Slide 1 11/7/2011 1 Assessment of Tophus Size: a Comparison Between Physical Measurement Methods and Dual Energy Computed Tomography Scanning Nicola Dalbeth, Opetaia Aati, Angela Gao, Meaghan House, Qiliang Liu, Anne Horne, Anthony Doyle, Fiona M McQueen Disclosures • This work was funded by the Health Research Council of New Zealand • Qiliang Liu was the recipient of a University of Auckland summer studentship • I have no other relevant financial disclosures. Background • The tophus is a pathognomonic feature of gout • Foreign body granulomatous response to monosodium urate (MSU) crystals – Innate and adaptive immune activation Dalbeth Arthritis Rheum 2010 Background • Impact of tophi – Disfiguring – Discharge with secondary infection – Obstruct joint movement – Disability – Joint damage • Tophus regression has been endorsed by OMERACT as a core domain for clinical trials of chronic gout Schumacher J Rheumatol 2009 Background • Many methods of tophus measurement described: – Vernier calipers (longest index tophus diameter)* – Tape measurement (index tophus area) – Counting of all visible tophi – Digital photography – Ultrasonography – Magnetic resonance imaging – Conventional computed tomography (CT) *OMERACT endorsed Dalbeth ARD 2011 Background • Dual energy CT (DECT) is a sensitive and specific method to detect urate deposits in patients with gout • DECT uses a specific display algorithm that assigns different colours to materials of different chemical composition (such as urate and hydroxyapatite) • The reliability of DECT for tophus measurement has not been reported to date Choi ARD 2009 Glazebrook Radiology 2011 11/7/2011 2 Aim • To compare the reliability and validity of various physical methods with DECT assessment of tophus size Methods: patients and tophus selection • Twenty-five patients with – a history of acute gout according to ACR classification criteria, and – at least one subcutaneous tophus • For each patient, up to three index tophi were selected for analysis (n=64 tophi, 55 in the feet) – sites in the feet were preferentially selected – if >3 tophi present in the feet, the largest tophi were selected – discharging, acutely inflamed or bursal tophi were not selected Methods: physical measurement • Each tophus was assessed by two independent observers – Vernier calipers (longest diameter) – Tape measure (area) • Tophus location was recorded in detail using a diagram and written description • The total number of subcutaneous tophi was also counted • Five patients returned within one week for repeat physical assessments Methods: DECT • All patients proceeded to DECT scanning of both feet (Somatom Definition Flash, Siemens Medical) • Index tophus DECT volume was assessed by two independent observers using automated volume assessment software • DECT scans from the returning patients were scored twice by both observers Methods • Each observer was blinded to the scores of the other observers and previous measures • Intra- and inter-observer reproducibility was assessed by intraclass correlation coefficient (ICC) and limits of agreement analysis (Bland and Altman). • For the purposes of these analyses the unit of investigation was assumed to be the tophus Results: patient characteristics Variable All patients (n=25) Patients returning for second visit (n=5) Age, years, median (range) 64 (40-85) 64 (44-74) Male gender, n (%) 23 (92%) 5 (100%) Ethnicity, n (%) Pacific New Zealand Maori New Zealand European/Other 10 (40%) 1 (4%) 14 (56%) 2 (40%) 0 (0%) 3 (60%) Aspirate proven gout 11 (44%) 2 (40%) Gout disease duration, years, median (range) 24 (3-50) 45 (21-49) Serum urate, mmol/L, median (range) 0.39 (0.18-0.71) 0.37 (0.35-0.49) On allopurinol, n (%) 18 (72%) 4 (80%) Total DECT urate volume (both feet), cm3, median (range) 1.65 (0.07-28.88) 8.02 (0.13-28.88) 11/7/2011 3 Results: Intraobserver reproducibility (Assessment 1 vs. Assessment 2) ICC, mean (95% CI) A Vernier calipers 0.75 (0.54-0.87) B Tape measure 0.91 (0.82-0.96) C Tophus count 0.94 (0.77-0.98) D DECT volume 1.00 (0.99-1.00) To allow comparison between measures, the y axis value approximates twice the mean score. Results: Interobserver reproducibility (Observer 1 vs. Observer2) ICC, mean (95% CI) A Vernier calipers 0.78 (0.66-0.86) B Tape measure 0.88 (0.82-0.93) C Tophus count 0.58 (0.25-0.79) D DECT volume 0.95 (0.92-0.97) To allow comparison between measures, the y axis value approximates twice the mean score. Results: Comparison of values between different methods (feet, n=55) Calipers Tape Calipers - rs=0.94 p<0.0001 Tape rs=0.94 p<0.0001 - DECT rs=0.46 p=0.004 rs=0.46 p=0.004 • In 20% of tophi recorded on physical assessment, no urate deposits were observed in the tophus by DECT • Those tophi without urate deposits on DECT had smaller caliper diameter (p=0.02) and tape area (p=0.01) Results: distribution of urate in tophi • Large variation was observed in the amount of urate deposits documented by DECT in tophi of similar physical size • Discrete urate collections were frequently scattered throughout the tophus, typically surrounded by soft tissue Example of two similar sized tophi from a single patient showing large variation in urate volume. The borders of the tophus (determined from CT images) are outlined. Summary • DECT reveals the composition of tophi which contain variable urate deposits embedded within soft tissue • DECT scanning is a highly reproducible method of assessing urate load within tophi – Overall higher reproducibility than physical tophus measurement methods – Relatively modest relationship between physical tophus size and DECT urate volume, reflecting the composition of the tophus Further questions/issues • Is DECT sensitive to change? • Is DECT feasible for use in clinical trials? – Cost – Availability – Radiation • What is the relative importance of urate load compared with total tophus size on clinically relevant outcomes?