Metastatic LAD ASHNR Final

Slide 1: 

Application of axial-based size criteria to coronal & sagittal planes to detect metastatic lymphadenopathy Eric S Bartlett MPH MD Thomas D Walters MBBS FRACP Eugene Yu MD University of Toronto Princess Margaret Hospital Toronto, Ontario Canada

Slide 2: 

Does classification of pathologically enlarged lymph nodes differ between imaging planes? Aim: 2

Slide 3: 

Methods: Retrospective, May 2007 – Nov. 2007 Single centre Pre-treatment staging head/neck CTs Mucosal-based SCCa and lymphadenopathy No previous treatment for head/neck cancer 3

Slide 4: 

Methods: RP = 0.8 cm JD = 1.5 cm 1b = 1.5 cm All others = 1.0 cm 2 head and neck radiologists Axial size criteria (longest dimension)*: Initial LN survey in the axial plane *Som PM, Curtin HD, Mancuso AA, “Imaging-based nodal classification for evaluation of neck metastatic adenopathy”. AJR:174, March 2000; 837-45. 4 Apply the axial-based size criteria to the coronal and sagittal planes

Slide 5: 

Methods: Axial: Selection of largest LN ≥ size criteria Measure LN in coronal and sagittal planes Separate review in coronal and sagittal planes: Selection of largest LN ≥ size criteria If selected node ≠ axial selection, > 1 node was allowed per level, measured in all 3 planes 5

Slide 6: 

Methods: Consensus conference, Independent measures Interclass Correlation Coefficient (ICC) Nodes classified as ‘enlarged’ or ‘normal’ Data for each plane were compared using: ICC Kappa statistic McNemar test Morphology: Normal Elongated Round / Suspicious 6

Slide 7: 

Results: n = 222 lymph nodes in 50 patients Median age = 59 yrs (IQR: 50.6 to 69.6 yrs) 70% male 7

Slide 8: 

Inter-observer agreement: Axial H/N radiologist 2 H/N radiologist 1 ICC = 0.98 (0.98-0.99) 8 H/N radiologist 2 H/N radiologist 1 ICC = 0.93 (0.91-0.95) Sagittal H/N radiologist 1 H/N radiologist 2 ICC=0.79 (95% CI=0.75, 0.83) ICC = 0.99 (0.98-0.99) Coronal

Slide 9: 

Classification: Sagittal Coronal Axial Normal Normal Normal Enlarged Enlarged Enlarged Lymph Nodes 66.7% 90.5% 88.3% 148 / 222 201 / 222 196 / 222 9

Slide 10: 

All Cor + Sag Ax + Sag (8) Ax + Cor (11) Ax only (11) Cor only (4) Sag only (2) 118 68 n = 222 Classification: 10

Slide 11: 

Results: 11

Slide 12: 

Results--Morphology: 12 Round / Suspicious Elongated Normal 27% (3) 27% 46% - 48% (101) 28% 24% + n 211 105 11 62 55 222 Coronal / Sagittal Round / Suspicious Elongated Normal 4% (3) 60% 36% - 69% (102) 12% 19% + n 148 105 74 62 55 222 Axial P<0.05 P<0.05

Slide 13: 

Results--Morphology: 13 27% (3) 27% 46% - 11 105 62 55 222 Round / Suspicious Elongated Normal 4% (3) 60% 36% + 72% (99) 11% 17% + n 137 74 Axial + - + Cor/Sag P<0.05

Slide 14: 

Nodes considered normal in the axial plane are frequently misclassified as enlarged when the axial-based size criteria are applied to the coronal and sagittal planes Lymph node morphology must be considered before classifying a node as metastatic, regardless of size Evaluation of lymph nodes in all 3 planes is helpful to determine morphology Conclusions: 14