Abstract
Objective:
Chronic rhinosinusitis with nasal polyps (CRSwNPs) remains a major challenge due to its high recurrence rate after endoscopic sinus surgery (ESS). We aimed to investigate the risk factors of recurrence among patients who underwent ESS for Chronic rhinosinusitis (CRS).
Methods:
Prospective cohort study including 391 cases in a single institution receiving ESS were included for analysis from 2014 and 2017. Baseline characteristics including rectal Staphylococcus aureus (S aureus) carriage in patients receiving ESS for CRSwNPs. The primary outcome was the recurrence of CRSwNPs. Multivariate regression model was established to identify independently predictive factors for recurrence.
Results:
Overall, 142 (36.3%) cases with recurrence within 2 years after ESS were observed in this study. After variable selection, multivariate regression model consisted of 4 variables including asthma (odds ratio [OR] = 3.41; P < .001), nonsteroidal anti-inflammatory drug allergy (OR = 2.27; P = .005), previous ESS (OR = 3.64; P < .001), and preoperative carriage of S aureus in rectum (OR = 2.34; P = .001).
Conclusions:
Based on our results, surgeons could predict certain groups of patients who are at high risk for recurrence after ESS. Rectal carriage of S aureus is more statistically related to the recurrence of CRSwNP after ESS compared with skin and nasal carriage.
Introduction
Chronic rhinosinusitis with nasal polyps is estimated to affect 5% to 10% of the overall population and the second most common chronic condition in the United states. 1,2 Patients with CRSwNP present with reduced quality of life, poor sleep and fatigue, resulting in a significant productivity loss. 3 The annual productivity cost associated with CRSwNP is more than $10 000 per patient in the United States. 4,5 After failure of medical treatment, endoscopic sinus surgery (ESS) is commonly performed with more than 1 000 000 cases per year in American and European countries. 6,7 Unfortunately, the disease recurrence rate could reach up to 35% in patients after ESS for CRSwNP. 8,9
Despite the significant disease burden of CRSwNP, the understanding of the pathogenesis of CRSwNP remains elusive. Previous studies have identified many risk factors for disease recurrence after ESS including previous ESS, asthma, nonsteroidal anti-inflammatory drug (NSAID) allergy, higher Lund-Kennedy (LK) score, elevated serum immunoglobulin E (IgE) and increased eosinophil levels. 10 -12 Recent studies demonstrated that the rate of nasal carriage of Staphylococcus aureus (S aureus) was also a risk factor and early sign of the disease recurrences after ESS for CRS, suggesting nasal carriage of S aureus might play an important role in CRSwNP pathogenesis. 13 This study also revealed that the prevalence of nasal carriage of S aureus increased after ESS compared with preoperative level. 13 The nasal S aureus decolonization which might reduce the recurrence rate usually had a very high failure rate possibly due to the overlapping carriage of S aureus in multiple host sites including skin and gut. 14 -16 These facts prompted us to ask the following questions: Is rectal or skin carriage of S aureus an independent risk factor for CRSwNP recurrence?
Methods
This study was approved by the Ethics Committee of our hospital. Informed consents were obtained from all participants. Patients were diagnosed according to the criteria in the European Position Paper on Rhinosinusitis and Nasal polyps 2012 definition were enrolled. 17 Adult patients elected to ESS (>18 years old) were prospectively enrolled from January 2014 to December 2017. Exclusion criteria included receiving topical or systemic antibiotic in 4 weeks prior to the surgical intervention, ciliary dyskinesia, mucoceles, any benign tumors in head and neck, and any history of malignancy were excluded from the study. Patients who did not complete the 2-year follow-up were excluded from analysis. The work has been reported in line with the Strengthening the Reporting of Cohort Studies in Surgery (STROCSS) criteria. 18
Patient Evaluation
Seasonal allergies, ethanol consumption, and smoking were based on patient report. All patients had the following laboratory tests before surgery: total IgE, C-reactive protein, and whole blood cell count. Endoscopic evaluation of the sinonasal cavities was performed and scored according to the modified LK scoring system. 19 Swab culture sampling was obtained from anterior nares, fingers and rectum one day before ESS and one month after ESS. Samples were processed by our institutional laboratory for culturing and bacterial identification for both aerobic and anaerobic bacteria.
Surgical and Postoperative Intervention
All surgeries were performed by the senior specialists in ESS. Surgical approach was determined by the intraoperative discretion of each treating surgeon. Typically, patients received frontal sinus procedure (frontal sinusotomy, modified Lothrop procedure, or Draf III), maxillary antrostomy, ethmoidectomy, and sphenoidotomy. Additional procedures included medial maxillectomy, septoplasty, and middle turbinate resection were performed if needed. All cases in this study were operated under general anesthesia with use of image guidance. Postoperative management including a 3-month 200-mL nasal saline rinses and fluticasone propionate nasal spray at 200 µg on daily basis. Oral prednisone was given as a rescue alternative therapy determined by the discretion of treating surgeons as needed.
Definition of Disease Recurrence
We used the same criteria used for preoperative diagnosis of CRSwNP, where objective evidence along with clinical symptoms of disease must be present together. Briefly, the disease status was defined as mucosal edema >1 on the modified LK endoscopic scoring system. 19 Any evidence of polyposis was considered to have CRSwNP even when the mucosal edema was 0 or 1. The patients were typically followed up at 1, 3, 6, 12, 18, and 24 months to collect patient-reported outcomes by treating surgeons. The evaluation on recurrence was conducted at 2 years after initial surgery for all patients.
Statistical Analysis
In all analyses of this study, statistical significance was defined as a P value <.05. All data were tabulated using EpiData, and all statistical analyses were performed using SPSS statistics 26.0 (IBM). Continuous and categorical variables are presented as mean ± standard deviation and number (proportion), respectively. Variables with continuous and categorical values were evaluated using the Mann-Whitney U test and Fisher exact test, respectively. All variables were included in multivariate analysis to establish the original model. The Bayesian information criterion was then used to reduce variables in the model to reduce overfitting. The discriminative performance of the model was measured using a concordance index (C-index) after variable selection.
Results
A total of 432 study participants completed enrollment procedures and received ESS from January 2014 to December 2017. Finally, 391 patients completed at least 2-year follow-up and included for analysis. The follow-up of 39 patients were lost within the 2 years after surgery. One patient died of car accident and one patient withdrew from the study due to malignancy. We performed a time-fixed analysis to evaluate the recurrence rate within 2 years after surgery. Overall, 142 (36.3%) cases with recurrence within 2 years after ESS were observed in this study.
Demographic and clinical variables are summarized in Table 1 after grouping patients by recurrence. To avoid missing any important variables, we did not set a P value threshold for multivariate model inclusion. Instead, we included all variables and using Bayesian information criteria for variable selection. After reducing variables, the final model (Table 2) consisted of 4 variables including asthma (odds ratio [OR] = 3.41; P < .001), NSAID allergy (OR = 2.27; P = .005), previous ESS (OR = 3.64; P < .001), and preoperative carriage of S aureus in rectum (OR = 2.34; P = .001). The C-index of this model was 0.82 suggesting a good accuracy in recurrence prediction. Based on the final model, the adjusted P values of skin (P = .474), nasal (P = .356), and any-site (P = .131) carriage of S aureus were all >0.05, indicating the more important role of rectal carriage compared with carriage at other sites.
Preoperative Demographics and Clinical Characteristics of 391 Patients Grouped by Recurrence.
Abbreviations: ESS, endoscopic sinus surgery; IgE, immunoglobulin E; NSAID, nonsteroidal anti-inflammatory drug; S aureus, Staphylococcus aureus.
a Asthma refers to the diagnosis of asthma irrespective of its control status.
Multivariate Logistic Regression Model After Variable Selection Using Bayesian Information Criteria.
Abbreviations: ESS, endoscopic sinus surgery; NSAID, nonsteroidal anti-inflammatory drug; S aureus, Staphylococcus aureus.
Discussion
This study investigated the prevalence of disease recurrence in a prospective cohort of patients receiving ESS for CRSwNP. The definition of disease recurrence varies among studies, including polyposis recurrence, 12 edema recurrence, 12 need for revision surgery 11 and the same criteria we adopted in current study. 13 Mild edema is extremely common after ESS and not entirely related to the severity of symptoms. 19 We thus used 2 or more points in edema subscale of LK endoscopic scoring system as our criteria. Clearly, polyposis recurrence is a sign of recurrence without controversies. It is generally agreed that NSAID allergy and asthma were risk factors for recurrence according to previous literatures. 13 Mendelsohn et al suggested that asthma and aspirin sensitivity were robust risk factor for the recurrence of nasal polyposis. 11 However, a more recent multicenter study involving 244 patients failed to find significant risk of these 2 factors on recurrent polyposis literatures. 12 The major critique on this study was its insufficient power. In this study, we included more cases and found these 2 factors were robustly related to the recurrence. Generally, the results of this study demonstrate that the recurrence rate after ESS for CRSwNP is currently unsatisfactory. Although postoperative medical management including tropical steroids can reduce recurrence rates, revision surgery remained inevitable for many patients.
The role for S aureus in CRS development or recurrence has been long assumed because the prevalence of nasal S aureus carriage is higher in patients before and after ESS than in healthy ones. 20 The biofilm-forming capacity of S aureus colonized in nasal cavity was associated with poor prognosis. 21 Previous studies all focused on nasal carriage alone but omitted the overlapping carriage. The novel and interesting finding of current study was that preoperative rectal S aureus carriage is an independent risk factor for disease recurrence and better in prediction than nasal carriage. Clearly, nasal S aureus might directly impact on the pathogenesis of CRS. However, rectal carriage was more statistically related to recurrence in this study. One possible explanation was that approximately 50% of adult nasal carriers were persistent, whereas most adult rectal carriers were persistent. 22 -24 Recent studies have revealed that gastrointestinal S aureus could colonize other host sites via classic fecal–oral route, transient bacteremia, and Trojan-horse mechanism. 16,25,26 This might explain the increase of nasal carriage after ESS. Further investigation is needed to elucidate this important point.
In addition, it remains unclear whether preoperative decolonization of rectal S aureus is beneficial to disease control and recurrence prevention. Future clinical study is needed to elucidate this issue. Because of the overlapping carriage of S aureus in multiple host sites including skin, nasal cavity, and gut, we believe the decolonization should involve multiple sites in future clinical studies. Isolated decolonization might result in failure because the site could be recolonized rapidly. The major strength of this study is its prospective design and large sample size. To our knowledge, this study has the largest sample size and we have objective evidences on recurrence instead of all relying on the patient-reported outcomes.
Limitations
This study has several limitations. First, 2 years may be considered as a short follow-up period after ESS. However, previous studies have demonstrated that the majority of recurrent CRS developed within 2 years after ESS. 11,12 Second, variation in the loss of follow-up might introduce a potential source of selection bias. Those patients who had larger symptom resolution after ESS could be less likely to complete follow-up compared with patients having disease recurrence. However, the rate of loss to follow-up was less than 10% and thus it is highly possible that the main conclusion would not be affected. Third, the subjective nature of LK endoscopy scoring system could inevitably lead to interrater and intrarater variation in rating.
Conclusions
Based on our results, surgeons could predict certain groups of patients who are at high risk for recurrence after ESS. Although nasal S aureus carriage might be the direct factor involved in the pathogenesis of CRS, rectal carriage of S aureus which is generally more persistent is more statistically related to the recurrence of CRSwNP after ESS compared with skin and nasal carriage. Future studies were needed to compare the efficacy of traditional nasal decolonization with that of multisite decolonization.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
