Normalised quantitative polymerase chain reaction for diagnosis of tuberculosis-associated uveitis.

Polymerase chain reaction (PCR)-based diagnosis of tuberculosis-associated uveitis (TBU) in TB-endemic countries is challenging due to likelihood of latent mycobacterial infection in both immune and non-immune cells. In this study, we investigated normalised quantitative PCR (nqPCR) in ocular fluids (aqueous/vitreous) for diagnosis of TBU in a TB-endemic population. Mycobacterial copy numbers (mpb64 gene) were normalised to host genome copy numbers (RNAse P RNA component H1 [RPPH1] gene) in TBU (n = 16) and control (n = 13) samples (discovery cohort). The mpb64:RPPH1 ratios (normalised value) from each TBU and control sample were tested against the current reference standard i.e. clinically-diagnosed TBU, to generate Receiver Operating Characteristic (ROC) curves. The optimum cut-off value of mpb64:RPPH1 ratio (0.011) for diagnosing TBU was identified from the highest Youden index. This cut-off value was then tested in a different cohort of TBU and controls (validation cohort, 20 cases and 18 controls), where it yielded specificity, sensitivity and diagnostic accuracy of 94.4%, 85.0%, and 89.4% respectively. The above values for conventional quantitative PCR (≥1 copy of mpb64 per reaction) were 61.1%, 90.0%, and 74.3% respectively. Normalisation markedly improved the specificity and diagnostic accuracy of quantitative PCR for diagnosis of TBU.

Effect of Adjunctive Pars Plana Vitrectomy on Focal Posterior Segment Inflammation: A Case-Control Study in Tuberculosis-Associated Uveitis.

PURPOSE: Pars plana vitrectomy (PPV) is commonly performed for managing complications of uveitis but the anti-inflammatory potential of PPV has not been extensively investigated beyond aqueous/vitreous inflammation. We studied the effect of PPV on resolution of focal posterior segment lesions in tuberculosis-associated uveitis (TBU). DESIGN: Case control study. PARTICIPANTS: Patients with bilateral TBU and active retinal/choroidal lesions in both eyes, and who received PPV in one eye were included. Fellow eyes of same patients, matched for patient characteristics and systemic therapy, were designated as controls. METHODS: Study eyes received 3-port 23-guage PPV, involving removal of nearly the entire vitreous. Part of vitreous sample was used for quantitative polymerase chain reaction (qPCR) for Mycobacterium tuberculosis. Post-operatively, anti-TB and/or systemic corticosteroid therapy was initiated depending on level of clinical suspicion of tubercular etiology, degree of intraocular inflammation and qPCR results. Focal lesions were documented in preoperative and postoperative fundus diagrams. Clinical photographs were taken whenever adequate media clarity was present. MAIN OUTCOME MEASURES: Primary outcome measures were rate of clinical resolution of focal posterior segment lesions and improvement in best-corrected visual acuity (BCVA), at 1 month post-surgery. RESULTS: Thirty-six patients with bilateral posterior segment lesions consistent with TBU were included. Possible and probable TBU (depending on radiographic evidence of TB) were diagnosed in 28 (77.7%) and 4 (11.1%) patients respectively, whereas remaining 4 patients were diagnosed only on basis of qPCR results. Focal posterior segment lesions included retinal vasculitis (n = 27), multifocal-serpigenoid choroiditis (n = 5), multifocal choroiditis (n = 3) and focal choroiditis (n = 1). At one month postvitrectomy, 28 eyes (73.7%) showed complete resolution of focal posterior segment lesions compared to 7 non-vitrectomised eyes (19.4%), while improvement in BCVA was significantly more in study eyes (0.38 logarithm of the minimum angle of resolution [logMAR], P = 0.04), compared to controls (0.12 logMAR, P = 0.17). Time to resolution following vitrectomy was unaffected by duration of disease, pre-operative systemic steroids or grade of vitritis. At 3 months, complete resolution was noted in 29 of 30 study eyes (96.7%) and 25 of 30 control eyes (83.3%). CONCLUSIONS: PPV facilitates faster resolution of focal posterior segment inflammation and BCVA improvement in TBU.