Design: Prospective single-centre cohort
Population: 154 patients with chronic rhinosinusitis with nasal polyps in Vietnam
Primary outcome: Discrimination of eosinophilic endotype by CT scores
Effect: Osteitis score AUC 0.777 vs Lund-Mackay 0.661; threshold ≥1: sensitivity 69.6%, specificity 83.3%

At a tertiary hospital in Ho Chi Minh City, 154 patients with chronic rhinosinusitis with nasal polyps had preoperative CT scored for osteitis (Kennedy Osteitis Score) and mucosal disease (Lund-Mackay), and tissue eosinophils counted after surgery (Otolaryngology-Head and Neck Surgery, 24 September 2026). Eosinophilic disease was defined as 10 or more eosinophils per high-power field.

Osteitis severity correlated with tissue eosinophil density (rs 0.45) and remained independently associated with it in adjusted analysis. Osteitis score discriminated eosinophilic disease better than Lund-Mackay (AUC 0.777 vs 0.661). A score of 1 or more gave 69.6% sensitivity and 83.3% specificity.

Eosinophilic polyps recur more and respond differently to steroids and biologics, so knowing the endotype before surgery helps planning and counselling. Osteitis on a CT already performed is a free clue, though not a substitute for histology.

Look for osteitis on preoperative sinus CT and record it, not just the Lund-Mackay score.
Treat osteitis as a pointer towards eosinophilic disease when planning and counselling.
Confirm the endotype with tissue eosinophil counts after surgery.
Consider closer follow-up for recurrence where osteitis and eosinophilia coexist.

Why it matters

It turns a finding often noted in passing into an early pointer to a harder-to-treat endotype.

The statistics, in plain English

An AUC of 0.78 is fair-to-good discrimination: useful as a pointer, not as a diagnosis. Sensitivity of 70% means about 3 in 10 eosinophilic cases had no osteitis.