Boris Buerke MD, Presenter: Nothing to Disclose

Christoph Schuelke, Abstract Co-Author: Nothing to Disclose

Anna Knauer, Abstract Co-Author: Nothing to Disclose

Anno Graser MD, Abstract Co-Author: Speakers Bureau, Siemens AG

Daniel dos Santos, Abstract Co-Author: Nothing to Disclose

Michael Fabel, Abstract Co-Author: Nothing to Disclose

Walter Leonhard Heindel MD, Abstract Co-Author: Nothing to Disclose

Johannes Wessling MD, Abstract Co-Author: Nothing to Disclose

Multicenter study for comparison of manual linear measurements with semi-automatic derived 3D measurements for therapy response evaluation of malignant lymphoma in CT follow up examinations.

MSCT data sets of 57 patients with malignant lymphoma were evaluated before (baseline) and after two cycles of chemotherapy (follow up) by two radiologists independently in radiological centers of four university hospitals. In 290 target lymph nodes (46cervical,122 thoracic und 120abdominal) the long-axis-.diameter(LAD), the short-axis-diameter(SAD) and the WHO-square were measured manally as well as semi-automatically using a dedicated software.Lymph node volumetry was performed semi-automatically only. Time expenditure for manual and semi-automatic lymph node analysis was documented digitally. Therapy response evaluation was performed based on each parameter separately, according to lymphoma-adapted RECIST criteria.Statistical analysis encompassed t-test and McNemar´s test.

Based on measurements of the single lymph node, semi-automatic volumetry and WHO-square were significantly superior to manual parameters for therapy response evaluation in all centers(e.g. volume 87.6%vs. manual SAD 47.7%,p<0.001).Evaluation of therapy response based on lymph node target groups per patient according to RECIST 1.1 a higher number of correct classifications was derived using semi-automatic parameters independently of the center, e.g. 91.7%/89.3% (WHO-square/volume) in comparison to manual measurements, e.g. 81.5/79.2%(LAD/SAD). In average, misclassifications were reduced up to 10.8 %. Evaluation of therapy response based on manual metric parameters showed significant deviations between the different radiological centers. Time expenditure was comparable between semi-automatic (30.1±18.1s)and manual measurements (30.4±14.6s) over all centers.

Semi-automatic derived parameters (i.e. volume and WHO-square) significantly reduce the number of misclassified patients in therapy response evaluation of malignant lymphoma in CT in comparison to manual 2D measurements. These results are of interest especially with regard to the increasing mobility of oncological patients between different hospitals and for quality management of multi-centric oncological studies.

Semi-automatic  volume and WHO-square of lymph nodes significantly reduce misclassifications of therapy response evaluation of malignant lymphoma.

Buerke, B, Schuelke, C, Knauer, A, Graser, A, dos Santos, D, Fabel, M, Heindel, W, Wessling, J, Malignant Lymphoma in Multi-Slice CT: Should Manual 2D Measurements be Replaced by 3D Measurements for Therapy Response Evaluation? Results of a Multicenter Study.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12024816.html