​​Joint space analysis is a measure of the distance between the proximal and distal bones of the MCP joints (Figure 1). A consensus method has been developed by the SPECTRA collaboration [1] [2] [3], and is proposed for universal use in ongoing and future clinical trials for rheumatology. is available for distribution from the manufacturer. Figure 1 - Joint space width analysis. The outcome measures of the joint space analysis include: joint space volume, joint space width, minimum joint space width, maximum joint space width, standard deviation of joint space width and the asymmetry of the joint space width, which is defined as the ratio of maximum and minimum joint space width.

Erosion analysis is a measure of the size and volume of breaks in the cortical bone (Figure 2). The following definition of an erosion observed in HR-pQCT scans has been developed by the SPECTRA collaboration [4]: Presence of a definite interruption in the cortical bone. The cortical break must extend over at least two consecutive slices. The cortical break must be detectable in two perpendicular planes. Loss of underlying trabecular bone at the cortical break. Nonlinear in shape (to differentiate from penetrating vascular channels). Figure 2 - Example of a cortical break meeting the SPECTRA definition of erosion by HR-pQCT imaging. A. Conventional radiography in the anterior-posterior view of the right hand of a 59-year-old female patient with rheumatoid arthritis diagnosed 6 months prior to imaging. B. Enlargement of the region of interest of the third MCP joint. C-E. HR-pQCT images of consecutive slices in the axial (C1 and C2), coronal (D1 and D2), and sagittal (E1 and E2) multiplanar reformations. The rhomboid orients to the radial side of the joint, and the arrow demonstrates the erosion location. MCP: metacarpophalangeal; HR-pQCT: high-resolution peripheral quantitative computed tomography. Methods for quantifying erosion volume Several methods for quantifying erosions have been proposed and are being investigated for use in clinical trials. These include semiquantitative scores, measures of maximal width, depth, surface area and volume. i. Manual measurements This method involves the manual measurement of the maximal width and depth of the erosion and approximate the erosion volume as a ellipsoidal shape [4]. The erosion volume can be calculated using a half-ellipsoid formula or by a full- ellipsoid formula [5]: ii. Volume by Osirix Dicom Viewer The volume is measured by outlining the erosion by a closed polygon on each axial 2D image with an erosion (Figure 3-A). Afterwards the volume is computed (Figure 3-B) [6] [7]. Figure 3 - Erosion volume measured using OSIRIX Medical imaging software. iii. Volume by Summarized Area Contours of the entire metacarpal head (A1) and the region not including the eroded bone (A2) are obtained. The erosion volume is then calculated using: V = (A1− A2) × n × T where n is the total number of images containing the erosion, and T is the image thickness [8] [9] [10]. iv. MIAF-Finger Medical Image Analysis Framework Software for the Finger (MIAF-Finger) allows for three-dimensional erosion segmentation to assess the volume and shape parameters of erosions using a level set segmentation technique. Erosions must be identified by an operator and manual corrections can be made using a series of tools embedded in the program [11] [12] [13]. The MIAF-Finger script are custom made by University of Erlangen-Nuremberg, Erlangen, Germany. The scripts are integrated into the Medical Image Analysis Framework. v. Modified Evaluation Script for Erosions The modified Evaluation Script for Erosions (mESE) is a semi-automated segmentation algorithm using thresholds of 250 mg/cm3 for cortical bone versus soft tissue and 50 mg/cm3 for the erosion versus the surrounding trabecular bone [13]. The script uses the manufacturer's Image Processing Language (IPL, Scanco Medical AG, Brüttisellen, Switzerland). The script is custom-made by University of Erlangen-Nuremberg, Erlangen, Germany and integrated in the Scanco system. vi. Cortical interruption detection algorithm This is a fully-automated algorithm that can quantify and visualize the volume of breaks in cortical bone without the need for operator erosion identification. Cortical interruptions are identified as a discontinuity of at least 3 consecutive slices, with a width of at least 3 voxels, and connected to both the endosteal and periosteal boundary. Once the cortical interruption has been identified, the trabecular void volume can be segmented [14] [15] [16]. Figure 4 - Steps incorporated in the automated cortical interruption detection algorithm and measurement of underlying loss of trabecular bone, as visualized on a 2D grayscale image. A) Detection of two cortical interruptions > 0.41 mm. B) Region of interest (ROI) identified by dilating the cortical interruptions by 48 voxels (corresponding to 3.936 mm), and masking with the periosteal contour. C) Only voids that are > 0.738 in diameter are selected by performing a distance transformation within the ROI. D) Voids are eroded by 2 voxels to detach connections of < 0.328 mm and prevent leakage into the trabecular bone. E) Inclusion of voids that remain connected to a cortical interruption after erosion. F) Dilation of voids to the original size, and inclusion of cortical interruptions that were originally detected [15]. The script uses the manufacturer's Image Processing Language (IPL, Scanco Medical AG, Brüttisellen, Switzerland). The script are custom made by Maastricht University Medical Centre, Maastricht, Netherlands. vii. Surface transformation algorithm The surface transformation algorithm quantifies surface deformity using a non-rigid generic reference surface produced from healthy subjects. This reference surface is then transformed to each patient surface where it can be used with the subject’s surface mesh to calculate the number, percentage surface area of erosion, and maximum positive and negative distance between the diseased surface and healthy reference [17]. Figure 5 - Surface deformation model(33) showing: (left) cross-section of MCP joint with the diseased surface (red) overlaid on the average healthy surface (blue). (Right) 3D heat map representing distances between diseased (top) and healthy (bottom) surfaces. Positive distances reflect bony proliferations, while negative distances represent erosive damage. Image provided courtesy of Worcester Polytechnic Institute (Kyle Murdock and Karen Troy). The script is custom made by the Worcester Polytechnic Institute, Worcester, Massachusetts, USA. Table 1 - Reported precision and reliability for different erosion analysis methods.

Density measurements of trabecular bone, and microarchitectural parameters such as the thickness and porosity of cortical bone can be evaluated from HR-pQCT scans. Recently, a joint working group between the International Osteoporosis Foundation, American Society of Bone and Mineral Research, and European Calcified Tissue Society have convened to produce the guidelines and recommendations to assess bone microarchitecture from HR-pQCT scans of the human radius. There is no similar set of recommendations for the MCP joints, which is required for standardization for multicenter clinical trials. This is key focus for future activities of the SPECTRA Collaboration.

There currently exists no consensus on how to define and analyze osteophytes, enthesophytes or bone spurs with HR-pQCT. Osteophytes are most commonly defined as: bony protrusions from the juxtaarticular cortical shell, bony proliferations at specific anatomical sites or bone formation arising from the periosteal bone cortex at the insertion sites of the capsule, ligament, or tendons or at the location of functional enthesis. Osteophyte size can either be graded semi quantitative according to the height which are measured between the highest surface of the lesion and the original surface of the cortical bone. The size has also been presented as height, or volume, which are manually segmented.

[1] Stok KS, Burghardt AJ, Boutroy S, Peters MPH, Manske SL, Stadelmann VA, Vilayphiou N, van den Bergh J, Geusens P, Li X, et al. Consensus approach for 3D joint space width of metacarpophalangeal joints of rheumatoid arthritis patients using high-resolution peripheral quantitative computed tomography. Quant Imaging Med Surg (2020). doi:10.21037/qims.2019.12.11 [2] Tom S, Frayne M, Manske SL, Burghardt AJ, Stok KS, Boyd SK, Barnabe C. Determining Metacarpophalangeal Flexion Angle Tolerance for Reliable Volumetric Joint Space Measurements by High-resolution Peripheral Quantitative Computed Tomography. J Rheumatol (2016). 43:1941–1944. doi:10.3899/jrheum.160649 [3] Burghardt AJ, Lee CH, Kuo D, Majumdar S, Imboden JB, Link TM, Li X. Quantitative in vivo HR-pQCT imaging of 3D wrist and metacarpophalangeal joint space width in rheumatoid arthritis. Ann Biomed Eng (2013) 41:2553–64. doi:10.1007/s10439-013-0871-x [4] Barnabe C, Toepfer D, Marotte H, Hauge EM, Scharmga A, Kocijan R, Kraus S, Boutroy S, Schett G, Keller KK, et al. Definition for rheumatoid arthritis erosions imaged with high resolution peripheral quantitative computed tomography and interreader reliability for detection and measurement. J Rheumatol (2016) 43:1935–1940. doi:10.3899/jrheum.160648 [5] Shimizu T, Choi HJ, Heilmeier U, Tanaka M, Burghardt AJ, Gong J, Chanchek N, Link TM, Graf J, Imboden JB, et al. Assessment of 3-month changes in bone microstructure under anti-TNFα therapy in patients with rheumatoid arthritis using high-resolution peripheral quantitative computed tomography (HR-pQCT). Arthritis Res Ther (2017) 19:222. doi:10.1186/s13075-017-1430-x [6] Keller KK, Thomsen JS, Stengaard-Pedersen K, Nielsen AW, Schiøttz-Christensen B, Svendsen L, Graakjær M, Petersen PM, Unger B, Kjær SG, Langdahl BL, Hauge EM. Local bone loss in patients with anti-citrullinated peptide antibody and arthralgia, evaluated with high-resolution peripheral quantitative computed tomography. Scand J Rheumatol (2018) 47(2):110-116. doi:10.1080/03009742.2017.1333629 [7] Ibrahim-Nasser N, Marotte H, Valery A, Salliot C, Toumi H, Lespessailles E. Precision and sources of variability in the assessment of rheumatoid arthritis erosions by HRpQCT. Jt Bone Spine (2018) 85:211–217. doi:10.1016/j.jbspin.2017.02.011 [8] Fouque-Aubert A, Boutroy S, Marotte H, Vilayphiou N, Bacchetta J, Miossec P, Delmas PD, Chapurlat RD, A. F-A, S. B, et al. Assessment of hand bone loss in rheumatoid arthritis by high-resolution peripheral quantitative CT. Ann Rheum Dis (2010) 69:1671–1676. doi:10.1136/ard.2009.114512 [9] Yue J, Griffith JF, Xiao F, Shi L, Wang D, Shen J, Wong P, Li EK, Li M, Li TK, et al. Repair of Bone Erosion in Rheumatoid Arthritis by Denosumab: A High‐Resolution Peripheral Quantitative Computed Tomography Study. Arthritis Care Res (Hoboken) (2017) 69:1156–1163. doi:10.1002/acr.23133 [10] Yue J, Griffith JF, Xu J, Xiao F, Shi L, Wang D, Wong PCH, Li EK, Li M, Li TK, et al. Effect of treat-to-target strategies on bone erosion progression in early rheumatoid arthritis: An HR-pQCT study. Semin Arthritis Rheum (2018) 48:374–383. doi:10.1016/j.semarthrit.2018.05.001 [11] Töpfer D, Finzel S, Museyko O, Schett G, Engelke K. Segmentation and quantification of bone erosions in high-resolution peripheral quantitative computed tomography datasets of the metacarpophalangeal joints of patients with rheumatoid arthritis. Rheumatology (Oxford) (2014) 53:65–71. doi:10.1093/rheumatology/ket259 [12] Töpfer D, Gerner B, Finzel S, Kraus S, Museyko O, Schett G, Engelke K. Automated three-dimensional registration of high-resolution peripheral quantitative computed tomography data to quantify size and shape changes of arthritic bone erosions. Rheumatology (Oxford) (2015) 54:2171–80. doi:10.1093/rheumatology/kev256 [13] Figueiredo CP, Kleyer A, Simon D, Stemmler F, D’Oliveira I, Weissenfels A, Museyko O, Friedberger A, Hueber AJ, Haschka J, et al. Methods for segmentation of rheumatoid arthritis bone erosions in high-resolution peripheral quantitative computed tomography (HR-pQCT). Semin Arthritis Rheum (2018) 47:611–618. doi:10.1016/j.semarthrit.2017.09.011 [14] Peters M, Scharmga A, De Jong J, van Tubergen A, Geusens P, Arts JJ, Loeffen D, Weijers R, Van Rietbergen B, van den Bergh J. An automated algorithm for the detection of cortical interruptions on high resolution peripheral quantitative computed tomography images of finger joints. PLoS One (2017) 12:1–15. doi:10.1371/journal.pone.0175829 [15] Peters M, de Jong J, Scharmga A, van Tubergen A, Geusens P, Loeffen D, Weijers R, Boyd SK, Barnabe C, Stok KS, et al. An automated algorithm for the detection of cortical interruptions and its underlying loss of trabecular bone; a reproducibility study. BMC Med Imaging (2018) 18:13. doi:10.1186/s12880-018-0255-7 [16] Peters M, van den Bergh JP, Geusens P, Scharmga A, Loeffen D, Weijers R, van Rietbergen B, van Tubergen A. Prospective Follow-Up of Cortical Interruptions, Bone Density, and Micro-structure Detected on HR-pQCT: A Study in Patients with Rheumatoid Arthritis and Healthy Subjects. Calcif Tissue Int (2019) 104:571–581. doi:10.1007/s00223-019-00523-2 [17] Henchie TF, Gravallese EM, Bredbenner TL, Troy KL. An image-based method to measure joint deformity in inflammatory arthritis: development and pilot study. Comput Methods Biomech Biomed Engin (2019) 22:942–952. doi:10.1080/10255842.2019.1607315