Bhattacharyya N, Jones DT, Hill M, Shapiro NL. Arch Otolaryngol Head Neck Surg. 2004;130:1029–1032
Purpose of the Study.
To determine the accuracy of computed tomography (CT) in the diagnosis of pediatric chronic rhinosinusitis (CRS).
The sinus CT scans of 2 cohorts of children were evaluated and compared prospectively. The “diseased” cohort consisted of 66 children (mean age: 8 years) who were scheduled to undergo endoscopic sinus surgery for medically refractory CRS. The second “nondiseased,” control cohort consisted of 192 children (mean age: 9 years) who were undergoing CT scans for reasons other than sinusitis.
Sinus CT scans were scored according to the Lund-MacKay system. The Lund-MacKay staging system scores each sinus (anterior ethmoid, posterior ethmoid, maxillary, frontal, and sphenoid sinuses) according to the following scale: 0, no opacification; 1, partial opacification; 2, complete opacification. The ostiomeatal complex is scored as 0 (not occluded) or 2 (occluded). The left and right sides are staged separately. The scores are summed so that the total Lund score may range from 0 to 24 for each patient. The authors adapted the Lund-Mackay staging system for young children by assigning a null value to undeveloped sinuses. The corresponding Lund score was then scaled up to range from 0 to 24 by scaling with the factor 12/n, where n represents a number of scoreable (pneumatized) sinuses. The Lund scores of the diseased and control groups were compared. The diagnostic accuracy of the CT scan in distinguishing diseased patients from control patients was established by using the receiver operating characteristic curve. Sensitivity, specificity, and predictive value analyses were also conducted. The authors calculated predictive values at different base-rate prevalence values.
The diseased group exhibited a mean Lund score of 10.4 (95% confidence interval [CI]: 9.2, 11.5), and the control group exhibited a mean Lund score of 2.8 (95% CI: 2.4, 3.2). The area under the curve for the receiver operating characteristic was 0.923 (P < .001), indicating excellent diagnostic accuracy for CT scans. Adopting a Lund score cutoff of 5 to represent true disease, the CT scan demonstrated a sensitivity and specificity of 86% and 85%, respectively. Lund scores of ≤2 have an excellent negative predictive value, whereas Lund scores of ≥5 have an excellent positive predictive value. The authors demonstrated a decline in diagnostic accuracy of the CT scan with decreasing base-rate prevalence of the disease by calculating positive and negative predictive values at base-rate prevalences of 0.2, 0.5, and 0.8.
The sinus CT scan demonstrates excellent diagnostic accuracy for the diagnosis of pediatric CRS, with excellent sensitivity and specificity. The predictive value depends substantially on the base-rate prevalence of CRS in the population being evaluated. The authors established Lund score ranges for CT scans of children with sinusitis: 0 to 2, normal; ≥5, positive for sinusitis; 3 to 4, equivocal.
Although CT is considered the gold standard for diagnosis of CRS, its sensitivity, specificity, and diagnostic accuracy have not been well established in children. This study uses quantitative Lund scores to differentiate pediatric patients with and without CRS based on radiographic criteria. Based on these data and analysis, we can use the CT scan to discriminate between children with and without CRS. Nevertheless, the positive and negative predictive values of this test are substantially dependent on the prevalence of CRS, and this must be factored into clinical decision-making. This study highlights the fact that CRS is primarily a clinical diagnosis, and both the decision to perform a sinus CT and the interpretation of the scan should include this clinical context.