To determine if thermography (in comparison with ultrasonography) may be helpful in detecting joint inflammation at the RA wrist categorised according to its clinical manifestations. Four wrist groups were derived from the right wrist of RA subjects as follows: (1) swollen; tender (S1T1); (2) swollen; non-tender (S1T0); (3) non-swollen; tender (S0T1); (4) non-swollen; non-tender (S0T0). Thermographic parameters included the maximum (Tmax), average (Tavg) and minimum (Tmin) temperatures. Ultrasound parameters included the Total PD (TPD) and Total GS (TGS) scores. One-way ANOVA and Kruskal-Wallis test (for normally and non-normally distributed imaging parameters, respectively) and subsequent post-hoc tests were carried out for the comparative analysis of the wrist groups. A total of 70 wrist joints of 70 RA subjects were included in this cross-sectional study. For all imaging parameters (Tmax, Tavg, Tmin, TPD and TGS), statistically significant differences (all p<0.05) were detected (a) between the 4 wrist groups using either the one-way ANOVA or Kruskal-Wallis test and (b) for subsequent pairwise comparison of wrist group 1 (S1T1) vs. group 4 (S0T0) and group 2 (S1T0) vs. group 4 (S0T0). No significant differences (all p>0.05) were found for pairwise comparison of wrist group 3 (S0T1) vs. group 4 (S0T0) for all imaging parameters. Thermography at the wrist appears promising in RA with its findings closely mirroring those from ultrasonography. Swollen joints (regardless of tenderness status) have higher joint surface temperatures and greater ultrasound-detected joint inflammation, findings which were not observed for tender only (non-swollen) joints.