Response to letter re: Carotid atherosclerotic plaques standardized uptake values: methodological issues on reproducibility and accuracy

Background

Dear Editor,

We would like to thank Dr. Siamak Sabour for his letter and comments relating to our recently published article [1]. In the manuscript, an investigation was conducted on whether or not carotid atherosclerotic plaque standardized uptake values (SUVs) are consistent and reproducible across software packages; therefore, the purpose of the analysis performed was to measure the reproducibility, rather than validity, of SUV measurements between two software packages (OsiriX MD® version 6.5.2, Pixmeo© SARL, Geneva, Switzerland and AquariusNet iNtuitionTM version 4.4.11, TeraRecon, Foster City, CA, USA) (Table 1).

Table 1 SUV measurements (mean and standard deviation) by location and software together with differences (mean and standard deviation)

Conclusions

We acknowledge that the p values reported in the manuscript previously submitted are dependent on the study sample size, and may not provide sufficient support of measurement reliability. Thus, we will now provide the intra-class coefficient (ICC) for the relevant variables (see Tables 2, 3, and 4) which was found to be supportive of our initial findings.

Table 2 ICC for SUVs mean with 95% confidence intervals

Table 3 ICC for SUVs max with 95% confidence intervals

Table 4 ICC for SUVs min with 95% confidence intervals

As expected, higher agreements (ICC) were found among SUV mean and maximum measurements. Effect size measurements also show that SUV max measurements were similar when compared (differences in mean values within the range: 0.02–0.05).

Once again we thank Dr. Sabour for his contribution to the important discussion around SUV measurements across software packages.