Potential conflict of interest: R.S. is a consultant for Medtronic and Nycomed, and on the speakers’ bureau for Merek Sharp Dolme and Arthrocare. R.J.H. has served on an advisory board for Schering Plough and Glaxo-Smith-Kline, is a previous consultant with Medtronic and Olympus, is on the speakers’ bureau for Merek Sharp Dolme and Arthrocare, and has received grant support from NeilMed.
The impact of neo-osteogenesis on disease control in chronic rhinosinusitis after primary surgery
Article first published online: 24 JUN 2013
© 2013 ARS-AAOA, LLC
International Forum of Allergy & Rhinology
Volume 3, Issue 10, pages 823–827, October 2013
How to Cite
How to Cite this Article: The impact of neo-osteogenesis on disease control in chronic rhinosinusitis after primary surgery. Int Forum Allergy Rhinol. 2013;3:823–827., , , , ,
- Issue published online: 15 OCT 2013
- Article first published online: 24 JUN 2013
- Manuscript Accepted: 16 MAY 2013
- Manuscript Revised: 14 MAY 2013
- Manuscript Received: 13 MAR 2013
- chronic rhinosinusitis;
- intranasal corticosteroid;
Osteitic bone is a feature of chronic rhinosinusitis (CRS), potentially playing a role in pathogenesis. Although seen after previous endoscopic sinus surgery (ESS), it is also a de novo feature in patients without prior interventions. In these patients, osteitis is associated with high tissue and serum eosinophilia. However, the impact of osteitis on prognosis is unclear. This study investigates the clinical and endoscopic outcomes between patients with and without osteitis after primary ESS.
A prospective study of a cohort of previously unoperated patients with CRS undergoing ESS was performed. The sinuses were scored radiologically for osteitis using the Global Osteitis Score (GOS) and Kennedy Osteitis Score (KOS) preoperatively and were also scored dichotomously for the presence or absence of osteitis. Disease-specific quality of life (22-item Sino-Nasal Outcomes Test [SNOT-22]), nasal symptom score (NSS), endoscopic score (Lund-Kennedy), and clinical outcomes—including oral steroid use, frequency of nasal steroid irrigation, and infective exacerbations—were collected at baseline and 1 year postsurgery. The presence and extent of osteitis was assessed relative to clinical outcome.
Fifty-three patients were included (41.5% female, age 47.4 ± 13.8 years), 42.9% of which had radiologic osteitis. There was no significant association between the presence or severity of osteitis at the time of surgery and SNOT-22, NSS, or endoscopy scores at 12 months postsurgery. However, the presence of osteitis was associated with the need for a course of oral steroid postsurgery (odds ratio [OR]=4.17; p = 0.026). High tissue eosinophilia could not predict this alone (p = 0.55). There was no significant relationship between osteitis and the frequency of steroid irrigations or infective exacerbations.
Osteitis in CRS is associated with the degree of eosinophilia and as a independent process it was associated with the need for a course of systemic corticosteroid over a 12-month period, but did it not affect overall disease control.
Chronic rhinosinusitis (CRS) is regarded to be a multifactorial chronic inflammatory condition of the nose and paranasal sinuses. Osteitic bone, whether truly osteitic or simply neo-osteogenic, has been well described as a feature of CRS, potentially playing a critical role in the pathogenesis of the inflammatory changes evident in the disease[1-3] (fig. 1). While many consider the term “osteitis” to be a misnomer and to instead represent an osteal or bony reaction to inflammation, the term osteitis is used herein to describe radiologic osteal thickening. Concurrent osteitis has been identified in 36% to 53% of patients with CRS using radiographic[3-5] and histopathologic criteria. While osteitis has been shown to be significantly more common in revision cases than in non-operated sinuses,[6, 7] primary cases also exhibit osteitic changes, with an incidence of 5% to 36%.[4, 5] This suggests that mucosal trauma secondary to surgery is not solely responsible for the pathogenesis of osteitis. Patients with high tissue and serum eosinophilia correlate with higher osteitis scores than those without[4, 5] and this is also the case with primary patients.
While several studies have investigated the endoscopic appearance and clinical implications of osteitis at the time of surgery, few studies have revealed the long-term implications of osteitic disease. Patients with osteitis have been shown to have greater disease severity endoscopically but not symptomatically at the time of surgery. In these studies, patients with osteitis have inferior outcomes post-ESS compared to those without any osteitis,[8, 9] but the cohorts include many patients with revision surgery, and thus a bias of poorer prognosis in the osteitis groups. This study aimed to investigate the clinical and endoscopic outcomes of osteitic disease at 1 year post-ESS in previously unoperated CRS patients.
Patients and methods
This was a prospective cohort study of patients undergoing primary ESS and ongoing treatment for CRS over a 12-month period. The study was approved by the St Vincents Hospital Human Research Ethics Committee.
Adult patients (age >18 years) with CRS with or without nasal polyps undergoing primary ESS were recruited from a tertiary referral clinic. CRS patients were defined according to the criteria set out by the European Position Paper on Rhinosinusitis and Nasal Polyps 2007 (EP3OS). Patients were excluded from the study if they had undergone previous ESS, had tumor pathology, or were immunocompromised. No patients had used oral corticosteroid for the 4 weeks prior to surgery. All patients underwent ESS after failed medical therapy and received postoperative therapy of once-daily nasal irrigation therapy of either budesonide (1 mg) or betamethasone dipropionate (1 mg) in 240 mL of normal saline solution. Treatment was continued daily for the first 3 months then tailored to disease control. Patients were reviewed every 3 months in the postoperative period.
Demographic data was recorded. A structured histopathology report was used to define inflammatory features of the disease. Histopathology reported tissue eosinophilia (<10 per high power field [HPF], >10 per HPF), Charcot-Leyden crystals (absent, present), eosinophil aggregates (absent, present), and severity of inflammation (absent, mild, moderate, or severe). Three HPFs were analyzed to reach a consensus. Asthmatic status was defined clinically if using an inhaled beta-agonist or corticosteroid.
Osteitis was defined as the process of bone thickening and was assessed at the time the patient was scheduled for surgery. Bony walls of all 10 sinuses were assessed for presence (dichotomously as present or absent), severity, and extent of osteitis using both the Global Osteitis Score (GOS) and Kennedy Osteitis Score (KOS). The cutoff between osteitic and non-osteitic bone was based on the original Kennedy grading system, in which osteitic bone was defined as any bony partition ≥3 mm thick involving any of the ethmoid, maxillary, or sphenoid sinuses. Normative values were interpreted as any values less than 3 mm. The thickness of the bone was measured by 2 separate researchers at varying points in the bony wall surrounding the maxillary and sphenoid sinuses as well as within the ethmoid partitions. The KOS is based on a previously described grading system for radiographic measures of ethmoid bone partitions and maxillary and sphenoid sinus walls. Georgalas et al. proposed the GOS based on a similar principle, adding that if greater than 50% of the sinus wall is involved, a higher score should be given.
Data was collected regarding endoscopic and patient-reported outcome measures (PROMs) at baseline preoperatively and 1 year postsurgery. Endoscopy scores were graded using the Lund-Kennedy endoscopic scoring system. For the PROMs, the 22-item Sino-Nasal Outcome Test (SNOT-22) was used for disease-specific quality of life assessment and a nasal symptom score (NSS) incorporating a 5-question, 5-point Likert scale for nasal obstruction, loss of smell, postnasal discharge, facial pain, and thick nasal discharge was also collected.
Treatment outcomes investigated included the number of oral steroid courses and number of infective exacerbations during the first postoperative year, evaluated at 1 year postsurgery. The frequency of use of steroid nasal irrigations was also collected and defined as no use, as needed (PRN), alternate days, daily, or twice daily use at 1 year postsurgery. The frequency of nasal irrigation after the first 3 to 6 months was determined by self-directed patient therapy and was a combination of compliance and symptom control. Patients were reviewed every 3 months; their endoscopy was also used to educate them and to guide the frequency of local pharmacotherapy. Most patients were instructed to continue a regular, even if infrequent, treatment regime.
Statistical analyses were performed using SPSS v20. Descriptive data was presented as percentage, mean, and standard deviation (SD) for parametric data, and median and interquartile range for nonparametric data. Student t test and Mann-Whitney U test (2-tailed) were used to compare unrelated groups of parametric and nonparametric data respectively. Chi-squared analysis was used for relationships of nominal variables. Pearson correlation coefficients were determined for linear relationship of scale variables. Spearman's correlation coefficients were used for ordinal values.
Fifty-three patients were included in this study; 41.5% were female (n = 22) and the mean age was 47.4 ± 13.8 years; 52.8% patients were diagnosed with CRS without nasal polyps (CRSsNP) (n = 28); 15.1% had a history of asthma (n = 8) and 13.2% were smokers (n = 7). The prevalence of osteitis in this primary patient group was 39.6% (n = 21). The mean interval between patients’ computed tomography (CT) scan and biopsy was 88 ± 121 days. There was no significant difference in the number of days between osteitis and non-osteitis patients nor for inflammation. The baseline PROMs, endoscopic mucosal and osteitis scores, and tissue eosinophilia are presented in table 1.
|Age, years||46.04 ± 15.14||46.17 ± 13.41||0.70|
|Gender female, n (%)||9 (39.1)||13 (43.3)||0.71|
|Diagnosis of CRSwNP, n (%)||12 (57.1%)||11 (34.4%)||0.09|
|SNOT||2.04 ± 1.02||1.80 ± 0.98||0.42|
|NSS||2.56 ± 1.35||2.11 ± 1.37||0.29|
|Endoscopy (LK)||6.58 ± 2.85||4.66 ± 3.52||0.04|
|CT (KOS)||4.35 ± 2.69||0|
|CT (GOS)||4.82 ± 3.62||0|
Total SNOT-22 score
Baseline SNOT-22 scores were 1.80 ± 0.98 without osteitis and 2.04 ± 1.02 with osteitis. At 12 months postsurgery the results were similar between groups; the mean SNOT-22 scores were 0.86 ± 0.61 without osteitis and 1.03 ± 0.80 with osteitis (p = 0.45). Total SNOT-22 scores at 12 months postoperation did not correlate with KOS (p = 0.29) nor GOS (p = 0.18). The changes in total SNOT-22 were 1.01 ± 1.02 and 0.95 ± 0.80, with and without osteitis, respectively (p = 0.77).
Mean baseline NSS were 2.56 ± 1.35 and 2.11 ± 1.37, with and without osteitis, respectively (p = 0.29). At 12 months postsurgery, the mean NSS was 1.05 ± 0.98 without osteitis and 1.14 ± 1.11 with osteitis (p = 0.77). NSS at 12 months postsurgery did not correlate with KOS (p = 0.74) nor GOS (p = 0.58). The change in NSS was 1.07 ± 1.13 without osteitis and 1.39 ± 1.21 with osteitis (p = 0.41). There was no statistical significance difference between groups.
Total endoscopy scores
At 12 months postsurgery, the total endoscopy score for patients without osteitis was 4.83 ± 3.97 and with osteitis was 4.06 ± 2.48 (p = 0.45). The total endoscopy scores did not correlate with KOS (p = 0.74) or GOS (p = 0.49).
Ten patients (18.9%) had used at least 1 course of oral corticosteroids. Of these patients, 7 (70%) had osteitis. Of the patients who had used no oral steroid in the first year postsurgery, 32.6% (n = 14) had osteitis (p = 0.05). The mean KOS and GOS in patients who had no courses of oral corticosteroids in the first year postsurgery were 1.28 ± 2.04 and 1.28 ± 2.04, respectively. In patients that had used at least 1 oral corticosterod course, the mean osteitis scores were KOS 3.70 ± 3.71 vs 1.28 ± 2.03 in those patients that used no oral corticosteroids (p = 0.028) (fig. 2). Similarly GOS were 4.50 ± 3.70 vs 1.26 ± 2.04 (p = 0.026). These results were statistically significant. High tissue eosinophilia could not predict this alone (p = 0.55).
The mean number of infective exacerbations in the first year postsurgery was similar, with 0.29 ± 0.90 vs 0.18 ± 0.48 (p = 0.68), with and without osteitis, respectively.
The number of steroid irrigations per week at 1 year postsurgery showed no statistical significance (table 2).
|Scoring scale||Steroid washes per week||Mean (SD)||p|
|Kennedy Osteitis Score||Never||0.50 (1.23)||p = 0.334|
|Alternate days||1.67 (3.20)|
|Twice daily||2.67 (3.06)|
|Global Osteitis Score||Never||0.50 (1.23)||p = 0.342|
|Alternate days||1.67 (3.20)|
|Twice daily||2.67 (3.06)|
While total endoscopy scores were significantly higher in osteitis patients at baseline, no difference in total endoscopy score was seen at 12 months postsurgery. Additionally, there was no difference at baseline or at 12 months postsurgery between patients with and without osteitis in mean SNOT-22 and NSS. Both patients with and without osteitis responded well to the combination treatment of ESS and postoperative corticosteroid irrigation.
Patients with osteitis were shown to be more likely to require an oral corticosteroid course in the 12 months postsurgery. High tissue eosinophilia could not predict this alone. It has been shown in the literature that the most challenging subgroup of high tissue eosinophilia achieves favorable outcomes after receiving postoperative corticosteroid irrigation.[13, 14] Such a treatment protocol for patients with high tissue eosinophilia results in disease control without the need for systemic corticosteroids. The reason patients with osteitis may require systemic corticosteroids is not yet known. While topical therapies may be beneficial for local disease control, they may lack total efficacy when the development of neo-osteogenesis is systemically driven or is a marker of systemic activation with serum eosinophilia. The pathogenesis of neo-osteogenesis in primary patients is poorly understood. While an association between T helper 2 (Th2) inflammation driven by interleukin 4 (IL4) and IL13 and osteitis in CRS patients has been suggested[5, 9, 15] and the concept of IL4 and IL13 playing an important role in the complex process of neo-osteogenesis has been proposed in animal studies, little human data exists. Oral corticosteroids may be required to suppress these cytokines and bone marrow–activated eosinophils when present.
In contrast to our findings, Kim et al. and Bhandarkar et al. have shown that patients with osteitis have inferior outcomes post-ESS compared to those without any osteitis. These studies investigated both primary and revision cases. Postoperative medications are not mentioned in their studies. The authors believe ESS alone cannot effectively control local inflammatory disease. The philosophy of ESS for eosinophilic CRS at our center is to create the access for topical therapies rather than a fundamental concept of relieving mechanical obstruction. Postoperative corticosteroid irrigation appears effective for diffuse CRS and the eosinophilic group in particular. Delivery of steroid via nasal irrigation combines the general therapeutic goals of topical management in providing pharmaceutical delivery and simultaneous mechanical lavage. Complete sinus distribution is achieved when a wide post-ESS corridor has been created. High-volume positive-pressure solutions allow pharmaceutical preparations to better contact sinus mucosa and may enhance the mechanical removal of mucus, inflammatory products, and bacteria/biofilms. Additionally, the inclusion of revision cases in these series potentially biases the patient selection to more severe disease and poor outcomes that have been associated with revision.
While osteitis is associated with a higher degree of eosinophilia, it has not been shown to have a significant impact on symptom control. Although more than 90% of patients were controlled with routine postoperative topical therapies, the presence of osteitis predicted the need for the use of a course of oral corticosteroid in the first year postsurgery. The prognostic value may be of benefit in treatment expectations for patients but did not prevent good long term mucosal control.