点击显示 收起
The Department of Laboratory Medicine and Pathology, The University of Alberta
The National Centre for Streptococcus, The Provincial Laboratory for Public Health (Microbiology)
Alberta Health and Wellness, Edmonton, Alberta, Canada
ABSTRACT
Invasive group A streptococcal (iGAS) disease was placed under surveillance in Alberta in August 1999. The purpose of this study was to determine the incidence rates of iGAS infections throughout Alberta over a 3-year period (2000 to 2002) and to better understand the epidemiology of iGAS in this province. There were a total of 441 cases of invasive GAS disease over the 3 years examined (average population over 3 years, 3,055,765) and 47 deaths. The incidence in Alberta was 5.0 (2000), 5.7 (2001), and 3.8 (2002) per 100,000. The two main metropolitan regions (Edmonton and Calgary) had the majority of iGAS disease cases (305 cases), producing incidence rates of 4.8 (Edmonton) and 6.9 (Calgary) in 2000, 6.9 (Edmonton) and 6.6 (Calgary) in 2001, and 4.1 (Edmonton) and 3.9 (Calgary) in 2002, as well as deaths attributable to GAS (31 deaths). The three most prevalent M types were M1 (71 cases), M3 (52 cases), and MPT2967 (44 cases). With respect to age, the highest incidence rates occurred in those less than 1 year old (11.7 per 100,000) and those 65 years or older (11.5 per 100,000). Varicella virus infection preceded iGAS disease in 25% of children 8 years of age and under. A seasonal association was observed during the 3 years studied, with the highest number of cases occurring in the winter months and the lowest occurring during the summer months. The data for years 2000 and 2001 show that the metropolitan regions of Alberta experienced some of the highest incidence rates reported in North America in the past decade.
INTRODUCTION
Invasive group A streptococcal (iGAS) disease has received increased attention worldwide in the last 20 years. This is due primarily to an increase in all forms of iGAS disease, most notably necrotizing fasciitis (NF) and streptococcal toxic shock syndrome (STSS) (3, 8, 10, 12, 22, 27). The increase in awareness of iGAS disease has lead to a better understanding of the epidemiology of GAS infections in general (4, 6, 12, 18, 21, 23, 28).
One facet of iGAS disease that has been documented by other investigators in North America is the gradual increase in the incidence rates of iGAS disease (4, 6, 12, 17, 18, 21, 23, 28). These increasing incidence rates and a perceived high rate of iGAS disease in the Calgary Health region in 1998 and 1999 prompted Alberta Health and Wellness to place iGAS disease under surveillance in August 1999. The data collected since that time provided an opportunity to examine the epidemiology of iGAS disease in Alberta.
The purpose of this study was to determine the incidence rates of iGAS infections throughout Alberta for all cases of invasive disease from 2000 to 2002 and to better understand the epidemiology of iGAS in this province during this time period. This information was correlated with the microbiologic data for GAS strains isolated from those cases.
MATERIALS AND METHODS
Data collection. iGAS disease was placed as a disease under surveillance in Alberta in August 1999. This resulted in all cases of iGAS disease in Alberta being reported to the public health official for the region in which the case occurred. The iGAS data presented for the purposes of this study encompass calendar years 2000 to 2002. Alberta, during the period reported, consisted of 17 Health regions, 2 metropolitan and 15 rural.
The surveillance case definition for iGAS infection was the isolation of GAS from a normally sterile site (e.g., blood, cerebrospinal fluid, joint fluid, pleural fluid, or pericardial fluid) with or without clinical evidence of invasive disease. A case was generated when GAS, isolated from a normally sterile site, was reported by the isolating laboratory to the regional public health official. The GAS isolate was then forwarded to the National Centre for Streptococcus (NCS) in Edmonton for M typing and antimicrobial susceptibility assays. To collect information on patients with iGAS disease, the public health official for the region was required to consult with the attending physician to collect clinical information, to report the case by fastest means possible to the Provincial Health Officer, and to ensure that the Notifiable Disease Report was completed and submitted to provincial public health within 7 days.
To adequately ensure that no cases of iGAS were missed, once a year a representative from provincial public health would meet with a representative from the NCS to review all cases for the year that had been documented to date by each agency. If cases had been missed by either agency, follow-up was initiated with the laboratory that had initially isolated and reported the GAS and/or with the attending physician associated with a missed case.
Strain characterization. Isolates of iGAS forwarded to the NCS were M typed as previously described (24). M and T typing and opacity factor determination were performed on all isolates by standardized methods (11). All antisera were prepared in-house. Due to the difficulty of producing antisera for some M types, antiopacity factor typing was used to predict the following M types: 9, 11, 25, 28, 48, 77, 78, and 92. All isolates were assayed for susceptibility to penicillin, vancomycin, chloramphenicol, erythromycin, and clindamycin by using the disk diffusion assay method as per NCCLS guidelines (16).
RESULTS
iGAS incidence rates. From 1 January 2000 to 31 December 2002, 441 cases of iGAS disease were identified in the province of Alberta (population, 3,055,765). There were no discernible clusters of iGAS disease noted during this time. The annual incidence rates for Alberta for each of years 2000, 2001, and 2002 were 5.0, 5.7, and 3.8, respectively (Table 1). The larger metropolitan areas (region 4, Calgary, population 998,277; region 10, Edmonton, population 880,569) experienced a greater number of cases than regions encompassing the rural areas (all other regions = rural, population 1,176,919) (Table 1). The highest incidence rates for these metropolitan regions were 6.9 per 100,000 (region 4 in 2000 and 2001) and 6.6 per 100,000 (region 10 in 2001). The highest incidence of disease over the 3 years surveyed occurred in children <1 year of age (11.7 per 100,000) and in individuals 65 years of age and older (11.5 per 100,000) (Fig. 1.).
Overall mortality for patients with iGAS in the province was 10.7, 13.2, and 6.8% for the years 2000, 2001, and 2002, respectively. The median ages for deaths associated with iGAS were 55.2 years (2000), 54.6 years (2001), and 64.5 years (2002). Fifty-three percent of all deaths occurred in individuals 65 years of age or older. There were five child deaths occurring in patients aged 1 month, 11 months, 12 months, 20 months, and 9 years. No single serotype predominated in cases where the outcome was death.
Clinical diagnosis. Bacteremia only was the most common clinical diagnosis, accounting for 25% of cases of iGAS infection (Table 2). This was followed by soft tissue infections (19%) and cellulitis (17%) (Table 2). There were 36 cases of STSS during the 3-year survey period, for an incidence rate of 1.2 per 100,000 (8.2% of all cases) (Table 2). STSS was associated with 34% of all deaths. The average age of cases with STSS for males was 47 years and for females it was 50 years. The overall median age was 48.5 years (age range, 1 to 89 years). Fifty-three cases presented with NF, for an incidence rate of 1.7 per 100,000 (12% of all cases) (Table 2). NF was associated with 19% of all deaths. The average age for males with NF was 42.5 years and for females it was 46.5 years. The overall median age for NF was 45 years (age range, 6 to 88 years). Examination of the M types associated with these two clinical entities for each of the 3 years studied showed that M1 was the most common M type associated with STSS (30.5% of cases) and NF (26.4% of cases) (Table 3). Interestingly, there were no cases of STSS in 2002 caused by M1 and there was a decrease in the M1 cases of NF in 2002 (Table 3).
Pneumonia was diagnosed in 46 patients (10.4%), with the majority being in patients 41 to 64 years of age (18 cases) or 65 years and older (20 cases). The incidence of GAS-associated pneumonia in 2000 was 0.74 per 100,000, in 2001 it was 0.33 per 100,000, and in 2002 it was 0.45 per 100,000. No single M type predominated in cases of pneumonia except for six M1 isolates in 2000. M1 was more likely to be associated with pneumonia than any other M type (10 of 46 [21.7%] versus 71 of 433 [16.4%]). All other M types numbered three or less each year. The large number of cases of GAS pneumonia in 2000 occurred predominantly in region 4 in 2000 (13 cases in 2000 versus 3 cases in 2001 and 4 cases in 2002).
There were 14 cases of postpartum iGAS infection (3.2% of 441 cases), for an incidence of 12.4 per 100,000 or 0.12 in 1,000 live births (total of 112,617 live births in 2000 to 2002) (Table 2) (1). Three cases occurred in 2000, four occurred in 2001, and seven occurred in 2002. Of these 14 cases, 4 were M28 (29%), 2 were M2 (14%), 2 were M77 (14%), and M1, M3, M4, M12, and MPT2967 were each single isolates (7%).
Underlying risk factors. Underlying risk factor information was reported for 117 (26.5%) of the 441 cases (Table 4). The top four risk factors were institutional acquisition, injection drug use, pregnancy, and varicella virus infection. Institutional acquisition was the most common reported underlying risk factor (17.1%). This category included patients whose onset of illness occurred in a hospital, nursing home, or extended care facility. There was no association with a specific M type for this category (>10 M types isolated from 20 cases). Cases involving injection drug use (12.8%) were in patients who ranged in age from 28 to 43 years old and were not associated with any single M type. Pregnancy-related risk factors also accounted for 12.8% of known cases reporting risk factors and were all related to postpartum iGAS infection as discussed previously. There were 14 cases of iGAS that presented with a history of chicken pox (3.2% of all iGAS cases and 25% of those in children 8 years of age). Eight were male. Ages ranged from 1 year to 11 years (average, 4.8 years), with one patient aged 79 years. Six cases were attributed to serotype M1, four were attributed to M3, two were attributed to M4, and one each was attributed to M6 and M12. All cases presented with bacteremia.
With respect to seasonal association of iGAS disease, cases occurred throughout the year but were more frequently seen during the winter and spring months (Fig. 2).
GAS isolate characterization. GAS isolates were available for characterization for 433 of the 441 of the cases. The most common M type for each of the 3 years was M1, accounting for 19.6% in 2000, 14.8% in 2001, and 14.7% in 2002 (Table 5). M1 is the most common M type seen in Canada overall (24). For some M types, the percentages varied substantially from year to year. For example, MPT2967 accounted for approximately 19% of GAS isolates in 2000 but decreased to 7.1% (2001) and 3.4% (2002). M3 accounted for 7.4% in 2000 and doubled in 2001 to 15.4%, with a slight decrease to 13% in 2002. Also of note was M28, which increased from 0.6% (2001) to 11.2% (2002), and M11, which decreased to 1.8% (2001) from 7.4% (2000) (Table 5).
All iGAS isolates were susceptible to penicillin and vancomycin. Sixty-five (15%) isolates were resistant to erythromycin. One isolate was resistant to clindamycin, and one was resistant to chloramphenicol. Of the 65 erythromycin-resistant isolates, 41 were MPT2967, 13 were M11, 6 were M58, and 1 each was M1 and M77. The single clindamycin-resistant isolate (M77) was also resistant to erythromycin. The single chloramphenicol-resistant isolate was an M12. Prevalence of erythromycin resistance in Alberta correlated with the prevalence of MPT2967, an M type that is typically resistant to this antibiotic. While erythromycin resistance is not limited to this M type, the overall erythromycin resistance rate for this collection paralleled the decreasing numbers of MPT2967 over the 3-year surveillance period. In 2000 erythromycin resistance for iGAS was 23.6%, decreasing to 11.2% in 2001 and to 5.1% in 2002.
DISCUSSION
The present study provides an overview of iGAS infections in Alberta from 2000 to 2002. The incidence rates of iGAS disease in the metropolitan areas of Alberta are some of the highest rates of iGAS disease in North America: 6.6 (2001 in Edmonton) and 6.9 (2000 and 2001 in Calgary) per 100,000. In comparison, a population-based survey for iGAS in the United States from 1995 to 1999 reported an incidence rate of 3.5 per 100,000 (17). Other studies have reported incidence rates of 4.1 per 100,000 in the San Francisco region (1989 to 1999) and incidence rates of 1.7 to 2.5 from 1993 to 1996 and 2.9 in 1996 and 1997, respectively, in Sweden (18, 28). With respect to other published Canadian rates, Davies et al. reported iGAS incidence rates of 1.3 per 100,000 in 1992 and 1.7 per 100,000 in 1993 in Ontario (6).
It is important to note that the high incidence in the metropolitan regions did not remain elevated throughout the surveillance period but decreased in 2002. Passaro et al. suggested that the incidence of iGAS in San Francisco peaks every 5 years (18). This may also be occurring in regions 4 and 10 in Alberta, with 2000 and 2001 possibly representing the peak incidence years. Continued surveillance should answer this question.
Why the incidence rates in the metropolitan regions of Alberta would be higher than in the overall rural population is unclear. It maybe due to closer contact of individuals in an urban setting in comparison to a rural setting. Another possible explanation for the increased incidence of disease may be due to the increase in awareness among primary care physicians in the health regions resulting in better reporting. It is also plausible that the high rates are the result of a new strain (M type) entering a region of susceptible individuals, followed soon after by its disappearance as a population develops immunity. However, the higher number of cases in 2001 does not correspond to an increase in any previously unseen M type in the region (Table 4). Rather, what is seen is elevated numbers of M types present the year before at lower rates, e.g., M3 at 7.4% (2000) and 15.4% (2001) and M4 at 2.0% (2000) and 6.5% (2001). These rates were then followed by a gradual decline in 2002. This fluctuation was reversed for other M types, e.g., M11 at 7.4% (2000) and 1.8% (2001) and M28 at 4.7% (2000) and 0.6% (2001). This decrease in incidence of a specific M type was most dramatically illustrated by MPT2967, which accounted for almost 19% of all M types in 2000 and dropped to 3.4% in 2002.
Also of note is the high rate of NF seen during the 3-year period. The rate of 12% is double that reported by both Davies et al. for Ontario (6) and by Factor et al. for metropolitan Atlanta, Baltimore, and Toronto in 1997 (6%) (9). The reason for the higher rate of NF in Alberta than elsewhere in North America is uncertain. It is possible that cases of NF were overreported during the surveillance period. We believe that this is unlikely, as the rates of STSS and overall iGAS incidence were also higher than those reported in other studies, suggesting that the high rate of NF in Alberta is real and not a reporting artifact.
The ages most greatly affected by iGAS are very similar to those reported in other studies, with the majority of iGAS cases occurring in the 1-year-old group and those 70 years of age and older. Also, there is a small increase in incidence seen in the middle-age population, 25 to 44 years of age. We have speculated in the past that the cases in the 25-to-44-year age group may be the result of household contacts, e.g., parents looking after small children or elderly with GAS infections, although this is still unclear (24). It has been documented that the risk for subsequent iGAS disease in household contacts is higher than the risk among the general population (6, 20). Still, these infections, while present, have been shown to be relatively infrequent, suggesting there are other explanations for the elevated incidences seen in the 25-to-44-year-old age group (20).
The tropism of specific M types for certain disease presentations is interesting. M28 accounted for close to 30% of all postpartum iGAS cases. This is consistent with the findings of Chuang et al., who also reported emm type 28 as the most prevalent type associated with all postpartum cases associated with iGAS (21%) (4). This was also noted in the United Kingdom, in which 26% of puerperal infections were due to M28 (5). In addition, emm 28 has also been reported in three outbreaks of postpartum iGAS (14, 15, 25).
The predominance of a single serotype for a single disease presentation was also noted for those patients with varicella virus infection. In children with a history of chicken pox and iGAS infection, M1 accounted for 43% of these cases. The association with iGAS and varicella virus has been well documented (2, 6, 19). In July 2001, Alberta began a universal infant varicella immunization program that should result in a decrease in iGAS infections associated with varicella virus in the coming years.
The results also show that the majority of erythromycin-resistant isolates were derived from one M type, MPT2967. It is probable that these isolates represent a single clone that entered the province in 2000 or earlier and is gradually decreasing in prevalence as the Alberta population most likely becomes immune to this serotype. While MPT2967 accounted for the majority of erythromycin-resistant isolates, the single clindamycin-resistant isolate, an M77, did not result in as many cases of iGAS disease as MPT2967.
The predominance of one M type expressing an erythromycin resistance phenotype in a defined geographical area in North America at a specific point in time is not unusual. Martin et al. reported the presence of an erythromycin-resistant M6 clone in throat cultures taken from school-age children in Pittsburgh (13). Erythromycin-resistant GAS in this setting accounted for 48% of the GAS isolates recovered. In addition, Weiss et al. reported the predominance of an erythromycin-resistant M28 (65% of erythromycin-resistant isolates were M28) in patients presenting with pharyngitis in 1998, and in Ontario in 1997 de Azavedo found that an M4 serotype accounted for 28% of the erythromycin-resistant strains studied (7, 26). What is interesting in our study is the observation that a particular erythromycin-resistant M type, e.g., MPT2967, can predominate early on and gradually diminish in prevalence over a period of 2 to 3 years, similar to what is seen for other non-erythromycin-resistant M types. Based on these observations, while erythromycin resistance may be high in a given year, it is possible that this will gradually decrease over the next 2 to 3 years provided no other erythromycin-resistant GAS strain enters the region.
In conclusion, the placement of iGAS as a disease under surveillance has allowed public health officials in Alberta to gain an understanding of the significance of iGAS infections in this province. Based on our results, there is significant morbidity and mortality associated with iGAS in Alberta in comparison to other regions in North America. Over the 3 years surveyed, an average of 144.3 cases occurred per year in a population of just over 3 million people. Continued enhanced surveillance of iGAS infections is important to assess and monitor the relative magnitude of iGAS disease in Alberta in relation to other diseases over the next few years. It is hoped that the information in this report will aid in providing a basis for rational vaccine development and will contribute to the implementation and evaluation of other intervention strategies for controlling this disease in Alberta and elsewhere.
ACKNOWLEDGMENTS
We are grateful to the analysis staff at Alberta Health and Wellness and the Regional Communicable Disease Nurses. We are also indebted to the Regional Medical Officers of Health in each region, microbiology laboratories, and physicians whose time and effort continue to provide important information regarding surveillance of GAS in Alberta.
Financial support for this work was from Alberta Health and Wellness, the Provincial Laboratory for Public Health (Microbiology), and the National Centre for Streptococcus.
We indicate no conflicts of interest.
REFERENCES
Anonymous. 2004. Vital statistics data. Alberta Government Services, Alberta, Canada. [Online.] http://www3.gov.ab.ca/gs/information/vs/statistical_info.cfm.
Centers for Disease Control and Prevention. 1997. Outbreak of invasive group A streptococcus associated with varicella in a childcare center—Boston, Mass. Morb. Mortal. Wkly. Rep. 46:944-949.
Chelsom, J., A. Halstensen, T. Haga, and E. A. Hoiby. 1994. Necrotising fasciitis due to group A streptococci in western Norway: incidence and clinical features. Lancet 344:1111-1115.
Chuang, I., C. Van Beneden, B. Beall, A. Schuchat, et al. 2003. Population-based surveillance for postpartum invasive group A streptococcus infections, 1995-2000. Clin. Infect. Dis. 35:665-670.
Colman, G., A. Tanna, A. Efstratiou, and E. T. Gaworzewska. 1993. The serotypes of Streptococcus pyogenes present in Britain during 1980-1990 and their association with disease. J. Med. Microbiol. 39:165-178.
Davies, H. D., A. McGeer, B. Schwartz, K. Green, D. Cann, A. Simor, D. E. Low, et al. 1996. Invasive group A streptococcal infections in Ontario, Canada. N. Engl. J. Med. 335:547-554.
de Azavedo, J., R. H. Yeung, D. J. Bast, C. L. Duncan, S. B. Borgia, and D. E. Low. 1999. Prevalence and mechanism of macrolide resistance in clinical isolates of group A streptococci from Ontario, Canada. Antimicrob. Agents Chemother. 43:2144-2147.
Eriksson, B., M. Norgen, 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.
Factor, S. H., O. S. Levine, B. Schwartz, L. H. Harrison, M. M. Farley, A. McGeer, and A. Schuchat. 2003. Invasive group A streptococcal disease: risk factors for adults. Emerg. Infect. Dis. 9:970-977.
Hoge, C. W., B. Schwartz, D. F. Talkington, R. F. Breiman, E. M. MacNeill, and S. J. Englender. 1993. The changing epidemiology of invasive group A streptococcal infections and the emergence of streptococcal toxic shock-like syndrome: a retrospective population-based study. JAMA 269:384-389.
Johnson, D. R., J. Sramek, E. I. Kaplan, R. Bicova, J. Havlicek, H. Havlickova, J. Motlova, and P. Kriz. 1996. Laboratory diagnosis of group A streptococcal infections. World Health Organization, Geneva, Switzerland.
Kaul, R., A. McGeer, D. E. Low, K. Green, B. Schwartz, the Ontario Group A Streptococcal Study Group, and A. E. Simor. 1997. Population-based surveillance for group A streptococcal necrotizing fasciitis: clinical features, prognostic indicators and microbiologic analysis of seventy-seven cases. Am. J. Med. 103:18-24.
Martin, J. M., M. Green, K. A. Baradora, and E. R. Wald. 2002. Erythromycin-resistant group A streptococci in schoolchildren in Pittsburgh. N. Engl. J. Med. 346:1200-1206.
Mastro, T. D., T. A. Farley, J. A. Elliott, R. R. Facklam, J. R. Perks, J. L. Hadler, R. C. Good, and J. S. Spika. 1990. An outbreak of surgical-wound infections due to group A streptococcus carried on the scalp. N. Engl. J. Med. 323:968-972.
McGregor, J., A. Ott, and M. Villard. 1984. An epidemic of "childhood fever." Am. J. Obstet. Gynecol. 150:385-388.
National Committee for Clinical Laboratory Standards. 2000. Performance standards for antimicrobial disk susceptibility tests, 7th ed. Approved standard M2-A7. Wayne, Pa.
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.
Passaro, D. J., D. S. Smith, E. C. Hett, A. L. Reingold, P. Daily, C. A. Van Beneden, and D. J. Vugia. 2002. Invasive group A streptococcal infections in the San Francisco Bay area, 1989-99. Epidemiol. Infect. 129:471-478.
Peterson, C. L., D. J. Vugia, H. B. Meyers, S. M. Chao, J. Vogt, J. Lanson, P.A Brunell, K. S. Kim, and L. Mascola. 1996. Risk factors for invasive group A streptococcal infections in children with varicella: a case-control study. Pediatr. Infect. Dis. J. 115:151-156.
Robinson, K. A., G. Rothrock, Q. Phan, B. Sayler, K. Stefonek, C. Van Beneden, O. S. Levine, et al. 2003. Risk for severe group A streptococcal disease among patients' household contacts. Emerg. Infect. Dis. 9:443-447.
Sharkawy, A., D. E. Low, R. Saginur, D. Gregson, B. Schwartz, P. Jessamine, K. Green, A. McGeer, et al. 2002. Severe group A streptococcal soft tissue infections in Ontario: 1992-1996. Clin. Infect. Dis. 34:454-460.
Stevens, D. L. 1994. Invasive group A streptococcal infections: the past, present and future. Pediatr. Infect. Dis. J. 13:561-566.
Svensson, N., S. Oberg, B. Henriques, S. Holm, G. Kallenues, V. Romanus, and J. Giesecke. 2000. Invasive group A streptococcal infections in Sweden in 1994 and 1995. Epidemiology and clinical spectrum. Scand. J. Infect. Dis. 32:609-614.
Tyrrell, G. J., M. Lovgren, B. Forwick, N. Hoe, J. M. Musser, and J. A. Talbot. 2002. M types of group A streptococcal isolates submitted to the National Centre for Streptococcus (Canada) from 1993 to 1999. J. Clin. Microbiol. 40:4466-4471.
Viglionese, A., V. F. Nottebart, H. A. Bodman, and R. Platt. 1991. Recurrent group A streptococcal carriage in a health care worker associated with widely separated nosocomial outbreaks. Am. J. Med. 91(Suppl.):S329-S333.
Weiss, K., J. de Azavedo, C. Restieri, L. A. Galarneau, M. Gourdeau, P. Harvey, J. F. Paradis. K. Salim, and D. E. Low. 2001. Phenotypic and genotypic characterization of macrolide-resistant group A streptococcus strains in the province of Quebec, Canada. J. Antimicrob. Chemother. 47:345-348.
Working Group on Severe Streptococcal Infections. 1993. Defining the group A streptococcal toxic shock syndrome: rationale and consensus definition. JAMA 269:390-391.
Zurawski, C., M. S. Bardsley, B. Beall, J. A. Elliot, R. R. Facklam, B. Schwartz, and M. M. Farley. 1998. Invasive group A streptococcal disease in metropolitan Atlanta: a population-based assessment. Clin. Infect. Dis. 27:150-157.