Glossary
Uveitis
This is the one eye problem worth knowing the warning signs for, and they're easy enough to remember. It comes on over hours rather than days. How fast it gets looked at changes what happens next, which is why this page says the urgent part more than once.
What the research found.
One review reports how often uveitis turns up, broken down by illness. It's 33 percent in ankylosing spondylitis, 25 percent in reactive arthritis, and 13 percent in undifferentiated spondyloarthritis, which is the high end of the range. It's 6 to 9 percent in psoriatic arthritis and 2 to 5 percent in bowel-related disease.
One study followed 301 people with spondyloarthritis, and at the point of diagnosis 11.5 percent had had an attack. Twenty years after diagnosis it was 39.3 percent. The HLA-B27 marker was independently linked to it, at a hazard ratio of 4.5 whose true value sits anywhere from 1.3 to 15.2.
Two trials pooled their double-blind periods and counted how many people developed uveitis in each arm. It occurred in 2 of 349 people on bimekizumab, which is 0.6 percent, against 11 of 237 on placebo, which is 4.6 percent. That came in at a nominal P value of 0.001.
One clinic worked up 912 consecutive people who were referred with uveitis, and it stayed unexplained in 46.9 percent of them. A whole-body disease accounted for 37.3 percent. The commonest of those were sarcoidosis at 17.1 percent, then HLA-B27-related disease at 12.5 percent, and then tuberculosis at 7.5 percent.
What it is
Uveitis means inflammation inside the eye, and the word comes from the uvea, which is the middle layer of the eyeball. Doctors sort it by where in the eye it sits. Anterior uveitis affects the front, posterior affects the back, intermediate sits between the two, and panuveitis means the whole thing is involved.
That distinction counts, because the two ends of the eye behave differently. In rheumatic disease most attacks are anterior and those usually do well. A minority reach the back of the eye, and one review notes those are more often complicated by swelling at the center of vision and by sight loss. That difference is the whole reason for the urgency below.
What an attack is like
It usually comes on over hours rather than days, and that speed is the most useful clue you have. Learn it and you'll recognize what's happening the first time it happens to you. The eye turns red and painful, bright light hurts more than it should, vision can blur, and it's typically one eye at a time rather than both.
Compare that with a slow, gritty dryness, which builds over weeks and has other causes entirely. Dry eye creeps up on you and this doesn't. Sudden and painful is the picture worth acting on, and hours rather than weeks is how you tell them apart.
Why it's a same-day problem
You can't tell from the outside whether an attack has reached the back of the eye, and neither can anybody else. It takes looking in with the right equipment. That's the whole reason for the urgency, because some attacks do well and some threaten sight, and from where you're standing the two start out looking alike.
So sudden eye pain, redness, blurred vision, or new trouble with bright light is a same-day problem. It doesn't wait for next week, and it doesn't wait for your next rheumatology appointment. That's the one sentence on this page worth remembering, and being told it once is usually enough to make it stick.
How often it happens
This is the commonest thing that happens outside the joints, and that holds right across this group of illnesses. How common it is depends entirely on which one of them you have. The spread between the diagnoses is wide enough that a single number for the whole group would mislead almost everybody who read it.
One review reports the figures by diagnosis, and ankylosing spondylitis sits at the top at 33 percent. Reactive arthritis comes in at 25 percent, undifferentiated spondyloarthritis at 13 percent, psoriatic arthritis at 6 to 9 percent, and bowel-related disease at 2 to 5 percent. The review adds that the rate rises both with the HLA-B27 marker and with how many years you've had the illness.
A Moroccan study approached the question from the other direction and got a matching answer. It followed 136 people with spondyloarthritis, and uveitis was the most frequent feature outside the joints, at 18.4 percent. Nothing else they counted came anywhere close to it.
The risk keeps building
One study makes the time point cleanly, and it's the one figure on this page that changes how the rest reads. It followed 301 people with spondyloarthritis and counted how many had ever had an attack. At the point of diagnosis, 11.5 percent had. Twenty years after diagnosis it was 39.3 percent, and over that stretch 82 of the 301 had at least one episode.
So a long quiet stretch means you've been lucky so far rather than that the question is closed. That's worth knowing without being worried by it. The useful response is to learn the warning signs once and then get on with your life, rather than to watch your eyes every morning.
What raises the risk
That study also looked at what predicted an attack, and two things came out independently. The HLA-B27 marker was linked to it at a hazard ratio of 4.5, with the true value running anywhere from 1.3 to 15.2. That range is very wide, so the size of the effect isn't well pinned down, though the direction is clear enough.
Heel pain was the second thing, at a hazard ratio of 1.8 from 1.1 to 2.9. That one is smaller and much better pinned down. It also sounds like nothing, which is why it's worth mentioning to whoever is looking after your spine.
Which drugs cut the attacks
They don't all behave alike, and the difference between them is one of the more useful things on this page. Plenty of people assume a drug that helps the joints must help the eyes. That study compared how many attacks people had before starting treatment with how many they had after.
Among 27 people on tumor necrosis factor antibodies, attacks fell from 1.83 per person to 0.41, at P equals 0.002. Among 19 people on etanercept they didn't fall, going from 0.44 to 0.79 instead. Those are small groups, and the review that follows reaches that conclusion from a much wider base.
That review names infliximab, adalimumab, and certolizumab as effective at reducing how often uveitis happens, and it names etanercept as not. It adds that one interleukin 17A blocker, secukinumab, hasn't been shown effective in uveitis. So the family a drug belongs to doesn't settle the question, and neither does how well it works on your joints.
A newer drug in that family points the other way, and its figures are worth having. Two trials pooled their double-blind periods, and uveitis occurred in 2 of 349 people on bimekizumab, which is 0.6 percent, against 11 of 237 on placebo, which is 4.6 percent. The nominal P value was 0.001, and across a wider pool of 848 patients the rate stayed low.
Those numbers are worth raising with your own team, and that goes double if you've had attacks before. They aren't figures to act on alone. What they're good for is turning a vague worry about your eyes into a specific question somebody can answer.
When uveitis comes first
Sometimes the eye trouble arrives before anybody has mentioned arthritis, and that turns out to be worth knowing. It changes a good deal about what happens afterwards. One registry held 2367 patients with spondyloarthritis, of whom 379 had this kind of uveitis.
Among those 379, the eye trouble came first or alongside for 59 people and afterwards for 229. The ones whose eyes came first did better on every measure the registry looked at. They had better function, at an odds ratio of 0.85 from 0.73 to 0.99, and less structural damage, at 0.88 from 0.79 to 0.99. They were also diagnosed faster, at 0.90 from 0.84 to 0.96.
That last figure probably explains a good deal of the rest. An eye attack sends somebody to a specialist quickly, and the arthritis then gets found while there's still less of it to find. It's an argument for speed rather than an argument about the eye itself.
If you've had uveitis
It's fair to ask whether anything else should be looked for, and one study set out to answer it. It scanned 102 people who'd had sudden front-of-eye uveitis. Bone marrow swelling in the sacroiliac joints showed in 52 of them, swelling strongly suggesting axial spondyloarthritis showed in 33, and 41 of the whole group met the classification criteria for the disease.
One detail from that study is worth keeping. Long-standing back pain was just as common in the comparison group of healthy people, so back pain on its own won't sort this out. A scan might, which is the point worth raising with whoever is looking after you.
Uveitis also has many causes that have nothing to do with rheumatic disease at all. One clinic worked up 912 consecutive referrals and found it stayed unexplained in 46.9 percent, while sarcoidosis accounted for 17.1 percent, ahead of HLA-B27-related disease at 12.5 percent and tuberculosis at 7.5 percent. The authors note that about 60 causes have been described, which is a reminder that an eye clinic is where this gets sorted out.
Common misconceptions.
Myth. Sore red eyes can wait for my next appointment.
Reality. Not these ones, because uveitis in rheumatic disease usually comes on over hours rather than days. Sudden eye pain, redness, blurred vision, or new trouble with bright light needs looking at the same day. Most attacks hit the front of the eye and do well, while a minority hit the back, and those are more often complicated by swelling at the center of vision and by sight loss.
Myth. It's a rare complication.
Reality. It's the commonest thing that happens outside the joints in this group of illnesses. One review reports it in 33 percent of people with ankylosing spondylitis, and one Moroccan study of 136 people with spondyloarthritis found it the most frequent feature outside the joints, at 18.4 percent. So this isn't a rarity, and it's the main one to know about.
Myth. If I haven't had it by now, I won't.
Reality. The risk keeps accumulating, which is the opposite of reassuring and still worth knowing. One study followed 301 people with spondyloarthritis and measured how many had ever had an attack, and at the point of diagnosis it was 11.5 percent. Twenty years later it was 39.3 percent, so a long quiet stretch means you've been lucky so far rather than that you're safe.
Myth. Any arthritis drug will protect my eyes.
Reality. They don't all behave alike here, and one study measured the difference. It compared attacks before and after treatment, and people on tumor necrosis factor antibodies had fewer, falling from 1.83 to 0.41 per person, at P equals 0.002. People on etanercept did not, going from 0.44 to 0.79, and one review lists three antibody drugs as effective at cutting attacks while naming etanercept as not.
Myth. Uveitis means I must have a rheumatic disease.
Reality. Often it means nothing of the kind, and the numbers on that are striking. One clinic worked up 912 people referred with uveitis, and it stayed unexplained in 46.9 percent while a whole-body disease explained 37.3 percent. The commonest of those was sarcoidosis at 17.1 percent, with HLA-B27-related disease next at 12.5 percent, and the authors note about 60 causes have been described.
Related terms.
- HLA-B27, the gene marker that raises the risk of this.
- Ankylosing spondylitis, where uveitis is commonest.
- Biologics, for the drugs that differ in whether they cut attacks.
Questions patients ask.
What is uveitis?
Inflammation inside the eye, with the word coming from the uvea, which is the middle layer of the eyeball. Doctors sort it by where in the eye it sits. Anterior means the front and that's much the commonest kind in rheumatic disease, posterior means the back, intermediate sits between the two, and panuveitis means the whole thing is involved.
What does an attack feel like?
It usually comes on over hours, and that speed is the useful clue. The eye goes red and painful, bright light hurts more than it should, and vision can blur, typically in one eye at a time. Speed is what separates this from the slow gritty dryness that builds over weeks and has other causes entirely. Sudden and painful means the same day rather than next week.
How urgent is it?
Same day rather than next week, and the reason is simple. Most attacks in rheumatic disease hit the front of the eye and settle with treatment, while a minority involve the back. One review notes those are more often complicated by swelling at the center of vision and by sight loss. You can't tell which kind you have from the outside, and neither can anybody else without the right equipment and a look inside the eye.
How likely am I to get it?
That depends on your illness and on how long you've had it. One review reports 33 percent in ankylosing spondylitis, 25 percent in reactive arthritis, and 13 percent in undifferentiated spondyloarthritis, which is the top of the range. It reports 6 to 9 percent in psoriatic arthritis and 2 to 5 percent in bowel-related disease. The review adds that the rate rises with HLA-B27 and with the years.
Does HLA-B27 raise my risk?
Yes, and by a fair amount on the evidence we have. One study followed 301 people with spondyloarthritis and found the marker linked to uveitis on its own, at a hazard ratio of 4.5 whose true value runs from 1.3 to 15.2. That interval is wide, so hold the size of it loosely. Heel pain was linked on its own too, at 1.8 from 1.1 to 2.9, and that one is better pinned down.
Do my arthritis drugs protect my eyes?
Some appear to and some don't, which is worth knowing before you assume. One study compared attacks before and after treatment, and people on tumor necrosis factor antibodies went from 1.83 attacks to 0.41, at a P value of 0.002. People on etanercept went from 0.44 to 0.79. One review lists infliximab, adalimumab, and certolizumab as effective, and it lists etanercept as not.
Can uveitis come first?
Yes, and people who take that route seem to do better. One registry held 2367 patients, of whom 379 had this kind of uveitis, and the 59 whose eye trouble came first or alongside had better function and less structural damage than the 229 whose came later. They were also diagnosed faster, at an odds ratio of 0.90 from 0.84 to 0.96.
I've had uveitis. Should I be checked for arthritis?
It's worth raising, and one study went looking for the answer. It scanned 102 people with sudden front-of-eye uveitis, and bone marrow swelling in the sacroiliac joints showed in 52 of them, with swelling strongly suggesting axial spondyloarthritis in 33. Of the whole group, 41 met the classification criteria. Long-standing back pain was just as common in the comparison group, so pain on its own won't tell you.
References.
- Sharma SM; Jackson D. Uveitis and spondyloarthropathies. Best Pract Res Clin Rheumatol. 2017;31:846-862. 10.1016/j.berh.2018.08.002Review of uveitis in the spondyloarthropathies
- Frantz C; Portier A; Etcheto A et al. Acute anterior uveitis in spondyloarthritis: a monocentric study of 301 patients. Clin Exp Rheumatol. 2019;37:26-31. PMID 30620268Cross-sectional single-centre observational study
- Brown MA; Rudwaleit M; van Gaalen FA et al. Low uveitis rates in patients with axial spondyloarthritis treated with bimekizumab: pooled results from phase 2b/3 trials. Ann Rheum Dis. 2024;83:1722-1730. 10.1136/ard-2024-225933Pooled analysis of uveitis incidence across phase 2b and phase 3 randomised controlled trials of bimekizumab in axial spondyloarthritis
- Bertrand PJ; Jamilloux Y; Ecochard R et al. Uveitis: Autoimmunity… and beyond. Autoimmun Rev. 2019;18:102351. 10.1016/j.autrev.2019.102351Review of the records of 912 consecutive patients referred to a hospital internal medicine department for diagnostic work-up of uveitis
- Gómez-García I; Ladehesa-Pineda ML; Puche-Larrubia MÁ et al. Uveitis as the first symptom in spondyloarthritis and its association with the evolution of the disease. Results from the REGISPONSER registry. Joint Bone Spine. 2021;88:105136. 10.1016/j.jbspin.2021.105136Cross-sectional analysis of the Spanish national spondyloarthritis registry REGISPONSER
- Bubova K; Hasikova L; Mintalova K et al. The Prevalence of MRI-Defined Sacroiliitis and Classification of Spondyloarthritis in Patients with Acute Anterior Uveitis: A Longitudinal Single-Centre Cohort Study. Diagnostics (Basel). 2022;12. 10.3390/diagnostics12010161Longitudinal single-centre cohort
- Essouiri J; Abourazzak FE; Kona I et al. Profile of Patients with Spondyloarthritis in Morocco. Curr Rheumatol Rev. 2018;14:258-263. 10.2174/1573397113666170406125338Retrospective observational study in one Moroccan rheumatology department analysing the records of 136 patients diagnosed with spondyloarthritis between January 2009 and June 2014
- Hong C; Kwan YH; Leung YY et al. Comparison of ankylosing spondylitis and non-radiographic axial spondyloarthritis in a multi-ethnic Asian population of Singapore. Int J Rheum Dis. 2019;22:1506-1511. 10.1111/1756-185X.13603Comparative study of 262 patients with axial spondyloarthritis in a Singapore registry
This page gathers the published research on this subject into one place. The studies behind it were published between 1989 and 2026, and every figure links to the paper it came from. Those studies were peer reviewed. This summary of them was not. Dr. Sarah Luebker is reviewing these pages one at a time and has not reached this one yet, so it carries no medical review date and nothing here is her opinion or her advice to you. Each page gets updated as she reaches it. It is here in the meantime because the science is worth having in one organized place that is easy to find and easy to read. Talk to your own clinician before acting on any of it.