Intra-articular and soft tissue corticosteroids injections and risk of infections

Galia Zacay, Anthony Heymann

Keywords: corticosteroids, infections, intra-articular injection

Background:

Intra-articular injection of corticosteroids was first reported in 1951. Since then, this treatment has become a common office-performed procedure for the treatment of articular and peri-articular inflammatory conditions and for pain control. There are reports of few systemic adverse events mainly hypothalamic-pituitary-adrenal axis suppression and transient elevation in blood glucose level in patients with diabetes. There are no reports regarding infection as a possible adverse event.

Research questions:

To assess the risk of infection after intra-articular or soft tissue steroid injections.

Method:

An historical prospective cohort study.
Participants: 7,088 individuals that met the inclusion criteria entered the study.
Intervention: The participants were self-matched and we analyzed the incidence of infection for three periods. We defined the exposure period as the 90 days that followed the injection, and two 90 days control periods.
Primary outcome: occurrence of the following infections; cellulitis, herpes zoster, influenza, osteomyelitis, pneumonia, septic arthritis, sinusitis, and urinary tract infection using the visit diagnosis field in the patient’s electronic medical record.

Results:

Self-matching analysis using conditional logistic regression showed significantly increased odds for the combined incidence of cellulitis, pneumonia, herpes zoster or urinary tract infection in the post exposure period compared with the control periods: OR 1.21, CI 1.05 to 1.32. For patients treated with betamethasone as opposed to methylprednisolone the odds ratio for infection were OR 1.4 CI 1.16 to 1.68, and OR 0.99 CI 0.65 to 1.49 respectively. The increased rate of infection appeared twenty days after the exposure date and peaked between 21-30 days.

Conclusions:

Intra-articular and soft tissue injections of betamethasone are associated with increased rates of infection that occur from three weeks after exposure. Further research should assess if methylprednisolone infections are safer.

Points for discussion:

#16