The onset of O-EIM symptoms vary broadly in relation to IBD symptoms. Patients with episcleritis tend to flare with bowel disease activity (53). On the other hand, patients with uveitis appear to flare independently of bowel disease activity (53). In pediatric patients, a cross-sectional prospective study found that there is no relationship between bowel disease activity and ocular inflammation (38). However, this data is limited.

Patients may present to an ophthalmologist with findings of eye inflammation prior to receiving a diagnosis of gut inflammation. A case report identified a patient with an initial presentation of posterior uveitis that went on to develop symptoms of IBD 6 months after the onset of uveitis (74). This is not the only case of ophthalmic presentation preceding gastrointestinal symptoms (70). Another case report described a patient who developed anterior uveitis eight years prior to developing bowel inflammation (67). For these patients who present initially with isolated uveitis, the differential diagnosis can be quite broad and an ongoing review of symptoms can help identify a new inflammatory disease that manifests over the course of treatment.

Patients with known IBD are referred to ophthalmology upon developing ocular symptoms (64). The vast majority will have symptoms of dry eyes, but a significant minority will go on to develop scleritis or uveitis. A study that performed ophthalmic screenings in a gastroenterology clinic in patients with IBD demonstrated occult eye disease which was defined as asymptomatic anterior chamber cell or flare (50, 67).

Unsurprisingly, HLA genes have also been found to be biomarkers of EIMs of IBD, and while some genetic associations, such as B27, B58, and DR 103, have been implicated, the phenotypic manifestations can vary broadly (63, 81). One study speculates that the disequilibrium between the HLA genes may be the culprit of EIMs (81); however, no studies have been done to date to assess this theory.

Various studies have proposed that patients having one EIM are at higher risk of developing another EIM (29, 71, 82). A prospective study looking at EIMs found a strong association of arthritis with having eye, skin, and mouth complications of IBD (29). A large cohort study evaluating the risk of uveitis in IBD patients with psoriatic arthritis found an increased risk of 3.55 of developing uveitis in patients with CD (20). Similarly, a cross-sectional study assessing patients with IBD in the ophthalmology clinic found arthritis was highly prevalent among the uveitis patients indicating a strong association between uveitis and arthritis EIMs (22). Another prospective study found that having uveitis was directly associated with the onset of arthritis and erythema nodosum in patients with IBD (41). They also proposed that the presence of uveitis may be used to predict disease severity. This finding has been replicated and research suggests that the gene encoding the RBM19 locus may serve as a biomarker for the concurrence of O-EIMs (83).

Anti-metabolites, namely methotrexate, azathioprine, and mycophenolate mofetil, prevent DNA synthesis and are frequently used as first line alternatives to corticosteroids. Methotrexate, which prevents DNA synthesis through inhibiting dihydrofolate reductase, has demonstrated effective results in the treatment of ocular inflammation as well as other autoimmune conditions. A multicenter, retrospective cohort study evaluating methotrexate’s treatment outcomes in patients with non-infectious ocular inflammation found that methotrexate is moderately effective for achieving a corticosteroid-sparing drug regimen (33). However, for patients with active luminal CD, methotrexate is more effective with a tapering dose of corticosteroids (86). Methotrexate is used more frequently than azathioprine and mycophenolate mofetil due to its once weekly dosing, and long term availability of safety data (86). However, in the settings of pregnancy or lactation, azathioprine is preferred (86). These medications are more inexpensive than the newer immune modulator treatments and have fewer systemic side effects than the corticosteroid medications; however, their broad mechanism of action still predisposes patients to a variety of other side effects. Newer agents such as biologic response modifiers have been noted to have fewer side-effects.

Anti-tumor necrosis factor (TNF) alpha agents comprise the main class of drugs referred to as biologics. The main stay of therapy for inflammatory bowel disease utilizes biologic therapies (86). TNF-alpha antagonist medications are monoclonal antibodies that bind to the TNF-alpha antigen to prevent pro-inflammatory effects (87). Infliximab (Remicade), adalimumab (Humira), and etanercept (Enbrel) comprise the most common TNF-alpha antagonists. These medications have been used to treat patients with IBD along with other non-ophthalmic autoimmune conditions and have been widely used off-label for the treatment of refractory uveitis.

Infliximab has been found to not only treat the underlying IBD symptoms, but also the EIMs (86). A cross-sectional and prospective study found that higher doses are required to effectively treat EIMs when compared to maintenance dosing for IBD symptoms due to an elevated inflammatory burden as measured by infliximab clearance and serum anti-drug metabolites (43). It is important to note, however, that the study (43) was done in children of whom only 8 had uveitis.

With a relatively low rate of treatment-ending adverse events at about 19%, infliximab has been shown to be a safe medication for patients who meet criteria (48, 51). There has been a case report, however, of a patient with UC who developed anterior uveitis after a second infusion of infliximab (69). Though unusual, the timing of the infusion and resolution with discontinuation point to infliximab as the likely cause, though this is a rare finding, and infliximab has demonstrated success in achieving remission from refractory uveitis with eventual and successful taper off the drug (51). A large observational cohort study found no difference in the risk of developing uveitis in patients who were taking infliximab versus adalimumab or an antimetabolite (azathioprine or methotrexate) thus indicating that antimetabolites and TNF-inhibitors are equally effective at suppressing uveitis (17). For patients who meet criteria for TNF-alpha antagonists and corticosteroid-sparing antimetabolites, the antimetabolites offer a more cost-effective option though they come with greater side-effects.

Adalimumab was shown to be effective at treating non-infectious uveitis (NIU) in VISUAL I and II studies which were multicenter randomized, controlled trials. In these trials (36, 37), the authors found a low risk of treatment failure in patients being treated with adalimumab versus placebo and the risk of adverse events was similar between the two groups as well. In the VISUAL III study (34), they found improved control of ocular inflammation, improvement, and stabilization of BCVA, and reduced corticosteroid usage. Additionally, a study analyzing 75 clinical trials of adalimumab in patients with varying inflammatory diseases found no incidence of IBD in patients with uveitis (88). Adalimumab has been widely used for IBD management, and with its effectiveness in treating uveitis as well, it is a common choice for patients who may benefit from a biologic response modifier.

Etanercept should be used cautiously in patients with ankylosing spondylitis since there have been case reports of new onset uveitis (66) and CD (19) in patients while on etanercept. However, no cases have been reported of uveitis development in a patient with IBD or vice versa while taking etanercept.

Vedolizumab is a gut-selective alpha4beta7 integrin blocker that has been approved for moderately to severely active UC and CD in adults. A multicenter retrospective study found that patients with EIMs had 50% improvement at 12 months. The study had 1 patient with uveitis who demonstrated an improvement at 6 months which continued at 12 months also (42). Similarly, a case report from 2018 of a 24-year-old male with UC and an anterior uveitis flare was started on vedolizumab (74). At his 1-year follow up, he had not developed another uveitis flare and a repeat colonoscopy demonstrated normal mucosa (74).

However, a retrospective study on patients with IBD comparing EIMs in patients receiving vedolizumab versus anti-TNF therapies demonstrated that patients with CD on vedolizumab were 28% more likely to develop any EIMs. The incidence rate ratio of episcleritis/scleritis was 2.51 (95% CI, 1.02-6.14) and uveitis was 2.89 (95% CI, 1.35-6.18). Patients with UC were not more likely to develop any type of EIMs, but were more likely to develop other, non-ophthalmic EIMs including aphthous stomatitis, pyoderma gangrenosum, and PSC (89). This indicates that vedolizumab may have limited efficacy outside of the gut and may explain why it is not commonly used (89).

IL-23 therapies have been explored to treat IBD and uveitis since IL-23 was found to be elevated in IBD and NIU (32). Ustekinumab (Stelara), which is an anti-interleukin-23, has been used to treat patients with uveitis and CD. Two female patients with ileocolonic CD who had undergone ileocecal resection and had multiple flares of severe bilateral non-infectious uveitis yearly despite adalimumab treatment were treated with an induction dose of ustekinumab and maintenance treatment every 8 weeks for at least 6 months without recurrence of NIU. Ustekinumab may, thus, be an effective treatment for patients with severe recurrent NIU and for patients in whom anti-TNFs are either contra-indicated or ineffective (55).

Similarly, in an international multicenter retrospective study of IBD patients who were treated with either ustekinumab or vedolizumab, one patient developed episcleritis while on vedolizumab while no patients developed new inflammatory eye disease while on ustekinumab after 1 year of treatment (13). Additionally, more patients developed arthralgias while on vedolizumab, and 2 patients on vedolizumab stopped treatment due to arthralgias while no patients on ustekinumab discontinued treatment (13).

These cytokine therapies have thus far demonstrated superiority to the prior medications; however, long term data in larger populations remains to be gathered.

JAK inhibitors are relatively new medications in the treatment of EIMs of IBD which are being explored due to the JAK pathway’s involvement in signal transduction for multiple cytokines in the setting of inflammation (25). The OCTAVE trial, evaluating tofacitinib, demonstrated either improvement of active disease from all EIMs or stability from baseline after 52 weeks of treatment (25). Based on this trial, the JAK inhibitors present a promising future for the treatment not only of uveitis, but for many EIMs.

The newest technologies that seek to treat EIMs in IBD are mesenchymal stem cell-derived exosomes (MSC-EXOs). While these treatments are still in preclinical trials, they demonstrate promising outcomes. Specifically, MSC-EXOs suppress the inflammatory response by decreasing CD3+ T cells in the retina and by preventing the migration of macrophages to the retina of mice (90). The studies also demonstrated a reduction in the inflammation of IBD and uveitis in mice when targeting Th1 and Th17 cells thus showing the importance of these cells for EIMs and IBD, and demonstrating a potential drug target (90). As these treatments progress in research, it will be important to ensure the accessibility of these medications to the patients for whom they were made.

Historically, rifabutin was used as a treatment for uveitis due to a theory (47, 49) suggesting that Mycobacterium paratuberculosis contributed to the pathogenesis of uveitis. However, there have been cases (62, 65) of patients developing ocular inflammation when on rifabutin and clarithromycin. This disease process has been coined “rifabutin-associated uveitis.” Worsening ocular inflammation arises leading to a hypopyon which is unusual in uveitis alone (62, 65). While the pathogenesis of rifabutin-associated uveitis is unknown, physicians should avoid using rifabutin in patients with IBD-associated uveitis for risk of worsening the ocular inflammation especially when other immunomodulators can be used instead. However, if a patient presents with rifabutin-associated uveitis, then the authors of the case reports suggest resolution of symptoms with systemic steroids (62, 65).

Though some have argued that recurrence of uveitis symptoms after colectomy is unlikely (54), there is a case report of patient who developed granulomatous uveitis on post-colectomy day 1 (58). This case also demonstrated the importance of rapid recognition and treatment of granulomatous uveitis to avoid enucleation (57). Similarly, another case report indicated that ocular inflammation after colectomy may be due to recurrent bowel inflammation which should be evaluated (59). A retrospective study found that there was no significant difference in ocular manifestations after undergoing or not undergoing a colectomy (82). Given the lack of definitive data, it is not clear that a colectomy should be performed to eliminate uveitic disease. Furthermore, uveitis should not be ruled out as a cause of ocular symptoms after a colectomy.