Uveitis is inflammation of the middle layer of the eye (uvea), located between the sclera and the retina, which is comprised of the iris, ciliary body, pars plana and/or choroid. In some cases, by extension, it can also affect the retina.

The incidence of uveitis is 45–52 cases per 100,000population/year, it peaks between the ages of 20 and 44 years, and the prevalence of the disease is 70 cases per 100000 population.1,2 It is the third leading cause of blindness in the working age population in developed countries, and 10% of the cases of total loss of vision are due to uveitis.

Anterior uveitis is the most common form of intraocular inflammation, representing between 50% and 92% of all cases of uveitis reported in a Western population.3 Anterior uveitis associated with HLA-B27 accounts for between 18% and 32% of the cases of anterior uveitis found in that population.4

It has been reported more frequently in men, in whom onset is earlier (20–40 years), it tends to be unilateral and there are higher rates of recurrences and complications. Anterior uveitis is more severe when associated with the HLA-B27 antigen, with a higher degree of inflammation and a more frequent use of systemic corticosteroids.5

The purpose of this article is to describe the clinical features, the need for systemic therapy and the frequency and types of ocular complications in a cohort of patients with anterior uveitis associated with HLA-B27 and in a cohort of patients with endogenous noninfectious anterior uveitis not associated with HLA-B27.

We also proposed to establish whether there were differences between the 2 cohorts in terms of clinical features, frequency and types of ocular complications, and the frequency of the use of systemic corticosteroids and other systemic immunosuppressive agents.

We included a total of 162 patients, 58 with anterior uveitis associated with HLA-B27 (HLA-B27+) and 104 with endogenous noninfectious anterior uveitis not associated with HLA-B27 (HLA-B27−).

Of the HLA-B27+ patients, 30 had a diagnosis of spondyloarthritis and 1 had a diagnosis of juvenile idiopathic arthritis (JIA). The different types of spondyloarthritis are shown in Table 1. We wish to point out the fact that 64.5% of these patients were diagnosed with spondyloarthritis after having been referred from the ophthalmology department to the rheumatology department after a first episode of uveitis. Most of these patients described symptoms characteristic of long-standing spondyloarthritis.

Most of the HLA-B27− patients had been diagnosed with idiopathic uveitis. The various diagnoses included and their frequency in this cohort can be seen in Table 2.

The demographic and clinical characteristics of both cohorts are shown in Table 3.

The mean age of the patients in the two cohorts was between 40 and 50 years, and there were no statistically significant differences between them.

We did not observe a significant predominance of one sex over the other in either of the 2 cohorts, although there was a somewhat higher number of men in the HLA-B27+ cohort and of women in the HLA-B27− cohort.

The median length of follow-up was greater in the HLA-B27+ cohort, but the difference did not reach statistical significance when compared with the HLA-B27− cohort.

Likewise, no significant differences were found with regard to laterality; unilateral involvement was more frequent in both cohorts.

However, differences were observed in terms of the clinical course. In the HLA-B27+ cohort, somewhat more than 2 thirds of the patients experienced recurrences and there were no cases of chronic uveitis. In the HLA-B27− cohort, half of the patients had had only 1 episode of uveitis, somewhat more than a fourth of the patients had a recurrent course, and the remainder, chronic.

The proportion of patients who developed some type of complication was similar in both cohorts, there being no statistically significant differences (P=.686). However, when the types of complications that were documented for the patients in the 2 cohorts were analyzed, we observed differences, but the statistical significance could not be determined because the sample size was not large enough (Table 4), except in the case of synechiae, which were more common in the HLA-B27+ cohort, although the difference was not statistically significant (P=.07). One patient in this cohort developed central serous choroiditis related to the use of systemic corticosteroids. In the HLA-B27− cohort, band keratopathy was observed in 6 patients, all of whom had been diagnosed with JIA. We also found a higher frequency of cataracts and ocular hypertension (OHT) in the HLA-B27− cohort. This fact was not associated with a more widespread use of systemic corticosteroids given that, as was seen in the case of cataracts, the higher frequency in the HLA-B27− cohort can be explained by a greater trend toward chronicity. In the case of OHT, the differences can be explained by the greater trend toward hypertension in some of the types of uveitis, such as Fuchs’ heterochromic iridocyclitis, included in the HLA-B27− cohort. The presence of OHT coincided with ocular inflammatory activity in all the patients.

Regarding the use of corticosteroids or other systemic immunosuppressive agents, again, there were no statistically significant differences, as these treatments were seldom required in either cohort (Table 5).

Most of the patients treated with systemic corticosteroids began the therapy with a dose of 30mg a day of prednisone or the equivalent. In every case, immunosuppressive therapy was introduced due to ocular inflammatory activity that could not be controlled with other treatments, in general, to prevent recurrences or chronicity.

The immunosuppressive agents utilized were sulfasalazine, methotrexate, adalimumab and tocilizumab (Table 6).

There are a considerable number of publications dealing with the clinical features and prognosis of HLA-B27-associated anterior uveitis.7–16

The description of the clinical features coincides in the majority of the publications, but the prognosis does not. We found articles in which the authors reported a better prognosis for uveitis associated with HLA-B27,7 whereas others affirmed the opposite,8 and still others who observed a prognosis similar to that of anterior uveitis not associated with HLA-B27.9–13

In contrast to earlier publications,4,5 in the present study, we did not observe male predominance in the cohort of patients with anterior uveitis associated with HLA-B27, an earlier onset or a greater trend toward unilaterality in that cohort. However, we did find a greater trend toward recurrence in the patients with anterior uveitis associated with HLA-B27, similar to that reported in previous publications.4,12 The rest of the clinical features were similar in both cohorts.

One noteworthy finding is the high percentage of patients who were diagnosed with spondyloarthritis after being referred from the ophthalmology department to the rheumatology department due to an episode of uveitis. This circumstance demonstrates the importance of the collaboration between the two specialties.

In this study, we did not take into account the degree of inflammatory activity at the time of the onset of the episode of uveitis because of the difficulty in standardizing the information, since the episodes were evaluated by different ophthalmologists and there was wide variability in terms of the establishment of the degree of inflammation.

Likewise, we did not take into account the visual prognosis of the patients, as we considered that the study design was not the most appropriate for that assessment or for establishing prognostic factors. We would like to continue the study, carrying out a prospective evaluation of the patients for the purpose of clarifying this point.

The fact that we found no statistically significant differences between the two cohorts in terms of the use of corticosteroids—in contrast to results published elsewhere5—and other systemic immunosuppressive agents, and the presence or absence of complications could indicate that the severity and the visual prognosis is comparable in the two groups, but we cannot reach any further conclusions because of the limitations of the study, which include the unavailability of data on the changes in the visual acuity of the patients. Moreover, in the HLA-B27− cohort, we observed an incidence of chronic uveitis of 21% and not a single case in the HLA-B27+ cohort. This difference would mean that, although the uveitides in the HLA-B27+ cohort were more severe, no significant differences were observed in terms of the use of systemic therapy on comparing the two cohorts, because in the patients with chronic uveitis, even if the activity was lower, systemic therapy would be initiated more often.

With respect to the differences observed in the various types of ocular complications, which, as we mentioned above, cannot be measured accurately because of the small sample size, we should consider the influence of the heterogeneity in the HLA-B27− cohort, which includes 8 different diagnoses. This would explain, for example, the elevated presence of band keratopathy, observed only in those patients of the sample with a diagnosis of JIA.

If we exclude synechiae, whose influence on the visual prognosis of the patient is low, the rate of complications is low in both cohorts.

In conclusion, in the present article, there are no significant differences between the clinical features of uveitides associated with HLA-B27 and those of uveitides not associated with HLA-B27, with the exception of a greater trend toward recurrence.

The study design does not enable us to establish differences with respect to the visual prognosis.

We would like to increase the sample size of the study and carry out a prospective follow-up to obtain information on the differences in the development of different types of ocular complications, and on differences and possible factors that intervene in the prognosis.

The authors declare they have no conflicts of interest.