18F-FDG-PET/CT showed 75 nodal stations involved and detected 10 extranodal lesions.
WB-MRI/DWIBS showed 80 nodal stations involved and detected 12 extranodal lesions,
revealing an overstaging in 2/25 (8%) patients and an understaging in 1/25 (4%).
A statistical evaluation with the Cohen k was performed to assess the degree of correlation between the two methods,
for each nodal stations and extranodal sites (see data in the table).
In 10 \ 25 patients we found differences in the attribution of disease in each of the stations described.
Of these 10 patients,
5 had a histological diagnosis of HL,
4 of NHL and 1 was a LLC.
The disagreements in total were 21: 4 for the cervical stations,
2 for the mediastinum,
5 for abdominal stations ,
3 for the pelvic stations,
3 for the femoral ones and finally 4 discrepancies were observed for extra-nodal localizations.
In 6 \ 21 disagreements observed,
18FDG-PET/CT was positive for lymph nodes with a diameter less than 1 cm,
with negative Whole Body-MRI/DWIBS (1 cervical,
1 mediastinal,
2 abdominal,
1 pelvic and 1 femoral).
In 2 \ 21 discrepancies 18FDG-PET/CT was positive,
with negative MRI,
respectively,
for a cervical lymph node localization with a diameter greater than 1 cm and a bone marrow localization.
In 13 \ 21 disagreements described MRI was positive,
with negative 18FDG-PET/CT ,
for nodal and extra-nodal localizations (2 cervical,
1 mediastinal,
3 abdominal,
2 pelvic,
2 femoral and 3 extranodal).
Then ,the two methods were concordant in the staging in 22/25 cases (88%).
The agreement between WB-MRI/DWIBS and 18F-FDG-PET/CT was moderate to good in evaluation of nodal involvement and staging,
in according to Ann Arbor criteria.