Bacteriologic findings in patients with chronic sinusitis.

Abstract

We studied the bacteriology of maxillary sinus aspirates obtained from patients diagnosed with chronic sinusitis. We recovered 659 strains from 510 aspirates; of these, 572 (86.8%) were aerobes and 87 (13.2%) were anaerobes. Aerobes only were recovered from 310 of the 510 specimens (60.8%) and anaerobes only from 31 (6.1%). Among the 572 aerobic bacteria, the most prevalent organisms were Streptococcus viridans (158 strains [27.6%]), Streptococcus pneumoniae (67 [11.7%1), Corynebacterium species (66 [11.5%]), Staphylococcus aureus (54 [9.4%]), Moraxella catarrhalis (38 [6.6%]), Hemophilus parainfluenzae (33 [5.8%]), and group C beta-hemolytic streptococci (26 [4.5%]). Among the 87 recovered anaerobes were species of Peptostreptococcus (32 strains [36.8%]), Prevotella (22 [25.3%]), Actinomyces (13 [14.9%]), Propionibacterium (11 [12.6%]), Fusobacterium (8 [9.2%]), and Veillonella (1 [1.1%1). Beta-lactamase production was detected in 115 of the 572 aerobic strains (20.1%) and in 10 of the 87 anaerobic strains (11.5%). We found that the prevalence and type of organisms we identified in chronic sinusitis did not differ substantially from those reported in previous studies. Our study is one of the more extensive reports on the type and prevalence of pathogens in chronic sinusitis that has been published to date.

Introduction

Sinusitis has a substantial impact on the health of children and adults. (1) It is classified as either acute or chronic, primarily on the basis of the duration of infection and pathologic findings. (2) Treatment with antibiotics is primarily empiric and based on pathogen prevalence. There has been a great deal of variation in the reporting of the bacteriology of chronic sinusitis; differences can be attributed to the different countries in which studies were performed, characteristics of study populations, and differences in sampling methods and laboratories. (3-5) The goal of our study was to determine the bacteriology of maxillary sinus aspirates obtained from patients diagnosed with chronic sinusitis.

Patients and methods

Between April 1, 1997, and Dec. 31, 2000, we conducted a prospective study of 510 outpatients--268 men and 242 women, aged 17 to 68 years (mean: 42)--who had been diagnosed with chronic sinusitis on the basis of clinical findings, imaging by x-ray and computed tomography (CT), and endoscopic exploration with both rigid and flexible instruments. Our study patients shared many clinical and imaging characteristics: (1) recurrent infections of the upper respiratory tract with persistent purulent nasal or postnasal discharge, nasal obstruction, cough, headache, pharyngitis, and/or concomitant otitis media, (2) smell, taste, and/or sleep disorders, (3) persistent clinical symptoms for at least 1 year, (4) initial response to medical treatment followed by a rapid recurrence, (5) mucosal thickening, purulent discharge, and ostiomeatalcomplex obstruction on endoscopic examination, and (6) opacity, osteitis, presence of an air-fluid level, and mucosal thickening on x-ray and CT. None of the patients had taken local or systemic antimicrobial therapy for at least 1 week prior to aspirate sampling. Patients with cystic fibrosis were excluded from the study. No patient had previously undergone sinus surgery.

Prior to puncture, the nares were disinfected with a cotton swab soaked in povidone-iodine solution for a minimum of 5 minutes. Material from the maxillary sinuses was aspirated into a syringe through an 18-gauge needle, which was used to puncture the inferior meatus. In cases when no material was obtained initially, 2 to 5 ml of sterile diluent was injected into the sinus and aspirated. In all cases, the contents of the aspirated material were injected into an anaerobic transport vial (TAB; Britania Laboratories; Buenos Aires), transported immediately to the laboratory, and held at room temperature until processing. At the laboratory, microscopic examination with Gram's staining was performed, and the density of each organism was categorized as either high, medium, or low.

Specimens were taken from the transport vial with a 0.01-ml calibrated loop and inoculated into a Columbia agar base supplemented with 5% sheep's blood, chocolate agar, eosin-methylene-blue agar, and Chapman agar (for aerobic and facultative bacteria) and onto a Brucella agar base supplemented with 5% sheep's blood, vitamin [K.sub.1], hemin with and without kanamycin, and thioglycolate medium without an indicator and supplemented with vitamin [K.sub.1] and heroin (for anaerobic bacteria). Plates were incubated at 35[degrees]C aerobically (eosin-methylene-blue agar and Chapman agar) and under 5% carbon dioxide (Columbia agar base and chocolate agar) and examined at 24 and 48 hours.

For anaerobic bacteria, plates were incubated in either an anaerobic bag or a GasPak jar, depending on the number of plates, for a minimum of 48 hours and a maximum of 7 days. Colonies in both aerobic and anaerobic plates were semiquantitated by the number of quadrants in which growth occurred (1 +, 2+, 3+, or 4+). Those organisms that appeared in low density in the Gram's stain and grew only at 1+ were considered to be contaminant flora. Using standard methods, most aerobic bacteria were identified to genus and species level and anaerobic bacteria were identified to genus level only. (6) Finally, antimicrobial susceptibility testing was carried out for aerobic bacteria according to the protocols established by the National Committee for Clinical Laboratory Standards. (7)

Results

Of the 510 aspirates studied, 10 (2.0%) were negative (table 1). In the remaining samples, 659 strains were recovered--572 (86.8%) were aerobes and 87 (13.2%) were anaerobes (table 2). The antimicrobial susceptibility of the more common gram-positive organisms was also determined (table 3). Beta-lactamase production was detected in 115 of 572 (20.1%) aerobic strains (enterobacteria, nonfermentative gram-negative rods, and one strain of Hemophilus influenzae) and in 10 of 87 (11.5%) anaerobic strains (nine strains of Prevotella species and one strain of Fusobacterium species).

Discussion

Many assessments of the bacteriology of chronic sinusitis have been made in several countries and by different sampling methods. Consequently, the results of these studies have been quite variable. The gold standard for diagnosing bacterial sinusitis is culture of infected secretions obtained by direct sinus puncture. Nevertheless, several other techniques are used to obtain sinusal specimens. (8) The primary reason for the discrepancies in results is contamination of samples with resident flora in nasal mucosa. Other important factors are differences in the methods of transportation and isolation of anaerobes.

Jiang et al obtained samples by inserting a cotton-tipped stick through a cannula that had been placed into the ethmoid cavity following removal of the ethmoid bulla. (9) They concluded that this method could decrease the chance of nasal contamination and thus might make study of the bacteriology of the sinus more accurate. Nasal endoscopy is indicated for patients with rhinosinusitis who do not respond to therapy as expected and when anatomic factors preclude adequate examination by anterior rhinoscopy. (10) A protected microswab can be used to sample these patients. (10) Liu et al reported on the use of a novel culture device (Accu-CulShure; MLA Systems; Pleasantville, N.Y.) to sample and transport sinusal material that avoids contamination with resident nasal flora and preserves the quality of the material during transport. (11)

Bacteriologic cultures of chronic sinusitis aspirates vary greatly from those of acute sinusitis aspirates. Besides the "infernal trio" (Streptococcus pneumoniae, H influenzae, and Moraxella catarrhalis), chronic sinusitis aspirates contain other bacteria, including anaerobes and gram-negative bacteria. (12)

In our study, we recovered aerobes in high numbers (86.8%), and the types of species were consistent with those reported by others. Doyle and Woodham evaluated the microbiology of chronic ethmoid sinusitis by obtaining biopsy specimens from the ethmoid sinus mucosa.t3 In their study, Staphylococcus aureus and members of the family Enterobacteriaceae were the most frequently isolated classic pathogens; no anaerobes were identified. In a study of 211 patients with chronic sinusitis, Busaba et al detected aerobic bacteria in 76% and an aerobic bacteria in 6.2%; the most common isolate was S aureus. (14) Biel et al obtained sinus samples endoscopically during surgery and found that the most common bacteria were coagulase-negative staphylococci (36%), followed by S aureus, Streptococcus viridans, Corynebacterium species, and anaerobes. (15) They speculated that coagulase-negative staphylococci might have been the causative pathogen and recommended that antimicrobial sensitivities be obtained for evaluation and possible treatment.

Bacteria isolated in chronic sinusitis aspirates usually do not have a clear pathogenic power--that is, their ability to cause infection is unknown. Therefore, several authors have studied the microbiology of the middle meatus and maxillary sinus in healthy adults to evaluate bacteria's role in sinusal disease. Gordts et al recovered organisms in 75% of the their patients, but growth was often poor and the predominant species suggested a commensal flora. (16) Nadel et al reviewed 507 endoscopically guided cultures obtained from 265 patients. (17) They found that gram-negative rods were present in 27% of cultures and were more common in patients who had previously undergone sinus surgery or who were using irrigations. Pseudomonas aeruginosa strains were more common in patients who were taking systemic steroids. Coagulase-negative staphylococci were present in the same proportions of patients and controls, but they were more prevalent in cultures obtained intraoperatively and in patients who were taking systemic steroids. (17)

In our study, S viridans and Corynebacterium species were two of the three most common isolates. However, these organisms cannot be considered pathogenic per se because they are also isolated from sinus mucosa in healthy patients. They could play a role in chronic sinusitis, but only in association with the presence of other bacteria, particularly anaerobes. Jiang et al also found that S viridans was the most common isolated bacteria, followed by Klebsiella pneumoniae, Proteus mirabilis, and Hemophilus parainfluenzae. (9) In aspirates obtained from 108 chronically inflamed maxillary sinuses, Brook and Frazier found that the predominant aerobic isolates were S aureus (12 isolates), alpha-hemolytic streptococci (10), M catarrhalis (8), Hemophilus species (6), and P aeruginosa (3). (18)

The prevalence of organisms recovered in our study is consistent with that reported by others, although there were substantial differences in the size of the study populations. Our study is one of the more extensive reports on pathogens in chronic sinusitis that has been published to date.

Based on the literature, it is possible to affirm that the isolation of anaerobes is the most variable finding among different studies, probably because of the diversity of sampling techniques and transportation methods. Despite different rates of isolation in previous studies, the particular species of pathogens were not substantially different. Although our rate of isolation of anaerobes was not high (13.2%), some authors did not recover any anaerobes, even though they used proper techniques). (13,19)

We observed, as did Mitchelmore and Tabaqchali, (20) that the main anaerobic genera involved in chronic sinusitis infections were Peptostreptococcus, Fusobacterium, and black-pigmented gram-negative rods, frequently in association with Streptococcus milleri. Similarly, Brook and Frazier found that the most common anaerobic isolates among 186 anaerobic strains were Peptostreptococcus species (61 isolates [32.8%]), Prevotella species (45 [24.2%]), Fusobacterium species (15 [8.1%]), and Propionibacterium acnes (14 [7.5%]). (18) They affirmed that the prevalence of anaerobes decreases following sinus surgery because the maxillary sinuses are then ventilated.

The cornerstones of treatment for chronic sinusitis are antibiotics, decongestants, and in some cases nasal steroids. Antimicrobial treatment often is prescribed empirically on the basis of the likely infecting pathogens, local bacterial antibiotic resistance rates, and antibiotics' pharmacologic profiles. (21) Injudicious use and indiscriminate overprescription of antibiotics has fostered the rapid development of penicillin-resistant organisms over the past 20 years, forcing physicians to consider alternatives in antibacterial management. (4) Because of the great variation in susceptibility patterns among organisms isolated from patients with chronic sinusitis, local surveillance of the susceptibility of individual isolates is desirable to enhance the benefit of empiric treatment and to influence prescribing habits so that physicians can prevent expansion of resistance. (22) Beta lactams, macrolides, and quinolones are the most frequently recommended drugs for treating both acute and chronic sinusitis because they are associated with a low rate of resistance among respiratory pathogens (23,24) In our study, the rates of resistance against these drugs was not significant.

The bacteria that predominate in sinusitis can resist penicillin by producing beta lactamase or by altering sites of penicillin binding. Beta-lactamase-producing bacteria can express their pathogenicity directly as a result of their ability to cause infections and indirectly by producing beta lactamase, thus protecting penicillin-susceptible pathogens from penicillin. (25) Brook et al found that aerobic and anaerobic beta-lactamase-producing bacteria play an important role in the polymicrobial etiology of chronic maxillary sinusitis. (3)

The three most common gram-positive bacteria isolated in our study--Sviridans,

S pneumoniae, and Corynebacterium species--were all completely susceptible to penicillin, although S aureus and coagulase-negative staphylococci were not. Individually, those organisms that have a natural resistance to certain beta lactams and beta-lactamase inhibitors--for example, some Enterobacteriaceae, Pseudomonas species, and Acinetobacter species--were not common in our study. However, together they accounted for almost 15% of the total number of aerobic isolates, and therefore they should be taken in account when prescribing empiric therapy in order to avoid therapeutic failures. Even though no S pneumoniae strain was resistant to penicillin in our study, this organism is associated with high rates of penicillin resistance in other countries. (26,27)

Table 1. Distribution of isolates from the 510
aspirates

Flora                                        Samples, n (%)

Pure aerobes                                   310 (60.8)
Mixed aerobes                                   115 (2.5)
Aerobes and anaerobes                           32 (6.3)
Pure anaerobes                                  31 (6.1)
Mixed anaerobes                                 12 (2.4)
None                                            10 (2.0)

Table 2. Aerobic and anaerobic isolates from 659 strains

Organism                                      No. isolates, n (%)

Aerobes                                           572 (86.8)

Anaerobes                                          87 (13.2)

Total                                              659 (100)

Aerobes (n = 572)

  Streptococcus viridans                          158 (27.6)
  Streptococcus pneumoniae                         67 (11.7)
  Corynebacterium species                          66 (11.5)
  Staphylococcus aureus                            54 (9.4)
  Moraxella catarrhalis                            38 (6.6)
  Hemophilus parainfluenzae                        33 (5.8)
  Group C beta-hemolytic streptococci              26 (4.5)
  Coagulase-negative staphylococci                 16 (2.8)
  Enterobacter aerogenes                           15 (2.6)
  Pseudomonas aeruginosa                           15 (2.6)
  Other *                                          84 (14.7)

Anaerobes (n = 87)

  Peptostreptococcus species                       32 (36.8)
  Prevotella species                               22 (25.3)
  Actinomyces species                              13 (14.9)
  Propionibacterium species                        11 (12.6)
  Fusobacterium species                             8 (9.2)
  Veillonella species                               1 (1.1)

* Includes: Klebsiella oxytoca (10 isolates), Klebsiella pneumoniae
(10), group A beta-hemolytic streptococci (9), Hemophilus influenzae
(9), Proteus mirabilis (8), Citrobacter kooseri (6), Neisseria species
(5), Candida albicans (4), Serratia species (4), group G beta-hemolytic
streptococci (3), Burkholderia gladioli (2), Enterobacter gergoviae
(2), Escherichia colt (2), Proteus vulgaris (2), Acinetobacter
baumannii (1), Arcanobacterium hemolyticum (1), Citrobacter freundii
(1), Enterococcus faecalis (1), Moraxella species (1), Morganella
morganii (1), Pseudomonas putida (1), and Stenotrophomonas maltophilia
(1).

Table 3. Antimicrobial susceptibility of the more common gram-positive
isolates

                                     Susceptible isolates, n (%)

Organism                                PEN *           ERY

Streptococcus viridans (n = 158)      158 (100)      147 (93.0)
Streptococcus pneumoniae (n = 67)     67 (100)        47 (70.1)
Corynebacterium species (n = 66)      66 (100)        66 (100)
Staphylococcus aureus (n = 54)            0           42 (77.8)
Coagulase-negative                        0           10 (62.5)
  staphylococci (n = 16)

Total (n =361)                       291 (80.6)      312 (86.4)

Organism                                 CLI            CIP

Streptococcus viridans (n = 158)      158 (100)     NT ([dagger])
Streptococcus pneumoniae (n = 67)     67 (100)        67 (100)
Corynebacterium species (n = 66)      55 (83.3)       66 (100)
Staphylococcus aureus (n = 54)        42 (77.8)       54 (100)
Coagulase-negative                    14 (87.5)       16 (100)
  staphylococci (n = 16)

Total (n =361)                       336 (93.1)      203 (100)
                                                     [section]
Organism                                 OXA

Streptococcus viridans (n = 158)         NT
Streptococcus pneumoniae (n = 67)     ([double
                                      dagger])
Corynebacterium species (n = 66)         NT
Staphylococcus aureus (n = 54)         50 (93)
Coagulase-negative                     16 (100)
  staphylococci (n = 16)

Total (n =361)                         66 (94.3)
                                      [section]

* PEN = penicillin; ERY = erythromycin; CLI = clindamycin; CIP =
ciprofloxacin; OXA = oxacillin.
([dagger]) NT = not tested.
([double dagger]) Used to determine susceptibility to penicillin.
[section] Percentages reflect substraction of NT values from the
overall total of 361 isolates.

References

(1.) Hamilos DL. Chronic sinusitis. J Allergy Clin Immunol 2000; 106: 213-27.

(2.) Infectious rhinosinusitis in adults: Classification, etiology and management. International Rhinosinusitis Advisory Board. Ear Nose Throat J 1997;76(Suppl):5-22.

(3.) Brook I, Thompson DH, Frazier EH. Microbiology and management of chronic maxillary sinusitis. Arch Otolaryngol Head Neck Surg 1994;120:1317-20.

(4.) Hadley JA. The microbiology and management of acute and chronic rhinosinusitis. Curr Infect Dis Rep 2001;3:209-16.

(5.) Jiang RS, Liang KL, Jang JW, Hsu CY. Bacteriology of endoscopically normal maxillary sinuses. J Laryngol Otol 1999;113: 825-8.

(6.) Summanem P, Baron EJ, Citron DM, et al. Wadsworth Anaerobic Bacteriology Manual. 5th ed. Belmont, Calif.: Star Publishing, 1993.

(7.) Performance Standards for Antimicrobial Disk Susceptibility Test. 7th ed. Document M2-A6. Wayne, Pa.: National Committee for Clinical Laboratory Standards, 1999.

(8.) Axelsson A, Brorson JE. The correlation between bacteriological findings in the nose and maxillary sinus in acute maxillary sinusitis. Laryngoscope 1973;83:2003-11.

(9.) Jiang RS, Hsu CY, Leu JF. Bacteriology of ethmoid sinus in chronic sinusitis. Am J Rhinol 1997:11:133-7.

(10.) Osguthorpe JD. Adult rhinosinusitis: Diagnosis and management. Am Fam Physician 2001;63:69-76.

(11.) Liu ES, Lebowitz RA, Jacobs JB, Tierno PM. The bacteriology of chronic rhinosinusitis: Results using a novel culture device. Am J Rhinol 2000;14:101-5.

(12.) van Cauwenberge PB, Vander Mijnsbrugge AM, Ingels KJ. The microbiology of acute and chronic sinusitis and otitis media: A review. Ear Arch Otorhinolaryngol 1993;250(Suppl 1):S3-6.

(13.) Doyle PW, Woodham JD. Evaluation of the microbiology of chronic ethmoid sinusitis. J Clin Microbiol 1991;29:2396-2400.

(14.) Busaba NY, Siegel N, Salman SD. Bacteriology of nontraumatic maxillary sinus mucoceles versus chronic sinusitis. Laryngoscope 2000;110:969-71.

(15.) Biel MA, Brown CA, Levinson RM, et al. Evaluation of the microbiology of chronic maxillary sinusitis. Ann Otol Rhinol Laryngol 1998;107(Pt 1):942-5.

(16.) Gordts F, Halewyck S, Pierard D, et al. Microbiology of the middle meatus: A comparison between normal adults and children. J Laryngol Otol 2000;114:184-8.

(17.) Nadel DM, Lanza DC, Kennedy DW. Endoscopically guided cultures in chronic sinusitis. Am J Rhinol 1998;12:233-41.

(18.) Brook I, Frazier EH. Correlation between microbiology and previous sinus surgery in patients with chronic maxillary sinusitis. Ann Otol Rhinol Laryngol 2001;110:148-51.

(19.) Chan J, Hadley J. The microbiology of chronic rhinosinusitis: Results of a community surveillance study. Ear Nose Throat J 2001;80: 143-5.

(20.) Mitchelmore IJ, Tabaqchali S. Anaerobes in sinusitis. Anaerobe 1999;5:355-7.

(21.) Brook I. Sinusitis--overcoming bacterial resistance. Int J Pediatr Otorhinolaryngol 2001;58:27-36.

(22.) Felmingham D. Antibiotic resistance. Do we need new therapeutic approaches? Chest 1995; 108(Suppl):70S-78S.

(23.) Goldstein EJ, Citron DM, Merriam CV. Comparative in vitro activities of amoxicillin-clavulanate against aerobic and anaerobic bacteria isolated from antral puncture specimens from patients with sinusitis. Antimicrob Agents Chemother 1999;43: 705 7.

(24.) Chatzimanolis E, Marsan N, Lefatzis D, Pavlopoulos A. Comparison of roxithromycin with co-amoxiclav in patients with sinusitis. J Antimicrob Chemother 1998;41(Suppl B):81-4.

(25.) Brook I. Microbiology of common infections in the upper respiratory tract. Prim Care 1998;25:633-48.

(26.) Granizo JJ, Aguilar L, Casal J, et al. Streptococcus pneumoniae resistance to erythromycin and penicillin in relation to macrolide and beta-lactam consumption in Spain (1979-1997). J Antimicrob Chemother 2000;46:767-73.

(27.) Paterick BD, Pollack CV, Jr. Geographic variation in penicillin resistance in Streptococcus pneumoniae: Selected sites, United States, 1997. Ann Emerg Med 2000;35:506-8.

From Laboratory Dr, Botello, Corrientes, Argentina (Dr. Merino), the Department of Bacteriology, Instituto de Medicina Regional, Universidad Nacional del Nordeste, Resistencia, Argentina (Dr. Merino, Dr. Ronconi, and Dr. Hrenuk), and the Institute of Otorhinolaryngology, Corrientes (Dr. de Pepe).

Reprint requests: Luis Antonio Merino, MSc, Av. Maipu 1550 2do A, 3400 Corrientes, Argentina. Phone: [phone]; fax: [phone]; e-mail: [e-mail]

Originally presented in part at a poster session during the 101st general meeting of the American Society for Microbiology; May 20-24, 2001; Orlando, Fla.