Reemergence of emm1 and a Changed Superantigen Profile for Group A Streptococci Causing Invasive Infections: Results from a Nationwide Study
http://www.100md.com
微生物临床杂志 2005年第4期
Department of Bacteriology, Mycology and Parasitology
Biostatistics Unit, Statens Serum Institut
Department of Infectious Diseases, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
ABSTRACT
Between 1999 and 2002, 496 invasive group A streptococcal (GAS) isolates from clinical microbiological departments in Denmark and subsequently 487 (98%) questionnaires from the clinicians treating the patients were received as part of a national surveillance. emm types and streptococcal superantigen (SAg) genes were determined. The incidence of invasive GAS infections was on average 2.3 per 100,000 per year. Bacteremia with no focal symptoms (27%) was together with erysipelas (20%) the most prevalent clinical diagnoses. Streptococcal toxic shock syndrome occurred in 10% of patients, of which 56% died. The overall case fatality rate within 30 days was 23%. In total, 47 different emm types were identified, of which emm1, emm3, emm4, emm12, emm28, and emm89 were identified in 72% of the 493 available isolates. During the 4-year period the presence of emm1 increased from 16% in 1999 to 40% in 2002. Concurrently, the presence of emm3 decreased from 23% in 1999 to 2% in 2002. The emm1 isolates predominantly carried speA, although the frequency decreased from 94% in 1999 to 71% in 2002, whereas the emm1-specific prevalence of speC increased from 25 to 53%. In a historical perspective, this could be interpreted as a reemergence of emm1 and could indicate a possible introduction of a new emm1 subclone. However, this reemergence did not result in any significant changes in the clinical manifestations during the study period. Our results show the complexity of invasive GAS infections, with time-dependent variations in the incidence and distribution of emm and SAg genes, which emphasizes the need for continuous epidemiological and molecular investigations.
INTRODUCTION
In the last two decades, Streptococcus pyogenes (group A streptococci [GAS]) has been identified as an emerging cause of severe infections: septic shock, organ failure, soft-tissue infections (myositis and necrotizing fasciitis [NF]), and streptococcal toxic shock syndrome (STSS) with high mortality.
Since the Working Group on Severe Streptococcal Infections proposed diagnostic criteria for STSS in 1993 (44), several studies of the epidemiological, microbiological, and clinical aspects of invasive GAS infection have been performed in various countries. However, many questions about the pathogenesis of invasive GAS infection still remain unanswered.
Traditional methods of T-agglutination typing (T typing) and M-precipitation typing (M typing) have been used in epidemiological studies for the last 50 years (10). Recently, new molecular methods have replaced these conventional methods, where emm sequencing, detection of the genes encoding M proteins, has been introduced. This has made epidemiological surveillance more detailed and revealed potential clusters (emm types) in specific clinical manifestations (17). In addition, emm subtypes have been introduced in recent surveillance papers (29, 38). However, a universal high prevalence of certain emm types in invasive GAS diseases may also reflect widespread transmission rather than an increased virulence and invasiveness.
Streptococcal exotoxins are presumed to play an important role in severe diseases, acting as superantigens (SAgs) and thereby inducing a devastating cytokine response in susceptible hosts (35). The number of identified potential SAgs has increased in the last few years, facilitated by the information obtained from the published whole genome of GAS (3, 18, 39). Despite numerous reports of SAg distributions, no previous study has, to our knowledge, made use of nationwide, longitudinal data from a population-based surveillance.
In the present study, epidemiological and disease-related data are reported in addition to the emm and SAg gene profiles, i.e., genes encoding pyrogenic exotoxins A to C, F to J, SSA, and SMEZ (speA to -C, speF to -J, ssa, and smeZ genes), to evaluate the differences in the clinical manifestations of GAS infections from the national surveillance of invasive GAS infections in Denmark from 1999 to 2002.
MATERIALS AND METHODS
Subjects and specimens. The Streptococcus Unit serves as the National Streptococcus Reference Centre and receives GAS isolates from normally sterile sites in patients admitted to all hospitals in Denmark (population, 5.34 million). The GAS isolates are received as pure cultures from all of the 15 Danish clinical microbiological departments as part of the national surveillance. From two-thirds of the clinical microbiological departments information was received, which enabled us to estimate that the Streptococcus Unit received on average 79% of the GAS blood isolates identified by the clinical microbiological departments and that this percentage remained constant during the study period (January 1999 to December 2002).
The reporting system from the clinical microbiological departments to the Streptococcus Unit has been the same since 1988, and since 1996 the Streptococcus Unit has distributed a detailed questionnaire to the clinical doctors treating the patients. In 1999 the questionnaire was redesigned to include information about the dates of admission, of discharge (or death), and of onset of primary symptoms of the infection and additionally to include a description of the type of primary symptoms, the course of the infection, treatment, and predisposing factors.
In the present study the following definitions were used. Bacteremia was defined as a clinical entity associated with identification of GAS in the blood culture without specific focus on the infection. NF was defined as diagnosis by the clinicians of necrosis of the fascia and of tissue (excluding muscle). A soft-tissue infection was defined as either NF or myositis. A patient with septic shock was defined as a patient with invasive GAS infection and a systolic blood pressure below 90 mm Hg, and finally the definition of STSS was based on the consensus definition from the Working Group on Severe Streptococcal Infections (44). An overall case fatality rate was assessed at day 30 after the culture was obtained (30-day CFR).
Date of death or a confirmation that the person was alive by 1 March 2004 was obtained from the Central Office of Civil Registration. The annual number of inhabitants in the period 1988 to 2002 was reported from Statistics Denmark. Incidence rates are presented as incidence per 100,000 inhabitants per year.
DNA preparation and emm sequencing. A few colonies from a fresh overnight culture on 5% blood agar were mixed with Chelex 100 (Bio-Rad, Richmond, Calif.) in TE buffer (21). A 0.5-μl portion of the supernatant containing the genomic DNA was used for emm sequencing and identification of exotoxin genes.
The emm gene types of GAS were determined by amplification and sequencing of the emm genes essentially as described elsewhere (2), although the PCR amplification was performed in a final volume of 25 μl containing 2.5 μl of 10x PCR buffer (200 mM Tris-HCl [pH 8.4], 500 mM KCl [Invitrogen, Carlsbad, Calif.]), 0.2 mM deoxynucleoside triphosphate mixture, 1.5 mM MgCl2 (final concentration), and a 0.4 μM concentration of each primer. Platinum Taq DNA polymerase (0.5 U; Invitrogen) was used in each reaction mixture. The following cycle parameters were used: 95°C for 15 min; 35 cycles of 94°C for 30 s, 48°C for 30 s, and 72°C for 1.5 min; and a final extension step at 72°C for 10 min. Sequencing reactions were performed by using the BigDye terminator cycle-sequencing ready-reaction kit version 2.0 (Applied Biosystems, Foster City, Calif.) according to the manufacturer's instructions, using emmseq2 as the sequencing primer. The emm sequences were obtained according to the recommendations (16) and aligned with sequences available in the Centers for Disease Control and Prevention emm-type database (http://www.cdc.gov/ncidod/biotech/strep/strepindex.html).
Identification of SAg genes (speA to -C, speF to -J, ssa, and smeZ). The primers used for identification of the exotoxin genes were found in the literature (5, 25, 30, 33, 34, 41). PCR was done as described above. Detection of speF to- J, ssa, and smeZ was performed using the following protocol: 2 min at 95°C and a total of 40 cycles of 94°C for 1 min, 50°C for 30 s, and 72°C for 1.5 min. Detection of speA to -C was carried out with the same protocol except that annealing was at 58°C for 35 cycles (25). The presence of the exotoxin gene was visually ascertained on an ethidium-stained agarose gel. Amplification of each of the genes located on mobile DNA (i.e., speA, -C, -H, and -I and ssa) was carried out in combination with amplification of one chromosomal gene (i.e., speB, -F, -G, or -J, smeZ, or the 16S rRNA gene) functioning as a positive control.
Statistical analyses. Data from the survey were analyzed with SAS System Release 8.02 (SAS Institute Inc. Cary, N.C.) and GraphPad Prism version 3.00 for Windows (GraphPad Software, San Diego, Calif.). A chi-square test of independence (in a two-by-four table) was conducted to analyze proportions of clinical manifestations (including all emm types and analysis of emm1 isolates exclusively) versus time (for the four years 1999, 2000, 2001, and 2002). The same type of analysis was used for comparisons of proportions of 30-day CFRs versus time and comparisons of proportions of SAg genes for emm1 and emm28 isolates, respectively, versus time. In the analysis of distributions of emm types (grouped emm1, emm3, emm4, emm12, emm28, emm89, and others) versus time, an overall chi-square test of independence (in a seven-by-four table) was performed initially. If an overall significant difference was observed, an analysis of the proportion of each emm type versus time was performed by a chi-square test of independence (in a two-by-four table). For each of the clinical manifestations the emm distributions in the two groups defined by the presence and absence, respectively, of the manifestation were compared in a likewise manner. Possible correlations between each of the clinical manifestations and the presence or absence of each of the SAg genes were examined by a chi-square test of independence (in a two-by-two table). Likewise, the distributions of emm1 isolates carrying the speA gene versus those lacking the speA gene and eventually of emm1 isolates carrying the speC gene versus those lacking the speC gene were examined in the absence or presence of clinical manifestations by a chi-square test (in a two-by-two table).
The proportions of predisposing factors were also compared with a chi-square test. In general, chi-square tests of independence were replaced by analogue exact tests in the case of sparse data.
For comparisons of age distributions between patient groups, Kruskal-Wallis or Mann-Whitney rank sum tests were used. Two-logistic regression analyses were performed in order to analyze which factors were significantly associated with STSS and with septic shock without fulfilling the criteria for STSS (44). The independent factors included in the analysis were sex, age (grouped as <40, 40 to 65, and >65 years), cancer, diabetes, chronic heart or lung diseases, skin lesions, alcohol abuse, immune incompetence, emm (grouped as emm1, emm3, emm28, and others), speA, speC, speH, and ssa. A similar logistic regression analysis was performed in order to analyze which factors were significantly associated with 30-day CFR (however, age was included without grouping). A standard backward elimination procedure was used in the logistic regression analyses. We attempted to eliminate the SAg genes (i.e., the variables speA, speC, speH, and ssa) from the model before emm. Interaction between the emm types and SAg genes was not included in our models.
A P value of <0.05 was considered significant. Odds ratios (ORs) are presented with a 95% confidence interval (95% CI).
The study was approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg (no. [KF] 11-156/00) and by the Danish Data Protection Agency (no. 2001-41-0807).
RESULTS
Incidence and demographics. Fig. 1 shows the incidence per 100,000 inhabitants per year of invasive GAS infections in Denmark during a 15-year period from 1988 to 2002 based on the isolates received in Statens Serum Institut. The average incidence in the present study period from 1999 to 2002 (2.3/100,000 per year) was similar to the average for the period 1988 to 1998.
A total of 496 GAS isolates from 492 patients were received from all of the clinical microbiological departments in Denmark in the 4-year study period. Of the received isolates, 454 (92%) were identified in blood, 10 (2%) were identified in cerebrospinal fluid, and 32 (7%) were from other specimens (synovial fluid, ascites, tissue, bones, etc.). The median age was 59.8 years (range, 0.4 to 97.4). Of the 496 GAS isolates, 268 (54%) were obtained from female patients.
Underlying diseases. Table 1 lists predisposing factors present before development of invasive GAS infection. Information was available from 472 (98%) patients, and in 348 (74%) of the patients at least one predisposing factor was registered (Table 1).
Clinical manifestations. Information was available from 487 cases. Bacteremia with no focal symptoms was registered as the only diagnosis in 132 (27%) patients and was the most frequent diagnosis, followed by erysipelas, identified in 98 (20%) patients. Septic shock was present in 124 (27%) patients. In 40% (50 of 124) of these patients, STSS was registered. Deep soft-tissue infections were found in 10% (50 of 487), including 34 patients with NF, 9 with myositis, and 7 with both NF and myositis.
Of the 487 patients, pneumonia, meningitis, and septic arthritis were identified in 50 (10%), 22 (5%), and 21 (4%) patients, respectively. Puerperal sepsis occurred in 29 women (12% of 268 women). Based on the information available, 118 of 472 (25%) of the patients were admitted to the intensive care unit, 68 of 486 (14%) were treated with mechanical ventilation, and 138 of 476 (29%) underwent surgical procedures during the admission. Intravenous immunoglobulins were administrated to 20 patients. There were no significant annual differences in the distribution of the different clinical manifestations from 1999 to 2002 (data not shown).
Case fatality rate. In total, 113 (23%) of the 492 patients died within 30 days after the culture was obtained and 56 (12%) died within the first day. There was no significant difference when comparing the annual 30-day CFRs in the study period. The age-specific incidence and outcome from 1999 to 2002 are shown in Fig. 2. The occurrence of invasive GAS infections was evident in childhood (0 to 9 years), but the incidence was progressively higher in the older age groups, along with an increasing case fatality rate: the 30-day CFRs varied greatly between the different clinical manifestations, and not surprisingly, it was highest in STSS (56% [28 of 50]). In comparison, 54% (40 of 74) of the patients with septic shock without STSS and 12% (43 of 363) of the patients without septic shock died.
The 30-day CFR was 10% (2 of 21), 10% (10 of 98), and 4% (1 of 28) in septic arthritis, erysipelas, and puerperal sepsis, respectively. Of the patients with bacteremia as the only diagnosis the 30-day CFR was 28% (37 of 132). The 30-day CFR was 38% (19 of 50) in the patients with soft-tissue infections compared to 32% (16 of 50) in pneumonia and 27% (6 of 22) in meningitis patients.
Distribution of emm types. Of the 496 GAS isolates, 493 were available for microbiological evaluation. We identified 47 different emm types (Fig. 3). The six most prevalent emm types constituted 72% (355 isolates) of all the isolates: emm1 (22%, 110 isolates), emm28 (18%, 89 isolates), emm3 (14%, 69 isolates), emm89 (9%, 44 isolates), emm12 (5%, 25 isolates), and emm4 (4%, 20 isolates). The majority (95%) of the emm1 isolates were T1, which enabled us to compare the typing results of the two typing methods relative to T1-emm1.
The distribution of the emm types varied significantly (P < 0.001) over the 4-year period: emm1 became almost three times as frequent (1999, 15% [16 of 104]; 2000, 10% [13 of 131]; 2001, 18% [19 of 103]; and 2002, 40% [62 of 162]) (P < 0.001). emm3 almost disappeared (1999, 23%; 2000, 30%; 2001, 5%; and 2002, 2%) (P < 0.001). emm28 showed a steady increase (1999, 11%; 2000, 16%; 2001, 19%; and 2002, 24%) (P = 0.043). The variations in the prevalence of emm4, emm12, and emm89 were insignificant.
Distribution of SAgs. The available 493 GAS isolates were analyzed by PCR in order to determine the presence of the genes coding the exotoxins. We identified the chromosomal genes in all the available isolates. speC was the most prevalent of the variable genes encoding the exotoxins and was identified in 62% (305 isolates) of all the invasive GAS episodes. ssa was found in 44% (217 isolates), speA was found in 34% (169 isolates), speH was found in 14% (69 isolates), and speI was found in 5% (24 isolates).
Relationship between emm and SAg genes. Among emm1 isolates 81% (89 of 110) carried the speA gene; 96% (67 of 71) of emm3 isolates and 90% (19 of 21) of the emm4 isolates carried ssa. speC was frequent in emm4 (19 of 21), emm28 (75 of 89), emm12 (23 of 27), and emm89 (35 of 44) isolates. speH was frequently carried by emm12 isolates (22 of 27), and speI was not prevalent in any of the emm isolates, except emm12 (10 of 27).
The relationship between emm1 and emm28 and the distributions of SAg genes changed from 1999 to 2002: for emm1 isolates, we found a significant decrease of speA (1999, 94% [15 of 16]; 2000, 92% [12 of 13]; 2001, 95% [18 of 19]; and 2002, 71% [44 of 62]) (P = 0.028) and ssa (1999, 25%; 2000, 38%; 2001, 32%; and 2002, 6%) (P = 0.006). In addition, a significant increase of speC was found (1999, 25%; 2000, 15%; 2001, 21%; and 2002, 53%) (P = 0.007). In the emm28 isolates from 1999 to 2002 there was a significant decrease in the carriage of speC (1999, 100% [11 of 11]; 2000, 95% [20 of 21]; 2001, 100% [20 of 20]; and 2002, 65% [24 of 37]) (P = 0.011). Likewise, in emm28 isolates speA also differed significantly (1999, 27%; 2000, 5%; 2001, 5%; and 2002, 35%) (P = 0.010). The emm1 SAg gene profiles comparing 1999 to 2001 with 2002 are shown in Table 2.
emm types and SAg genes in different clinical manifestations. For emm1 isolates exclusively, we did not detect any significant difference in the prevalence of clinical manifestations (STSS, soft-tissue infections, pneumonia, and erysipelas) over time (the years 1999, 2000, 2001, and 2002). Thus, in particular, the data did not support any difference between 1999 to 2001 and 2002 in relation to STSS (10 of 48 versus 9 of 62), soft-tissue infections (6 of 48 versus 7 of 62), pneumonia (2 of 48 versus 10 of 62), or erysipelas (8 of 48 versus 9 of 62). These comparisons were performed in order to detect whether the sudden increase of emm1 in 2002 had any clinical consequences.
The emm-type distributions in the isolates from patients with or without the different clinical manifestations are shown in Table 3. There was a significantly (P = 0.027) different emm-type distribution in the isolates obtained from patients with STSS compared to isolates from patients without STSS. Also, isolates from erysipelas patients showed a significantly (P = 0.014) different emm-type distribution compared to isolates obtained from patients without erysipelas. Women younger than 40 years old with puerperal sepsis were infected with emm types significantly different (P < 0.001) than the emm types in women younger than 40 years without puerperal sepsis. In contrast, there were no significant differences in the distribution of the emm types in patients with or without bacteremia, arthritis, soft-tissue infections, or pneumonia.
We found no significant differences when we compared the distribution of emm1/speA+ versus emm1/speA or emm1/speC+ versus emm1/speC in combination with the presence or absence of the clinical manifestations (data not shown). Table 4 shows the results of the logistic regression analyses of the factors associated with STSS and septic shock without STSS.
The median age of STSS patients was lower (58.6 years; 95% CI, 52.2 to 62.4 years) compared to the median age of patients with septic shock without STSS (68.6 years; 95% CI, 63.2 to 71.1 years) (P = 0.002). However, there was no significant difference in the proportion of patients with predisposing factors between the two groups (16 of 50 versus 13 of 75) (P = 0.083).
In a logistic regression analysis of factors associated with 30-day CFR, we found that increasing age (P < 0.001) and chronic heart or lung diseases (OR = 3.2; 95% CI, 1.3 to 7.8; P = 0.010) were independent risk factors. When patients with were compared to those without skin lesions in the analysis, the odds ratio was less than 1 (OR = 0.5; 95% CI, 0.3 to 0.9; P = 0.022), implicating skin lesions as negatively associated with 30-day CFR. No emm types were significantly associated with the 30-day CFR.
DISCUSSION
emm1 was among the most prevalent types (constituting 10 to 18%) of the registered invasive GAS isolates in Denmark from1999 to 2001 but increased to constitute 40% in 2002. Subsequently, the occurrence of noninvasive T1 infections in 2003 reflected this type-specific increase (11). In a historical perspective the present increase could be interpreted as a reemergence of emm1. In the winter of 1988 to 1989 in Denmark, an increase in the occurrence and severity of invasive GAS infections was predominantly caused by T1 isolates, which increased from 1 to 50% (22). In 1988 to 1998 the prevalence of this serotype varied between 5 and 49% (26). Coincident with these changes similar experiences with T1M1 were reported from Sweden, the United Kingdom, and the United States (19, 37, 40), initially explained by the spread of certain virulent clones (31). Recently, isolates from invasive and noninvasive infections have been investigated without identifying any clones that were statistically associated with invasive infections (27). The prevalence of any emm (or sero-) type is strain specific and is likely to result from a combination of the prevalence of the strain circulating within a community and the invasiveness of that strain, and the degree of individual and population-level immunity to that strain plays a pivotal role (15, 27). The present proportional increase of emm1 could therefore be explained by a general increased susceptibility in the population. Alternatively, in addition to the identified proportional increase of emm1, we found a change in the percentage of emm1 isolates harboring speA and speC. Furthermore, we were unable to detect any corresponding increase in the severity of the clinical presentation associated with the emm1 type. Although speculative, we suggest that this could also indicate a possible introduction of a new emm1 subclone in the susceptible population.
The incidence of invasive GAS infection ranged from 1.6 to 3.8 and 2.0 to 3.0 per 100,000 inhabitants per year in Denmark in 1988 to 1998 and 1999 to 2002, respectively. Both the fluctuating pattern and the average incidence are similar to what have been described previously (8, 12, 32). We received the majority and a constant fraction of GAS blood isolates identified in the clinical microbiologic departments and almost all the questionnaires from patients admitted to all hospitals in Denmark with invasive GAS infections and consider our results to be the best possible estimate of invasive GAS disease burden in a nationwide study.
During the study period the prevalence of the different clinical manifestations did not change, and the incidence of STSS, soft-tissue infections, pneumonia, and erysipelas lies within the ranges described in the literature (8, 12, 42). Given that our study included only invasive GAS infections, the finding of erysipelas and pneumonia as two of the most frequent clinical manifestations was unexpected. The frequency of pneumonia could be misjudged because we did not include any objective distinction between pneumonia and invasive GAS disease complicated by adult respiratory distress syndrome. Erysipelas represents one of the primary clinical manifestations in patients with skin lesions and superficial wound infections and can probably be seen as a consequence of the increasing age and number of patients with underlying diseases with potentially impaired peripheral blood supply (diabetes mellitus or chronic heart and lung diseases) (9). A previous study found that only 10% of all erysipelas cases caused by GAS yielded a positive blood culture (13).
Two recent European studies investigated correlations between the severity of GAS infections and toxin profiles or genes encoding fibronectin binding protein and concluded that observed differences basically reflected the linkage to certain emm types (36, 43). In accordance, we did not identify any of the SAg genes to be significantly associated with STSS or septic shock without STSS. However, emm1 was compared to other emm types (excluding emm3) associated with STSS, but neither the predisposing factor nor age was found to be significantly associated with STSS, which was in contrast to the risk factors associated with septic shock without STSS. Despite the fact that emm1 had a low prevalence among isolates from the infections with the best prognosis, completely opposite to the distribution of emm28 isolates, as confirmed by others (14, 32), we were unable to detect any significant difference in the 30-day CFRs between infections with emm1 isolates and those with emm28 isolates. Although there was no difference in the fatality rates, STSS patients were significantly younger than septic shock patients without STSS, which suggests different pathogenesis for these two related clinical manifestations.
We did not examine the expression of the detected SAgs, because toxin production in vitro may not accurately reflect the regulation of protein expression in vivo (20). This is of course a limitation in our conclusions since we did not evaluate the role of chromosome-encoded SAgs in the present study. Some of the most potent SAgs (i.e., SPEJ and SMEZ) are present in virtually all GAS isolates (1, 35). In addition, other bacterial virulence properties have been reported to be associated with severe GAS infections (4). At the same time, increasingly more focus has been directed towards the central role of host immunity, where the lack of protective neutralizing antibodies against individual SAgs is an essential risk factor for the development of severe SAg-mediated infections (1, 24, 28).
The overall 30-day CFR (23%) is in agreement with some previous studies (10), although somewhat higher compared to other results (8, 14, 32). We did not consider the causes of deaths but obtained only the information regarding deaths of the patients from the Central Office of Civil Registration. It has previously been estimated that approximately 90% of deaths within 30 days after cultures were obtained is directly related to the infection (23). The only risk factors significantly associated with 30-day CFR were chronic heart or lung diseases together with increasing age, and patients with skin lesions had a significantly lower fatality rate than patients without skin lesions. One obvious explanation could be that invasive infections following an impairment of the skin barrier may represent a secondary bacteremia and is probably associated with early identification and treatment.
The importance of a fast and correct diagnosis and the need for optimal treatment of patients is emphasized since more than one in four of the patients with invasive GAS infections in our study developed septic shock and more than half of these patients died within 30 days after the culture was obtained. An inadequate number of patients received intravenous immunoglobulins or clindamycin to draw any conclusions regarding the efficacy of these treatments, which previously have been associated with improved outcome (7). Prevention through vaccination against GAS might have a great impact on the incidence of GAS infection together with rheumatic fever and acute glomerulonephritis, and several vaccine candidates are being tested (6). A potential vaccine strategy is obviously based on multivalent vaccines similar to that used against pneumococcal infections, but the multiplicity of emm types causing invasive infection and the increase and disappearance of certain emm types/strains emphasize the importance of continued country- and region-specific surveillance in order to develop and maintain optimal vaccines.
In conclusion, we observed a sudden proportional increase of emm1 during the study period. Simultaneously, we observed a change in the SAg profile in the emm1 isolates. This may possibly be a consequence of an introduction of a new subclone. We were not able to identify any increase in the incidence of invasive GAS infections in Denmark in 1999 to 2002 or any change in the disease spectrum. Based on the degree of multiorgan failure, cases with septic shock can be divided into STSS and septic shock without STSS, associated with different risk factors. However, the 30-day CFRs are equally high. Further investigations are warranted to identify potential target-directed treatment strategies.
ACKNOWLEDGMENTS
We acknowledge Kirsten Burmeister, Kimette Nielsen, and Kirsten Linboe for skillful technical assistance, Annemarie Jrgensen for secretarial assistance, Asger Mortensen for data management, and Jessica Darenberg and Lotte Hougs for valuable help with emm sequencing. We acknowledge the clinical microbiologic departments for submitting the invasive isolates and the medical doctors for filling in the questionnaires and returning them to Statens Serum Institut.
There were no conflicts of interest in relation to this study.
REFERENCES
Basma, H., A. Norrby-Teglund, Y. Guedez, A. McGeer, D. E. Low, O. El Ahmedy, B. Schwartz, and M. Kotb. 1999. Risk factors in the pathogenesis of invasive group A streptococcal infections: role of protective humoral immunity. Infect. Immun. 67:1871-1877.
Beall, B., R. Facklam, and T. Thompson. 1996. Sequencing emm-specific PCR products for routine and accurate typing of group A streptococci. J. Clin. Microbiol. 34:953-958.
Beres, S. B., G. L. Sylva, K. D. Barbian, B. Lei, J. S. Hoff, N. D. Mammarella, M. Y. Liu, J. C. Smoot, S. F. Porcella, L. D. Parkins, D. S. Campbell, T. M. Smith, J. K. McCormick, D. Y. Leung, P. M. Schlievert, and J. M. Musser. 2002. Genome sequence of a serotype M3 strain of group A Streptococcus: phage-encoded toxins, the high- virulence phenotype, and clone emergence. Proc. Natl. Acad. Sci. USA 99:10078-10083.
Bisno, A. L., M. O. Brito, and C. M. Collins. 2003. Molecular basis of group A streptococcal virulence. Lancet Infect. Dis. 3:191-200.
Chatellier, S., N. Ihendyane, R. G. Kansal, F. Khambaty, H. Basma, A. Norrby-Teglund, D. E. Low, A. McGeer, and M. Kotb. 2000. Genetic relatedness and superantigen expression in group A streptococcus serotype M1 isolates from patients with severe and nonsevere invasive diseases. Infect. Immun. 68:3523-3534.
Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13:470-511.
Darenberg, J., N. Ihendyane, J. Sjolin, E. Aufwerber, S. Haidl, P. Follin, J. Andersson, and A. Norrby-Teglund. 2003. Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome: a European randomized, double-blind, placebo-controlled trial. Clin. Infect. Dis. 37:333-340.
Davies, H. D., A. McGeer, B. Schwartz, K. Green, D. Cann, A. E. Simor, and D. E. Low. 1996. Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N. Engl. J. Med. 335:547-554.
Dupuy, A., H. Benchikhi, J. C. Roujeau, P. Bernard, L. Vaillant, O. Chosidow, B. Sassolas, J. C. Guillaume, J. J. Grob, and S. Bastuji-Garin. 1999. Risk factors for erysipelas of the leg (cellulitis): case-control study. BMJ 318:1591-1594.
Efstratiou, A. 2000. Group A streptococci in the 1990s. J. Antimicrob. Chemother. 45:3-12.
Ejlertsen, T., K. Ekelund, and H. C. Schonheyder. 2004. Distribution of sites of infection and T-types of beta-haemolytic streptococci group A during a period of hyperendemicity in a Danish region, abstr. P1676. In Program and abstracts of the 14th European Congress of Clinical Microbiology and Infectious Diseases. European Society of Clinical Microbiology and Infectious Diseases, Basel, Switzerland.
Eriksson, B. K., J. Andersson, S. E. Holm, and M. Norgren. 1998. Epidemiological and clinical aspects of invasive group A streptococcal infections and the streptococcal toxic shock syndrome. Clin. Infect. Dis. 27:1428-1436.
Eriksson, B. K., C. Jorup-Ronstrom, K. Karkkonen, A. C. Sjoblom, and S. E. Holm. 1996. Erysipelas: clinical and bacteriologic spectrum and serological aspects. Clin. Infect. Dis. 23:1091-1098.
Eriksson, B. K., M. Norgren, K. McGregor, B. G. Spratt, and B. H. Normark. 2003. Group A streptococcal infections in Sweden: a comparative study of invasive and non-invasive infections and analysis of dominant T28 emm28 isolates. Clin. Infect. Dis. 37:1189-1193.
Eriksson, B. K., A. Villasenor-Sierra, M. Norgren, and D. L. Stevens. 2001. Opsonization of T1M1 group A streptococcus: dynamics of antibody production and strain specificity. Clin. Infect. Dis. 32:E24-E30.
Facklam, R., B. Beall, A. Efstratiou, V. Fischetti, D. Johnson, E. Kaplan, P. Kriz, M. Lovgren, D. Martin, B. Schwartz, A. Totolian, D. Bessen, S. Hollingshead, F. Rubin, J. Scott, and G. Tyrrell. 1999. emm typing and validation of provisional M types for group A streptococci. Emerg. Infect. Dis. 5:247-253.
Facklam, R. F., D. R. Martin, M. Lovgren, D. R. Johnson, A. Efstratiou, T. A. Thompson, S. Gowan, P. Kriz, G. J. Tyrrell, E. Kaplan, and B. Beall. 2002. Extension of the Lancefield classification for group A streptococci by addition of 22 new M protein gene sequence types from clinical isolates: emm103 to emm124. Clin. Infect. Dis. 34:28-38.
Ferretti, J. J., W. M. McShan, D. Ajdic, D. J. Savic, G. Savic, K. Lyon, C. Primeaux, S. Sezate, A. N. Suvorov, S. Kenton, H. S. Lai, S. P. Lin, Y. Qian, H. G. Jia, F. Z. Najar, Q. Ren, H. Zhu, L. Song, J. White, X. Yuan, S. W. Clifton, B. A. Roe, and R. McLaughlin. 2001. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. USA 98:4658-4663.
Gaworzewska, E., and G. Colman. 1988. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol. Infect. 100:257-269.
Hauser, A. R., D. L. Stevens, E. L. Kaplan, and P. M. Schlievert. 1991. Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J. Clin. Microbiol. 29:1562-1567.
Helweg-Larsen, J., J. S. Jensen, T. Benfield, U. G. Svendsen, J. D. Lundgren, and B. Lundgren. 1998. Diagnostic use of PCR for detection of Pneumocystis carinii in oral wash samples. J. Clin. Microbiol. 36:2068-2072.
Henrichsen, J., and T. Rnne. 1992. Reemergence of servere group A streptococcal infections in Denmark 1988-89. Zentbl. Bakteriol. Suppl. 22:8-10.
Hindsholm, M., and H. C. Schonheyder. 2002. Clinical presentation and outcome of bacteraemia caused by beta-haemolytic streptococci serogroup G. APMIS 110:554-558.
Holm, S. E., A. Norrby, A. M. Bergholm, and M. Norgren. 1992. Aspects of pathogenesis of serious group A streptococcal infections in Sweden 1988-89. J. Infect. Dis. 166:31-37.
Jasir, A., A. Tanna, A. Efstratiou, and C. Schalen. 2001. Unusual occurrence of M type 77, antibiotic-resistant group A streptococci in southern Sweden. J. Clin. Microbiol. 39:586-590.
Johansen, K. E. 2000. Group A streptococcal T-protein in invasive and non-invasive isolates in Denmark 1987-98, abstr. P9.10. In Program and abstracts of the 14th Lancefield International Symposia on Severe Streptococcal Diseases, Auckland, New Zealand.
Johnson, D. R., J. T. Wotton, A. Shet, and E. L. Kaplan. 2002. A comparison of group A streptococci from invasive and uncomplicated infections: are virulent clones responsible for serious streptococcal infections J. Infect. Dis. 185:1586-1595.
Kotb, M., A. Norrby-Teglund, A. McGeer, H. El Sherbini, M. T. Dorak, A. Khurshid, K. Green, J. Peeples, J. Wade, G. Thomson, B. Schwartz, and D. E. Low. 2002. An immunogenetic and molecular basis for differences in outcomes of invasive group A streptococcal infections. Nat. Med. 8:1398-1404.
Li, Z., V. Sakota, D. Jackson, A. R. Franklin, and B. Beall. 2003. Array of M protein gene subtypes in 1064 recent invasive group A streptococcus isolates recovered from the active bacterial core surveillance. J. Infect. Dis. 188:1587-1592.
McCormick, J. K., A. A. Pragman, J. C. Stolpa, D. Y. Leung, and P. M. Schlievert. 2001. Functional characterization of streptococcal pyrogenic exotoxin J, a novel superantigen. Infect. Immun. 69:1381-1388.
Musser, J. M., V. Kapur, J. Szeto, X. Pan, D. S. Swanson, and D. R. Martin. 1995. Genetic diversity and relationships among Streptococcus pyogenes strains expressing serotype M1 protein: recent intercontinental spread of a subclone causing episodes of invasive disease. Infect. Immun. 63:994-1003.
O'Brien, K. L., B. Beall, N. L. Barrett, P. R. Cieslak, A. Reingold, M. M. Farley, R. Danila, E. R. Zell, R. Facklam, B. Schwartz, and A. Schuchat. 2002. Epidemiology of invasive group A streptococcus disease in the United States, 1995-1999. Clin. Infect. Dis. 35:268-276.
Proft, T., V. L. Arcus, V. Handley, E. N. Baker, and J. D. Fraser. 2001. Immunological and biochemical characterization of streptococcal pyrogenic exotoxins I and J (SPE-I and SPE-J) from Streptococcus pyogenes. J. Immunol. 166:6711-6719.
Proft, T., S. L. Moffatt, C. J. Berkahn, and J. D. Fraser. 1999. Identification and characterization of novel superantigens from Streptococcus pyogenes. J. Exp. Med. 189:89-102.
Proft, T., S. Sriskandan, L. Yang, and J. D. Fraser. 2003. Superantigens and streptococcal toxic shock syndrome. Emerg. Infect. Dis. 9:1211-1218.
Schmitz, F. J., A. Beyer, E. Charpentier, B. H. Normark, M. Schade, A. C. Fluit, D. Hafner, and R. Novak. 2003. Toxin-gene profile heterogeneity among endemic invasive European group A streptococcal isolates. J. Infect. Dis. 188:1578-1586.
Schwartz, B., R. R. Facklam, and R. F. Breiman. 1990. Changing epidemiology of group A streptococcal infection in the USA. Lancet 336:1167-1171.
Shulman, S. T., R. R. Tanz, W. Kabat, K. Kabat, E. Cederlund, D. Patel, Z. Li, V. Sakota, J. B. Dale, and B. Beall. 2004. Group A streptococcal pharyngitis serotype surveillance in North America, 2000-2002. Clin. Infect. Dis. 39:325-332.
Smoot, J. C., K. D. Barbian, J. J. Van Gompel, L. M. Smoot, M. S. Chaussee, G. L. Sylva, D. E. Sturdevant, S. M. Ricklefs, S. F. Porcella, L. D. Parkins, S. B. Beres, D. S. Campbell, T. M. Smith, Q. Zhang, V. Kapur, J. A. Daly, L. G. Veasy, and J. M. Musser. 2002. Genome sequence and comparative microarray analysis of serotype M18 group A Streptococcus strains associated with acute rheumatic fever outbreaks. Proc. Natl. Acad. Sci. USA 99:4668-4673.
Stromberg, A., V. Romanus, and L. G. Burman. 1991. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J. Infect. Dis. 164:595-598.
Unnikrishnan, M., D. M. Altmann, T. Proft, F. Wahid, J. Cohen, J. D. Fraser, and S. Sriskandan. 2002. The bacterial superantigen streptococcal mitogenic exotoxin Z is the major immunoactive agent of Streptococcus pyogenes. J. Immunol. 169:2561-2569.
Vlaminckx, B., W. van Pelt, L. Schouls, A. Van Silfhout, C. Elzenaar, E. Mascini, J. Verhoef, and J. Schellekens. 2004. Epidemiological features of invasive and noninvasive group A streptococcal disease in The Netherlands, 1992-1996. Eur. J. Clin. Microbiol. Infect. Dis. 23:434-444.
Vlaminckx, B. J., E. M. Mascini, J. Schellekens, L. M. Schouls, A. Paauw, A. C. Fluit, R. Novak, J. Verhoef, and F. J. Schmitz. 2003. Site-specific manifestations of invasive group A streptococcal disease: type distribution and corresponding patterns of virulence determinants. J. Clin. Microbiol. 41:4941-4949.
Working Group on Severe Streptococcal Infections. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 269:390-391.(Kim Ekelund, Peter Skinhj)
Biostatistics Unit, Statens Serum Institut
Department of Infectious Diseases, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
ABSTRACT
Between 1999 and 2002, 496 invasive group A streptococcal (GAS) isolates from clinical microbiological departments in Denmark and subsequently 487 (98%) questionnaires from the clinicians treating the patients were received as part of a national surveillance. emm types and streptococcal superantigen (SAg) genes were determined. The incidence of invasive GAS infections was on average 2.3 per 100,000 per year. Bacteremia with no focal symptoms (27%) was together with erysipelas (20%) the most prevalent clinical diagnoses. Streptococcal toxic shock syndrome occurred in 10% of patients, of which 56% died. The overall case fatality rate within 30 days was 23%. In total, 47 different emm types were identified, of which emm1, emm3, emm4, emm12, emm28, and emm89 were identified in 72% of the 493 available isolates. During the 4-year period the presence of emm1 increased from 16% in 1999 to 40% in 2002. Concurrently, the presence of emm3 decreased from 23% in 1999 to 2% in 2002. The emm1 isolates predominantly carried speA, although the frequency decreased from 94% in 1999 to 71% in 2002, whereas the emm1-specific prevalence of speC increased from 25 to 53%. In a historical perspective, this could be interpreted as a reemergence of emm1 and could indicate a possible introduction of a new emm1 subclone. However, this reemergence did not result in any significant changes in the clinical manifestations during the study period. Our results show the complexity of invasive GAS infections, with time-dependent variations in the incidence and distribution of emm and SAg genes, which emphasizes the need for continuous epidemiological and molecular investigations.
INTRODUCTION
In the last two decades, Streptococcus pyogenes (group A streptococci [GAS]) has been identified as an emerging cause of severe infections: septic shock, organ failure, soft-tissue infections (myositis and necrotizing fasciitis [NF]), and streptococcal toxic shock syndrome (STSS) with high mortality.
Since the Working Group on Severe Streptococcal Infections proposed diagnostic criteria for STSS in 1993 (44), several studies of the epidemiological, microbiological, and clinical aspects of invasive GAS infection have been performed in various countries. However, many questions about the pathogenesis of invasive GAS infection still remain unanswered.
Traditional methods of T-agglutination typing (T typing) and M-precipitation typing (M typing) have been used in epidemiological studies for the last 50 years (10). Recently, new molecular methods have replaced these conventional methods, where emm sequencing, detection of the genes encoding M proteins, has been introduced. This has made epidemiological surveillance more detailed and revealed potential clusters (emm types) in specific clinical manifestations (17). In addition, emm subtypes have been introduced in recent surveillance papers (29, 38). However, a universal high prevalence of certain emm types in invasive GAS diseases may also reflect widespread transmission rather than an increased virulence and invasiveness.
Streptococcal exotoxins are presumed to play an important role in severe diseases, acting as superantigens (SAgs) and thereby inducing a devastating cytokine response in susceptible hosts (35). The number of identified potential SAgs has increased in the last few years, facilitated by the information obtained from the published whole genome of GAS (3, 18, 39). Despite numerous reports of SAg distributions, no previous study has, to our knowledge, made use of nationwide, longitudinal data from a population-based surveillance.
In the present study, epidemiological and disease-related data are reported in addition to the emm and SAg gene profiles, i.e., genes encoding pyrogenic exotoxins A to C, F to J, SSA, and SMEZ (speA to -C, speF to -J, ssa, and smeZ genes), to evaluate the differences in the clinical manifestations of GAS infections from the national surveillance of invasive GAS infections in Denmark from 1999 to 2002.
MATERIALS AND METHODS
Subjects and specimens. The Streptococcus Unit serves as the National Streptococcus Reference Centre and receives GAS isolates from normally sterile sites in patients admitted to all hospitals in Denmark (population, 5.34 million). The GAS isolates are received as pure cultures from all of the 15 Danish clinical microbiological departments as part of the national surveillance. From two-thirds of the clinical microbiological departments information was received, which enabled us to estimate that the Streptococcus Unit received on average 79% of the GAS blood isolates identified by the clinical microbiological departments and that this percentage remained constant during the study period (January 1999 to December 2002).
The reporting system from the clinical microbiological departments to the Streptococcus Unit has been the same since 1988, and since 1996 the Streptococcus Unit has distributed a detailed questionnaire to the clinical doctors treating the patients. In 1999 the questionnaire was redesigned to include information about the dates of admission, of discharge (or death), and of onset of primary symptoms of the infection and additionally to include a description of the type of primary symptoms, the course of the infection, treatment, and predisposing factors.
In the present study the following definitions were used. Bacteremia was defined as a clinical entity associated with identification of GAS in the blood culture without specific focus on the infection. NF was defined as diagnosis by the clinicians of necrosis of the fascia and of tissue (excluding muscle). A soft-tissue infection was defined as either NF or myositis. A patient with septic shock was defined as a patient with invasive GAS infection and a systolic blood pressure below 90 mm Hg, and finally the definition of STSS was based on the consensus definition from the Working Group on Severe Streptococcal Infections (44). An overall case fatality rate was assessed at day 30 after the culture was obtained (30-day CFR).
Date of death or a confirmation that the person was alive by 1 March 2004 was obtained from the Central Office of Civil Registration. The annual number of inhabitants in the period 1988 to 2002 was reported from Statistics Denmark. Incidence rates are presented as incidence per 100,000 inhabitants per year.
DNA preparation and emm sequencing. A few colonies from a fresh overnight culture on 5% blood agar were mixed with Chelex 100 (Bio-Rad, Richmond, Calif.) in TE buffer (21). A 0.5-μl portion of the supernatant containing the genomic DNA was used for emm sequencing and identification of exotoxin genes.
The emm gene types of GAS were determined by amplification and sequencing of the emm genes essentially as described elsewhere (2), although the PCR amplification was performed in a final volume of 25 μl containing 2.5 μl of 10x PCR buffer (200 mM Tris-HCl [pH 8.4], 500 mM KCl [Invitrogen, Carlsbad, Calif.]), 0.2 mM deoxynucleoside triphosphate mixture, 1.5 mM MgCl2 (final concentration), and a 0.4 μM concentration of each primer. Platinum Taq DNA polymerase (0.5 U; Invitrogen) was used in each reaction mixture. The following cycle parameters were used: 95°C for 15 min; 35 cycles of 94°C for 30 s, 48°C for 30 s, and 72°C for 1.5 min; and a final extension step at 72°C for 10 min. Sequencing reactions were performed by using the BigDye terminator cycle-sequencing ready-reaction kit version 2.0 (Applied Biosystems, Foster City, Calif.) according to the manufacturer's instructions, using emmseq2 as the sequencing primer. The emm sequences were obtained according to the recommendations (16) and aligned with sequences available in the Centers for Disease Control and Prevention emm-type database (http://www.cdc.gov/ncidod/biotech/strep/strepindex.html).
Identification of SAg genes (speA to -C, speF to -J, ssa, and smeZ). The primers used for identification of the exotoxin genes were found in the literature (5, 25, 30, 33, 34, 41). PCR was done as described above. Detection of speF to- J, ssa, and smeZ was performed using the following protocol: 2 min at 95°C and a total of 40 cycles of 94°C for 1 min, 50°C for 30 s, and 72°C for 1.5 min. Detection of speA to -C was carried out with the same protocol except that annealing was at 58°C for 35 cycles (25). The presence of the exotoxin gene was visually ascertained on an ethidium-stained agarose gel. Amplification of each of the genes located on mobile DNA (i.e., speA, -C, -H, and -I and ssa) was carried out in combination with amplification of one chromosomal gene (i.e., speB, -F, -G, or -J, smeZ, or the 16S rRNA gene) functioning as a positive control.
Statistical analyses. Data from the survey were analyzed with SAS System Release 8.02 (SAS Institute Inc. Cary, N.C.) and GraphPad Prism version 3.00 for Windows (GraphPad Software, San Diego, Calif.). A chi-square test of independence (in a two-by-four table) was conducted to analyze proportions of clinical manifestations (including all emm types and analysis of emm1 isolates exclusively) versus time (for the four years 1999, 2000, 2001, and 2002). The same type of analysis was used for comparisons of proportions of 30-day CFRs versus time and comparisons of proportions of SAg genes for emm1 and emm28 isolates, respectively, versus time. In the analysis of distributions of emm types (grouped emm1, emm3, emm4, emm12, emm28, emm89, and others) versus time, an overall chi-square test of independence (in a seven-by-four table) was performed initially. If an overall significant difference was observed, an analysis of the proportion of each emm type versus time was performed by a chi-square test of independence (in a two-by-four table). For each of the clinical manifestations the emm distributions in the two groups defined by the presence and absence, respectively, of the manifestation were compared in a likewise manner. Possible correlations between each of the clinical manifestations and the presence or absence of each of the SAg genes were examined by a chi-square test of independence (in a two-by-two table). Likewise, the distributions of emm1 isolates carrying the speA gene versus those lacking the speA gene and eventually of emm1 isolates carrying the speC gene versus those lacking the speC gene were examined in the absence or presence of clinical manifestations by a chi-square test (in a two-by-two table).
The proportions of predisposing factors were also compared with a chi-square test. In general, chi-square tests of independence were replaced by analogue exact tests in the case of sparse data.
For comparisons of age distributions between patient groups, Kruskal-Wallis or Mann-Whitney rank sum tests were used. Two-logistic regression analyses were performed in order to analyze which factors were significantly associated with STSS and with septic shock without fulfilling the criteria for STSS (44). The independent factors included in the analysis were sex, age (grouped as <40, 40 to 65, and >65 years), cancer, diabetes, chronic heart or lung diseases, skin lesions, alcohol abuse, immune incompetence, emm (grouped as emm1, emm3, emm28, and others), speA, speC, speH, and ssa. A similar logistic regression analysis was performed in order to analyze which factors were significantly associated with 30-day CFR (however, age was included without grouping). A standard backward elimination procedure was used in the logistic regression analyses. We attempted to eliminate the SAg genes (i.e., the variables speA, speC, speH, and ssa) from the model before emm. Interaction between the emm types and SAg genes was not included in our models.
A P value of <0.05 was considered significant. Odds ratios (ORs) are presented with a 95% confidence interval (95% CI).
The study was approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg (no. [KF] 11-156/00) and by the Danish Data Protection Agency (no. 2001-41-0807).
RESULTS
Incidence and demographics. Fig. 1 shows the incidence per 100,000 inhabitants per year of invasive GAS infections in Denmark during a 15-year period from 1988 to 2002 based on the isolates received in Statens Serum Institut. The average incidence in the present study period from 1999 to 2002 (2.3/100,000 per year) was similar to the average for the period 1988 to 1998.
A total of 496 GAS isolates from 492 patients were received from all of the clinical microbiological departments in Denmark in the 4-year study period. Of the received isolates, 454 (92%) were identified in blood, 10 (2%) were identified in cerebrospinal fluid, and 32 (7%) were from other specimens (synovial fluid, ascites, tissue, bones, etc.). The median age was 59.8 years (range, 0.4 to 97.4). Of the 496 GAS isolates, 268 (54%) were obtained from female patients.
Underlying diseases. Table 1 lists predisposing factors present before development of invasive GAS infection. Information was available from 472 (98%) patients, and in 348 (74%) of the patients at least one predisposing factor was registered (Table 1).
Clinical manifestations. Information was available from 487 cases. Bacteremia with no focal symptoms was registered as the only diagnosis in 132 (27%) patients and was the most frequent diagnosis, followed by erysipelas, identified in 98 (20%) patients. Septic shock was present in 124 (27%) patients. In 40% (50 of 124) of these patients, STSS was registered. Deep soft-tissue infections were found in 10% (50 of 487), including 34 patients with NF, 9 with myositis, and 7 with both NF and myositis.
Of the 487 patients, pneumonia, meningitis, and septic arthritis were identified in 50 (10%), 22 (5%), and 21 (4%) patients, respectively. Puerperal sepsis occurred in 29 women (12% of 268 women). Based on the information available, 118 of 472 (25%) of the patients were admitted to the intensive care unit, 68 of 486 (14%) were treated with mechanical ventilation, and 138 of 476 (29%) underwent surgical procedures during the admission. Intravenous immunoglobulins were administrated to 20 patients. There were no significant annual differences in the distribution of the different clinical manifestations from 1999 to 2002 (data not shown).
Case fatality rate. In total, 113 (23%) of the 492 patients died within 30 days after the culture was obtained and 56 (12%) died within the first day. There was no significant difference when comparing the annual 30-day CFRs in the study period. The age-specific incidence and outcome from 1999 to 2002 are shown in Fig. 2. The occurrence of invasive GAS infections was evident in childhood (0 to 9 years), but the incidence was progressively higher in the older age groups, along with an increasing case fatality rate: the 30-day CFRs varied greatly between the different clinical manifestations, and not surprisingly, it was highest in STSS (56% [28 of 50]). In comparison, 54% (40 of 74) of the patients with septic shock without STSS and 12% (43 of 363) of the patients without septic shock died.
The 30-day CFR was 10% (2 of 21), 10% (10 of 98), and 4% (1 of 28) in septic arthritis, erysipelas, and puerperal sepsis, respectively. Of the patients with bacteremia as the only diagnosis the 30-day CFR was 28% (37 of 132). The 30-day CFR was 38% (19 of 50) in the patients with soft-tissue infections compared to 32% (16 of 50) in pneumonia and 27% (6 of 22) in meningitis patients.
Distribution of emm types. Of the 496 GAS isolates, 493 were available for microbiological evaluation. We identified 47 different emm types (Fig. 3). The six most prevalent emm types constituted 72% (355 isolates) of all the isolates: emm1 (22%, 110 isolates), emm28 (18%, 89 isolates), emm3 (14%, 69 isolates), emm89 (9%, 44 isolates), emm12 (5%, 25 isolates), and emm4 (4%, 20 isolates). The majority (95%) of the emm1 isolates were T1, which enabled us to compare the typing results of the two typing methods relative to T1-emm1.
The distribution of the emm types varied significantly (P < 0.001) over the 4-year period: emm1 became almost three times as frequent (1999, 15% [16 of 104]; 2000, 10% [13 of 131]; 2001, 18% [19 of 103]; and 2002, 40% [62 of 162]) (P < 0.001). emm3 almost disappeared (1999, 23%; 2000, 30%; 2001, 5%; and 2002, 2%) (P < 0.001). emm28 showed a steady increase (1999, 11%; 2000, 16%; 2001, 19%; and 2002, 24%) (P = 0.043). The variations in the prevalence of emm4, emm12, and emm89 were insignificant.
Distribution of SAgs. The available 493 GAS isolates were analyzed by PCR in order to determine the presence of the genes coding the exotoxins. We identified the chromosomal genes in all the available isolates. speC was the most prevalent of the variable genes encoding the exotoxins and was identified in 62% (305 isolates) of all the invasive GAS episodes. ssa was found in 44% (217 isolates), speA was found in 34% (169 isolates), speH was found in 14% (69 isolates), and speI was found in 5% (24 isolates).
Relationship between emm and SAg genes. Among emm1 isolates 81% (89 of 110) carried the speA gene; 96% (67 of 71) of emm3 isolates and 90% (19 of 21) of the emm4 isolates carried ssa. speC was frequent in emm4 (19 of 21), emm28 (75 of 89), emm12 (23 of 27), and emm89 (35 of 44) isolates. speH was frequently carried by emm12 isolates (22 of 27), and speI was not prevalent in any of the emm isolates, except emm12 (10 of 27).
The relationship between emm1 and emm28 and the distributions of SAg genes changed from 1999 to 2002: for emm1 isolates, we found a significant decrease of speA (1999, 94% [15 of 16]; 2000, 92% [12 of 13]; 2001, 95% [18 of 19]; and 2002, 71% [44 of 62]) (P = 0.028) and ssa (1999, 25%; 2000, 38%; 2001, 32%; and 2002, 6%) (P = 0.006). In addition, a significant increase of speC was found (1999, 25%; 2000, 15%; 2001, 21%; and 2002, 53%) (P = 0.007). In the emm28 isolates from 1999 to 2002 there was a significant decrease in the carriage of speC (1999, 100% [11 of 11]; 2000, 95% [20 of 21]; 2001, 100% [20 of 20]; and 2002, 65% [24 of 37]) (P = 0.011). Likewise, in emm28 isolates speA also differed significantly (1999, 27%; 2000, 5%; 2001, 5%; and 2002, 35%) (P = 0.010). The emm1 SAg gene profiles comparing 1999 to 2001 with 2002 are shown in Table 2.
emm types and SAg genes in different clinical manifestations. For emm1 isolates exclusively, we did not detect any significant difference in the prevalence of clinical manifestations (STSS, soft-tissue infections, pneumonia, and erysipelas) over time (the years 1999, 2000, 2001, and 2002). Thus, in particular, the data did not support any difference between 1999 to 2001 and 2002 in relation to STSS (10 of 48 versus 9 of 62), soft-tissue infections (6 of 48 versus 7 of 62), pneumonia (2 of 48 versus 10 of 62), or erysipelas (8 of 48 versus 9 of 62). These comparisons were performed in order to detect whether the sudden increase of emm1 in 2002 had any clinical consequences.
The emm-type distributions in the isolates from patients with or without the different clinical manifestations are shown in Table 3. There was a significantly (P = 0.027) different emm-type distribution in the isolates obtained from patients with STSS compared to isolates from patients without STSS. Also, isolates from erysipelas patients showed a significantly (P = 0.014) different emm-type distribution compared to isolates obtained from patients without erysipelas. Women younger than 40 years old with puerperal sepsis were infected with emm types significantly different (P < 0.001) than the emm types in women younger than 40 years without puerperal sepsis. In contrast, there were no significant differences in the distribution of the emm types in patients with or without bacteremia, arthritis, soft-tissue infections, or pneumonia.
We found no significant differences when we compared the distribution of emm1/speA+ versus emm1/speA or emm1/speC+ versus emm1/speC in combination with the presence or absence of the clinical manifestations (data not shown). Table 4 shows the results of the logistic regression analyses of the factors associated with STSS and septic shock without STSS.
The median age of STSS patients was lower (58.6 years; 95% CI, 52.2 to 62.4 years) compared to the median age of patients with septic shock without STSS (68.6 years; 95% CI, 63.2 to 71.1 years) (P = 0.002). However, there was no significant difference in the proportion of patients with predisposing factors between the two groups (16 of 50 versus 13 of 75) (P = 0.083).
In a logistic regression analysis of factors associated with 30-day CFR, we found that increasing age (P < 0.001) and chronic heart or lung diseases (OR = 3.2; 95% CI, 1.3 to 7.8; P = 0.010) were independent risk factors. When patients with were compared to those without skin lesions in the analysis, the odds ratio was less than 1 (OR = 0.5; 95% CI, 0.3 to 0.9; P = 0.022), implicating skin lesions as negatively associated with 30-day CFR. No emm types were significantly associated with the 30-day CFR.
DISCUSSION
emm1 was among the most prevalent types (constituting 10 to 18%) of the registered invasive GAS isolates in Denmark from1999 to 2001 but increased to constitute 40% in 2002. Subsequently, the occurrence of noninvasive T1 infections in 2003 reflected this type-specific increase (11). In a historical perspective the present increase could be interpreted as a reemergence of emm1. In the winter of 1988 to 1989 in Denmark, an increase in the occurrence and severity of invasive GAS infections was predominantly caused by T1 isolates, which increased from 1 to 50% (22). In 1988 to 1998 the prevalence of this serotype varied between 5 and 49% (26). Coincident with these changes similar experiences with T1M1 were reported from Sweden, the United Kingdom, and the United States (19, 37, 40), initially explained by the spread of certain virulent clones (31). Recently, isolates from invasive and noninvasive infections have been investigated without identifying any clones that were statistically associated with invasive infections (27). The prevalence of any emm (or sero-) type is strain specific and is likely to result from a combination of the prevalence of the strain circulating within a community and the invasiveness of that strain, and the degree of individual and population-level immunity to that strain plays a pivotal role (15, 27). The present proportional increase of emm1 could therefore be explained by a general increased susceptibility in the population. Alternatively, in addition to the identified proportional increase of emm1, we found a change in the percentage of emm1 isolates harboring speA and speC. Furthermore, we were unable to detect any corresponding increase in the severity of the clinical presentation associated with the emm1 type. Although speculative, we suggest that this could also indicate a possible introduction of a new emm1 subclone in the susceptible population.
The incidence of invasive GAS infection ranged from 1.6 to 3.8 and 2.0 to 3.0 per 100,000 inhabitants per year in Denmark in 1988 to 1998 and 1999 to 2002, respectively. Both the fluctuating pattern and the average incidence are similar to what have been described previously (8, 12, 32). We received the majority and a constant fraction of GAS blood isolates identified in the clinical microbiologic departments and almost all the questionnaires from patients admitted to all hospitals in Denmark with invasive GAS infections and consider our results to be the best possible estimate of invasive GAS disease burden in a nationwide study.
During the study period the prevalence of the different clinical manifestations did not change, and the incidence of STSS, soft-tissue infections, pneumonia, and erysipelas lies within the ranges described in the literature (8, 12, 42). Given that our study included only invasive GAS infections, the finding of erysipelas and pneumonia as two of the most frequent clinical manifestations was unexpected. The frequency of pneumonia could be misjudged because we did not include any objective distinction between pneumonia and invasive GAS disease complicated by adult respiratory distress syndrome. Erysipelas represents one of the primary clinical manifestations in patients with skin lesions and superficial wound infections and can probably be seen as a consequence of the increasing age and number of patients with underlying diseases with potentially impaired peripheral blood supply (diabetes mellitus or chronic heart and lung diseases) (9). A previous study found that only 10% of all erysipelas cases caused by GAS yielded a positive blood culture (13).
Two recent European studies investigated correlations between the severity of GAS infections and toxin profiles or genes encoding fibronectin binding protein and concluded that observed differences basically reflected the linkage to certain emm types (36, 43). In accordance, we did not identify any of the SAg genes to be significantly associated with STSS or septic shock without STSS. However, emm1 was compared to other emm types (excluding emm3) associated with STSS, but neither the predisposing factor nor age was found to be significantly associated with STSS, which was in contrast to the risk factors associated with septic shock without STSS. Despite the fact that emm1 had a low prevalence among isolates from the infections with the best prognosis, completely opposite to the distribution of emm28 isolates, as confirmed by others (14, 32), we were unable to detect any significant difference in the 30-day CFRs between infections with emm1 isolates and those with emm28 isolates. Although there was no difference in the fatality rates, STSS patients were significantly younger than septic shock patients without STSS, which suggests different pathogenesis for these two related clinical manifestations.
We did not examine the expression of the detected SAgs, because toxin production in vitro may not accurately reflect the regulation of protein expression in vivo (20). This is of course a limitation in our conclusions since we did not evaluate the role of chromosome-encoded SAgs in the present study. Some of the most potent SAgs (i.e., SPEJ and SMEZ) are present in virtually all GAS isolates (1, 35). In addition, other bacterial virulence properties have been reported to be associated with severe GAS infections (4). At the same time, increasingly more focus has been directed towards the central role of host immunity, where the lack of protective neutralizing antibodies against individual SAgs is an essential risk factor for the development of severe SAg-mediated infections (1, 24, 28).
The overall 30-day CFR (23%) is in agreement with some previous studies (10), although somewhat higher compared to other results (8, 14, 32). We did not consider the causes of deaths but obtained only the information regarding deaths of the patients from the Central Office of Civil Registration. It has previously been estimated that approximately 90% of deaths within 30 days after cultures were obtained is directly related to the infection (23). The only risk factors significantly associated with 30-day CFR were chronic heart or lung diseases together with increasing age, and patients with skin lesions had a significantly lower fatality rate than patients without skin lesions. One obvious explanation could be that invasive infections following an impairment of the skin barrier may represent a secondary bacteremia and is probably associated with early identification and treatment.
The importance of a fast and correct diagnosis and the need for optimal treatment of patients is emphasized since more than one in four of the patients with invasive GAS infections in our study developed septic shock and more than half of these patients died within 30 days after the culture was obtained. An inadequate number of patients received intravenous immunoglobulins or clindamycin to draw any conclusions regarding the efficacy of these treatments, which previously have been associated with improved outcome (7). Prevention through vaccination against GAS might have a great impact on the incidence of GAS infection together with rheumatic fever and acute glomerulonephritis, and several vaccine candidates are being tested (6). A potential vaccine strategy is obviously based on multivalent vaccines similar to that used against pneumococcal infections, but the multiplicity of emm types causing invasive infection and the increase and disappearance of certain emm types/strains emphasize the importance of continued country- and region-specific surveillance in order to develop and maintain optimal vaccines.
In conclusion, we observed a sudden proportional increase of emm1 during the study period. Simultaneously, we observed a change in the SAg profile in the emm1 isolates. This may possibly be a consequence of an introduction of a new subclone. We were not able to identify any increase in the incidence of invasive GAS infections in Denmark in 1999 to 2002 or any change in the disease spectrum. Based on the degree of multiorgan failure, cases with septic shock can be divided into STSS and septic shock without STSS, associated with different risk factors. However, the 30-day CFRs are equally high. Further investigations are warranted to identify potential target-directed treatment strategies.
ACKNOWLEDGMENTS
We acknowledge Kirsten Burmeister, Kimette Nielsen, and Kirsten Linboe for skillful technical assistance, Annemarie Jrgensen for secretarial assistance, Asger Mortensen for data management, and Jessica Darenberg and Lotte Hougs for valuable help with emm sequencing. We acknowledge the clinical microbiologic departments for submitting the invasive isolates and the medical doctors for filling in the questionnaires and returning them to Statens Serum Institut.
There were no conflicts of interest in relation to this study.
REFERENCES
Basma, H., A. Norrby-Teglund, Y. Guedez, A. McGeer, D. E. Low, O. El Ahmedy, B. Schwartz, and M. Kotb. 1999. Risk factors in the pathogenesis of invasive group A streptococcal infections: role of protective humoral immunity. Infect. Immun. 67:1871-1877.
Beall, B., R. Facklam, and T. Thompson. 1996. Sequencing emm-specific PCR products for routine and accurate typing of group A streptococci. J. Clin. Microbiol. 34:953-958.
Beres, S. B., G. L. Sylva, K. D. Barbian, B. Lei, J. S. Hoff, N. D. Mammarella, M. Y. Liu, J. C. Smoot, S. F. Porcella, L. D. Parkins, D. S. Campbell, T. M. Smith, J. K. McCormick, D. Y. Leung, P. M. Schlievert, and J. M. Musser. 2002. Genome sequence of a serotype M3 strain of group A Streptococcus: phage-encoded toxins, the high- virulence phenotype, and clone emergence. Proc. Natl. Acad. Sci. USA 99:10078-10083.
Bisno, A. L., M. O. Brito, and C. M. Collins. 2003. Molecular basis of group A streptococcal virulence. Lancet Infect. Dis. 3:191-200.
Chatellier, S., N. Ihendyane, R. G. Kansal, F. Khambaty, H. Basma, A. Norrby-Teglund, D. E. Low, A. McGeer, and M. Kotb. 2000. Genetic relatedness and superantigen expression in group A streptococcus serotype M1 isolates from patients with severe and nonsevere invasive diseases. Infect. Immun. 68:3523-3534.
Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13:470-511.
Darenberg, J., N. Ihendyane, J. Sjolin, E. Aufwerber, S. Haidl, P. Follin, J. Andersson, and A. Norrby-Teglund. 2003. Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome: a European randomized, double-blind, placebo-controlled trial. Clin. Infect. Dis. 37:333-340.
Davies, H. D., A. McGeer, B. Schwartz, K. Green, D. Cann, A. E. Simor, and D. E. Low. 1996. Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N. Engl. J. Med. 335:547-554.
Dupuy, A., H. Benchikhi, J. C. Roujeau, P. Bernard, L. Vaillant, O. Chosidow, B. Sassolas, J. C. Guillaume, J. J. Grob, and S. Bastuji-Garin. 1999. Risk factors for erysipelas of the leg (cellulitis): case-control study. BMJ 318:1591-1594.
Efstratiou, A. 2000. Group A streptococci in the 1990s. J. Antimicrob. Chemother. 45:3-12.
Ejlertsen, T., K. Ekelund, and H. C. Schonheyder. 2004. Distribution of sites of infection and T-types of beta-haemolytic streptococci group A during a period of hyperendemicity in a Danish region, abstr. P1676. In Program and abstracts of the 14th European Congress of Clinical Microbiology and Infectious Diseases. European Society of Clinical Microbiology and Infectious Diseases, Basel, Switzerland.
Eriksson, B. K., J. Andersson, S. E. Holm, and M. Norgren. 1998. Epidemiological and clinical aspects of invasive group A streptococcal infections and the streptococcal toxic shock syndrome. Clin. Infect. Dis. 27:1428-1436.
Eriksson, B. K., C. Jorup-Ronstrom, K. Karkkonen, A. C. Sjoblom, and S. E. Holm. 1996. Erysipelas: clinical and bacteriologic spectrum and serological aspects. Clin. Infect. Dis. 23:1091-1098.
Eriksson, B. K., M. Norgren, K. McGregor, B. G. Spratt, and B. H. Normark. 2003. Group A streptococcal infections in Sweden: a comparative study of invasive and non-invasive infections and analysis of dominant T28 emm28 isolates. Clin. Infect. Dis. 37:1189-1193.
Eriksson, B. K., A. Villasenor-Sierra, M. Norgren, and D. L. Stevens. 2001. Opsonization of T1M1 group A streptococcus: dynamics of antibody production and strain specificity. Clin. Infect. Dis. 32:E24-E30.
Facklam, R., B. Beall, A. Efstratiou, V. Fischetti, D. Johnson, E. Kaplan, P. Kriz, M. Lovgren, D. Martin, B. Schwartz, A. Totolian, D. Bessen, S. Hollingshead, F. Rubin, J. Scott, and G. Tyrrell. 1999. emm typing and validation of provisional M types for group A streptococci. Emerg. Infect. Dis. 5:247-253.
Facklam, R. F., D. R. Martin, M. Lovgren, D. R. Johnson, A. Efstratiou, T. A. Thompson, S. Gowan, P. Kriz, G. J. Tyrrell, E. Kaplan, and B. Beall. 2002. Extension of the Lancefield classification for group A streptococci by addition of 22 new M protein gene sequence types from clinical isolates: emm103 to emm124. Clin. Infect. Dis. 34:28-38.
Ferretti, J. J., W. M. McShan, D. Ajdic, D. J. Savic, G. Savic, K. Lyon, C. Primeaux, S. Sezate, A. N. Suvorov, S. Kenton, H. S. Lai, S. P. Lin, Y. Qian, H. G. Jia, F. Z. Najar, Q. Ren, H. Zhu, L. Song, J. White, X. Yuan, S. W. Clifton, B. A. Roe, and R. McLaughlin. 2001. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. USA 98:4658-4663.
Gaworzewska, E., and G. Colman. 1988. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol. Infect. 100:257-269.
Hauser, A. R., D. L. Stevens, E. L. Kaplan, and P. M. Schlievert. 1991. Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J. Clin. Microbiol. 29:1562-1567.
Helweg-Larsen, J., J. S. Jensen, T. Benfield, U. G. Svendsen, J. D. Lundgren, and B. Lundgren. 1998. Diagnostic use of PCR for detection of Pneumocystis carinii in oral wash samples. J. Clin. Microbiol. 36:2068-2072.
Henrichsen, J., and T. Rnne. 1992. Reemergence of servere group A streptococcal infections in Denmark 1988-89. Zentbl. Bakteriol. Suppl. 22:8-10.
Hindsholm, M., and H. C. Schonheyder. 2002. Clinical presentation and outcome of bacteraemia caused by beta-haemolytic streptococci serogroup G. APMIS 110:554-558.
Holm, S. E., A. Norrby, A. M. Bergholm, and M. Norgren. 1992. Aspects of pathogenesis of serious group A streptococcal infections in Sweden 1988-89. J. Infect. Dis. 166:31-37.
Jasir, A., A. Tanna, A. Efstratiou, and C. Schalen. 2001. Unusual occurrence of M type 77, antibiotic-resistant group A streptococci in southern Sweden. J. Clin. Microbiol. 39:586-590.
Johansen, K. E. 2000. Group A streptococcal T-protein in invasive and non-invasive isolates in Denmark 1987-98, abstr. P9.10. In Program and abstracts of the 14th Lancefield International Symposia on Severe Streptococcal Diseases, Auckland, New Zealand.
Johnson, D. R., J. T. Wotton, A. Shet, and E. L. Kaplan. 2002. A comparison of group A streptococci from invasive and uncomplicated infections: are virulent clones responsible for serious streptococcal infections J. Infect. Dis. 185:1586-1595.
Kotb, M., A. Norrby-Teglund, A. McGeer, H. El Sherbini, M. T. Dorak, A. Khurshid, K. Green, J. Peeples, J. Wade, G. Thomson, B. Schwartz, and D. E. Low. 2002. An immunogenetic and molecular basis for differences in outcomes of invasive group A streptococcal infections. Nat. Med. 8:1398-1404.
Li, Z., V. Sakota, D. Jackson, A. R. Franklin, and B. Beall. 2003. Array of M protein gene subtypes in 1064 recent invasive group A streptococcus isolates recovered from the active bacterial core surveillance. J. Infect. Dis. 188:1587-1592.
McCormick, J. K., A. A. Pragman, J. C. Stolpa, D. Y. Leung, and P. M. Schlievert. 2001. Functional characterization of streptococcal pyrogenic exotoxin J, a novel superantigen. Infect. Immun. 69:1381-1388.
Musser, J. M., V. Kapur, J. Szeto, X. Pan, D. S. Swanson, and D. R. Martin. 1995. Genetic diversity and relationships among Streptococcus pyogenes strains expressing serotype M1 protein: recent intercontinental spread of a subclone causing episodes of invasive disease. Infect. Immun. 63:994-1003.
O'Brien, K. L., B. Beall, N. L. Barrett, P. R. Cieslak, A. Reingold, M. M. Farley, R. Danila, E. R. Zell, R. Facklam, B. Schwartz, and A. Schuchat. 2002. Epidemiology of invasive group A streptococcus disease in the United States, 1995-1999. Clin. Infect. Dis. 35:268-276.
Proft, T., V. L. Arcus, V. Handley, E. N. Baker, and J. D. Fraser. 2001. Immunological and biochemical characterization of streptococcal pyrogenic exotoxins I and J (SPE-I and SPE-J) from Streptococcus pyogenes. J. Immunol. 166:6711-6719.
Proft, T., S. L. Moffatt, C. J. Berkahn, and J. D. Fraser. 1999. Identification and characterization of novel superantigens from Streptococcus pyogenes. J. Exp. Med. 189:89-102.
Proft, T., S. Sriskandan, L. Yang, and J. D. Fraser. 2003. Superantigens and streptococcal toxic shock syndrome. Emerg. Infect. Dis. 9:1211-1218.
Schmitz, F. J., A. Beyer, E. Charpentier, B. H. Normark, M. Schade, A. C. Fluit, D. Hafner, and R. Novak. 2003. Toxin-gene profile heterogeneity among endemic invasive European group A streptococcal isolates. J. Infect. Dis. 188:1578-1586.
Schwartz, B., R. R. Facklam, and R. F. Breiman. 1990. Changing epidemiology of group A streptococcal infection in the USA. Lancet 336:1167-1171.
Shulman, S. T., R. R. Tanz, W. Kabat, K. Kabat, E. Cederlund, D. Patel, Z. Li, V. Sakota, J. B. Dale, and B. Beall. 2004. Group A streptococcal pharyngitis serotype surveillance in North America, 2000-2002. Clin. Infect. Dis. 39:325-332.
Smoot, J. C., K. D. Barbian, J. J. Van Gompel, L. M. Smoot, M. S. Chaussee, G. L. Sylva, D. E. Sturdevant, S. M. Ricklefs, S. F. Porcella, L. D. Parkins, S. B. Beres, D. S. Campbell, T. M. Smith, Q. Zhang, V. Kapur, J. A. Daly, L. G. Veasy, and J. M. Musser. 2002. Genome sequence and comparative microarray analysis of serotype M18 group A Streptococcus strains associated with acute rheumatic fever outbreaks. Proc. Natl. Acad. Sci. USA 99:4668-4673.
Stromberg, A., V. Romanus, and L. G. Burman. 1991. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J. Infect. Dis. 164:595-598.
Unnikrishnan, M., D. M. Altmann, T. Proft, F. Wahid, J. Cohen, J. D. Fraser, and S. Sriskandan. 2002. The bacterial superantigen streptococcal mitogenic exotoxin Z is the major immunoactive agent of Streptococcus pyogenes. J. Immunol. 169:2561-2569.
Vlaminckx, B., W. van Pelt, L. Schouls, A. Van Silfhout, C. Elzenaar, E. Mascini, J. Verhoef, and J. Schellekens. 2004. Epidemiological features of invasive and noninvasive group A streptococcal disease in The Netherlands, 1992-1996. Eur. J. Clin. Microbiol. Infect. Dis. 23:434-444.
Vlaminckx, B. J., E. M. Mascini, J. Schellekens, L. M. Schouls, A. Paauw, A. C. Fluit, R. Novak, J. Verhoef, and F. J. Schmitz. 2003. Site-specific manifestations of invasive group A streptococcal disease: type distribution and corresponding patterns of virulence determinants. J. Clin. Microbiol. 41:4941-4949.
Working Group on Severe Streptococcal Infections. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 269:390-391.(Kim Ekelund, Peter Skinhj)