Invasive Group A Streptococcal Infections in Children. A Nationwide Survey in Finland

Transcription

1 Original Studies Invasive Group A Streptococcal Infections in Children A Nationwide Survey in Finland Terhi Tapiainen, MD, PhD,* Saana Launonen, MD,* Marjo Renko, MD, PhD,* Harri Saxen, MD, PhD, Eeva Salo, MD, PhD, Matti Korppi, MD, PhD, Leena Kainulainen, MD, PhD, Tarja Heiskanen-Kosma, MD, PhD, Laura Lindholm, MSc, Jaana Vuopio, MD, PhD, ** Tiina Huotari, MSc, Jarmo Rusanen, PhD, and Matti Uhari, MD, PhD* Background: The incidence of invasive group A streptococcus (igas) infections varies in time and geographically for unknown reasons. We performed a nationwide survey to assess the population-based incidence rates and outcomes of children with igas infections. Methods: We collected data on patients from hospital discharge registries and the electronic databases of microbiological laboratories in Finland for the period We then recorded the emm types or serotypes of the strains. The study physician visited all university clinics and collected the clinical data using the same data entry sheet. Results: We identified 151 children with igas infection. Varicella preceded igas infection in 20% of cases and fasciitis infection in 83% of cases. The annual incidence rate of igas infection was 0.93 per 100,000 in , 1.80 in and 2.50 in The proportion of emm 1.0 or T1M1 strains peaked in and again in , to 44% and 37% of all typed isolates. The main clinical diagnoses of the patients were severe soft-tissue infection (46%), sepsis (28%), empyema (10%), osteoarticular infection (9%) and primary peritonitis (5%). Severe pain was the most typical symptom for soft-tissue infections. More than half of the patients underwent surgery and received clindamycin. The readmission rate was 7%, and the case fatality rate was 2%. Conclusions: The incidence rate of pediatric igas infections tripled during our study. The increase was not, however, the result of a change in the strain types causing igas. Varicella immunization would likely have prevented a significant number of the cases. Key Words: Streptococcus pyogenes, epidemiology, microbiology, fasciitis, streptococcal M protein, chickenpox, chickenpox vaccine (Pediatr Infect Dis J 2016;35: ) Accepted for publication September 2, From the *Department of Pediatrics and Adolescents, Medical Research Center and PEDEGO Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland; Children s Hospital, University of Helsinki and Helsinki University Hospital, Helsinki, Finland; Center for Child Health Research, Tampere University and Tampere University Hospital, Tampere, Finland; Department of Pediatrics, Turku University Hospital, Turku, Finland; Department of Pediatrics, Kuopio University Hospital, Kuopio, Finland; Bacterial Infections Unit, National Institute of Health and Welfare, Turku, Finland; **Department of Medical Microbiology and Immunology, University of Turku, Turku, Finland; and Department of Geography, University of Oulu, Oulu, Finland. The authors have no funding or conflicts of interest to disclose. Address for correspondence: Terhi Tapiainen, MD, PhD, Department of Pediatrics and Adolescents, Medical Research Center and PEDEGO Research Center, University of Oulu and Oulu University Hospital, P.O. Box 5000, Oulu FIN-90014, Finland. terhi.tapiainen@oulu.fi. Copyright 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: /16/ DOI: /INF Group A streptococcus (GAS, Streptococcus pyogenes) typically causes impetigo and tonsillitis in children. Severe invasive infections, such as pneumonia, empyema, endocarditis, osteomyelitis, fasciitis, necrotizing fasciitis, sepsis and streptococcal toxic shock syndrome, are less common in children but cause significant morbidity and mortality. 1 5 The incidence of invasive GAS (igas) infections varies in time and geographically. 1,4 The reasons for changes in their occurrence are not fully understood. Our clinical experience indicates that the incidence of igas infections in children has been increasing in Finland since the 1990s. In particular, severe igas soft-tissue infections such as fasciitis seemed to be more common in northern Finland. Thus, we designed a nationwide survey to assess the population-based incidence rates of severe GAS infections in children over the past 15 years. To clarify more closely the epidemiology of GAS bacteria, we typed the igas strains in cases where the bacteria were available. We aimed to analyze possible differences in treatment practices to determine how important it is to combine surgical treatment with antimicrobials in the treatment of severe soft-tissue infections. PATIENTS AND METHODS We identified possible pediatric patients treated for igas infection in the 5 university pediatric hospitals in Finland (Helsinki, Kuopio, Oulu, Tampere and Turku) from hospital discharge registries (January 1, 1996 January 31, 2010) and electronic databases of the microbiological laboratories of each hospital (January 1, 2000 December 31, 2010). The following ICD-10 codes served to identify possible cases for further medical record review: B95.0, streptococcus, group A, as a cause of diseases classified in other chapters; J15.4, pneumonia because of other streptococci than Streptococcus pneumoniae or group B streptococci; A40.0, sepsis because of streptococcus, group A; A40.9, streptococcal sepsis, unspecified; L03, cellulitis; J86, pyothorax; M63, disorders in muscle diseases classified elsewhere; M002, other streptococcal arthritis and polyarthritis and A49.1, streptococcal infection, unspecified site. We also identified possible cases from the databases of each clinical microbiological laboratory of the 5 university hospitals for the period and reviewed the medical records of all hospitalized cases with GAS isolated in the pharynx, blood, pleural fluid, spinal fluid, joint fluid or any other deep tissue samples. From the microbiology laboratory databases, we also included patients who died before hospitalization and whose cultures were taken post mortem. Typing of GAS strains in Finland is performed routinely for all blood culture isolates and occasionally for other isolates. We recorded the emm types (gene encoding for the surface M protein) and/or serotypes of the streptococcal strains from the patients medical records and from the database of the national reference laboratory (National Institute of Health and Welfare, Turku, Finland). 6,7 Inclusion criteria required the isolation of GAS from blood or a deep tissue sample (confirmed cases) or from a throat or skin swab (probable cases) during the acute illness. A separate definition for a probable case was used to control for the potential confounding because of different surgery rates, that is, to make incidence rates comparable in all hospitals The Pediatric Infectious Disease Journal Volume 35, Number 2, February

2 Tapiainen et al The Pediatric Infectious Disease Journal Volume 35, Number 2, February 2016 during the whole study period independently on the rate of obtaining deep tissue cultures. The study physician visited all pediatric university clinics and systematically collected the data using the same data entry sheet and IBM SPSS software (IBM, Armonk, NY). A study physician, together with the treating physicians in 5 university hospitals, reviewed the medical records of all possible cases and recorded any underlying conditions, antimicrobial and surgical treatment, intensive care, outcome, recurrences and sequelae. For our patient series presenting the clinical details and outcome of our patients, all children treated in pediatric university clinics during the study era were included. Pediatric university clinics in Finland treat all children younger than 16 years and occasionally patients of years of age. There were 3 patients aged years who were treated in pediatric clinics during the study era. For our population-based calculations, we used the mean annual population of children younger than 15 years during the study period in each hospital district to calculate the populationbased incidence rates (per 100,000 children per year). Mean annual population of children in each hospital district in each study year was obtained from Official Statistics of Finland (OSF), which is a comprehensive collection of statistics describing the development and state of society. The basic data of the OSF are available to all users free of charge. The cut-off age of 15 years was chosen as it is used in OSF databases to report the number of children. Thus children older than 15 years and/or referred to the university hospital from other hospital districts (11 cases with any igas infection and 3 cases with positive blood culture infection) were excluded from the population-based analyses. To compare clinical manifestations and incidence rates in different geographical areas within Finland, the data are also presented separately for each hospital. Oulu University Hospital is located 610 km north of Helsinki and Kuopio University Hospital, 390 km northeast of Helsinki; Turku, Tampere and Helsinki University Hospitals are located within 180 km of each other in the southern part of Finland (Fig. 1). The driving times (minutes) from patients homes to the nearest university hospital were calculated with a geoinformatics software program (ArcGIS ESRI Inc., Redlands, CA) by using Digiroad database available for this purpose in the Department of Geography, University of Oulu, Finland. The research plan was accepted in all pediatric university clinics before the study, and the data registry announcement was sent to the national Office of the Data Protection Ombudsman. In accordance with regulations in Finland, no ethical committee review or informed consent is required for registry-based studies conducted by the physicians of institutions that have treated the patients in the study. All medical data reviewed in this study served scientific purposes only and remain confidential. RESULTS We identified a total of 151 cases of confirmed or probable igas infections in children younger than 18 years between 1996 and 2010 (Table 1). The diagnosis of confirmed cases (n = 123) was based on a positive blood culture (n = 51), a positive deep tissue culture (n = 60) or both (n = 12; Table 2). The diagnosis of probable TABLE 1. Characteristics of the Children With Invasive GAS Infection (N = 151) FIGURE 1. The home address of each child with igas infection in is marked with a black spot in 5 university hospital districts (Oulu, Kuopio, Tampere, Turku and Helsinki) in the map of Finland. Altogether 14 children were transferred to university clinics from other hospital districts; these children were excluded from the populationbased incidence analysis but are shown on the map. The annual incidence of pediatric igas infections was highest (3.01/100,000) in the Oulu area. Age, mean (SD) 7.0 yr (4.4) Gender, boys, 86 (57) Underlying conditions, Atopic eczema, asthma or allergy 19 (13) Mental retardation 6 (4) Diabetes mellitus type I 1 (0.6) Renal insufficiency 1 (0.6) Chronic lung disease 1 (0.6) Preceding events, Varicella within 1 mo 30 (20) Family member with streptococcal pharyngitis 22 (15) Surgery within 1 mo 6 (4) Minor skin trauma 74 (49) Distance from home to the nearest university hospital, mean (SD) Kilometers 41.9 (53.4) Driving time (min) 30.9 (34.7) No. of children >1-h drive, 23 (15) Symptoms, Fever 146 (90) Pain 105 (65) Erythema 66 (41) Shortness of breath 17 (11) CRP, mean (SD) At entry 128 (99) Maximum 180 (104) CRP indicates C-reactive protein; SD, standard deviation Wolters Kluwer Health, Inc. All rights reserved.

3 The Pediatric Infectious Disease Journal Volume 35, Number 2, February 2016 Invasive Group A Streptococci TABLE 2. Positive Microbiological Findings and the Mean Duration (Days) of Any Antimicrobial Treatment for GAS Among Children With Invasive GAS Infection (n = 151) According to the Treating Physician s Main Clinical Diagnosis Cellulitis* (n = 57), Sepsis (n = 46), Empyema (n = 15), Fasciitis (n = 13), Arthritis (n = 10), Peritonitis (n = 7), Osteomyelitis (n = 3), Blood culture 9 (16) 42 (100) 5 (33) 0 (0) 5 (50) 0 (0) 2 (67) Deep tissue culture 34 (60) 1 (2) 7 (47) 12 (92) 9 (90) 7 (100) 2 (67) Throat or skin swab 17 (30) 5 (11) 4 (7) 1 (8) 0 (0) 0 (0) 0 (0) Antimicrobials (day, SD) 15.0 (8.0) 13.4 (8.5) 22.6 (8.5) 16.4 (4.1) 21.4 (9.6) 10.0 (11.5) 38 (2.5) *Includes both patients with a positive deep tissue culture or blood culture (n = 40) and patients with a positive skin or throat swab (n = 17). Includes 4 patients with sepsis-like illness but a negative blood culture. One child can be included in more than 1 group. SD indicates standard deviation. FIGURE 2. Age distribution of children with invasive GAS infection treated in pediatric university clinics in Finland in The patients aged years are only occasionally treated in pediatric clinics (*). FIGURE 3. The annual population-based incidence rates of invasive GAS infections in children (0 14 years old) in Finland in cases (n = 28) was based on clinical findings of invasive disease and a positive culture obtained from a throat swab (n = 19), the skin (n = 8) or both (n = 1; Table 2). The mean age of the patients was 7.0 years (standard deviation, 4.4; range, 2 days to 17.8 years; Table 1; Fig. 2), and 57% (86 cases) were boys. Most children were previously healthy before the igas infection and lived within a 1-hour drive from the university hospital (Table 1). Altogether 30 (20%) children had had a recent varicella infection before (<1 month) the invasive GAS infection (Table 1) and in children with fasciitis, 83% of families (10/12) reported a preceding chickenpox infection. In total, 22 (15%) children had had a family member with recent streptococcal pharyngitis. Six (4%) cases had undergone surgery before invasive GAS, and many (49%) families reported that a child had suffered a minor skin trauma, such as a scratch or cut before igas, according to medical records. Almost all the children had a fever before arriving at the hospital. The mean duration of fever was 2.8 days (standard deviation, 2.7) before admittance. Records also contain frequent reports of local pain and skin erythema (Table 1). The main clinical diagnoses of the patients were cellulitis (n = 57), sepsis (n = 42), empyema (n = 15), fasciitis (n = 13), arthritis (n = 10), peritonitis (n = 7), osteomyelitis (n = 3) and a sepsis-like illness without a positive blood culture (n = 4; Table 2). Of the 57 cases with cellulitis, 40 involved a positive deep tissue culture, and 17 cases, a positive skin or throat swab. Severe pain was the most typical symptom of severe igas soft-tissue infection (Table 1). The population-based incidence rate of igas infections was 1.60 per 100,000 per year, and the population-based incidence rate of blood culture-positive igas infections was 0.67/100,000 per year during the study period. The annual incidence rate of igas infections among children younger than 15 years was 0.93 per 100,000 [95% confidence interval (CI): ] in , 1.76 (95% CI: ) in and 2.50 (95% CI: ) in (Fig. 3). If only the confirmed cases were included, the incidence rates were 0.71, 1.53 and 1.87, respectively (Fig. 3). The annual incidence rate of blood culture-positive igas infections almost tripled from 0.37 per 100,000 (95% CI: ) in to 1.16 (95% CI: ) in (Fig. 3). The incidence of igas infection was highest in December to February. Serotypes were available from 8 isolates before the emm typing era, and emm types were available from 52 isolates (Fig. 4). The proportion of emm 1.0 or T1M1 strains among the typed isolates was 44% (4/9) in , 15% (2/13) in and 37% (14/38) in The incidence rates in the 2 northernmost hospital districts (Oulu and Kuopio) were the highest (Table 3; Fig. 1). In Oulu University Hospital, the majority of children had severe soft-tissue 2015 Wolters Kluwer Health, Inc. All rights reserved

4 Tapiainen et al The Pediatric Infectious Disease Journal Volume 35, Number 2, February 2016 FIGURE 4. Bacterial typing data of 60 pediatric invasive GAS isolates in in Finland. In total, 52 strains were emm typed, and 8 isolates were serotyped before the emm typing. infection (fasciitis or cellulitis; 24/39, 62%) or empyema (10/39, 26%), whereas in Tampere University Hospital, most children had sepsis (14/22, 64%). Emm 1.0 strains were present in all university hospitals and in all clinical manifestations. Emm 28.0 strains caused mainly sepsis (7/9) and were not detected in Oulu. A total of 29 (19%) children received treatment in an intensive care unit, 17 (11%) patients required ventilator treatment, and 14 (9%) children had hypotension treated with ionotropic drugs. One child received treatment with extracorporeal membrane oxygenation, and 7 children received intravenous (i.v.) immunoglobulin. All children admitted to the hospital (n = 149) were treated with antimicrobials. The most commonly used antimicrobial treatment (after diagnosis of igas) included i.v. β-lactam antibiotics (usually penicillin or cefuroxime) combined with i.v. clindamycin (n = 72; 48%). All patients with fasciitis (n = 13) and 13 (87%) of 15 empyema patients received clindamycin. Patients who received clindamycin had a longer mean length of stay than did those treated without clindamycin (11.2 vs. 5.8 days; 95% CI of the difference: 2.9 to 7.8 days, P < ). Altogether 58 (39%) patients received β-lactam antibiotics, and 4 children received clindamycin as monotherapies; 15 children received other combinations. The mean length of any i.v. antibiotic treatment was 8.4 days (range 1 44), and the mean length of any antibiotic treatment (including both i.v. and oral treatment) was 15.6 days (range, 1 48; Table 2). Of 151 patients, 83 (55%) underwent at least 1 surgical procedure (Table 4). Surgical treatment was most common in children with fasciitis, 92% (n = 12/13) of whom underwent fasciotomy. All children with empyema (n = 15) had either a chest tube or underwent thoracotomy. The most typical clinical picture of igas softtissue infection was unusually severe pain in one or more anatomic regions of the body after a varicella infection that stopped promptly after the incision or fasciotomy. Three children underwent appendectomy because of unusual severe pain from suspected appendicitis. Most children found severe soft-tissue infection to be very painful and had a toxic appearance before surgery, but the necrotizing tissue was rarely observed in surgery. Ten children were readmitted after the first treatment period (n = 10, 7 %), approximately 1 week after the discharge. The diagnoses of the readmitted patients were cellulitis (n = 5), arthritis (n = 2), empyema (n = 1) and sepsis (n = 2). Half of the readmitted patients had received β-lactams without clindamycin (n = 5/10), and rest of them (n = 5/10) received clindamycin alone or combined with β-lactams during their first hospitalization (range, 1 21 days). Five of the readmitted patients had undergone surgical procedure before the discharge and 2 patients after they were readmitted. At least 1 follow-up visit after the acute treatment was scheduled for 83 children of 148 (56%) discharged patients. One child had restricted joint movements in 1 finger; 1 child had still pleural thickening on chest radiography after empyema but was clinically asymptomatic; 2 children had suspected reactive arthritis on several joints, and one child had Clostridium difficile diarrhea treated with oral metronidazole. Of the 151 children, 3 (2.0%) died of igas infection. One patient admitted to the hospital died there. This 2-year-old girl was previously healthy but deteriorated during the hospital treatment and died of septic shock (caused by serotype T1, emm 1.0) on day 3 after receiving i.v. penicillin as a monotherapy for the treatment of pneumonia, empyema and sepsis. The other 2 patients died before reaching the hospital: one of them, an 18-month-old boy, was found dead in bed at home after 1 day of febrile illness; S. pyogenes was TABLE 3. Incidence and Clinical Manifestations of igas Infections in 5 Pediatric University Clinics in Finland University Hospital Oulu Tampere Helsinki Kuopio Turku Incidence rate* per 100,000 children/year All (N) 3.01 (37) 1.75 (21) 0.92 (36) 2.96 (17) 2.53 (28) Confirmed invasive (N) 2.44 (30) 1.66 (20) 0.71 (28) 2.96 (17) 1.63 (18) Blood culture positive (N) 0.73 (9) 1.33 (16) 0.38 (15) 1.05 (6) 0.99 (11) Diagnosis, Sepsis 4 (10) 13 (59) 11 (28) 7 (35) 7 (23) Empyema 10 (26) 0 (0) 3 (8) 1 (5) 1 (3) Fasciitis 11 (28) 0 (0) 1 (3) 0 (0) 1 (3) Arthritis 1 (3) 1 (5) 3 (8) 4 (20) 1 (3) Peritonitis 0 (0) 2 (9) 2 (5) 2 (10) 1 (3) Osteomyelitis 0 (0) 2 (9) 0 (0) 0 (0) 1 (3) Sepsis-like illness 0 (0) 1 (5) 1 (3) 0 (0) 2 (7) Cellulitis 13 (33) 3 (14) 19 (48) 6 (30) 16 (53) Total 39 (100) 22 (100) 40 (100) 20 (100) 30 (100) *Incidence rate was calculated among children younger than 15 years (139 igas cases). We used the mean population (children <15 years) during the study period per year in 5 hospital districts in the calculations (Oulu, 82,022; Tampere, 80,117; Helsinki, 261,503; Kuopio, 38,247; Turku, 73,705). Blood or deep tissue culture positive Wolters Kluwer Health, Inc. All rights reserved.

5 The Pediatric Infectious Disease Journal Volume 35, Number 2, February 2016 Invasive Group A Streptococci TABLE 4. Primary Surgical Treatment of Children With igas infection (n = 151) According to the Treating Physician s Main Clinical Diagnosis Cellulitis* (n = 57) Sepsis (n = 46) Empyema (n = 15) Fasciitis (n = 13) Arthritis (n = 10) Peritonitis (n = 7) Osteomyelitis (n = 3) Incision or revision Chest tube Fasciotomy Laparotomy Thoracotomy Joint puncture Appendectomy Total 35 (62%) 1 (2%) 15 (100%) 12 (92%) 10 (100%) 4 (57%) 3 (100%) Altogether 14 children underwent more than 1 operation (8 children had 2 operations, 5 had 3 operations and 1 child with severe soft-tissue infection had 5 revisions). *Includes patients with either a deep tissue culture (n = 40) or a positive skin or throat swab (n = 17) that tested positive for GAS. Includes 4 patients with sepsis-like illness. Because of suspected appendicitis (false diagnosis). TABLE 5. Summary of Main Findings in Clinical Studies of igas Infections in Children Including This Study Author Years Region Design Patients (N) Incidence* Fasciitis (%) Blood culture+ CFR igas Minodier et al Quebec Patient series 68 NR 27 56% 4.4% Henriet et al Paris Patient series 28 NR 25 61% 3.6% Laupland et al Ontario Prospective surveillance % 4.1% Tapiainen et al (this study) Finland Patient series % 2/0.6% GAS necrotizing fasciitis Moss et al Albuquerque Patient series 7 NR 100 NR 14% Eneli and Davies Canada Patient series NR 0% The reported incidence and CFRs are not directly comparable because of different study designs and variable case definitions. *Incidence rate per 100,000 children/year. Fasciitis and severe soft-tissue infections combined. Of the 151 children in this study, 3 (2%) died, 2 of whom died before reaching the hospital; 1 (0.6%) of 149 died after admittance. Only necrotizing fasciitis included in the study. CFR indicates case fatality ratio; NR, not reported. cultured post mortem from cerebrospinal fluid, blood and lung tissue samples. The other, a 3-year-old girl, died of septic shock in transit to the hospital after 5 days of febrile illness and S. pyogenes (emm 1.0) was cultured post mortem from blood. The driving time from home to the nearest university hospital was 11.5 minutes and 121 minutes, respectively, for these 2 children who died before admission. DISCUSSION Our nationwide survey shows that the incidence rate of pediatric igas infections tripled in Finland in In the last years of our study, the annual incidence of pediatric igas infections was similar (2.5/100,000 all cases, 1.9/100,000 confirmed cases) to that reported in Canada (1.9/100,000) in the 1990s, when igas infections in the country increased markedly (Table 5). 4,8 Earlier studies have shown that the mortality of children with igas infection varies between 0% and 14 % depending on the characteristics of the study population (Table 5). 1 5 The case fatality rate in our patients was low; of the 151 children in our study, 3 died because of igas infection. Our patients were actively treated with surgery, often received clindamycin and had access to pediatric intensive care. Clindamycin is recommended as an initial antimicrobial agent for the treatment of igas infections, as it is a protein synthesis inhibitor and probably suppresses the synthesis of S. pyogenes antiphagocytic M protein and bacterial toxins In a retrospective series of 56 children with igas infection, combining clindamycin with β-lactam antimicrobials seemed to be more effective than β-lactam treatment alone. 10 Most children in our study received clindamycin combined with β-lactams. None of our patients who received clindamycin died. Treating children with igas infections must take into account local clindamycin resistance. 13 In Finland during the study period, the proportion of S. pyogenes strains resistant to clindamycin was low (0% 2%). 14 Aggressive surgical debridement of any deep-seated GAS infection is recommended. 9 In one earlier patient series among children with suspected necrotizing fasciitis, only those who underwent active surgery within hours of their arrival at the hospital survived. 5 Accordingly, our patients underwent active surgery, with 55% undergoing at least 1 surgical procedure. In our patients undergoing soft-tissue surgery, however, clear necrotizing fasciitis was rare. GAS infections recurred in 10% of our patients even after receiving long antimicrobial treatment. Viable intracellular GAS residing inside macrophages have been present in human biopsies even after prolonged i.v. antimicrobial treatment (i.v. β-lactam and clindamycin) of GAS soft-tissue infections. 15 Researchers have previously proposed similar mechanisms to explain the frequent recurrences of streptococcal pharyngotonsillitis. 16 Thus, the relapses of igas infections in our patients may have derived from an intracellular reservoir after stopping the antimicrobials. Preceding chickenpox has reportedly increased the risk for igas infections in children from 50-fold to 60-fold within the first month after the infection. 4 The national immunization program for children in Finland has not yet included varicella immunization despite the recommendations of the experts. More than 80% of 2015 Wolters Kluwer Health, Inc. All rights reserved

6 Tapiainen et al The Pediatric Infectious Disease Journal Volume 35, Number 2, February 2016 igas fasciitis cases and 20% of all igas cases in our study were associated with recent chickenpox and could have been prevented with varicella vaccination. The findings of this study further support the use of universal varicella immunization for children. The most prevalent GAS strain namely emm1/t1m1 accounted for one third of the cases in our study. Streptococcal M protein is located on the surface of S. pyogenes and, together with pyrogenic exotoxins, is an important virulence factor and capable of inducing septic shock. 17 The immune response to GAS develops against surface structures, which are thus under selective pressure. Emm 1 strains have been associated with the increase of igas infections in many countries. 8,18 21 Based on the whole genome sequencing of more than 3600 emm 1 strains collected in 8 countries in , the acquisition of specific genetic elements in 1 GAS cell in the early 1980s might be the origin of the igas epidemic observed in many countries since the late 1980s. 22 In our study, the relative proportion of emm 1 strains showed a typical fluctuation. 7 Emm 28.0 strains were not found in Oulu where the occurrence of igas infections was the highest. In our study, the northernmost university hospital district, Oulu (64 70 northern latitude), had the highest annual incidences of igas infection, especially severe soft-tissue infections. The strain type distribution did not clearly differ between the university hospitals and did not explain the differences observed. In adult populations, igas infections are most common in northern European countries, where the annual incidence rate of igas infections has reached 3 per 100, In North America, the high incidence rates have been observed in Canada. 3,4,8 The reasons for the higher risk for igas infections in northern countries are not yet understood. Strengths of this study are its systematic collection of nationwide clinical data on pediatric igas infections over a 15-year period. This patient series of igas infections in children is one of the largest available (Table 5), and we present the temporal changes in the population-based incidences of igas together with the bacterial typing data. The main limitation of the study is that the efficacy of different treatments cannot be analyzed in this study, as our study design was observational. Yet, the active surgery and active use of clindamycin may explain the low case fatality rate in our patients. In addition, the electronic microbiological databases were not available before year 2000, which may partly explain the increasing number of found cases since In conclusion, the incidence of pediatric igas infections has tripled in Finland over the past 15 years, a rise that cannot be explained only by the rise of emm 1.0 (M1T1) strains in the community. Varicella immunization would have prevented a large number of the pediatric igas cases observed. REFERENCES 1. Minodier P, Bidet P, Rallu F, et al. Clinical and microbiologic characteristics of group A streptococcal necrotizing fasciitis in children. Pediatr Infect Dis J. 2009;28: Henriet S, Kaguelidou F, Bidet P, et al. Invasive group A streptococcal infection in children: clinical manifestations and molecular characterization in a French pediatric tertiary care center. Eur J Clin Microbiol Infect Dis. 2010;29: Eneli I, Davies HD. Epidemiology and outcome of necrotizing fasciitis in children: an active surveillance study of the Canadian Paediatric Surveillance Program. J Pediatr. 2007;151:79 84, 84.e1. 4. Laupland KB, Davies HD, Low DE, et al. Invasive group A streptococcal disease in children and association with varicella-zoster virus infection. Ontario Group A Streptococcal Study Group. Pediatrics. 2000;105:E Moss RL, Musemeche CA, Kosloske AM. Necrotizing fasciitis in children: prompt recognition and aggressive therapy improve survival. J Pediatr Surg. 1996;31: Siljander T, Toropainen M, Muotiala A, et al. emm typing of invasive T28 group A streptococci, , Finland. J Med Microbiol. 2006;55(Pt 12): Siljander T, Lyytikäinen O, Vähäkuopus S, et al. Epidemiology, outcome and emm types of invasive group A streptococcal infections in Finland. Eur J Clin Microbiol Infect Dis. 2010;29: Davies HD, McGeer A, Schwartz B, et al. Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N Engl J Med. 1996;335: Committee on Infectious Diseases American Academy of Pediatrics. Red Book: 2012 Report of the Committee on Infectious Diseases. 29th ed. IL: Elk Grove Village; Zimbelman J, Palmer A, Todd J. Improved outcome of clindamycin compared with beta-lactam antibiotic treatment for invasive Streptococcus pyogenes infection. Pediatr Infect Dis J. 1999;18: Stevens DL, Ma Y, Salmi DB, et al. Impact of antibiotics on expression of virulence-associated exotoxin genes in methicillin-sensitive and methicillin-resistant Staphylococcus aureus. J Infect Dis. 2007;195: Stevens DL, Gibbons AE, Bergstrom R, et al. The eagle effect revisited: efficacy of clindamycin, erythromycin, and penicillin in the treatment of streptococcal myositis. J Infect Dis. 1988;158: Chen M, Yao W, Wang X, et al. Outbreak of scarlet fever associated with emm12 type group A Streptococcus in 2011 in Shanghai, China. Pediatr Infect Dis J. 2012;31:e158 e Gunell M, Hakanen A, Aittoniemi J, et al. Mikrobilääkeresistenssi Suomessa Finres (Finnish). Finland: National Institute of Health and Welfare; Thulin P, Johansson L, Low DE, et al. Viable group A streptococci in macrophages during acute soft tissue infection. PLoS Med. 2006;3:e Osterlund A, Popa R, Nikkilä T, et al. Intracellular reservoir of Streptococcus pyogenes in vivo: a possible explanation for recurrent pharyngotonsillitis. Laryngoscope. 1997;107: Bisno AL, Brito MO, Collins CM. Molecular basis of group A streptococcal virulence. Lancet Infect Dis. 2003;3: Bucher A, Martin PR, Høiby EA, et al. Spectrum of disease in bacteraemic patients during a Streptococcus pyogenes serotype M-1 epidemic in Norway in Eur J Clin Microbiol Infect Dis. 1992;11: Carapetis J, Robins-Browne R, Martin D, et al. Increasing severity of invasive group A streptococcal disease in Australia: clinical and molecular epidemiological features and identification of a new virulent M-nontypeable clone. Clin Infect Dis. 1995;21: Colman G, Tanna A, Efstratiou A, et al. The serotypes of Streptococcus pyogenes present in Britain during and their association with disease. J Med Microbiol. 1993;39: O Brien KL, Beall B, Barrett NL, et al. Epidemiology of invasive group a streptococcus disease in the United States, Clin Infect Dis. 2002;35: Nasser W, Beres SB, Olsen RJ, et al. Evolutionary pathway to increased virulence and epidemic group A Streptococcus disease derived from 3,615 genome sequences. Proc Natl Acad Sci U S A. 2014;111:E1768 E Lamagni TL, Darenberg J, Luca-Harari B, et al; Strep-EURO Study Group. Epidemiology of severe Streptococcus pyogenes disease in Europe. J Clin Microbiol. 2008;46: Wolters Kluwer Health, Inc. All rights reserved.