SUMMARY: Computed tomography-based depth-specific image processing is able to precisely identify regional differences between healthy patellae and patellae with osteoarthritis. INTRODUCTION: This study aims to assess the precision errors and potential differences in regional, depth-specific subchondral bone mineral density (BMD) in normal and osteoarthritic (OA) human patellae in vivo using CT-based density analyses. METHODS: Fourteen participants (2 men and 12 women; mean age, 51.4; SD, 11.8 years) were scanned using clinical quantitative CT (QCT) three times over 2 days. Participants were categorized as either normal (n = 7) or exhibiting radiographic OA (n = 7). Average subchondral BMD was assessed at three depths relative to the subchondral surface. Regional BMD analysis included: total lateral facet BMD, total medial facet BMD, and superior/middle/inferior BMD of lateral and medial facets at normalized depths of 0-2.5, 2.5-5, and 5-7.5 mm from the subchondral surface. We assessed precision using root mean square coefficients of variation (CV%). We evaluated differences between OA and normal BMD by (1) calculating percentage differences between the groups (in relation to normal BMD) (2) relating percentage differences to respective CV% errors and (3) determining effect sizes using Cohen's d. RESULTS: Root mean square CV% precision errors ranged from 1.1 to 5.9 %. Percentage differences between OA and normal BMD varied from -1.6 to -30.1 % (BMD lower in OA patellae). In relation to precision errors, percentage differences were, on average, 5.5× greater than CV% errors. Cohen's d effect sizes ranged from -1.7 to -0.1. Largest differences were noted at depths of 2.5-5 and 5-7.5 mm from the subchondral surface. CONCLUSIONS: Patellar subchondral BMD measures were precise (average CV%, ≤3 %). This region- and depth-specific CT-based imaging tool characterized regional standardized BMD differences between normal and OA patellae in vivo.
SUMMARY: Computed tomography-based depth-specific image processing is able to precisely identify regional differences between healthy patellae and patellae with osteoarthritis. INTRODUCTION: This study aims to assess the precision errors and potential differences in regional, depth-specific subchondral bone mineral density (BMD) in normal and osteoarthritic (OA) human patellae in vivo using CT-based density analyses. METHODS: Fourteen participants (2 men and 12 women; mean age, 51.4; SD, 11.8 years) were scanned using clinical quantitative CT (QCT) three times over 2 days. Participants were categorized as either normal (n = 7) or exhibiting radiographic OA (n = 7). Average subchondral BMD was assessed at three depths relative to the subchondral surface. Regional BMD analysis included: total lateral facet BMD, total medial facet BMD, and superior/middle/inferior BMD of lateral and medial facets at normalized depths of 0-2.5, 2.5-5, and 5-7.5 mm from the subchondral surface. We assessed precision using root mean square coefficients of variation (CV%). We evaluated differences between OA and normal BMD by (1) calculating percentage differences between the groups (in relation to normal BMD) (2) relating percentage differences to respective CV% errors and (3) determining effect sizes using Cohen's d. RESULTS: Root mean square CV% precision errors ranged from 1.1 to 5.9 %. Percentage differences between OA and normal BMD varied from -1.6 to -30.1 % (BMD lower in OA patellae). In relation to precision errors, percentage differences were, on average, 5.5× greater than CV% errors. Cohen's d effect sizes ranged from -1.7 to -0.1. Largest differences were noted at depths of 2.5-5 and 5-7.5 mm from the subchondral surface. CONCLUSIONS: Patellar subchondral BMD measures were precise (average CV%, ≤3 %). This region- and depth-specific CT-based imaging tool characterized regional standardized BMD differences between normal and OA patellae in vivo.
Authors: W Burnett; S Kontulainen; C McLennan; D Hazel; C Talmo; D Hunter; D Wilson; J Johnston Journal: J Musculoskelet Neuronal Interact Date: 2016-03 Impact factor: 2.041