*2.7. Statistical Analysis*

Statistical analysis was performed, using SPSS Statistics 22.0 (IBM Corp.). Accuracy was measured at the patient and lesional levels, separately for groups A and B. The difference between FDG PET parameters (SUVmax, SUVpeak, SUVmean and TLG) across the different categories in the two groups (true positives, false positives and false negatives, group A and B) was assessed using the Kruskal–Wallis test. The association between the tracer uptake in the primary tumor and accuracy of staging was assessed by using the Mann–Whitney U test. The association between histopathological features of the primary tumor and accuracy of staging was assessed by using a chi-square test or a Mann–Whitney U test. The association between semi-quantitative FDG PET parameters and histopathological features of the primary tumor was investigated, using a mixed model analysis with an intercept on patient level. Results were considered significant for a *p*-value of <0.05.
