2018 Section 5 - Rhinology and Allergic Disorders

Otolaryngology–Head and Neck Surgery 150(2)

Figure 2. Two-dimensional (2D) coronal osteomeatal complex (OMC) scoring system. The 2D coronal OMC scoring system uses a volu- metric segmentation of coronal computed tomography image standardized by review of the coronal cut through the OMC.

Pre-Treatment Comparison

Post-Treatment Comparison

0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

0.915

0.904

0.864

0.837

0.824

0.803

0.767

0.768

0.586

0.58

0.56

0.553

3D vs 2D OMC

3D vs Zinreich

Lund-Mackay vsZinreich

3Dvs Lund- Mackay

2DOMC vs Zinreich

2DOMC vs Lund-Mackay

3D vs 2D OMC

3D vs Zinreich

Lund-Mackay vsZinreich

3Dvs Lund- Mackay

2DOMC vs Zinreich

2DOMC vs Lund-Mackay

Upper CI

CCC Lower CI

Upper CI

CCC Lower CI

Figure 4. Posttreatment comparison of computed tomography staging systems.

Figure 3. Pretreatment comparison of computed tomography sta- ging systems.

Discussion We evaluated disease severity and response to treatment intervention of CRS patients’ CT scans with the most popular and user-friendly method currently used (Lund-Mackay) and a variation of this method created to better subgrade diseases and thus increase sensitivity (Zinreich) against 2 novel objec- tive computerized methods (3D volumetric and 2D coronal OMC) to determine how well results of these systems corre- lated in determining severity of disease and response to treat- ment. There is little doubt that the 3D volumetric scoring method, which is able to yield a numerical score of total dis- ease percentage by complete volumetric measurement of all sinuses, is the most accurate and objective scoring tool yet created. By setting the 3D volumetric scoring system as the standard against which to compare the other systems, we can determine how well results of these other methods depict dis- ease status and sensitivity to change. All methods tested were found to be replicable and with satisfactory intersystem agreement when compared with the 3D method on pretreatment scoring. However, there was a

Zinreich and Lund-Mackay systems had the third highest correlation (0.864). In general, confidence intervals increased with decreasing CCC. Comparison of posttreatment correlation ( Figure 4 ) revealed the Zinreich modification to have the best correlation to the 3D volumetric Scoring system (0.837) ahead of the 2D OMC method (0.767). This may have been secondary to the 2D OMC method’s being less inclusive. The remaining com- parisons showed only moderate correlation and increased con- fidence intervals between values. Comparison of the difference between pre- and posttreat- ment results ( Figure 5 ) demonstrated that both the 2D cor- onal OMC (0.824) and Zinreich (0.778) systems had strong agreement with the 3D volumetric scoring system and were very sensitive to change, whereas the Lund-Mackay method (0.545) had only moderate agreement. The 2D OMC and Zinreich methods also have strong correlation with each other (0.788). The mean, range, and standard deviation for each method’s results on both pretreatment and posttreat- ment analysis can be found in Table 3 .

23