Abstract
Background
Endoscopically guided cultures are frequently employed to guide antimicrobial therapy in refractory chronic rhinosinusitis (CRS) patients. The objective of this study was to determine the impact of culture-directed antibiotics on patient symptoms.
Methods
Retrospective review was conducted of 105 adult CRS patients undergoing evaluation in the ambulatory clinic of tertiary care academic medical center.
Results
The most common microbes were Staphylococcus aureus (29.5%), Pseudomonas aeruginosa (23.8%) and methicillin-resistant S. aureus (11.4%). Normal respiratory flora or no growth was found in 19% of patients. Culture results changed antibiotic choices in 77% of patients. Statistically significant change in total SNOT-20 scores and all 4 subdomains was noted, with improvement being clinically meaningful in the rhinologic subdomain (− 1.10, p < 0.0001). Repeat purulence was only noted in 5 cases (4.8%). Multivariate regression analysis demonstrated that concurrent use of oral steroids was independently associated with improvement in the rhinologic subdomain (p = 0.0041). The mean length of follow-up was 37 days. Length of follow-up (14–30, 31–60, 61–90 days) did not statistically impact SNOT-20 scores.
Conclusion
Endoscopic-derived sinus cultures are associated with clinically meaningful change in the rhinologic subdomain of SNOT-20 scores, and repeat purulence was infrequently noted at follow-up. Further prospective studies are needed to better delineate the role of cultures in CRS management.
Level of evidence
4.
1
Introduction
Chronic rhinosinusitis (CRS) represents an inflammatory disorder of the nose and paranasal sinuses characterized by sinonasal symptoms with evidence of disease by endoscopy or radiographic imaging of greater than 12 weeks duration . A variety of environmental and host mechanisms have been implicated in the etiology of CRS including presence of microbes (bacteria, fungus), allergy, ciliary dysfunction, derangements in innate and adaptive immunity, biofilm formation, and osteitis . The exact role of bacterial pathogens remains to be fully elucidated to date.
The utility of antibiotics in the management paradigm of CRS has been a cause of significant debate. Nonetheless, they are frequently employed to treat infectious exacerbations of CRS . The microbiology of CRS, especially in the post-sinus surgery setting can be quite disparate, with frequent presence of Staphylococcus aureus , coagulase negative Staphylococcus (CNS), and gram negative rods . Thus, endoscopically guided cultures (EGC) are commonly used to guide antimicrobial therapy, with one previous study noting EGCs resulting in change in antibiotic choice in 51.4% of cases . However, there is a paucity of data addressing the essential question if antibiotics derived from EGCs impact objective symptom scores. The present study was conducted to better ascertain the impact of endoscopically driven antibiotic therapy on patient symptoms and endoscopic findings in CRS patients.
2
Materials and methods
The study was conducted at the University of Texas Southwestern Medical Center from July 2010 to October 2012. Patients for inclusion were identified from the senior author’s (PSB) clinical practice using ICD-9 codes for CRS (473.0, 473.1, 473.2, 473.3, 473.8, and 473.9). The patient list was then cross-referenced to the central microbiology registry to identify patients that underwent sinus cultures. Institutional review board approval was obtained from UT Southwestern Medical Center prior to commencing the study.
The inclusion criteria included diagnosis of CRS with or without polyposis in patients with age ≥ 18 years. Acute exacerbation of CRS, defined by minimum SNOT-20 score of 1.0 on scale of 0 to 5, was required for inclusion. This was based on 0.8 being deemed a clinically meaningful change by Picirillo et al. and would afford the ability to detect a discernable difference post-treatment . All patients used nasal saline irrigations and topical nasal steroid for maintenance therapy. Patients were excluded who received oral, topical and/or intravenous antibiotic therapy within one week of presentation or who did not have purulence on presenting nasal endoscopy. One week constitutes more than 4.5 half-lives for most antibiotics and, thus, it was assumed that after one week, the level of antibiotics in patient’s serum and soft tissue would be clinically negligible. All initial and follow-up endoscopies were performed by the senior author. Patients with follow-up shorter than 2 weeks were excluded to ensure adequate time for antibiotic efficacy to manifest in clinical changes. Similarly, follow-up greater than 90 days were also excluded to minimize the likelihood that the clinical would be influenced by other factors.
All patients presented with acute exacerbation of sinonasal symptoms with evidence of purulent secretions on endoscopy. A rayon-tipped swab (Bactiswab®, Remel Products, Lenexa, KS) or Lukens trap (Cardinal Health, Dublin, OH) was used to collect secretions from the site of purulence on nasal endoscopy. The microbiology laboratory protocol involved special handling of the sinus cultures. The cultures were initially handled on broad spectrum media. Specific colonies of certain bacteria commonly seen in CRS, such as S. aureus , Pseudomonas aeruginosa , Escherichia coli , Serratia marcescens , and Stenotrophomonas maltophilia , were assayed and reported when present. If two or three microbes were equally present, then the results were reported as polymicrobial. Normal respiratory flora was reported if none of these bacteria were identified. For patients undergoing multiple cultures, only the initial culture result was included.
Other data collected included patient demographics, co-morbidities (asthma, inhalant allergy, aspirin exacerbated respiratory disease, polyps), surgical history, site of culture, key pathogens, and antibiotic choices. SNOT-20 scores were prospectively recorded at the initial visit and then on follow-up visit after completion of treatment. Evidence of purulence on follow-up endoscopy was also ascertained.
2.1
Statistical analysis
Descriptive statistics of the study population were performed with mean and standard deviation being calculated for continuous variables and frequencies with percentage being calculated for categorical variables. The change in overall and sub-scale (rhinologic, ear/facial, psychological, sleep) SNOT scores were determined by using the following formula: delta score = post-treatment score − pre-treatment score. Univariate analysis was conducted to determine the impact of age, gender, number of previous surgeries, smoking history, asthma, aspirin sensitivity, polyps, organism type, and use of concurrent oral steroids on SNOT-20 scores. Association between each risk factor and overall and sub-scale SNOT-20 scores was calculated using Pearson correlation for continuous variables and two-sample t-test for categorical variables.
Multivariate stepwise logistic regression model was used to identify significant independent risk factors impacting post-treatment SNOT-20 scores while adjusting for all other risk factors. ANOVA test was used to test the association between SNOT-20 scores and time interval between exams (14–30, 31–60, and 61–90 days). Two-sided significance level of 0.05 was used for all statistical analyses. All analyses were carried out using SAS version 9.2 (SAS Institute Inc, Cary, NC).
2
Materials and methods
The study was conducted at the University of Texas Southwestern Medical Center from July 2010 to October 2012. Patients for inclusion were identified from the senior author’s (PSB) clinical practice using ICD-9 codes for CRS (473.0, 473.1, 473.2, 473.3, 473.8, and 473.9). The patient list was then cross-referenced to the central microbiology registry to identify patients that underwent sinus cultures. Institutional review board approval was obtained from UT Southwestern Medical Center prior to commencing the study.
The inclusion criteria included diagnosis of CRS with or without polyposis in patients with age ≥ 18 years. Acute exacerbation of CRS, defined by minimum SNOT-20 score of 1.0 on scale of 0 to 5, was required for inclusion. This was based on 0.8 being deemed a clinically meaningful change by Picirillo et al. and would afford the ability to detect a discernable difference post-treatment . All patients used nasal saline irrigations and topical nasal steroid for maintenance therapy. Patients were excluded who received oral, topical and/or intravenous antibiotic therapy within one week of presentation or who did not have purulence on presenting nasal endoscopy. One week constitutes more than 4.5 half-lives for most antibiotics and, thus, it was assumed that after one week, the level of antibiotics in patient’s serum and soft tissue would be clinically negligible. All initial and follow-up endoscopies were performed by the senior author. Patients with follow-up shorter than 2 weeks were excluded to ensure adequate time for antibiotic efficacy to manifest in clinical changes. Similarly, follow-up greater than 90 days were also excluded to minimize the likelihood that the clinical would be influenced by other factors.
All patients presented with acute exacerbation of sinonasal symptoms with evidence of purulent secretions on endoscopy. A rayon-tipped swab (Bactiswab®, Remel Products, Lenexa, KS) or Lukens trap (Cardinal Health, Dublin, OH) was used to collect secretions from the site of purulence on nasal endoscopy. The microbiology laboratory protocol involved special handling of the sinus cultures. The cultures were initially handled on broad spectrum media. Specific colonies of certain bacteria commonly seen in CRS, such as S. aureus , Pseudomonas aeruginosa , Escherichia coli , Serratia marcescens , and Stenotrophomonas maltophilia , were assayed and reported when present. If two or three microbes were equally present, then the results were reported as polymicrobial. Normal respiratory flora was reported if none of these bacteria were identified. For patients undergoing multiple cultures, only the initial culture result was included.
Other data collected included patient demographics, co-morbidities (asthma, inhalant allergy, aspirin exacerbated respiratory disease, polyps), surgical history, site of culture, key pathogens, and antibiotic choices. SNOT-20 scores were prospectively recorded at the initial visit and then on follow-up visit after completion of treatment. Evidence of purulence on follow-up endoscopy was also ascertained.
2.1
Statistical analysis
Descriptive statistics of the study population were performed with mean and standard deviation being calculated for continuous variables and frequencies with percentage being calculated for categorical variables. The change in overall and sub-scale (rhinologic, ear/facial, psychological, sleep) SNOT scores were determined by using the following formula: delta score = post-treatment score − pre-treatment score. Univariate analysis was conducted to determine the impact of age, gender, number of previous surgeries, smoking history, asthma, aspirin sensitivity, polyps, organism type, and use of concurrent oral steroids on SNOT-20 scores. Association between each risk factor and overall and sub-scale SNOT-20 scores was calculated using Pearson correlation for continuous variables and two-sample t-test for categorical variables.
Multivariate stepwise logistic regression model was used to identify significant independent risk factors impacting post-treatment SNOT-20 scores while adjusting for all other risk factors. ANOVA test was used to test the association between SNOT-20 scores and time interval between exams (14–30, 31–60, and 61–90 days). Two-sided significance level of 0.05 was used for all statistical analyses. All analyses were carried out using SAS version 9.2 (SAS Institute Inc, Cary, NC).
3
Results
3.1
Demographics
A total of 105 patients were included based on the inclusion criteria. The mean age was 46.3 years and 43.8% were males. Polyps were noted in 53 (51.0%) patients. The presence of inhalant allergy, asthma, and ASA sensitivity was noted in 58 (55.2%), 39 (37.1%), and 5 (4.8%) cases, respectively. The mean number of previous endoscopic sinus surgeries (ESS) was 1.9 (range 0–8). Overall, 96 (91.4%) patients had undergone previous sinus surgery. The mean follow-up after initial antibiotic treatment was 37 days (range: 14–85 days).
3.2
Culture and treatment data
The location of the sinonasal cultures included middle meatus in 56 (53.3%), maxillary sinus in 32 (30.5%), nasal cavity in 7 (6.7%), sphenoid sinus in 3 (2.9%), ethmoid cavity in 3 (2.9%), frontal recess in 2 (1.9%), and unspecified in 2 (1.9%) cases. The mean number of pathogens on the aerobic culture was 1.2 (range 0–3), with 32 (30.5%) cultures being polymicrobial ( Table 1 ). No growth and normal respiratory flora was reported in 6 (5.7%) and 14 (13.3%), respectively.
