Background/Purpose

Tuberculosis (TB) is a major concern in patients receiving TNF inhibitors (TNFi), especially in countries with a high TB burden.  Careful TB screening is mandatory before TNFi initiation, and patients with latent TB receive prophylactic treatment. This study aimed to assess the incidence of active TB and the efficacy of TB prevention measures in a cohort of patients with rheumatoid arthritis (RA), psoriatic arthritis (PsA) and ankylosing spondylitis (AS)receiving TNFi, in an academic Rheumatology center.

Methods

Data of all patients who received treatment with TNFi in our clinic between Jan 1^st2001 and June 1^st2014 have been retrospectively analyzed. Demographics, baseline TB screening, TB prophylaxis, duration of TNFi treatment and reported active TB data were extracted from clinical records. The cohort was divided into 3 groups according to the mandatory TB screening method at baseline: tuberculin skin test (TST) with a positive threshold of either >10mm (01.2001-12.2005) = group TST1, or >5mm (01.2006-08.2010) = group TST2, and QuantiFERON^®-TB Gold test (09.2010-06.2014) = group QTF. All patients tested positive for latent TB received prophylaxis with isoniazide for 9 months. Cases of TB reactivation were defined as TB occurring at <12 months since TNFi initiation. The incidence of active TB was analyzed for each group and compared to TB incidence data in Romanian general population, retrieved from the World Health Organization (WHO) reports. The impact of the screening method on the incidence of active TB was assessed using Cox proportional hazard regression.

Results

A total of 550 patients were included (304 RA, 43 PsA and 203 AS). According to the initial TB screening method, 77 patients belonged to the TST1, 251 to the TST2 and 222 to the QTF group. The number of active TB cases/ time of exposure to TNFi in person-years (PY) in the 3 groups was 9/477.3, 9/1081.5 and 3/445.5 respectively, accounting for an incidence of 1885.7, 832.2 and 673.4 cases per 100,000 PY. These numbers are significantly higher than the incidence of TB in Romania,which decreased from 147 to 94 per 100,000 PY from 2001 to 2013.

Cases classified as latent TB reactivation/total TB cases in the TST1, TST2 and QTF groups were only 2/9, 2/9 and 1/3 respectively, while the remaining TB cases were more likely new TB infection. Using Cox regression adjusted for age, sex and disease, we found no influence of the TB screening method on the risk of latent TB reactivation. However, the TST1 group had a significantly higher TB risk, with the QTF group as reference, when all cases of active TB were analyzed: hazard ratio [95% confidence interval] were 9.33[1.64-53.10], p=0.012 for TST1 and 1.68 [0.43-6.59], p=0,445 for TST2, suggesting that the period of TNFi initiation in TST1, when the TB incidence in general population was higher than in the later years, determined a higher risk for active TB.

Conclusion In a country with high TB burden, where all patients initiated on TNFi are screened for latent TB at baseline, new TB infection exceeds latent TB reactivation. TB incidence in these patients is much higher than in the general population and baseline screening does not solve the problem of later infection, suggesting the necessity of yearly TB re-screening.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/risk-of-active-tuberculosis-in-patients-with-arthritis-receiving-tnf-%ce%b1-inhibitors-a-look-beyond-the-baseline-tuberculosis-screening-protocol/