Accuracy of CT Imaging for Diagnosis of Acute Invasive Fungal Rhinosinusitis in Children.
In a retrospective cohort of 106 at-risk children, including 40 diagnosed with acute invasive fungal rhinosinusitis, sinus CT showed 83% sensitivity, 97% specificity, and an AUC of 0.90, but missed seven cases, often involving the middle turbinate.
Open original publication →What the AI sees
In a retrospective cohort of 106 at-risk children, including 40 diagnosed with acute invasive fungal rhinosinusitis, sinus CT showed 83% sensitivity, 97% specificity, and an AUC of 0.90, but missed seven cases, often involving the middle turbinate.
Research significance
The evidence supports CT as a clinically useful but insufficiently sensitive diagnostic tool; by inference, maintaining nasal endoscopy when suspicion remains despite a negative CT could reduce missed or delayed AIFRS diagnosis in immunocompromised children, although improved treatment outcomes were not evaluated.
Source abstract
PURPOSE: The purpose of this research was to determine the diagnostic accuracy of CT imaging of the paranasal sinuses for the diagnosis of acute invasive fungal rhinosinusitis (AIFRS) in children. METHODS: A retrospective review of sinus CTs performed in patients at risk for AIFRS was performed. A multi-anatomic site evaluation of CT imaging was performed with attention to soft tissue and bone abnormalities. Diagnostic accuracy of sinus CT findings was calculated for diagnosis of AIFRS with gold standard obtained from surgical biopsy and/or clinical follow-up. RESULTS: A total of 106 pediatric patients were included with a mean age was 9.4 years. The most common clinical risk factor for AIFRS was leukemia. A total of 40 children were diagnosed with AIFRS. The nasal turbinates and nasal septum were the most common sites of disease. Presence of facial symptoms, duration of ANC < 500, and duration of ANC < 200 were significantly different in AIFRS (+) versus AIFRS (-) patients (p<0.001; p=0.02; p=0.01). The sensitivity, specificity, and AUC for diagnosis of AIFRS from one or more abnormal CT findings were 83%, 97%, and 0.9 respectively. A total of seven patients (17.5%) with AIFRS were not detected by CT imaging, and the most common site of missed AIFRS was at the middle turbinate. CONCLUSION: CT imaging of the sinuses demonstrates moderate to high level of overall diagnostic accuracy for AIFRS, but with greater specificity than sensitivity. AIFRS in the nasal cavity was not identified by CT in a substantial number of children such that nasal endoscopy remains the diagnostic gold standard in the setting of a negative sinus CT when the clinical concern for AIFRS remains.