Follow up of coiled intracranial aneurysms 105 To determine test characteristics for CEMRA the ratings of the aneurysm occlusion status were first dichotomized. R&R class 1 and 2 are regarded as adequately occluded, not requiring retreatment; class 3 is regarded as inadequately occluded, requiring retreatment. Calculations were repeated after dichotomization between complete aneurysm occlusion (R&R class 1) on the one hand, and incomplete occlusion (R&R class 2 and 3) on the other. Calculations were also performed using only the ratings of a single follow-up study, in patients who had undergone two CEMRA follow-ups. Sensitivity, specificity, negative and positive predictive values (NPV and PPV) were calculated with 95% confidence intervals (CI), using the DSA ratings as standard of reference. Analysis of the results of the comparison with initial occlusion status was done after dichotomization between “new or increased recanalization” (category 1) on the one hand, and “no change” and “disappeared or decreased remnant” (categories 2 and 3) on the other. Results Seventy-one patients were included in the study. There were 69 patients with one aneurysm, one patient with two aneurysms and one patient with three aneurysms. 14 patients were included two times, the first time at six months follow-up and the second time at 24 months follow-up; among these 14 were the two patients with multiple aneurysms, thus resulting in a total of 91 aneurysm assessments. There were no stent-assisted coilings. For four patients with four aneurysms the initial post-coiling DSA images were not available. Patient characteristics are given in Table 1. Complete occlusion of the aneurysm was found at follow-up DSA (the standard of reference) in 39 of 91 aneurysm assessments (42.9%), while a neck remnant was found in 30 (33.0%) and incomplete occlusion of the aneurysm lumen was found in 22 (24.2%) assessments. Compared with the initial post-coiling DSA study increased aneurysm remnant or new recanalization was found in 36 of 87 assessments (41.4%), occlusion was unchanged in 45 (51.7%) and the aneurysm remnant had decreased in 6 (6.9%) aneurysms at follow-up DSA. Confidence in scoring per aneurysm was very good: the first observer scored ‘good’ confidence in 89 aneurysm assessments, ‘moderate’ in 1 and ‘poor’ confidence in 1 assessment. For the second observer these figures were 79, ten and two, respectively. Linear weighted Kappa for interobserver agreement in assessing occlusion status was 0.57 (95% CI: 0.43 – 0.72). For data dichotomized between adequate occlusion (R&R 1 and 2) and inadequate occlusion (R&R 3) Kappa was 0.52 (95% CI: 0.35 – 0.69); if the data were dichotomized between complete (R&R 1) and incomplete occlusion (R&R 2 and 3), Kappa was 0.65 (95% CI: 0.49 – 0.82).
proefschrit Van Zwam inhoud met kaft en stellingen.indd
To see the actual publication please follow the link above