prophylaxis42/58 (72%)18/42 (43%)24/42 (57%)0.37
       Previous IE2/69 (3%)1/2 (50%)1/2 (50%)1
       Central venous catheter9/69 (13%)4/9 (44%)5/9 (56%)0.73
      Postoperative IE, n total (%)48/69 (70%)21/48 (44%)27/48 (56%)0.26
       Foreign material used44/48 (92%)20/44 (45%)24/44 (55%)0.62
       RV-PA conduit32/69 (46%)13/32 (41%)19/32 (59%)0.12
       Surgery within 30 days before diagnosis9/69 (19%)3/9 (33%)6/9 (67%)0.31
      Localization of IE, n total (%)0.70
       Left heart15/69 (22%)10/15 (66%)5/15 (33%)
       Right heart31/69 (45%)14/31 (45%)17/31 (55%)
       Other23/69 (33%)11/23 (48%)12/23 (52%)
      Pathogen found, n total (%)0.02
      Staphylococcus aureus19/69 (28%)14/19 (74%)5/19 (26%)0.02
       Coagulase negative Staphylococcus7/69 (10%)5/7 (71%)2/7 (29%)0.43
      Streptococcus13/69 (19%)5/13 (38%)8/13 (62%)0.33
      HACEK6/69 (9%)3/6 (50%)3/6 (50%)1
      Enterococcus8/69 (12%)5/8 (63%)3/8 (37%)0.71
      Clinical course
       CRP level at admission, median concentration (mg/l) (IQR)75 (32–157.2)101.5 (47–206)55.6 (28–106)0.11
      Abbreviations: CRP = c-reactive protein; HACEK = Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, Kingella; RV-PA = Right-ventricular to pulmonary artery conduit.

      4. Discussion

      This study constitutes the first nationwide analysis of pediatric infective endocarditis in Switzerland and based on the findings of 69 patients treated for IE between 2011 and 2020. The incidence for pediatric IE of 0.33 per 100′000 per year was comparable to prior published single center cohorts [
      • Knirsch W.
      • Schuler S.K.
      • Martin C.
      • Roland W.
      Time-trend population analysis of the clinical and epidemiologic effect on pediatric infective endocarditis after change of antibiotic prophylaxis guidelines.
      ]. In our cohort, more than half of the patients suffered from severe complications leading to a high demand for surgical treatment in 46%.
      The most frequent pathogen causing IE in our study population was S. aureus (26%) followed by Viridans Streptococci (14%). The impact of both pathogen is well-known since decades as the causative pathogen for pediatric IE [
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Shamszad P.
      • Khan M.S.
      • Rossano J.W.
      • Fraser C.D.
      Early surgical therapy of infective endocarditis in children: a 15-year experience.
      ,
      • Jomaa W.
      • et al.
      Clinical features and prognosis of infective endocarditis in children: insights from a Tunisian multicentre registry aspects.
      ]. In recent decades, however, there has been a shift towards an increasing number of S. aureus associated IE [
      • Rosenthal L.B.
      • et al.
      The changing epidemiology of pediatric endocarditis at a children’s hospital over seven decades.
      ]. A contributing factor is that Viridans Streptococci are most common in patients with underlying CHD, whereas Staphylococcus is increasingly found in the absence of predisposing CHD [
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Esposito S.
      • et al.
      Infective endocarditis in children in Italy from 2000 to 2015.
      ,
      • Marom D.
      • Ashkenazi S.
      • Samra Z.
      • Birk E.
      Infective endocarditis in previously healthy children with structurally normal hearts.
      ]. The latter group seems to have increased in the last decades [
      • Elder R.W.
      • Baltimore R.S.
      The changing epidemiology of pediatric endocarditis.
      ]. Overall, the increasing prevalence of more “aggressive” pathogens seem to change the clinical picture with frequent complications and severe clinical course.
      The overall mortality in our study population was 7% and is comparable with recently published results for pediatric IE, ranging between 3% and 24% [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Rosenthal L.B.
      • et al.
      The changing epidemiology of pediatric endocarditis at a children’s hospital over seven decades.
      ,
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ,
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ,
      • Day M.D.
      • Gauvreau K.
      • Shulman S.
      • Newburger J.W.
      Characteristics of children hospitalized with infective endocarditis.
      ,
      • Jomaa W.
      • et al.
      Clinical features and prognosis of infective endocarditis in children: insights from a Tunisian multicentre registry aspects.
      ,
      • Esposito S.
      • et al.
      Infective endocarditis in children in Italy from 2000 to 2015.
      ,
      • Marom D.
      • Ashkenazi S.
      • Samra Z.
      • Birk E.
      Infective endocarditis in previously healthy children with structurally normal hearts.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ,
      • Sun L.C.
      • et al.
      Risk factors for infective endocarditis in children with congenital heart diseases - a nationwide population-based case control study.
      ,
      • Johnson J.A.
      • Boyce T.G.
      • Cetta F.
      • Steckelberg J.M.
      • Johnson J.N.
      Infective endocarditis in the pediatric patient: a 60-year single-institution review.
      ]. Despite the mortality, high morbidity of IE may be determined by the occurrence of major complications in more than half (51%) of our study population. All our patients dying from pediatric IE suffered from severe complications. In contrast to other studies investigating severe courses of pediatric IE, but focusing more on embolic complications or death [
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ], we analyzed three main complications including sepsis, embolism, and severe congestive heart failure. Therefore, the high number of complicated courses in our study population may be overestimated.
      In our study population, S. aureus was found as a risk factor for severe complicative course (Table 4), which has been described by other groups as risk factor not only for complicated courses, but also for death [
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ,
      • Johnson J.A.
      • Boyce T.G.
      • Cetta F.
      • Steckelberg J.M.
      • Johnson J.N.
      Infective endocarditis in the pediatric patient: a 60-year single-institution review.
      ]. However, we could not find an association between coagulase-negative Staphylococcus (CoNS) infection and severe complications, although CoNS were identified as the causative agent in two of the five patients dying due to postoperative IE. Other studies already described CoNS as a risk factor for severe complications and death in pediatric IE with prosthetic valves [
      • Khoo B.
      • et al.
      Outcomes of surgery for infective endocarditis in children: a 30-year experience.
      ] as well as for the adult population [
      • Lalani T.
      • et al.
      Prosthetic valve endocarditis due to coagulase-negative staphylococci: findings from the international collaboration on endocarditis merged database.
      ].
      Nearly half of our study population (47%) needed early surgical intervention for IE during antibiotic treatment, which is comparable with literature ranging from 30 to 60% [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ,
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Shamszad P.
      • Khan M.S.
      • Rossano J.W.
      • Fraser C.D.
      Early surgical therapy of infective endocarditis in children: a 15-year experience.
      ,
      • Esposito S.
      • et al.
      Infective endocarditis in children in Italy from 2000 to 2015.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ,
      • Johnson J.A.
      • Boyce T.G.
      • Cetta F.
      • Steckelberg J.M.
      • Johnson J.N.
      Infective endocarditis in the pediatric patient: a 60-year single-institution review.
      ,
      • Lin Y.T.
      • Hsieh K.S.
      • Chen Y.S.
      • Huang I.F.
      • Cheng M.F.
      Infective endocarditis in children without underlying heart disease.
      ]. In 47% of our patients requiring surgical therapy (15/32), Staphylococcus were found as causative organisms. Other studies already found Staphylococcus infections more frequent in patients who require surgical treatment [
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ,
      • Khoo B.
      • et al.
      Outcomes of surgery for infective endocarditis in children: a 30-year experience.
      ]. Additionally, these patients seem to need earlier surgical treatment [
      • Shamszad P.
      • Khan M.S.
      • Rossano J.W.
      • Fraser C.D.
      Early surgical therapy of infective endocarditis in children: a 15-year experience.
      ,
      • Khoo B.
      • et al.
      Outcomes of surgery for infective endocarditis in children: a 30-year experience.
      ]. In our study population, 75% of all surgically treated patients required early surgery during antibiotic treatment. This proportion is difficult to compare with the literature, as many surgical publications defined early surgery as surgical intervention within 3 to 7 days of diagnosis [
      • Shamszad P.
      • Khan M.S.
      • Rossano J.W.
      • Fraser C.D.
      Early surgical therapy of infective endocarditis in children: a 15-year experience.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ,
      • Lee J.H.
      • et al.
      Surgical outcomes of infective endocarditis in children: should we delay surgery for infective endocarditis?.
      ].
      For the adult population with IE, CRP levels at admission were found as predictive risk factor for in-hospital mortality, severe complications such as embolism or congestive heart failure, and need for early surgical intervention [
      • Nunes M.C.P.
      • et al.
      Outcomes of infective endocarditis in the current era: early predictors of a poor prognosis.
      ,
      • Mohanan S.
      • et al.
      Baseline C-reactive protein levels and prognosis in patients with infective endocarditis: a prospective cohort study.
      ,
      • Cornelissen C.G.
      • Frechen D.A.
      • Schreiner K.
      • Marx N.
      • Krüger S.
      Inflammatory parameters and prediction of prognosis in infective endocarditis.
      ]. Although CRP levels at admission in patients with severe complications were twice as high compared to patients without severe complications, we could not determine CRP levels as a predictor of severe complications (Table 4).
      In our study population 84% had CHD (Table 1), most of them were cyanotic (32/58; 55%) and acquired postoperative IE (48/58, 83%) with a high rate of prosthetic material infections (29/48; 60%). Other studies found CHD as the major predisposing factor for IE [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Jortveit J.
      • Eskedal L.
      • Klcovansky J.
      • Døhlen G.
      • Holmstrøm H.
      Prevalence of infective endocarditis in children.
      ,
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ,
      • Esposito S.
      • et al.
      Infective endocarditis in children in Italy from 2000 to 2015.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ], with increasing numbers of CHD associated IE during the last decades [
      • Rushani D.
      • et al.
      Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors.
      ]. Especially complex and cyanotic CHD seem to be at risk [
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Rushani D.
      • et al.
      Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors.
      ,
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ,
      • Sun L.C.
      • et al.
      Risk factors for infective endocarditis in children with congenital heart diseases - a nationwide population-based case control study.
      ]. Also the incidence of postoperative IE was found to be increasing [
      • Rosenthal L.B.
      • et al.
      The changing epidemiology of pediatric endocarditis at a children’s hospital over seven decades.
      ] with higher risk for IE within the first months postoperative as well as after use of foreign material [
      • Rushani D.
      • et al.
      Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors.
      ,
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ]. The high percentage of patients with CHD could also be due to the fact that there is a higher level of suspicion for IE in patients with a predisposition. This is especially required for diagnosis in children with unspecific clinical symptoms. [
      • Vicent L.
      • Luna R.
      • Martínez-Sellés M.
      Pediatric infective endocarditis: a literature review.
      ]
      The RV-PA conduit was the most frequent localization for pediatric IE overall (35%). Pulmonary valve in pediatric IE is reported in the literature between 17 and 38%, mostly in association with postoperative IE in children with CHD [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ,
      • Esposito S.
      • et al.
      Infective endocarditis in children in Italy from 2000 to 2015.
      ,
      • Johnson J.A.
      • Boyce T.G.
      • Cetta F.
      • Steckelberg J.M.
      • Johnson J.N.
      Infective endocarditis in the pediatric patient: a 60-year single-institution review.
      ]. Overall, RV-PA infections seemed to have increased in recent years. A systematic review describes the incidence of IE after Melody valve implantation between 3 and 14% [
      • Dixon G.
      • Christov G.
      Infective endocarditis in children: an update.
      ]. Another systematic review found the incidence of IE in adult populations higher in bovine jugular vein grafts (BJV) than other types of RV-PA conduits (5.4% vs. 1.2%, p < 0.0001), but no difference was found between surgical (Contegra) or catheter-based implanted RV-PA BJV grafts (Melody). The BJV grafts may be more sensitive due to their specific tissue characteristics predisposing to bacterial adhesion, while the implantation method itself (surgery vs. catheter-based) does not alter the risk for IE, as we had pediatric cases in both [
      • Sharma A.
      • Cote A.T.
      • Hosking M.C.K.
      • Harris K.C.
      A systematic review of infective endocarditis in patients with bovine jugular vein valves compared with other valve types.
      ].
      As other foreign material associated IE, we found 4 patients after VSD respectively one patient after ASD closure, either due to infected VSD patches (n = 3) or infected Amplatzer devices (n = 2). This type of pediatric IE after VSD or ASD correction using Amplatzer devices has rarely been described in case reports [
      • Knirsch W.
      • Nadal D.
      Infective endocarditis in congenital heart disease.
      ]. Jalal et al. found incomplete endothelialization in two thirds of Amplatzer device related IE >6 months after implantation. This delayed or incomplete endothelialization increases the risk of delayed infection so that the authors discussed a prolongation of postoperative AB prophylaxis [
      • Jalal Z.
      • et al.
      Long-term complications after Transcatheter atrial septal defect closure: a review of the medical literature.
      ].
      Interestingly, neither CHD nor foreign material was found to be a risk factor for severe courses in our study. This is in contrast to the findings of various other studies, which state that CHD, especially severe or cyanotic CHD, is associated with increased mortality [
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Day M.D.
      • Gauvreau K.
      • Shulman S.
      • Newburger J.W.
      Characteristics of children hospitalized with infective endocarditis.
      ,
      • Sun L.C.
      • et al.
      Risk factors for infective endocarditis in children with congenital heart diseases - a nationwide population-based case control study.
      ]. Furthermore, prosthetic valves have already been described as a risk factor for death [
      • Khoo B.
      • et al.
      Outcomes of surgery for infective endocarditis in children: a 30-year experience.
      ].
      Comparable to some recent findings that 28 to 33% of children with IE have septic embolism [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Luca A.C.
      • et al.
      Difficulties in diagnosis and therapy of infective endocarditis in children and adolescents—cohort study.
      ], 28% of our study population suffered this severe complication. On the contrary, our rates of thromboembolic complications are higher than other reported numbers of 6% and 15% [
      • Gupta S.
      • Sakhuja A.
      • McGrath E.
      • Asmar B.
      Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ]. These lower values could be explained by the fact that in both studies only stroke was considered as a thromboembolic complication, whereas in our study thromboembolic complications beyond the central nervous system were also included. However, a long-term Canadian study conducted over a 30-year period found that 40% of children with IE had thromboembolic complications in the first decade of this century, with thromboembolic events associated with age over 3 years [
      • Thom K.
      • et al.
      Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
      ]. As the median age of the patients in our study was younger (6.4y compared to 12.3y) and also more than a quarter of the patients in our study population were younger than one year, this could partly explain the different findings. Furthermore, an American study of neurological complications in pediatric infective endocarditis found that more than one fifth of the patients with neurological complications were asymptomatic and only found incidentally on screening neuroimaging [
      • Jacobwitz M.
      • et al.
      Neurologic complications of infective endocarditis in children.
      ]. In our study patients were not systematically screened for thromboembolic complications. This could lead to an underestimation of the events.
      The rate of congestive heart failure in our study population (23%) as a complication of IE is higher than the 12 to 15% reported [
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ]. The comparison with both studies should be taken with caution, as neither study describes a definition of heart failure.
      Right-sided infections occurred in almost half of the patients (45%), which is comparable to results from Belgium [
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ]. In a multicenter registry in Spain, right-sided infections were found to be more common in children with CHD, with infections of the RV-PA being most frequent [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ]. Additionally, IE in pulmonary valve position has been found to increase in patients with complex CHD [
      • Dixon G.
      • Christov G.
      Infective endocarditis in children: an update.
      ,
      • Kelchtermans J.
      • et al.
      Clinical characteristics of infective endocarditis in children.
      ,
      • Esposito S.
      • et al.
      Infective endocarditis in children in Italy from 2000 to 2015.
      ]. On the other hand, several other studies presented smaller numbers of right-sided infections (15–33%) [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ]. In these studies either a lower percentage of CHD was found to be a predisposing risk factor [
      • Vicent L.
      • et al.
      Infective endocarditis in children and adolescents: a different profile with clinical implications.
      ,
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ] or the reported CHD were less complex and mostly involved VSD [
      • Cao G.
      • fang & Bi, Q.
      Pediatric infective endocarditis and stroke: a 13-year single-center review.
      ].
      In 16 of our patients (22%) localization was unknown, all had CHD pre-diagnosed, and of them 81% (13/16) had postoperative IE with implanted foreign material in 75% (12/16). In other studies, no localization of IE was found in 14–33% of the patients [
      • Mahony M.
      • et al.
      Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
      ,
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Jortveit J.
      • et al.
      Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
      ,
      • Lin Y.T.
      • Hsieh K.S.
      • Chen Y.S.
      • Huang I.F.
      • Cheng M.F.
      Infective endocarditis in children without underlying heart disease.
      ], most of them having inconclusive echo due to CHD with complex anatomy or the use of prosthetic material [
      • Webb R.
      • et al.
      Infective endocarditis in New Zealand children 1994-2012.
      ,
      • Lin Y.T.
      • Hsieh K.S.
      • Chen Y.S.
      • Huang I.F.
      • Cheng M.F.
      Infective endocarditis in children without underlying heart disease.
      ]. A study on IE in adults found no vegetation on echo after implantation of pulmonary valves in up to 50% of cases, even after TOE [
      • Cheung G.
      • et al.
      Infective endocarditis following percutaneous pulmonary valve replacement: diagnostic challenges and application of intra-cardiac echocardiography.
      ]. Thus, an estimated number of unknown locations can be assumed, with the RV-PA conduits probably being affected even more frequently.
      In 2015 the European Society of Cardiology (ESC) included 18F-fluorodeoxyglucose (18F-FDG) PET-CT as a diagnostic criteria for infective endocarditis as an “important supplementary method for patients with suspected IE and diagnostic difficulties” [
      • Habib G.
      • et al.
      2015 ESC guidelines for the management of infective endocarditis.
      ], for example in patients with suspected prosthetic valve IE [
      • Dixon G.
      • Christov G.
      Infective endocarditis in children: an update.
      ]. In one of our patients definite diagnosis could be made by a positive PET-CT after inconclusive echo findings.
      The main limitations of this study are the retrospective approach and low statistical power due to the sample size. Therefore, further studies with larger numbers of patients are needed to identify risk factors. Although the study was conducted nationwide with the participation of the four major tertiary cardiac centers, 69 patients could be included in the study over the last ten years. As only tertiary hospitals participated in the study, there may be a bias towards more severe courses. To avoid this bias, the smaller pediatric cardiology centers in Switzerland were contacted before the start of the study, showing that the patients with IE were all treated in collaboration with the large centers. Nevertheless, some patients might not have been included in the study, for example because they were treated by adult specialists. It would also be interesting to compare our pediatric data with data from adults in Switzerland. Unfortunately, there does not exist such comparable data yet. Also, this is the first study to look at the combination of severe complications including sepsis, congestive heart failure and embolism, which makes it difficult to compare our findings with other studies and the comparisons done should be taken with caution.

      5. Conclusion

      IE is still a severe cardiac disease in childhood with relevant mortality. It is also associated with relevant morbidity, including high rates of severe complications associated with Staphylococcus aureus infections and often involvement of surgical intervention. Congenital heart diseases represent a risk factor for IE, in particular the high number of cases associated with prosthetic pulmonary valve needs further evaluation and therapeutic alternatives.

      Declaration of Competing Interest

      The authors have no conflicts of interest to disclose.

      Acknowledgement

      The “Verein Kinderherzforschung Schweiz” (Association of Children's Heart Research Switzerland) supported the project. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Appendix A. Appendix

      Table APossible transient bacteremia.
      Possible transient bacteremian = 29 (42%)
      Dental sourcen = 8 (12%)
       Carious tooth lesionsn = 2
       Dental procedure before diagnosisn = 4 (within 2 weeks to 2 months)
       Gingivitisn = 1
       Aphthaen = 1
      Skin sourcen = 10 (14%)
       Sternotomy with wound infectionn = 1
       Necrotizing fasciitis after Varicella Zoster infectionn = 1
       Bacterial superinfection after Herpes Zoster infectionn = 1
       Atopic dermatitis with excoriated lesionsn = 1
       Laceration woundn = 1
       Abrasion woundn = 1
       Panaritiumn = 1
       Cellulitis and folliculitisn = 1
       Acnen = 1
       Bee stingn = 1
      Central venous linen = 9 (13%)
       Temporary central venous linen = 8
       Permanent central venous linen = 1
      Other entry sitesn = 2 (3%)
       Infectious sacroiliitisn = 1
       Scoliosis correction surgeryn = 1 (3 days before diagnosis)
       Necrotizing enterocolitisn = 1
      Table BInfectious diseases associated with manifested IE.
      Associated infectious diseases
      multiple findings possible.
      n = 14 (20%)
      Pneumonian = 11
      Meningitisn = 3
      Osteomyelitisn = 1
      Epiglottitisn = 1
      Cholecystitisn = 1
      Pyelonephritisn = 1
      Polyarthritisn = 1
      Peritonitis after peritoneal dialysis due to heart-failure associated renal failuren = 1
      low asterisk multiple findings possible.
      Table CCardiac and extracardiac findings and complications.
      Cardiac findings and complicationsn = 36 (52%)
      Symptoms of heart failure
      multiple findings possible.
      without impaired cardiac function.
      n = 16 (23%)
       Hemodynamic instabilityn = 6
       Pulmonary oedema due to left heart failuren = 8
       Hepatomegaly due to right heart failuren = 3
       Reduced myocardial function in echocardiogramn = 2
      Cardiac surgery associated complicationsn = 2 (3%)
       AV-Block grade II (Wenckebach) with hemodynamic instabilityn = 1
       Pericardial tamponaden = 1
      Othern = 1 (1%)
       Aneurysmatic RV-PA conduitn = 1
      Extracardiac findings and complications
      Neurologic complications
      multiple findings possible.
      n = 11 (16%)
       Embolismn = 9
       Intracranial bleeding (2× combined with cerebral embolism)n = 3
       Status epilepticusn = 1
       Grand mal seizure (complex fever seizure)n = 1
      Renal complications
      multiple findings possible.
      n = 20 (29%)
       Kidney injury / renal failuren = 11
       Glomerulonephritisn = 15
       Renal embolismn = 2
      Pulmonal complications
      multiple findings possible.
      n = 18 (26%)
       ARDSn = 3
       Embolismn = 11
       Pulmonary oedeman = 8
       Pulmonary hemorrhagesn = 1
      Gastrointestinal complications
      multiple findings possible.
      n = 14 (20%)
       Hepatomegaly due to right heart failuren = 3
       Shock livern = 1
       Splenomegalyn = 8
       Splenic embolismn = 3
      Hematologic complications
      multiple findings possible.
      n = 61 (88%)
       Anemian = 58
       Thrombocytopenia (1× with purpura fulminans)n = 40
      Allergic complicationsn = 2 (3%)
       Type I allergic reaction to Penicillinn = 1
       Potential Stevens-Johnson-Syndrome due to Vancomycinn = 1
      Endocrinological complicationsn = 2 (3%)
       Infection-triggered SIADHn = 1
       Stress induced hyperglycemian = 1
      Other complicationsn = 5 (7%)
       Rhabdomyolysisn = 1
       Vasculitisn = 1
       Conjunctival bleedingn = 1
       Retinal embolismn = 1
       Embolism into extremitiesn = 1
      Abbreviations: AV = atrio-ventricular; RV-PA = right ventricular to pulmonary artery; ARDS = acute respiratory distress syndrome; SIADH = syndrome of inadequate ADH-secretion.
      low asterisk multiple findings possible.
      low asterisklow asterisk without impaired cardiac function.

      References

        • Dixon G.
        • Christov G.
        Infective endocarditis in children: an update.
        Curr. Opin. Infect. Dis. 2017; 30: 257-267
        • Knirsch W.
        • Nadal D.
        Infective endocarditis in congenital heart disease.
        Eur. J. Pediatr. 2011; 170: 1111-1127
        • Vicent L.
        • et al.
        Infective endocarditis in children and adolescents: a different profile with clinical implications.
        Pediatr. Res. 2022; 1–7https://doi.org/10.1038/s41390-022-01959-3
        • Mahony M.
        • et al.
        Infective endocarditis in children in Queensland, Australia: epidemiology, clinical features and outcome.
        Pediatr. Infect. Dis. J. 2021; 40: 617-622
        • Jortveit J.
        • Eskedal L.
        • Klcovansky J.
        • Døhlen G.
        • Holmstrøm H.
        Prevalence of infective endocarditis in children.
        Tidsskr. Nor. Laegeforen. 2019; 1–5
        • Gupta S.
        • Sakhuja A.
        • McGrath E.
        • Asmar B.
        Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States.
        Congenit. Heart Dis. 2017; 12: 196-201
        • Webb R.
        • et al.
        Infective endocarditis in New Zealand children 1994-2012.
        Pediatr. Infect. Dis. J. 2014; 33: 437-442
        • Sakai Bizmark R.
        • Chang R.K.R.
        • Tsugawa Y.
        • Zangwill K.M.
        • Kawachi I.
        Impact of AHA’s 2007 guideline change on incidence of infective endocarditis in infants and children.
        Am. Heart J. 2017; 189: 110-119
        • Baltimore R.S.
        • et al.
        Infective endocarditis in childhood: 2015 update - a scientific statement from the American Heart Association.
        Circulation. 2015; 132: 1487-1515
        • Rushani D.
        • et al.
        Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors.
        Circulation. 2013; 128: 1412-1419
        • Rosenthal L.B.
        • et al.
        The changing epidemiology of pediatric endocarditis at a children’s hospital over seven decades.
        Pediatr. Cardiol. 2010; 31: 813-820
        • Kelchtermans J.
        • et al.
        Clinical characteristics of infective endocarditis in children.
        Pediatr. Infect. Dis. J. 2019; 38: 453-458
        • Jortveit J.
        • et al.
        Endocarditis in children and adolescents with congenital heart defects: a Norwegian nationwide register-based cohort study.
        Arch. Dis. Child. 2018; 103: 670-674
        • Cao G.
        • fang & Bi, Q.
        Pediatric infective endocarditis and stroke: a 13-year single-center review.
        Pediatr. Neurol. 2019; 90: 56-60
        • Day M.D.
        • Gauvreau K.
        • Shulman S.
        • Newburger J.W.
        Characteristics of children hospitalized with infective endocarditis.
        Circulation. 2009; 119: 865-870
        • Weber R.
        • et al.
        Interventions using foreign material to treat congenital heart disease in children increase the risk for infective endocarditis.
        Pediatr. Infect. Dis. J. 2008; 27: 544-550
        • Knirsch W.
        • Schuler S.K.
        • Martin C.
        • Roland W.
        Time-trend population analysis of the clinical and epidemiologic effect on pediatric infective endocarditis after change of antibiotic prophylaxis guidelines.
        Infection. 2020; https://doi.org/10.1007/s15010-020-01433-4
        • Li J.S.
        • et al.
        Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis.
        Clin. Infect. Dis. 2000; 30: 633-638
        • Wilson W.
        • et al.
        Prevention of infective endocarditis: guidelines from the American Heart Association.
        Circulation. 2007; 116: 1736-1754
        • Flückiger U.
        • Jaussi A.
        Revidierte schweizerische Richtlinien für die Endokarditis-Prophylaxe.
        Kardiovaskuläre Medizin. 2008; 11: 392-400
        • Habib G.
        • et al.
        2015 ESC guidelines for the management of infective endocarditis.
        Eur. Heart J. 2015; 36
        • Knirsch W.
        • Mackenzie C.R.
        • Schäfers H.-J.
        • Baumgartner H.
        • Kramer H.-H.
        Infektiöse Endokarditis und Enokarditisprophylaxe - Leitlinien der deutschen Gesellschaft für Pädiatrische Kardiologie.
        2014
        • Shamszad P.
        • Khan M.S.
        • Rossano J.W.
        • Fraser C.D.
        Early surgical therapy of infective endocarditis in children: a 15-year experience.
        J. Thorac. Cardiovasc. Surg. 2013; 146: 506-511
        • Jomaa W.
        • et al.
        Clinical features and prognosis of infective endocarditis in children: insights from a Tunisian multicentre registry aspects.
        Arch. Cardiovasc. Dis. 2017; 110: 676-681
        • Esposito S.
        • et al.
        Infective endocarditis in children in Italy from 2000 to 2015.
        Expert Rev. Anti-Infect. Ther. 2016; 14: 353-358
        • Marom D.
        • Ashkenazi S.
        • Samra Z.
        • Birk E.
        Infective endocarditis in previously healthy children with structurally normal hearts.
        Pediatr. Cardiol. 2013; 34: 1415-1421
        • Elder R.W.
        • Baltimore R.S.
        The changing epidemiology of pediatric endocarditis.
        Infect. Dis. Clin. N. Am. 2015; 29: 513-524
        • Thom K.
        • et al.
        Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years.
        Int. J. Cardiol. 2018; 252: 74-79
        • Sun L.C.
        • et al.
        Risk factors for infective endocarditis in children with congenital heart diseases - a nationwide population-based case control study.
        Int. J. Cardiol. 2017; 248: 126-130
        • Johnson J.A.
        • Boyce T.G.
        • Cetta F.
        • Steckelberg J.M.
        • Johnson J.N.
        Infective endocarditis in the pediatric patient: a 60-year single-institution review.
        Mayo Clin. Proc. 2012; 87: 629-635
        • Khoo B.
        • et al.
        Outcomes of surgery for infective endocarditis in children: a 30-year experience.
        J. Thorac. Cardiovasc. Surg. 2019; 158: 1399-1409
        • Lalani T.
        • et al.
        Prosthetic valve endocarditis due to coagulase-negative staphylococci: findings from the international collaboration on endocarditis merged database.
        Eur. J. Clin. Microbiol. Infect. Dis. 2006; 25: 365-368
        • Lin Y.T.
        • Hsieh K.S.
        • Chen Y.S.
        • Huang I.F.
        • Cheng M.F.
        Infective endocarditis in children without underlying heart disease.
        J. Microbiol. Immunol. Infect. 2013; 46: 121-128
        • Lee J.H.
        • et al.
        Surgical outcomes of infective endocarditis in children: should we delay surgery for infective endocarditis?.
        Eur. J. Cardiothorac. Surg. 2021; 60: 920-927
        • Nunes M.C.P.
        • et al.
        Outcomes of infective endocarditis in the current era: early predictors of a poor prognosis.
        Int. J. Infect. Dis. 2018; 68: 102-107
        • Mohanan S.
        • et al.
        Baseline C-reactive protein levels and prognosis in patients with infective endocarditis: a prospective cohort study.
        Indian Heart J. 2018; 70: S43-S49
        • Cornelissen C.G.
        • Frechen D.A.
        • Schreiner K.
        • Marx N.
        • Krüger S.
        Inflammatory parameters and prediction of prognosis in infective endocarditis.
        BMC Infect. Dis. 2013; 13
        • Vicent L.
        • Luna R.
        • Martínez-Sellés M.
        Pediatric infective endocarditis: a literature review.
        J. Clin. Med. 2022; 11
        • Sharma A.
        • Cote A.T.
        • Hosking M.C.K.
        • Harris K.C.
        A systematic review of infective endocarditis in patients with bovine jugular vein valves compared with other valve types.
        JACC Cardiovasc. Interv. 2017; 10: 1449-1458
        • Jalal Z.
        • et al.
        Long-term complications after Transcatheter atrial septal defect closure: a review of the medical literature.
        Can. J. Cardiol. 2016; 32: 1315.e11
        • Luca A.C.
        • et al.
        Difficulties in diagnosis and therapy of infective endocarditis in children and adolescents—cohort study.
        Healthc. 2021; 9
        • Jacobwitz M.
        • et al.
        Neurologic complications of infective endocarditis in children.
        Cardiol. Young. 2022; 1–10https://doi.org/10.1017/s1047951122001159
        • Cheung G.
        • et al.
        Infective endocarditis following percutaneous pulmonary valve replacement: diagnostic challenges and application of intra-cardiac echocardiography.
        Int. J. Cardiol. 2013; 169: 425-429