- Design
- Prospective single-centre cohort
- Population
- 154 patients with chronic rhinosinusitis with nasal polyps
- Primary outcome
- Discrimination of eosinophilic endotype (10 or more eosinophils per high-power field) by CT osteitis score
- Effect
- AUC 0.777 versus 0.661 for Lund-Mackay; score of 1 or more: sensitivity 69.6%, specificity 83.3%
In a prospective cohort at a Vietnamese tertiary hospital, 154 patients with chronic rhinosinusitis with nasal polyps had CT osteitis graded with the Kennedy score. Tissue eosinophils were counted, and eosinophilic disease was defined as 10 or more per high-power field.
Osteitis severity correlated with tissue eosinophil density (Spearman 0.45) and with the Lund-Mackay score (0.42). For identifying eosinophilic disease, the osteitis score had an AUC of 0.777, better than the Lund-Mackay score at 0.661. A threshold of 1 or more gave 69.6% sensitivity and 83.3% specificity.
The authors suggest osteitis as a noninvasive marker for endotype stratification before surgery. An AUC near 0.78 is moderate, the cut-off was derived in the same cohort, and a single centre may not match patients elsewhere. Biopsy remains the reference for endotype.
- Osteitis severity correlated moderately with tissue eosinophils.
- AUC was 0.777 for osteitis versus 0.661 for Lund-Mackay score.
- A score of 1 or more gave 69.6% sensitivity and 83.3% specificity.
- Cut-off came from the same cohort, so it needs validation.
- Look at bone changes on the CT, not only mucosal disease, when planning surgery.
Why it matters
A free feature of an existing scan could help predict who has the endotype that tends to recur.
Don't overread it
A single-centre prospective cohort shows a moderate association and does not validate a decision rule.
The statistics, in plain English
An AUC of 0.78 means that, picking one eosinophilic and one non-eosinophilic patient at random, the score is higher in the eosinophilic one about 78% of the time. It is useful but far from definitive, and a cut-off tuned in the same data usually looks better than it will elsewhere.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for ent & head and neck surgery, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free