Uveitis accounts for 15% of the causes of legal blindness. The etiological diagnosis of uveitis is difficult because of the poor bacteriological performance of aqueous or vitreous fluid analysis. At the end of a medical and paramedical check-up, oriented by the typology of uveitis, a clinical situation is frequently encountered: idiopathic uveitis with a Quantiferon test (QFN) positive orienting to an old or recent contact with tuberculosis. Ocular tuberculosis is often characterized by a partial and transient response to corticosteroid therapy (local or general), due to predominant hypersensitivity phenomena and low inoculum. Therefore, antitubercular treatment is recommended for idiopathic posterior uveitis with positive QFN. This treatment of 6-9 months has shown, in combination with systemic corticosteroids, its effectiveness on ocular inflammation and significant decrease in recurrence frequency.

For previous uveitis with QFN positive, there is no study or recommendation in the low endemic countries on the indication of anti-tuberculosis drugs and practices are variable.

Tuberculous anterior uveitis is distinguished by high rate of relapses and chronic uveitis upon discontinuation of topic corticosteroid therapy that exposes to broad posterior synechiae leading to an ocular functional impairment. Optimizing the management of recurrent anterior uveitis is therefore crucial.

The aim of this prospective, randomized, controlled, open, two parallel arm trial is to compared antitubercular treatment "add-on "of local corticosteroid therapy to Local Corticosteroid Therapy Only in patients with recurrent or chronic anterior uveitis.

Primary outcome is the treatment succes defined as uveitis recovery at 3 months and the absence of recurrence at 18 months of follow-up.

Inclusion Criteria:

1. Age≥18 years.

2. For women of childbearing age (unless confirmed postmenopausal or sterile), βHCG negative.

3. For subjects of childbearing age, the willingness to use adequate contraceptive measures to prevent the subject or partner of the subject from becoming pregnant during the first 6 months of the study in case of randomization in the experimental group.

4. Recurrent anterior Uveitis (≥ 2 episodes of ocular inflammation within the past 2 years before inclusion with a free-interval of at least 3 months between ocular inflammations, patients with a second episode of ocular inflammation may be included in the study) or chronic anterior Uveitis (persistence of ocular inflammation = partial response after 3 months of well-conducted local treatment) .

5. Positive Quantiferon test (QFN) performed after the first episode of ocular inflammation (accepted tests: Quantiferon-TB-Gold, Quantiferon-TB-Gold in tube or Quantiferon plus) with a threshold ≥ 1 IU / ml or associated with a positive ELISPOT test if the QFN level is between 0.7 and 1UI / l.

6. Absence of other etiology that may explain anterior uveitis during etiological investigations

   1. Serology of herpes group viruses (HSV,, CMV, VZV) negative or old immunity (achieved after the first episode of ocular inflammation).
   2. TPHA, negative VDRL (performed after the 1st episode of ocular inflammation).
   3. Serologies HIV, HBV and HCV, negative (performed within the 3 months before inclusion).
   4. Negative Lyme serology (performed after the first episode of ocular inflammation) or medical history not supporting this etiology
   5. HLA B27 negative (achieved after the first episode of ocular inflammation) if recurrent or non-granulomatous uveitis
   6. Negative PCR from anterior chamber fluid for Herpes group viruses, Toxoplasma gondii and Mycobacterium tuberculosis if severe inflammation (Tyndall Cellular and / or Flare> 2+) and / or posterior synechiae .
   7. Non-contributory pulmonary imaging (performed within the last month before inclusion) (radiography or chest CT scan left to the discretion of the clinician).

   Note: The non-granulomatous character uveitis during clinical examination is not an exclusion criterion.

7. If 4+ severity score (Tyndall and / or Flare of aqueous humor) an expert opinion is required (internist / ophthalmologist pair): with no indication to initiate an anti-tuberculosis treatment without delay.

8. Signature of informed consent to participate in the study.

9. Patients affiliated to the French health care insurance

Exclusion Criteria:

1. Weight strictly less than 50 kg

2. Weight strictly greater than 185 kg

3. History of cancer 5 years before inclusion (except in situ cervical cancer or non-metastatic baso or squamous cell carcinoma) or progressive malignant hemopathy.

4. Liver failure or ALTgreater than three times the normal value or severe renal impairment (GFR <30ml / min).

5. Neutropenia <1000 / mm3, Thrombocytopenia <50,000 / mm3, Hemoglobin <8g / dL

6. Pulmonary or active visceral tuberculosis.

7. Associated posterior and intermediate uveitis (indication for almost constant systemic corticosteroid therapy, and de facto contraindication to a control arm without TB treatment).

8. Monophthalmic patient

9. Intervention with general anesthesia during the first 6 months

10. Clinical presentation of acute anterior uveitis type HLA B27.

11. History of tuberculous disease treated.

12. Systemic corticosteroid therapy or immunosuppressive therapy received within 3 months before inclusion.

13. Local corticotherapy received for more than 15 days in the 2 months before inclusion.

14. Hypersensitivity to the family of rifamycin, isoniazid, pyrazinamide and known ethambutol or to any of the excipients present in the medicinal products of this trial (presence, in particular, of excipients with known effect: sucrose, sodium)

15. Known hypersensitivity to fluorometholone or any of the excipients, in particular with benzalkonium chloride.

16. Known hypersensitivity to dexamethasone phosphate or to any of the excipients

17. Known hypersensitivity to tropicamide, atropine or its derivatives,

18. Known hypersensitivity to phenylephrine, thiomersal

19. Antecedent of optic neuritis.

20. Patients with wheat allergy (other than celiac disease).

21. Association with praziquantel, voriconazole, which cannot be interrupted for clinical research study.

22. Porphyries known.

23. Patient under Valaciclovir

24. Hyperuricemic subjects with symptomatic joint involvement

25. Eye infections not controlled by antiinfectives, such as:

    • acute purulent bacterial infections, including Pseudomonas and Mycobacteria infections,
    • fungal infections,
    • epithelial keratitis due to Herpes simplex virus (dendritic keratitis), vaccinia virus, varicella zoster virus and most other viral infections of the cornea and conjunctiva,
    • amoebic keratitis,

26. Perforation, ulceration and corneal injury associated with incomplete reepithelialization

27. Known ocular hypertension caused by glucocorticoids, risk of angle closure glaucoma,

28. Pregnancy or breastfeeding.

29. Psychiatric disorder and / or patient under guardianship.