Park DW, Folk JC, Whitcup SM, Polk TD, Kansupada K, Fountain C, Brown J, Nussenblatt RB. Phakic Patients With Cystoid Macular Edema, Retinal Periphlebitis, and Vitreous Inflammation. Arch Ophthalmol. 1998;116(8):1025-1029. doi:10.1001/archopht.116.8.1025
To characterize a group of phakic patients with idiopathic intermediate uveitis as defined by vitritis, cystoid macular edema, and retinal periphlebitis.
Nineteen phakic patients (35 eyes) with vitreous inflammation, cystoid macular edema, and/or retinal periphlebitis of unknown cause.
Main Outcome Measures
Best-corrected final visual acuities, standardized clinical examinations, photographic and fluorescein angiographic evaluations, and class I and II HLA analysis on all 19 patients.
Fifteen of the 19 patients were women. The mean age was 38 years, the mean follow-up was 104 months, and the mean duration of symptoms was 154 months. All 35 affected eyes had significant vitritis; 21 eyes (60%) had cystoid macular edema, 21 eyes (60%) had retinal periphlebitis. The median initial visual acuity was 20/30. The median final visual acuity was 20/20 with 32 (91%) of 35 eyes having 20/40 or better visual acuity at the final visit. No patient developed "snowbanks" or evidence of systemic disease, including multiple sclerosis or sarcoidosis, during the follow-up period. There were no statistically significant HLA associations in these patients compared with controls from another study from Iowa, but the Iowa phakic patients with cystoid macular edema did differ from the Iowa patients with pars-planitis at loci HLA-B8, HLA-B51, and HLA-DR2.
We describe a disease entity of idiopathic intermediate uveitis that affects primarily young to middle-aged women and usually causes bilateral vitritis, cystoid macular edema, and retinal periphlebitis. Most patients retained good vision over a prolonged follow-up period. Multiple sequential examinations and HLA associations suggest that these conditions are distinct from other syndromes of intermediate uveitis, particularly pars-planitis.
THE ETIOLOGY and pathogenesis of most uveitis remains elusive. Characterizing uveitic patients as having distinct syndromes such as acute multifocal posterior placoid pigment epitheliopathy, multiple evanescent white dot syndrome, bird shot chorioretinopathy, and others, is helpful for counseling patients as to the probable course and prognosis of their disease. It is also helpful as a first step in investigating the cause of the disease.
We have monitored 19 patients at The University of Iowa Hospitals and Clinics, Iowa City, with idiopathic intermediate uveitis defined by vitreous inflammation and usually cystoid macular edema (CME) or retinal periphlebitis. The main goals of this study were to determine the long-term visual prognosis of this group of patients, to determine whether the CME in these patients resembled CME in pseudophakic patients or patients with classic pars-planitis, and to determine whether there were any HLA antigen associations. Specifically, we wanted to determine whether this group of patients represented perhaps a more benign form of pars-planitis. We hoped to answer this question by determining how many of the patients developed pars plana exudates (ie, snowbanks) on follow-up and whether they had the same associations of HLA antigens and the risk of multiple sclerosis as reported by Malinowski et al1 and Tang et al.2
Nineteen patients (35 eyes) who had idiopathic intermediate uveitis as defined by vitreous inflammation with CME and/or retinal periphlebitis were followed up at The University of Iowa Hospitals and Clinics. All patients and affected eyes initially developed the uveitis while they were phakic. Each patient underwent an extensive evaluation including detailed history, physical examination, and laboratory testing. Each patient was provided a detailed explanation of the study and signed a consent form approved by The University of Iowa Hospitals and Clinics institutional review board. No patient had a history of skin rashes, gastrointestinal tract diseases, oral or genital ulcerations, vitiligo, or multiple sclerosis.
We defined intermediate uveitis as ocular inflammation that primarily involves the vitreous and peripheral retina.3 Pars-planitis is used to describe a subset of patients with intermediate uveitis of unknown cause. Patients with pars-planitis have vitreous cells, often CME, and snowbanks that occur over the pars plana and ora serrata.3 Sarcoidosis is a systemic granulomatous disease that commonly involves the lungs, lymph nodes, eyes, and other organs. While there is no one test that is diagnostic for sarcoidosis, a combination of clinical features and abnormal laboratory test results (including chest x-ray, serum angiotensin-converting enzyme levels, serum lysozyme levels, and lung or conjunctival biopsies) are used to make the diagnosis. Multiple sclerosis is demyelinating disease in which the clinical diagnosis is made by a reliable history of at least 2 episodes of neurologic deficits at more than 1 site in the central nervous system. Syphilis is caused by infection with Treponema pallidum and can be seen with a wide variety of clinical features. The diagnosis of syphilis is made using history, clinical signs, and laboratory tests of which the nonreactive VDRL and fluorescent treponemal antibody-adsorption assay (FTA-ABS) tests are particularly useful.
All patients in this study had normal serum angiotensin-converting enzyme values, complete blood cell counts, and nonreactive VDRL and fluorescent treponemal antibody-adsorption assay testings. Most patients underwent further laboratory testing, which showed either negative or normal values; these tests included erythrocyte sedimentation rate in 11 of 19 patients, chest x-ray in 10 of 19 patients, antinuclear antibody titer in 10 of 19 patients, purified protein derivative skin test in 7 of 19 patients, and rheumatoid factor in 7 of 19 patients. A few patients had other laboratory testing results that were either normal or negative, including serum lysozyme levels, Lyme disease serum titers, anti-Toxocara titers, serum antineutrophil cytoplasmic antibodies, Sjögren antibodies, and conjunctival biopsy. Patients with historical, physical, or laboratory evidence of other diseases, including pars-planitis, multiple sclerosis, sarcoidosis, syphilis, Vogt-Koyanagi syndrome, Behçet disease, or systemic lupus erythematosus, were excluded from the study.
Each patient underwent a complete ophthalmologic examination including best-corrected final visual acuity on the ETDRS chart, slitlamp examination, tonometry, slitlamp fundus biomicroscopy, and dilated indirect ophthalmoscopy with scleral depression. A careful examination with indirect ophthalmoscopy and scleral depression did not show snowballs or snowbanking at the final visit (or in any of the follow-up visits) in any of the 19 patients. No patient initially had any sign of multifocal choroiditis or any chorioretinal scars. One patient (patient 10) was found to have chorioretinal scars inferiorly during follow-up examination.
Blood samples from these 19 patients were sent to the National Institutes of Health, Bethesda, Md, for analysis of class I and class II HLA typing. The HLA frequencies from the 19 patients with phakic CME were compared with the HLA frequencies from a previously described group of Iowa patients with pars-planitis and an Iowa control group.1 Each locus was examined with a 2-way contingency table. These contingency tables were treated to a log-linear analysis and results from the log-linear analyses were partitioned into 3 pairwise comparisons of the antigens at each locus between patients with phakic CME and Iowa controls, patients with pars-planitis and Iowa controls, and patients with phakic CME and patients with pars-planitis.
High-quality fluorescein angiograms and color fundus photographs were available for 16 of the 19 patients. These photographs and fluorescein angiograms were compared with those from 7 patients with active pars-planitis and CME and 13 patients with CME after cataract extraction (ie, pseudophakic CME). The diagnoses were made at the time that the patients were examined in the clinic using the findings of the history, clinical examination, and laboratory testing. The photographs were graded by 2 masked readers who did not know the patients' diagnoses (J.C.F. and T.D.P.) and adjudicated by a third person (D.W.P.). Photographic standards for mild leakage in the fovea, severe leakage in the fovea, retinal venule dilatation, and other criteria were determined before grading the photographs. Photofiles of the patients with CME due to pseudophakia and pars-planitis were selected at random from those listed under these diagnoses at the University of Iowa and then choosing those with high-quality color photographs and fluorescein angiograms. Statistical analysis for group comparisons was performed using the Fisher exact test.
Table 1 lists the clinical characteristics of the patients. Fifteen of the 19 patients were women. The mean age at time of presentation was 38 years old (range, 10-66 years), and the mean follow-up was 104 months (range, 8-329 months). The mean duration of symptoms, including the time before the initial visit to The University of Iowa, was 154 months (range, 32-408 months). The disease was bilateral in 16 (92%) of 19 of patients.
All 35 affected eyes had significant vitritis. Twenty-one (60%) of 35 eyes had CME and 21 (60%) of 35 eyes had retinal periphlebitis observed sometime during the follow-up period. Three of 35 affected eyes (patient 11, left eye; patient 17, right eye; patient 18, right eye) had neither periphlebitis nor CME.
The median initial visual acuity was 20/30. The median final visual acuity was 20/20, with 32 (91%) of 35 eyes seeing 20/40 or better at the final visit. Eight (23%) of the 35 eyes had persistent CME at the last follow-up visit. In these 8 eyes, the median final visual acuity was 20/32. Of the eyes that did not have CME at the final visit, the median final visual acuity was 20/20. We reported the median rather than the mean visual acuity. The mean visual acuity did not accurately represent the visual acuity because there was a significant outlier visual acuity in the initial (patient 18) and final (patient 3) visual acuities (Table 1).
Thirteen (37%) of 35 eyes had anterior chamber inflammation, 8 (23%) of 35 eyes had keratic precipitates, and 5 (14%) of 35 eyes had posterior synechiae ob served during the course of their disease. In the eyes with anterior chamber inflammation, there were relatively few (1+) cells. Eight (35%) of 35 eyes had retinal hemorrhages, 2 had cotton-wool spots, 1 had vitreous hemorrhages, and 1 had iris rubeosis. Five (14%) of 35 eyes developed elevated intraocular pressures due to either corticosteroid use or open-angle glaucoma. No patient developed snowbanks during the follow-up period or by the final visit.
No patient developed multiple sclerosis during the follow-up period. Three patients did, however, have neurologic symptoms. Patient 19 had numbness in both lower arms due to carpal tunnel syndrome and patient 4 had a 6-week history of left-sided neck and arm numbness. Both patients underwent extensive evaluations by neurologists who concluded that these patients did not have multiple sclerosis. A referring physician believed that patient 16 possibly had an episode of optic neuritis but her visual fields were normal and there was no relative afferent pupillary defect. Patient 16 was also evaluated by a neurologist who performed a computed tomographic scan and the neurologist did not believe that this patient had multiple sclerosis.
The results of the color photograph and fluorescein angiogram comparison are presented in Table 2. The number of total eyes in the various subgroups of Table 2 varies because some of the grading criteria could not be assessed in all eyes because of limitations in the quality of photographs. Patients with phakic CME and vitritis (ie, idiopathic intermediate uveitis) in this study were more likely to have mild (55%) rather than severe leakage into the fovea on fluorescein angiography when compared with patients with pseudophakic CME (14%) and CME due to pars-planitis (38%); patients with phakic CME were also less likely to have severe leakage (23%) than patients with pseudophakic CME (50%) and CME due to pars-planitis (62%), (P=.002). The size of the foveal avascular zone (as measured with a reticle) was smaller in the group with phakic CME and vitritis (390 µm) than the group with CME due to pars-planitis (558 µm) (P=.03). However, the timing of the color photographs and fluorescein angiograms were not matched for duration of disease or level of visual acuity among the different groups, and thus, there may been a bias of ascertainment.
Patients with phakic CME and vitritis (ie, idiopathic intermediate uveitis) also had a significantly greater proportion of posterior pole fluorescein angiographic leakage (P=.005) (defined as the area from 1 disc diameter away from the foveal avascular zone center to the major retinal vascular arcades), retinal venule sheathing (P=.004), and dilated retinal venules (P<.01) when compared with the eyes with pseudophakic CME.
The results of the class I and II HLA frequencies in the Iowa patients with phakic CME, pars-planitis, and controls are presented in Table 3. There were no statistically significant associations in the patients with phakic CME as compared with the Iowa control group. The HLA associations in patients with pars-planitis, however, differed from both the Iowa control group and the phakic CME group at the loci HLA-B8, HLA-B51, and HLA-DR2. In comparing the patients with pars-planitis with the control group, there were statistically significant associations at HLA-B8, HLA-B51, and HLA-DR2 (all P<.001). In comparing the pars-planitis group with the phakic CME group, there were statistically significant associations for the pars-planitis group at HLA-B8 (P=.002), HLA-B51 (P<.001), and HLA-DR2 (P=.01). Therefore, although there were no statistically significant HLA differences between the Iowa phakic CME patients and the Iowa control group, the Iowa patients with phakic CME did differ from the Iowa patients with pars-planitis.
All the patients benefited from treatment with topical corticosteroids and cycloplegics during times of flare-up of anterior segment inflammation. Seven of the 19 patients needed treatment only with topical drops throughout their follow-up and retained excellent vision. Six other patients required courses of oral or depot corticosteroid treatment. Five of the 6 were treated for less than a year whereas one required intermittent treatment for 312 years. All of these patients, however, have remained stable without the need for oral or depot corticosteroids for about 4 years in 4 patients and for 10 and 13 years in the other 2 patients. Two of the 19 patients have required long-term oral corticosteroid treatment for disease control. These 2 are now controlling the disease with low doses of prednisone, 5 mg/d,and 7.5 mg every other day.
One of the 19 patients has had severe visual loss from severe anterior segment inflammation and glaucoma. Her treatment was hampered by a severe ocular hypertensive response to corticosteroids. This patient's disease perhaps could have been controlled with other immunosuppressive agents. Another patient has lost vision due to a cataract and posterior synechiae even with intermittent oral and depot corticosteroids. So far, the patient has declined treatment, but this eye appears to visual potential if the cataract were removed. The disease in this patient's other eye is controlled with subtenons depot corticosteroids given approximately once a year. One patient was unavailable for follow-up after 8 months. His visual acuity was 20/16 OU and he was receiving a tapering oral corticosteroid dose. Another patient was treated with both depot and oral corticosteroids and neither seemed to have any effect. This patient's vision has gradually improved to 20/25 OD and 20/32 OS. Overall, 5 of the 19 patients developed elevated intraocular pressure that was thought to be related to treatment with topical, depot, or oral corticosteroids.
We describe a disease entity of idiopathic intermediate uveitis that affects primarily young to middle-aged women with vitritis, CME, and retinal periphlebitis. The disease is typically bilateral and affects mostly the posterior segment of the eye although some eyes had anterior chamber inflammation, keratic precipitates, and posterior synechiae.
Patients with this type of idiopathic intermediate uveitis had a median final visual acuity of 20/20, with 91% of eyes having a final visual acuity of 20/40 or better. The excellent visual acuity is impressive since patients were symptomatic for an average of 154 months and 60% had CME at some time during the course of their disease. At the final visit, 8 of the 31 eyes still had CME but the final median visual acuity in these eyes was also relatively good (ie, 20/32).
The photographic analysis revealed that these patients often had less leakage into the center of the fovea than patients with pseudophakic CME and CME due to pars-planitis. This may in part account for the good long-term visual prognosis. The study patients, like those with pars-planitis but unlike those with pseudophakic CME, also often had retinal venule sheathing, leakage outside the fovea in the posterior pole, and dilated retinal venules.
One of the purposes of this study was to determine whether these patients represented a milder form of pars-planitis.4- 7 None of the patients, however, developed snowbanking or definite multiple sclerosis despite a prolonged follow-up. In addition, these patients did not have the same HLA associations as did the patients with pars-planitis reported by Malinowski et al1 and Tang et al.2 Although there were no statistically significant HLA differences between the Iowa phakic CME patients and the Iowa control group, the Iowa phakic CME patients did differ from the Iowa pars-planitis group of patients at the loci HLA-B8, HLA-B51, and HLA-DR2. Moreover, HLA-DR2 has been associated with multiple sclerosis,1 and there was no statistical association to HLA-DR2 in our patients. These findings do not support that these patients simply represent a milder form or subset of pars-planitis.
We describe a group of patients with characteristics of idiopathic intermediate uveitis. Idiopathic intermediate uveitis has been described as a larger category of which pars-planitis and multiple sclerosis may be a subset.3 Our study patients appear to represent a distinct subset within the larger idiopathic intermediate uveitis classification.
Overall, however, these patients appear to have an excellent long-term visual prognosis. Most of the 19 patients in this study did well with only topical drops or with short courses of oral or depot corticosteroids. The short courses of corticosteroids were helpful in some patients to reduce CME, retinal phlebitis, or severe anterior segment inflammation. We believe, however, that the good visual outcome in most patients was due more to a mild disease process rather than the treatment. Systemic periocular and topical steroids were used intermittently in these patients and appeared to reduce the CME and improve the vision in a few. However, these patients also had a significant risk of developing steroid-related glaucoma or cataract. These complications and the good visual prognosis speak for caution in using corticosteroids, and perhaps only in eyes with severe CME and visual loss.
Accepted for publication April 22, 1998.
Supported in part by an unrestricted grant from Research to Prevent Blindness Inc, New York, NY, to the Department of Ophthalmology and Visual Sciences at the University of Iowa Hospitals and Clinics and The Heed/Knapp Ophthalmic Foundation, Cleveland, Ohio.
Reprints: Donald W. Park MD, Retinal Consultants of Arizona, 2720 N 20th St, Suite 225, Phoenix, AZ 85006.