Validation of Handheld Ultrasound Devices in Rheumatology

This study is a clinical trial testing the concurrent validity of Clarius handheld ultrasound devices against the gold-standard device to detect characteristics (i.e., anatomical structures and vascular flow) of healthy and rheumatic joints in adult Psoriatic Arthritis patients. The study will be conducted in 3 centers and 10 patients will be included in each center. Study patients will not be randomized to any group. Both B mode and Doppler images of the participants' predetermined nail, joint, tendon and enthesitis areas will be recorded with both handheld (Clarius Mobile Health Inc, HD3 L15 scanner, HD3 L20 scanner) and the gold standard device (GE LogicE9/E10) at baseline visit. Two paper CRFs, patient and physician CRFs, will be filled for each patient during the visit and these will be the source document. The CRFs (eCRFs) will be scanned and uploaded to the SharePoint at each center, within 3 days after the visit. In addition, US images of each patient will be uploaded to the SharePoint. The paper CRFs (source documents) will be kept on site. US images and CRFs uploaded from each center will be reviewed by the research assistant at Ottawa Hospital Research Institute (OHRI) center. If there are any missing or erroneous data in the CRF copy, research assistant will contact the site to ensured that the errors and deficiencies are corrected in source document according to Good Documentation Practice. Site should then re-scan the CRF to the SharePoint, ensuring sequential versioning. Any queries will be confirmed with the site within a week of the data entry. Then the analyzed CRFs data will be transferred to Research Electronic Data Capture (REDCap) ( version 12.4.18 - © 2023 Vanderbilt University) by the research assistant at OHRI. Before each analysis (interim and final), all the paper CRFs and REDCap data will be compared for quality assurance. The scoring of the US images will be done blindly by the principal investigator at OHRI as stated below. At the central site (OHRI), the research assistant will give a unique identifier number to each image, for a random quality control and for cross referencing whenever needed. The cropped images, as detailed below, will not have the subject ID visible to the PI at the time of reading but will be accessible for the quality control. (read-only access). The research assistant at OHRI is the only site personnel who has the capacity to uncrop the images in the PowerPoint file (password protected files). Images will not contain any identifiable information such as Date of Birth (DOB) or initials. The US images will be transferred to a PowerPoint file by the research assistant at OHRI in JPEG format. The research assistant will generate an unblinded master list, inaccessible to other site personnel, to link the slide numbers with the patients and scanned anatomical sites and the slide will have no other information on the patient number or ID. For scoring the images by the PI, a random order slide show will be conducted, irrespective of the machine used or the anatomical site or patient assessed, to ensure blindness to data related to the patient identifiers (The PI will not be blinded to the machine that the image was taken with as the JPEG format that is achieved from different machines are identifiable, but due to the random order scoring, images that belong to the same joint by the different machines are not to be scored consecutively). There will be nine separate powerpoint files, for images of joints, tendons, entheses, nail including power Doppler and gray scale findings; and grey scale file for erosions. If there are missing images for some sites for any of the probes, the images that were obtained for the same site using other probe(s) will also be excluded. The number of missing joint/tendon/entheses and nail images will be reported. Missing data will not be imputed. After the images are evaluated blindly, statistical analysis will be completed as described below. The primary endpoint analysis will be the interrater agreement of detecting any synovitis in B mode with the Clarius and gold standard machine. The kappa coefficients will be evaluated using the guideline outlined by Landis and Koch, where the strength of the kappa coefficients are: 0.01-0.20 slight; 0.21-0.40 fair; 0.41-0.60 moderate; 0.61-0.80 substantial; 0.81-1.00 almost perfect. For secondary outcomes, the interrater agreement for the presence of Doppler signals within the joints, tenosynovitis, erosions, nail, as well as features of enthesitis (hypoechogenicity, thickening, erosions, enthesophytes, calcifications) will also be evaluated using the same method. The agreement of the semiquantitative grading of the intraarticular findings' severity (synovitis in B mode, Doppler signals, erosions, each being on a scale between 0-3) will be done using weighted kappa analysis. The study will be completed after the imaging of all patients is completed and the analysis of data is done.