Estimating the incidence of autoimmune inflammatory arthritis after Lyme disease

Objective

A previous case series described the development of new autoimmune, inflammatory arthritis (IA) developing within 2 years of Lyme disease (LD). This study aimed to estimate the incidence of IA following LD using administrative claims data. Influenza, an infection not typically associated with post-infectious IA, was used for comparison.

Methods

We conducted a retrospective cohort study using the Johns Hopkins Health System administrative claims data from 01/2013–05/2024. Patients with LD and influenza were identified using International Classification of Diseases (ICD) codes, with LD cases further defined by requiring an antibiotic prescription within 30 days of the ICD code. IA cases were identified using 2+ ICD codes for rheumatoid arthritis, psoriatic arthritis or spondyloarthritis. Comparisons between infection groups were conducted using chi-square tests and Z-tests as appropriate. Logistic regression analyses were conducted to estimate the odds of IA development within two years of Lyme or influenza diagnosis, controlling for age and sex.

Results

The incidence of IA was significantly higher following LD compared to influenza (1.67 % vs 0.45 %, p < 0.0001), with the highest incidence of IA occurring within the first year after LD. Regression analysis showed LD was associated with increased odds of IA compared to influenza (OR 3.76, 95 % CI: 2.41–5.86, p < 0.001) after adjusting for age and sex.

Conclusion

The incidence of IA was higher within a year of LD infection, higher than that expected in the general population. The temporal association and elevated incidence support the need for prospective studies to elucidate mechanisms linking infection and autoimmunity in LD.