Pos1360 patients are always right! association between humidity level and pain experience in hand osteoarthritis

Background While many patients with joint disease report pain changes according to meteorological factors (1), this common belief has not been clearly demonstrated and never been documented in hand osteoarthritis (HOA). Objectives We aimed to investigate the cross-sectional association between meteorological factors that are ambient temperature and humidity on HOA symptoms. Methods We used the baseline data from the prospective DIGICOD HOA cohort including patients with established symptomatic HOA. The clinical outcomes were AUSCAN-pain, AUSCAN-stiffness and AUSCAN-function subscores, VAS-scale for pain during activity, number of spontaneous tender joints and tender joints at palpation. Meteorological data (from Météo France), ambient temperature and humidity, are defined as the mean value observed during the 3 days before inclusion (day of inclusion, D-1 and D-2). Considering non-normal distribution, each outcome was binarized by the median value, in order to define high and low symptoms intensity. Association between each outcome and meteorological data were studied using two logistic regression models (separate models for temperature and humidity), adjusted for age, sex, sum of Kellgren-Lawrence score of all hand joints and Hospital Anxiety Depression scale. Assumption of linearity between temperature or humidity and the logit of the outcomes being not verified, those two parameters were introduced in the models as categorical variables according to quartiles. Results are presented as odds ratio (OR) with 95% confidence intervals (CIs). The reference groups were respectively lower humidity and lower temperature quartile. Results We have studied 377 patients in whom all variables of interest were available (mean age ± standard deviation (SD) 66.5 ± 7.4 years, 85% women). Median AUSCAN-pain subscore was 20/100 (IQR [8; 38]), median AUSCAN-stiffness subscore was 22/100 (IQR [7; 53]), median AUSCAN-function subscore was 32/100 (IQR [13; 55]), median VAS-pain during activity was 42/100 (IQR [22; 66]), median number of spontaneous tender joints was 0/30 (IQR [0; 2]), median number of tender joints at palpation was 3/30 (IQR [2; 6]). Results of models for humidity are presented in Table 1. There was an association between humidity percentage and AUSCAN pain ≥ 20 for third quartile vs. first quartile (OR [79-85%] vs. [43-68%] = 1.99 (1.08 to 3.68), p = 0.03) and between humidity and VAS pain during activity ≥ 42 for second quartile vs. first quartile (OR [ 68-79%] vs. [43-68%] = 1.97 (1.06 to 3.63) p = 0.03). Considering patients with at least 1 spontaneous hand joint pain, an association was found for all 3 higher quartiles of humidity ([68-79%], [79-85%] and [85-96%]) compared to the lower one ([43-68%]), but without dose-effect. There was no effect of humidity on AUSCAN stiffness or function, or number of tender joints at palpation. Temperature was not associated with any clinical outcomes (pain, function or stiffness). Conclusion This is the first study about impact of meteorological factors in HOA. While we did not find any association between temperature and symptoms, we found an association between humidity and pain. Although we cannot conclude on any causality between humidity and pain, our results support patient’s belief about influence of humidity on joint pain. Reference [1]Timmermans EJ, J Rheumatol 2015;42:1885–92. Table 1: Association between humidity and HOA symptoms: logistic regression models with systematic adjustment on sex, age, sum of Kellgren-Lawrence score for all hand joint and HAD score. OR (IC 95% ) p-value AUSCAN-pain subscore ≥ 20 Humidity (%) 0.14 1. [43 ; 68[ 1 2. [68 ; 79[ 1.76 (0.96;3.25) 0.07 3. [79 ; 85[ 1.99 (1.08;3.68) 0.03 4. [85 ; 96[ 1.52 (0.83;2.79) 0.18 At least 1 spontaneous tender joint Humidity (%) 0.02 1. [43 ; 68[ 1 2. [68 ; 79[ 1.74 (0.94;3.21) 0.08 3. [79 ; 85[ 2.67 (1.44;4.94) 0.002 4. [85 ; 96[ 1.95 (1.06;3.61) 0.03 VAS-scale for pain during activity ≥ 42 Humidity (%) 0.1832 1. [43 ; 68[ 1 2. [68 ; 79[ 1.97 (1.06;3.63) 0.03 3. [79 ; 85[ 1.58 (0.86;2.91) 0.14 4. [85 ; 96[ 1.56 (0.85;2.86) 0.15 Acknowledgements: NIL. Disclosure of Interests None Declared.