Education

A customized protocol to assess bone quality in the metacarpal head, metacarpal shaft and distal radius: A high resolution peripheral quantitative computed tomography precision study

Description
A customized protocol to assess bone quality in the metacarpal head, metacarpal shaft and distal radius: A high resolution peripheral quantitative computed tomography precision study
Categories
Published
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  RESEARCH ARTICLE Open Access A customized protocol to assess bone quality inthe metacarpal head, metacarpal shaft and distalradius: a high resolution peripheral quantitativecomputed tomography precision study Lynne Feehan 1,4,5* , Helen Buie 2 , Linda Li 1,4 and Heather McKay 3,5 Abstract Background:  High Resolution-Peripheral Quantitative Computed Tomography (HR-pQCT) is an emergingtechnology for evaluation of bone quality in Rheumatoid Arthritis (RA). However, there are limitations with  standard  HR-pQCT imaging protocols for examination of regions of bone commonly affected in RA. We developed acustomized protocol for evaluation of volumetric bone mineral density (vBMD) and microstructure at the metacarpalhead (MH), metacarpal shaft (MS) and ultra-ultra-distal (UUD) radius; three sites commonly affected in RA. The purposewas to evaluate short-term measurement precision for bone density and microstructure at these sites. Methods:  12 non-RA participants, individuals likely to have no pre-existing bone damage, consented to participate[8 females, aged 23 to 71 y [median (IQR): 44 (28) y]. The  custom  protocol includes more comfortable/stable positioningand adapted cortical segmentation and direct transformation analysis methods. Dominant arm MH, MS and UUD radiusscans were completed on day one; repeated twice (with repositioning) three to seven days later. Short-term precisionfor repeated measures was explored using intraclass correlational coefficient (ICC), mean coefficient of variation (CV%),root mean square coefficient of variation (RMSCV%) and least significant change (LSC% 95 ). Results:  Bone density and microstructure precision was excellent: ICCs varied from 0.88 (MH 2  trabecular number) to .99(MS 3  polar moment of inertia); CV% varied from< 1 (MS 2  vBMD) to 6 (MS 3  marrow space diameter); RMSCV% variedfrom<1 (MH 2  full bone vBMD) to 7 (MS 3  marrow space diameter); and LSC%  95 varied from 2 (MS 2  full bone vBMD to21 (MS 3  marrow space diameter). Cortical porosity measures were the exception; RMSCV% varying from 19 (MS 3 ) to 42(UUD). No scans were stopped for discomfort. 5% (5/104) were repeated due to motion during imaging. 8% (8/104) of final images had motion artifact graded>3 on 5 point scale. Conclusion:  In our facility, this  custom  protocol extends the potential for in vivo HR-pQCT imaging to assess, with highprecision, regional differences in bone quality at three sites commonly affected in RA. Our methods are easy to adoptand we recommend other users of HR-pQCT consider this protocol for further evaluations of its precision and feasibilityin their imaging facilities. Keywords:  HR-pQCT, Bone microstructure, Volumetric bone mineral density, Precision, Metacarpal head, Metacarpalshaft, Ultra-ultra-distal radius, Early rheumatoid arthritis * Correspondence: lynne.feehan@gmail.com 1 Department of Physical Therapy, Faculty of Medicine, University of BritishColumbia (UBC), Vancouver, BC, Canada 4 Arthritis Research Centre of Canada, Richmond, BC, CanadaFull list of author information is available at the end of the article © 2013 Feehan et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the srcinal work is properly cited. Feehan  et al. BMC Musculoskeletal Disorders  2013,  14 :367http://www.biomedcentral.com/1471-2474/14/367  Background Despite marked improvements in the clinical managementof systemic inflammatory joint-disease in early rheumatoidarthritis (RA), people with RA remain at risk for developingunderlying systemic inflammatory mediated bone-changes[1-4]. Changes can include progressive periarticular bone thinning (osteopenia) and development of resorptive bone le-sions (erosions) [5,6]. Periarticular bone damage, most com- monly seen in the bone near the metacarpal phalangeal andwrist joints, can contribute to the development of hand de-formities and profound functional limitations in people livingwith RA [6,7]. Additionally, systemic extra-articular inflam- matory bone changes contribute to a two-fold increase infracture risk with aging in people living with RA [8-11]. Currently, radiography and several clinical imagingsystems, such as magnetic resonance imaging (MRI),computed tomography (CT), ultrasonography (US), dual-energy X-ray absorptiometry (DXA) and digital X-ray radiogrammetry (DXR) are used clinically to monitorbone changes in RA [12-17]. While these tools are useful for capturing later macro-structural joint and bone damagethat occurs in RA, their abilities to identify the earlier bonemicrostructural bone changes are poor. Thus, there is anurgent need for new imaging technologies and methodsto be developed that can reliably identify and characterizethese early changes  before  permanent macro structuralbone damage occurs. This is especially important given thatearly microstructural changes are potentially modifiableif they are reliably identified and treated early.High Resolution Peripheral Quantitative CT (HR-pQCT;SCANCO Medical AG, Brüttisellen, Switzerland) is apromising imaging technology capable of imaging finebone internal  ‘ micro ’  detail at a resolution similar tothe thickness of a human hair (75 to 100 microns) [18].Thus, HR-pQCT imaging is a promising tool for evaluatingthe changes in bone quality that accompany RA. However,research that uses this tool in RA is limited and just emer-ging [19-32]. Further, it is not possible to compare and synthesize findings from studies in RA that used HR pQCTas image location, acquisition and evaluation proceduresare not standardized and vary widely  [33].There are a number of possibilities for these inconsisten-cies with the primary reason related to applying  standard  protocols developed specifically for one region of interest(ROI) to another ROI without consideration of the tech-nical limitations for doing this. Secondly, although a posi-tioning device is available to support  standard   positioningof the arm, this device is not designed to position andstabilize the hand during imaging near the metacarpalphalangeal or wrist joint regions. Thirdly,  standard  semi-automated image evaluation protocols cannot reliably separate (segment) cortical and trabecular bone compart-ments in the periarticular metacarpal head and very distalradius bone regions that have very thin cortical shells.This is notable as these regions are commonly affected ininflammatory arthritis [34]. Finally,  standard   image evalu-ation protocols were not designed to evaluate regions thatare comprised primarily of compact lamellar cortical bonesuch as found in the extra-articular metacarpal mid-shaftregion which is also commonly affected in inflammatory arthritis [3,35,36]. Recently, HR pQCT semi-automated image analysiscapabilities were advanced to allow more accurate seg-mentation of the cortical bone compartment [37,38]. This relatively new approach was developed to evaluateregions of bone with a thin cortical shell and thereforeovercomes some of the limitations associated with the  standard   imaging protocols. In addition, direct transform-ation image analyses methods developed for microCTanalyses ex vivo were recently adapted to evaluate corticalbone density, morphometry and porosity in vivo  ,  using HR-pQCT [38-41]. Importantly, these advances permit evalu- ation of several micro-structural and macro-structural boneparameters within the integral, trabecular and cortical bonecompartments that could not previously be assessedusing  standard   HR-pQCT evaluation protocol, in vivo.There is a need, however, to assess the precision of adaptedsemi-automated cortical compartment segmentation andadapted direct transformation image analyses methods forHR-pQCT assessment in vivo  ,  generally and at bone sitescommonly affected by RA (e.g. periarticular distal radiusand metacarpal head regions and extra-articular metacarpalmid-shaft region).Therefore, the purpose of this study was to determinethe short term precision of an HR-pQCT imaging protocol,in vivo customized for the hand and distal radius. Thenovel features of this protocol include: 1) comfortable posi-tioning and better stabilization of the head, trunk and upperarm, 2) standardized positioning of the hand and forearmusing a custom-made positioning device, and 3) adaptedsemi-automated cortical segmentation and direct trans-formation image analyses methods that permit assessmentof integral, cortical and trabecular bone macro- and micro-structural morphometry and bone mineral density atthe Metacarpal Head (MH), Metacarpal Shaft (MS) andthe Ultra-Ultra-Distal (UUD) radius bone regions. Weuse the term Ultra-Ultra-Distal (UUD) radius to differentiatethe more distal periarticular distal radius location examinedin our study, from the  standard   ultra-distal radius scan loca-tion [42]. Our secondary objectives were to explore partici-pant tolerance to the novel positioning protocol as well asrates for re-scanning due to motion during imaging andexcessive image motion artifact (e.g. graded > 3 on themanufacturer 5 point rating scale) in the final images [43]. Methods This precision study was conducted in a medical imagingresearch centre setting and received academic institutional Feehan  et al. BMC Musculoskeletal Disorders  2013,  14 :367 Page 2 of 12http://www.biomedcentral.com/1471-2474/14/367  ethical approval from the University of British Columbia,Vancouver Canada. Community-dwelling adults wererecruited from a large urban metropolitan setting. Par-ticipants received no financial remuneration for partici-pation and provided informed consent to participate.With the exception of a physician diagnosis of inflam-matory arthritis, participants were not screened for any other self-reported health (e.g. diabetes, osteoporosis)or lifestyle (e.g. smoking, alcohol consumption, physicalinactivity) condition that may have affected their bonehealth. We specifically excluded individuals with a diagnosisof inflammatory arthritis as we were not be able to de-termine a priori if they may already have underlyingmacro-structural bone damage in the regions of bonewe were examining. Participants were also excluded if they: 1) had any physical condition that would preventthem from sitting motionless with their arm in thescanner supported by a positioning device for up to 6minutes, 2) had metal or surgical implants in the hand orforearm of interest, 3) were pregnant or possibly pregnant,4) had sustained a fracture in their dominant arm hand orforearm in the previous 12 months, and 5) were unable toread or understand the consent form.Prior to scanning we assessed height (cm) using awall mounted stadiometer (SECA corp. Chino, CA)and weight (kg) using a medical grade digital floorscale (Tanita Corporation of America, Inc. ArlingtonHeights, Ill) using standard techniques. We derivedbody mass index (BMI) as wt/ht 2 (kg/m 2 ) [44]. Follow-ing these anthropometric measures, the hand and fore-arm were positioned in a custom-made positioningdevice made of rigid thermoplastic splinting material.The forearm was aligned parallel to the long axis of thesplint and the metacarpal phalangeal joints positionedin 0 degrees of flexion. The splint-supported hand andforearm were then positioned within a holder that wasmodified from manufacturer specifications to suit thehand (Scanco Medical AG, Switzerland). The hand andforearm were then stabilized with additional strapping(Figure 1A). Participants were positioned to face theimaging system. Pillows were placed behind partici-pants ’  hips and in front of them so that the participantcould lean forward and rest on the pillows with theiropposite arm, upper body and head comfortably sup-ported. The holder, with the arm correctly positionedwithin it, was then placed inside the HR pQCT unit forscan acquisition (Figure 1B).A single trained operator (author LF) performed allscans using standard in vivo imaging parameters (82  μ mnominal isotropic resolution, 60 kV  p  effective energy,900  μ A current, and 100 ms integration time). The traininginvolved a rigorous and standardized training protocol de- veloped by the facility for the safe operation of the scanner.Manufacturer specifications for the scanner define that forevery 110 slices acquired the measurement time is 2.8minutes with an effective dose of 3  μ Sv at distal extremity sites. This estimate of effective dose is based on a weightedcomputed tomography dose index (CTDIw) of 6.1 mGy and a local dose of 3.2 mGy using standard HR-pQCTin vivo image acquisition parameters [45]. A trained oper-ator also performed daily density calibrations and weekly geometry calibrations of the HR-pQCT imaging systemusing the manufacturer ’ s calibration phantom.Three scans of the dominant arm were completed inseries during a single scanning session. The ROIs includedthe metacarpal head (MH), metacarpal mid-shaft (MS) andultra-ultra-distal (UUD) radius sites. To assess short-termprecision with repositioning, we acquired two additionalseries of three scans with repositioning between each series.The additional two series were completed during a singlescanning session, three to seven days after the initial scans.Prior to each scan, we performed a 150 mm length scout view of the hand and distal forearm which is the maximumavailable length for a scout view. The reference line for theradius scan was located at the medial edge of the distal ra-dius; the scan region was 1 mm proximal to this referenceline and extended 9.02 mm (110 slices) proximally. For themetacarpal head scan, the reference line was the tip of themost distal second or third metacarpal head; the scanstarted 2 mm distal to this reference line and extended18.04 mm (220 slices) proximally. For the metacarpal shaftscan, the reference line was half (50%) the total length of the metacarpal shaft assessed on the scout view. The meta-carpal shaft scan region of interest extended from 4.5 mmdistal to the reference line to 9.02 mm (110 slices) proximalto the reference line (Figure 2 A, B, C).The operator visually assessed all images for motionartifact at the completion of the three-scan series. If mo-tion artifact was apparent in only one image the operatorrepeated the scan. If there was motion artifact in two ormore of the scans across the series, the operator repeatedthe scan at one site only. Our image order of priority wasthe distal radius followed by the metacarpal head.Images were then independently analysed by 1 of 2trained and experienced operators, one of whom wasthe same person as the image acquisition operator in thisstudy (first author LF), the other a study research assist-ance. Before conducting any image analysis in this study,each operator was required to obtain an intra-rater reliabil-ity coefficient (Pearson R) of  ≥ 0.90 for measures of UUDtrabecular bone fraction from at least 10 images assessedtwice by the same operator within 7 to 10 days [46].Prior to analysis, each image was graded visually formotion artifact using the 5-point manufacture gradingsystem [47]. We included images graded 3 or less by bothoperators for final data analysis [43]; any disagreement wasresolved by consensus. Image analyses were conductedbased on operator availability; operators did not use image Feehan  et al. BMC Musculoskeletal Disorders  2013,  14 :367 Page 3 of 12http://www.biomedcentral.com/1471-2474/14/367  registration to evaluate repeated scans. Operators wereblinded to previous image analyses data; we allowed atleast 10 days between image analyses of a repeated scan inany individual by the same operator. Both operatorsassessed the same numbers of scan images.Using the manufacturer evaluation software (V 6.0), theoperator analyzed five sub-regions of interest [1 - UUDradius (110 slices); 2 - MH2 & MH3 (110 slices); 2 - MS2and MS3 (110 slices)] (Figure 2, A,B,C). They performedsemi-automated contouring of the periosteal bone surfaceand segmented bone from surrounding soft tissue usingstandard manufacturer evaluation script protocols [48].The operator extracted cortical and trabecular regionsusing the semi-automated segmentation method [37,38], but applied a modified boundary condition for analysis of the metacarpal head.Following initial segmentation, the operator mademinor adjustments to endosteal and periosteal contoursas needed [39]. This step included a visual inspection of the computer generated lines for delineation of the cor-tical region segmentation in all slices, making minormanual corrections to any deviations from accurate peri-osteal or endosteal surface delineation (Figure 2, D,E,F).Manual correction at this step was rarely indicated; usually only required for the correction of the endosteal edgedelineation in a limited number of slices in any image.The most common reason for the need for any manualcorrection was in instances when there were very lar-ger intra-cortical pores or large bi-cortical breaks cre-ated by vascular channels. These manual adjustmentprocedures have been described in further detail by Burghardt et al., [38].The operator then ran a series of evaluation scripts usingthe manufacturer evaluation software for assessmentof the full, cortical and trabecular bone regions usingdirect transformation image analyses scripts adapted fromstandard microCT evaluation scripts recently developed forcortical bone and described in more detail by NishiyamaKK et al. [40], and Liu XS et al., [41]. These adopted direct transformation evaluation scripts for HR-pQCT arenow included in current upgrades of manufacturerevaluation software.For the periarticular UUD Radius, MH2 and MH3regions we examined apparent volumetric bone min-eral density (vBMD) for the full (vBMD full  - mgHA/cm 3 ),cortical (vBMD Cort  - mgHA/cm 3 ) and trabecular(vBMD Trab  - mgHA/cm 3 ) bone regions. We also ex-amined selected microstructural morphometric boneparameters, including:   Cortical bone : thickness (CtTh - mm) and porosity (CtPo - %). Figure 1  Custom image acquisition positioning. A)  Shows the standardized positioning of the hand and forearm (left or right) in a custom-madeinsert (top) with additional stabilization and placement in a modified manufacturer ex-vivo holder (bottom).  B)  Shows the modified positioning forimaging with an individual seated on a chair facing scanner with their head, upper body and opposite arm resting on pillows with the hand to bescanned in the holder and positioned inside the scanner for scanning. Feehan  et al. BMC Musculoskeletal Disorders  2013,  14 :367 Page 4 of 12http://www.biomedcentral.com/1471-2474/14/367    Trabecular bone : volume fraction (BV/TV  trab  - %),number (TbN  –  1/mm), thickness (TbTh - mm) andseparation (TbSp - mm). At the extra-articular MS2 and MS3 mid-shaft siteswe examined full and cortical bone apparent volu-metric BMD (vBMD full  & vBMD cort  - mgHA/cm 3 ), aswell as, cortical bone material bone mineral density (vTMD cort  - mgHA/cm 3 ). In addition we examined thefollowing selected micro- and macro-structural mor-phometric parameters:   Full bone : volume (BV  full -  mm 3 ), volume fraction(BV/TV   full -  %), section modulus  –  major direction(SM full -  mm 3 ), polar moment of inertia(pMOI full -  mm 4 ), and marrow space diameter(MSdia - mm).   Cortical bone : thickness (CtTh - mm), porosity (CtPo - %), volume (BV  cort -  mm 3 ), volume fraction(BV/TV  cort  - %), section modulus  –  major direction(SM cort -  mm 3 ), polar moment of inertia(pMOI cort -  mm 4 ). Direct transformation evaluation methods applied toimages acquired using HR-pQCT, in vivo tend to over-estimate some trabecular bone outcomes (TbTh, TbSpand BV/TV  trab)  [49,50]. Therefore, the  standard   manu-facturer HR-pQCT evaluation script applies a correctionfactor to these parameters to adjust for known differences.We also applied this correction factor to variables acquiredat the UUD Radius, MH2 and MH3 sites so as to directly compare our data with values acquired using  standard  image evaluation methods at other bone regions [41].Trabecular bone volume fraction (BV/TV  trab_s ) was de-rived using a standard approach [trabecular bone appar-ent volumetric bone mineral density (vBMD trab ) dividedby 1200 mg/cm 3 )]. Trabecular thickness (TbTh s ) andtrabecular separation (TbSp s ) were derived using a standard Figure 2  Scan locations and cortical segmentation.  Top Row  (A,B,C)  shows the reference line, scan location and Region of Interest (ROI)analyses overlaid on a 150 mm scout view for the Ultra-Ultra-Distal Radius  (A) , Metacarpal Head  (B)  and Metacarpal Shaft  (C)  scans. Bottom Row (D,E,F)  shows examples of semi-automated cortical compartment segmentation in one HR-pQCT slice for the UUD radius  (D) , Metacarpal Head (E)  and Metacarpal shaft  (F)  ROIs. Feehan  et al. BMC Musculoskeletal Disorders  2013,  14 :367 Page 5 of 12http://www.biomedcentral.com/1471-2474/14/367
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks