ORIGINAL ARTICLE Bacteriology of Acute and Chronic Frontal

Itzhak Brook, MD, MSc

spirates of 15 acutely and 13 chronically infected frontal sinuses were processed for aerobic and anaerobic bacteria. A total of 20 isolates (1.3 per specimen) were recov- ered from the 15 cases of acute frontal sinusitis, 16 aerobic and facultative isolates (1.1 per specimen) and 4 anaerobic isolates (0.3 per specimen). Aerobic and faculta- Ative organisms alone were recovered in 13 specimens (87%), and mixed aerobic and anaerobic bac- teria were recovered in 2 (13%). The predominant aerobic and facultative organisms were Hae- mophilus influenzae (6), Streptococcus pneumoniae (5), and Moraxella catarrhalis (3). A total of 32 isolates were recovered from the 13 cases (2.5 per patient) of chronic frontal sinusitis, 12 aerobic and facultative isolates (0.9 per specimen) and 20 anaerobic isolates (1.5 per specimen). Aerobic and facultative organisms only were recovered in 3 instances (23%), anaerobes only in 7 instances (54%), and mixed aerobic and anaerobic bacteria in 3 instances (23%). The predominant aerobic bacteria were gram-negative bacilli (H influenzae, Klebsiella pneumoniae, and Pseudomonas aeruginosa). The predominant anaerobes included species (8), species (6), and Fu- sobacterium species (4). These findings illustrate the microbiologic features of acute and chronic frontal sinusitis. Arch Otolaryngol Head Neck Surg. 2002;128:583-585

Frontal sinusitis is a potentially devastat- ACUTE FRONTAL SINUSITIS ing infection with a high frequency of in- tracranial complication. In contrast to max- A total of 20 isolates were recovered from illary sinusitis, the microbiologic features the 15 cases (1.3 per specimen), 16 aero- of frontal sinusitis are not well estab- bic and facultative isolates (1.1 per speci- lished, and only a few reports1-3 document men) and 4 anaerobic isolates (0.3 per the organisms isolated. The role of anaero- specimen). The number of isolates per speci- bic bacteria in this infection was also not men varied from 1 to 3. Aerobic and facul- well studied, although their recovery was tative organisms alone were recovered in 13 recorded in a few cases.4-6 This report de- specimens (87%), and mixed aerobic and scribes my experience during a 25-year anaerobic bacteria were recovered in 2 period of studying the aerobic and anaero- (13%). The predominant aerobic and fac- bic microbiologic characteristics of fron- ultative isolates were Haemophilus influen- tal sinusitis. zae (6), Streptococcus pneumoniae (5), and Moraxella catarrhalis (3) (Table). Seven ␤-lactamase–producing bacteria (BLPB) RESULTS were recovered from 5 specimens (33%). No differences were noted in the micro- CHRONIC FRONTAL SINUSITIS biologic findings between children and adults, the year of the infection, the ad- A total of 32 isolates were recovered from ministration of previous antimicrobial the 13 cases (2.5 per patient), 12 aerobic therapy, or the surgical approach (osteo- and facultative isolates (0.9 per speci- plastic vs endoscopic). men) and 20 anaerobic isolates (1.5 per specimen). The number of isolates var- From the Department of Pediatrics, Georgetown University School of Medicine, ied from 2 to 4. Aerobic and facultative Washington, DC. organisms only were recovered in 3 in-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 128, MAY 2002 WWW.ARCHOTO.COM 583

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 PATIENTS AND METHODS persisted for more than 1 month. Frontal surgery was per- formed by the osteoplastic flap approach in 13 patients and by an intranasal endoscopic approach in 15. Other si- The 28 patients included in the report were studied be- nuses were involved at the time of surgery in 6 patients (4 tween June 1975 and June 2001. Excluded are 9 addi- ethmoid and 2 maxillary). tional patients whose frontal sinusitis showed no bacterial The specimens were obtained during surgery, using growth. The patients were seen in the following hospitals: strict asepsis to avoid any contamination, and were trans- University of California Medical Center and County Medi- ported to the laboratory in a syringe sealed with a rubber cal Center in Los Angeles (1974-1977), Children’s Hospi- stopper after evacuation of the air or in an anaerobic trans- tal National Medical Center in Washington, DC (1977- port tube (Port-A-Cul; Baltimore Biological Laboratories, 1980), and Naval Hospital in Bethesda, Md (1977-1998). Cockeysville, Md). The time between the collection of ma- Similar microbiologic methods for collection and trans- terials and inoculation of the specimen was generally less portation of specimens were used in these institutions. than 60 minutes for syringes and less than 3 hours for the Patients’ ages ranged from 11 to 72 years (mean age, anaerobic transport tube. 38 years 8 months), and 18 were male. Five patients Specimens were inoculated onto 5% sheep’s blood, were children (younger than 18 years). Included in the chocolate agar, and MacConkey agar plates for aerobic and study were 15 patients with acute and 13 with chronic facultative organisms. The plates were incubated at 37°C sinusitis. Antimicrobial therapy was administered to 15 aerobically (MacConkey) or under 5% carbon dioxide (5% patients (54%) in the month before sample collection. sheep’s blood and chocolate agar) and examined at 24 and These patients included 10 with acute and 5 with chronic 48 hours. For anaerobes, the material was plated onto pre- sinusitis. reduced vitamin K1–enriched Brucella blood agar, an an- Only patients who fulfilled the following criteria were aerobic blood agar plate containing kanamycin sulfate and included in the study: typical clinical symptoms of sinus- vancomycin hydrochloride, an anaerobic blood plate con- itis (headache, fever, nasal drainage); positive radio- taining colistin sulfate and nalidixic acid, and an enriched 7 graphic findings; bacterial growth on cultures; biopsy speci- thioglycolate broth (containing hemin and vitamin K1). mens demonstrating acute or chronic inflammation of the The anaerobic plates were incubated in anaerobic jars (Gas- sinus mucosal lining; or clinical and radiologic findings com- Pak jars; Baltimore Biological Laboratories) and examined patible with frontal sinusitis followed by clinical and ra- at 48 and 96 hours. diologic improvement following surgery and treatment with Anaerobes were identified by techniques described pre- antibiotics. viously.7 Aerobic bacteria were identified by conventional Sinusitis was considered acute if the duration of symp- methods.8 ␤-Lactamase activity was determined by use of toms was less than 1 month and chronic if symptoms the chromogenic cephalosporin analog 87/312 method.9

stances (23%), anaerobes only in 7 instances (54%), and chronic maxillary sinusitis, where anaerobic bacteria are mixed aerobic and anaerobic bacteria in 3 instances (23%). the main isolates.6,10-12 These were mainly Peptostrepto- The predominant aerobic bacteria were gram-negative ba- coccus species, Fusobacterium species, and pigmented Pre- cilli (H influenzae, Klebsiella pneumoniae, and Pseudomo- votella and Porphyromonas species, all members of the nas aeruginosa) (Table). None of the patients who had oropharyngeal flora. P aeruginosa were diagnosed with cystic fibrosis. The pre- The frequent involvement of anaerobes in chronic dominant anaerobes included Prevotella species (8), Pep- frontal sinusitis may be related to the poor drainage and tostreptococcus species (6), and Fusobacterium species (4). increased intranasal pressure that develops during in- Twelve BLPB were recovered from 8 patients (62%). These flammation.13 This can reduce the oxygen tension in the included all Staphylococcus aureus isolates, 3 (75%) of 4 inflamed sinus14 by decreasing the mucosal blood flow15 of Fusobacterium species, and 4 (50%) of 8 of Prevotella and depressing the ciliary action.16 The lowering of the species. oxygen content and pH of the sinus cavity supports the growth of anaerobic organisms by providing them with 16 COMMENT an optimal oxidation-reduction potential. ␤-Lactamase–producing bacteria were isolated in 13 This study demonstrates the microbiologic features of (46%) of 28 patients. The recovery of BLPB is not surpris- acute and chronic frontal sinusitis. Since the number of ing, since more than half of our patients received antimi- patients included in this report was small and was col- crobial agents, including the ␤-lactams within the past 3 lected during a period of more than 25 years, prospec- months, which might have selected for these organisms. tive studies are required. The small number of patients Surgical drainage is essential in most cases, and an may also account for the lack of correlations between clini- initial, empiric, broad spectrum antimicrobial coverage cal (eg, age and previous antibiotic therapy) and micro- is required. However, the unique microbiologic features biologic findings. Similar to the study by Ruoppi et al,2 of acute and chronic frontal sinusitis and the recovery the present study also recovered S pneumoniae, H influ- of BLPB in approximately half of the specimens require enzae, and S aureus from patients with acute frontal si- adjusting an initial empiric therapy to a specific one when- nusitis. These findings are similar to the microbiologic ever possible. features of acute maxillary sinusitis, where S pneumo- The antimicrobial agents most commonly used to treat niae, H influenzae, and M catarrhalis predominate, and acute sinusitis include (with and without cla-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 128, MAY 2002 WWW.ARCHOTO.COM 584

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 bination of and a macrolide or the combi- Organisms Isolated From 28 Aspirates of Frontal Sinusitis* nation of a (eg, amoxicillin) and a ␤-lacta- mase inhibitor (eg, ), and the newer Acute Chronic quinolones (eg, trovafloxacin mesylate and moxifloxacin). Sinusitis Sinusitis Total Bacteria (n = 15) (n = 13) (n = 28) Other effective agents are available only in parenteral form (eg, sodium, cefotetan disodium, and cef- Aerobic bacteria metazole sodium). If gram-negative organisms, such as ␣ Hemolytic streptococcus . . . 2 2 Microaerophilic streptococci 1 3 4 P aeruginosa, are involved, parenteral therapy with an Streptococcus pneumoniae 5 ... 5 aminoglycoside, a fourth-generation cephalosporin Streptococcus pyogenes 1 ... 1 (cefepime hydrochloride or ceftazidime sodium), or oral Staphylococcus aureus ... 2(2) 2 (2) or parenteral treatment with a fluoroquinolone (only in Staphylococcus epidermidis ... 1 1 postpubertal patients) is also used. Parenteral therapy with Klebsiella pneumoniae ... 1(1) 1 (1) a carbapenem (eg, ) is more expensive, but pro- Pseudomonas aeruginosa ... 1(1) 1 (1) Moraxella catarrhalis 3(3) ... 3 (3) vides coverage for most potential pathogens, both an- Haemophilus influenzae 6 (3) 2 (1) 8 (4) aerobes and aerobes. Subtotal 16 (6) 12 (5) 28 (11) Prospective studies are warranted to elucidate the role Anaerobic bacteria of anaerobic bacteria in acute and chronic sphenoid sinus- Peptostreptococcus species 1 5 6 itis. It is, however, recommended that specimens are ob- Propionibacterium acnes 112 tained for culture from infected sphenoid sinuses for both Fusobacterium species 1 2 (1) 3 (1) Fusobacterium nucleatum ... 2(2) 2 (2) aerobic and anaerobic bacteria and fungi so that appropri- species . . . 2 (1) 2 (1) ate antimicrobial therapy can be determined. Prevotella melaninogenica ... 3(1) 3 (1) Prevotella oralis ... 1(1) 1 (1) Accepted for publication October 26, 2001. Prevotella intermedia 1 (1) 3 (1) 4 (2) Corresponding author and reprints: Itzhak Brook, Porphyromonas asaccharolytica ... 1 1 MD, MSc, PO Box 70412, Chevy Chase, MD 20812-0412 Subtotal 4 (1) 20 (7) 24 (8) Total 20 (7) 32 (12) 52 (19) (e-mail: [email protected]).

*Numbers within parentheses indicate ␤-lactamase producers; ellipses REFERENCES indicate not applicable. vulanic acid), cephalosporins, and macrolides. Amoxi- 1. Antila J, Suonpaa J, Lehtonen OP. Bacteriological evaluation of 194 adult pa- tients with acute frontal sinusitis and findings of simultaneous maxillary sinus- cillin is often used for sinusitis therapy, is safe and in- itis. Acta Otolaryngol Suppl. 1997;529:162-164. expensive, and is still, when given in a high dose, the drug 2. Ruoppi P, Seppa J, Nuutinen J. Acute frontal sinusitis: etiological factors and of choice for intermediately penicillin-susceptible S pneu- treatment outcome. Acta Otolaryngol. 1993;113:201-205. moniae. However, the growing resistance of H influen- 3. Suonpaa J, Antila J. Increase of acute frontal sinusitis in southwestern Finland. zae and M catarrhalis to amoxicillin increases the risk that Scand J Infect Dis. 1990;22:563-568. 4. Moon T, Lin RY, Jahn AF. Fatal frontal sinusitis due to Neisseria sicca and Eu- it will fail to clear the infection. The addition of clavu- bacterium lentum. J Otolaryngol. 1986;15:193-195. lanic acid (a ␤-lactamase inhibitor) to amoxicillin or 5. Brook I, Friedman EM, Rodriguez WJ, Controni G. Complications of sinusitis in the use of antimicrobial agents resistant to ␤-lactamase children. Pediatrics. 1980;66:568-572. activity is effective against resistant organisms. The in- 6. Brook I. Bacteriologic features of chronic sinusitis in children. JAMA. 1981;246: crease in resistance of S pneumoniae to penicillin re- 967-969. 7. Summanen P, Barron EJ, Citron DM, Strong C, Wexler HM, Finegold SM. Wads- quires an increase in the amount of amoxicillin admin- worth Anaerobic Bacteriology Manual. 5th ed. Belmont, Calif: Star Publishing; 1993. istered to patients (up to 90 mg/kg daily in children 8. Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH. Manual of Clinical Mi- and 3.0 g/d in adults). This requires the addition crobiology. 6th ed. Washington, DC: American Society for Microbiology; 1995. of an equal amount of amoxicillin to amoxicillin– 9. O’Callaghan DH, Morris A, Kirby SM, Shingler AH. Novel method for detection of beta-lactamase by using a chromogenic cephalosporin substrate. Antimicrob clavulanic acid. Agents Chemother. 1972;1:283-288. The second-generation cephalosporins (cefurox- 10. Frederick J, Braude AI. Anaerobic infection of the paranasal sinuses. N Engl J ime axetil, cefprozil, and cefpodoxime proxetil) are ac- Med. 1974;290:135-137. tive against penicillin-resistant Haemophilus and Mo- 11. Brook I. Bacteriology of chronic maxillary sinusitis in adults. Ann Otol Rhinol Laryn- raxella species and intermediately penicillin-resistant gol. 1989;98:426-428. 12. Nord CE. The role of anaerobic bacteria in recurrent episodes of sinusitis and S pneumoniae. The newer quinolones (eg, levofloxacin, tonsillitis. Clin Infect Dis. 1995;20:1512-1524. gatifloxacin, and moxifloxacin hydrochloride) are effec- 13. Drettner B, Lindholm CE. The borderline between acute rhinitis and sinusitis. Acta tive against penicillin-sensitive and penicillin-resistant Otolaryngol (Stockh). 1967;64:508-513. S pneumoniae and are also active against Haemophilus and 14. Carenfelt C, Lundberg C. Purulent and non-purulent maxillary sinus secretions Moraxella species. with respect to pO2, pCO2 and pH. Acta Otolaryngol (Stockh). 1977;84:138-144. 15. Aust R, Drettner B. Oxygen tension in the human maxillary sinus under normal Antimicrobial agents used for chronic sinusitis therapy and pathological conditions. Acta Otolaryngol (Stockh). 1974;78:264-269. should be effective against aerobic and anaerobic BLPB. 16. Carenfelt C. Pathogenesis of sinus empyema. Ann Otol Rhinol Laryngol. 1979; These include , , the com- 88:16-20.

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 128, MAY 2002 WWW.ARCHOTO.COM 585

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021