Lecture: Diagnostic Work Up of the Uveitis Patient

This lecture reviews how to approach the diagnostic work up of a uveitis patent based on clinical history and presentation, pattern recognition, and laboratory work up. Cases will be presented with didactic material.

Lecturer: Dr. Ramana Moorthy


(To translate please select your language to the right of this page)

DR RAMANA MOORTHY: Good morning. Can everybody hear me okay? This is Dr. Ramana Moorthy. I’m holding this conference from Indianapolis, Indiana. I want to thank everyone from Cybersight, Orbis, and Lawrence in particular for putting this all together, and thank you, Dr. Neely, for inviting me to talk. Today the topic is going to be on uveitis. And this is one of the first lectures on uveitis that we’re doing. So we will keep the questions to the end, I think, just in the interests of time. We have a lot of slides to go through. And then we’ll try to anticipate any questions, et cetera, that you may have. There are a few questions that we’ll do during the talk as well. So today the talk is going to be on trying to work up a uveitis patient. How do you approach a patient that has intraocular inflammation? How do you do a laboratory workup, and what is the sequence of events that we go through when we see a patient who has inflammation in and around their eye? We classify uveitis today into the broad categories of non-infectious uveitis or infectious uveitis. Non-infectious refers to those where there may be some underlying systemic immunologic abnormality, or as we see in most cases, idiopathic, where something has initiated the inflammatory event in the eye, and this could be an environmental antigen or pathogen. But that is long gone, but this ongoing inflammatory disease process in the eye, because of the immune privileged environment of the intraocular structures, results in propagation and potentiation of a lot of inflammatory cells, and the inflammation continues, and needs to be treated, because it causes structural damage. Infectious uveitis, on the other hand, has a direct cause. A viral, bacterial cause, such as syphilis, TB, Lyme disease, for example, fungal cause, protozoal cause, helminthic cause, et cetera. We also can classify uveitis into masquerade syndromes, like retinitis pigmentosa, that can look like uveitis, or primary CNS or intraocular lymphoma. We can also classify uveitis anatomically, or based on time course. For example, anterior uveitis, the inflammatory cells are primarily in the anterior chamber. And in panuveitis, there’s inflammation of all the structures, with no preponderance for any one particular location. And of course, the time course can be separated into acute, chronic, or recurrent. And acute cases are those cases that last less than 3 months. Recurrent cases — and this is based on the criteria that were developed by the standardization of uveitis nomenclature, back in the early 2000s — this recurrence indicates that the patient has gone more than 3 months from the initiation of the first episode, has been off of medications, and then has a recurrence of disease, 3 months after the first episode. But if they have recurrence of disease, stopping medications and then they get a recurrence immediately, well, that’s more of a chronic or relapsing disease. So chronicity indicates that the inflammation never really completely went away. This is a picture of a patient who has anterior uveitis. This picture is from Dr. Emmet Cunningham. And this is a beautiful picture of anterior chamber cells. And you can see the tremendous amount of cellular inflammation, which should be differentiated from flare, which is caused by the so-called Tindal effect, which is actually the substance to the actual beam of light that is traversing the eye. And this is a good way to judge the amount of cells. 2×1 millimeter wide beam, on the highest intensity, using a slit lamp, is a good way to judge the number of cells in the anterior chamber. So here is the first poll question I have. What is the most common anatomic type of uveitis encountered in clinical practice? And this is for general practice. We’ll see what the poll shows us here. So the most common cause of… Or the most common anatomic type, excuse me, of uveitis that is encountered in clinical practice, is anterior uveitis. Probably 2/3 to 3/4 of cases that we see in clinical practice are anterior uveitis. In specific subsets, for example, pediatric, it’s idiopathic pars planitis. I see you guys got that, so you’re beyond that topic. I won’t spend too much time on that. The most common posterior uveitis we see, posterior infectious uveitis, is toxoplasmosis. The key to determining what causes the uveitis is really to first obtain an accurate and thorough history and perform a thorough ophthalmic and physical examination on the patient. And I can’t stress this enough. When I speak to residents, I tell them that we want to all jump in and say well, this person has uveitis, and I’m gonna do so and so labs. Well, the labs don’t mean anything, when they’re not put in the proper clinical context. So it is very important to obtain a thorough history of the ocular events, what has happened, what has transpired in this patient, and try to tie in with a thorough review of systems. Not just the ophthalmic review of systems, but the non-ophthalmic review of systems, to try to help you determine what is the underlying cause of the inflammation. Because determining the cause — and specifically, specifically making sure that you know this is not infectious uveitis — is very, very important. Because if it’s infectious, the treatment should be obviously geared towards using the appropriate antimicrobial agents to treat the infection, and then initiating corticosteroid therapy, after you’ve started the antiinfective therapy first, and getting the patient better. Laboratory testing, of course, can be helpful in confirming and narrowing the spectrum of the etiology. But they’re not a substitute for this history and physical examination. If you suspect an infection but you can’t prove an infection, it behooves you to do appropriate testing, and sometimes that involves invasive testing, such as obtaining vitreous or aqueous cultures. And if neoplasia is suspected, biopsy is essential. So when I do a comprehensive medical history, I really do a thorough evaluation of, for example, what drugs the patient is getting. What is currently active in the patient? Is there a history of autoimmune disease? Prior ocular surgery? This patient had multiple retinal surgeries and now presenting with uveitis in both eyes. This could be sympathetic ophthalmia. A complete review of systems is important. And let me explain what that involves for me. I start with the integumentary system. Any painful skin rashes? Lumps or bumps, painful red nodules under the skin? Any history of mucosal ulcerations such as oral aphthous ulcers or genital ulcers? Any history of arthritic symptoms, axial, distal, in the hands, fingers, toes, feet — where are the arthritic symptoms? Where are they? How long have you had them and how long have they been treated? History of chronic cough, sputum production, chest x-ray, positive PPD skin test, or quantiferon test? Recurrent diarrhea, blood in stool, change in bowel habit — things like that, that may point towards inflammatory bowel disease? History of recurrent urinary tract infections? History of travel, significant travel to any unusual locations in the world? For example, have you spent the past 6 months camping in the High Sonoran Desert of North America? Have you been to areas where you have drunk unfiltered water, where toxoplasmosis may be endemic? Those things are very useful. Any existing pets? Cats at home, or any unusual veterinary exposure? Those are all parts of the review of systems I do. Family history is systems helpful. For example when we’re looking at patients who have Blau’s syndrome, the familial form of sarcoidosis. There may be siblings that have inflammatory disease at a young age in the family already. A social history of smoking of course is very important, occupational exposure, IV drug use, et cetera, illicit drug use, is also very important. Smoking obviously potentiates and makes inflammation worse, so we need to know that for the patient as well. I talked about the review of systems here. I asked about systemic illnesses, current or ongoing. Fevers, chills, weight loss, and I also talked about the organ system involvement. Immunocompromised state. Are they being treated currently for cancer or have they had smokers have they had exposure to HIV or that may put them at risk for HIV? Are they malnourished? These are all parts of the etiologic examination. In the uveitis patients, you need to put all of what you learned in ophthalmology together. Uveitis, unlike other entities, involves and could produce complications in all of the broad categories and subspecialties of ophthalmology. It can cause cataracts, glaucoma, retinal structural abnormalities, including retinal detachment, cystoid macular edema, neovascularization. All these things are important. So we have to do a thorough examination of the patient. Best corrected vision, motility, confrontational examination of fields. And specifically looking for conjunctival nodules or granulomas that might indicate sarcoidosis. Corneal disease, epithelial dendrites or stromal infiltrates. Presence of KPs. If they’re diffuse or small or dendritic, this may indicate cytomegalovirus posteriorly. What are the color? Pigmented may indicate herpetic disease or old chronic inflammation. The anterior chamber depth, whether there’s presence of angle closure, iris Bombay, presence of broad corneal adhesions, or synechiae, for example in the iris, we need to look at cells in the anterior chamber, and look for evidence of hypopyon. Are there cells that should not be confused for inflammatory cells? Significant flare, that might tell us a fibrinous cyclitis — may be associated with low back pain. So these are all important considerations when viewing the anterior chamber. You want to look for rubeosis, iris neovascularization. Heterochromia, atrophy. Heterochromia can be difficult to see. So you should turn on the lights in the room, and get a gestalt, to see if there’s heterochromia, if you suspect iridocyclitis. Document, so you know what things look like in follow-up. Whether there are Koeppe or Busacca’s nodules. And presence of cataract. In addition, of course, applanation tonometry, looking at intraocular pressure, and evaluation of patients by gonioscopy, particularly if there’s elevated IOP and structural damage to the iris with posterior synechiae. Look in the angle, because these patients may be at risk for developing glaucoma, with combined mechanism. Look at the macula for cystoid macular edema, retinal thinning or atrophy. You may use the OCT to determine how much macular edema there may be. Look at the retinal vessels for sheathing, vascular occlusion, look at the retina for retinal whitening, retinal edema, retinal inflammation. Look at the choroid for old choroidal scars or active scars that might look white and creamy. Look at the optic disc for swelling or neovascularization. And the periphery. Retinal detachments, evidence of pars plana exudates in patients who may have pars planitis, a vitreous snowball in the periphery. Still, if you’re confused, after doing the history and physical examination, what is this inflammation? Obviously you anatomically categorized it at this point. You might be able to differentiate what this may be caused by, based on laterality. For example, in unilateral cases, herpetic disease, from cytomegalovirus or toxoplasmosis is more common. There are many cases of infectious entities that are bilateral, or can be unilateral. For example, the most common identifiable cause of hypopyon in uveitis is unilateral in presentation. Scleritis can be unilateral and is not infectious. Just because something is unilateral doesn’t always mean it’s going to be infectious, but infection rises to the top of the list. Especially herpetic infection. If you have identifying features that suggest sectoral atrophy, corneal scarring, things like that suggest herpetic disease as the underlying cause. Use non-ocular clues as well to help you in the diagnosis. So once you’ve done the ocular exam, look at the patient. Do they have a vesicular dermatomal rash to suggest zoster ophthalmicus, like the patient you see in the top picture? The middle picture is is the exanthematous rash of secondary syphilis. Be careful shaking hands with these patients. There are active spirochetes in that rash. Look for evidence of a bite that may be associated with Lyme borreliosis. You can see the classic picture of Lyme borreliosis with the target lesion. Here’s a patient who has very severe psoriasis, with large plaque-like psoriatic lesions. And this patient had anterior uveitis in both eyes and was HLA-B27 positive. This can give you a clue as to the etiology. Look at the lacrimal gland for inflammation and enlargement, as with this patient with dacryoadenitis from sarcoid. Or parotitis. This patient had Heerfordt’s syndrome. You can see the dramatic appearance of the patient’s enlarged wood write parotid gland and the active inflammation. She had panuveitis in both eyes as well at the same time. Is the hypopyon sterile? Infectious? Neoplastic? You can’t assume. There may be other things going on. If they’re five days postcataract surgery and have hypopyon and vision loss, it’s pretty easy to suggest they have infectious endophthalmitis. But if the vitreous is clear and they have isolated hypopyon, you might have to think of HLA-B27. So we see patterns. Toxoplasmosis, for example, has a pattern of unilateral focal choroiditis. But that could be Bartonella henselae. Its most common presentation is not neuroretinitis, but unilateral focal choroiditis. Varicella zoster virus and HSV can cause anterior changes, corneal or iris changes. These are all clinical features that are very typical of unilateral cases of viral anterior uveitis, or viral anterior uveitis associated with other changes that suggest VZV or HSV. Late onset endophthalmitis based on history, somebody who is pseudophakic who can’t get off corticosteroid drops for two years, this may suggest that the patient has endophthalmitis. If the patient has been sick or ill, has been an IV drug user, you might think about the possibility of aspergillus endophthalmitis. A patient who has a history of HIV, who presents with vision loss and floaters, and has evidence of necrotizing retinitis with retinal hemorrhages and a pizza pie appearance, that’s cytomegalovirus. Toxocariasis you can see as a granuloma or diffuse endophthalmitis picture as well. If you’re still not sure, combine what you know about the history with the pattern recognition, to try to determine the underlying cause. So here’s a picture of hypopyon. How would I know what’s causing the hypopyon, unless I ask the patient — I’ve been treated with rifabutin for the last six months for a mycobacterium avium infection. So this patient got a rifabutin or drug-induced hypopyon on the top left. And this is a patient on the top right that had cataract surgery three days ago and presents with pain and vision loss. Severe redness, ecchymosis, excuse me, chemosis of the conjunctiva. And this is all suggestive of infectious endophthalmitis. Here’s a patient on the bottom left, who had an intravitreal injection of Kenalog. Literally two days ago. And presents with vision loss. This patient actually just has a pseudohypopyon, from presence of Kenalog particles in the anterior chamber. See the white stark beautiful clinical appearance. There’s a little bit of fibrin in the anterior chamber too, which is probably why he had received the Kenalog injection. And you can see on the bottom right a patient who has the classic hypopyon of Behcet’s disease. This patient had a shifting hypopyon. If you have the patient sit up and shake their head, the hypopyon disappears. It’s not a congealed hypopyon you may see with HLA-B27 or endophthalmitis. And this is significant vitritis. The snowball appearance. Here’s toxoplasma retinochoroiditis in a 10-year-old boy. You can see adjacent vasculitic infiltration of the retinal vessels. You can see along the retinal vessels here — my arrow — this vascular sheathing — this is called Kyrieleis vasculitis, which is associated with toxoplasmosis. But Bartonella should be a consideration in children who have such changes as well. It should be serologically evaluated in those patients. Here’s a patient with tuberculous uveitis. But I’ve seen patients with syphilis that look the same way. This patient has a history of active TB, multifocal choroidal miliary nodules in the posterior pole, the choroid, significant vitritis, and these miliary granulomas are on the macula. Vitreous hemorrhage, neovascularization on the disc. This is a complex patient with tuberculous panuveitis. This is something we see rarely in the United States, but it’s more common in places like India, where TB is endemic, where we see multifocal tuberculous serpiginous choroiditis. This patient clearly had active tuberculosis. This patient had severe inflammation along — in the vitreous, along with changes of this multifocal choroidal inflammation. But this is a pattern of TB that should not be missed, and anybody with serpiginous choroiditis you should be considering TB in the differential diagnosis. Here’s the classic star formation in ocular Bartonellosis. This is a classic, but not the most common for Bartonella. This is the classic pizza pie that you see in association with cytomegalovirus in a patient with HIV/AIDS. Here’s a person with necrotizing herpetic retinitis, in an immunocompetent person, where you see evidence of perivascular clearing, giving this cracked mud appearance in the later stages of the disease. And here’s a clinical picture of a patient who had all the findings classic for what? Behcet’s disease. Hypopyon, and here almost a necrotizing retinitis appearance because of vasculitis from Behcet’s disease. This is based on multiple clinical criteria, and not a blood test. This patient may have been HLA-B51 positive, but that’s not what led to the diagnosis of Behcet’s disease. It’s a clinical diagnosis. This patient has Vogt-Koyanagi-Harada syndrome. Evidence of severe thickening of the choroid, serous retinal detachment, and this starry sky pattern of leakage from multiple spots with pooling of fluorescein, and here you can see in the later stages of the disease, the convalescent phases of the disease, vitiligo of the eyelid skin, in this patient, who has a sunset glow fundus, and evidence of small punched out lesions in the periphery, especially inferiorly. This is typical. Here’s a patient that I showed you earlier, with psoriasis. Severe psoriasis. Who is HLA-B27 positive, who presents with dense fibrinous hypopyon, associated with fibrinous iridocyclitis. He also has ankylosing spondylitis. This is a tragic case. Presents with severe vision loss, with white, quiet eye. What’s the diagnosis? Exactly. This is juvenile idiopathic arthritis. Articular disease, young patient, evidence of keratopathy, chronic anterior segment inflammation that has been untreated or poorly treated with structural changes and damage and blindness. Very high risk for severe vision loss and even loss of the eye in these kinds of settings. These kinds of patients need to be caught early, screened frequently, caught early, and treated. So let’s talk about what kind of lab test you do for a patient who presents with uveitis. So every patient should be worked up for which of the following conditions? One, rheumatoid arthritis and systemic lupus? HSV, VZV, CMV infection, toxoplasmosis, and sarcoidosis, syphilis, and tuberculosis. What is the right answer? I think we’ll look at what people have responded here. Give a minute or two here. But I think that as we talked about, all of these things are potential things to consider in the right setting. But there are three entities that should always be evaluated in patients who have uveitis. And that is sarcoidosis, syphilis, and TB. These are important, because syphilis and tuberculosis are treatable. And important from both a public health perspective — let’s talk about this a little bit more, because it looks like we kind of have a smattering of responses. So any patient who has uveitis — let me ask. Rheumatoid arthritis and systemic lupus — typically will not cause anterior chamber inflammation. Will not cause vitreous inflammation. These diseases can cause scleritis. They may rarely cause retinal vasculitis. But rarely alone do they cause inflammation with cells in the anterior chamber or vitreous or the so-called uveitic inflammatory disease. Now, HSV, VZV, and cytomegalovirus infection — lab testing for those would probably not involve serologic lab testing. I should have been more specific in my question. Certainly these are things to consider based on the clinical appearance, but these are not typical labs that we’ll obtain on a patient. Toxoplasmosis — I’ll check that if it’s appropriate, based on the clinical appearance of the patient. But every patient who has inflammatory cells — if you’re gonna work them up, you need to rule out the possibility of sarcoid, syphilis, and TB. Because these diseases, again, they can masquerade as multiple different forms of uveitis. And syphilis and TB in particular have broad implications in terms of their curability, and their social treatability, and the associated social aspects of these diseases. So these are very important to identify. These systemic or infectious inflammatory conditions. So every uveitis patient that has a workup should be tested for sarcoid, syphilis, TB. SST. Remember that. Well, syphilis testing involves serologic tests that are specific for treponemes. In other words, you want to get treponemal-specific testing, such as FTA, MHA-tp, or now commonly what’s available is antitreponemal antibodies. Which has replaced the FTA in most places in the world. At least in Europe and North America. So syphilis testing should be specific for treponemes. There’s a lot of false positives for syphilis testing. It’s beyond the scope a little bit of this. Obviously patients who have lupus, other autoimmune diseases, women in particular may have a higher rate of false positivity for this testing. But the FTA is important. Remember, syphilis, if it causes uveitis, it’s going to be postprimary. Either secondary or tertiary. These patients have already had primary syphilis. So non-specific markers such as VDRL — it’s only going to be transiently serologically positive. The FTA, however, starting within a couple weeks, will be positive and remain positive for the lifetime of the patient. So that’s why it’s important to look for those, because secondary and tertiary syphilis — we’re going to be seeing positive FTA and not necessarily positive VDRL. So when you say I’ve checked for syphilis, make sure you’re talking about treponemal-specific testing. Sarcoidosis testing is some controversy, but if the patient has pulmonary symptoms, chronic cough, weight loss, you know, they have evidence of enlargement of the salivary gland, or of lymphadenopathy, or evidence that they have lacrimal gland enlargement, an angiotensin converting enzyme level, serum calcium level, and chest x-ray can certainly be part of the workup, that can be helpful in ruling out sarcoidosis. The blood test alone can’t be diagnostic. Tissue is diagnostic for sarcoid, and you need to show conjunctival granulomas, for example, that are non-caseating granulomas, based on the histopathologic evaluation. And of course, in a histopathologic evaluation of sarcoid granulomas from the conjunctiva, you need to rule out other things like foreign body and fungus and tuberculosis. By using acid fast testing on the histologic specimen. TB… How do we test for TB? There was a question about — how about quantiferon? In endemic areas of the world, we still tend to rely on simple testing like tuberculin skin testing. If the patient has been treated with BCG vaccination, as they are in many places like India and Sub-Saharan Africa, it may be necessary to get a Quantiferon Gold — that testing has in the United States replaced PPD testing. And it can be very, very useful. If a patient has had previous BCG vaccination, a PPD may not be reliable. And in that case, a chest x-ray is very important. But does the clinical appearance of the uveitis match tuberculosis, for you to suggest that this is TB? That can be very difficult, because TB is a masquerader. It can present as anterior uveitis, intermediate uveitis or pars planitis. So if these patients have a quantiferon gamma that’s positive, no previous history of vaccination, I would check their PPD. Do they have a significant skin reaction of the PPD? Don’t bypass the PPD, even though it’s an old test, to confirm if the quantiferon gamma assay is of significance. We need to look at the chest x-ray, the PPD, and the clinical appearance of the patient before making the decision that this is tuberculous uveitis. In addition, I think it’s very, very important to check in specific disease conditions for HLA-B27. Now, I normally do not obtain an antinuclear antibody testing in an adult who presents with anterior uveitis, because lupus doesn’t cause that. But in a child who has chronic iritis or chronic iridocyclitis, we’re considering that juvenile idiopathic arthritis is an important condition to rule out. We’re going to check an antinuclear antibody, because especially girls will be ANA positive — those patients are gonna be the ones who have anterior uveitis. This is a picture of the fluorescent treponemal antibody test, where the treponemes are illuminated by the antibody. This is an overview of the assays that are done for cell function, ELISA, the quantiferon TB or T spot TB test, where we obtain blood samples from the patient and stimulate them with mycobacterium tuberculosis protein, and look at the interferon gamma that’s produced in a couple of different ways to determine if the patient has revved up T-cells, to determine if the patient has more than just previous TB exposure, but actually latent or active disease. This is very helpful for latent disease. Active disease, you’re going to see signs of active disease clinically. That needs to be put together. Sarcoidosis, a chest x-ray can be very helpful to grade the stage of the disease. And the same goes for tuberculosis as well. We can also use, if we’re still suspicious of infectious etiologies, and we want to look specifically for underlying pathogens in the aqueous fluid or vitreous fluid, many of us now rely on the use of polymerase chain reaction testing. PCR testing, it can be very, very useful if you suspect that the patient has viral disease, such as HSV, VZV, CMV, or if they have toxoplasmosis. The usefulness of aqueous taps for patients who have infectious posterior uveitis is of some question. And in those cases, you need to go to where the action is. If the patient has vitritis and posterior uveitis, the best samples are going to be derived from the vitreous. The diagnostic testing — once you ruled out TB and syphilis and sarcoid, what else can you do? Based on the clinical appearance and history, I would taper what other test you get. Do I get toxoplasma? Is there a history of going into endemic areas of Lyme, such as in Western Wisconsin in the United States, or in the Northeastern seaboard region of the United States, where Lyme is more endemic or are there other non-ocular or non-uveitic clues in the cornea or the iris that may suggest that the patient has HSV disease? In those cases, we may obtain specific testing for viral etiology. Remember that serologic testing for HSV, VZV, and CMV is only helpful in ruling out a negative diagnosis, because most adults are going to have positive serologies to these viruses from previous exposure. I have a few questions. I’m just gonna break off here. So yes, many of you don’t have all of these testings readily available where you are. If you’re in remote Mongolia, for example, you don’t have all of the testing, et cetera. And rely more on history and physical examination. But ultimately, at least if you had to pare down to the most minimal things that you need to check for, the sarcoid, syphilis, TB is the bare minimum. So if you can’t get a blood test for sarcoid, it’s not available, a chest x-ray may be very useful. A TB skin test is very useful, and certainly minimal testing for syphilis, using treponemal-specific antibody testing, is really very important. TORCH screening may not be as useful in cases where reliable testing for uveitis, especially in adults — and of course, the testing that you do for diagnosis of Bartonella — we do Bartonella serology. We look for Bartonella henselae, quintana and IgG titers. That can be very helpful for these diseases. I’m gonna move on to get to some of the cases I want to share with you. Getting a little bit late into the course here. So, you know, I think there’s other additional serologic things that can be helpful in delineating what the patient has. You know, we talked about the ACE and lysozyme being not as useful, and only combined with a chest x-ray that’s positive, and a histopathologic evidence of sarcoid that’s positive. In teenagers who have bilateral ocular inflammatory disease, who present with uveitis, I think about the possibility of tubulointerstitial nephritis. The interstitial nephritis can be transient, and the uveitis can be persistent. There are no serologic tests for things like sympathetic ophthalmia and Behcet’s. These are clinical diagnoses. And in scleritis, I do check for lupus and for specific vasculitic conditions. And I use ANA and rheumatoid factor testing in a child who has chronic anterior segment inflammation, if I’m suspecting JIA-associated disease. PCR testing we’ve already touched on, but there are numerous things that are available. The most common readily available PCR testing in most labs in North America and Europe are for CMV and Epstein-Barr virus and toxoplasma gondii. There are large academic centers in Europe and North America and Southeast Asia — we can have access to panfungal and panbacterial primers, with multiplex PCR technology now being available, with very tiny quantities being able to deduce underlying forensic DNA analysis of the aqueous, if you will. It can be very helpful in delineating what the etiology of the inflammation may be. I’m gonna move quickly on a little bit and talk a little bit about therapy. Because I think it’s important. Because if you’re in the midst of a workup of a patient with uveitis, the question — the statement here is a true or false statement. Corticosteroids should never be used alone in the treatment of infectious uveitis. Is that true or false? Let’s see what you guys all thought. Yeah. So, you know, corticosteroids should never be used alone and the key is alone. If you suspect that the patient has infectious inflammatory disease, you don’t want to give them — I’m talking about systemic intraocular or periocular corticosteroids. You should never, ever do that. To treat a patient who you think has infectious uveitis. It’s probably safe to put them on topical corticosteroid, but you need to add some systemic antimicrobial agent, if you think it’s viral or HSV or VZV, you may want to add some variant of acyclovir. If it’s Bartonella, add an appropriate antimicrobial agent. And then go on to treat a patient with something more aggressive, in terms of corticosteroids, based on what you find in the evaluation of the infection. So once you’ve found that the patient is infectious, you have them on antibiotics, do corticosteroids have any role? The answer is: Corticosteroids have no role in the treatment of infectious uveitis. Is that true or false? This is a little bit of a trick question. It’s a fine point, and I’ll try to clarify this. Obviously any time you have inflammation, you’re going to use corticosteroids. But the question is… When to initiate corticosteroids. If the patient has infectious uveitis, always, always, always treat them with the infectious agent first. And then use the corticosteroids. That’s right. Use the corticosteroids, then, to treat the inflammatory component. We know that the host inflammatory disease is what causes the damage in cases of infectious uveitis. So, in other words, if you see a patient with endophthalmitis, you don’t just give them intravitreal antibiotics alone. You follow that up with intensive topical and sometimes more aggressive corticosteroid therapy to help reduce the inflammation that results from the death of the bacteria. So therapy should be, in the case of infectious uveitis, directed towards the agent. And then later addition of corticosteroids. Sometimes you have to use very aggressive periocular intravitreal or systemic corticosteroids to control the infectious uveitis, once you put them on antibiotic therapy. Neoplastic masquerades have to be diagnosed based on histopathology. Pars plana vitrectomy, sometimes biopsy for patients with primary CNS lymphoma is essential. And you have to communicate with the pathologist when you’re obtaining the specimen, in order to make the diagnosis. If you don’t have a pathologist who is competent, they won’t know what to look for, and you’ll miss the opportunity to make a live saving, potentially, live changing diagnosis. And non-neoplastic masquerades — we confirm based on history and exam findings. Is there an interocular foreign body in the angle? Do a gonioscopy and look for that, if there’s a history of trauma. Chronic retinal detachment — these are clinical diagnoses. Next full question. The patient has poor vision, severe anterior segment inflammation, small pupil with very limited view of the fundus, and the labs are pending. What should you do to treat this patient? Their inflammation is very severe. Should you, number one, start interocular steroids? Start periocular steroids? Start systemic corticosteroids? Or start topical steroids? So the worst thing you can do when you don’t have a view of the posterior pole and severe inflammation, granulomatous, posterior synechiae, can’t see anything in the back, and it’s a unilateral disease process, you haven’t looked in the back of the eye, the worst thing you can do is give them interocular, periocular, or systemic corticosteroids. If they have necrotic retinal syndrome, you’ve finished the eye. So if you’re not sure of the etiology, take the safe path initially, until the workup comes back, and then proceed with more aggressive therapy based on your evaluation of the patient. Sometimes you don’t have a good idea. Choose an empiric systemic antimicrobial therapy or even interocular antimicrobial, for the most serious infectious condition that you think it possibly is. That way you’re covering yourself and then start topical steroids. If they don’t respond, add oral corticosteroids with the antibiotics on board and observe them closely and see how they respond and how they improve. It’s very, very important that you don’t make something that is bad irretrievably worse by doing something that will get you into trouble. And that specifically is treating infectious uveitis inappropriately, with aggressive corticosteroids, without appropriate antimicrobial coverage. So if you have a lack of response or worsening to corticosteroids, you need to think about the fact that — hey, this could be infectious, or this could be a masquerade syndrome. We’ll talk more about — in a different lecture — about therapeutics. I just wanted to touch on it, and touch on the importance of why we need to rule out infection versus non-infection. Now, I have a few cases. We’re getting late into the talk here. Into the last 10 minutes or so. And I will try to wrap this up, but these are important cases I wanted to present, not because I think that these are cases with take-aways and really common things, but they’re rather really distal cases really difficult cases that I’ve seen, that I struggled over making a diagnosis. Sometimes these diagnoses take months to make. There was a question about toxoplasmosis. So if you suspect toxoplasmosis, if I’m gonna do serologic testing, I will check for IgG antibody, and IgM both. But if the patient has an old scar, and there’s adjacent reactivation, then only the IgG is going to be positive. Remember, the IgM is an acute phase kind of — from initial exposure, the IgM level will go up. But it will not remain positive for the life of the patient, and it will gradually become negative again. However, that IgG will start going up after initial exposure, and go down. But then when reexposure occurs, only the IgG antibody response is mounted by the body, because of immune memory. And as a result, IgG testing is very important. So I always obtain toxoplasma IgG antibody. Occasionally will obtain IgM if I think it’s a primary acquired toxoplasmosis. More than 50% of worldwide toxoplasmosis is primarily acquired. It’s not congenital disease, as we used to think. So this is a very important question. How long do you wait before starting systemic steroids? Well, if you think it’s infectious, and you put them on antimicrobials, I would start systemic corticosteroids probably within a couple of days. About 48 hours after initiation of antimicrobial therapy. If you don’t know whether it’s infectious or not, and the laboratory test is unrevealing and the patient is getting worse and you’ve just got them on topicals, try to get them on systemic steroids and watch them carefully, especially if you have a good view of posterior segment. If you don’t, that presents a diagnostic and physical challenge. A lot of times the vision can help us. If it’s really poor and there’s no view of the posterior segment, you have to assume there’s significant problems going on in the posterior segment as well and not just opacities causing problems. I think about masquerade syndromes in the presence of, for example, clinical scenarios and history. I didn’t talk about this much, because this is a general topic, but if a 70-year-old patient presents with a new onset of vitritis, bilaterally, with significant vitreous inflammation, I would think about the possibility of primary ocular lymphoma. If the patient has a history of nyctalopia, and has cells in both eyes and bone spiculing, this patient has retinitis pigmentosa. So these are based on a clinical diagnosis, clinical appearance, history, and physical examination and ocular examination — is going to tell you when you need to consider the possibility of masquerade syndromes. The last question is… I assume from Turkey here, where I have a question about the IGRA test for TB. It is reliable. I think it’s an excellent test. The CDC in the United States has recommended that replace PPD skin testing completely. But in endemic areas, it can be a little bit tricky. Because patients have exposure to TB all the time. Places like India, where 25% of patients are walking around, potentially, who have had some form of latent TB, the IGRA test is going to be positive in those patients. How do you then know that the TB is the underlying cause of uveitis? In many cases, you are stuck with treating the tuberculosis, the latent form of the TB, at least with appropriate antituberculous therapy with the help of an infectious disease specialist, and then at the same time, simultaneously, putting the patient on appropriate systemic immunomodulatory therapeutic agents like corticosteroids to get the inflammation under control. So you have to use your clinical judgment in those cases. The first case that I have is a 27-year-old male who presents with floaters in both eyes and shortness of breath. He’s had a 30-pound weight loss over 6 months. He lives in a trailer, a mobile home, with several cats. His vision is 20/50 in both eyes, granulomatous KPs, and cystoid macular edema in both eyes. Here’s the fundus examination. You can see the vitritis, disc edema, and choroidal inflammation on the fluorescein angiogram. Small spots of hyperfluorescent multifocal choroiditis in both eyes. And here’s white fleck-like deposits in the peripheral retina that this patient has, preretinal deposits. The laboratory workup is negative for the patient, but radiology testing and chest CT shows severe hilar adenopathy. PPD skin test is negative, everything else is negative on this patient. So there are multiple enlarged paratracheal lymph nodes, mediastinal lymph nodes are enlarged, massively enlarged pulmonary artery from pulmonary hypertension. We were thinking this could be sarcoid. He’s also go non-contrasted changes on CT with the liver. So he has hepatomegaly as well, and periaortic lymph nodes that are present here. There’s a dark arrow pointing to those lymph nodes. So this lymphadenopathy was biopsied. We don’t have any serologic testing to help us. And there was granulomatous inflammation in the lymph nodes that were biopsied from the chest. This patient was thought to have sarcoid, and the patient was started on systemic steroids. But inflammation improved initially, but then as the steroids were tapered, the patient worsened. Fevers developed. Fevers. Now we’re stuck. We’ve got somebody who… You know, kind of got better on steroids, but is getting worse again. Did we really miss what was going on? Is this infectious? Did we miss it? It was some sort of indolent infectious process that we’re covering up with the steroids? These are everyday problems that we see with really severe cases of uveitis where we’re trying to figure out what’s causing the inflammation. Should we biopsy something else? Look again at the lymph nodes? What’s our next step? So because we have already done the lymph node biopsy, we felt like there was nothing else externally to biopsy. Perhaps we should look at the vitreous. So we did a vitreous biopsy and got some of that preretinal white material biopsied. And this showed lymphocytes and large macrophages that contained PAS-positive material. And the PCR was performed, because we had multiplex PCR available. We were able to scream for tropheryma whipplei. This was Whipple’s disease! He had some chronic pain, mild diarrhea and had all these PS-positive material in the substantia propria, the intracellular material in the jejunal biopsy that you see here. And this was consistent with Whipple’s disease. This patient was managed with rifampin and Bactrim, the corticosteroids were in place with the antibiotic therapy, and this patient has been off meds for more than 10 years now from the onset of the disease. This second case — 79-year-old man with a unilateral severe granulomatous anterior uveitis and vitritis with a hazy view of the posterior pole. This is straight out of the question I asked earlier. Hazy view of the back, 20/40 vision, and the patient apparently had been treated with intraocular steroids and got a lot worse. The light perception at presentation — rapid worsening after that injection of intraocular steroids. Now the antenna is up. He’s pseudophakic. And the right eye is normal. But he has a couple of little funny scars in the periphery in the right eye. Here is the left eye. Look at this left eye. Hazy view. You see the optic disc — these are hazy. This is what I saw. Looked just as bad. And in the periphery, I saw these large white areas. What’s happening here? I was concerned. This could be necrotizing retinitis. Necrotizing herpetic retinitis. Or could this be something else? Could this be toxoplasmosis? It’s in one eye. It’s possible. So I sent off serology. What am I gonna treat this patient with? He’s had intravitreal steroids that made things worse. This is bad. This patient had very poor vision. The possibilities at the top of the list are necrotizing herpetic retinitis and toxoplasmosis. Should I treat him for both? Well, you could do that. I chose the worst thing that it could be. I chose necrotizing herpetic retinitis, because that’s what I was most concerned about. And I started doing intravitreal injections of antivirals in this patient. I also put the patient on oral antitoxoplasma medication. I didn’t want to take any chances. But in the mean time I set him up to have a pars plana vitrectomy and biopsy performed, because I thought it was important for us to rule out the possibility of toxoplasmosis or viral etiologies for the retinitis. And indeed, this patient had PCR that was positive for toxoplasmosis. At the time of the vitrectomy, I gave him clindamycin and ganciclovir empirically. But a few days later, he was known to have toxoplasmosis. There was no view of the posterior pole and the patient was inappropriately given corticosteroids. Made it much worse, and the patient ended up with — not a blind eye. He got up to 20/70 when I saw him a few years ago. So he had chronic cystoid macular edema. But at least better than light perception vision. But this patient’s retina looked like this, after we were all done with the vitrectomy. There are some laser scars around the edge of the retinitis here, large spots of retina. Fortunately he didn’t wipe out the entire posterior pole. But you can really make somebody much worse by putting them on inappropriate corticosteroid therapy, if it’s infectious. So treat with antimicrobials, and then add the corticosteroids. Never give intravitreal periocular corticosteroids without appropriate antibiotic coverage if infection is even remotely suspected. And in the case of infectious uveitis, if you don’t have serologic diagnosis, tissue is often the issue to make the diagnosis correctly. Here’s another… This is the last case that we’ll do. So we can wrap things up. And I’ll check on the questions and answers after this case. 69-year-old white female with painless subacute vision loss, metamorphopsia and a few floaters. But on review of systems, she complains of severe pleuritic pain, dyspnea, and arthralgias of hands and fees. There’s 1+ vitritis in both eyes, and the retina has evidence of disc hyperemia and serous retinal detachment of posterior pole in each eye. There’s no vasculitis that’s seen, but there’s a serous detachment that these arrows are trying to point out. It’s not a great picture, because it’s a little hazy. And there’s serous detachment in the macula on the fellow eye as well. Here’s what the fluorescein looked like. Starry sky pattern of posterior pole choroidal lesions that leak and pooling of fluorescein in the subretinal space. Is this VKH? That’s what it looks like. But she has dyspnea and weight loss. That’s not typical for VKH. And no history of trauma or previous surgery or anything. So what is this? This is weird? What kinds of things should I do in the workup? Sarcoid, syphilis, TB, still important to rule out, obviously. Here’s the other eye. Same kind of clinical picture. Very interesting picture here. She had no history of hearing loss, vitiligo, polio, anything to suggest VKH. Everything we did was all negative. But the chest x-ray I got for dyspnea, which revealed bilateral pleural effusions, and I went back and obtained an antinuclear antibody on this patient. Very, very high. And the double stranded DNA antibody was high as well. The pleural fluid was biopsied and showed a very high positive antinuclear antibody titer, and this patient had lupus! And she presented with lupus choroidopathy. She had met more than 4 out of the 11 diagnostic criteria. So this is very important to keep in mind. That the diagnostic criteria are what’s used to make the clinical diagnosis of lupus. The blood tests were an afterthought, frankly. I normally don’t get blood testing for lupus for uveitis patients, unless they’re a child and I’m expecting that they have pauciarticular JRA. In this case, with vasculitis, lupus choroidopathy is a very rare condition. You can see that we need to use pattern recognition to help us go in the right direction for diagnosis. And then choose the appropriate testing based on what the clinical appearance is. The patient was treated with systemic corticosteroids. Has had durable remission and is disease free now for 15 years. I’m going to stop there and look at the questions. So somebody asked — what is the percentage of uveitis with negative labs? The party line, among uveitis specialists in the United States, I would say 50 to 60% of patients that we see have a laboratory workup that is negative. And have so-called idiopathic uveitis. So this is an excellent question. The vast majority of patients are going to have a negative workup. That doesn’t mean that there’s no cause for the inflammation, or there’s some mysterious thing. Many of these patients may have had an initial exposure, as I said, to some environmental antigen, or some environmental pathogen that induced the inflammatory process to begin with, and because they were genetically predisposed to the inflammation, and then that started a series of cyclical immunologic events inside the eye, that caused chronic inflammation to develop. And that’s the way I kind of understand it, when I can’t — and that’s why I try to tell patients — when I don’t have an underlying cause, and we’re treating the disease, but not really treating the underlying cause. A few more questions here. Why would you — why you do every patient with uveitis intermedia with an MRT to exclude multiple sclerosis? Oh, would you do? Yeah. So intermediate uveitis and its association with multiple sclerosis is an excellent point that somebody brought up. We know ophthalmic literature has suggested that between 5% to 15% of patients with intermediate uveitis may have associated multiple sclerosis. And similarly, the data works the other way. About 10% of patients with multiple sclerosis actually have intermediate uveitis. So in these cases, I am careful to ask the patient on review of systems about neurologic symptoms. If they have intermediate uveitis, I’ll go back to review of systems and ask them. Have you had evidence of a neuropathy that lasted for six weeks and went away? Weakness or numbness in an extremity that lasted for some time and went away? History of optic neuritis? If they give a clinical history of that, I would say… Yes, we need to then evaluate the patient for multiple sclerosis. And I would refer them to a patient — to a neurologist for evaluation. I would — in that situation — not do the MRI. Not evaluate them myself. Because there are numerous questions that then arise, if they have MS, in terms of how they should be best managed, and a neurologist who is an expert in the treatment of multiple sclerosis should do the testing and evaluation and treatment of that patient. I use cycloplegics always, when the patient has acute anterior uveitis. I almost never not use cycloplegics. We will talk more about this in the therapeutics section. But I use it religiously. Cycloplegics usually for milder inflammation — I’ll use cyclopentolate. Or I’ll use atropine initially and switch to something milder once the inflammation is better controlled. Is there any alternative to HLA testing for resource-poor countries? If you suspect that the patient has spondyloarthropathies, seronegative spondyloarthropathies, HLA testing isn’t essential, but an x-ray of the involved joints, for example, a sacroiliac joint film, can be very helpful in determining whether the patient has underlying sacroiliitis. Up to 50% of patients who have HLA-B27 hypopyon have sacroiliitis. So that would be a reasonable screening tool if you do not have HLA testing available. For severe inflammation in TB uveitis, of course you have to use steroids, but you have to use steroids only in the presence of appropriate antituberculous therapy. What blood test for lymphoma or leukemia choroiditis? There aren’t any blood tests available. If the patient has severe leukemia and has a complete blood count, a differential and actual plated differential — where a pathologist actually looks at the smear, peripheral smear of the blood — it’s very helpful in making the diagnosis of leukemia, if there’s leukemic infiltrates in the eye. Sometimes we’ll use aqueous or vitreous sample, of course, to make the definitive diagnosis, if we don’t have any systemic things going on, and the lymphoma is isolated only to the eye. Remember that primary interocular lymphoma presents only in the eye in 20%, 25% of patients. So that may be the first presentation. So we rely on biopsy, vitreous biopsy, in those cases. If the cornea is involved, how will it affect management? I think that the question probably is related to herpetic disease, or corneal epithelial defects, and using corticosteroids. So this is an excellent question. So if the patient has herpetic disease and has a corneal epithelial defect, and has significant anterior segment inflammation, usually I will put those patients on antiviral — systemic antivirals. Because those are highly effective. Especially against HSV, for example, and cause the epithelium to heal, so that then we can use the topical steroids. I would be very careful about using topical steroids in the presence of an epithelial defect, if I’m considering the possibility that the epithelial defect is due to herpetic disease or other infectious entities. And of course if you have a hypopyon in a patient who has an epithelial defect and mild corneal infiltrate, this could be infectious keratitis, and the uveitis could be secondary to the keratitis. Keep that in mind. Optic nerve head swelling is quite common in patients who have intermediate and even anterior uveitis. I’ve seen it as well. So this is an effect, probably because the peripapillary choroid has been inflamed, and perhaps some of the peripapillary retinal vessels have been inflamed, resulting in disc edema, in some cases. There are some cases where the disc edema is very, very severe, and in those cases, the patient may truly have a second disease in addition to the uveitis. I had a JIA patient with pseudotumor, and that patient required pressure evaluation, to rule out the pseudotumor, and indeed, he had pseudotumor. Had to be treated with systemic Diamox along with immunosuppressives. What investigation can you do if you have someone who has a postflu vaccine shot associated neuroretinitis? I don’t know of any. In this case it’s your clinical impression, clinical gestalt, that is going to make you go along the lines of — this neuroretinitis is not associated with the typical thing. But I would still rule out the possibility of Bartonella. Syphilis. And even TB. Those kinds of things can certainly cause neuroretinitis as well. How do we diagnose intermediate uveitis in the setting of sarcoidosis? You know, sarcoid can certainly cause intermediate uveitis. Remember that sarcoid is, quote-unquote, an autoimmune disease. How you treat it — whether it’s idiopathic or sarcoid uveitis is not going to make the difference. But the diagnosis is going to be based on obtainable tissue biopsy. Before you put the patient on topical corticosteroids, which can cause the conjunctival granulomas to melt away. Look in the inferior fornix, flip the lid, look at the bulbar conjunctiva. Look for granulomas. If you can make a tissue diagnosis of sarcoid, it’s immensely helpful going forward with systemic therapy. For VKH patients with bilateral optic nerve head swelling, do we need brain MRI? In that setting, no. Unless the patient has other neurologic symptoms. If they have classic features of VKH and optic nerve head swelling I don’t do MRI nor do I recommend lumbar puncture. So this is a clinical diagnosis. Would you take conj biopsy if there are no nodules? No, I always look for nodules. I don’t do blind conjunctival biopsy. If you read reports from Ralph Eagle, the yield is very, very low. So you should look for granulomas. These granulomas are gonna be different from the typical conjunctival follicles. They’re going to look larger. They’re going to be more fleshy-looking. And they’ll be pretty obvious. These are conjunctival granulomas. And it’s surprising how often that’s missed. So I always look. Look at the conjunctiva. Dr. Ron Smith, my old mentor, used to tell me — you know, if you don’t look at the conjunctiva, as a uveitis specialist, it’s like not doing a peripheral retinal examination with scleral depression, being a retinal specialist. It’s very, very essential in uveitis to look at the conjunctiva. The role of doxycycline in Chikungunya retinitis. I have very little experience, but doxycycline as an antibiotic cover is very important before the initiation of corticosteroid therapy for the treatment of optic nerve head inflammation. Which part of the conjunctiva would you find nodules? Most of the time in the inferior fornix. Usually in the forniceal, between the bulbar — in that area. But I also look on the upper lid. I’m going to break it off here. I want to thank you for your excellent questions and involvement. I hope you could hear me well. I really enjoyed this. I always learn something from all your questions. So I look forward to the next session. Thank you very much.

Download Slides


January 6, 2017

Last Updated: October 31, 2022

3 thoughts on “Lecture: Diagnostic Work Up of the Uveitis Patient”

Leave a Comment