TRANSIENT BONE MARROW EDEMA SYNDROME AND SUBCHONDRAL INSUFFICIENCY FRACTURE

Quantitative Assessment of Bone Marrow Edemas (BME) of the Knee: Comparison of Two Methods
M Breitenseher¹, M Mayerhoefer¹, J Kramer², N Aigner³, S Hofmann⁴.
¹Department of Radiology, University of Vienna, Vienna, Austria, ²Institute of CT and MRI Diagnostics Schillerpark, Linz, Austria, ³First Orthopaedic Department, Orthopaedic Hospital Vienna-Speising, Vienna, Austria, and ⁴Department of Orthopaedics, LKH Stolzalpe, Stolzalpe, Austria

Purpose: To compare two methods for the quantification of BME bone marrow edemas of the knee.

Materials and Methods: 41 patients with BME of the knee due to osteonecrosis, osteoarthritis, bone bruise or stress were included in the study. Coronal STIR images of the affected knees were obtained using either a 1.0 Tesla or a 1.5 Tesla MR scanner. To monitor the BME's progression, every knee was examined twice at a 3-month interval. Size and signal intensity of BME were semi-quantitatively assessed in consensus by two radiologists and two orthopaedic surgeons. Independently, size and signal intensity were calculated using a new computer assisted method based on grey value analysis and calculation of a threshold value between normal and edematous bone marrow. The results of the semi-quantitative method were correlated with those of the computer assisted method.

Results: The correlation coefficient was 0.89 ("strong") for the BME size and 0.72 ("medium to strong") for the BME signal intensity. For the progression of size and signal intensity, correlation coefficients of 0.78 ("strong") and 0.67 ("medium") were found.

Conclusion: Good correlation between semi-quantitative and quantitative methods of assessment of both size and intensity of BME was found. While the computer assisted method is time-consuming and reserved for scientific purposes, the semi-quantitative method is simple and offers to the experienced examiner a fast and simple means for BME recording in clinical practice.