Group A streptococcal puerperal sepsis: initial characterization of virulence factors in association with clinical parameters

Journal of Reproductive Immunology 82 (2009) 74–83

Group A streptococcal puerperal sepsis: initial characterization of virulence factors in association with clinical parameters Janice L.B. Byrne a , Kjersti M. Aagaard-Tillery b,∗ , Jason L. Johnson c , Larry J. Wright d , Robert M. Silver a a

Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, University of Utah Health Sciences Center, Salt Lake City, UT, USA b Baylor College of Medicine, Division of Maternal-Fetal Medicine, Houston, TX, USA c Departments of Obstetrics and Gynecology at LDS Hospital, Salt Lake City, UT, USA d Department of Pathology at LDS Hospital, Salt Lake City, UT, USA Received 19 October 2008; received in revised form 30 March 2009; accepted 10 June 2009

Abstract Group A ␤-hemolytic streptococcus (GAS) is an uncommon but potentially fatal source of postpartum infection. Pathogenesis in invasive GAS infections has been linked to bacterial virulence factors. In this study, we sought to provide an initial description of potential virulence factors in association with puerperal morbidity by virtue of specific M-protein type antigens. Women with confirmed GAS puerperal infection in the Salt Lake City region were prospectively identified over a 6-year interval (1991–1997). From this cohort, GAS isolates were analyzed with respect to M-serotype and presence of genes encoding the Streptococcal Pyogenic Exotoxins A and B (SPE-A and SPE-B). Bacterial isolates from 18 subjects with GAS puerperal infection underwent M-serotyping and PCR-based genotyping for the speA and speB genes. Among these, 8/18 subjects manifest criteria of severe disease. All 18 isolate strains expressed speB; 6/18 isolates expressed speA. Of the M-serotypes, 8/8 severe disease isolates expressed M-types 1 (N = 3) or 28 (N = 5). Pulse-field gel electrophoresis did not indicate an outbreak strain among similar isolates. We conclude that in this initial characterization, morbidity among women with GAS puerperal infection is associated with M-types 1 and 28, but not speB genotype. © 2009 Elsevier Ireland Ltd. All rights reserved. Keywords: Invasive group A streptococcus; Puerperal sepsis; Infection in pregnancy; Streptococcal M-protein; Streptococcal pyrogenic exotoxins

1. Introduction Over the past few decades there has been a dramatic increase in the number of serious group A streptococcal (GAS) infections reported in Western countries (Abouzeid et al., 2005; Stevens et al., 1989; Obrien et al., 2002; Carapetis et al., 2005; Chuang et al., 2002). ∗

Corresponding author at: Baylor College of Medicine, Department of Obstetrics & Gynecology, Division of Maternal-Fetal Medicine, One Baylor Plaza, Houston, TX 77030, USA. Tel.: +1 713 798 8467. E-mail address: [email protected] (K.M. Aagaard-Tillery).

The reemergence of this age-old pathogen has been widely publicized in the lay media as the “flesh-eating bacteria”. However, the clinical spectrum associated with puerperal group A streptococci infection actually ranges from mild endometritis that quickly resolves with antibiotic therapy, to invasive GAS. Major morbidity and mortality associated with invasive puerperal GAS includes necrotizing fasciitis and myositis requiring surgical debridement, and fulminate streptococcal toxic shock syndrome (STSS) similar to the toxic shock syndrome caused by Staphylococcus aureus (Stevens et al., 1989; Obrien et al., 2002; Carapetis et al.,

0165-0378/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jri.2009.06.126

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

2005; Chuang et al., 2002; Alnaes-Katjavivi and Kahn, 2005). Group A streptococci is similarly responsible for nonpuerperal suppurative infections both local (pharyngitis and pyoderma) and invasive (bacteremia, necrotizing fasciitis, myositis, and scarlet fever). Efforts to better understand factors which modulate the clinical spectrum of GAS disease have focused on variations in expression of bacterial virulence factors. One such major virulence factor of Streptococcus pyogenes is the M-protein type antigen (M-type), which is encoded by the emm gene. Anchored to the cell membrane, the cell-bound proximal region is highly conserved among isolates, with marked type-specific variation localizing to antigenic epitopes (MacArthur and Walker, 2006). There are 86 different antigenic subtypes of M-types which allow group A streptococcus to thwart the immune system by repelling phagocytes and inhibiting activation of the alternate complement cascade (MacArthur and Walker, 2006). Each of these M-types can be identified and classified either by their emm gene sequencing, or serotyping; ergo, emm sequencing serves as a specific typing method (Carapetis et al., 2005; Chuang et al., 2002; AlnaesKatjavivi and Kahn, 2005; MacArthur and Walker, 2006; Vlaminckx et al., 2005). Exotoxins constitute another major virulence factor for group A streptococcus. Streptococcal pyrogenic exotoxins (SPEs) induce the release of a variety of inflammatory mediators such as tumor necrosis factor␣, interleukin-1-␤, and interleukin-6 (Chatellier et al., 2000). In turn, these cytokines mediate fever, shock, and tissue injury. Also, several of the group A streptococcal exotoxins function as superantigens and thereby result in a fulminant STSS (Stevens et al., 1989; Obrien et al., 2002; Carapetis et al., 2005; Chuang et al., 2002; AlnaesKatjavivi and Kahn, 2005; MacArthur and Walker, 2006; Vlaminckx et al., 2005; Chatellier et al., 2000; Working Group on Severe Streptococcal Infections, 1993; Proft et al., 2003; Thomas et al., 2008; Marrack and Kappler, 1990). Of note, although the superantigens are primary mediators contributing to the development of toxic shock, they are by no means sole determinants of severity or susceptibility to disease. Rather, the combined factors of exotoxin expression, host T cell repertoire, and underlying genetic and epigenetic susceptibility are likely to ultimately determine disease severity and susceptibility (Chatellier et al., 2003; Kasper et al., 2008; Abdeltawab et al., 2008; Datta et al., 2005). In order to understand the pathogenesis of puerperal sepsis, a thorough characterization of GAS at the molecular level is required, including not only a detailed analysis of virulence factors, but profiling of

75

superantigens and all surface proteins. As an alternate approach, an initial step in the molecular characterization would entail description of the primary and best-studied virulence factors and their association with clinical parameters. Our objective in this study was therefore to provide an initial characterization of the primary virulence factors, to determine whether specific bacterial M-types or the detected presence of specific SPE exotoxin genes are associated with morbidity in patients with group A streptococcal puerperal infection. 2. Methods We prospectively identified cases of group A streptococcal puerperal infection occurring in the Salt Lake City region between 1991 and 1997. All patients had a febrile illness, clinical evidence of endometritis, or both during the postpartum period. Subjects also had at least one positive culture for group A streptococcus from a normally sterile body site. Cases were identified by voluntary reporting by physicians and infection control nurses and by routine surveillance of culture logs (as audited by L.J.W.) in the microbiology laboratories at the LDS and University of Utah hospitals in Salt Lake City, Utah. However, reporting of group A streptococcal infection is not mandatory in the state of Utah and the study was not population-based. Characteristics of cases, medical and obstetric histories were obtained from review of the medical records and communication with the attending physician. Severe disease was considered to be present in patients with at least one of the following previously published criteria for severe (i.e., disseminated or invasive) disease (Abouzeid et al., 2005; Stevens et al., 1989; Alnaes-Katjavivi and Kahn, 2005; Chatellier et al., 2000; Working Group on Severe Streptococcal Infections, 1993): (1) disease requiring surgical exploration or debridement; (2) admission to the intensive care unit; (3) group A streptococcal toxic shock syndrome (Working Group on Severe Streptococcal Infections, 1993); or (4) hospitalization for ≥14 days. Renal insufficiency was defined as a serum creatinine ≥2 mg/dL, liver involvement as alanine aminotransferase, aspartate aminotransferase, or total bilirubin levels greater than or equal to twice the upper limit of normal for age, and coagulopathy with platelets ≤100 ␮L−1 or disseminated intravascular coagulation (Working Group on Severe Streptococcal Infections, 1993). Adult respiratory distress syndrome was defined as the acute onset of diffuse pulmonary infiltrates and hypoxemia in the absence of cardiac failure (Working Group on Severe Streptococcal Infections, 1993).

76

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

Group A streptococcal isolates were characterized by M-type and T agglutination pattern using classical serologic techniques (Moody et al., 1965). The Ouchterlony double-diffusion method of Rotta et al. (1971) was used for M-typing precipitin reactions. Antisera available for M-typing included types 1–6, 8, 12, 14, 15, 17–19, 23–26, 29–33, 36–41, 43, 47, 49, 51–53, and 55–57. Opacity factor detection and serotyping were assessed using techniques developed by Johnson and Kaplan (1988). Opacity factor typing antisera included types 2, 4, 9, 11, 22, 25, 28, 48, 49, 58, 59, 60 - 64, 66, 68, 73, 75, 78 and 81. The opacity factor serum is easier to obtain than M-type and was preferentially used when antisera were available for both M-type and opacity factor for a single serotype (Johnson and Kaplan, 1988). Serotypes determined by opacity factor inhibition technique are considered equivalent to the corresponding M-type (Maxted et al., 1973). The presence of the streptococcal pyrogenic exotoxin A and B genes was determined by the polymerase chain reaction. Briefly, chromosomal DNA was extracted from group A streptococcal isolates using previously described methods (Carroll et al., 1996). Two microliter aliquots of DNA preparations from each isolate were amplified with primers specific for the streptococcal pyrogenic exotoxin A and B genes (Tyler et al., 1992) using an Idaho Technology Rapidcycler 1608 (Idaho Falls, ID). The concentrations of components in the reaction mixture were 50 mM Tris (pH 8.3), 250 ␮g/mL bovine serum albumin, 4 mM MgCl2 , 2% sucrose, 100 ␮M Cresol Red, 200 ␮M (each) dATP, dTTP, dCTP, and dGTP, 0.5 ␮M (each) primer, and 0.4 U of Taq polymerase (AmpliTaq DNA polymerase, PerkinElmer, Branchburg, NJ). Conditions for detection of streptococcal exotoxin A gene (speA) included denaturation for 0 s at 94 ◦ C, annealing for 12 s at 50 ◦ C and primer extension for 8 s at 72 ◦ C for 40 cycles. Thirty cycles of denaturation for 0 s at 94 ◦ C, annealing for 0 s at 50 ◦ C, and primer extension for 6 s at 94 ◦ C were used to detect the streptococcal exotoxin B gene (speB). Amplicons were analyzed by standard gel electrophoresis in 2% agarose. The group A streptococcal isolate DLS 88003 (previously shown to express streptococcal speA and speB genes) was used as a positive control for exotoxins A and B. The group A streptococcal isolate DLS 90-400 was used as a negative control for exotoxin A and a positive control for exotoxin B. Both isolates were kindly provided by Dennis L. Stevens, M.D., V.A. Medical center, Boise, ID. Several strains of group A streptococcus also underwent analysis by pulsed-field gel electrophoresis (PFGE) to determine if an outbreak strain was present (Tenover et al., 1995).

Characteristics and outcome of patients with severe and mild disease and between those infected with group A streptococcus of differing M-types and exotoxins were compared using Student’s t-test, Mann–Whitney U test, and contingency tables as appropriate. Differences between groups were considered significant when P < 0.05. 3. Results Eighteen patients had group A streptococcal puerperal sepsis and concomitantly underwent M-type serotyping and speA and speB detection of the group A streptococcal isolates. Two strains did not have an identifiable M-type detectable by serotyping. The severity of illness and the extent and duration of therapy for subjects are shown in Tables 1 and 2. Patients were stratified by the severity of their disease according to the criteria outlined in the methods section. Ten of 18 patients (65.6%) had uncomplicated endometritis that resolved with antimicrobial therapy alone, and 8/18 patients (44.4%) manifested criteria of severe disease. Mean hospitalization for the entire cohort was 14.8 days. Most cases presented within 3 days of delivery, with the latest re-admission occurring on the 13th postpartum day. The majority of patients had high fevers and most had either elevated white blood cell counts or an increased proportion of immature leukocytes. Complications such as renal insufficiency, coagulopathy, and acute respiratory distress syndrome only occurred in women with severe disease (Table 1). All of the individuals with mild disease improved on antibiotic therapy and 6 required 3 days or fewer of intravenous treatment. Given the prospective observational nature of this study, decision to proceed with definitive surgical management was at the attending physician’s discretion. However, in review of the outcomes a number of interesting observations arise. Six of 8 patients with severe disease underwent attempts at conservative management; each of the 6 among this group was initially treated either with medical therapy or fertility-sparing surgical therapy. Two of the 6 patients were successfully managed in this fashion, although one subsequently required emergency surgery secondary to a bleeding gastric ulcer. Four of the 6 patients failed conservative therapy and ultimately required hysterectomy for worsening clinical disease with evidence or suspicion of myositis, necrotizing fasciitis, or expanding abscess formation. Among these women, patients with the shortest delay to definitive surgical therapy had uniformly briefer hospitalizations with overall improved outcomes (Table 2). Moreover, composite morbidity

Patient #

Degree of severity

Onset of symptoms (# days postpartum)

Maximal temperature (◦ C)

Systemic rash

Renal insufficiency

Liver involvement

Coagulopathy

ARDS

WBC on hospital admission (thousands)

1 3 5 6 8 12 13 14 16 18 2 4 7 9 10 11 15 17

Mild Mild Mild Mild Mild Mild Mild Mild Mild Mild Severe Severe Severe Severe Severe Severe Severe Severe

13 3 2 8 3 2 2 5 6 4 3 9 2 3 5 2 2 2

39.2 40.7 39.7 38.7 39.4 39.5 39.1 39.6 39.6 37.2 41.0 39.4 37.1 39.0 38.3 39.7 40.0 38.9

N N Y N Y N N N N N Y N N N Y Y Y N

N N N N N N N N N N Y Y N N Y N N N

N N N N Y N N N N N Y N N N N Y N N

N N N N N N N N N N Y N N N Y Y Y N

N N N N N N N N N N Y Y N Y N Y Y N

14.3 8.9 26.3 14.8 13.2 8.9 16.7 24.3 18.8 16.8 1.4 11.5 9.2 27.9 11.3 8.3 9.0 4.1

N, No; Y, yes; ARDS, Acute Respiratory Distress Syndrome.

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

Table 1 Severity of illness: systemic involvement.

77

78

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

Table 2 Severity of illness: extent and duration of therapy. Patient #

Degree of severity

Total days in hospital

Total days intravenous antibiotics

Intravenous antibiotic therapy

Days on ventilator (#days)

Surgical procedures

1 3 5 6 8 12 13 14 16 18 2

Mild Mild Mild Mild Mild Mild Mild Mild Mild Mild Severe

4 7 4 2 13 1 2 6 3 2 96

2.3 5.5 3.0 4.0 13.0 0.7 1.7 5.6 2.7 1.8 42.0

Triple therapy Triple therapy Triple therapy Triple therapy Nafcillin Triple therapy Ticarcillin-clavulanate Imipenem-cilastatin Triple therapy Cefoxitin Vancomycin/gentamicin/ clindamycin

0 0 0 0 0 0 0 0 0 0 17

4

Severe

15

10.5

Vancomycin/gentamicin/ clindamycin

10

7

Severe

3

2.3

Triple therapy

0

9

Severe

5

8.0

Triple therapy

0

10

Severe

9

9.9

0

11

Severe

19

14.6

Ticarcillinclavulanate/clindamycin Imipenemcilastatin/nafcillin/ gentamicin

None None None None None None None None None None HD# 1 exploratory laparotomy, HD# 38 TAH/BSO, HD# 70 bilateral below the knee amputations HD# 1 exploratory laparotomy, HD# 3 TAH/BSO, HD# 4 exploratory laparotomy HD# 1 salpingooopherectomy HD# 1 Episiotomy debridement HD# 2 TAH/BSO

15

Severe

65

28.5

Imipenemcilastatin/vancomycin

39

17

Severe

11

10.1

Ampicillin-sulbactam

0

8

HD# 9 exploratory laparotomy, vagotomy and pyloroplasty (stress ulcer) HD# 10 TAH/BSO, DI# 15 exploratory laparotomy HD# 1 exploratory laparotomy

HD, Hospital day #; TAH/BSO, total abdominal hysterectomy/bilateral salpingo-oophorectomy; “Triple therapy”, ampicillin/gentamicin/ clindamycin.

among severely ill patients was generally less in those undergoing aggressive surgical debridement compared to conservative management (Table 2). However, given the limited number of individuals in this prospective cohort, no inferences can be made regarding whether initial aggressive surgical management is preferential over attempted conservative management. Traditional risk factors for endomyometritis alongside demographic characteristics for patients with mild and severe disease are outlined in Table 3. Women with severe and mild disease had a similar mean duration of ruptured membranes (8.1 ± 1.9 h versus 6.8 ± 2.06 h, P = 0.646; Table 3) and number of vaginal examinations (9.0 ± 0.6 h versus 8.0 ± 0.6 h, P = 0.238; Table 3). No subject had intrapartum fever or a clinical diagnosis of

intra-amniotic infection. It is of interest to note that the mean maternal age differed significantly among the mild and severe disease cohorts, with (mean ± SEM) maternal age = 25.1 ± 2.1 versus 31.9 ± 1.3 for mild and severe disease respectively (P = 0.021). Table 4 depicts positive cultures for group A streptococcus and the M-types and detected presence of exotoxin genes for group A streptococcal isolates. Sixteen of 18 women had endometrial cultures and were positive for group A streptococcus; the remaining two women had positive blood cultures. All 8 patients with severe disease had group A streptococcus with either Mtype 1 (N = 3) or 28 (N = 5). Three patients (38%) with severe disease had detectable speA, as compared to two (20%) with mild puerperal infections (Table 4).

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

79

Table 3 Patient and labor characteristics. Patient #

Degree of severity

Age

Parity

Gestational age (weeks)

Ruptured membranes (h)

Vaginal examinations during labor (number)

Delivery mode

1 2 5 6 8 12 13 14 16 18 2 4 7 9 10 11 15 17

Mild Mild Mild Mild Mild Mild Mild Mild Mild Mild Severe Severe Severe Severe Severe Severe Severe Severe

18 23 25 27 26 29 19 27 40 17 27 34 37 34 35 27 30 31

1 2 2 2 0 2 0 1 4 0 3 1 7 0 0 2 1 4

39.4 38.9 40.3 32.4 40.0 39.7 38.0 38.3 39.6 39.0 40.0 39.7 39.1 40.4 41.6 40.6 41.0 38.9

10.2 0.1 4.4 9.5 16.8 4.3 11.7 3.1 3.8 17.1 3.8 0.9 4.8 12 9.2 18 2.8 2.9

7 3 4 7 Unknown 7 7 5 3 7 7 4 5 9 9 6 7 6

SVD SVD SVD FAVD SVD SVD SVD SVD SVD VAVD VAVD SVD SVD FAVD SVD SVD SVD VAVD

SVD, Spontaneous vaginal delivery; FAV, forceps assisted vaginal delivery; VAVD, vacuum assisted vaginal delivery.

Clinical outcome as stratified by initial molecular characterization of the GAS strains included M-type, speA, and disease severity is shown in Table 5. Eight of 11 (73%) isolates with M-types 1 and 28 were associ-

ated with severe disease. In contrast, no cases of severe disease were associated with M-types other than 1 or 28 (P < 0.01). It is noteworthy that each of the 3 isolates from patients with mild disease with M-types 1

Table 4 Characteristics of group A streptococcal isolates. Patient #

Disease severity

Endometrial culture

Blood culture

Urine culture

Other culture

M-type

Exotoxin A (speA)

Exotoxin B (speB)

1 3 5 6 8 12 13 14 16 18 2

Mild Mild Mild Mild Mild Mild Mild Mild Mild Mild Severe

+ + + + + NA + + + + +

− − − − − + + + NA NA +

− − + + + NA + − + − NA

NT 11 11 28 3 12 22 NT 28 28 28

+ − + − + − − − − − −

+ + + + + + + + + + +

4

Severe

NA

+

NA

1

+

+

7 9 10 11 15 17

Severe Severe Severe Severe Severe Severe

+ + + + + +

NA + − + + −

NA + NA + NA NA

NA NA NA NA NA NA NA NA NA NA Ascites (+) and CSF (+) Sputum (+) and CSF (+) NA Throat (−) Ascites (+) NA NA Ascites (+) and sputum (−)

28 28 1 28 1 28

− − + − + −

+ + + + + +

NT, Not typable; NA, culture not done; +, positive; −, negative.

80

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

Table 5 M-type, speA and disease severity.

M-type 1 or 28 Other M-types or untypable Exotoxin A (speA) No exotoxin A (speA)

Severe disease

Mild disease

8/11 (73%)a 0 3/6 (50%) 5/12 (42%)

3/11 (27%) 7/7 (100%) 3/6 (50%) 8/12 (67%)

a Isolates with M-type 1/28 are more likely to be associated with severe disease than other M-types (P < 0.01; Chi-square).

or 28 were M-type 28 (Table 5). speA was present in 6 isolates and was thus not significantly associated with disease severity. However, it is noteworthy that of the 3 isolates from patients with severe disease, all 3 manifested criteria of STSS. speB was detected in all group A streptococci isolates. Because of the prevalence of M-type 28 strains in our series, we tested 10 isolates, including strains from the 2 children of case #11 with PFGE to determine if an outbreak strain was present. Four distinct chromosomal DNA fingerprint patterns were identified when defined by virtue of a 10-band identical match (Tenover et al., 1995). The first pattern included three separate strain isolates from patient # 11 as well as her 2 children. The second pattern included 4 patients hospitalized at 4 different hospitals between July 1991 and May 1997 (2 with severe and 2 with mild disease). Among this second pattern group, none of these patients were hospitalized within less than 7 months of each other, and all were delivered by different physicians. The third pattern was seen in 2 patients admitted to 2 hospitals 1 month apart. Finally, there was a single case that had a unique chromosomal DNA fingerprint pattern. 4. Discussion To date, this is the largest series to focus on associative bacterial genes and toxins with puerperal GAS severity. Although current estimates note an increasing prevalence of the disease in recent decades, the delay in bringing forth such a series is likely due to the relative low overall incidence. Indeed, an active population-based surveillance for postpartum invasive GAS conducted by the CDC from 1995 to 2000 over 9 regions in the U.S. placed current disease estimates at 220 cases/year (Chuang et al., 2002). Of those with confirmed GAS infection, 46% manifest bacteremia without focus, 28%, endometritis, and the remaining a spectrum ranging from peritonitis (8%) and cellulitis (3%) to necrotizing fasciitis (3%) and streptococcal toxic shock syndrome (STSS, 3%) (Chuang et al., 2002). In

this cohort, the case-fatality rate approximated 3.5% (Chuang et al., 2002). The findings in our prospective series mirror those of the CDC: women with severe infections were extremely ill, suffering life threatening illness, multiple organ failure, and in some cases, permanent morbidity. Five met the case definition of STSS (Working Group on Severe Streptococcal Infections, 1993), and several others had individual criteria such as renal impairment, coagulopathy, liver involvement, acute respiratory distress syndrome, and a systemic rash. These findings are consistent with reports of serious group A streptococcal infections, although the absence of maternal deaths is in contrast with higher relative mortality rates in non-puerperal cases (Stevens et al., 1989; Chuang et al., 2002; Mencarelli et al., 2005; Tyrrell et al., 2005; Davies et al., 1996). Our data suggest that timely surgical intervention may reduce morbidity in severe cases, although no definitive inferences regarding management can be made. In general, patients with severe disease who were managed with aggressive surgical debridement faired better than those who were managed conservatively. Several women required repeat laparotomies after failed attempts at fertility-sparing surgery and patients with the longest interval from the onset of severe disease to definitive surgical therapy had the most complications. Improved clinical outcome has also been reported after prompt surgical debridement of dead or devitalized tissues in non-obstetric patients with invasive group A streptococcal infection (Stevens et al., 1989; Alnaes-Katjavivi and Kahn, 2005; Norrby-Teglund et al., 2005; Zimbelman et al., 1999). This observation is not surprising, given the tendency for group A streptococcus to cause tissue necrosis, allowing the organism to escape antimicrobial chemotherapy. Although women with group A streptococcal puerperal sepsis may not require hysterectomy, it is important to consider surgical intervention in all cases manifesting necrotizing fasciitis or myositis (Baxter and McChesney, 2000; Norrby-Teglund et al., 2005). In addition to surgical management, adjunctive medical therapy has been shown to potentially alter the course of GAS disease. Although our prospective series does not address such management issues per se, recent data from other studies suggest that both the choice of antibiotics and employment of intravenous immunoglobulin (IVIG) are of relative clinical importance. Group A streptococci remains sensitive to penicillin, albeit a reduction in the efficacy of penicillin has been observed with high colonization (Zimbelman et al., 1999). Several observational studies and retrospective reviews support the use of high dose penicillin G (24 million units in divided doses, assuming normal renal function) and clindamycin

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

(600–900 mg i.v., every 8 h) for suspected severe GAS (Baxter and McChesney, 2000; Zimbelman et al., 1999). Similarly, although IVIG therapy was not employed in the management of our prospectively acquired cohort, several recent trials have reported improved outcomes in instances of non-puerperal invasive disease at doses of 1–2 g/kg initial dose, followed by 0.4–0.5 g/kg for 3–5 days (Darenberg et al., 2003; Kaul et al., 1999). The mechanism of action of such therapy has not been fully elucidated. However, lower levels of opsonic antiM-type and neutralizing anti-superantigen antibodies among patients with invasive GAS infections when compared with mild disease controls have been observed; levels of these antibodies are present in tested formulations of IVIG (Endobuline S/D, Gamunex, and Gamimune) (Baxter and McChesney, 2000). It is therefore reasonable to postulate that IVIG therapy might be beneficial in instances of invasive puerperal disease manifesting STSS. The high prevalence of cases associated with Mtype 28 Group A streptococci suggested the potential for an outbreak strain in the community. However, the identification of four distinct patterns of chromosomal DNA fragment typing on pulsed-field gel electrophoresis makes this unlikely. Although four patients had an identical pattern, they failed to share common epidemiological factors nor risk of nosocomial infection. All were delivered in separate hospitals by different physicians several months to years apart. Moreover, none of the currently identified strains had a clear predilection to cause severe disease. Accordingly, these data weigh against an outbreak strain. While it was outside the initial scope and intent of our analysis to provide a complete molecular characterization of potential virulence factors, superantigens, and surface moities of group A strep, a number of interesting observations arose with respect to our reported initial characterization. First, morbidity in women with GAS puerperal infection was significantly associated with Mtypes 1 and 28. Specifically, it was our observation that all patients requiring admission to an intensive care unit, hospitalized for >14 days, requiring surgical debridement, or manifesting STSS were infected with isolates expressing either of these two M-types. These data are consistent with other investigators who have reported a strong association between M-type 1 and morbidity in patients with non-puerperal group A streptococcal infections (Stevens et al., 1989; Musser et al., 1991; Loubinoux et al., 2004; Forni et al., 1995; Mencarelli et al., 2005; Tyrrell et al., 2005; Davies et al., 1996; Talkington et al., 1993; Baxter and McChesney, 2000; Eriksson et al., 1999). Moreover and in accordance with

81

other investigators reports regarding puerperal sepsis, we similarly observed an association with M-type 28 and morbidity (Chuang et al., 2002; Green et al., 2005; Areschoug et al., 2004). This association may be due to a disproportionately high rate of asymptomatic vaginal colonization with this serotype (Chuang et al., 2002; Davies et al., 1996). It is additionally worth commenting that the M-3 serotype has similarly been shown to be associated with severe disease (Forni et al., 1995; Davies et al., 1996). The one patient infected with an M-3 isolate in our study had disease not necessitating surgical debridement, although she had a rash, liver involvement and a prolonged hospital course. Given that several patients with M-type 28 had uncomplicated endometritis, it is likely that factors other than virulent M-antigen serotype are important in the pathogenesis of severe group A streptococcal infection. Other authors have similarly noted that strain-specific virulence factors that differ within a given serotype may contribute to the varied clinical outcomes after infection with S. pyogenes (Chuang et al., 2002; Davies et al., 1996; Talkington et al., 1993). For example, Cleary and colleagues reported distinctive DNA patterns in patients with severe and mild infections associated with M1 strains (Baxter and McChesney, 2000). It is also likely that differences in host susceptibility are important determinants of disease severity in response to GAS infection (Stevens et al., 1989; Obrien et al., 2002; Carapetis et al., 2005; Chuang et al., 2002; Thomas et al., 2008; Forni et al., 1995; Mencarelli et al., 2005; Tyrrell et al., 2005; Davies et al., 1996; Talkington et al., 1993; Baxter and McChesney, 2000; Darenberg et al., 2004; Eriksson et al., 1999; Norrby-Teglund et al., 2005; Zurawski et al., 1998; Ekelund et al., 2005). In our case #11, the mother had severe disease, while both of her children had mild cases of streptococcal pharyngitis. In a similar vein, numerous reports have linked the detected presence of SPE-A to severe or invasive disease in patients with group A streptococcal infection (Stevens et al., 1989; Carapetis et al., 2005; Tyler et al., 1992; Musser et al., 1991; Loubinoux et al., 2004; Forni et al., 1995; Mencarelli et al., 2005; Tyrrell et al., 2005; Davies et al., 1996; Talkington et al., 1993). In our study, although most strains with the detectable presence of speA caused severe illness, overall morbidity did not demonstrate a significant association. Although the small number of patients in our study does not permit us to exclude such an association, several other studies have similarly failed to demonstrate a correlation between SPE-A gene presence and morbidity (AlnaesKatjavivi and Kahn, 2005; Loubinoux et al., 2004; Forni et al., 1995; Mencarelli et al., 2005; Tyrrell et al., 2005;

82

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83

Davies et al., 1996; Talkington et al., 1993). As was anticipated, speB was detected in all strains tested and was thus considered unrelated to morbidity (Tyler et al., 1992). However, it is of importance to note that given the established association of M1 and M3 to speA, we cannot definitively attribute our observed association to speA genotype. When we consider specifically those patients manifesting criteria for STSS (patients 2, 4, 10, 11, and 15) (Working Group on Severe Streptococcal Infections, 1993), 3 isolates demonstrated the detectable presence of speA (Table 4). The STSS is thought to be mediated by the release of SPEs and other antigens, which function as superantigens (Proft et al., 2003; Marrack and Kappler, 1990; Talkington et al., 1993; Baxter and McChesney, 2000; Eriksson et al., 1999; Darenberg et al., 2004; Norrby-Teglund et al., 2005; Zurawski et al., 1998; Ekelund et al., 2005). Superantigens stimulate T cells in a manner which bypasses classical antigen processing and presentation by binding simultaneously with the V␤ region of the T cell receptor and the class II MHC complex of the antigen presenting cell. This results in marked antigen-independent cellular proliferation with massive cytokine production-hallmarks of the clinically recognizable STSS (Proft et al., 2003). Several studies have shown a strong association between infection with strains possessing speA and non-puerperal STSS (Proft et al., 2003; Ekelund et al., 2005; Zimbelman et al., 1999; Darenberg et al., 2003; Kaul et al., 1999). Our series is consistent with these findings, and suggests that among patients manifesting criteria of severe disease, speA presence associated with STSS. In summary, our prospective cohort series provides an initial description of morbidity in women with puerperal group A streptococcal infection in association with limited characterization of virulence factors. Further complete molecular characterization is indicated at both the level of the microbiata, alongside interrogation of the host immune repertoire in order to fully understand susceptibility to puerperal sepsis. Acknowledgments We are indebted to Dr. Edward L. Kaplan and associates at the World Health Organizing Collaborating Center for Reference and Research on Streptococci, University of Minnesota, Minneapolis, MN, for streptococcal serotyping, to Dr. Donald Anderson Jr. and associates at the Sacred Heart Medical Center, Spokane, WA, for exotoxin gene assays and Mr. Jess Dalton, MSPH, for expert assistance in the preparation of the manuscript.

References Abdeltawab, N.F., Aziz, R.K., Kansal, R., Rowe, S.L., Su, Y., Gardner, L., Brannen, C., Nooh, M.M., Attia, R.R., Abdelsamed, H.A., Taylor, W.L., Lu, L., Williams, R.W., Kotb, M., 2008. An unbiased systems genetics approach to mapping genetic loci modulating susceptibility to severe streptococcal sepsis. PLoS Pathog. 18 (4), e1000042. Abouzeid, H., Wu, P., Mohammed, N., Al-Samarrai, M., 2005. Group A streptococcal puerperal sepsis: the return of a potentially fatal disease. J. Obstet. Gynaecol. 25, 806–808. Alnaes-Katjavivi, P., Kahn, J., 2005. Postnatal group A streptococcal infection in the upper genital tract. Acta Obstet. Gynecol. Scand. 84, 1212–1213. Areschoug, T., Carlsson, F., Stålhammar-Carlemalm, M., Lindahl, G., 2004. Host–pathogen interactions in Streptococcus pyogenes infections, with special reference to puerperal fever and a comment on vaccine development. Vaccine 6 (22 Suppl. 1), S9–S14. Baxter, F., McChesney, J., 2000. Severe group A streptococcal infection and streptococcal toxic shock syndrome. Can. J. Anesth. 47, 1129–1140. Carapetis, J.R., Steer, A.C., Mulholland, E.K., Weber, M., 2005. The global burden of group A streptococcal diseases. Lancet Infect. Dis. 5, 685–694. Carroll, K.C., Leonard, R.B., Newcomb-Gayman, P.L., Hillyard, D.R., 1996. Rapid detection of the staphylococcal mecA gene from BACTEC blood culture bottles by the polymerase chain reaction. Am. S. Chin. Pathol. 106, 600–605. Chatellier, S., Ihendyane, N., Kansal, R.G., Khambaty, F., Basma, H., Norrby-Teglund, A., Low, D.E., McGeer, A., Kotb, M., 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. Chuang, I., Van Beneden, C., Bernard, B., Schuchat, A., the ABC Surveillance/Emerging Infections Program Network, 2002. Population based surveillance for postpartum invasive group A streptococcus infections, 1995–2000. Clin. Infect. Dis. 35, 665–670. Darenberg, J., Soderquist, B., Normark, B.H., Norrby-Teglund, A., 2004. Differences in potency of intravenous polyspecific immunoglobulin G against streptococcal and staphylococcal superantigens: implications for therapy of toxic shock syndrome. Clin. Infect. Dis. 38, 836–842. Darenberg, J., Ihendyane, N., Sjolin, J., Aufwerber, E., Haidl, S., Follin, P., Andersson, J., Norrby-Teglund, A., StreptIg Study Group, 2003. Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome: a European randomized, double-blind, placebocontrolled trial. Clin. Infect. Dis. 37, 333–340. Datta, V., Myskowski, S.M., Kwinn, L.A., Chiem, D.N., Varki, N., Kansal, R.G., Kotb, M., Nizet, V., 2005. Mutational analysis of the group A streptococcal operon encoding streptolysin S and its virulence role in invasive infection. Mol. Microbiol. 56, 681–695. Davies, H.D., McGeer, A., Schwartz, B., Green, K., Cairn, D., Simor, A.E., Low, D.E., 1996. Invasive group A streptococcal infections in Ontario, Canada. N. Engl. J. Med. 335, 547–554. Ekelund, K., Skinhoj, P., Madsen, J., Konradsen, H.B., 2005. Reemergence of emm1 and a changed superantigen profile for group A streptococci causing invasive infections: results from a nationwide study. J. Clin. Microbiol. 43, 1789–1796. Eriksson, B.K.G., Andersson, J., Holm, S.E., Norgren, M., 1999. Invasive group A streptococcal infections: T1M1 isolates expressing

J.L.B. Byrne et al. / Journal of Reproductive Immunology 82 (2009) 74–83 pyrogenic exotoxins A and B in combination with selective lack of toxin-neutralizing antibodies are associated with increased risk of streptococcal toxic shock syndrome. J. Infect. Dis. 180, 410–418. Forni, A.L., Kaplan, E.L., Schlievert, P.M., Roberts, R.B., 1995. Clinical and microbiological characteristics of severe group A streptococcus infections and streptococcal toxic shock syndrome. Clin. Infect. Dis. 21, 333–340. Green, N.M., Beres, S.B., Graviss, E.A., Allison, J.E., McGeer, A.J., Vuopio-Varkila, J., LeFebvre, R.B., Musser, J.M., 2005. Genetic diversity among type emm28 group A Streptococcus strains causing invasive infections and pharyngitis. J. Clin. Microbiol. 43, 4083–4091. Johnson, D.R., Kaplan, E.L., 1988. Microtechnique for serum opacity factor characterization of group A streptococci adaptable to the use of human sera. J. Clin. Microbiol. 26, 2025–2030. Kasper, K.J., Xi, W., Nur-Ur Rahman, A.K., Nooh, M.M., Kotb, M., Sundberg, E.J., Madrenas, J., McCormick, J.K., 2008. Molecular requirements for MHC class II alpha-chain engagement and allelic discrimination by the bacterial superantigen streptococcal pyrogenic exotoxin C. J. Immunol. 181, 3384–3392. Kaul, R., McGeer, A., Norrby-Tegglund, A., Kotb, M., Schwartz, B., O’Rourke, K., Talbot, J., Low, D.E., 1999. Intravenous immunoglobulin therapy for streptococcal toxic shock syndromea comparative observational study. Canadian Streptococcal Study Group. Clin. Infect. Dis. 28, 800–807. Loubinoux, J., Florent, M., Merad, B., Collobert, G., Bouvet, A., 2004. Epidemiological markers of group A streptococcal infections in France. Indian J. Med. Res. 119, 152–154. MacArthur, J.D., Walker, M.J., 2006. Domains of group A streptococcal M protein that confer resistance to phagocytosis, opsonization, and protection: implications for vaccine development. Mol. Microbiol. 59, 1–4. Marrack, P., Kappler, I., 1990. The staphylococcal enterotoxins and their relatives. Science 248, 705–711. Maxted, W.R., Widdowson, J.P., Fraser, C.A.M., Ball, S.C., Bassett, D.C.J., 1973. The use of the serum opacity reaction in the typing of group A streptococci. J. Med. Microbiol. 6, 83–90. Mencarelli, M., Corbisiero, R., Padula, M.G., Galgai, I., Stolzuoli, L., Cellesi, C., 2005. Group A streptococcal infections: trend and strain emm typing in an area of central Italy, 1985–2002. Epidemiol. Infect. 133, 1107–1111. Moody, M.D., Padula, J., Lizana, D., Hall, C.T., 1965. Epidemiologic characterization of group A streptococci by T-agglutination and M-precipitation tests in the public health laboratory. Health Lab. Sci. 2, l49–162. Musser, S.M., Hauser, A.R., Kim, M.H., Schlievert, P.M., Nelson, K., Selander, R.K., 1991. Streptococcus pyogenes causing toxicshock-like-syndrome and other invasive diseases: clonal diversity and pyrogenic exotoxin expression. Proc. Natl. Acad. Sci. U.S.A. 88, 2668–2672. Norrby-Teglund, A., Muller, M.P., McGeer, A., Gan, B.S., Guru, V., Bohnen, J., Thulin, P., Low, D.E., 2005. Successful management of severe group A streptococcal soft tissue infections using an aggressive medical regimen including intravenous polyspecific immunoglobulin together with a conservative surgical approach. Scand. J. Infect. Dis. 37, 166–172.

83

Obrien, K.L., Beall, B., Barret, N.L., Cieslak, P.R., Reingold, A., Farley, M.M., Danila, R., Zell, E.R., Facklam, R., Schwartz, B., Schuchat, A., 2002. Epidemiology of invasive group A streptococcus disease in the United States, 1995–1999. Clin. Infect. Dis. 35, 268–276. Proft, T., Sriskandan, S., Yang, L., Fraser, J.D., 2003. Superantigens and streptococcal toxic shock syndrome. Emerg. Infect. Dis. 9, 1211–1218. Rotta, I., Krause, R.M., Lancefield, R.C., Everly, W., Lackland, H., 1971. New approaches for the laboratory recognition of M types of group A streptococci. J. Exp. Med. 134, 1298–1315. Stevens, D.L., Tanner, M.H., Winship, J., Swartz, R., Ries, K.M., Schlievert, P.M., Kaplan, E., 1989. Severe group A streptococcal infections associated with a toxic-shock-like syndrome and scarlet fever toxin A. N. Engl. J. Med. 321, 1–7. Talkington, D.F., Schwartz, B., Black, C.M., Todd, J.K., Elliott, J., Brieman, R.F., Facklam, R.R., 1993. Association of phenotypic and genotypic characteristics of invasive Streptococcus pyogenes isolates with clinical components of streptococcal toxic shock syndrome. Infect. Immun. 61, 3369–3374. Tenover, F.C., Arbeit, R.D., Goering, R.V., Mickelsen, P.A., Murray, B.E., Persing, D.H., Swaminathan, B., 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol. 33, 2233–2239. Thomas, D., Perpoint, T., Dauwalder, O., Lina, G., Floccard, B., Richard, J.C., Bouvet, A., Peyramond, D., Allaouchiche, B., Chidiac, C., Vandenesch, F., Etienne, J., Ferry, T., 2008. In vivo and in vitro detection of a superantigenic toxin Vbeta signature in two forms of streptococcal toxic shock syndrome. Eur. J. Clin. Microbiol. Infect. Dis., doi:s10096-008-0671-7. Tyler, S.D., Johnson, W.M., Huang, I.C., Ashton, F.E., Wang, G., Low, D.E., Rozee, K.R., 1992. Streptococcal erythrogenic toxin genes: detection by polymerase chain reaction and association with disease in strains isolated in Canada from 1940 to 1991. J. Clin. Microbiol. 30, 3127–3131. Tyrrell, G.J., Lovren, M., Kress, B., Grimsrud, K., 2005. Invasive group A streptococcal disease in Alberta, Canada 2000–2002. J. Clin. Microbiol. 43, 1678–1683. Vlaminckx, B.J., van Pelt, W., Schouls, L.M., van Silfhout, A., Mascini, E.M., Elzenaar, C.P., Fernandes, T., Bosman, A., Schellekens, J.F., 2005. Long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1994–2003. Clin. Microbiol. Infect. 11, 226–231. Working Group on Severe Streptococcal Infections, 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 269, 390–439. Zurawski, C.A., Bardsley, M.S., Beall, B., Elliott, J.A., Facklam, R., Schwartz, B., Farley, M.M., 1998. Invasive group A streptococcal disease in Metropolitan Atlanta: a population-based assessment. Clin. Infect. Dis. 27, 150–157. Zimbelman, J., Palmer, A., Todd, J., 1999. Improved outcome of clindamycin compared with beta-lactam antibiotic treatment for invasive Streptococcus pyogenes infection. Pediatr. Infect. Dis. J. 18, 1096–1100.