Bonatti H; Stelzmueller I

Research Article

Austin J Surg. 2018; 5(2): 1126.

Group Milleri Streptococci in Surgical Patients: A Review of 367 Patients from a Single Center

Bonatti H¹* and Stelzmueller I²

¹Unversity of Maryland Community Medical Group Surgical Care, Easton, USA

²Praxis DOZ. DR. Ingrid STELZMüLLER, Salzburg, Austria

*Corresponding author: Hugo Bonatti, Unversity of Maryland Community Medical Group Surgical Care, USA

Received: December 13, 2017; Accepted: January 12, 2018; Published: February 09, 2018

Abstract

Background: Group Milleri streptococci (GMS) also known as Group anginosus streptococci comprise a heterogeneous subgroup of Gram-positive cocci, which are capable to cause severe purulent infections and abscesses.

Patients and Methods: This is a retrospective study of infections due to 637 consecutive GMS isolates in 448 patients during a four year period. Eight patient subgroups were created: primary non surgical (18%), ENT/dental (16%), upper GI tract (5%), lower GI tract diseases (8%), appendicitis (9%), anorectal (10%), biliary (27%), thoracic (3%) and miscellaneous infections (4%).

Results: Overall, 346 patients had a single isolate, 71 had two, 18 had three and 21 had four or more isolates. Median patient age was 51.1 years (range 0.1-94), 58.6% were male. Significant co-morbidities were present in 19.2% of patients, 17.9% were immunocompromised, 11.6% had cancer and 10.9% were solid organ recipients. Swabs accounted for 51.8% of isolates, bile for 19%, aspirates for 13.7%, blood cultures for 7.8% and the remaining specimens were from miscellaneous sites. Mixed infections were found in 48% with E. coli and Bacteroides fragilis being the predominant secondary pathogens. Of the surgically treated patients, 70% also received antibiotics [mean duration of treatment was 6 days). Recurrent infections developed in 5.8% and persistent infection requiring repeat surgery and/or prolonged antibiotic therapy in 11.2%. During the study period, 20 patients died, GMS infection accounted for a 2.5% mortality rate. Resistance rates to penicillin G, piperacillin/tazobactam, carbapenems and clindamycin was <5%, for cephalosporins and quinolones resistance ranged between 5% and 40%.

Conclusion: GMS are significant surgical pathogens. They should be distinguished from viridans streptococci. Most GMS infections require combined surgical/interventional treatment and antimicrobial chemotherapy.

Keywords: Streptococcus group Milleri; Streptococcus group anginosus; Sepsis; Abscess; Antibiotic; Surgical infection, Immunosuppression

Introduction

Pyogenic infections may be caused by a multitude of different microorganisms, may affect everybody site [1-3] and are commonly polymicrobial [2,4,5]. Some infections only require antibiotics and many can be treated with drainage alone but for most pyogenic infections treatment includes drainage, source control and antimicrobial therapy [6,7].

Until the 1970s viridans streptococci were rarely associated with purulent infections except endocarditis [8]. Viridans streptococci including Group Milleri Streptococci (GMS) can cause purulent infections of various internal organs [8-11]. GMS - in the US referred as to Group anginosus streptococci - comprise a heterogeneous group of streptococci including the species Streptococcus anginosus, Streptococcus constellatus, Streptococcus intermedius and certain β-haemolytic streptococci from Lancefield groups C, F and G [1,12]. The relevance of GMS within viridans streptococci with regard to surgical infections may have been underestimated. GMS have a propensity to cause purulent infections and form deep seated abscesses, are frequently isolated as part of polymicrobial infections and have a high recurrence rate if not treated properly [13-19]. In 1956, Guthof first described Streptococcus milleri in abscesses of the oral cavity [20]; subsequently, GMS were isolated from liver and central nervous system abscesses and in patients suffering from appendicitis, peritonitis, pleural empyema, Ear-Nose-Throat (ENT) and dental infections [20-25]. They also may cause endocarditis and bacteraemia [10,26,27]. Very recently, worldwide an increasing interest in GMS can be observed with multiple new reports being available [28-36]. Some of these reports focus on previously described GMS infections others emphasize the significance of GMS in thus far poorly studied patient populations such as those with pulmonary disease or GMS in the setting of vascular graft infection [30,36].

In this retrospective study the significance of GMS in a cohort of patients from a single center was analyzed (637 consecutive GMS isolates in 448 patients).

Patients and Methods

Patient identification, data collection, statistical analysis

For identification of patients the computerized database of the Institute for Hygiene and Microbiology was screened for GMS isolates. Clinical data were retrieved from electronic and paper charts and entered into an MS EXCEL database. The work was conducted according to the rules of the local ethics committee; subsets of this series have been reported previously.

Infection was assumed if GMS was isolated from otherwise sterile body sites together with clinical signs of infection such as fever, cough or purulent discharge and/or radiological signs such as abscess formation and/or laboratory findings indicative for acute infection including leucocytosis and/or elevated C-reactive protein.

Eight patient subgroups were created: primary non surgical (18%), ENT/dental (16%), upper GI tract (5%) and lower GI tract diseases (8%), appendicitis (9%), anorectal (10%), biliary (27%), thoracic (3%) and miscellaneous infections (4%).

For non surgical patients only limited clinical data could be obtained. Data are reported as total numbers and percentage for discreet and median/mean with range for continuous parameters. Statistical analysis was performed using SPSS. A p-value of <0.05 was considered statistically significant.

Microbiological analysis

The microbiology database was queried for all isolates of GMS from a collection of approximately 430.000 processed samples during a four year period. Isolation and identification of GMS were performed according to the Clinical and Laboratory Standards Institute (CLSI). Aerobic and anaerobic testing was performed. Specimens were incubated on Columbia agar containing 5% sheep red blood cells and incubated in a CO2-enriched atmosphere for 48 hours. Streptococci were identified by morphology, colony size, and hemolysis pattern. GMS were identified using the API 20 Strep Bio Merieux^® (Marcyl ´Etoile, France) assay. Antimicrobial susceptibility testing was performed using the disc diffusion assay according to CLSI guidelines (CLSI 2004 Performance Standards for Antimicrobial Susceptibility Testing; Fourteenth Informational Supplement CLSI, Wayne, Pa). Subsets of isolates were subtyped; Streptococcus intermedius and constellatus were identified but not Streptococcus anginosus.

Results

Clinical Data

A total of 637 GMS isolates in 448 patients could be identified. The majority of patients (65%) were treated at various surgical services. Twenty percent of isolates originated from non-surgical services including internal medicine (12%), gynecology and urology (4%), neurology (2%), dermatology and pediatrics (1% each). Figure 1a shows distribution within surgical services. Approximately 80% of all isolates were isolated from general surgical and transplant patients. Figure 1b displays distribution of GMS isolates according to the site of infection. The biliary tract was the most common infection site with 27% followed by ENT/dental infections with 16%; all other infections accounted for <10% of cases.

Figure 1: Study population: 1a: Distribution of surgical patients; 1b: Distribution according to infection site for all patients.

    
    
    Figure 1:  Study population: 1a: Distribution of surgical patients; 1b:
Distribution according to infection site for all patients.

The median age of the surgical cohort was 50.3 years. The youngest patient was a seven-day-old baby with necrotizing enterocolitis; the oldest was a 95 year old woman with perforated appendicitis. Overall, 57.7% of patients were male and 42.3% female. Median age of females was 54.9 years as compared to 50.6 years for the male cohort, p=0.01. Patients with appendicitis had the lowest median age (38 years) followed by patients with ENT/dental and anorectal infections (39 years). Patients with thoracic infections and infections associated with perforated lower gastrointestinal tract were significantly older. Significant co-morbidities could be identified in 18% of patients, 17% were immunocompromised (HIV, Crohn´s disease, chronic steroid treatment, diabetes mellitus), 16.5% had undergone solid organ transplantation (40 liver, one renal, five pancreatic, three intestinal recipients) and 12% had malignancies.

Microbiological data

The majority of the samples were obtained from swabs (n=328, 51%), followed by bile (n=125, 20%), aspirates (n=85, 13%), miscellaneous specimens (n=53, 8%) and blood cultures [n=46, 7%). When comparing surgical and non surgical patients, no differences with regard to type of specimen was found with the exception of blood cultures, which accounted for only 2.5% of surgical isolates (n=26 in 14 patients) but 18.7% of non-surgical isolates (n=20 in 12 patients), p<0.0001. There was an equal number of patients who had multiple GMS isolates (23.1% surgical vs. 25.3 % non-surgical patients, p=0.376). Within surgically treated patients, those with pleural empyma had the highest percentage of multiple isolates. Surgical patients had significantly more mixed infections when compared to non-surgical patients (52.7% vs. 19.8%, p<0.0001). Mixed infection was found in >80% of appendicitis, 50-60% in thoracic, hepatobiliary, anorectal and upper/lower gastrointestinal tract infections and <30% in soft tissue as well as ENT/dental infections (p=0.001).

GMS subtyping was performed in 268 cases (42%) including 193 from surgical patients, of which 29 isolates (6.2%) were Streptococcus constellatus and 164 specimens (35.5%) were Streptococcus intermedius. Out of the 637 isolates, GMS was the single pathogen in 331 cases (52%), whereas polymicrobial infection was found in 48% (306 isolates); for surgical patients mixed infection was present in 58% of cases. In total, 392 additional microorganisms were identified including 102 Gram positive bacteria, 181 Gram negative bacteria, 90 anaerobes and 19 fungi (Figure 2). Escherichia coli was the single most common coinfecting pathogen (n=79) followed by Bacteroides fragilis (n=74) and enterococci (n=40); there were 31 isolates of Klebsiella spp. and 28 of coagulase negative staphylococci (CNS), all other pathogens were cultured in <20 cases. Table 1displays the spectrum of the most common secondary pathogens according to the infection site. Staphylococcus aureus was a coinfecting pathogen in 3.3% with no major differences between the different infection sites. CNS was predominantly found in chest infections, Enterococcus spp originated mainly from the biliary tract. E. coli and Bacteroides fragilis were most frequently cultured from patients with gastrointestinal perforation. Pseudomonas aeruginosa was predominantly isolated in infections of the upper and lower gastrointestinal tract and Candida albicans was mainly found in upper gastrointestinal tract and chest infections.

Table 1: Common co-infecting organisms according to infection site.




  
    
    
    
    
    
    
    
    
    
    Soft 
    
    
  
  
    
    non 
    ENT/ 
    
    
    Appendi 
    Ano 
    Hepato 
    
    tissue/ 
    
    total # of 
  
  
    
    surgical 
    dental 
    Upper GI 
    Lower GI 
    citis 
    rectal 
    biliary 
    Thoracic 
    various 
    p-value 
    isolates 
  
  
    Staphylococcus 
    
    
    
    
    
    
    
    
    
    
    
  
  
    aureus 
    4.4 
    4.3 
    4.8 
    0 
    0 
    0 
    0.8 
    7.7 
    0 
    0.257 
    10 
  
  
    Coagulase negative 
    
    
    
    
    
    
    
    
    
    
    
  
  
    staphylococci 
    3.3 
    5.7 
    4.8 
    11.8 
    0 
    0 
    9.2 
    23.1 
    8.7 
    0.026 
    28 
  
  
    Enterococcus spp. 
    4.4 
    1.4 
    9.5 
    14.7 
    4.9 
    0 
    21 
    0 
    4.3 
    <0.0001 
    40 
  
  
    Escherichia coli 
    1.1 
    1.4 
    28.6 
    38.2 
    61 
    27.9 
    16 
    7.7 
    4.3 
    <0.0001 
    79 
  
  
    Klebsiella spp. 
    1.1 
    0 
    23.8 
    11.8 
    4.9 
    0 
    13.4 
    7.7 
    8.7 
    <0.0001 
    31 
  
  
    Enterobacter spp. 
    1.1 
    2.9 
    4.8 
    8.8 
    2.4 
    2.3 
    5.9 
    7.7 
    4.3 
    0.588 
    18 
  
  
    Pseudomonas 
    
    
    
    
    
    
    
    
    
    
    
  
  
    aeruginosa 
    1.1 
    4.3 
    14.3 
    14.7 
    4.9 
    2.3 
    1.7 
    0 
    0 
    0.004 
    17 
  
  
    Bacteroides fragilis 
    2.2 
    0 
    9.5 
    55.9 
    46.3 
    37.2 
    11.8 
    0 
    8.7 
    <0.0001 
    74 
  
  
    Candida albicans 
    1.1 
    4.3 
    9.5 
    2.9 
    0 
    0 
    9.2 
    7.7 
    0 
    0.036 
    19



Table 1: Common co-infecting organisms according to infection site.

Download PDF

Citation: Bonatti H and Stelzmueller I. Group Milleri Streptococci in Surgical Patients: A Review of 367 Patients from a Single Center. Austin J Surg. 2018; 5(2): 1126.

Instruction for Authors

Submit Your Article

Group Milleri Streptococci in Surgical Patients: A Review of 367 Patients from a Single Center

Abstract

Introduction

Patients and Methods

Results

Discussion

References