Serpiginoid choroiditis associated with presumed ocular tuberculosis.

The purpose of this study is to present a case with serpiginoid choroiditis with possible ocular tuberculosis. The intraocular inflammation and choroiditis were successfully controlled by systemic antituberculosis treatment. A 63-year-old female presented with progressive bilateral blurred vision for over a year. At presentation, her best-corrected visual acuity was 6/20 in her right eye and counting fingers at 10 cm in her left eye. A fundus examination showed diffuse patchy geographic retinal pigment epithelium (RPE) changes with some pigmentation in both eyes. Fluorescein angiography disclosed leakage from RPE lesions and discs as well as retinal vasculitis. Systemic survey results for rheumatic and infectious diseases were normal except for a positive QuantiFERON-TB Gold test result. Her uveitis improved and chorioretinal lesions stabilized from the 2nd month of antituberculosis treatment. The antituberculosis treatment was discontinued after a 12-month course. No recurrence of uveitis was noted during the following 2 months. Diagnosing ocular tuberculosis is challenging. The clinical presentation, interferon-gamma release assay test, and clinical response to antituberculosis therapy can support a presumed diagnosis of tubercular uveitis. This case highlights that serpiginoid choroiditis can be a clinical presentation of ocular tuberculosis. Clinicians should pay attention to this etiology when facing a serpiginous-like retinal appearance.

Introduction

Ocular features of tubercular uveitis are diverse. Broad-based posterior synechiae, anterior chamber cells, vitritis, and retinal vasculitis with or without choroiditis are common intraocular signs in patients with tubercular uveitis.[1] However, a diagnosis of tubercular uveitis is often made clinically as the obtained ocular tissues are insufficient to make an accurate diagnosis in most cases. A presumed diagnosis is made when ocular findings and systemic findings are consistent with tuberculosis. Interpretation of interferon-gamma release assay (IGRA) together with tuberculin skin test (TST) has shown high accuracy for diagnosing tubercular uveitis.[2] Furthermore, a presumed diagnosis can be supported by clinical response to antituberculosis therapy (ATT).

In this report, we present a case of ocular tuberculosis who had negative polymerase chain reaction (PCR) findings for tuberculosis DNA in the aqueous but a positive QuantiFERON-TB Gold test result. The intraocular inflammation and choroiditis were eventually well controlled after 12 months of ATT.

Case Report

A 63-year-old female presented with progressive bilateral blurred vision for over a year. She did not have a medical history of ocular trauma, surgery, hypertension, or diabetes mellitus. Systemic diseases including rheumatic disease, infections, and intraocular lymphoma were ruled out based on blood tests, chest X-rays, and a vitreous biopsy at another hospital. Serpiginous choroiditis was diagnosed for which systemic corticosteroids and cyclosporine were initially prescribed. However, symptoms including blurred vision and visual field defect progressed in the following year despite these treatments.

At the first presentation at our clinic, her best-corrected visual acuity (BCVA) was 6/20 in her right eye and counting fingers at 10 cm in her left eye, with intraocular pressures of 22 and 21 mmHg. No keratic precipitates were noted on her cornea. No inflammatory cells were noted; however, 0.5+ flare was noted in her right anterior chamber. The vitreous in her left eye was clear, but 0.5+ vitreous cells and grade 1 vitreous haze were found in her right eye. A fundus examination showed diffuse patchy geographic retinal pigment epithelium (RPE) changes with some pigmentation in both eyes. Some para-vascular whitish exudates and hemorrhage could also be seen [Figure 1]. Fundus autofluorescent photos showed diffuse hypo-autofluorescent lesions, some of which were heterogeneous with hyper-autofluorescent lesions. These lesions were compatible with the chorioretinitis lesions seen in fundoscopic examinations. Fluorescein angiography revealed leakage from RPE lesions and discs as well as retinal vasculitis in both eyes [Figure 2]. Disruptions of the ellipsoid zone and RPE as well as some sub-RPE lesions were noted by optical coherence tomography in both eyes, and the condition was worse in the left eye than in the right eye [Figure 3].

Figure 1

(a and b) Funduscopic examination showed diffuse patchy geographic retinal pigment epithelium changes with some pigmentation in both eyes. Some para-vascular whitish exudates and hemorrhage could also be seen. (c and d) Fundus autofluorescent photos showed diffused hypo-autofluorescent lesions, some of which were heterogeneous with hyper-autofluorescent lesions (c, arrow). These lesions were compatible with the chorioretinitis lesions seen in fundoscopic examinations and became totally dark after 12 months of treatment (d, arrow head)

Figure 2

Leakage from retinal pigment epithelium lesions and disc in both eyes. Retinal vasculitis could also be seen

Figure 3

Optical coherence tomography of the right (a) and left eye (b) showed disruptions of the ellipsoid zone and retinal pigment epithelium, as well as some subretinal pigment epithelial lesions (arrow)

Blood examinations and chest X-rays were evaluated again, but all showed negative (normal) results. A sputum examination was performed, and both acid-fast bacilli (AFB) smear and culture of Mycobacterium tuberculosis showed negative results. However, a serum QuantiFERON-TB Gold test showed a positive (2.83 IU/mL) result. Paracentesis of the anterior chamber was then performed; however, tuberculosis DNA could not be detected in PCR of the aqueous.

Despite the negative PCR test results, oral cyclosporine and prednisolone were tapered to discontinuation, and antituberculosis drugs with rifampin, isoniazid and pyrazinamide were prescribed. Two months after the initial treatment with antituberculosis drugs, the BCVA in her right eye improved to 6/10 but her left eye remained the same. Anterior chamber flares and vitreous haze in both eyes subsided, and the choroidal retinal lesions became dry and pigmented without any progression [Figure 4]. After 12 months of treatment, no anterior or vitreous cells were noted, and the ATT was discontinued. During the following of 2 months, her vitreous was clear, and choroiditis was not seen. Stabilized geographic RPE pigmentation remained in both eyes even after discontinuation of all systemic and topical medications.

Figure 4

Funduscopic examination at 2 months after antituberculosis treatment showed that all of the lesions were dry without any progression

Discussion

The manifestation of infection caused by M. tuberculosis in the eye can be variable. Ocular tuberculosis is often caused by the hematogenous spread of M. tuberculosis from pulmonary or extrapulmonary sites, and the uveal tract is most commonly affected. Anterior chamber cells, vitreous cells and opacity, and retinal vasculitis with or without choroiditis are common intraocular findings in patients with tubercular uveitis.[1] In addition, posterior synechiae, peripheral anterior synechiae and granuloma on the iris, optic nerve head or choroid can also been seen.

The ocular involvement in tuberculosis serpiginoid choroiditis is usually unilateral, typically with multifocal choroiditis in a centrifugally spreading serpiginous pattern.[3] Evidence of tuberculosis in patients presenting with multifocal serpiginoid choroiditis is increasing from both endemic and nonendemic areas.[34] Bansal et al. reported M. tuberculosis within ocular tissue of multifocal serpiginous choroiditis detected using multi-targeted PCR.[5] In another study by Bansal et al., 105 patients had active serpiginoid choroiditis and positive TST or QuantiFERON-TB Gold test results, and all 93 who received ATT had resolution of the lesions, with nine (9.7%) having recurrence. In addition, of the 12 patients who were given corticosteroids alone, none showed progression, but nine (75%) developed recurrence.[6]

The definite diagnosis of tubercular uveitis is made when ocular tissue shows positive AFB smear and culture results. Using PCR to detect the DNA of M. tuberculosis from ocular tissue supports the diagnosis of tubercular uveitis. However, PCR is highly specific but of variable sensitivity.[7] In a clinical setting, the diagnosis of uveitis is challenging, as obtaining ocular tissue for microbiologic or molecular tests is not always possible. A presumed diagnosis of tubercular uveitis is made when ocular and systemic findings are consistent with tuberculosis. Systemic findings suggestive of tuberculosis include positive AFB smear and culture results, consistent chest X-ray findings, or a positive IGRA or TST. Resolution of the lesions after the administration of ATT further supports a presumed diagnosis of ocular tuberculosis.[6]

Our patient presented with serpiginoid choroiditis with a positive QuantiFERON-TB Gold test result. Resolution of her lesions after discontinuation of systemic steroids and cyclosporine and administration of ATT further supported a diagnosis of presumed tubercular uveitis. The QuantiFERON-TB Gold test is a blood test and a commonly used IGRA. It measures the amount of interferon-gamma released by T cells after they are stimulated by antigens unique to M. tuberculosis.[8] IGRA is not influenced by tuberculosis vaccination status because the antigens used are not encoded by Mycobacterium bovis. However, as with TST, IGRA cannot distinguish between active tuberculosis and latent infection. In addition, the sensitivity of IGRA is lower in patients with human immunodeficiency virus infection due to a lower CD4 count.[9] Apart from latent tuberculosis, recent studies have reported an association between false-positive QuantiFERON-TB test results in patients with sarcoidosis.[1011] Rina proposed that sarcoidosis may be associated with elevated interferon-gamma levels in some patients with uveitis, and that this may result in false-positive results. In a previous study from India, an area endemic for tuberculosis, Gupta et al. demonstrated that patients with sarcoidosis were more likely to have a negative TST compared to those with tuberculosis (89.5% vs. 23.3%, P < 0.001).[12] This may be correlated with anergy to tuberculin in patients with sarcoidosis. To improve the accuracy of the diagnosis of tubercular uveitis, it is important to interpret QuantiFERON-TB together with a TST, as this combination has a higher accuracy than either one alone to diagnose tubercular uveitis.[2] In combination with clinical signs, radiographic and histopathological findings, a negative TST result may help to differentiate tubercular uveitis from ocular sarcoidosis. Unfortunately, our patient did not receive a TST skin test at our hospital or previous hospitals. However, two factors supported our diagnosis in this patient. First, she mainly suffered from chorioretinitis without obvious anterior or intermediate uveitis at presentation. This is different from ocular sarcoidosis, in which signs of anterior and intermediate inflammation are often seen rather than serpiginoid lesions. Second, the ineffectiveness of systemic steroids and immunosuppressive agents suggested an infectious etiology in this patient.

ATT for tubercular uveitis is the same as for pulmonary tuberculosis.[13] In addition to ATT, topical or systemic steroids can be used concomitantly, and this has been shown to reduce inflammation-induced damage compared to ATT alone.[14] Nevertheless, treatment with immunosuppressants should be given with caution, especially when the patients are at risk of having active or latent tuberculosis. Recently, anti-tumor necrosis factor-alpha has been proven to be an effective treatment for nonanterior noninfectious uveitis.[1516] However, patients treated with anti-tumor necrosis factor-alpha therapy have also been demonstrated to have a higher risk of opportunistic infection and reactivation of tuberculosis.[1617] Screening for active or latent infection of tuberculosis before initiating this therapy is highly recommended.[17]

Conclusion

Ocular manifestations of tubercular uveitis are variable. It remains challenging to make a definite diagnosis as it is not always possible to obtain ocular tissue to yield demonstrable mycobacteria from acquired aqueous or vitreous samples. A presumed diagnosis of tubercular uveitis can be made, as the ocular findings and serum IGRA findings are consistent with tuberculosis. A clinical response to ATT further supports a presumed diagnosis of tubercular uveitis.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

The authors declare that there are no conflicts of interests of this paper.