The diagnosis of ocular tuberculosis remains a challenge in the absence of a definitive diagnostic criterion, even in the immunocompetent patients. Ocular tuberculosis does not have a typical or characteristic clinical picture with which it presents. Most other causes of uveitis may also present with similar clinical picture. Also, the definitive diagnosis of ocular tuberculosis requires microbiological isolation of the organism from the tissue or body fluid, which can be difficult and highly invasive in case of ocular disease [77]. Another point to note is that ocular tuberculosis may often exist without other systemic manifestations or foci of infection; thus, it is not possible to rule it out completely if systemic investigations like radiological imaging and sputum exam are normal [78].

Gupta et al. enumerated the following clinical signs as consistent with intraocular tuberculosis in the immunocompetent patients, namely, broad posterior synechiae , retinal perivasculitis with or without discrete choroiditis or scars, multifocal serpiginous choroiditis, single or multiple choroidal granuloma, optic disc granuloma, and optic neuropathy . Among these broad posterior synechiae, retinal vasculitis with or without choroiditis or scars and serpiginous-like choroiditis were shown to have the higher specificity for diagnosis. They have also proposed a classification system for intraocular tuberculosis (IOTB) which classifies the disease in confirmed, possible, and probable IOTB [79].

There is a lack of consensus on the guidelines to diagnose and treat ocular tuberculosis, including the use of immunosuppressive agents. However, with rise in the multidrug-resistant cases of tuberculosis, it has become imperative to make an accurate and early diagnosis of the condition to avoid inappropriate treatment [80–82]. Currently, diagnosis of ocular tuberculosis is made on the basis of ocular lesions, evaluation of intraocular fluid, evidence of tuberculosis elsewhere in the body, and response to antitubercular treatment. In immunocompromised patients, ocular lesions may precede other systemic findings, and hence, detecting them may be critical to early diagnosis of the disease and commencement of treatment [83]. Traditionally, systemic tuberculosis was diagnosed on the basis of tests like chest X-ray, Mantoux tuberculin skin sensitivity test, sputum examination, etc. However, these do not work well for ocular tuberculosis, especially in the immunocompromised. Tests like Mantoux test may be negative due to altered immune status, and chest X-ray may be normal since these present many times as extra-pulmonary tuberculosis. Also, being a paucibacillary infection, the ocular samples are rarely positive in microbial culture tests. Therefore, detection of the ocular lesions clinically becomes very crucial [62, 83–85].

The newer tests to diagnose tuberculosis include the immunological tests such as interferon gamma release assay (IGRA) , a test which measures the IFN-gamma levels, and the polymerase chain reaction which detects the mycobacterial genomic material. IGRA, unlike tuberculin sensitivity test, is not affected by the BCG vaccine status. Commonly used IGRA tests include QuantiFERON-TB Gold (QFT-G) test and T-SPOT.TB test. QFT-G test has been found to be useful in detecting intraocular tuberculosis with good sensitivity and may also be helpful in deciding to start steroid therapy in uveitis patients. It also has the advantage of being able to detect latent tuberculosis in BCG-vaccinated patients with fewer false positives as compared to Mantoux tuberculin skin sensitivity test (TST) . However, it has been noted to have a lower specificity in the immunocompromised as compared to the immunocompetent [86–89]. QFT-G test has a higher sensitivity and specificity for detection of ocular tuberculosis as compared to systemic disease [90]. In fact, QFT-G test may be diagnostically superior to tuberculin skin sensitivity test especially in the immunocompromised patients, though some claim the difference between the two to be only marginal [91–93].

However, there have been studies which have reported that QFT-G test may be indeterminate in patients with immunosuppression [94]. In these patients, the T-SPOT.TB test may show better sensitivity to detect the disease [95]. T-SPOT.TB test relies on quantitating the IFN-gamma-releasing T cells, rather than the cytokine levels directly [96]. The T-SPOT.TB test has been found to be a useful adjunct in the diagnosis of ocular tuberculosis in non-endemic areas [97]. Ariga et al. found IFN-gamma release to correlate with peripheral blood T cell population, which may be altered in an immunocompromised patient. They also prescribed a lower cut-off value for the IFN-gamma levels for diagnosing these patients in the low endemic settings [98]. In fact, higher levels of interferon gamma on this test have been found to correlate with successful response to treatment [99]. Interferon gamma assay has been prescribed for aiding the diagnosis of ocular tuberculosis in conjunction with the clinical features and positive TST [100, 101]. Recent studies have shown QFT-G test to be superior to T-SPOT.TB test for diagnosing ocular tuberculosis [102].

PCR has been known to detect mycobacterial gene from aqueous and vitreous samples with good specificity [103]. Babu et al. used PCR to diagnose ocular tuberculosis successfully in HIV-positive patients [29]. The newer quantitative PCR which allows real-time assessment of the process and enables better quantification of genetic material has also been used to diagnose ocular tuberculosis [104, 105]. A newer nucleic acid amplification test called the loop-mediated isothermal amplification assay PCR (LAMP PCR) has also been found to have a very good sensitivity and specificity in detecting mycobacterial genome in uveitis patients and may be an alternative modality to the conventional PCR [106]. Recently, Rishi et al. reported the use of RT-PCR to demonstrate Mycobacterium tuberculosis RNA in patients with culture-positive tubercular endogenous endophthalmitis. They emphasized that in contrast to the microbial culture method which takes at least 2–3 weeks to provide results, RT-PCR can provide evidence of the microbe in a much shorter time [107].

Currently, most ophthalmologists treat ocular tuberculosis with a standard ATT regime of 9–12 months, depending on other systemic involvements. The use of steroids along with ATT has also been prescribed to control the inflammation especially in the acute phase, and it has been shown to be beneficial in reducing the damage due to immune hypersensitivity. The use of steroids is not advisable without concurrent ATT. However, in immunocompromised patients, a risk of worsening of the tubercular inflammation is seen when HAART and ATT are started together. This can often lead to panophthalmitis and thus the loss of the eye. To avoid this, it is suggested that the HAART be started a few weeks after initiating ATT [83, 108, 109]. Shorter duration of ATT and concurrent use of immunosuppressives have been associated with higher failure rates. Other factors like non-compliance and drug resistance affect the treatment outcomes in an adverse manner [80, 110]. The use of ATT for cases of tubercular uveitis has also been associated with a reduced rate of disease recurrence [111].

The diagnosis and management of ocular tuberculosis remains a clinical challenge in the absence of standardized criteria, even in the immunocompetent. In immunocompromised patients, the clinical ocular features are less typical and may deteriorate faster than the immunocompetent patients if early diagnosis is not made. Other factors like immune recovery uveitis may also lead to a rapid worsening of the disease. Diagnostic tests like TST and radiological imaging are often inconclusive in these patients making the diagnosis even more challenging. Thus, early diagnosis based on clinical features such as choroidal lesions may help in the timely diagnosis and appropriate treatment, thus preventing complications.