Skip Navigation

Prevention of Invasive Group A Streptococcal Disease among Household Contacts of Case Patients and among Postpartum and Postsurgical Patients: Recommendations from the Centers for Disease Control and Prevention

  1. The Prevention of Invasive Group A Streptococcal Infections Workshop Participantsa
  1. Reprints or correspondence: Dr. Matthew R. Moore, Foodborne and Diarrheal Disease Branch, Centers for Disease Control and Prevention, 1600 Clifton Rd., Mailstop D-63, Atlanta, GA 30333 (mmoore4{at}cdc.gov).

Abstract

The Centers for Disease Control and Prevention hosted a workshop to formulate recommendations for the control of invasive group A streptococcal (GAS) disease among household contacts of persons with invasive GAS infections and for responding to postpartum and postsurgical invasive GAS infections. Experts reviewed data on the risk of subsequent invasive GAS infection among household contacts of case patients, the effectiveness of chemoprophylactic regimens for eradicating GAS carriage, and the epidemiology of postpartum and postsurgical GAS infection clusters. For household contacts of index patients, routine screening for and chemoprophylaxis against GAS are not recommended. Providers and public health officials may choose to offer chemoprophylaxis to household contacts who are at an increased risk of sporadic disease or mortality due to GAS. One nosocomial postpartum or postsurgical invasive GAS infection should prompt enhanced surveillance and isolate storage, whereas ⩾2 cases caused by the same strain should prompt an epidemiological investigation that includes the culture of specimens from epidemiologically linked health care workers.

Invasive group A streptococcal (GAS) infection is defined by the isolation of GAS from a normally sterile site (e.g., blood) or by the isolation of GAS from a nonsterile site in the presence of the streptococcal toxic shock syndrome or necrotizing fasciitis [14]. An estimated 8800 cases of invasive GAS disease and 1000 deaths due to invasive GAS infection occurred in the United States in 2000 (3.1 cases and 0.4 deaths per 100,000 population). The overall case-fatality rate for invasive GAS infections is 12%–13% [1, 5, 6]; the rate is highest (30%–80%) among persons with streptococcal toxic shock syndrome [3, 512]. Because of the burden and severity of invasive GAS infection, it is critical to identify opportunities for prevention of this disease.

To discuss new information [5, 13] (K. Robinson, personal communication) and to update previous recommendations [14], the Centers for Disease Control and Prevention (CDC) hosted a meeting in October 2000 to formulate recommendations for the control of the disease among household contacts of persons with invasive GAS infections and to formulate guidelines for responding to postpartum and postsurgical GAS infections. The CDC invited experts in the epidemiology and management of GAS infection and representatives of leading public health and clinical associations. The present report states the positions of the CDC and not the official policy of other organizations.

Prevention of Subsequent Invasive Gas Infections Among Household Contacts of Persons With Invasive Gas Disease

Epidemiological Features of Invasive GAS Infection among Household Contacts of Case Patients

Risk of subsequent invasive GAS disease. For the purposes of the present article, a household contact is defined as a person who spent at least 24 h in the same household as the index patient during the 7 days before the onset of the case patient's symptoms [15]. This definition is intended to distinguish household contacts from other close contacts, such as children in day-care centers or schools or residents of long-term-care facilities. A case of subsequent invasive GAS disease is defined as invasive GAS infection that develops after exposure to a person with a confirmed case. Two studies have systematically estimated the risk of subsequent invasive GAS disease among household contacts of persons with culture-confirmed invasive GAS infections (table 1) [6] (K. Robinson, personal communication). Population-based active surveillance for invasive GAS infections in Ontario, Canada, from January 1992 through June 1995 identified 4 cases of subsequent invasive GAS disease among 1360 household contacts of persons with invasive GAS infections [6]. All 4 subsequent cases occurred among spouses or adult siblings of index patients; there were no deaths.

The second study was performed between January 1997 and May 1999 among a surveillance population of 12.1 million (in Connecticut, Minnesota, and selected counties in Oregon and California) (K. Robinson, personal communication). The investigators identified 1 confirmed case of subsequent invasive GAS disease and 1 probable case (i.e., a culture-negative clinical syndrome consistent with invasive GAS disease) among 1514 household contacts. Both patients with subsequent invasive GAS disease were close adult contacts of the index patients, and neither died. In summary, 2 prospective studies that were designed to identify subsequent cases among household contacts (who were observed for a total of 66.5 million person-years) identified only 5 confirmed cases of subsequent invasive disease (table 1).

Potential burden of chemoprophylaxis. On the basis of these 2 prospective studies, we estimate that 12,000–22,000 household contacts per year would be candidates for chemoprophylaxis. If chemoprophylaxis were 100% effective, 8–64 cases of subsequent invasive GAS infection per year would be prevented in the United States.

Antimicrobial therapy can have undesirable effects, including adverse drug reactions and selection for resistant organisms [16]. One means of minimizing antibiotic use while maximizing its benefit would be to recommend prophylaxis only for those household contacts who are at the highest risk of subsequent invasive GAS infection and/or for those at the highest risk of death from invasive infection. It is not possible to identify such risk factors on the basis of the 5 documented cases of subsequent invasive GAS disease. However, epidemiological studies of invasive GAS infection have identified several risk factors for sporadic disease (table 2) [1, 6, 17] (S. Factor, personal communication). The only risk factor consistently associated with mortality due to sporadic invasive GAS infection is being ⩾65 years of age [1, 5, 6, 17].

Figure 1

Recommended public health response to cases of postpartum and postsurgical group A streptococcal (GAS) disease. HCW, health care worker.

Figure 2

Recommended management for health care workers (HCWs) colonized with group A Streptococcus

Table 2

Factors that increase the risk of sporadic invasive group A streptococcal disease.

Potential Effectiveness of Chemoprophylactic Regimens

No controlled trials have evaluated the effectiveness of chemoprophylaxis in preventing invasive GAS disease among household contacts of persons with invasive GAS infections. In studies of eradication of upper respiratory tract carriage of GAS, intramuscular administration of benzathine penicillin G in combination with orally administered rifampin was the only penicillin-containing regimen that eradicated chronic, asymptomatic pharyngeal carriage of GAS [18]. Orally administered clindamycin is as effective as intramuscular benzathine penicillin G plus rifampin [19, 20]. Orally administered azithromycin was 95% effective in eradicating asymptomatic pharyngeal carriage of GAS among school-aged children [13]. Among US Marine Corps recruits, orally administered azithromycin prevented acquisition of pharyngeal colonization with GAS [21] (table 3).

Table 1

Summary of 2 studies of the risk of subsequent invasive group A streptococcal (GAS) disease among household contacts of case patients.

Table 3

Summary of trials that used antimicrobial agents for the eradication of group A streptococci (GAS) from the upper respiratory tract.

Recommendations for the Prevention of Invasive GAS Disease among Household Contacts of Persons with Invasive GAS Infections

Antimicrobial chemoprophylaxis for any infectious disease is most desirable if disease is severe; if defined risk groups can be identified; and if a safe, affordable, and effective chemoprophylactic regimen is available [23]. Although the risk of subsequent invasive GAS disease among household contacts is higher than the risk among the general population, subsequent invasive GAS infections among household contacts are rare. Given the infrequency of these infections and the lack of a clearly effective chemoprophylactic regimen, the available data do not support a recommendation for routine testing for GAS colonization or for routine administration of chemoprophylaxis to all household contacts of persons with invasive GAS disease. The CDC recommends that health care providers routinely inform all household contacts of persons with invasive GAS disease about the clinical manifestations of pharyngeal and invasive GAS infection (e.g., fever, sore throat, and localized muscle pain) and emphasize the importance of seeking immediate medical attention if contacts develop such symptoms. Studies have suggested that a heightened index of suspicion for subsequent GAS disease should be maintained for 30 days after the diagnosis is made for the index patient [2429] (K. Robinson, personal communication).

Although routine chemoprophylaxis for all household contacts is not recommended, subsequent invasive GAS infections do occur, albeit rarely [6, 2428] (K. Robinson, personal communication). Certain underlying illnesses and other host factors are consistently associated with an increased risk of sporadic invasive GAS infection in persons exposed to the organism [6, 17] (S. Factor, personal communication). Once infected, persons aged ⩾65 years are at increased risk of death [1, 5, 6, 17]. Therefore, although chemoprophylaxis is not recommended routinely for household contacts, health care providers may choose to offer chemoprophylaxis to household members aged ⩾65 years or those at increased risk for sporadic invasive GAS infection (table 2). The CDC does not recommend routine use of culture to identify household contacts who are colonized.

Clustering of asymptomatic carriage of GAS among members of a household is common, and the source of GAS in households is not necessarily the person with invasive GAS infection; therefore, providers who choose to prescribe chemoprophylaxis for an elderly or high-risk member of a household should prescribe chemoprophylaxis for all members of that household. If chemoprophylaxis is prescribed, the CDC recommends any 1 of 3 regimens (table 4). All are appropriate for nonpregnant patients who are not allergic to penicillin. There is limited, indirect evidence that first- and second-generation cephalosporins are effective in eradicating pharyngeal colonization with GAS [31, 32]. Therefore, these agents could be considered for patients allergic to penicillin whose allergic reactions are not anaphylactic [33]. All persons who receive chemoprophylaxis should watch for signs and symptoms of invasive GAS disease for 30 days after the diagnosis of invasive disease in the household contact.

Table 4

Recommended regimens for chemoprophylaxis against group A streptococcal infection.

Although penicillin resistance in GAS has never been described [34], clindamycin resistance occurs rarely. The prevalence of macrolide resistance among invasive strains of GAS varies regionally [35] but remains <8% in most areas [36, 37]. If available, antibiotic susceptibility data should be used to select the most appropriate chemoprophylactic agent.

Prevention of Postpartum and Postsurgical Invasive Gas Infections

Epidemiological Features of Postpartum and Postsurgical Invasive GAS Infections

Burden of infection. A recent study has estimated that ∼220 cases of postpartum invasive GAS disease occur annually in the United States (0.06 cases/1000 live births) [38]. These data likely underestimate the true incidence, because most cases are not invasive [39], and microbiological diagnoses of noninvasive postpartum infections are frequently not available. For example, although some experts recommend that samples of endometrial contents be obtained and cultured during the postpartum period for women suspected of having endomyometritis [40], neither blood nor endometrium cultures are done routinely.

In 2000, 1.5% of cases of invasive GAS disease detected by the CDC's Active Bacterial Core surveillance (ABCs) program were classified as postsurgical cases (i.e., invasive GAS infections that occurred during the first 7 days after surgery). On the basis of 1999 national census data, an estimated 135 postsurgical cases of invasive GAS disease occur in the United States annually (CDC, unpublished data). This estimate is conservative, because GAS infections do occur >7 days after surgery [41, 42].

Descriptive features of investigations of infection clusters. To understand the epidemiology of outbreaks of postpartum and postsurgical GAS infection, workshop participants reviewed reports of postpartum and postsurgical GAS infection clusters published during 1990–1999 (table 5) [4247]. Case definitions required isolation of GAS from case patients but not necessarily from sterile sites. Microbiology laboratories were the primary resource for identification of cases, but other sources included employee records that listed persons who provided preoperative and postoperative care [44] and operating room and medical records [42, 44, 45, 47].

Table 5

Descriptive features of published investigations of clusters of postpartum and postsurgical group A streptococcal infections, 1990–1999.

In 6 [4245, 47] of 8 investigations, an asymptomatic health care worker (HCW) carried the same strain of GAS as did the case patients. Each cluster of infections subsided after effective treatment of the HCW who was the carrier. In most reports, a focused epidemiological investigation limited screening for GAS carriage to HCWs who had contact with case patients. Among epidemiologically linked asymptomatic HCWs, outbreak strains of GAS have been isolated from the throat, the anus, the vagina, and skin lesions [4145, 47, 48] (table 5).

In some instances, household contacts of epidemiologically linked, asymptomatic, colonized HCWs have been shown to be asymptomatic carriers of outbreak strains of GAS [42, 45, 48]. In at least 1 of these investigations, a household contact of the HCW may have served as the reservoir for GAS that led to the outbreak [42]. These carriers may play a role in recolonizing treated HCWs [45, 48].

Recommendations for the Prevention of Postpartum and Postsurgical Invasive GAS Disease

Case definitions. A postpartum case of invasive GAS is defined as isolation, during the postpartum period, of GAS in association with a clinical postpartum infection (e.g., endometritis) or from either a sterile site or a wound infection. To increase the likelihood of identifying nosocomial cases of postpartum GAS infection, the postpartum period of interest includes all inpatient days and the first 7 days after discharge. A case of postsurgical GAS infection is defined as isolation, during the hospital stay or the first 7 days after discharge, of GAS from a sterile site or a surgical wound in a postsurgical patient for whom the indication for surgery was not a preexisting GAS infection. The incubation period of severe GAS infections is usually short (1–3 days [49]), and, therefore, cases that occur >7 days after discharge are more likely to be of community origin. In states where invasive GAS infection is reportable, all cases of invasive postpartum and postsurgical GAS disease should be reported to the local or state health department [50].

Strategies for the investigation of a single case of infection. Given the potential for prevention of additional cases, even 1 case of postpartum or postsurgical GAS infection should prompt an epidemiological investigation by the hospital's infection control personnel (figure 1), which should include enhanced surveillance and storage of GAS isolates from the index patient and any additional cases.

Enhanced surveillance should include one or both of the following: (1) review of microbiology records and autopsy reports from the previous 6 months and/or (2) review of operative, labor and delivery, and medical records from within the hospital. To improve the identification of cases, obstetricians and surgeons should be encouraged, during the ensuing months, to perform appropriate pretreatment cultures for patients suspected of having postpartum or postsurgical infections. Isolates from the index case and any additional cases should be stored for at least 6 months to allow comparison of strains isolated at different times. If laboratory resources are available to type GAS isolates, some infection control practitioners might choose to screen HCWs after the occurrence of a single case of postpartum or postsurgical GAS disease. However, screening of HCWs is not a substitute for enhanced disease surveillance. If infection-control personnel choose to screen HCWs, screening should be considered for HCWs who were present at delivery and for those who performed vaginal examinations before delivery (for postpartum cases) and for all HCWs present in the operating room during surgery and those who changed dressings on open wounds (for postsurgical cases).

If screening of HCWs is undertaken, sites from which specimens should be obtained and cultured include the throat, anus, vagina, and any skin lesions [45]. Screened HCWs may return to work pending the culture results. However, HCWs identified as colonized should be suspended from patient care duties until they have received chemoprophylaxis for 24 h [51].

Strategies for the investigation of ⩾2 cases of infection. If ⩾2 cases are identified within a 6-month period, they may have a common source of GAS transmission. Isolates should be compared using PFGE [52], serotyping, emm typing [53], or other molecular methods. Isolates that differ probably indicate a community source rather than a common source in an HCW [46]. Enhanced surveillance should be initiated, regardless of whether the strains are identical. Identification of 2 cases caused by identical strains should lead to enhanced surveillance and to an investigation of possible epidemiological links between cases.

If 2 cases are found to be caused by the same strain within a 6-month period, screening of HCWs is strongly recommended to prevent further cases of serious infection. For all HCWs epidemiologically linked to the case patients, specimens from the anus, skin lesions, throat, and vagina should be cultured. If no colonized HCW is identified or if HCWs are colonized with strains unrelated to the outbreak strain, the search for colonized HCWs should be broadened to include those HCWs without immediate epidemiological links to all case patients. This might include, for example, HCWs who had direct contact with most but not all of the case patients [42]. The use of standard precautions for infection control should supplement any investigation of postpartum or postsurgical GAS infections.

Management of epidemiologically linked, colonized HCWs. Because most HCWs associated with a given outbreak will not be colonized, HCWs may return to work pending culture results (figure 2). However, colonized HCWs should be suspended from patient care for the first 24 h that they receive chemoprophylaxis [51], and HCW strains should be compared with patient strains by use of the same typing method(s). If an HCW is epidemiologically linked to the case patients and the strain the HCW is carrying is the same as the strains isolated from patients, follow-up cultures should be done for the HCW 7–10 days after the completion of therapy.

In contrast to the recommendations for household contacts of persons with invasive GAS disease, it is recommended that HCWs included in investigations of postpartum or postsurgical GAS infection clusters have cultures performed if results of follow-up cultures from the implicated HCW remain positive 7–10 days after the completion of therapy [42, 45, 48]. For children who are household contacts of the HCW, cultures of specimens from the throat and skin lesions should be performed. For adult household contacts, cultures of specimens from the throat, any skin lesions, the anus, and the vagina should be performed. Colonized household contacts of an HCW implicated in an outbreak should receive chemoprophylaxis.

Treatment of epidemiologically linked HCWs carrying outbreak strains. Treatment options for asymptomatic colonized HCWs have not been rigorously studied. Previous studies [13, 1820] have suggested that 3 regimens may be effective (table 4). Any of these regimens is appropriate for nonpregnant HCWs who are not allergic to penicillin and for their colonized household contacts. Clindamycin or azithromycin is recommended for HCWs and colonized household contacts who are allergic to penicillin. Rectal carriage of GAS is difficult to eradicate with penicillin-based regimens [43, 48, 5456]. Oral therapy with vancomycin in combination with rifampin has been recommended in such cases [57, 58]; however, no controlled trials have been done to support this recommendation. Given the well-documented effects of clindamycin on bowel flora, oral clindamycin is recommended for the treatment of HCWs and their household contacts who have rectal carriage of GAS. If azithromycin or clindamycin is used, susceptibility testing of the HCW strain of GAS against macrolides and clindamycin should be performed.

Summary

Although the risk of subsequent invasive GAS disease among household contacts of persons with invasive GAS infections is higher than the risk among the general population, subsequent invasive GAS infections are rare. As a result, routine screening for colonization with GAS and routine administration of chemoprophylaxis against GAS are not recommended for household contacts of index patients. However, because of the increased risk of sporadic invasive GAS disease among certain groups (table 2) and the risk of death among persons aged ⩾65 years who develop invasive GAS disease, health care providers may choose to offer chemoprophylaxis to members of the households of patients with invasive GAS infection that include persons aged ⩾65 years or other high-risk individuals (table 4). Performance of cultures is not recommended for potential GAS carriers.

Postpartum and postsurgical GAS infections frequently are undetected and are underreported, and some of these cases present opportunities for prevention. Enhanced surveillance should be implemented after identification of a single case of postpartum or postsurgical GAS infection, and all GAS isolates from suspected cases should be stored and compared by serotyping or molecular techniques. Some infection control practitioners might choose to screen HCWs after a single case has occurred, but this should not interfere with the conduct of enhanced surveillance. The occurrence of ⩾2 cases of infection with the same GAS type within a 6-month period suggests that an HCW might be the source of the cluster; therefore, screening of HCWs who are epidemiologically linked to the case patients is strongly recommended (figures 1 and 2).

It is important to indicate the settings that are not included in the present recommendations. Invasive GAS infections and clusters of noninvasive infections (e.g., pharyngitis or uncomplicated cellulitis) that occur in day-care centers, schools, military training facilities, and nursing homes present unique challenges. These recommendations do not apply to those settings.

The workshop participants identified several research issues that deserve attention. These include antibiotic treatment of GAS carriage in adults, the prevalence of macrolide and clindamycin resistance among isolates of GAS, and the effectiveness of antimicrobial agents for the eradication of rectal carriage of GAS. Finally, because subsequent GAS disease accounts for <1% of all invasive cases and also because preventable postpartum and postsurgical infections constitute a small portion of disease burden, primary prevention of GAS infections remains a research priority. The development of GAS vaccines may offer the ultimate solution to this problem.

The Prevention of Invasive Group A Streptococcal Infections Workshop Participants

The participants in the workshop were as follows, in alphabetical order: Bernard Beall (Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, Centers for Disease Control and Prevention [CDC], Atlanta, Georgia), John Besser (Minnesota Department of Health, Minneapolis), Alan Bisno (Miami Veteran's Affairs [VA] Medical Center and University of Miami School of Medicine, FL), Ilin Chuang (Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, CDC), Allen S. Craig (Tennessee Department of Health, Nashville; representative, Active Bacterial Core Surveillance program), Richard Facklam (Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, CDC); Janice Fetter (Northside Hospital, Atlanta, GA; representative, Association for Professionals in Infection Control and Epidemiology), Michael A. Gerber (Children's Hospital Medical Center, Cincinnati, OH; representative, American Academy of Pediatrics), Gregory Gray (University of Iowa College of Public Health, Iowa City; representative, US Navy [retired]), Harry Hill (University of Utah School of Medicine, Salt Lake City), Lisa Lepine (Emory University School of Medicine, Atlanta, GA), Orin Levine (National Institute for Allergy and Infectious Diseases, Bethesda, MD), Allison McGeer (Mt. Sinai Hospital, and Laboratory Medicine and Public Health Sciences, University of Toronto, ON, Canada), Matthew Moore (Epidemic Intelligence Service Program, Division of Applied Public Health Training, Epidemiology Program Office, and Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, CDC), Michele Pearson (Division of Healthcare Quality Promotion, CDC), Katherine O'Brien (Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD), Anne Schuchat (Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, CDC), Mack Sewell (New Mexico Department of Health, Santa Fe; representative, Council of State and Territorial Epidemiologists), Stanford Shulman (Children's Memorial Hospital, and Department of Pediatrics, Northwestern University, Chicago, Illinois), Jane Siegel (University of Texas Southwestern Medical Center, Dallas, TX; representative, Healthcare Infection Control Practices Advisory Committee), Dennis L. Stevens (Boise VA Medical Center, ID; and University of Washington School of Medicine, Seattle), Larry Strausbaugh (Portland VA Medical Center and Oregon Health and Sciences University School of Medicine, Portland; representative, Infectious Diseases Society of America), and Chris Van Beneden (Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, CDC).

acknowledgments

We gratefully acknowledge the contributions of Katherine Robinson.

Footnotes

  • a Workshop participants are listed at the end of the text.

  • This article presents the findings of the Prevention of Invasive Group A Streptococcal Infections Workshop held at the Centers for Disease Control and Prevention in October 2000.

  • Financial support: Centers for Disease Control and Prevention.

  • Received March 12, 2002.
  • Revision received May 28, 2002.

References

| Table of Contents