The Significance of Streptococcus Anginosus Group in Intracranial Complications of Pediatric Rhinosinusitis

ORIGINAL ARTICLE The Significance of Streptococcus anginosus Group in Intracranial Complications of Pediatric Rhinosinusitis

Michael W. Deutschmann, MD; Devon Livingstone, BSc; John J. Cho, MD; Otto G. Vanderkooi, MD; James T. Brookes, MD

Objective: To assess the significance of the Streptococ- sinusitis due to S anginosus group bacteria compared with cus anginosus group in intracranial complications of pe- other bacteria. diatric patients with rhinosinusitis. Results: Infection caused by the S anginosus group re- Design: Retrospective cohort study. sulted in more severe intracranial complications (P=.001). In addition, patients with S anginosus group–associated Setting: Tertiary pediatric hospital. infections were more likely to require neurosurgical intervention (PϽ.001) and develop long-term neurologic Patients: A 20-year review of medical records identi- deficits (P=.02). Intravenous antibiotics were adminis- fied patients with intracranial complications resulting from tered for a longer duration (PϽ.001) for S anginosus rhinosinusitis. In the 50 cases identified, S anginosus was group–associated infections. the most commonly implicated bacterial pathogen in 14 (28%). Documented data included demographics, cul- Conclusions: Rhinosinusitis associated with the S an- tured bacteria, immune status, sinuses involved, type of ginosus group should be considered a more serious in- intracranial complication, otolaryngologic surgical and fection relative to those caused by other pathogens. Strep- neurosurgical intervention, type and duration of antibi- tococcus anginosus group bacteria are significantly more otics used, and resulting neurologic deficits. Complica- likely than other bacteria to cause more severe intracra- tions and outcomes of cases of S anginosus group– nial complications and neurologic deficits and to re- associated rhinosinusitis were compared with those of quire neurosurgical intervention. A low threshold for in- other bacteria. tervention should be used for infection caused by this pathogen. Main Outcome Measures: The severity and outcomes of intracranial complications of pediatric rhino- JAMA Otolaryngol Head Neck Surg. 2013;139(2):157-160

EDIATRIC RHINOSINUSITIS IS cus and Staphylococcus are frequently as- relatively common. Most sociated with these diseases.5 cases are caused by Strepto- The Streptococcus anginosus group, coccus pneumoniae, Hae- previously known as the Streptococcus mophilus influenzae, Morax- milleri group, is from the Streptococcus ella catarrhalis, and, less commonly, family, which is known to be a cause of P 6 Author Affiliations: Sections of Staphylococcus aureus and Streptococcus serious infections. This group includes Otolaryngology–Head and Neck pyogenes.1 The increasing number of chil- S anginosus, S intermedius, and S constel- Author Aff Surgery (Drs Deutschmann, dren attending day care facilities has led latus. The S anginosus group has been Otolaryngo Cho, and Brookes) and to an increasing incidence of this dis- shown5 to be a major causative factor of Surgery (Dr Pediatric Surgery (Dr Brookes), ease.2 Complications associated with intracranial infections in adults, particu- Cho, and B Department of Surgery, and Pediatric Su Section of Infectious Diseases, rhinosinusitis, although potentially dev- larly those in which acute bacterial Departmen Department of Pediatrics, and astating, are rare. Intracranial and orbital rhinosinusitis is the inciting infection. Section of I Departments of Microbiology complications occur via direct extension The Streptococcus family produces many Departmen and Infectious Diseases and from the sinuses or hematogenous spread. enzymes and works synergistically with Departmen Pathology and Laboratory Approximately 3% of pediatric patients anaerobic bacteria to cause tissue necro- and Infectio Medicine (Dr Vanderkooi) hospitalized with rhinosinusitis develop sis and abscess formation.7,8 Pathology a University of Calgary, Calgary, 3 Medicine (D Alberta, Canada. an intracranial complication. These in- At the Alberta Children’s Hospital, University Mr Livingstone is a medical fections include meningitis, epidural ab- Calgary, we have observed several severe Alberta, Ca student at the University scess, subdural abscess, brain abscess, and intracranial complications from rhino- Livingstone of Calgary. cavernous sinus thrombosis.4 Streptococ- sinusitis caused by the S anginosus at the Univ

JAMA OTOLARYNGOL HEAD NECK SURG/ VOL 139 (NO. 2), FEB 2013 WWW.JAMAOTO.COM 157

No. (%) Streptococcus anginosus Other Organisms Characteristic (n = 14) (n = 36) P Value Age, mean (SD), y 10.8 (3.6) 9.1 (5.4) .28 Sex Male 8 (57) 28 (78) .17 Female 6 (43) 8 (22) Previous adenoidectomy No 14 (100) 32 (89) Yes 0 3 (8) .55 Adenoids not present 0 1 (2) Duration of hospital stay, mean (SD), d 14.4 (11.2) 15.4 (11.8) .79 Causative organismb Streptococcus anginosus 14 (100) Streptococcus pneumoniae 10 (28) Staphylococcus aureus 2 (6) Group B Streptococcus 2 (6) Group A ␤-hemolytic Streptococcus 1 (3) Coagulase-negative Staphylococcus 2 (6) Propionobacterium 1 (3) Other bacterium 5 (14) No growth 17 (47) Sinus involvementc Frontal 9/14 (64) 14/24 (58) .13 Maxillary 9/14 (64) 26/36 (72) .73 Sphenoid 6/14 (42) 13/26 (50) .75 Ethmoid 10/14 (71) 27/36 (75) Ͼ.99

a One patient was excluded because of death from intracranial complications of sinusitis upon admission, and 4 were excluded because of a compromised immune system. b Some patients’ cultures grew multiple organisms. c Patients with rhinosinusitis localization information unavailable, those younger than 5 years (sphenoid), and those younger than 7 years (frontal) were excluded.

group. We reviewed the medical records of all patients Descriptive statistics were tabulated for all variables in the who had an intracranial complication of rhinosinusitis presence and absence of S anginosus group rhinosinusitis. A to determine whether the disease severity and outcomes 2-sided Fisher exact test was conducted on all categorical vari- were worse when the S anginosus group was the caus- ables (sex, adenoidectomy, sinus involvement, intracranial com- ative organism. plication, concurrent orbital complication, otolaryngologic surgical intervention, neurosurgical intervention, and neurologic deficits). Frequency distribution tables were created for con- METHODS tinuous variables (length of hospital stay, age, and duration of antibiotic use) including analysis of variance, skew, and kurto- A review of medical records from patients at the Alberta Chil- sis. A 2-sample unpaired t test was conducted to determine the dren’s Hospital was conducted. Ethics approval was received statistical significance of relationships between S anginosus group– from the Conjoint Health Research Ethics Board of the Uni- associated infection and continuous primary outcome vari- versity of Calgary. ables. Ninety-five percent confidence intervals were calculated Demographic and personal health information was col- on the basis of binomial distribution. Simple logistic regression lected from hospital records and health records, including data analysis was performed to assess variables that may have con- regarding intracranial complications of pediatric rhinosinusi- founded a significant outcome on univariate analysis. tis as well as the medical and surgical interventions under- taken to treat the infection in each patient. RESULTS Children aged 1 day to less than 18 years were included. All patients must have been admitted to Alberta Children’s Hos- pital between January 1, 1991, and December 31, 2010, with Fifty-five patients meeting the inclusion criteria were iden- an intracranial complication of acute bacterial rhinosinusitis. tified; 4 patients were excluded because they were im- Patients who were immunocompromised when they devel- munocompromised, and 1 patient was excluded be- oped the infection were excluded from analysis. cause of death during admission to the hospital. There In all statistical analyses, variables and outcomes of pa- were no significant differences in sex, age, duration of tients with rhinosinusitis caused by the S anginosus group were compared with those caused by other organisms. Primary out- hospitalization, or location of rhinosinusitis between the come variables included age, sex, type of intracranial compli- S anginosus group and the comparative group. Fourteen cation, duration of hospital stay, duration of antibiotic use, sur- patients had rhinosinusitis due to the S anginosus group gical intervention requirement, and permanent neurologic (Table 1). It was the most commonly implicated bac- deficits due to intracranial complications of the infection. terial pathogen, accounting for 14 of the 50 cases (28%)

JAMA OTOLARYNGOL HEAD NECK SURG/ VOL 139 (NO. 2), FEB 2013 WWW.JAMAOTO.COM 158

No. (%) Streptococcus Other anginosus Organisms P Streptococcus Other Characteristic (n = 14) (n = 36) Value anginosus Organisms P Intervention (n = 14) (n = 36) Value Intracranial complications Neurosurgical Meningitis 2 (14) 21 (58) Craniotomy with 4 (28) 5 (14) I&D Epidural abscess 5 (36) 5 (14) Craniectomy with 4 (28) 0 Ͻ.001 Subdural abscess 5 (36) 3 (8) I&D Brain abscess 1 (7) 2 (6) Other 1 (7) 0 Cavernous sinus 1 (7) 1 (3) .001 Total No. 9 5 thrombosis ENT surgical Pott puffy tumor 2 (14) 0 Endoscopic sinus 5 (36) 3 (8) Other 2 (14) 7 (20) surgery Total excluding 16 18 Orbital 00 meningitis .15 a decompression Total No. 18 39 Frontal sinus surgery 0 1 (3) Concurrent orbital CSF leak closure 0 1 (3) complication Other 1 (7) 2 (6) Periorbital cellulitis 2 (14) 6 Total No. 6 7 Orbital cellulitis 1 (7) 1 Ͼ Antibiotic duration, Subperiosteal 01.99 mean (SD), d abscess Intravenous 41.6 (4.6) 21.8 (16.0) Ͻ.001 Orbital abscess 0 0 Oral 7.4 (15.1) 8.2 (12.6) .85 Total No. 3 8 Antibiotic used Permanent neurologic deficits Amoxicillin- 1 (7) 6 (17) clavulanate Vision loss 1 (7) 0 Cefotaxime 5 (36) 29 (80) Reduced EOM 1 (7) 0 Ceftriaxone 4 (28) 7 (19) Unilateral paralysis 1 (7) 0 .02 Clindamycin 2 (14) 4 (11) Death 0 0 Meropenem 6 (43) 1 (3) Total No. 3 0 Metronidazole 11 (78) 8 (22) Penicillin G or V 5 (36) 6 (17) Abbreviation: EOM, extraocular movement. a Vancomycin 12 (86) 20 (56) Some patients had multiple intracranial complications.

Abbreviations: CSF, cerebrospinal fluid; ENT, otolaryngologic; associated infections were more likely to develop I & D, incision and drainage. complications other than meningitis (P = .001) (Table 3). Only 2 of 14 intracranial complications (14%) resulting from rhinosinusitis due to the S anginosus group were men- of acute rhinosinusitis with intracranial complication. ingitis compared with 21 of 36 patients (58%) when the Table 1 reports the descriptive statistics of patient de- complications were caused by other organisms. Patients mographics and characteristics analyzed. with S anginosus group–associated rhinosinusitis were also Rhinosinusitis with intracranial complications of more likely to develop permanent neurologic deficits all causes was significantly more common in males (36 (P = .02) (Table 3). There was no significant difference be- patients [72%]) than in females (14 [28%]) (P = .03). tween the groups for concurrent orbital infection. Streptococcus pneumoniae was the second most common Eight of the 10 S pneumoniae cases occurred after 2002, bacterial cause of rhinosinusitis with intracranial com- when the 7-valent pneumococcal conjugate vaccine was plications, accounting for 10 cases (20%). A descriptive introduced in Alberta. Interestingly, 10 of the 14 S an- statistical analysis of the neurosurgical, otolaryngologic ginosus cases occurred after 2002. surgical, and antibiotic interventions is reported Multivariable analysis was performed to look at the in Table 2. risk of surgical intervention during infections caused by Patients with S anginosus group–associated rhinosinu- S anginosus compared with all other bacteria. A simple sitis were more likely to require neurosurgical interven- logistic regression controlled for severity of the infection compared with patients with infections caused by tion (ie, abscess vs nonabscess) showed increased odds other organisms (9 patients [64%] vs 5 [14%]; P Ͻ .001). of surgical intervention when S anginosus was the patho- There was no significant difference between the groups gen (odds ratio, 5.5; 95% CI, 0.8-40.5; P = .09). for otolaryngologic surgical intervention. Intravenous antibiotics were administered a mean of 20 days longer in the S anginosus group (95% CI, 11-28 days; t48 = 4.53; COMMENT P Ͻ .001). There was no significant difference between the groups for oral antibiotic administration. Streptococcus anginosus group is known to be particu- The complications and sequelae resulting from rhino- larly aggressive and is generally susceptible to penicil- sinusitis are listed in Table 3. All patients had intracra- lin, ampicillin, erythromycin, and tetracycline.5 Our study nial complications; however, those with S anginosus group– shows that intracranial complications resulting from pe-

JAMA OTOLARYNGOL HEAD NECK SURG/ VOL 139 (NO. 2), FEB 2013 WWW.JAMAOTO.COM 159

©2013 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 diatric rhinosinusitis caused by S anginosus are signifi- dren because of its prevalence in severe intracranial com- cantly more likely to require neurosurgical intervention plications, with more neurologic deficits, greater and a longer duration of intravenous antibiotics. In ad- likelihood of neurosurgical intervention, and longer du- dition, these infections are more likely to cause more se- rations of intravenous antibiotic therapy compared with vere intracranial complications and permanent neuro- other bacteria. Early and aggressive interventions are re- logic deficits. Simple logistic regression analysis showed quired when treating these virulent infections. that the risk of requiring surgical intervention is higher in the S anginosus group regardless of whether an ab- Submitted for Publication: August 10, 2012; final revi- scess is present. To date, there have been no studies com- sion received October 8, 2012; accepted October 29, 2012. paring the severity of S anginosus with other bacteria in Correspondence: James T. Brookes, MD, Section of Pe- causing intracranial complications associated with pediatric Surgery, Department of Surgery, Alberta Chil- diatric rhinosinusitis. One study5 found that patients with dren’s Hospital, 2888 Shaganappi Trail NW, Calgary, AB S anginosus–associated intracranial infections caused by T3B 6A8, Canada. rhinosinusitis were common; however, further vari- Author Contributions: All authors had full access to all the ables associated with S anginosus and other bacteria be- data in the study and take responsibility for the integrity yond the hospital length of stay were not analyzed. of the data and the accuracy of the data analysis. Study con- Extracellular enzymes are generated by the bacteria cept and design: Deutschmann, Cho, Vanderkooi, and and might be a factor in its potent virulence. It has been Brookes. Acquisition of data: Deutschmann and Living- shown9-11 that this group is able to produce several en- stone. Analysis and interpretation of data: Deutschmann, Liv- zymes, including hyaluronidase, deoxyribonuclease, and ingstone, Vanderkooi, and Brookes. Drafting of the manu- chondroitin sulfatase. These enzymes are capable of caus- script: Deutschmann, Livingstone, Cho, Vanderkooi, and ing tissue liquefaction and leading to abscess formation. Brookes. Critical revision of the manuscript for important in- It also has been shown7,8 that S anginosus group organ- tellectual content: Deutschmann, Cho, Vanderkooi, and isms can act synergistically with anaerobic bacteria to Brookes. Statistical analysis: Livingstone and Vanderkooi. cause worse abscesses. This may explain why the S an- Obtained funding: Deutschmann. Administrative, technical, ginosus group caused significantly worse intracranial in- and material support: Deutschmann. Study supervision: Cho, fections compared with other organisms. A recent study12 Vanderkooi, and Brookes. has shown that brain abscesses caused by the S angino- Conflict of Interest Disclosures: None reported. sus group can be treated successfully with cefotaxime and Previous Presentation: This work was presented as a poster metronidazole, as well as rifampin, if the Glasgow Coma at the spring meeting of the American Society of Pediatric Scale score is less than 11. Otolaryngology; April 19, 2012; San Diego, California. A likely explanation for S anginosus–associated infections requiring a longer duration of intravenous antibiotic therapy is the increased likelihood of abscesses. At REFERENCES our institution, patients with abscesses routinely re- 1. Wald ER. Microbiology of acute and chronic sinusitis in children and adults. Am ceive 4 to 6 weeks of intravenous antibiotics compared J Med Sci. 1998;316(1):13-20. with only 2 weeks for those with meningitis. This vari- 2. Johnson DL, Markle BM, Wiedermann BL, Hanahan L. Treatment of intracranial ability makes the type of infection a confounding factor abscesses associated with sinusitis in children and adolescents. J Pediatr. 1988; 113(1, pt 1):15-23. in determining whether the duration of intravenous 3. Clayman GL, Adams GL, Paugh DR, Koopmann CF Jr. Intracranial complica- antibiotic therapy is significantly different between the tions of paranasal sinusitis: a combined institutional review. Laryngoscope. 1991; S anginosus group and other organisms. 101(3):234-239. 4. Oxford LE, McClay J. Complications of acute sinusitis in children. Otolaryngol The H influenzae type b vaccine has dramatically re- Head Neck Surg. 2005;133(1):32-37. duced the incidence of meningitis caused by this organ- 5. Rankhethoa NM, Prescott CA. Significance of Streptococcus milleri in acute rhino- ism. Our study did not have any such cases, so we can- sinusitis with complications. J Laryngol Otol. 2008;122(8):810-813. 6. Ruoff KL. Streptococcus anginosus (“Streptococcus milleri”): the unrecognized not comment on whether the vaccine has changed the pathogen. Clin Microbiol Rev. 1988;1(1):102-108. incidence of H influenzae–associated meningitis result- 7. Nagashima H, Takao A, Maeda N. Abscess forming ability of Streptococcus mil- ing from rhinosinusitis. leri group: synergistic effect with Fusobacterium nucleatum. Microbiol Immunol. 1999;43(3):207-216. The widespread use of the multivalent pneumococ- 8. Shinzato T, Saito A. A mechanism of pathogenicity of “Streptococcus milleri group” cal vaccine has dramatically changed the bacteriology of in pulmonary infection: synergy with an anaerobe. J Med Microbiol. 1994;40 rhinosinusitis.13,14 This could explain the recent promi- (2):118-123. 9. Unsworth PF. Hyaluronidase production in Streptococcus milleri in relation to nence of S anginosus as a cause of intracranial infections infection. J Clin Pathol. 1989;42(5):506-510. secondary to rhinosinusitis. However, these studies did 10. Jacobs JA, Stobberingh EE. Hydrolytic enzymes of Streptococcus anginosus, Strep- not comment on the S anginosus group. Interestingly, our tococcus constellatus and Streptococcus intermedius in relation to infection. Eur J Clin Microbiol Infect Dis. 1995;14(9):818-820. study had more cases of S pneumoniae and S anginosus 11. Shain H, Homer KA, Beighton D. Degradation and utilisation of chondroitin sul- after the introduction of the 7-valent pneumococcal con- phate by Streptococcus intermedius. J Med Microbiol. 1996;44(5):372-380. jugate vaccine. It is possible these more recent S pneu- 12. Kowlessar PI, O’Connell NH, Mitchell RD, Elliott S, Elliott TS. Management of patients with Streptococcus milleri brain abscesses. J Infect. 2006;52(6):443-450. moniae cases were caused by subtypes that do not ex- 13. Benninger MS, Manz R. The impact of vaccination on rhinosinusitis and otitis press the antigens found in the vaccine. Studies would media. Curr Allergy Asthma Rep. 2010;10(6):411-418. be necessary to determine this. 14. Brook I, Gober AE. Frequency of recovery of pathogens from the nasopharynx of children with acute maxillary sinusitis before and after the introduction of vac- In conclusion, S anginosus group–associated rhino- cination with the 7-valent pneumococcal vaccine. Int J Pediatr Otorhinolaryngol. sinusitis should be considered a serious infection in chil- 2007;71(4):575-579.

JAMA OTOLARYNGOL HEAD NECK SURG/ VOL 139 (NO. 2), FEB 2013 WWW.JAMAOTO.COM 160