Bhattacharyya N, Kepnes LJ. The Microbiology of Recurrent Rhinosinusitis After Endoscopic Sinus Surgery. Arch Otolaryngol Head Neck Surg. 1999;125(10):1117-1120. doi:10.1001/archotol.125.10.1117
To determine the microbiology of recurrent sinus infections occurring in patients after endoscopic sinus surgery (ESS).
Retrospective review of sinus cultures obtained over a 4-year period from a consecutive series of patients who underwent ESS.
An academic general otolaryngology practice.
A total of 290 cultures were performed in 125 patients after ESS. The female-male ratio of cultures was 2.5:1 with an average patient age of 47.3 years. This group of patients represents 14.5% of 860 patients who underwent ESS during the same period. A total of 65 patients had 1 culture performed, and 60 patients had multiple cultures. Of the 290 culture specimens, 87 (30.0%) demonstrated no growth. Gram-positive cocci predominated, accounting for 37.9% of culture results. Gram-negative rods constituted 14.8% of the isolates. Of the cultures yielding gram-negative rods, 90.7% occurred in patients who had multiple cultures (P=.03). Fungal forms were cultured in 1.7% of the specimens. None of the Streptococcus pneumoniae isolates demonstrated penicillin-based resistance. The percentages of β-lactamase–producing strains for Haemophilus influenzae and Branhamella (Moraxella) catarrhalis were 45.4% and 81.8%, respectively. Staphylococcal species also exhibited significant antibiotic resistance patterns, but no statistical association with multiple cultures was noted (P=.23).
A wide range of bacteria may be present in the infected post-ESS sinus cavity, with a considerable population of gram-negative organisms, including Pseudomonas species. β-Lactamase–producing organisms continue to be prevalent in postoperative sinus infections. Culture and sensitivity analyses of pathologic secretions may identify drug-resistant organisms or organisms related to difficult-to-treat infections in exacerbations of chronic rhinosinusitis in the postoperative setting.
THE PRIMARY objective of endoscopic sinus surgery (ESS) is the ventilation of obstructed sinuses. Despite the clinical success demonstrated with ESS, surgical therapy may fail in a significant portion of patients. Such failure can be manifested by persistent symptoms, recurrent infections, or the need for revision surgery. Depending on the study methods used and duration of follow-up, this failure rate ranges from 3% to 20%.1,2 In addition, a subgroup of the post-ESS patients continue to experience recurrent infectious exacerbations of their chronic sinus disease, often with purulent drainage despite patent sinus ostia. While much has been written on the microbiology of chronic rhinosinusitis in the presurgical setting, less is known about the bacteriology of chronic rhinosinusitis after ESS. This study was undertaken to characterize the microbiology of exacerbations of chronic rhinosinusitis in patients after ESS.
Patients with recurring bouts of rhinosinusitis after ESS often require multiple courses of antibiotics for extended durations. This, coupled with the fact that these patients have usually been treated with multiple different antibiotics prior to sinus surgery, makes these patients especially susceptible to antibiotic resistance. Antibiotic resistance is becoming an increasing problem in otolaryngology and many areas of medicine. This study also sought to determine the nature of antibiotic resistance in these postoperative patients.
The microbiology database of an academic tertiary care hospital was searched for all cases of sinus cultures submitted between January 1, 1994, and December 31, 1998. All types of cultures including aerobic, anaerobic, and fungal culture specimens were included. These were then cross-referenced with a procedural database consisting of a consecutive series of patients undergoing ESS during the same period. Sinus cultures associated with patients who had undergone ESS were identified and tabulated, and the records were reviewed. For each culture specimen, the corresponding demographic, microbiologic, and surgical data were extracted from the medical record for each patient. Indications for culture were the presence of mucopurulent sinonasal secretions in conjunction with at least 1 of the major or minor symptoms of chronic rhinosinusitis defined by the American Academy of Otolaryngology–Head and Neck Surgery Task Force on Rhinosinusitis.3 Cultures of mucopurulent secretions were obtained from the sinonasal cavity under endoscopic guidance using a swab (CultureSwabtransport system, DIFCO Laboratories, Detroit, Mich) technique, as described by Nadel et al.4 Care was taken to prevent contamination with the anterior nasal or nasopharyngeal secretions. The data were examined for the presence of bacteria on culture, the type(s) of bacteria isolated, and antibacterial sensitivity patterns for various antibiotics. Specifically, resistance patterns were examined for streptococcal, staphylococcal, and Pseudomonas species. For consistency in reporting, the distribution of various bacterial types is reported as a percentage of the total cultures submitted rather than as a percentage of positive isolates. Further statistical analysis with the Pearson χ2 method was performed to determine the association between cultures yielding gram-negative rods and the frequency of repeated cultures (SPSS Inc, Chicago, Ill). Similar analysis was conducted to determine if an association existed between repeated cultures and resistance for staphylococcal species.
A total of 372 sinus culture specimens were initially identified. When these were cross-referenced with the procedural database for ESS, 290 cultures were performed in patients after ESS. The female-to-male ratio of cultures was 2.5:1. These cultures were performed among a total of 125 patients with an average age of 47.3 years, representing 14.5% of a total 860 patients who underwent ESS during the same period. A total of 65 patients had 1 culture, and 60 patients (7.2% of the 860 ESS patients) had more than 1 culture performed during this period. The average number of cultures per patient was 2.3.
Of the 290 culture specimens, 87 (30.0%) demonstrated no growth. The distribution of bacteria among the remaining 203 specimens is presented in Table 1. Gram-positive cocci predominated, being recovered from 37.9% of the 290 cultures. Gram-negative rods, such as Enterobacter species and Escherichia coli, usually associated with infradiaphragmatic sources, comprised 14.8% of the isolates. Of the cultures yielding gram-negative rods, 90.7% occurred in patients who had multiple cultures and 9.3% occurred in patients who had only 1 culture (P=.03). Yeast or fungal forms were encountered in 1.7% of the specimens. No patient with cultures yielding Pseudomonas species had a concurrent diagnosis of cystic fibrosis. None of the Streptococcus pneumoniae isolates demonstrated penicillin-based resistance. The percentages of β-lactamase–producing strains for Haemophilus influenzae and Branhamella (Moraxella) catarrhalis were 45.4% and 81.8%, respectively. The remainder of results for the sensitivity analysis is displayed in Table 2. For staphylococcal species, no association was found between multiple cultures and increased antibiotic resistance (P=.23).
Antimicrobial therapy is a key element in the management of both acute and chronic rhinosinusitis.5 The microbiology of acute rhinosinusitis has been extensively studied. Several reports have confirmed the involvement of typical upper respiratory tract organisms in this disease process, such as S pneumoniae, H influenzae, or M catarrhalis. Several other studies have delineated the common organisms involved in chronic rhinosinusitis.6- 8 These include staphylococcal and streptococcal species, as well as gram-negative organisms. Studies by Brook et al9 and Erkan et al10 have also propounded the potential importance of anaerobic species in chronic rhinosinusitis. Thus, chronic rhinosinusitis appears to be a polymicrobial infection with the potential for involvement with a diverse range of bacteria. The antimicrobial therapy for chronic rhinosinusitis is based in large part on these microbiologic studies. The American Academy of Otolaryngology–Head and Neck Surgery Task Force on Rhinosinusitis has recommended broad-spectrum antibiotic coverage for an extended period of time as an important component of the medical management of chronic rhinosinusitis.11 Despite these data and recommendations, limited information is available regarding the microbiology of chronic rhinosinusitis in the postoperative setting.
Our data illuminate several interesting findings. First, despite the findings of purulence on endoscopic examination, a considerable portion of patients with seemingly active infections after ESS may have no growth from culture specimens. Several possible explanations for this exist, including the absence of bacteria, therapy with oral antimicrobial drugs at the time of culture, infection with atypical or difficult-to-culture bacteria, or the presence of anaerobes. As these cultures were not routinely submitted for anaerobic culture, this remains an area that requires further investigation. It is common for patients with recurrent postoperative infections to have been treated with multiple courses of oral antibiotics. This reinforces the need to withdraw oral antimicrobial therapy prior to obtaining the sinus culture. Patients with "no growth" from endoscopic culture despite the findings of purulence can be especially difficult to manage. Further study of these patients with cytological evaluation for eosinophilia or specialized cultures for anaerobes may be beneficial, although the yield for anaerobes may be as low as 10%.4
Not unexpectedly, these culture data differ from the organisms typically encountered in acute rhinosinusitis. Together, S pneumoniae, H influenzae, and M catarrhalis accounted for only 10.8% of the culture results. This study clearly demonstrates the continued prevalence of staphylococcal species in infections occurring in patients after sinus surgery. Staphylococcus aureus has been implicated as a significant pathogen in (preoperative) chronic rhinosinusitis, and it remains a significant pathogen when recurrent postoperative infections occur.6 Therefore, empiric antimicrobial therapy should possess activity against this organism, which is usually β-lactamase producing and prone to antibiotic resistance. Culture data with sensitivity analysis may be required to properly treat infections related to this organism. Involvement with coagulase-negative staphylococcal species (Staphylococcus epidermidis) is more difficult to interpret. Some controversy exists as to the pathogenicity of this organism, as some believe it is more likely a contaminant and not a significant pathogen in chronic rhinosinusitis.8 However, according to our sensitivity analysis, this organism may manifest multiple drug resistances, and if it is indeed a pathogen, the related infection may be difficult to treat.
Another striking finding is the prevalence of Pseudomonas species. As this organism has a strong propensity for multidrug resistance, these findings underscore the potential value of targeting antimicrobial therapy.12 The resistance patterns for Pseudomonas seen here may explain the difficulties encountered in its eradication in chronic rhinosinusitis. Other less commonly encountered bacteria such as Serratia and Enterobacter species may also surface as pathogens in the post-ESS setting, but fortunately not in large numbers. These organisms often carry resistance to multiple antibiotics, and sensitivity data should be obtained.12 Overall, these data implicate the same organisms commonly encountered in preoperative chronic rhinosinusitis with the exception of an increased prevalence of Pseudomonas and diphtheroids. Our data parallel those of Bolger,13 who found a 34.1% incidence (reported as a percentage of total isolates) of gram-negative bacteria in sinus culture isolates. In Bolger's study, specimens were obtained from patients prior to ESS and from other patients after ESS. Similarly, Nadel et al4 found a prevalence of 27.0% and 16.0% (reported as a percentage of total cultures) for gram-negative rods and Pseudomonas species, respectively, in their study of patients with chronic sinusitis and pathologic secretions. In addition, a significantly higher percentage of gram-negative isolates (30.0%) was found in patients with prior sinus surgery compared with those without prior surgery (9.5%). In our study, the significant statistical association between gram-negative rod isolates and repeated cultures suggests that these species, often with their multiple drug resistances, may be difficult to eradicate in the postoperative setting. These studies as well as ours further emphasize the emerging role of gram-negative pathogens in chronic sinusitis.
The fact that 14.5% of our ESS population received postoperative cultures highlights another potential indicator of clinical success or failure after ESS. Recently, several studies have documented long-term clinical outcomes for ESS by focusing on patient satisfaction and revision rates. These studies have demonstrated failure rates ranging from 3% to 20%, depending on the follow-up period and methods used.1,2 Interestingly, our culture rate of 14.5% falls within this range. Patients who have recurrent infections and who have been further evaluated with sinonasal cultures constitute a portion of the ESS population for whom, to some degree, surgical therapy has failed. This is especially true for patients who received multiple cultures (7.2% of those who underwent ESS), generally indicating recurrent postoperative infections and therefore recurrent symptomatic sinus disease. While examining culture rates will not capture all patients whose symptoms fail to respond to ESS, it does provide insight into those patients who will continue to have recurrent purulent sinus infections despite ESS. Paradoxically, one advantage of ESS is that it provides access to the ethmoid labyrinth and the maxillary antrum for obtaining a representative culture specimen.
Antibiotic resistance is an emerging problem in all fields of medicine. Our data highlight the potential problems of antibiotic resistance in patients with chronic rhinosinusitis and recurrent infections. The frequent use of extended periods of antibiotics in this group of patients is probably responsible for some of the resistance encountered. This study confirms that β-lactamase production remains a significant problem in the postoperative sinus infection. In addition, many patients will manifest staphylococcal infections resistant to standard penicillin- or cephalosporin-based therapy. Others have shown the increasing prevalence of β-lactamase–producing organisms in chronic rhinosinusitis.9,12 With the emergence of penicillin-resistant streptococci, and the increased use of the fluoroquinolone family of antibiotics, the risk for further antibiotic resistance is likely to increase. The prevalence of a 12.0% resistance rate (and a 20.0% rate of intermediate resistance) of Pseudomonas species to ciprofloxacin is especially distressing in this patient population. As the fluoroquinolones are one of the few oral antibiotic agents with activity against Pseudomonas, creation of resistance should be considered in selecting antibiotic therapy. Using culture-directed therapy may be one way to help prevent the emergence of resistant strains, and promote longer-lasting efficacy of the broad-spectrum antibiotics. Furthermore, nonantibiotic therapy such as saline irrigations should be considered as adjunctive elements in the management of these patients.14
In summary, our study demonstrates the following conclusions: (1) A wide range of bacteria may be present in the infected post-ESS sinus cavity. This includes a significant population of gram-negative organisms including Pseudomonas species. (2) β-Lactamase–producing organisms continue to be prevalent in postoperative sinus infections. (3) Given the prevalence of antibiotic resistance in this group of patients, further study of the bacteriology and antibiotic utilization in postoperative exacerbations of chronic rhinosinusitis is warranted.
Accepted for publication June 18, 1999.
Reprints: Neil Bhattacharyya, MD, Division of Otolaryngology, 333 Longwood Ave, Boston, MA 02115.