Identification of initial severity determinants to predict arthritis after chikungunya infection in a cohort of French gendarmes

Background: The objective was to identify severity characteristics of initial chikungunya infection (CHIKV) stages associated with post-CHIKV arthritis and arthralgia. Methods: French gendarmes exposed to the 2005-2006 CHIKV epidemic in Reunion Island who completed the 2006 (self-reporting acute and early chronic [median: 6 months] symptoms) and 2008 (Endpoint [median: 30 months]: self-perceived recovery and rheumatic disorders (RDs)) surveys were included. Multinomial logistic regression and multiple correspondence analysis (MCA) were used. Arthralgia was defined by joint pain and/or stiffness and arthritis by joint swelling in addition to pain and/or stiffness. Results: In 2008, 124 (31.3%)/403 participants (101 CHIKV+/302 CHIKV-) reported arthralgia and 57 (14.1%) arthritis. The multivariate model kept CHIKV infection, comorbidity and acute stage depressed mood as independent prognostic factors for both arthralgia and arthritis, but found early chronic stage RD as the main determinant of the same RD two years later. The MCA performed with the 85 CHIKV + patients who answered the question on self-perceived recovery enabled the calculation of severity scores based on initial symptoms that were strongly associated with persistent arthritis and, to a lesser extent, to arthralgia in bivariate analyses. The MCA graph clearly distinguished arthritis as the only RD associated with early severity indicators represented by sick leave, joint swelling and depressed mood during the acute stage, and early chronification of arthritis and depressed mood. Conclusion: Initial CHIKV severity predicted recovery, with higher severity associated with arthritis and lower severity with arthralgia. More interestingly, specific markers of post-CHIKV arthritis, which can easily be used by clinicians for case management, were identified.