Background: The risk of tuberculosis (TB) has decreased in biologic disease modifying anti-rheumatic drugs (bDMARDs) treated rheumatoid arthritis (RA) patients, but remains unaltered 4-fold increased in bio-naïve RA patients compared to the general population in Sweden (1). In absolute numbers, most TB cases in contemporary RA patients occur in the group of bio-naïve patients. Knowledge about risk factors for TB and TB characteristics in bio-naïve RA patients is still limited.

Objectives: To investigate risk factors for TB and TB characteristics in bio-naïve RA patients.

Methods: Population-based case-control study. A national bio-naïve RA cohort was identified from the National Patient Register and the Swedish Rheumatology Quality Register. RA cases with TB were identified by linkage to the Swedish Tuberculosis Register (with mandatory TB registration) 2001-2014 (n=42). For each case, four matched RA controls without TB were identified. Clinical data were obtained from medical records. Univariate and multivariable logistic regression analyses were used to estimate risk for TB expressed as adjusted (adj) odds ratio (OR) with 95% confidence intervals (CI).

Results: After review of the medical records and validation of diagnoses, 31 cases with RA and TB and 122 controls remained in the study. The TB cases had a median of 3 (1-6) reported TB risk factors, and almost 90% were born before 1950. Only one case was screened for TB (with negative result of tuberculin skin test). Active TB occurred at a mean of 15 years after RA diagnosis, and all except three cases were considered as reactivation of latent TB. Exposure to leflunomide (5 cases, 4 controls) (adj OR 6.02; 95% CI 1.47-24.65) and azathioprine (5 cases, 6 controls) (adj OR 3.85; 95% CI 1.06-13.79) were associated with increased risk for TB. Methotrexate, used in 67.7% of cases and 73.9% of controls, was not associated with increased risk of TB (adj OR 0.83; 95% CI 0.34-1.98). Exposure to corticosteroids was more common among cases than controls (74.2% vs 53.8%, p= 0.04), and was associated with an adj OR for TB of 2.44 (95% CI 1.00-5.92). No significant differences were identified between prednisolone-treated cases and controls in terms of maximum dose ever of prednisolone, treatment duration before TB, or cumulative dose of prednisolone during the last year before diagnosis of TB. Obstructive pulmonary disease was the only comorbidity linked to an increased TB risk (adj OR 3.94; 95% CI 1.45-10.69). Pulmonary TB dominated (84%) followed by TB lymphadenitis (19%). Treatment success was 94%, comparable to TB patients in general.

Conclusion: Several RA-associated risk factors may contribute to increased TB risk in bio-naïve RA patients (treatment with leflunomide, azathioprine, or prednisolone and concomitant obstructive lung disease). We could not confirm previous findings of an association with the use of moderate to high doses of prednisolone (≥15 mg). TB risk seems difficult to predict with precision in the individual bio-naïve patient based on RA-associated risk factors. To further decrease the TB risk in RA patients TB screening should also be considered in the group of bio-naïve patients.