Slice thickness in the assessment of medial and lateral tibial cartilage volume and accuracy for the measurement of change in a longitudinal study.

Objective: The optimal magnetic resonance image (MRI) slice thickness required to assess cartilage volume accurately and efficiently in cross-sectional and longitudinal studies is unknown. We compared cartilage volume measured from MRI of the knees using different slice thicknesses (1.5 to 7.5 mm) and assessed longitudinal change.

Methods: A total of 123 subjects with osteoarthritis had baseline and followup MRI on their symptomatic knee at 2 years. Medial and lateral tibial cartilage volumes were calculated using increasing slice thickness by extracting each second, third, fourth, or fifth slice area to calculate total volume, which was compared to the "gold standard" volume calculated from the original 1.5 mm slices.

Results: There was little difference in the average medial and lateral tibial cartilage volume observed as the slice thickness increased from 1.5 to 7.5 mm; medial tibial cartilage volume ranged from 1750 microl to 1787 microl and lateral tibial cartilage volume ranged from 1949 microl to 2007 microl. There was also little absolute difference in the average change in medial and lateral tibial cartilage volume measured over 2 years. However, with increasing slice thickness, there was a decreased correlation between the tibial cartilage volume change calculated from the increased slice thickness, with the lowest correlation being 0.77 (p < 0.001) when the lateral cartilage volume calculated from the 7.5 mm slice was compared to the 1.5 mm slices.

Conclusion: Increasing slice thickness may provide sufficiently accurate measurement of tibial cartilage volume and change over time in some studies. This would result in reduction in MRI scanning and postimaging processing time, which has the potential of increasing the feasibility of this technique.