The study cohort consisted of 26 eyes in 21 children, of which 13 were female and 8 were male. The mean age was 10.5 years (SD 3.3). Sixteen patients (76.2%) had unilateral macular edema (Table 1). Twenty-two eyes out of 26 eyes (84.6%) had active uveitis at the time of initial presentation of ME at UCSF. The average duration of follow-up was 35.3 months (SD 25.7).
Table 1: Demographic Characteristics of Patients with Non-infectious Uveitic Macular Edema (n=21)
Age
|
Mean 10.5 years, SD 3.3
|
Follow-Up
|
Mean # of follow-up visits
|
27.5, SD 18
|
Mean months of follow-up at UCSF
|
35.3, SD 25.7
|
|
# of people (%)
|
Sex
|
Male
|
8 (38.0%)
|
Female
|
13 (62.0%)
|
Laterality of uveitis
|
Unilateral
|
5 (23.8%)
|
Bilateral
|
16 (76.2%)
|
Race
|
Caucasian
|
4 (19.0%)
|
Hispanic/Latino
|
6 (28.6%)
|
African-American
|
2 (9.5%)
|
Asian
|
2 (9.5%)
|
Unknown
|
7 (33.3%)
|
The mean duration of uveitis at the time of ME diagnosis was 9.6 months (SD 11.2). Nineteen eyes (73.1%) had undifferentiated uveitis, 4 eyes (15.4%) had pars planitis, and 3 eyes (11.5%) had juvenile idiopathic arthritis-associated uveitis. Four eyes (15.3%) had anterior uveitis, 14 eyes (53.8%) had anterior/intermediate uveitis, 2 eyes (7.7%) had intermediate uveitis, and 6 eyes (23.1%) had panuveitis (Table 2). There was no association between the anatomical designation of uveitis and rate of macular edema resolution (P = 0.28). The morphology of the macular edema was cystoid in 9 eyes (34.6%), cystoid/diffuse in 5 eyes (19.2%), cystoid/sub-retinal fluid in 2 eyes (7.7%), diffuse in 6 eyes (23.1%), diffuse/sub-retinal fluid in 1 eye (3.8%) and unknown morphology in 3 eyes (11.5%). There was also no association between macular edema morphology and rate of resolution (Table 2, P = 0.16).
Table 2: Uveitis and Macular Edema Characteristics (n=26 eyes)
|
# of eyes (%)
|
Anatomic Designation
|
Anterior
|
4 (15.4 %)
|
|
Anterior-Intermediate
|
14 (53.8%)
|
|
Intermediate
|
2 (7.7%)
|
|
Panuveitis
|
6 (23.1%)
|
Uveitis Diagnosis
|
Idiopathic
|
19 (73.1%)
|
|
Pars Planitis
|
4 (15.4%)
|
|
JIA-associated
|
3 (11.5%)
|
Complications at baseline
|
Cataract
|
9 (34.6%)
|
|
Band Keratopathy
|
5 (19.2%)
|
|
Amblyopia
|
5 (19.2%)
|
|
Posterior Synechiae
|
3 (11.5%)
|
|
Retinal Vasculitis
|
3 (11.5%)
|
|
Glaucoma
|
1(3.8%)
|
Morphology of Macular Edema
|
Cystoid
|
9 (34.6%)
|
|
Cystoid/diffuse
|
5 (19.2%)
|
|
Cystoid/sub-retinal fluid
|
2 (7.7%)
|
|
Diffuse
|
6 (23.1%)
|
|
Diffuse/sub-retinal fluid
|
1 (3.8%)
|
|
Unknown
|
3 (11.5%)
|
Patients presented with one or more complications at baseline.
The most common pre-existing ocular complications were cataracts (9 eyes, 34.6%), followed by band keratopathy (5 eyes, 19.2%), amblyopia (5 eyes, 19.2%), posterior synechiae (3 eyes, 11.5%), retinal vasculitis (3 eyes, 11.5%), and glaucoma (1 eye, 3.8%) (Table 2). Some patients had a history of ocular surgery: 2 eyes (7.7%) with a history of cataract surgery, 2 eyes (7.7%) with retinal surgery, and 1 eye (3.8%) having undergone surgery for glaucoma.
Best-corrected visual acuity (BCVA) at baseline had a median of 0.5 logMAR (IQR 0.3 -1.0, n = 26 eyes) (Snellen 20/69, IQR 20/40 to 20/200). The median BCVA subsequently improved to 0.3 logMAR at 3 months (IQR 0.1 - 0.5, n = 26 eyes) (Snellen 20/40, IQR 20/25 to 20/69), 0.2 logMAR at 6 months (IQR 0.1 - 0.6, n = 26 eyes) (Snellen 20/32, IQR 20/25 to 20/79), and 0.2 logMAR at 9 months (IQR 0 - 0.5, n = 24 eyes) (Snellen 20/28, IQR 20/20 to 20/69). At the final 12-month follow-up visit, the median visual acuity was 0.1 logMAR (IQR 0 -0.4, n = 21) (Snellen 20/25, IQR 20/20 to 20/50) (Table 3, Figure 1).
Table 3: Median Visual Acuity and Central Macular Thickness (CMT) Over 12 Months Follow-Up
Outcome
|
Baseline
|
3 Months
|
6 Months
|
9 Months
|
12 Months
|
Median BCVA in logMAR
|
0.54
(IQR 0.3-1.0, n = 26)
|
0.30
(IQR 0.1-0.5, n = 26)
|
0.20
(IQR 0.1-0.6, n = 26)
|
0.15
(IQR 0-0.5,
n = 24)
|
0.1
(IQR 0-0.4,
n = 21)
|
Median BCVA in Snellen
|
20/69,
IQR 20/40 to 20/200
|
20/40,
IQR 20/25 to 20/69
|
20/32,
IQR 20/25 to 20/79
|
20/28,
IQR 20/20 to 20/69
|
20/25,
IQR 20/20 to 20/50
|
CMT (microns)
|
424
(IQR 352-620.5,
n = 23)
|
325
(IQR 283-347,
n = 21)
|
310
(IQR 298-325,
n=22)
|
305
(IQR 285-319,
n=19)
|
302
(IQR 281-340,
n = 18)
|
The median central macular thickness (CMT) decreased from 424 microns at baseline (IQR 352 - 620.5, n = 23 eyes) to 302 microns at 12 months (IQR 281.3 - 340.5, n = 18 eyes). The largest decrease in median CMT had occurred by the 3rd month, to 325 microns (IQR 283-347, n = 21 eyes) (Table 3, Figure 2). Eighteen of the 26 original eyes (69.2%) achieved resolution of their macular edema by the 12-month point. Among the eyes that achieved resolution, the median time to resolution of ME was 3 months (IQR 3 - 6). Ten (55.6%) of the 18 eyes with resolution of ME were free of ME by the 3-month visit. Six more eyes resolved by the 6-month visit and 2 more were free of ME by the 9th month of observation.
Treatment regimens varied by patient and over time, with some patients switching treatment over the duration of this study. Furthermore, some patients were treated with multiple types of medications simultaneously. Of the 26 eyes, 23 (88%) were treated with biologics or immunosuppressive treatment (i.e. antimetabolites) at some point in their treatment. A Cox proportional hazards multivariable model adjusting for each of the treatments found the following hazard ratios (HR) for macular edema resolution: biologics or immunosuppressive treatment HR (95% CI 0.4-21.2, P-value 0.32), oral prednisone (11 eyes, 42%) HR 0.76 (95% CI 0.3-2, P-value 0.56), topical steroid drops (15 eyes, 58%) HR 0.52 (95% CI 0.2-1.3, P-value 0.16) and intravitreal steroids injections/implants (5 eyes, 19%) HR 4.0 (95% CI 1.3-12.2, P-value 0.01). Patients who were ever treated with steroid injections had a significantly higher rate of ME resolution. The remaining types of treatment did not have a significant association with macular edema resolution.