Thesis

158 Part II Chapter 6 statistical inference was made based on parametric tests. In case of severe deviation of (log)normality, defined as 0 not being within the 95%-confidence interval of both distribution metrics, parameters are described as median and IQR and non-parametric tests were used. Correlations are presented as R2, describing the proportion of the variance in the dependent variable that is predictable from the independent variable(s). As this R2 loses the sign of the correlation, it was added for completion (i.e. R2=R*R for R≥0 and R2=-R*R for R<0). Interobserver agreement was computed using the intraclass correlation coefficient (ICC), alpha two-way random model on absolute agreement and interpreted according to Koo, et al. [24]. Furthermore, agreement was graphically described using Bland-Altman plots including systematic error (mean absolute difference) and random error (95%-limits of agreement), which also visualizes heteroscedasticity. All statistical analysis was performed using SPSS software (version 23, IBM, New York, New York, USA). A type I error (p-value) below 0.05 was considered statistically significant. In case of multiple comparisons this p-value was adjusted for the number of parallel tests (n) by pcritical=0.05/n (i.e. Bonferroni correction). Results Twenty patients met our eligibility criteria. Mean age was 46 years (range 16 to 68), 14 were female. Sixteen patients suffered from PFD of whom four were diagnosed with McCune-Albright syndrome (MAS). All MAS patients suffered precocious puberty in the past, two suffered from hyperthyroidism, one also from growth hormone excess, but none suffered an untreated endocrinopathy during the time of Na[18F]F-PET/CT). The remaining four patients were diagnosed with MFD. All Na[18F]F-PET/CT scans were made before start of denosumab treatment or planned surgery. At the time of the Na[18F]F-PET/CT, eighteen used bisphosphonates. Patient characteristics are displayed in table 1.

RkJQdWJsaXNoZXIy MjY0ODMw