Presumed tuberculosis-associated uveitis: rising incidence and widening criteria for diagnosis in a non-endemic area.

PurposeTo assess the incidence, clinical ocular involvement and effectiveness of anti-tuberculous treatment in patients with chronic uveitis presumed to be associated with tuberculosis in a non-endemic community.Patients and methodsRetrospective case series of patients with uveitis and evidence of tuberculosis, with no other identified cause of uveitis, who underwent a 6-month course of standard anti-tuberculosis treatment between 2008 and 2015. The response to treatment was assessed at 6 and 12 months after initiation of treatment.ResultsForty-eight patients were included of whom 36 (75%) were born outside the United Kingdom. Only five had concurrent active pulmonary or nodal tuberculosis. There were 85 affected eyes, including 25 with granulomatous anterior uveitis, 32 with retinal vasculitis (occlusive in 21), and 20 with multifocal choroiditis or serpiginous-like retinochoroiditis. Gamma-interferon testing was positive in 95%. Complete resolution at end point was seen in only 60%, but a further 19% were inflammation-free on topical steroid only. Resolution was lower (50%) in those with panuveitis compared to other anatomical types (75%). Sixty-four eyes (75%) had a LogMAR visual acuity of 0.1 or better at the end of the study.ConclusionsThe incidence of presumed tuberculosis-associated uveitis (TBU) has almost quadrupled in this region. The efficacy of treatment has not been enhanced by the introduction of gamma-interferon testing to support diagnosis. Some patients may require more prolonged antibiotic therapy to ensure quiescence, but chronic non-infective anterior uveitis may in any case follow treated TBU.