Abstract
Purpose:
To estimate the incidence and predictive factors for cataract in eyes with episcleritis and scleritis, and to evaluate the outcome of cataract surgery in those eyes.
Design:
Retrospective cohort study at uveitis subspecialty centers.
Methods:
One thousand three hundred eighty-four eyes with non-infectious scleritis and episcleritis at risk of cataract were included. Predictive factors for cataract development were assessed by multivariable Cox regression. The main outcomes were development of cataract, defined as the first reduction of presenting visual acuity <20/40 attributed to cataract or else occurrence of cataract surgery itself. A second cohort of eyes with episcleritis and scleritis that underwent cataract surgery was evaluated for postoperative outcomes. Logistic regression was utilized to assess variables associated with visual acuity 20/40 or better one year after cataract surgery.
Results:
Seventy-six eyes developed cataract (incidence=0.025/eye-year, 95% confidence interval: 0.019–0.031). Age≥65 years, elevated intraocular pressure≥30 mmHg, use of oral corticosteroids at the preceding visit, and anterior chamber inflammatory activity were associated with increased cataract incidence. Race/ethnicity, type of scleritis, bilaterality were unassociated with cataract risk after adjustment. Among 79 cataractous eyes that underwent cataract surgery, median presenting visual acuity improved by 6 ETDRS lines. Pre-operative factors including duration of inflammation, immunotherapy use, and corticosteroid use were not significantly associated with odds of post-operative visual acuity 20/40 or better.
Conclusions:
Under subspecialty management, the incidence of cataract was low in eyes with episcleritis and scleritis. Cataract surgery was associated with large and sustained improvements in visual acuity.
Keywords: scleritis, episcleritis, cataract, surgery, visual outcome
Introduction:
Scleritis is an uncommon and potentially destructive inflammatory disorder of the sclera; remission usually requires many years so chronic treatment typically is required1, 2. In one large retrospective study, nearly 40% of scleritis cases were associated with an underlying systemic inflammatory disease, most often rheumatoid arthritis3. The majority of cases require treatment with systemic corticosteroids and/or immunosuppressive drugs4. However, given less use of topical therapy in scleritis, the risk of cataract might be less than in other forms of ocular inflammation since topical corticosteroid therapy is associated with more risk of cataract than systemic therapy5. Episcleritis is the inflammation of the episclera that often is benign, but the redness and irritation can be disturbing to patients and is frequently treated with chronic topical corticosteroids. Given the severe inflammatory nature of scleritis and chronic use of corticosteroids required for treatment of episcleritis and scleritis, we anticipated that patients would have a high incidence of cataract formation. Known risk factors for cataract include age, use of corticosteroids, diabetes mellitus, and sunlight exposure, amongst others6.
Knowledge of the magnitude of cataract risk and of potentially modifiable predictive factors for eyes with scleritis/episcleritis is limited and could aid in clinical management choices and counseling. Here, we estimate the incidence of cataract and predictive factors for cataract formation in a large cohort of eyes with scleritis and episcleritis at risk of cataract seen at US tertiary academic referral centers. We also evaluated the visual outcome after cataract surgery and associated factors in patients with scleritis and episcleritis.
Methods:
Study design/primary outcome:
A large retrospective cohort study, the Systemic Immunosuppressive Therapy for Eye Disease (SITE), was conducted at tertiary care centers from center inception (earliest date 1978) to the end of 2010 inclusive; the study protocol that has been published previously7,8 was extended through December 31, 2010.
This study examined two cohorts of these eyes/patients. The first cohort consisted of those with non-infectious scleritis and episcleritis and at risk of cataract (phakic with presenting visual acuity better than 20/40 at cohort entry). Patients with infectious ocular disease or diagnosis of human immunodeficiency virus infection had been excluded from the parent study. Individuals were followed in this retrospective cohort study from initial visit until the end of follow-up or the visit at which development of cataract was documented, with either of the following qualifying as the outcome - either the reduction of visual acuity worse than 20/40 which extant records attributed to cataract, or documented cataract surgery.
The second cohort, an incompletely overlapping group, consisted of eyes with scleritis and episcleritis that underwent cataract surgery with at least one pre-operative visit within three months of surgery and at least one post-operative visit during the year after cataract surgery. This group was independently evaluated to examine visual outcomes after cataract surgery, including potential predictive factors.
The research was conducted under approval by the institutional review board at each of the participating centers and adhered to the tenets of the Declaration of Helsinki.
Data collection:
Data for eligible patients were obtained in the parent study for each eye at each visit from medical records and entered into a computer-based standardized data entry form by trained expert reviewers as previously described7,8. Covariates obtained (see Table 1a,b) included patient demographic characteristics, type of scleral inflammation, bilateral versus unilateral involvement, degree of inflammatory activity (graded as inactive, active, and slightly inactive—the latter defined as “activity that is minimally present, described as mild, few, trace, etc.”), anterior chamber inflammation4 (graded using criteria later adopted by the SUN group9, intraocular pressure, use of topical and/or oral corticosteroids, and smoking status. Specifically, the eyes were categorized into episcleritis, anterior scleritis or sclerokeratitis, posterior scleritis, and necrotizing scleritis (including anterior and posterior necrotizing scleritis). Prednisolone acetate 1%, or bioequivalent doses of alternative topical corticosteroids, was studied. Cases treated with difluprednate were excluded from the primary analysis since the bioequivalent dose is not yet clear for that drug. The large majority of oral corticosteroids used were prednisone; bioequivalent doses of alternative systemic corticosteroids were calculated when prednisone itself was not used. The topical and oral corticosteroid covariates both were coded as any use vs. no use given the limited number of events. Usages of aspirin, statins, non-steroidal anti-inflammatory drugs (NSAIDs), and angiotensin converting enzyme inhibitors (ACEIs) at any time during follow-up also were evaluated as we had hypothesized they may potentially alter cataract incidence.
Table 1.
Baseline eye-specific characteristics of patients with scleritis and episcleritis.
| Right eye | Left eye | All eyes | |
|---|---|---|---|
| N (%) | N(%) | N (%) | |
| Anterior chamber cells | |||
| Quiet or 0.5+ | 660 (98) | 685(96) | 1345 (97) |
| 1–4+ | 14 (2) | 25 (4) | 39 (3) |
|
| |||
| Bilateral disease | |||
| Ever | 514 (72) | 514 (68) | 1028 (70) |
| Never | 201 (28) | 244 (32) | 445 (30) |
|
| |||
| Intraocular pressure (mmHg) | |||
| <10 | 13 (2) | 8 (1) | 21 (2) |
| 10–20 | 523 (78) | 557 (79) | 1080 (79) |
| 21–29 | 120 (18) | 126 (18) | 246 (18) |
| ≥ 30 | 12 (2) | 14 (2) | 26 (2) |
|
| |||
| Periocular corticosteroid | |||
| None | 712 (100) | 753 (99) | 1465 (99) |
| Any | 3 (0) | 5 (1) | 8 (1) |
|
| |||
| Systemic corticosteroid (mg/day) | |||
| None | 515 (72) | 547 (72) | 1062 (72) |
| ≤ 7.5 | 28 (4) | 33 (4) | 61 (4) |
| > 7.5 | 172 (24) | 178 (23) | 350 (24) |
|
| |||
| Topical corticosteroids (number of drops/day) | |||
| None | 544 (76) | 566 (75) | 1110 (75) |
| 1–4 | 109 (15) | 134 (17) | 243 (16) |
| 5–8 | 42 (6) | 39 (5) | 81 (6) |
| > 8 | 20 (3) | 20 (3) | 39 (3) |
|
| |||
| Scleritis type | |||
| Episcleritis | 181 (25) | 191 (25) | 372 (25) |
| Anterior Scleritis | 443 (62) | 472 (62) | 915 (62) |
| Sclerokeratitis | 21 (3) | 22 (3) | 43 (3) |
| Posterior scleritis | 37 (5) | 32 (4) | 69 (5) |
| Necrotizing scleritis | 33 (5) | 41 (5) | 74 (5) |
|
| |||
| Area of involvement | |||
| Diffuse | 286 (76) | 305 (76) | 591 (76) |
| Nodular | 88 (24) | 95 (24) | 183 (24) |
For the analysis of visual outcomes in eyes undergoing cataract surgery, additional variables were recorded from the pre-operative visit for analysis, including topical corticosteroid, systemic corticosteroid and other immunosuppressive therapy use, duration of inflammation, pre-operative presenting visual acuity, cataract surgery method, and year of surgery. For the primary analysis, the post-operative visual acuity at the visit closest to one year after cataract surgery was used. For eyes with visual acuity worse than 20/40, the primary cause for decreased vision was obtained from the patient record and tabulated.
Statistical analysis:
Survival analysis with univariable and multivariable Cox regression models were utilized to assess the incidence and possible predictive factors of cataract development using unadjusted and adjusted hazard ratios (HR), accounting for correlation between eyes of the same patient. In preliminary analyses counting cataract as present based on either the visual acuity or occurrence of surgery definition, overall activity of scleritis was associated with lower cataract incidence, corresponding to the usual ophthalmologist practice of operating on quiet eyes—a potential bias in the analysis. Therefore, we limited the predictive factor analysis to cases of cataract based on the visual acuity definition alone so that we could conduct an unbiased assessment of inflammatory activity as a potential predictive factor for cataract development. Analyses adjusted for age as well as other covariates in a full multivariable model, and a multivariable model that went through stepwise variable selection.
Univariable and multivariable logistic regression models were utilized to determine the factors associated with a post-operative visual outcome of 20/40 or better one year after cataract surgery in eyes with episcleritis and scleritis. Factors with p<0.05 in univariable analysis were included into the multivariable logistic regression model.
All statistical analyses were performed with SAS software version 9.4 (SAS Inc., Cary, NC).
Results:
One thousand three hundred eight-four eyes of 1008 patients with scleritis and episcleritis were included in the study of cataract incidence. Initial characteristics are given in Table 1a for patient medical history and systemic medications that could potentially accelerate cataract development, and Table 1b for eye-specific characteristics. Among patients with one or more eyes at risk for cataract, 659 (65%) were female and 706 (70%) were white. Among eyes at risk, 915 (62.1%) had non-necrotizing anterior scleritis, 43 (3%) had sclerokeratitis, 69 (5%) had posterior scleritis, 74 (5.0%) had necrotizing scleritis, and 372 (25%) had episcleritis. Of those with anterior scleritis and episcleritis, the majority were diffuse (n=591, 76%) versus nodular (n=183, 24%).
A total of 76 eyes developed cataract over 3030 eye-years (incidence=0.025/eye-year, 95% confidence interval (CI): 0.019, 0.031) (Figure 1). Among these, 39 were identified due to decline in presenting visual acuity to worse than 20/40, which was confirmed on chart review to be due primarily to cataract; and 37 eyes underwent cataract surgery.
Figure 1.
Cumulative cataract-free survival among eyes with scleritis initially free of cataract with 95% confidence interval in hatched lines.
Eyes with necrotizing scleritis, which were few in number, tended to have a higher incidence of cataract compared to other patterns of (epi-) scleral inflammation (crude hazard ratio (HR)=4.46; 95% CI 1.15–17.34, see Table 2). The incidence of cataract did not differ significantly among the other forms of scleral inflammation.
Table 2.
Cox regression analysis of risk factors to cataract development in eyes with episcleritis and scleritis.
| Age-adjusted analysis | Stepwise multivariate analysis | |||
|---|---|---|---|---|
| HR (95% CI) | p-value | HR (95% CI) | p-value | |
| Age | ||||
| <65 | Reference | 0.0001 | ||
| 65+ | 3.85 (1.94–7.63) | |||
|
| ||||
| Topical corticosteroids | ||||
| None | Reference | 0.237 | Reference | 0.542 |
| 1+ drops | 1.50 (0.77–2.93) | 1.29 (0.57–2.95) | ||
|
| ||||
| Oral corticosteroids | ||||
| None | Reference | 0.0004 | Reference | 0.0004 |
| Any | 3.35 (1.71–6.55) | 3.42 (1.73–6.74) | ||
|
| ||||
| Race | ||||
| White | Reference | 0.874 | ||
| Non-white | 1.06 (0.54–2.08) | |||
|
| ||||
| Diabetes | ||||
| No | Reference | 0.008 | Reference | 0.0460 |
| Yes | 2.77 (1.30–5.88) | 2.40 (1.01–5.69) | ||
|
| ||||
| Intraocular pressure | ||||
| <10 | 4.08 (1.19–14.01) | 0.025 | 3.23 (0.87–11.35) | 0.079 |
| 10–19 | Reference | Reference | ||
| 20–29 | 1.08 (0.42–2.80) | 0.873 | 0.93 (0.35–2.46) | 0.875 |
| ≥ 30 | 5.90 (1.27–27.34) | 0.023 | 6.13 (1.16–32.27) | 0.033 |
|
| ||||
| Scleritis type | ||||
| Episcleritis | Reference | |||
| Anterior/sclerokeratitis | 1.87 (0.65–5.42) | 0.247 | ||
| Posterior | 1.67 (0.31–8.97) | 0.551 | ||
| Necrotizing | 4.46 (1.15–17.34) | 0.031 | ||
|
| ||||
| Smoker | ||||
| Never | Reference | |||
| Former | 1.50 (0.70–3.21) | 0.300 | ||
| Current | 0.81 (0.28–2.33) | 0.700 | ||
|
| ||||
| Concurrent Aspirin use | 1.85 (0.79–4.35) | 0.160 | ||
|
| ||||
| Concurrent Statin use | 1.36 (0.60–3.09) | 0.456 | ||
|
| ||||
| Concurrent NSAID use | 1.16 (0.57–2.35) | 0.680 | ||
|
| ||||
| Concurrent ACE inhibitor use | 1.33 (0.54–3.29) | 0.538 | ||
|
| ||||
| Anterior chamber cells | ||||
| Quiet, 0.5+ | Reference | 0.0005 | Reference | 0.022 |
| 1–4+ | 9.15 (0.6–32.03) | 4.79 (1.26–18.27) | ||
|
| ||||
| Bilateral disease | 0.73 (0.37–1.45) | 0.365 | ||
|
| ||||
| Inflammatory activity | ||||
| Inactive | Reference | |||
| Slightly active | 2.16 (0.90–5.16) | 0.085 | ||
| Active | 1.41 (0.68–2.92) | 0.361 | ||
As expected, increased age (greater than or equal to vs less than 65 years - aHR 3.85; 95% CI 1.94–7.63) was associated with increased risk of cataract. Time-updated intraocular pressure greater or equal to 30mmHg also was associated with increased cataract incidence (aHR 6.13; 95% CI 1.16–32.37), although no association was seen with more moderate elevations in intraocular pressure. Time-updated use of oral corticosteroids (aHR 3.42; 95% CI 1.73–6.74) was significantly associated with increased cataract risk compared with non-use (association for oral corticosteroids was similar for those taking 7.5 mg/day prednisone-equivalent and higher doses versus lower doses; data not shown). However, no association was seen between use of topical corticosteroid and cataract formation (aHR 1.29; 95% CI 0.57–2.95). Anterior chamber cell inflammation (aHR 4.79; 95% Cl 1.26–18.27) and a history of diabetes (aHR 2.40; 95% CI 1.01–5.69) both were associated with development of cataract.
Race, patient smoking history and use of aspirin, NSAIDs, statins, or ACEIs all were unassociated with cataract incidence. Neither bilaterality versus unilaterality of scleritis nor overall inflammatory activity of scleritis (adjusted for anterior chamber cell grade) had a statistically significant association with cataract development (see Table 2).
In an overlapping but non-identical subset of the SITE cohort, 79 cataractous eyes of 65 patients with episcleritis and scleritis, meeting pre-operative and post-operative follow-up criteria underwent cataract surgery. Average patient age at time of surgery was 65.6 (SD 10.3) years. Patients/eyes were seen an average of 1.2 months (SD 1.24 months) prior to surgery and followed for a median of 35.7 (IQR 19.5 – 53.7) months post-operatively. The median duration of episcleritis or scleritis prior to surgery was 32.4 (IQR 21.0 – 51.6) months. Eye-specific characteristics at the pre-operative visit are shown in Table 3. Median presenting visual acuity improved by a median of 6 ETDRS lines 12 months postoperatively. None of the covariates studied showed any significant positive or negative association with visual acuity 20/40 or better one year after cataract surgery, including preoperative status of inflammatory activity, duration of inflammation, pre-operative visual acuity, and topical and systemic corticosteroid use (Table 4). The documented primary cause for decreased vision one year after surgery were heterogeneous, with macular scarring (n = 4, 18.2%) and corneal sequelae (n = 3, 13.6%) being most common (Table 5). The latter included one patient with history of corneal transplant and angle closure, one with a history of Descemet Stripping Automated Endothelial Keratoplasty (DSAEK), and one with a history of repeated pars plana vitrectomies with a scleral buckle and use of silicone oil. The distribution of visual acuity one year post-operatively remained stable up until at least 60 months, though some eyes were lost to follow-up (Figure 2).
Table 3.
Pre-operative characteristics of scleritis and episcleritis eyes that underwent cataract surgery.
| All eyes, N (%) | |
|---|---|
| Age (years) | |
| <65 | 37 (46.8) |
| 65+ | 42 (53.2) |
|
| |
| Gender | |
| Female | 63 (79.7) |
| Male | 16 (20.3) |
|
| |
| Race | |
| White | 57 (72.2) |
| Black | 11 (13.9) |
| Hispanic | 5 (6.3) |
| Other | 6 (7.6) |
|
| |
| Smoker | |
| Never | 54 (68.4) |
| Former | 11 (13.9) |
| Current | 9 (11.4) |
|
| |
| Topical corticosteroids/day | |
| None | 53 (67.1) |
| 1–2 drops | 17 (21.5) |
| >2 drops | 9 (11.4) |
|
| |
| Systemic corticosteroid (mg/day) | |
| None | 40 (50.60) |
| ≤ 7.5 | 19 (24.1) |
| > 7.5 | 20 (25.3) |
|
| |
| Immunotherapy use | |
| Never | 33 (41.8) |
| IST | 35 (44.3) |
| Biologic | 3 (3.8) |
| IST and biologic | 8 (10.1) |
|
| |
| Inflammatory activity | |
| Inactive | 69 (87.3) |
| Slightly active | 3 (3.8) |
| Active | 6 (7.6) |
| Unknown | 1 (1.3) |
|
| |
| Duration of inflammation | |
| < 2 years | 54 |
| > 2 years | 25 |
|
| |
| Preoperative visual acuity | |
| 20/50 – 20/80 | 26 (32.9) |
| 20/100 – 20/200 | 19 (24.1) |
| > 20/200 | 34 (43.0) |
|
| |
| Cataract surgery method | |
| Pars plana lensectomy | 3 (3.8) |
| ECCE | 8 (10.1) |
| Phacoemulsification | 64 (81.0) |
| Unknown | 4 (5.1) |
|
| |
| Intraocular placement | |
| Yes | 75 (94.9) |
| No | 4 (5.1) |
|
| |
| Year surgery performed | |
| < 2000 | 14 (17.7) |
| 2000+ | 65 (82.3) |
|
| |
| Type of scleritis | |
| Necrotizing scleritis | 12 (15.2) |
| Posterior scleritis | 5 (6.3) |
| Anterior scleritis | 58 (73.4) |
| Episcleritis | 4 (5.1) |
|
| |
| Follow-up duration (months) | |
| 6 – 12 | 9 (11.4) |
| 13 – 24 | 16 (20.3) |
| 25 – 36 | 15 (19.0) |
| 37 – 48 | 13 (16.5) |
| 49 – 60 | 12 (15.2) |
| > 60 | 14 (17.7) |
Table 4.
Logistic regression analysis of factors associated with visual acuity 20/40 or better one year after cataract surgery in eyes with episcleritis and scleritis.
| Unadjusted OR | Multivariate | |||
|---|---|---|---|---|
| HR (95% CI) | p-value | HR (95% CI) | p-value | |
| Age (years) | ||||
| <65 | Reference | 0.563 | ||
| 65+ | 0.72 (0.23–2.21) | |||
|
| ||||
| Gender | ||||
| Female | Reference | 0.766 | ||
| Male | 1.23 (0.31–4.81) | |||
|
| ||||
| Smoker | ||||
| Never | Reference | 0.237 | ||
| Former | 0.19 (0.02–1.62) | |||
| Current | 0.53 (0.17–1.69) | |||
|
| ||||
| Topical corticosteroids | ||||
| None | Reference | 0.3436 | ||
| 1+ drops | 0.61 (0.22–1.69) | |||
|
| ||||
| Systemic corticosteroid (mg/day) | ||||
| None | Reference | 0.956 | ||
| ≤ 7.5 | 1.06 (0.30–3.78) | |||
| > 7.5 | 0.89 (0.26–3.04) | |||
|
| ||||
| Immunotherapy use | ||||
| Never | Reference | 0.1050 | ||
| IST or biologic | 0.34 (0.11–1.04) | |||
| IST and biologic | 1.56 (0.15–15.70) | |||
|
| ||||
| Inflammatory activity | ||||
| Inactive | Reference | 0.061 | Reference | 0.115 |
| Active/slightly active | 0.26 (0.06–1.06) | 0.35 (0.08–1.50) | ||
|
| ||||
| Duration of inflammation | ||||
| < 2 years | Reference | 0.642 | ||
| > 2 years | 1.33 (0.40–4.49) | |||
|
| ||||
| Preoperative visual acuity | ||||
| 20/50 – 20/80 | Reference | 0.060 | Reference | 0.156 |
| 20/100 – 20/200 | 0.37 (0.08–1.72) | 0.37 (0.08–1.74) | ||
| > 20/200 | 0.19 (0.05–1.06) | 0.21 (0.05–0.92) | ||
|
| ||||
| Cataract surgery method | ||||
| Not phacoemulsification | Reference | 0.267 | ||
| Phacoemulsification | 2.00 (0.59–6.80) | |||
|
| ||||
| Intraocular lens placement | ||||
| Yes | Reference | 0.165 | ||
| No | 0.36 (0.09–1.52) | |||
|
| ||||
| Year surgery performed | ||||
| < 2000 | Reference | 0.168 | ||
| 2000+ | 2.30 (0.71–7.48) | |||
Table 5.
Documented primary cause for visual acuity 20/50 or worse one year after cataract surgery.
| N(%) | |
|---|---|
| Corneal sequelae | 3 (13.6) |
| Posterior capsular opacification | 1 (4.6) |
| Glaucoma | 2 (9.1) |
| Cystoid macular edema | 2 (9.1) |
| Epiretinal membrane | 1 (4.6) |
| Macular scar | 4 (18.2) |
| Inflammatory haze | 1 (4.6) |
| Refractive error | 2 (9.1) |
| Other/Unknown | 5 (22.7) |
Figure 2.
Median visual acuity in logMAR months after surgery and number of eyes.
Discussion:
The 0.025/eye-year incidence of cataract in episcleritis/scleritis was relatively low, less than seen in eyes with anterior (0.043/eye-year)10 and intermediate (0.076 /eye-year) uveitis2 from the same parent study, despite an older age distribution in our episcleritis/scleritis population compared to those in the anterior and intermediate uveitis studies2,10. The majority of our eyes had anterior scleritis or sclerokeratitis, neither of which had an increased risk of cataract compared to eyes with episcleritis. Eyes with posterior scleritis did not have an increased risk either. In those with necrotizing scleritis, there was a statistically significant increase in cataract incidence when adjusting only for age, but in an initial analysis adjusting for all other variables including corticosteroid use, this was no longer seen, suggesting that the apparent association might be due to confounding (data not shown).
Use of oral corticosteroids significantly increased incidence of cataract, but no difference in effect was observed between low and high oral doses (greater than prednisone 7.5mg/day) in sub-analyses (data not shown). Oral corticosteroids are well-known to predispose to cataract formation, although the dose and duration required to induce clinically important cataract have been unclear6,11 and may vary among different persons. Notably, the centers participating in our study were pioneers of early corticosteroid-sparing immunomodulatory therapy in the treatment of scleritis, so it is unlikely that corticosteroid use was excessive.
Topical corticosteroids at doses of two drops per day or higher have been associated with increased risk of cataract in anterior10 and intermediate uveitis cases1. In our study, any use of topical corticosteroids was not associated with increased cataract risk though confidence intervals allowed for the possibility of moderately increased risk given the moderate sample size. Notably, topical corticosteroids in our study included prednisolone acetate 1% and bioequivalent doses of alternatives, excluding difluprednate. Eyes treated with difluprednate, which was introduced in 2008, were excluded from our study due to the higher potency and unclear bioequivalency of difluprednate compared with prednisolone acetate 1%12. Further assessment of whether there is increased risk with difluprednate is warranted since a recent observational small cohort study suggested difluprednate as a potentially effective non-systemic therapy for scleritis13.
Other factors that increased cataract incidence were the well-known risk factor of age6, as well as time-updated elevation of intraocular pressure to 30 mmHg or higher and anterior chamber inflammation. Prior studies in children14,15 and adults with anterior uveitis have shown a positive association between cataract formation and anterior chamber inflammatory activity10, although our study in intermediate uveitis did not find the same1. While anterior chamber activity can be thought of as a marker of disease severity intraocularly, it is likely that its effect in scleritis may primarily occur if there is an associated anterior uveitis (where the inflammation is adjacent to the lens).
As prior studies have shown, a history of diabetes6 is associated with greater risk of cataract. Some have found the use of statins and ACEIs to increase risk of cataract16,17, although others18, like our study, have not. Use of aspirin and NSAIDs also had no impact, which is supported by most studies6. Prior studies have shown a history of smoking19 to increase cataractogenesis, but our study did not confirm this, which could be due to insufficient power to detect smaller effects. The differences in conclusions amongst studies may be due, at least in part, to differing study designs, limited and different study populations, and varied selection of covariates that results in variable control of confounders or effect mediators. However, the drugs in question probably do not have large beneficial effects of a degree sufficient to warrant further study for the prevention of cataracts.
Once patients with episcleritis/scleritis develop cataract, our study is encouraging in that there is no significant difference in the odds of post-operative visual acuity 20/40 or better at one year, despite differences in pre-operative factors like duration of inflammation, immunotherapy use, and corticosteroid use. With inflammatory activity, there was also no statistical significance between the two groups compared, those with inactive inflammation and those with slightly active or active inflammation, which were combined due to small group size. In general, the distribution of post-operative visual acuity at three months remained stable for up to at least five years overall. While sterile cornea melts have been known to occur post-cataract surgery in patients with rheumatoid arthritis and collagen vascular disease, and both conditions associated with scleritis20, corneal and/or scleral melts requiring surgical stabilization was not observed in our cohort. The causes of reduced visual acuity in the relatively few cases with postoperative visual acuity 20/50 or worse were heterogeneous.
Our study of the incidence of cataract in eyes with episcleritis/scleritis benefits from a relatively large initial population with long follow-up time for a rare disease. However, the number of patients who developed cataract over the course of the study was still relatively few, which is good news for patients but limited our ability to study potentially modifiable predictive factors. Nevertheless, several predictive factors were identified. A number of eyes which underwent cataract surgery did not qualify as having a cataract under our definition based on visual acuity, which reduced the number of events for the predictive factor analyses in order to avoid biases related to the practice pattern of generally avoiding surgery in eyes with inflammatory activity. Similarly, the number of cataractous eyes was relatively few, which may have hampered our ability to detect predictive factors. Like all retrospective studies, our study has issues of incomplete follow up and missing data that might have skewed results if the ocular observations available systematically and substantially differed from those unavailable during follow-up, which is not likely to have occurred. The type of cataract was not assessed in our study; this characterization would be valuable in future studies related to elucidation of cataract etiology but are less important to guide clinical management. The type of intraocular lens used, perioperative corticosteroid dosing and use of intracameral medications and any intraoperative complications were not captured by our database, and are limitations of this analysis. In addition, the SITE Cohort Study was performed at tertiary uveitis referral centers wherein patients may have a greater average degree of disease severity than in the general population, which may limit its generalizability to non-tertiary populations, though results should be generalizable to tertiary centers wherein outcomes of cataract surgery appear to be considerably better21.
In summary, the incidence of cataract in episcleritis and scleritis is relatively low, less than in uveitis1,10 even though corticosteroids frequently are used in episcleritis and scleritis treatment. Factors that increased risk for cataract development in our study include use of oral corticosteroids, higher age, highly elevated intraocular pressure, and anterior cell inflammation. Wider use of corticosteroid-sparing immunosuppression might help to avoid cataract in scleritis management, particularly if complete cessation of corticosteroids can be achieved. Visual outcome of cataract surgery was generally positive, with a median six-line improvement in the presenting visual acuity distribution that was maintained through at least five years. Regardless of the course of the disease, patients with episcleritis and scleritis undergoing cataract surgery can be reassured that the average outcome is favorable and sustained with respect to visual outcomes.
Funding/Support:
Primary support from grants EY014943 and EY026717 (Dr. Kempen), EY022445 (Dr. Rosner), and 1R21 EY032592-01 (Dr. Gangaputra), National Eye Institute, Bethesda, MD, USA. Additional support from: Laboratory of Immunology, National Eye Institute, Bethesda, MD, USA (Drs. Levy-Clarke, Sen); Research to Prevent Blindness, New York, NY, USA (Drs. Kempen, Jabs, Rosenbaum, unrestricted department support); Sight for Souls, Bellevue, WA, USA (Dr. Kempen), Portland Veterans Affairs Medical Center, Portland, WA, USA (Dr. Suhler)
Footnotes
Financial disclosures:
JHK: Equity owner – Betaliq, Tarsier Pharma
CSF: Consultant, Grant support – Aldeyra, Bausch and Lomb, Novartis, pSivida, Consultant- Allakos, Genentech
JET: Consultant, AbbVie, Canfield, Clearside, Gildead, Guidepoint, Roche, UptoDate, F.Hoffmann-La Roche AG, Scientific Advisory Board, Equity Owner – Tarsier Pharma.
DAJ: Research funding – AbbVie
EBS: Consultant – Eyevensys, Santen, Consultant, Financial Support – EyeGate, AbbVie, Clearside, EyePoint, Other non-financial interests – Gilead
JTR: Employment – Corvus Pharmaceuticals, Consultant – AbbVie, Gilead, Janssen, Eyevensys, UptoDate, Novartis, Roche, Celgene, UCB, Santen, UCB, Financial support – Pfizer, Alcon Research Institute Consultant and Financial support – Horizon, Education – Pfizer
HNS: Employee – Sanofi Research and Development
GAL-C: Consultant, Lecture Fees – AbbVie, Grant Support – Allergan, Consultant, Grant support – Mallinckrodt, Grant support, Lecture fees – Sanofi
NPB: Other non-financial disclosures – Allergan
All other authors have no conflicting interests to disclose
References
- 1.Minkus C, Begum H, Kaçmaz RO, Jabs DA, Nussenblatt RB, Rosenbaum JT. Risk of Cataract in Intermediate Uveitis. Am J Ophthalmol 2021;229:200–209. doi: 10.1016/j.ajo.2021.02.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kempen JH, Pistilli M, Begum H, et al. Remission of Non-Infectious Anterior Scleritis: Incidence and Predictive Factors. Am J Ophthalmol 2021;223:377–395. doi: 10.1016/j.ajo.2019.03.024 [DOI] [PubMed] [Google Scholar]
- 3.Braithwaite T, Adderley NJ, Subramanian A, et al. Epidemiology of Scleritis in the United Kingdom From 1997 to 2018: Population-Based Analysis of 11 Million Patients and Association Between Scleritis and Infectious and Immune-Mediated Inflammatory Disease. Arthritis and Rheumatology. 2021;73(7):1267–1276. doi: 10.1002/art.41709 [DOI] [PubMed] [Google Scholar]
- 4.Jabs DA, Nussenblatt RB, Rosenbaum JT, et al. Standardization of uveitis nomenclature for reporting clinical data. Results of the first international workshop. Am J Ophthalmol 2005;140(3):509–516. doi: 10.1016/j.ajo.2005.03.057 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Papaliodis GN, Rosner BA, Dreger KA, et al. Incidence of and Risk Factors for Cataract in Anterior Uveitis. Am J Ophthalmol 2023;254:221–232. doi: 10.1016/j.ajo.2023.06.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hodge WG, Whitcher JP, Satariano W. Risk Factors for Age-related Cataracts. Epidemiol Rev 1995;17(2):336–346. [DOI] [PubMed] [Google Scholar]
- 7.Kempen JH, Daniel E, Gangaputra S, et al. Methods for identifying long-term adverse effects of treatment in patients with eye diseases: The Systemic Immunosuppressive Therapy for Eye Diseases (SITE) Cohort Study. Ophthalmic Epidemiol 2008;15(1):47–55. doi: 10.1080/09286580701585892 [DOI] [PubMed] [Google Scholar]
- 8.Kempen JH. Overall and cancer related mortality among patients with ocular inflammation treated with immunosuppressive drugs : retrospective cohort study. Published online 2009. doi: 10.1136/bmj.b2480 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Jabs DA, Nussenblatt RB, Rosenbaum JT S of UN (SUN) WG. Standardization of Uveitis Nomenclature for Reporting Clinical Data. Results of the First International Workshop. Am J Ophthalmol 2005;140(3):509–516. doi: 10.1016/j.ajo.2005.03.057.Standardization [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Artornsombudh P, Kothari S, Gangaputra SS, et al. Incidence of and Risk Factors for Cataract in Anterior Uveitis. Am J Ophthalmol 2023;254:221–232. doi: 10.1016/j.ajo.2023.06.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wang JJ, Rochtchina E, Tan AG, Cumming RG, Leeder SR, Mitchell P. Use of Inhaled and Oral Corticosteroids and the Long-term Risk of Cataract. Ophthalmology. 2009;116(4):652–657. doi: 10.1016/j.ophtha.2008.12.001 [DOI] [PubMed] [Google Scholar]
- 12.Jamal KN, Callanan DG. The role of difluprednate ophthalmic emulsion in clinical practice. Clinical Ophthalmology. 2009;3:381–390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Liberman P, Burkholder BM, Thorne JE, Berkenstock MK. Effectiveness of Difluprednate for the Treatment of Anterior Scleritis. Am J Ophthalmol 2022;235:172–177. doi: 10.1016/j.ajo.2021.09.008 [DOI] [PubMed] [Google Scholar]
- 14.Blum-Hareuveni T, Seguin-Greenstein S, Kramer M, et al. Risk Factors for the Development of Cataract in Children with Uveitis. Am J Ophthalmol 2017;177:139–143. doi: 10.1016/j.ajo.2017.02.023 [DOI] [PubMed] [Google Scholar]
- 15.Thorne JE, Woreta FA, Dunn JP, Jabs DA. Risk of cataract development among children with juvenile idiopathic arthritis-related uveitis treated with topical corticosteroids. Ophthalmology. 2010;117(7):1436–1441. doi: 10.1016/j.ophtha.2009.12.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Macías Saint-Gerons D, Bosco Cortez F, Jiménez López G, Castro JL, Tabarés-Seisdedos R. Cataracts and statins. A disproportionality analysis using data from VigiBase. Regulatory Toxicology and Pharmacology. 2019;109:104509. doi: 10.1016/J.YRTPH.2019.104509 [DOI] [PubMed] [Google Scholar]
- 17.Kanthan GL, Wang JJ, Rochtchina E, Mitchell P. Use of antihypertensive medications and topical beta-blockers and the long-term incidence of cataract and cataract surgery. British Journal of Ophthalmology. 2009;93(9):1210–1214. doi: 10.1136/bjo.2008.153379 [DOI] [PubMed] [Google Scholar]
- 18.Becker C, Jick SS, Meier CR. ACE inhibitor use and risk of cataract: A case-control analysis. British Journal of Ophthalmology. 2019;103(11):1561–1565. doi: 10.1136/bjophthalmol-2018-312980 [DOI] [PubMed] [Google Scholar]
- 19.Beltran-Zambrano E; Garcia-Lozada D; Ibanez-Pinilla E Risk of Cataract in Smokers: A Meta-Analysis of Observational Studies. Arch Soc Esp Oftalmol 2019;94(2):60–74. [DOI] [PubMed] [Google Scholar]
- 20.Perez VL, Azar DT, Foster CS. Sterile corneal melting and necrotizing scleritis after cataract surgery in patients with rheumatoid arthritis and collagen vascular disease. Semin Ophthalmol 2002;17(3–4):124–130. doi: 10.1076/soph.17.3.124.14786 [DOI] [PubMed] [Google Scholar]
- 21.Pistilli M, Gangaputra SS, Pujari SS, et al. Contemporaneous Risk Factors for Visual Acuity in Non-infectious Uveitis. Ocul Immunol Inflamm 2021;29(6):1056–1063. doi: 10.1080/09273948.2020.1828493 [DOI] [PubMed] [Google Scholar]