CT perfusion measurements of head and neck carcinoma from single section with largest tumor dimensions or average of multiple sections: agreement between the two methods and effect on intra- and inter-observer agreement.

Purpose: To evaluate the agreement between quantitative CT perfusion measurements of head and neck squamous cell carcinoma (SCC) obtained from single section with maximal tumor dimension and from average values of multiple sections, and to compare intra- and inter-observer agreement of the two methods.

Methods: Perfusion was measured for 28 SCC cases using a region of interest (ROI) inserted in the single dynamic CT section showing maximal tumor dimension, then using average values of multiple ROIs inserted in all tumor-containing sections. Agreement between values of blood flow (BF), blood volume (BV), mean transit time (MTT) and permeability surface area product (PS) calculated by the two methods was assessed. Intra-observer agreement was assessed by comparing repeated calculations done by the same radiologist using both methods after 2 months blinding period. Perfusion measurements were done by another radiologist independently to assess inter-observer agreement of both methods.

Results: No significant differences were observed between the means of the 4 perfusion parameters calculated by both methods, all p values >0.05 The 95% limits of agreement between the two methods were (-33.9 to 43)ml/min/100g for BF, (-2.5 to 2.8)ml/100g for BV, (-4.9 to 3.9)s for MTT and (-17.5 to 18.6)ml/min/100g for PS. Narrower limits of agreement were obtained using average of multiple sections than with single section denoting improved intra- and inter-observer agreement.

Conclusion: Agreement between both methods is acceptable. Taking the average of multiple sections slightly improves intra- and inter-observer agreement.