Temporal Association Between Rhinovirus Activity and Kingella kingae Osteoarticular Infections

To determine whether the seasonal distribution of Kingella kingae osteoarticular infections is similar to that of common respiratory viruses.

Between October 2009 and September 2016, we extracted the results of K kingae-specific real-time polymerase chain reaction analyses performed for bone or joint specimens in patients from 2 pediatric tertiary care centers in Paris. We used data of respiratory virus detection from the Réseau National des Laboratoires network with coordination with the National Influenza Center of France. The Spearman rank correlation was used to assess a correlation between weekly distributions, with P < .05 denoting a significant correlation.

During the 7-year study period, 322 children were diagnosed with K kingae osteoarticular infection, and 317 testing episodes were K kingae-negative. We observed high activity for both K kingae osteoarticular infection and human rhinovirus (HRV) during the fall (98 [30.4%] and 2401 [39.1%] cases, respectively) and low activity during summer (59 [18.3%] and 681 [11.1%] cases, respectively). Weekly distributions of K kingae osteoarticular infection and rhinovirus activity were significantly correlated (r = 0.30; P = .03). In contrast, no significant correlation was found between the weekly distribution of K kingae osteoarticular infection and other respiratory viruses (r = -0.17, P = .34 compared with respiratory syncytial virus; r = -0.13, P = .34 compared with influenza virus; and r = -0.22, P = .11 compared with metapneumovirus).

A significant temporal association was observed between HRV circulation and K kingae osteoarticular infection, strengthening the hypothesis of a role of viral infections in the pathophysiology of K kingae invasive infection.