Get access

Correlation of the Kennedy Osteitis Score to clinico-histologic features of chronic rhinosinusitis


  • Potential conflict of interest: R.J.H. has served on an advisory board for Schering Plough and Glaxo-Smith-Kline, was a consultant with Medtronic, was on the speakers bureau for Merek Sharp Dolme and Arthrocare, and has received grant support from NeliMed. R.S. is a consultant for Medtronic and on the speakers bureau for Merek Sharp Dolme.

  • Presented orally at the Annual ARS Meeting on September 8, 2012, Washington, DC.

Correspondence to: Kornkiat Snidvongs, MD, The Australian School of Advanced Medicine, 2 Technology Place, Macquarie University, Sydney, NSW 2109, Australia; e-mail:,



Osteitis is a feature of chronic rhinosinusitis (CRS) and often associated with recalcitrant disease. Radiological characteristics of osteitic sinus changes are commonly reported in practice but the clinical and pathologic significance is poorly defined. The objective of this study was to correlate the Kennedy Osteitis Score (KOS) to clinico-histologic features of CRS.


A cross-sectional study of CRS patients undergoing sinus surgery was conducted. Osteitis was scored radiologically using the KOS. Associations between osteitis and histopathology, symptoms, 22-item Sino-Nasal Outcomes Test (SNOT-22), endoscopy, computed tomography (CT) mucosal score, and seromarkers were assessed. Interobserver correlation coefficient was performed. Additionally, the KOS was correlated to an alternate Global Osteitis Score.


A total of 88 patients were assessed (45.5% female, age 50.3 ± 13.6 years); 45 (51.1%) patients had osteitis. Patients with KOS >0, had greater endoscopy score (6.1 ± 2.9 vs 4.4 ± 3.6, p = 0.03) and CT score (14.0 ± 6.0 vs 10.1 ± 5.7, p < 0.01) than those without osteitis. There was no difference in symptom score (2.4 ± 1.3 vs 2.4 ± 1.1, p = 0.89) and SNOT-22 (2.0 ± 1.0 vs 1.9 ± 1.1, p = 0.56) in patients with and without osteitis. KOS was higher in patients with tissue eosinophilia >10/high-power field (HPF) (median 3.0 [IQR, 1.0–5.3] vs 0.0 [0.0–4.0], p = 0.03) and serum eosinophilia >0.3 × 109/L (4.0 [2.0–7.0] vs 1.0 [0.0–4.0], p < 0.01). Importantly, this was also true for those without prior surgery. The interobserver correlation coefficient was good (R = 0.86, p < 0.001). There was a significant correlation between the KOS and the Global Osteitis Score (R = 0.93, p < 0.001).


The KOS is a simple, easy, and reproducible scale in assessing osteitic bones in patients with CRS and can predict measures of severity in eosinophilic rhinosinusitis.