RATIONALE AND OBJECTIVES: To determine the volume of articular cartilage in cadavers, patients, and healthy volunteers by using a volumetric, fat-suppressed spoiled gradient-recalled signal acquisition in the steady state (SPGR) magnetic resonance (MR) sequence. METHODS: Sagittal MR images were obtained with a fat-suppressed SPGR sequence (repetition time, 52 msec; echo time, 10 msec; 60 degrees flip angle; 3.0-3.5-mm partitions, 256 x 192 matrix, two signals acquired). The cartilaginous surfaces of the tibia, femur, and patella were planimetrically defined with a three-dimensional workstation. A three-dimensional model volume was created by threshold segmenting the cartilage from the adjacent tissues. The volume as calculated by using MR imaging was compared with the actual volume of the cartilage specimens. RESULTS: Observed measurements correlated with actual weight and volume displacement measurements with an accuracy of 82%-99% and linear correlation coefficients of 0.99 (P = 2.5e-15) and 0.99 (P = 4.4e-15). Precision of segmentation in healthy volunteers yielded a coefficient of variation of 0.4% for interobserver variability and 0.3% for intraobserver variability. CONCLUSION: This pilot study suggests that accurate volumetric calculations of knee articular cartilage are possible with currently available MR imaging pulse sequences and a commercially available work station.
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging