Post-infectious Uveitis

Infectious uveitis is common, particularly in referral centers and in the developing world, where up to one-third or more of all patients with ocular inflammation have an identified infectious cause for their uveitis.^1–5 In addition, less direct contributions of infections to the development or continuation of ocular inflammation can also occur.^6–10 Perhaps best recognized is the so-called Jarisch–Herxheimer Reaction (JHR), named after Adolf Jarisch and Karl Herxheimer, two dermatologists who independently described transient worsening of syphilitic skin lesions following treatment with mercury in 1895 and 1902, respectively. Today, the JHR occurs most often following treatment of syphilis with systemic penicillin.¹¹ Such paradoxical worsening of inflammation following antimicrobial treatment has been described in patients with other treponemal infections, including leptospiroisis¹² and Lyme disease,¹³ and can also occur during treatment of tuberculosis.¹⁴ Moreover, at least five mechanisms of post-infectious activation of autoreactive lymphocytes have also been hypothesized, with varying degrees of scientific support, including: (1) molecular mimicry; (2) microbial expression of lymphocyte-activating superantigens; (3) enhanced processing and presentation of autoantigens resulting in epitope spreading; (4) bystander activation; and (5) lymphocyte activation by lymphotropic viruses.¹⁰ Such theoretical possibilities notwithstanding, it is often difficult to determine with any degree of certainty whether, and to what extent, inflammation following infection may be autoimmune versus the direct effect of persistent microbes and/or their antigens. This is particularly true for chronic or latent infections, such as those caused by spirochetes, mycobacteria, and herpes viruses. To complicate matters further, Gram-negative enteric infections can trigger¹⁵ – whereas intestinal colonization with helminthic parasites may protect against¹⁶ – the development of autoimmune disease. These and other seminal observations support a growing appreciation for the modulatory role of the intestinal microbiome in the pathogenesis of uveitis.^17–19 Five original articles^20–24 and one letter to the editor²⁵ in this issue of Ocular Immunology & Inflammation (OII) directly or indirectly address issues related to post-infectious uveitis.

Filloy et al.²⁰ described an 8-year-old boy from Barcelona, Spain, who developed bilateral panuveitis with retinal vasculitis, vascular sheathing, and cystoid macular edema (CME) following acute bilateral tonsillitis treated with oral antibiotics immediately prior to presentation. Testing revealed a markedly elevated anti-streptolysin-O titer and was otherwise unremarkable. The inflammation was controlled following 1 month of treatment with topical and systemic corticosteroids, but recurred 6 weeks following discontinuation of therapy. The authors discussed previously reported cases of post-streptococcal uveitis (PSU) and emphasized that most patients have non-granulomatous anterior chamber inflammation, that up to one-third have some degree of posterior segment involvement, and that most do well with treatment. Recurrent PSU has been reported previously.²⁶ While the precise pathogenesis of PSU is unknown, molecular mimicry with induced lymphocytic cross-reactivity to normally non-immunogenic ocular antigens appears to be the mostly widely accepted mechanism.⁶

Goldhardt et al.²¹ described two patients, a 34-year-old Jamaican woman and an 8-year-old girl from Miami, both of whom developed bilateral acute posterior multifocal placoid pigment epitheliopathy (APMPPE) following serologic evidence of dengue virus infection. The Jamaican woman developed visual symptoms 3 weeks following a febrile illness believed to be due to repeat infection with a second serovar of dengue, whereas the young girl from Miami was found to have APMPPE-like lesions nearly 6 years following what were suspected to have been multiple dengue virus infections. In the younger patient, past ocular history was complicated by what the authors described as consecutive episodes of multiple evanescent white dot syndrome (MEWDS) that produced scarring. While chorioretinal scarring remained, inflammation resolved in both patients following treatment with systemic corticosteroids. The authors suggested that inflammation and occlusion of the choriocapillaris of the type proposed to produce acute posterior multifocal placoid pigment epitheliopathy (APMPPE)²⁷ may occur following dengue infection. Of note, acute macular neuromaculopathy (AMN) has also been described following dengue infection,²⁸ a syndrome similarly linked to localized occlusion of the choriocapillaris.²⁹

Bajgai et al. used²² a polymerase chain reaction (PCR)-based assay directed against three Mycobacterium tuberculosis (MTB) genes (IS6110, MPB6, and protein b) to test whether MTB DNA was present in subretinal fluid, and by inference in retinal pigment epithelial (RPE) cells contained therein, from 16 individuals with latent tuberculosis (TB) who underwent pars plana vitrectomy to repair a rhegmatogenous retinal detachment (RRD). Six (37.5%) were found to have evidence of MTB DNA in the collected subretinal fluid specimens. A total of 84 patients with RRD and no evidence of latent TB infection, none of which contained MTB DNA on PCR, served as negative controls. While others have shown MTB DNA in clinically and histologically normal lung and adipose tissue from patients with latent TB,^30,31 and Rao et al. found histologic evidence of acid-fast bacilli confirmed to be MTB by PCR in the RPE of an enucleated eye of a patient with progressive uveitis and a 4 mm purified protein derivative (PPD) skin test,³² the study by Bajgai et al.²² was the first to report MTB DNA in eyes without clinical evidence of tuberculosis. The authors suggested that the RPE may act as a reservoir for dormant MTB, a plausible notion, given both the phagocytic nature of the RPE and the relative immune-privileged nature of both the RPE and peri-RPE space – a preferred site for other infections, such as Aspergillus^33,34 and Nocardia,^35,36 and for infiltrations, including lymphoma and retinoblastoma.^37–39 Given the identification of MTB DNA in eyes with Eales-like peripheral retinal vasculitis with non-perfusion,^40,41 one has to wonder whether pure, immune-mediated ocular inflammation can ever be confirmed in patients with latent tuberculosis.

Agarwal et al.²³ used optical coherence tomography angiography (OCT-A) to image multifocal serpiginoid choroiditis (MSC)^42,43 lesions in five patients with tuberculosis whose lesions showed paradoxical worsening, a form of JHR, following initiation of anti-tubercular therapy (ATT). All patients were seen at a uveitis referral center in Chandigarh, India, and each received both oral prednisone, 1 mg/kg in a tapering dose, together with four-drug ATT, which included isoniazid, rifampicin, ethambutol, and pyrazinamide. Worsening was defined either clinically or based on new or increased early hypofluorescence followed by late hyperfluorescence on fluorescein angiography, and occurred at a mean of 5 ± 3.81 weeks (range: 2–10 weeks; median: 3 weeks). Management included treatment with intravenous methylprednisolone and/or the addition of azathioprine in all patients. Prior to treatment, OCT-A of MSC lesions showed variable degrees of flow void in the choriocapillaries consistent with decreased perfusion at that level, an interpretation supported by corresponding areas of hypocyanescence on indocyanine green angiography (ICG-A). With paradoxical worsening, these areas of inner choroidal flow void showed either enlargement or coalescence. Progression tended to occur at lesion edges and corresponded to areas of late hyperautofluorescence and hyperfluorescence. Such leading-edge progression was often noted concurrent with central healing and loss of choriocapillaris and increased visualization of deeper, medium-size choroidal vessels. Lesions in two eyes showed the development of distinctive vascular tufts at the level of the choriocapillaris, which the authors suggested might have been localized choroidal angiogenesis – although simple restoration of choriocapillaris blood flow seems equally possible. As demonstrated by the same group in a companion paper in this issue of OII, paradoxical worsening and other findings that can influence treatment decisions can occur beyond the posterior pole and near-periphery, an area best visualized using ultra-wide field imaging.²⁴ Yee et al. described the use of OCTA in this issue of OII to demonstrate choroidal neovascularization in a patient with tuberculous MSC.²⁵ Agarwal et al.²³ cited an earlier study by their group that reported the occurrence of paradoxical worsening in one in seven patients with tubercular MSC following initiation of both ATT and systemic corticosteroids, and noted that management of inflammation should include continuation of ATT and addition of either intravenous methylprednisolone and/or the addition of a systemic immunosuppressive agents, such as azathioprine used in their subjects.⁴⁴

Together, these studies highlight the often complex causality of ocular inflammation in patients with infectious uveitis, particularly in those with chronic or latent infections and in settings where antimicrobial resistance is common. When possible, culture and/or PCR-based testing of ocular fluid should be used both to identify the presence of microbes or their DNA and to sequence for specific, resistance-conferring mutations. Use of either systemic⁴⁵ or local^46–48 corticosteroids or non-corticosteroid immunosuppressive agents together with microbe-directed antimicrobial therapy is often both indicated and effective.⁴⁵ This combined treatment approach, together with close clinical monitoring and early therapeutic response to changing clinical signs, are likely to achieve best visual outcomes.

DECLARATION OF INTEREST

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.

FUNDING

Supported in part by The Pacific Vision Foundation (ETC) and The San Francisco Retina Foundation (ETC).