Relevance of Group Milleri Streptococci in Thoracic Surgery: A Clinical Update

By Stelzmueller, I Biebl, M; Berger, N; Eller, M; Et al

Group Milleri streptococci (GMS), a heterogeneous group of streptococci, are associated with purulent infections. This study was a retrospective analysis of all consecutive thoracic infections of GMS between 2001 and 2004. Of 246 surgical GMS infections, thoracic infections accounted for 4.5 per cent, including 10 pleural infections (eight empyemas and two infected pleural effusions) and one mediastinal infection. The etiology of pleural infection was parapneumonic (7), second to esophageal perforation (1), liver transplantation (1), and liver resection (1). Polymicrobial infections were present in 64 per cent. All patients underwent removal of the infected masses, including drainage (3), thoracoscopic decortication (5), thoracotomy with debridement (2), and incision with drainage (1). The case fatality rate was 9 per cent (there was one patient with congestive heart disease unfit to undergo surgical empyema evacuation) and the recurrence rate was 27.3 per cent (three patients). Combined antibiotic/surgical treatment was successful in all other cases. GMS isolates were susceptible to clindamycin and all beta-lactam antibiotics except ceftazidime, but were resistant to aminoglycosides. If found intrathoracically, GMS frequently progress to severe empyema. Therefore, timely removal of pleural collection by percutaneous drainage or surgical intervention seems indicated. If surgery is required, thoracoscopic decortication may be the preferred approach. THE SPECTRUM OF PATHOGENS associated with intrathoracic bacterial infections is rather diverse, including Staphylococci, Streptococci (including Streptococcus pneumoniae), Pseudomonas spp., Klebsiella spp., and Hemophilus influenza,1-3 and largely depends on the source of the infection. Although Streptococci other than S. pneumoniae were one of the leading pathogens cultured from pleural empyema 50 years ago, in more recent series, their importance apparently has declined and were reported to account for approximately 14 per cent of all cases of pleural infections.4, 5

The most relevant subgroup are members of the Group Milleri streptococci (GMS), which are characterized by a high potential to cause purulent infection and abscess formation.6, 7 Although GMS are not listed in the approved lists of bacterial names, they comprise a subset of streptococci such as the species S. intermedius, S. constellatus, S. anginosus, together with rare beta-hemolytic streptococci of Lancefield groups G and F. Most authors only consider the three species 5. intermedius, S. constellatus, and 5. anginosus as GMS because of their clinical significance.5-9

Parapneumonic pleural effusion alone may remain asymptomatic; however, without proper antibiotic treatment, in >50 per cent of cases, this may progress to parapneumonic empyema (defined as purulent or Gram stain-positive effusion associated with lower respiratory tract infection).1-2 Until today, pleural empyema remains a major cause of morbidity and mortality after bacterial pneumonia. Intrathoracic infection can complicate all surgical interventions of thoracic organs, in particular in case of dehiscence of esophageal anastomosis or leakage of a bronchial stump. Other causes of intrathoracic infections are penetrating trauma, spread of infection from the neck, sternal, or vertebral osteomyelitis and discitis, perforation of the larynx, trachea, pharynx, or esophagus, or cardia of the stomach in the presence of a hiatal hernia and transdiaphragmatic spread of pathogens in the case of peritonitis. Moreover, spontaneous infections of pleural effusions are well known to occur in immunosuppressed individuals. GMS have been reported in all of these thoracic infections, which can be explained by the fact that GMS colonize the upper gastrointestinal tract and oral cavity. Any purulent pleural effusion can progress to systemic infection and/or can cause localized complications. The late sequelae of a contained empyema of the pleural space are progressive organization of the purulent material with ingrowths of fibroblasts, which often results in long- term morbidity such as recurrent airway infection from entrapment and impaired ventilation of the ipsilateral lung.1, 2, 10 Therefore, all pleural empyemas mandate aggressive therapy, including broad- spectrum antibiotics combined with surgical intervention and drainage with the aim to evacuate any purulent fluid from the thoracic cavity.1, 2, 10

In the present study, we report on our experience with intrathoracic infections caused by GMS observed during a 5-year period.

Patients and Methods

Identification of Patients, Data Collection, and Statistical Analysis

All results of microbiologic specimen testing at out hospital are archived in a computerized database maintained by the Department of Microbiology, which serves as a core facility for the western part of Austria. This database was searched for all isolates of GMS identified between January 2001 and December 2004. A total of 436,000 samples were processed at our microbiological facility during the study period. A total 637 GMS-positive isolates in 452 patients were identified.

Twelve patients with GMS pleural, intrapulmonal, or mediastinal infections were identified, which represents 2.4 per cent of all patients with GMS infection. One patient with a solitary peripheral nodule on chest X-ray underwent CT-guided biopsy and tested positive for GMS but did not present with a purulent infection, and therefore was excluded from analysis. Clinical data of the remaining 11 patients were retrieved from electronic records and hospital paper charts and were entered into a computerized database. A database containing all patients who were surgically treated for pleural empyema was searched to determine the frequency of GMS in this clinical setting. Data are reported as total numbers and percentage or mean +- SD.

Microbiological Analysis

Isolation and identification of GMS followed the Clinical and Laboratory Standards Institute (CLSI) at our microbiological laboratory. All specimens were cultured aerobically and anaerobically. Specimens were incubated on Columbia agar containing 5 per cent sheep red blood cells and incubated in a CO^sub 2^- enriched atmosphere for 48 hours. Isolated streptococci were identified by morphology, colony size, and hemolysis pattern. Identification of GMS was made with API 20 Strep Bio Merieux(R) (Marcy-l’Etoile, France), which includes S. anginosus, S. constellatus, S. intermedius, and certain beta-hemolytic streptococci from Lancefield groups C, F, and G. Subtyping was not performed. Antimicrobial susceptibility testing was performed using the disk diffusion assay according to CLSI guidelines (CLSI 2004 Performance Standards for Antimicrobial Susceptibility Testing; 14th Informational Supplement; CLSI, Wayne, PA).

A subgroup analysis between polymicrobial versus monomicrobial GMS infection was performed (Table 1).

Results

Eleven patients with thoracic GMS infection were identified, including four (36.4%) women and seven (63.6%) men. Mean age at time of diagnosis was 58.5 +- 11.5 years. Presenting clinical condition were pleura empyema in eight patients (72.7%), infected pleural effusion in two patients (18.2%), and mediastinal infection in one patient (9.1%; Table 2).

In six (54.5%) cases, mixed aerobic/anaerobic infections were detected. Other isolated organisms were gram-negative bacilli in five cases (Proteus spp., Enterobacter spp. Escherichia coli, and Lactobacillus), gram-positive cocci in six cases (coagulase- negative Staphylococci, 5. viridians, S. cremoris), and one fungus (Candida spp.). Table 2 compares patients with monomicrobial GMS infection and patients with polymicrobial infection. There was no difference in treatment strategies or outcome between the two groups.

The etiology of the empyema was esophageal perforation in one patient, and parapneumonic in the other seven cases. Both GMS- infected pleural effusions were diagnosed in patients after abdominal surgery, including liver resection for Echinococcus alveolaris of the liver and liver transplantation, respectively. One patient suffered from severe caries and gingivitis and subsequently developed multiple abscesses at the neck and mediastinum, and required several consecutive incisions and drainage procedures.

Seven (63.6%) patients were inpatients at time of diagnosis, and five (45.5%) were being treated on an intensive care unit, including two (18.2%) patients who required mechanical ventilation. All patients had significant comorbidities, including chronic obstructive pulmonary disease (3), congestive heart disease (1), diabetes mellitus (1), bronchiectasia (1), alcohol abuse and smoking (1), polyarthritis (1), CREST, and Mallory-Weiss-lesion (1). The liver recipient received immunosuppressive therapy with Tacrolimus, mycophenol-mofetil, and steroids.

TABLE 1. Comparison of Patients with Monomicrobial versus Polymicrobial Group Miller Streptococcal Infection

Treatment

Empiric intravenous antibiotic treatment was initiated upon diagnosis of intrathoracic infection and was modified according to susceptibility testing. Additional surgical interventions were performed in eight (72.7%) patients, including video-assisted thoracoscopic empyema evacuation (VATS) in five patients, thoracotomy in two patients, and abscess incision with drainage in one patient. Three (27.3%) patients were treated with pleural drainage, including both patients with infected pleural effusions and one 91-year-old patient with pleural empyema, who was not considered suitable for any more invasive procedure. Outcome

One patient died from the infection, case fatality rate of 9 per cent. This 91-year-old man with congestive heart disease suffered from a parapneumonic pleural empyema and was treated by pleural drainage. Despite intensive antibiotic treatment (ceftriaxone and imipenem/cilastatin), the patient developed multiorgan failure and died.

All other patients recovered, but three (27.3%) patients developed recurrent infections. The patient with the multiple odontogenic neck and mediastinal abscesses was primarily treated with amikacin and was later successfully switched to imipenem/ cilastatin and clindamycin. He required several abscess incisions. The other two recurrences occurred in patients with parapneumonic empyemas, who had primarily undergone thoracotomy. Recurrent empyema only occurred in one patient with a CREST (9.1%), who required rethoracotomy with creation of a thoracostoma. The third patient developed a thoracic wall abscess after thoracotomy and was treated with abscess incision and drainage. All of these patients recovered without further relapse after reinterventions.

Results of Antibiotic Susceptibility Testing

GMS isolates were susceptible to penicillin, imipenem/ cilastatin, and cephalosporins in 100 per cent of cases. Clindamycin tested active in 96 per cent, fosfomycin in 90 per cent, and ciprofloxacin in 71 per cent. All strains were resistant to aminoglycosides with the exception of amikacin (susceptibility of 61%). The most commonly used antimicrobial agents were broad- spectrum penicillins, carbapenems, cephalosporins, and clindamycin. The median time of antibiotic use was 15 days (range 2-AS).

Discussion

GMS are commensals of the respiratory, intestinal, and urogenital tract. The clinical relevance of GMS in thoracic surgery is based on the propensity for deep abscess formation requiring surgical intervention and drainage.7 Only limited information has been published on GMS infections in thoracic surgery. The few larger series published have to be considered outdated as bacterial spectrum and antibiotic treatment have changed substantially over the last decades. In older series, thoracic infections comprised between 10 per cent and 32 per cent of all GSM infections,11-13 and more recently, Porta et al.5 reported up to 20 per cent of all GMS infection located in the thoracic cavity. In the present series, intrathoracic GMS infections were rarer than previously reported, accounting for only 2.5 per cent of all GMS infections. The finding of 13.3 per cent of all pleural empyemas treated surgically at our institution between 2001 and 2004 caused by GMS blends in the figures reported in the literature (14%57%).4, 5, 9

Pleural empyema can develop from any pulmonary or intrathoracic infection. In our series, the most common way of infection was second to pneumonia (72%), however, all other reported ways of infection were also encountered (Table 1). Interestingly, two (18.2%) of our patients acquired a pleural infection after abdominal surgery, all without opening of the gastrointestinal tract (one liver transplantation and one liver resection). Although the bile duct was opened during liver transplantation and the postoperative immunosuppressive therapy may facilitate progression of the infection through spilled contaminated bile, no intra-abdominal infection occurred during the liver resection. Thoracic GMS infections have also been described in association with ascites or subdiaphragmatic hematomas alone.1 Although a Mallory-Weiss lesion can be an entrance for infection into the mediastinum,1,5 we observed three patients in our cohort with contiguity of the infection through an intact muscle layer, accounting for almost one- third (27%) of all cases in this series. Dental trauma is a recognized entrance path for streptococci into the blood stream, and GMS have also been described in 3 per cent to 15 per cent of all Streptococcus viridians-causmg endocarditis.6 In addition, anecdotic reports in the literature have described severe intrathoracic infections secondary to odontogenic abscesses.14,15 In our patient, infection extended directly from the oral cavity into the mediastinum, resulting in multiple abscess formations.

TABLE 2. Demographic and Clinical Data of the Study Cohort

All intrathoracic infections mandate broadspectrum, high-dose antibiotic treatment. In addition, evacuation of infected collections from the thoracic cavity is important to enable restitution and prevent recurrent infection or late lung entrapment.1-3,10 In case of a parapneumonic pleural effusion alone, which often represents a (sterile) exsudative fluid collection because of increased capillary permeability,1 antibiotic treatment of the underlying pneumonia seems sufficient to obviate progression to complicated pleural effusion or empyema.1-3,10 In the presence of an infected pleural fluid collection, however, drainage is usually required to prevent progression to empyema.1 During the early stages (referred as phase one of pleural empyema), evacuation of the liquid effusions by large-bore thoracic drainage is sufficient.1,3,10-15 A more progressed empyema with liquid fibropurulent masses is often too viscous for simple drainage. To reduce the risk of open entrance to the thoracic wall through thoracotomy, VATS is recommended as the treatment of choice.10,15,16

Evacuation of the infected material becomes more difficult during the different stages of empyema, and the optimal timing for surgical intervention is not clearly defined.1,3,4 Although drainage of any infected fluid is recommended as soon as possible, most authors would advocate a more invasive approach with VATS after 3 to 7 days of unsuccessful drainage. 1-3,10 If a thoracoscopic approach is possible, VATS has excellent results with low perioperative morbidities, and recurrence rates range from O per cent to 9 per cent.16-19

In addition to mechanical removal of the infected masses, adequate antibiotic treatment based on sensitivity testing is mandatory.5 In our patients, the median duration of antibiotic therapy was 15 days, with a range from 10 to 65 days. In most cases, antibiotic treatment consisted of at least two different groups of antimicrobial drugs-mainly beta-lactams and clindamycin. The combination of two or more antibiotic agents was often started empirically on the basis of suspected polymicrobial purulent infection, and also in patients with critical illnesses. After identification of the causative bacteria, therapy was adapted as indicated. As polymicrobial infection with GMS occurred in more than half of our patients (6/11), initial broad-spectrum antibiotic treatment seems key to successful infection control. With good susceptibility of the isolated GMS strains to ureidopenicillins and carbapenems and cephalosporin/clindamycin combination, these substances can be recommended for empiric initial treatment of severe plural empyema, especially when a polymicrobial infection is suspected.5,7,9 Gram-negative bacilli and Gram-positive cocci should be covered and, of note, clindamycin is the treatment of choice against anaerobes; Clindamycin might have some advantages in terms of pharmacokinetics, as the agent accumulates intracellularly within white blood cells and by that can be delivered to the site of infection even in the case of abscess membranes, which are commonly found in GMS infections. 1,5 In general, poor susceptibility to aminoglycosides was found, and in our series, amikacin and fosfomycin were only used in two patients. We observed three ongoing infections after initial intervention in our patients, but only one recurrent empyema. All patients with recurrent infection required additional surgical interventions. One patient had multiple abscesses and therefore required repeat drainage, whereas both other patients underwent open rethoracotomy and thoracostomy. We had one death in our series, where surgical treatment was not possible because of the poor condition of this 91-year-old patient.

In conclusion, we found that GMS infection of the thoracic cavity occurred with a lower incidence in our series compared with the literature. If transmitted into the pleural space, GMS continues to be a relevant pathogen because of the high incidence of empyema formation. Broad-spectrum empiric antibiotic treatment is recommended. In the presence of any infected pleural collection, patients benefit from timely removal by drainage alone or thoracoscopic empyema evacuation, rather than initial antibiotic treatment and late decortication, which often requires thoracotomy.

REFERENCES

1. Shields TW. General Thoracic Surgery, 6th ed. Baltimore: Lippincott Williams & Wilkins, 2004, pp 674-713.

2. Chapman SJ, Davies RJO. The management of pleural space infections. Respirology 2004;9:4-11.

3. Davies CWH, Gleeson FV, Davies RJO, on behalf of the BTS Pleural Disease Group. BTS guidelines for the management of pleural infection. Thorax 2003;58:ii18-ii28.

4. Lindstrom S, Kolbe J. Community-acquired parapneumonic thoracic empyema: Predictors of outcome. Respirology 1999;4: 173-9.

5. Porta G, Rodriguez-Carbellaira M, Gomez L, et al. Thoracic infection caused by Streptococcus milleri. Eur Respir J 1998; 12: 357-62.

6. Piscitelli SC, Shwed J, Schreckenberger P, Danziger LH. Streptococcus miller group: Renewed interest in an elusive pathogen. Eur J Clin Microbiol Infect Dis 1992;11:491-8.

7. Clarridge JE, Attorri S, Musher DM, et al. Streptococcus intermedius, Streptococcus constellatus, and Streptococcus anginosus (“Streptococcus milleri group”) are of different clinical importance and are not equally associated with abscess. J Clin Inf Dis 2001;32:1511-5. 8. Jacobs JA, Pietersen HG, Stobberingh EE, Soeters PB. Streptococcus anginosus, Streptococcus constellatus, and Streptococcus intermedius: Clinical relevance, hemolytic and serologic characteristics. Clin Microbiol Inf Dis 1995;104:547-53.

9. Belko J, Goldmann DA, Macone A, Zaidi AK. Clinically significant infections with organisms of the streptococcus miller group. Pediatr Infect Dis J 2002;21:715-25.

10. Kolditz M, Halank M, Hoeffken G. Parapneumonic effusion and pleural empyema: Topical aspects of classification, diagnosis and treatment. Pneumologie 2004;58:83-91.

11. Jacobs JA, Pietersen HG, Slobbering EE, Soeters PB. Bacteremia involving the “streptococcus Milleri” group: Analysis of 19 cases. Clin Inf Dis. 1994; 19:704-13.

12. Molina JM, Leport C, Bure A, et al. Clinical and bacterial features of infections caused by Streptococcus miller. Scand J Infect Dis 1991;23:659-66.

13. Murray HW, Gross KC, Masur H, Roberts RB. Serious infections caused by Streptococcus miller. Am J Med 1978;64: 759-64.

14. Economopoulos GC, Scherzer HH, Gryboski WA. Successful management of mediastinitis, pleural empyema, and aortopulmonary fistula from odontogenic infection. Ann Thorac Surg 1983;35:184-7.

15. Bonapart IE, Stevens HP, Kerver AJ, Rietveld AP. Rare complications of an odontogenic abscess: Mediastinitis, thoracic empyema and cardiac tamponade. J Oral Maxillofac Surg 1995; 53:610- 3.

16. Roberts JR. Minimally invasive surgery in the treatment of empyema: Intraoperative decision making. Ann Thorac Surg 2003; 76:225-30.

17. Gossot D, Stern JB, Galetta D, et al. Thoracoscopic management of postpneumonectomy empyema. Ann Thorac Surg 2004;78:273- 6.

18. Hope WW, Bolton WD, Stephenson JE. The utility and timing of surgical intervention for parapneumonic empyema in the era of video- assisted thoracoscopy. Am Surg 2005;71:512-4.

19. Luh SP, Chou MC, Wang LS, et al. Video-assisted thoracoscopic surgery in the treatment of complicated parapneumonic effusions or empyemas: Outcome of 234 patients. Chest 2005;127: 1427-32.

I. STELZMUELLER, M.D.,* M. BIEBL, M.D.,* N. BERGER, M.D.,* M. ELLER, M.D.,[dagger] J. MENDEZ, M.D.,[section] M. FILLE, M.D.,[dagger] K. ANGERER, M.D.,* T. SCHMID, M.D.,* I. LORENZ, M.D.,[double dagger] R. MARGREITER, M.D.,* H. BONATTI, M.D.*[section]

From the Departments of * General, Thoracic, and Transplant Surgery and [dagger] Hygiene, Microbiology, and Social Medicine, [double dagger]Anaesthesiology and Critical Care Medicine, Innsbruck Medical University, Innsbruck, Austria, and [section] Infectious Diseases, Mayo Clinic, Jacksonville, Florida

I.S. and M.B. authors contributed equally to this article.

Address correspondence and reprint requests to Ingrid Stelzmueller, M.D., Innsbruck Medical University, Department of General, Thoracic, and Transplant Surgery, Anichstrasse 35 6020, Innsbruck, Austria.

Copyright Southeastern Surgical Congress May 2007

(c) 2007 American Surgeon, The. Provided by ProQuest Information and Learning. All rights Reserved.