CASE

In June 2020, a previously healthy 11-year-old boy presented to the eye clinic in Bari, Italy with visual loss in his left eye. He was fully immunized and denied travel outside Italy. He lived in an industrial town in Southern Italy. There was a family history of systemic lupus erythematosus, thyroiditis, and asthma. He reported blunt trauma to his left eye while playing football a few days earlier. There was no history of weight loss, night sweats, headaches, or any neurological symptoms. The child appeared well, and his body weight was 30 kg. Ophthalmological examination showed corrected visual acuity in the right eye of 20/20 while corrected visual acuity in the left eye was 4/20; intraocular pressure was 16 mmHg bilaterally. The left eye had a responsive pupil and transparent lens but slight cellularity in the anterior chamber and inflammation in the pars plana (with snowballs and dense snowbanking). Both eyes had vitritis, though only slight on the right. Fundus examination showed a normal nerve head and no macular edema but peripheral vascular sheathing. The remainder of the physical examination was normal. An initial diagnosis of intermediate uveitis of the left eye was made.

Fluorescein angiography showed mild optic nerve head leakage in both eyes, but extensive staining and late leakage of the retinal vessels in the left eye, whereas the right eye had only discrete fluorescein leakage, indicating chronic inflammation. Indocyanine green angiography (ICG), used for the study of choroidal vasculature, showed no sign of choroidal inflammation. Macular optical CT showed a normal macular profile and thickness in both eyes. B-scan ultrasonography showed bilateral vitritis, more evident in the left eye. In addition, the patient’s diagnostic work-up included the following tests: HLA B51, autoimmunity antibody panel (anti-nuclear, anti-dsDNA, ENA, and ANCA antibodies), Widal-Wright, fluorescent treponemal antibody absorption (FTA-ABS), ACE level, and antibodies against Bartonella henselae, Borrelia burgdorferi, toxocariasis, toxoplasma, herpes simplex virus, rubella, cytomegalovirus, mumps virus, and varicella-zoster virus. All these tests were negative.

He was discharged with the diagnosis of uveitis of both eyes (left>right) and was treated with prednisone, 12.5 mg per day, with little improvement. An additional test was performed which pointed to the underlying infectious cause.

DENOUEMENT

Interferon-gamma release assay (IGRA) for tuberculosis was positive. Thus, the patient was sent to the Pediatric Infectious Diseases Unit for further management. No specific risk factors or exposure for TB were identified. Tuberculin skin test (TST) performed while on prednisone was negative. Additional TB diagnostic evaluation was completed including chest radiograph and CT, brain MRI, and bronchoscopy with bronchoalveolar lavage, all of which were negative for TB involvement. Based on the bilateral uveitis and positive IGRA, a final diagnosis of ocular tuberculosis (OTB) was made. Of note, his mother’s TST was positive, but she had no evidence of active disease so was diagnosed with latent TB infection. The remainder of the patient’s household had negative TSTs. No source case was identified.

Multi-drug anti-TB therapy with isoniazid (300 mg per day), rifampicin (600 mg per day), and pyrazinamide (750 mg per day) was started. Prednisone (12.5 mg per day) was continued, along with lansoprazole (10 mg per day) and intraocular nepafenac. On day 15 of anti-TB therapy, the boy was discharged with no major change in his visual acuity or his ocular findings.

Anti-TB therapy was continued for one year. Attempts at discontinuation of corticosteroid therapy led to worsening of the visual acuity; therefore, steroids were progressively tapered beginning at month 4 of therapy until discontinuation after 1 year. With this treatment, he had clear improvement by 3 months of therapy and complete resolution at 10 months and the patient’s best corrected visual acuity in both eyes was 20/20 after 1 year. The vitreous was clear, and fluorescein angiography only showed slight leakage phenomena in the peripheral lower retinal sectors.

As in adults, pediatric uveitis incidence, prevalence, and etiology are affected by geographical variations and ethnicity.1 About 10% of uveitis occurs in children under 16. The yearly incidence of pediatric uveitis is about 4-6 cases per 100,000, and the prevalence rate is about 27 cases per 100,000. Most cases (70-90%) of uveitis in children are of noninfectious origin, and only a minority are due to infectious causes.2 Infectious pathogens associated with uveitis include herpes viruses (HSV, VZV, CMV), parasites (protozoa, nematodes, cestodes, trematodes, and ectoparasites), syphilis, toxoplasmosis, toxocariasis as well as tuberculosis, which is a great mimicker of all the various uveitis entities.3

According to 2019 data on tuberculosis from European Center for Disease Prevention and Control and WHO Europe, Italy has a very low TB incidence of <20/100,000.4 The mean age of new cases was 52.9 years for cases reported among native Italians and 37.2 years for foreign people; 60.6% of cases occurred in adults between 25 and 64 years of age, 19.1% among those >64 years, and 14.8% among those 15-24 years of age. There were 80 cases (2.5%) among children under 5 years of age, and 96 cases (3.0%) among children between 5 and 14 years of age.5 Ocular TB (OTB) is very unusual, especially in children, and few cases have been reported in Italy.

Almost all intraocular segments may be involved in OTB, and patients may develop acute or chronic inflammation, either unilateral or bilateral. Anterior uveitis is usually granulomatous, with posterior synechiae, Koeppe and Busacca nodules, mutton-fat keratic precipitates, and/or complicated cataracts. In children, band keratopathy can occur. Intermediate uveitis appears like pars planitis. Features include vitritis with snowball, snow banking, peripheral vascular sheathing, and/or peripheral retino-choroidal granulomas. Posterior uveitis is the most common presentation of OTB with solitary or multiple tubercles, miliary choroidal tubercles, or multifocal choroiditis that may progress to a serpiginous-like lesion. Retinal lesions may take the form of either focal tubercles, subretinal abscesses, or diffuse retinitis with retinal hemorrhagic vasculitis. Exudative retinal hemorrhagic periphlebitis with uveitis is highly suggestive of OTB.

While OTB is an extrapulmonary form of TB, in most patients, clinical manifestations at other sites are absent6,7 as in the present pediatric case. Due to the variety of ocular manifestations, a clinical diagnosis of OTB is quite challenging and requires a complete history, physical examination, and fundoscopy. In most cases, a presumptive diagnosis is made based on epidemiological factors, clinical manifestations, and corroborating immunologic evidence. Loss of visual acuity is the most frequent ophthalmological symptom (87.5%); posterior uveitis is the most frequent localization (72.9%). Gupta et al.6 identified broad-based posterior synechiae, retinal vasculitis without choroiditis, retinal vasculitis with choroiditis, and serpiginous-like choroiditis as features of OTB with specificities of 93%, 97%, 99%, and 98%, respectively. However, all these features have poor sensitivity. Other highly suspicious lesions include choroidal granulomas (granulomatous uveitis), multifocal serpiginous choroiditis, occlusive retinal periphlebitis, or vasculitis.6 Confirmation of the diagnosis of OTB is difficult as direct microscopic examination of lesional tissue to find Mycobacterium tuberculosis (MTb) is impractical, and ocular manifestations may also represent a delayed hypersensitivity reaction rather than direct infection.8,9 Polymerase chain reaction (PCR) is considered a rapid method for detecting infectious causes of uveitis, including MTb,10 but the sensitivity and specificity of PCR using ocular samples remains uncertain.11 Diagnosis of tuberculous uveitis (TBU) is also usually made after excluding other causes of uveitis.

TBU can cause permanent damage to visual acuity, particularly in patients with a delayed diagnosis. Therefore, prompt initiation of systemic TB treatment is essential to improve prognosis as significant delays are associated with increased morbidity.3 Typically, there is improvement in visual acuity in 74.4% of eyes, but complete response is achieved in only 56.4%.3,12

The Collaborative Ocular Tuberculosis Study 1 (COTS-1) represents the largest collaborative multicenter dataset of ocular TB, including patients from both endemic and nonendemic countries. The diagnostic and inclusion criteria defined by COTS-1 include (1) clinical signs suggestive of ocular tuberculosis and (2) exclusion of other uveitis entities where relevant based on clinical manifestations of disease and regional epidemiology, plus either investigations documenting the mycobacteria or its genome or corroborative investigations such as positive TST, positive IGRA, or evidence of healed or active tuberculosis on chest radiography. The COTS-1 Consensus Group also proposed criteria for treatment13 based on TST, IGRA, and radiology results, type of uveitis (intermediate versus pan-uveitis), and endemic versus nonendemic area. In all scenarios, radiologic evidence of TB disease is required before treatment is recommended. It is interesting to note that based on the inclusion criteria, our patient qualified for the diagnosis of OTB; however, he did not meet the treatment criteria for antitubercular therapy because his chest radiograph was normal. Therefore, strictly following these criteria could have resulted in the inadvertent withholding of specific TB therapy in our patient, which could have had negative prognostic implications. Thus, these guidelines should be applied with caution, particularly for pediatric patients.14

REFERENCES 1. Testi I, Agrawal R, Mahajan S, et al. The Collaborative Ocular Tuberculosis Study (COTS)-1: a multinational descriptive review of tubercular uveitis in paediatric population. Ocular Immunol Inflamm. 2020;28(sup1):58–64. 2. Tugal-Tutkun I. Pediatric Uveitis. J Ophthalmic Vis Res. 2011;6:259–269. 3. Anibarro L, Cortés E, Chouza A, et al. Early treatment of tuberculous uveitis improves visual outcome: a 10-year cohort study. Infection. 2018;46:549–554. 4. Global tuberculosis report 2019. World Health Organization; 2019. Available at: https://www.ecdc.europa.eu/en/publications-data/presentation-tuberculosis-situation-eueea-2019. Accessed August 17, 2021. 6. Gupta A, Gupta V, Bambery P, et al. Diagnosis and management of intraocular tuberculosis. Natl Acad Med Sci (India). 2010;46:164–175. 7. Distia Nora R, van Velthoven ME, Ten Dam-van Loon NH, et al. Clinical manifestations of patients with intraocular inflammation and positive QuantiFERON-TB gold in-tube test in a country nonendemic for tuberculosis. Am J Ophthalmol. 2014;157:754–761. 8. Petrushkin H, Sethi C, Potter J, et al. Developing a pathway for the diagnosis and management of ocular tuberculosis. The pan-London Ocular tuberculosis Pathway-LOOP. Eye (Lond). 2020;34:805–808. 9. Agarwal A, Agrawal R, Raje D, et al. Twenty-four month outcomes in the Collaborative Ocular Tuberculosis Study (COTS)-1: defining the “Cure” in ocular tuberculosis. Ocul Immunol Inflamm. 2019;27:465–473. 10. Mochizuki M, Sugita S, Kamoi K, et al. A new era of uveitis: impact of polymerase chain reaction in intraocular inflammatory diseases. Jpn J Ophthalmol. 2017;61:1–20. 11. La Distia NR, Putera I, Khalisha DF, et al. The diagnostic value of polymerase chain reaction for ocular tuberculosis diagnosis in relation to antitubercular therapy response: a meta-analysis. Int J Infect Dis. 2021;110:394–402. 12. Patel SS, Saraiya NV, Tessler HH, et al. Mycobacterial ocular inflammation: delay in diagnosis and other factors impacting morbidity. JAMA Ophthalmol. 2013;131:752–758. 13. Agrawal R, Testi I, Bodaghi B, et al.; Collaborative Ocular Tuberculosis Study Consensus Group. Collaborative ocular tuberculosis study consensus guidelines on the management of tubercular uveitis-report 2: guidelines for initiating antitubercular therapy in anterior uveitis, intermediate uveitis, panuveitis, and retinal vasculitis. Ophthalmology. 2021;128:277–287. 14. Agrawal T, Multani P, Shetty SB, et al. Bilateral ocular tuberculosis in a child with negative TuberculiSkin Test (TST): a diagnostic dilemma. Ocul Immunol Inflamm. 2022;30:1199–1202.