A COMPARATIVE ANALYSIS BETWEEN THERMAL AND ULTRASOUND IMAGING AT THE ELBOW IN PATIENTS WITH RHEUMATOID ARTHRITIS

Y. K. Tan, R. Sultana, J. ThumbooSingapore General Hospital, Department of Rheumatology and Immunology, Singapore, Singapore Duke-NUS Medical School, Centre for Quantitative Medicine, Singapore, Singapore  Background Thermal imaging (TI) is a relatively low cost, non-invasive imaging technique that offers a quick and objective measurement of joint surface temperature. However, TI data at the elbow in patients with rheumatoid arthritis (RA) is presently lacking. Objectives The aim of this study is to compare TI outcomes with ultrasound (US) joint inflammation findings at the elbow among patients with RA. Methods TI and US imaging were performed at the same study visit. TI was carried out in a draft free room following a standardized protocol. The maximum (Tmax), average (Tavg) and minimum (Tmin) temperatures at the elbow (anterior, posterior, lateral and medial aspects) were summed up to obtain the respective MAX, AVG and MIN temperatures for the right and left elbows. Ultrasound power Doppler (PD) and greyscale (GS) joint inflammation were graded semi-quantitatively (0-3) at the anterior (humeroradial) and posterior fossa recesses of the elbow joint using previously validated scoring methods; and these were summed to obtain the respective PD and GS scores at the right and left elbows per patient. Pearson’s correlation coefficient was used to correlate the findings from TI and US imaging, while simple linear regression was used to describe the relationship between parameters. Results In this cross-sectional study, 60 elbows were evaluated by TI and US imaging among 30 adult RA patients with the following patient baseline characteristics: 76.7% female; 76.7% Chinese; mean (SD) disease duration 7.3 (6.8) months; mean (SD) DAS28 3.83 (1.19). Table 1 shows the results of the comparative analysis between the TI parameters (MAX, AVG and MIN) and the US PD and GS scores. For Pearson’s correlation, the TI parameters (MAX, AVG and MIN) were all significantly correlated (P<0.05) with the US PD scores (Table 1) at both the right and left elbows. For US GS scores (Table 1), significant correlation (P<0.05) were observed with all the TI parameters (MAX, AVG and MIN) only at the right elbow but not at the left elbow (with P-values all >0.05). The simple linear regression estimates between TI and US imaging parameters at the right and left elbows are summarized in Table 1. Conclusion To the best of our knowledge, our study is the first to report on the correlation analysis between TI parameters and US joint inflammation outcomes at the elbow from an RA cohort. Between US PD and GS joint inflammation, TI is more consistently associated with the former at the elbow in RA. TI of the elbow in RA appears promising and will require further validation in independent RA cohorts. Table 1. Comparative analysis between the thermal and ultrasound imaging parameters Thermal imaging parameter Pearson’s correlation Simple linear regression vs. GS score vs. PD score vs. GS score vs. PD score Correlation Coefficient (95% CI) P-value Correlation coefficient (95% CI) P-value Estimates (95% CI) P-value Estimates (95% CI) P-value Right elbow MAX 0.40 (0.04, 0.66) 0.029* 0.47(0.12, 0.70) 0.009** 0.16(0.02, 0.3) 0.030* 0.11(0.03, 0.18) 0.009** AVG 0.42(0.07, 0.68) 0.020* 0.43(0.07, 0.68) 0.017* 0.15(0.03, 0.28) 0.020* 0.09(0.02, 0.16) 0.018* MIN 0.39(0.03, 0.66) 0.032* 0.40(0.04, 0.66) 0.028* 0.15(0.01, 0.28) 0.033* 0.09(0.01, 0.16) 0.029* Left elbow MAX 0.32(-0.05, 0.61) 0.086 0.55(0.22, 0.75) 0.002** 0.11(-0.02, 0.23) 0.086 0.08(0.03, 0.13) 0.002** AVG 0.26(-0.12, 0.56) 0.174 0.52(0.19, 0.74) 0.003** 0.08(-0.04, 0.19) 0.172 0.07(0.03, 0.11) 0.003** MIN 0.15(-0.23, 0.48) 0.433 0.45(0.10, 0.69) 0.012* 0.04(-0.07, 0.16) 0.430 0.06(0.01, 0.10) 0.012* Statistically significant: *P<0.05, **P<0.01 REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests None Declared. Keywords: Imaging, Rheumatoid arthritis, Ultrasound DOI: 10.1136/annrheumdis-2023-eular.140Citation: , volume 82, supplement 1, year 2023, page 2005Session: Diagnostics and imaging procedures (Publication only)