Research Article, J Otol Rhinol Vol: 2 Issue: 3
Quantitative Evidence Against Bacterial Infection Alone as a Cause of Chronic Rhinosinusitis
|1Department of Otology & Laryngology, Harvard Medical School, UK|
|2Division of Otolaryngology, Brigham and Women’s Hospital, Boston, USA|
|Corresponding author : Neil Bhattacharyya
Division of Otolaryngology, 45 Francis St. Boston, MA 02115, USA
Tel: 617-525-6540; Fax: 617-525-6544
E-mail: [email protected]
|Received: March 15, 2013 Accepted: May 10, 2013 Published: May 18, 2013|
|Citation: Bhattacharyya N (2013) Quantitative Evidence Against Bacterial Infection Alone as a Cause of Chronic Rhinosinusitis. J Otol Rhinol 2:2. doi:10.4172/2324-8785.1000120|
Quantitative Evidence Against Bacterial Infection Alone as a Cause of Chronic Rhinosinusitis
Objective: Determine if the differential presence of bacteria within individual patients with CRS supports a bacterial role in the pathogenesis of CRS.
Methods: A prospective series of adult patients with unilateral sinus disease on computed tomography undergoing primary endoscopic sinus surgery was studied. From the diseased sinus on one side and the contralateral analogous non-diseased sinus, bacterial cultures were obtained and analyzed. Quantitative assessment of bacterial burden was obtained and compared between the paired non-diseased (control) and the diseased sinuses. Results: 82 patients completed analysis with microbiological data for paired sinuses. When considering positive cultures for any bacteria excluding oral flora, 71/82 (86.6%) of the radiographically diseased sinuses were positive on culture versus 71/82 (86.6%) of the paired equivalent control sinuses on the contralateral side (p=0.999). Quantitatively similar bacterial recovery was obtained between control and diseased sinuses (1.1[1.1] versus 1.3[1.2], (serial dilution [standard deviation]) respectively, p=0.183). These results held true when coagulase negative Staphylococcus species were excluded from the analysis (p=0.333). Degree of bacterial resistance was also not different between control and diseased sinuses (p=0.851). Conclusion: Based on this prospective study with an intra?patient control, very similar rates of quantitative bacterial recovery were found between radiographically normal (control) and diseased paranasal sinuses within patients with unilateral CRS. These data argue against a significant role for persistent bacterial infection alone in the pathogenesis of unilateral CRS. These data may also have implications for bilateral CRS.
Materials/Methods: We surveyed both NES and English-speakers (ES) before and two months after receiving voice therapy at our institution for functional dysphonia. NES received therapy via translation. The survey instrument was the 30-item Voice Handicap Index (VHI). NES subjects were provided with a professionally translated version of the VHI in their native language.
Results: Change scores were calculated for the 20 subjects (12 NES and 7 ES) who completed both pre- and post-treatment surveys, and were normalized to a maximal score of 120 to account for incomplete questionnaires. ES improved from a baseline mean of 49.1 to 30.5 (p=0.015), and NES from 56.9 to 47.3 (p=0.065). The mean for the entire group improved from 54.0 to 41.1 (p=0.002). No significant between-group difference was found.
Conclusion: Overall, this mixed population showed improvement in the VHI with voice therapy. Analyzed individually, English-speakers had significant improvement, while non-English speakers did not. Further studies of voice therapy via translation are warranted.
Keywords: Chronic rhinosinusitis; Pathogenesis; Nondiseased sinuses
|Chronic rhinosinusitis; Pathogenesis; Nondiseased sinuses|
|Based on recent figures, chronic rhinosinusitis (CRS) affects approximately 6% of the adult United States population . Unfortunately, the etiology and pathogenesis of CRS remains elusive. Traditionally, CRS has been thought of as a chronic infectious process, perhaps as a temporal extension of acute to subacute rhinosinusitis. More recently, however, over the past decade the pendulum has shifted toward an impression of CRS as a chronic inflammatory rather than a chronic infectious disease process [2,3]. Again, the drivers of that putative inflammation within the paranasal sinuses have yet to be convincingly determined.|
|With respect to infection, bacteria may yet still be one of the primary drivers of the inflammation CRS. Alternatively, they may in fact be an epiphenomenon, colonizing and/or infecting the paranasal sinuses when the host defense mechanisms become deranged during the chronic inflammation that characterizes CRS . Proving or disproving a partial bacterial etiology in the pathogenesis of CRS is difficult, in part because of the chronic nature of the disease, the frequency of positive microbiological cultures in CRS and the multifactorial etiology of the disease itself.|
|Previously, we reported on a series of adult patients manifesting unilateral sinus disease in whom intra-patient paired cultures of affected and unaffected left versus right paranasal sinuses were obtained . In that study, we found very similar rates of bacterial culture positivity in control versus radiographically diseased paired sinuses. Those findings argued against a strict bacterial etiology and pathogenesis for CRS. That study, however, was insufficiently powered for quantitative analysis of bacterial recovery rates in paired control-affected sinuses. Therefore, we undertook this study to extend the previous findings to a quantitative analysis of bacterial organism recovery rates in both diseased and paired, undiseased paranasal sinuses within patients with CRS. Our goal was to determine if there was a quantitative difference in bacterial recovery among paired sinuses which, if indeed a greater quantity of pathogenic bacteria were obtained from the diseased sinuses relative to the control sinuses, might still argue for a bacterial role in the pathogenesis of CRS.|
|Over a five year period, a consecutive series of adult patients undergoing primary endoscopic sinus surgery (ESS) for medically refractory CRS was tracked for patients with strictly unilateral sinus disease based on Lund-MacKay staging of the preoperative paranasal sinus CT scan. Patients with recurrent acute rhinosinusitis were excluded as were patients with suspicion for inverting papilloma, sinonasal malignancy or other unusual causes of unilateral sinus disease (e.g. Wegener’s granulomatosis). This study was approved by our institutional review board for clinical investigations. At the time of ESS, microbiological swabs with a micro-tip culturette were taken of the radiographically diseased sinus on one side; an analogous sinus culture was taken of the same non-diseased paranasal sinus on the contralateral side. Care was taken to endoscopically obtained the cultures without cross-contamination from the nasal vestibule, turbinates or other paranasal sinus structures. When necessary, sterile nasal specula were used to shield the culture tip from these structures. Cultures were immediately transported to microbiology and submitted for aerobic and anaerobic culture analysis with standard microbiological sensitivity analysis.|
|For each control-diseased sinus pair culture results were tabulated and entered into SPSS version 16.0 (Chicago, Illinois). The first two bacterial species isolated were tabulated along with the first two isolated anaerobic species. Quantitative values for the volume of a covered bacteria were determined by serial dilution and recorded on a range from 0 (equivalent to no growth) through 4+. Also tabulated were the bacterial resistance patterns for recovered isolates. Sensitivity or resistance of the bacterial isolate was determined according to the cut points elaborated by the National Committee for Clinical Laboratory Standards. For the purposes of this study, intermediate resistance to an antibiotic was considered as frank resistance.|
|Power analysis with data regarding proportion of microbiologically positive sinuses from our previous study indicated that 79 diseasedcontrol paired sinuses would be required for quantitative analysis of bacterial culture results in order to detect a 0.5 difference in quantitative bacterial recovery between the diseased-control sites. A target sample size of 85 patients was designed to allow for errors in sample collection and/or incomplete microbiological data. Statistical analysis was conducted with paired sample nonparametric testing for the quantitative culture outcomes with statistical significance set at p=0.05. Quantitative comparisons for the paired sinuses were conducted for (1) all positive cultures excluding oral flora considered as pathologically positive and repeated with (2) positive cultures excluding both oral flora and coagulase negative staphylococcal species, considered as pathologically positive. Finally, tabulated resistance levels for recovered bacteria were compared between the diseased side and the control side to determine if on the disease side, more pathologically purulent (i.e. greater antimicrobially resistant) bacteria might be recovered.|
|A total of 87 adult patients with paired non-diseased and diseased sinus sides were enrolled; 5 patients were excluded due to incomplete microbiological data leaving a net of 82 patients for statistical analysis. . Cultures were obtained from the ostiomeatal complex region (55 patients) posterior ethmoid (10 patients), maxillary sinus (11 patients) and sphenoid sinus (5 patients). When considering positive cultures for any bacteria excluding oral flora, 71/82 (86.6%) of the radiographically diseased sinuses were positive on culture versus 71/82 (86.6%) of the paired equivalent control sinuses o me and him n the contralateral side (p=0.999). When coagulase negative staphylococcal species were considered non-pathological, 44/82 (53.7%) of radiographically diseased sinuses were culture positive versus 44/82 (53.7%) among non‑diseased paired sinuses (p=0.999). Table 1 presents the relative frequencies of recovered bacteria among the diseased and control sinuses.|
|Table 1: Distribution of bacterial isolates between diseased and control paranasal sinuses.|
|Figure 1 compares the quantitative bacterial recovery rate for diseased versus control paired sinuses considering all bacteria pathological except oral flora. For this pairing, there was no statistically significant difference in quantitative bacterial recovery between the diseased and controlled paired sinuses (p=0.183). Similarly, Figure 2 compares the quantitative bacterial recovery rate for diseased versus controlled paired sinuses, considering all bacteria pathologically positive except oral flora and coagulase negative staphylococcal species. For this pairing, there was no statistically significant difference in quantitative bacterial recovery rates either (p=0.333). Finally, Figure 3 compares the quantitative resistance (average number of resistances to antibiotics tested for each bacterium recovered) between the diseased sinus and its control pair. Again, no statistically significant difference was found between the two sides (p=0.851).|
|Figure 1: Comparison of quantity of bacteria recovered from bilaterally paired diseased versus non-diseased paranasal sinuses (all bacteria considered pathogenic except for oral flora; p=0.183).|
|Figure 2: Comparison of quantity of bacteria recovered from bilaterally paired diseased versus non-diseased paranasal sinuses (all bacteria considered pathogenic except for oral flora and coagulase negative staphylococcal species; p=0.333).|
|Figure 3: Comparison of number of antimicrobial resistances encountered among bacteria recovered from bilaterally paired diseased and non-diseased paranasal sinuses (p=0.851).|
|Despite its relative prevalence, our understanding of CRS in terms of its pathogenesis, modifying factors and optimal treatment patterns remains poorly delineated. Over the past two decades, the general consideration of CRS is an infectious process has transitioned to consideration of CRS as an inflammatory process, possibly with bacterial superinfection or epiphenomenon. Complex interactions between various microbiological organisms and the host paranasal sinus tissues have also been conceived in the pathogenesis of CRS . We hypothesized that if bacterial infection were involved in the pathogenesis of CRS, we should be able to recover quantitatively higher amounts of bacteria from radiographically diseased paranasal sinuses versus those sinuses without any evidence of disease. We designed a prospective, paired study with intra-patient controls to ascertain differences in quantitative bacterial recovery in this regard.|
|With the adequately powered current study, we were unable to demonstrate a significant difference in overall bacterial recovery rates and further unable to show a difference in quantitative bacterial recovery rates between diseased and non-diseased (control) sinuses within the same patient. The use of the intrapatient comparison controls for the presence of modifying factors such as cigarette smoking, the presence of asthma, and other conditions that may affect bacterial recovery rates. The current results are remarkably consistent with our previously published studies regarding gross bacterial presence in diseased versus nondiseased sinuses . However, one potential criticism of our previous study was that it was inadequately powered to demonstrate quantitative difference in bacterial recovery amounts when comparing diseased versus nondiseased sinuses.|
|Although the current data fail to demonstrate significant differences in bacterial recovery between diseased and nondiseased sinuses, several added dimensions deserve mention in the interpretation of the results. Although the data argue against bacterial infection as a primary pathogenic cause of CRS, bacterial presence and/or infection may still play a role in the evolution of CRS, perhaps as a catalyst for inflammation or as an accelerant of the inflammatory state, once that inflammatory state in fully resident. Certainly, several elegantly conducted studies have demonstrated that complex interactions take place between various bacterial species and the local immune system within the paranasal sinuses that may further stimulate paranasal sinus mucosal inflammation . However, as indicated in Table 1, we paradoxically recovered more S aureus species from control sinuses than from diseased sinuses. Other investigators have derived similar results across CRS affected and unaffected patients .|
|The current data also corroborate previous investigations that have shown that the paranasal sinuses beyond the middle meatus, even in normal individuals, is not a sterile environment . As noted herein, different forms of bacterial species can be recovered from radiographically normal sinuses, further casting doubt on the role of bacteria in the pathogenesis of CRS. The role of biofilms also deserves specific mention. As a more advanced state of bacterial colonization or infection, biofilms are more durable than planktonic bacteria, and research has demonstrated higher levels of disease burden and poorer outcomes in the presence of bacterial biofilms . However, other investigations have also demonstrated that bacterial biofilms may be present in non-diseased, control sinonasal mucosa [10,11]. Thus, while our data suggest that the presence of bacteria alone does not likely drive chronic rhinosinusitis, transformation (perhaps due to alterations of the mucosal state) of the bacterial surface presence to a biofilm state may still accelerate and worsen the disease process. Finally, although bacteria may be recovered at similar rates between control and disease sinuses, one might hypothesize that a more virulent form of the bacterium might be present in the disease sinus as a driver of the inflammation. For example, S. aureus may indeed be present on both the control and diseased sides, but perhaps the variant of S. aureus on the diseased side may be a multi-drug resistant strain, adding to its pathogenicity. However, using degree of bacterial antimicrobial resistance as a surrogate marker of virulence, we did not see differential expression of bacterial resistance between control and diseased sinuses, making that less likely an explanation for the role of bacteria in CRS.|
|A few limitations of our study merit mention. First, we conceived the unilateral disease intra‑patient control study design to reduce the potential for disease modifying confounders, of which there are many, in CRS. However, this in turn may limit the generalizability of these results, i.e. extending from unilateral CRS cases to the bilateral CRS population. Nonetheless, all of these patients met rhinosinusitis task force criteria for the diagnosis of CRS with symptoms for greater than three months prior to surgery and had positive radiographic disease (Lund-MacKay score greater than 3 on the diseased side). Qualitatively, the type of mucosal disease encountered at the time of surgery on the diseased side was very akin to what we typically encountered when we do ESS for bilaterally diseased cases. Similarly, the pathological findings from the histopathology of the sinus tissue removed were qualitatively quite similar to those we have encountered in bilateral CRS cases. Nonetheless, it has to be recalled that unilateral cases of CRS may be different than bilateral CRS cases in their clinical behavior, pathogenesis and outcome. Our findings need to be taken in this context. A second potential limitation is the difficulty in correlation between the patients’ symptoms and the radiological findings. We use the Rhinosinusitis Symptom Inventory to quantify patients’ symptoms prior to surgery. However that does not distinguish between unilateral and bilateral disease symptomatology. Furthermore, we have found that patients often have difficulty in identifying a dominant sidedness for their unilateral sinus disease. This is often because CRS does not exist always alone; there can be concurrent diagnoses of allergic rhinitis, etc. This is an area that would benefit from future study. However, the diagnostic capability of computed tomography in identification of CRS has been well studied and was based upon these diagnostic criteria that patients were selected, in addition to symptoms and duration, for this study .|
|Based on this prospective study with an intra‑patient control, very similar rates of quantitative bacterial recovery were found between radiographically normal (control) and diseased paranasal sinuses in patients with unilateral CRS. These data argue against a significant role for persistent bacterial infection as a primary source of pathogenesis of unilateral CRS. These data may have implications as well for the pathogenesis of bilateral CRS.|