CT Findings and Significance in Chronic Rhinosinusitis
CT scan remains the gold standard for radiologic confirmation of CRS, with a Lund-Mackay score ≥5 providing excellent positive predictive value for true disease, though CT findings correlate poorly with symptom severity and should primarily guide surgical planning rather than medical management decisions. 1
Key CT Findings That Confirm CRS
CT demonstrates several characteristic findings that objectively document sinonasal inflammation:
- Mucosal thickening within the paranasal sinuses 1
- Sinus opacification (partial or complete) 1
- Polyps or retention cysts 1
- Sclerosis and thickening of sinus walls (particularly in chronic disease) 1
- Ostiomeatal complex obstruction 1
Diagnostic Performance and Scoring
The Lund-Mackay scoring system is the validated standard for quantifying CT findings, scoring each sinus (maxillary, anterior ethmoids, posterior ethmoids, frontal, sphenoid) as 0 (no opacification), 1 (partial), or 2 (complete), plus 0 or 2 for ostiomeatal complex, yielding a maximum score of 24 (12 per side). 1
Critical diagnostic thresholds:
- LMS ≤2: Excellent negative predictive value; essentially rules out CRS (sensitivity 94%, specificity 41%) 1
- LMS ≥5: Excellent positive predictive value for true disease (pediatric data: sensitivity 86%, specificity 85%) 1
- LMS 3-4: Clinically relevant disease warranting treatment consideration 1
When CT Is Essential vs. Optional
CT is essential after:
- Failure of appropriate medical treatment in secondary care with continued symptoms AND abnormal endoscopy 1
- Before functional endoscopic sinus surgery for surgical planning 1
- When complications are suspected (orbital, intracranial) - use CT with contrast or MRI 1
- When neoplasia or invasive fungal sinusitis is suspected 1
CT is NOT essential at:
- Initial presentation to ENT/secondary care, even with highly suggestive symptoms (consensus unclear among experts, with 52% rating it as moderately important and 41% rating it as not essential) 1
- Acute rhinosinusitis (diagnosis is clinical; CT is overused in this context) 1
Critical Pitfall: Poor Symptom Correlation
The most important caveat is that CT findings correlate weakly or not at all with symptom severity in most CRS patients. 1, 2
- No association exists between total SNOT-22 scores and Lund-Mackay scores in CRS without nasal polyps 1
- The only symptom that significantly correlates with CT scores is olfaction 1
- Facial pain shows inverse correlation with CT findings in CRS without polyps (higher pain scores = lower LMS, p=0.022) 1
- Some correlation exists in CRS with nasal polyps subgroup, but remains weak overall 1
This means CT severity does not predict quality of life impairment or guide medical treatment intensity - it primarily serves to confirm diagnosis and plan surgery. 1, 2
Surgical Planning Value
CT is critical for identifying anatomic variants and abnormalities that increase surgical risk:
- Cribriform plate dehiscence or asymmetry 1
- Lamina papyracea integrity (orbital wall) 1
- Anterior ethmoidal artery canal location 1
- Onodi cells and sphenoid sinus variants 1
- Carotid artery dehiscence 1
Low-dose CT protocols are inadequate for visualizing these surgically critical structures, particularly in patients with nasal polyps or prior surgery. 1
Contrast Enhancement
Contrast is NOT needed for:
Contrast IS needed for:
- Suspected orbital or intracranial complications 1
- Suspected neoplasia 1
- Suspected invasive fungal sinusitis 1
Alternative Imaging Modalities
- Cone beam CT: Accurate for CRS diagnosis with lower radiation, but limited soft-tissue evaluation; not preferred if extrasinus disease suspected 1
- MRI: Superior for differentiating inflammation from neoplasia; useful when complications suspected or radiation exposure is a concern 1, 3, 4
- Plain radiography: Inadequate; no longer recommended 1, 4
Stability of CT Findings Over Time
CT findings in untreated CRS patients remain relatively stable over short to middle time intervals (mean 338 days), with Lund-Mackay scores showing strong correlation (r=0.86) between serial scans regardless of initial disease severity. 5 This suggests repeating CT scans in symptomatic untreated patients should be seriously questioned unless clinical status has significantly changed. 5