Retropharyngeal abscess in children: the emerging role of group A beta hemolytic Streptococcus.

Background: Because of a recent increase in the number of cases of retropharyngeal abscess (RPA) admitted to our hospital, we reviewed the incidence, microbiology, and treatment outcome of RPA during an 11-year period (1993-2003).

Methods: A retrospective review of medical records of children with RPA.

Results: Sixty-seven children (46 males) with RPA were identified, representing a 4.5-fold increase in incidence over a previous 12-year period. The majority (66%) of patients presented during the last 4 years. Computed tomography revealed inflammatory or ring enhancing lesion in all patients. Abscess drainage was performed in 51 (76%) patients. A total of 101 isolates (84 aerobes, 17 anaerobes) were recovered from 41 specimens (a mean of 2.5 isolates per specimen). Group A beta hemolytic streptococcus (GABHS) was recovered from 22 (54%) of 41 specimens compared with 6 (35%) of 17 over the previous 12 years. Treatment included IV antibiotics: ampicillin/sulbactam or clindamycin plus either cefuroxime or ceftri-axone, followed by oral amoxicillin/clavulanate or clindamycin. All patients recovered.

Conclusions: RPA, an aerobic/anaerobic polymicrobial infection, is increasing in frequency and is associated with increased recovery of GABHS in our patients. Whether this rise in incidence is due to increased invasiveness of GABHS strains is to be determined.

Key Words: retropharyngeal abscess, group A streptococcus, pharyngitis 

**********

Deep neck infections include those involving the retropharyngeal and parapharyngeal spaces. Retropharyngeal abscess (RPA) is a relatively uncommon infection found more frequently in children and has potential serious complications. Abscess formation usually follows upper respiratory tract infection with subsequent suppuration of the lymph nodes in the retropharyngeal space. These nodes usually atrophy by 4 years of age. (1) In children, retropharyngeal lymphadenitis and RPA occur secondary to an adjacent focus of infection such as pharyngitis, tonsillitis, dental infection and adenitis. Other, less frequent causes include penetrating trauma, vertebral osteomyelitis, parotitis, mastoiditis and sinusitis. (2-4) RPA in children is polymicrobial and isolated organisms usually reflect mouth and nasopharyngeal flora. (5-7)

During the last 2 years, we noted an increase in the number of cases of RPA in children at our institution. This study was undertaken to identify the incidence of RPA among children in our institution, to identify the predisposing factors and the microbiology of RPA and to evaluate the treatment and clinical outcome. We present the results of a retrospective investigation of children with retropharyngeal abscess during 11 consecutive years.

Patients and Methods

Children admitted to our hospital with the diagnosis of RPA between February 1993 and December 2003 are the subjects of this study. Children's Hospital of Michigan is a tertiary care teaching pediatric institution with a 220-bed capacity. For this study, we examined the patient's demographic data, laboratory characteristics including microbiologic and radiographic data, clinical course and response to treatment.

Purulent material obtained at surgery from the retropharyngeal abscess was transported anaerobically to the microbiology laboratory within 30 minutes and inoculated onto plates supportive for aerobic and anaerobic growth. Gram-stained smears were obtained on all specimens. For aerobic growth, the material was plated onto sheep blood agar plates, colistin-nalidixic acid (CNA) plates, chocolate agar plates and MacConkey agar plates. All plates were incubated at 37[degrees]C, the MacConkey plates aerobically and the other plates under 5% C[O.sub.2]. All plates were examined at 24 and 48 hours. Aerobes were identified using conventional methods. Susceptibility testing was performed using the minimal inhibitory concentration method.

For anaerobes, the material was plated on prereduced kanamycin-vancomycin agar plates, anaerobic blood agar plates, CNA plates and into brain heart infusion broth. The plates were incubated in anaerobic jars (Gas Pak) and examined at 48 and 96 hours. The broth was incubated for 4 days. Anaerobes were identified by colony morphology, Gram stain, automated biochemical testing (Innovative Diagnostic Systems) and gas chromatography.

Results

During the 11-year period from 1993 through 2003, 67 children (46 males, 21 females) were admitted to our hospital with the diagnosis of retropharyngeal abscess. Twelve had concomitant parapharyngeal abscess and one had concomitant peritonsillar abscess. The age range was 5 months to 17 years (mean 58 months, median 47 months). Thirty six (54%) were 4 years of age or younger. The mean number of all hospital admissions during the same period was 12,314 per year. The 67 cases of RPA represent an incidence of 4.94 per 10,000 patients. This also represents a 4.5-fold increase in incidence when compared with a similar study conducted at our institution for the 12-year period of 1978 to 1989. (5) The majority of cases (44 [66%]) presented during the last 4 years of the study period (Fig.).

The majority of our patients had associated pharyngitis or tonsillitis (97%) and cervical lymphadenitis (87%). One patient had a concomitant dental abscess, one had a foreign body and another had tuberculous osteomyelitis of the cervical spine. The child with tuberculosis was previously reported and excluded from the current analysis. (6)

The most common signs and symptoms were fever in 91% of patients, neck pain, swelling or torticollis in 91%, dysphagia and difficulty swallowing in 85%, sore throat, pharyngitis or tonsillitis in 97%, cervical lymphadenitis in 87% and difficulty in breathing or stridor in 21%. Surgical management that included aspiration or incisional drainage was performed in 51 of the 67 (76%) patients. Pus was obtained from 41 children. A total of 101 bacterial isolates were recovered from 41 specimens (Table). Gram stained smears were positive on 27 of 41 specimens for 1 to 3 organisms each. Cultures confirmed the interpretation of the Gram stained smears in 20 of 27 specimens. The number of isolates per specimen varied from 1 to 6 with a mean of 2.5 isolates/specimen. Eighty four of the bacterial isolates were aerobes and 17 were anaerobes. All anaerobes recovered were mixed with aerobes. Group A beta hemolytic streptococcus (GABHS) was recovered from 22/41 (54%) specimens and was second only to viridans streptococcus as the most commonly isolated aerobe. GABHS was the only isolate recovered from 7 samples. In addition, GABHS was isolated from 4/6 throat cultures including two children from whom attempted aspirations of the RPA yielded no pus. Another child who was managed medically and who had no cultures obtained was later diagnosed with poststreptococcal reactive arthritis.

[FIGURE OMITTED]

The mean blood leukocyte count was 22,500/[mm.sup.3] (range: 7,700-44,900). Lateral neck films were obtained on 45 children and all but 2 showed bulging or enlarged retropharyngeal space. Computed tomography (CT) of the neck showed hypodensity suggestive of inflammation or ring enhancing abscess/fluid collection in all study patients. One patient had a CT scan that showed a nonenhancing mass; however, pus was obtained by incisional drainage. Five children had CT findings of abscess formation; however, no fluid was obtained on attempted aspiration in these patients.

Eighteen patients had received oral antibiotics for a period of 1 to 7 days before admission. One patient received a dose of intramuscular ceftriaxone. During hospitalization, all patients were treated with intravenous (IV) antibiotics before undergoing surgery. Antibiotics were administered from 1 hour to 7 days before sample collection. The most commonly used antibiotics were ampicillin/sulbactam or clindamycin in combination with either cefuroxime or ceftriaxone. The most frequently used oral antibiotics were amoxicillin/clavulanate and clindamycin. One child required intubation to maintain airway patency and another required emergency tracheostomy because of failure to intubate following incisional drainage. Two other children were intubated because of postoperative respiratory distress. All patients recovered without sequelae.

Discussion

Retropharyngeal abscess in children results from suppuration of the paramedian retropharyngeal chains of lymph nodes that receive drainage from the nasopharynx, posterior paranasal sinuses and adenoids. Infection of the retropharyngeal lymph nodes progresses through the stages of cellulitis, phlegmon and finally, abscess formation. The predisposing factors for development of RPA in children are classically upper respiratory tract infections, especially oropharyngeal infections. This most frequently involves extension of infection from pharyngitis, tonsillitis and adenitis to the retropharyngeal space. In contrast, RPA in adults is usually secondary to trauma, foreign body or occurs as a complication of dental infection.

Our results support previous reports and underscore the mixed aerobic/anaerobic polymicrobial nature of RPA in children. The predominant aerobes isolated from RPA in two previous studies were viridans streptococci (alpha and gamma hemolytic), Staphylococcus aureus, Haemophilus influenzae, Neisseria species and GABHS. The predominant anaerobes were Bacteroides species, Peptostreptococcus species and Fusobacterium species. (5,7,8) Brook reported a mean of 5.6 isolates/specimen from 14 patients and all had anaerobic isolates.7 Aerobes were isolated, mixed with anaerobes in 12 of 14 patients (mean, 1.9/specimen). In a previous study from our institution, aerobes were recovered from all 17 patients and anaerobes were isolated in 9 of 17 patients. (5) The reason for the relatively lower number of isolates per specimen in the present study is not clear. This could be due to the effect of the IV antibiotic treatment therapy that our patients had received before collection of abscess specimens. However, specimen collection and handling may have influenced the recovery of some anaerobic organisms.

Our results suggest that the incidence of RPA is increasing in our patient population. A sharp increase was noted in the last 4 years of the study period. GABHS was second only to viridans streptococcus as the most common organism recovered from our patients. It was isolated from 54% of patients, a rate that was higher than the previously reported rate of 35% from our institution (although the difference was not statistically significant, P = 0.32) and also higher than that in any previous study. (2,5,7) Brook reported isolation of GABHS from 3 of 15 (21%) patients with RPA. (7) The clinical manifestations of GABHS infection in children vary with age. (8,9) Pharyngitis is the most common presentation among children 5 to 15 years old. Infection in children younger than 3 years of age is frequently associated with "streptococcosis" manifested by low-grade fever, protracted nasal symptoms and cervical lymphadenopathy. In our study, GABHS was isolated from children with RPA of different age groups, including those as young as 5 months of age.

The recent increase in the incidence of RPA in our patient population is possibly secondary to increased rates of GABHS oropharyngeal infection. The incidence of severe invasive GABHS disease such as necrotizing fasciitis and toxic shock syndrome has increased in recent years. (10,11) Several virulence factors have been attributed to the invasiveness of GABHS. The main factor appears to be the antiphagocytosis M-protein and the extracellular toxins. (12) Invasiveness of GABHS has been primarily associated with serotypes MI and M3 and with GABHS strains producing streptococcal pyogenic exotoxins A and B. (13,14) Serotyping and characterization of GABHS strains that were recovered from our patients were not performed. Whether the recent increase in incidence of RPA in our patients was related to increased incidence of infection with invasive GABHS strains is not known.

Lateral neck x-ray findings suggestive of RPA include soft tissue swelling and loss of cervical lordosis. (15) Wholey et al (16) suggested evaluating abnormalities by measuring the retropharyngeal and retrotracheal spaces from the anterior inferior aspects of C2 and C6 respectively. In a child 15 years old or younger, a retropharyngeal space exceeding 7 mm (normal 2-7 mm) and a retrotracheal space exceeding 14 mm (normal 5-14 mm) suggests a mass, an inflammatory process, or bleeding. The presence of gas in the soft tissue is suggestive of RPA. Only one of our patients had gas in the prevertebral soft tissue. Contrast-enhanced CT is particularly helpful in identifying the different stages of retropharyngeal inflammation. (17-19) However, limitations of CT include inability to differentiate between phlegmon and abscess. In one study, the positive predictive value was 84% and the negative predictive value was 44%. (4)

Antibiotic treatment of RPA should be directed against the likely causative organisms. Because these infections are often mixed and [beta]-lactamase producing organisms are frequently isolated, the use of penicillin or ampicillin alone is probably inadequate. (2,4,5,7,20) Similarly, use of clindamycin alone in empiric treatment of RPA that has not been drained may not be adequate due to lack of coverage of potential Gram negative organisms such as H. parainfluenzae or non-typeable H. influenzae. However, Gram negative aerobic isolates were not isolated frequently in our study patients (7 of 101). Al-Sabah et al (21) reported a clinical success rate of 100% with clindamycin use alone in 68 patients with retropharyngeal infections. The most optimal antibiotic treatment of RPA in children has not been established. All our patients received a second or third generation cephalosporin with clindamycin or alternatively received ampicillin/sulbactam. All our patients recovered. Both regimens provide excellent in vitro coverage for mixed infections. (22) It is possible that clindamycin use alone could have provided adequate antibiotic coverage in our patients with RPA.

The increased incidence of GABHS infection in our patients may have implications regarding empiric antibiotic coverage. Whether a penicillin-containing regimen such as ampicillin/sulbactam could provide better coverage of the RPA that was preceded by GABHS infection than other regimens remains unclear. However, aggressive GABHS infections, such as necrotizing fasciitis, respond less well to penicillin treatment alone. (23,24) Experimental and clinical studies suggest that patients with deep soft tissue infections caused by GABHS have a more favorable prognosis if initial treatment included a protein-inhibiting antibiotic such as clindamycin compared with treatment with a cell wall-inhibiting agent alone. (10) This was explained by the "Eagle effect" which is believed to be associated with decreased effectiveness of penicillin in high inoculum GABHS infections secondary to decreased concentrations of penicillin-binding proteins in the stationary phase of growth. Whether clindamycin use alone or in combination with a second or third generation cephalosporin provides superior coverage to ampicillin/sulbactam in RPA induced by GABHS infection remains unclear.

The role of surgical drainage in the management of RPA is controversial. A relatively large number of our patients (24%) were treated with antibiotic therapy alone. The rate of success with medical therapy alone has been variable and probably reflects the stage of inflammation in the retropharyngeal space or the size of the abscess at the time of initiating therapy. Patients with cellulitis without abscess formation can be managed with antibiotics alone. (2,19) Broughton et al (25) described 6 patients who had round or oval cystic lesions consistent with an abscess and who were managed successfully without any drainage procedure. More recently, in a series of 64 patients with RPA, Craig and Schunk have demonstrated successful conservative management without surgery in 58% of patients. (26) Even in patients with defined abscess, 37% were managed successfu lly with antibiotics alone. (26) In their series of retropharyngeal infections, Al-Sabah et al have also demonstrated a high response rate of 75% with medical therapy alone. (21) In many cases, both in our series and other studies, when surgical drainage was performed only small amounts of pus were obtained. It is possible that more patients in our study could have been treated successfully without drainage procedures. Alternatively, patients who appear to have an abscess on CT scan and who have significant respiratory distress should undergo drainage of the abscess after securing the airway. Definite guidelines on the management of children who have CT findings consistent with an abscess but who are not in respiratory distress are not clear. No controlled studies have been performed to compare antibiotic treatment alone versus combined surgical drainage and antibiotic therapy in these patients. A reasonable approach, as has been done for some of our patients, would be to give a trial of IV antibiotics in stable patients and to monitor clinical response over 24 to 48 hours. If there is no clinical improvement, then surgical drainage should be considered.

Suppurative complications of RPA include rupture of the abscess with subsequent aspiration pneumonia or asphyxia. Extension of the inflammatory process can result in mediastinitis or pneumonia. Vascular complications include jugular thrombophlebitis. Erosion of the carotid or vertebral arteries may result in hemorrhage. Asphyxia is a rare complication of RPA and occurs as a result of direct airway pressure or sudden rupture.

Our findings support the concept that retropharyngeal abscess in children results from extension of oropharyngeal infection into the retropharyngeal lymph nodes. In addition, the increased frequency of GABHS infection in our patients suggests that this organism triggers the inflammatory process that leads to formation of polymicrobial retropharyngeal abscess. Whether increased invasiveness of GABHS infection is contributing to the high incidence of retropharyngeal abscess during the last 4 years is not clear. Further prospective studies are needed to clarify the role of GABHS in the pathogenesis of retropharyngeal abscess in children.

References

1. Seid AB, Dunbar JS, Cotton RT. Retropharyngeal abscesses in children revisited. Laryngoscope 1979;89:1717-1724.

2. Thompson JW, Cohen SR, Reddix P. Retropharyngeal abscess in children: a retrospective and historical analysis. Laryngoscope 1988;98 (6 Pt 1):589-592.

3. Goldenberg D, Golz A, Joachims HZ. Retropharyngeal abscess: a clinical review. J Laryngol Otol 1997;111:546-550.

4. Barratt GE, Koopmann CF Jr, Coulthard SW. Retropharyngeal abscess--a ten-year experience. Laryngoscope 1984;94:455-463.

5. Asmar BI. Bacteriology of retropharyngeal abscess in children. Pediatr Infect Dis J 1990;9:595-597.

6. Abuhammour WM, Asmar BI. A sixteen-year-old male youth with fever, dysphagia and weight loss. Pediatr Infect Dis J 1999; 18:732, 748-749.

7. Brook I. Microbiology of retropharyngeal abscesses in children. Am J Dis Child 1987;141:202-204.

8. Wald ER. Expanded role of group A streptococci in children with upper respiratory infections. Pediatr Infect Dis J 1999;18:663-665.

9. Quinn RW, Vander Zwaag R, Lowry PN. Acquisition of group A streptococcal M protein antibodies. Pediatr Infect Dis 1985;4:374-378.

10. Zimbelman J, Palmer A, Todd J. Improved outcome of clindamycin compared with beta-lactam antibiotic treatment for invasive Streptococcus pyogenes infection. Pediatr Infect Dis J 1999;18:1096-1100.

11. Bronze MS, Dale JB. The reemergence of serious group A streptococcal infections and acute rheumatic fever. Am J Med Sci 1996;311:41-54.

12. Marrack P, Kappler J. The staphylococcal enterotoxins and their relatives. Science 1990;248:705-711.

13. Zervas SJ, Zemel LS, Romness MJ, et al. Streptococcus pyogenes pyomyositis. Pediatr Infect Dis J 2002;21:166-168.

14. Kaplan EL. Recent epidemiology of group A streptococcal infections in North America and abroad: an overview. Pediatrics 1996;97 (6 Pt 2):945-948.

15. Wong YK, Novotny GM. Retropharyngeal space--a review of anatomy, pathology, and clinical presentation. J Otolaryngol 1978;7:528-536.

16. Wholey MH, Bruwer AJ, Baker HL Jr. The lateral roentgenogram of the neck; with comments on the atlanto-odontoid-basion relationship. Radiology 1958;71:350-356.

17. Morrison JE Jr, Pashley NR. Retropharyngeal abscesses in children: a 10-year review. Pediatr Emerg Care 1988;4:9-11.

18. Endicott JN, Nelson RJ, Saraceno CA. Diagnosis and management decisions in infections of the deep fascial spaces of the head and neck utilizing computerized tomography. Laryngoscope 1982;92 (6 Pt 1): 630-633.

19. Holt GR, McManus K, Newman RK, et al. Computed tomography in the diagnosis of deep-neck infections. Arch Otolaryngol 1982;108:693-696.

20. Yeoh LH, Singh SD, Rogers JH. Retropharyngeal abscesses in a children's hospital. J Laryngol Otol 1985;99:555-566.

21. Al-Sabah B, Bin Salleen H, Hagr A, et al. Retropharyngeal abscess in children: 10-year study. J Otolaryngol 2004;33:352-355.

22. Reinhardt JF, Johnston L, Ruane P, et al. A randomized, double-blind comparison of sulbactam/ampicillin and clindamycin for the treatment of aerobic and aerobic-anaerobic infections. Rev Infect Dis 1986;8 (Suppl 5):S569-575.

23. Brogan TV, Nizet V, Waldhausen JH, et al. Group A streptococcal necrotizing fasciitis complicating primary varicella: a series of fourteen patients. Pediatr Infect Dis J 1995;14:588-594.

24. Doctor A, Harper MB, Fleisher GR. Group A beta-hemolytic streptococcal bacteremia: historical overview, changing incidence, and recent association with varicella. Pediatrics 1995;6 (3 Pt 1):428-433.

25. Broughton RA. Nonsurgical management of deep neck infections in children. Pediatr Infect Dis J 1992;11:14-18.

26. Craig FW, Schunk JE. Retropharyngeal abscess in children: clinical presentation, utility of imaging, and current management. Pediatrics 2003; 111 (6 Pt 1): 1394-1398.

Horse sense is the thing a horse has which keeps it from betting on
people.
--W.C. Fields

Nahed M. Abdel-Haq, MD, Ashraf Harahsheh, MD, and Basim I. Asmar, MD

From Children's Hospital of Michigan and the Carman and Ann Adams Department of Pediatrics, Wayne State University School of Medicine, Detroit MI.

Reprint requests to Nahed Abdel-Haq, MD, Division of Infectious Diseases Children's Hospital of Michigan 3901 Beaubien Boulevard Detroit, MI 48201. Email: [e-mail]

This work was presented in part at the annual meeting of the Infectious Diseases Society of America, San Diego, CA, October 9-12, 2003. Abstract # 785

Accepted March 22, 2006.

RELATED ARTICLE: Key Points

* Retropharyngeal abscess (RPA) in children is an aerobic/anaerobic polymicrobial infection.

* RPA is increasing in frequency and is associated with increased recovery of group A beta hemolytic streptococcus (GABHS).

* GABHS pharyngitis may cause inflammation of the retropharyngeal lymph nodes and subsequent polymicrobial abscess development.

* Ampicillin/sulbactam or clindamycin plus cefuroxine or ceftriaxone are effective antibiotics in combination with surgical drainage when indicated.

Table. Bacterial isolates in 67 children with retropharyngeal abscess

Isolates                                          No. of Isolates

Aerobic and facultative
  Gram-positive cocci
    GABHS                                         22
    Viridans streptococci                         24
    Streptococcus nonspeciated                     6
    Micrococcus species                            9
    Coagulase negative Staphylococcus              3
    Staphylococcus aureus                          2
    Streptococcus mitis                            1
  Gram-negative cocci
    Neisseria species (nonpathogenic)              6
  Gram-negative bacilli
    Eikenella species                              2
    Haemophilus parainfluenzae                     2
    Haemophilus influenzae (nontypeable)           1
    Klebsiella pneumoniae                          1
    Gram-negative bacilli (nonlactose fermenter)   1
  Gram-positive bacilli
    Corynebacterium species                        4
Total number of aerobes                           84
Anaerobic
  Anaerobic cocci
    Peptostreptococcus species                     2
  Gram-negative bacilli
    Prevotella species                             8
    Prevotella intermedia                          2
    Bacteroides species                            2
    Fusobacterium varium                           2
    Prevotella oralis                              1
Total number of anaerobes                         17