Desley van Zoggel

PET/CT response after induction chemotherapy 131 CHAPTER 7 Inter-reader variability The level of agreement between the response assessment by the blinded dedicated nuclear medicine specialist and the response assessment by a second specialist was moderate. The weighted kappa value was 0.71 (95 percent CI 0.60–0.83, P < 0.001). Correlation with MRI response assessment The major histopathological response was correctly identified by PET/CT and MRI in 15 and 17 out of 36 cases, respectively. The evaluation overlapped in 11 cases. A partial histopathological response was correctly identified by PET/CT and MRI in 23 and 15 out of 43 cases, respectively. The evaluation overlapped in 10 cases. A poor histopathological response was correctly identified by PET/CT and MRI in 10 and 20 out of 26 cases, respectively. The evaluation overlapped in nine cases (Figure 5). The weighted kappa value for a correlation between the two tests was 0.38 (95 percent CI 0.23–0.52), P <0.001. Therewas a significant correlation and linear relationship between the MRI response and the likelihood of R0 resection (P = 0.041). Figure 5. Correlation between PET/CT and MRI response assessment Mandard 1-2 Mandard 3 Mandard 4-5 36 total 43 total 26 total PET 4 MRI 6 11 PET 13 MRI 5 10 PET 1 MRI 11 9

RkJQdWJsaXNoZXIy MTk4NDMw