Background Several studies have indicated that High Resolution peripheral CT (HR-pQCT) scanning is more sensitive than radiography in detecting cortical breaks in destructive joint diseases like rheumatoid arthritis (RA)[1–3]. Cortical breaks are also seen in healthy controls, but the exact nature of these breaks is not known, and might represent vascular channels (VCs). No previous study has compared histology to HR-pQCT images in finger joints. We hypothesized that VCs seen on histology can also be detected by HR-pQCT imaging.
Objectives To identify histologically defined VCs in cadaveric hand joints on HR-pQCT imaging.
Methods Based on HR-pQCT, three regions in metacarpophalangeal joints from female cadavers with an unknown medical history (mean age 84.7, SD 5.5 years) were selected. These regions were extracted, embedded undecalcified in methylmetacrylate and histologically sectioned (thickness 15μm) parallel to the axial plane. Every second section (n=450) was stained with Goldner Trichrome. VCs were identified as a cortical break in one histological section which contained one or more vessels. HR-pQCT images (thickness 82μm) were independently scored by two trained readers for the presence of cortical breaks and if applicable categorized as VC. A break on HR-pQCT was defined as an interruption of the cortex seen on 2 consecutive slices in at least 2 orthogonal planes. A VC was defined as a break that is linear in shape with parallel lining. Finally, the histological sections were matched visually to corresponding axial HR-pQCT images.
Results A total of 56 VCs were identified on histology. On HR-pQCT 20 breaks were identified, of which 7 were categorized as VC. Only 3 VCs matched with VCs on histology. Of the remaining 53 histologically identified VCs, 34 could be detected on HR-pQCT, but the interruption of the cortex was not seen on 2 consecutive slices therefore they were not classified as a break. Eight histologically VCs fulfilled the definition of a break on HR-pQCT, but were not categorized as a VC. Eleven histologically VCs could not be identified on HR-pQCT. Figure 1 demonstrates histology and HR-pQCT images.
Conclusions VCs were frequently seen on histology. Only a minority of histologically defined VCs is interpreted as VC using a pre-specified definition on HR-pQCT images. Small histological VCs were often identified as an interruption but rarely considered a break according to the current definition of a VC on HR-pQCT. Therefore, additional criteria in order to diagnose the presence of VCs on HR-pQCT are warranted.
Stach CM, A&R.2010 Feb; 62(2):330–339.
Srikhum W, JRheum.2013 Apr; 40(4):408–416.
Fouque-Aubert A, ARD.2010 Sep; 69(9):1671–1676.
Disclosure of Interest A. Scharmga: None declared, K. Keller: None declared, M. Peters: None declared, A. van Tubergen: None declared, J. van den Bergh: None declared, B. van Rietbergen Consultant for: Scanco Medical AG, R. Weijers: None declared, D. Loeffen: None declared, E. M. Hauge: None declared, P. Geusens: None declared