Chest Computerized Tomography in the Evaluation of Uveitis in Elderly Women

PETER K. KAISER, MD, CAREEN Y. LOWDER, MD, PHD, PRESTON SULLIVAN, MD, STEVEN R. SANISLO, MD, GREGORY S. KOSMORSKY, DO, MOULAY A. MEZIANE, MD, THOMAS W. RICE, MD, SCOTT D. SMITH, MD, MPH, AND DAVID M. MEISLER, MD

PURPOSE:

To describe the usefulness of chest computerized tomography (CT) in the evaluation of uveitis in elderly women and the clinical characteristics of patients with an abnormal chest CT scan. DESIGN: Prospective noncomparative case series. METHODS: We evaluated 30 elderly women, aged 61 83 years, with chronic iritis, vitritis, or choroiditis and with no denitive cause for their uveitis. All patients underwent a battery of diagnostic laboratory studies and chest CT. RESULTS: The diagnostic examination in most patients included serum angiotensin converting enzyme level, serum lysozyme, rapid plasma reagin level, uorescent treponemal antibody-absorption test, puried protein derivative skin test, and chest x-rays. Chest CT performed on all patients showed parenchymal, mediastinal, and/or hilar adenopathy in 17 patients (57%). Histopathologic conrmation of sarcoidosis with noncaseating granulomas in the biopsy specimens was obtained in 14 patients: eight by mediastinoscopy, two by bronchoscopy, two by conjunctival biopsy, one by nasal biopsy, and one by vitreous biopsy. CONCLUSIONS: Chest CT can be useful in elderly female patients with chronic uveitis for identifying mediastinal lymphadenopathy and other lesions suggestive of sarcoidosis, as well as to help guide tissue conrmation and to rule out other diagnoses including lymphoma. (Am J Ophthalmol 2002;133:499 505. 2002 by Elsevier Science Inc. All rights reserved.)

ANY DIFFERENT UVEITIC ENTITIES CAN PRESENT

for the rst time in elderly women.13 In one report of a heterogeneous uveitis population, 10.4% of patients were older than 60 years of age at diagnosis.2 The nal diagnosis was reported as idiopathic in 31.2% of patientsthe largest category in the report. Diagnosing the underlying cause of uveitis in this age group can be problematic. Despite undergoing a standard battery of diagnostic tests, the underlying cause is often not discovered, making treatment difcult. In this study, we expand upon our previously reported4 experience with chest computerized tomography (CT) for evaluation of chronic uveitis in elderly women. This test can facilitate the diagnosis of sarcoidosis when other diagnostic tests are negative or equivocal.4 Based on this initial experience, we prospectively ordered chest CT scans on all elderly women with chronic uveitis who had no other denitive cause for their uveitis. Herein, we describe our experience with 30 patients who underwent chest CT for evaluation of chronic uveitis.

METHODS
FIFTY-TWO PATIENTS (44 WOMEN, 8 MEN) UNDERWENT

Accepted for publication Dec 14, 2001. From the Division of Ophthalmology, Cole Eye Institute, The Cleveland Clinic Foundation, Cleveland, Ohio (P.K.K., C.Y.L., P.S., S.R.S., G.S.K., S.D.S., D.M.M.); Division of Radiology, The Cleveland Clinic Foundation, Cleveland, Ohio (M.A.M.); Division of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio (T.W.R.). Each author states that he/she has no proprietary interest in the development or marketing of this or any competing piece of equipment. Reprint requests to Careen Y. Lowder, MD, PhD, Division of Ophthalmology, Desk i-32, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195; fax (216) 445-2226; e-mail: lowderc@ ccf.org
0002-9394/02/$22.00 PII S0002-9394(02)01333-8

chest CT as part of their laboratory evaluation between June 1997June 2001 in the Uveitis Department of the Cole Eye Institute. Thirty-two patients (29 women, 3 men) were over age 60. This series includes all patients who met the following inclusion criteria: chronic uveitis dened as chronic iridocyclitis, vitritis, or choroiditis, age greater than 60 years, female gender, and no denitive cause for the uveitis. Also included are our two previously reported cases (patients 16,17) that also met these criteria.4 All patients underwent a laboratory evaluation that included most of the following tests: serum angiotensin converting enzyme (ACE) level, serum lysozyme level, rapid plasma reagin (RPR) level, uorescent treponemal antibody-absorption (FTA-ABS) test, and puried protein derivative (PPD) skin test. All patients had chest CT that was considered positive if it demonstrated parenchymal
RIGHTS RESERVED.

2002 BY

ELSEVIER SCIENCE INC. ALL

499

lesions, mediastinal or hilar adenopathy consistent with sarcoidosis as interpreted by a radiologist. Chest x-ray was performed in 25 patients. The chest x-ray was considered positive if it revealed mediastinal or hilar adenopathy. Further diagnostic testing for tissue conrmation either by bronchoscopy/mediastinoscopy, conjunctival biopsy, nasal mucosal biopsy, or vitreous biopsy was performed in 15 patients. Four patients were excluded: one woman had serpiginous choroiditis and of the three men who were excluded, one had a uveitis masquerade syndrome associated with lung cancer and the other two patients had diffuse uveitis characterized by iritis, vitritis, and macular edema. One of these patients had an elevated serum lysozyme; otherwise, the laboratory evaluation including chest CT was negative.

RESULTS
TWENTY-EIGHT PATIENTS WITH CHRONIC UVEITIS OVER

the age of 60 years were prospectively evaluated and combined with our two previously reported patients (median 72 5.2 years; range 61 83 years; Table 1).4 Characteristics of all 30 patients are listed in Table 1. All patients were elderly women and each had received a careful clinical history and complete ophthalmologic examination. Twenty-ve (83%) patients were Caucasian, three (10%) were African American, and two (7%) were Indian. Six patients had a history of cancer. Laboratory studies (Table 1) were performed on all patients. RPR or FTA-ABS serologies were negative in 24 patients. A PPD skin test was negative in six patients, positive in one patient, and seven patients were anergic. The patient with a positive PPD skin test had previously received antituberculosis medications and, despite treatment, continued to have recurrent uveitis. Six patients had both elevated serum ACE and lysozyme levels but only one (patient 7) of the ve tested had both a positive chest x-ray and chest CT. Concern for central nervous system lymphoma led to magnetic resonance imaging of the brain in 13 of our patients. All but one of these brain scans were read as negative for parenchymal disease or masses. In this patient, sarcoidosis could not be ruled out. In addition, six patients had cerebrospinal uid analysis that was also normal. Seventeen of 25 (68%) patients had no ndings suggestive of any disease process on standard posteroanterior and lateral chest x-rays, despite the radiologists being specically directed to rule out granulomatous lung disease. Evidence of hilar lymphadenopathy was seen in three of 25 chest x-rays. Nonspecic old granulomas or parenchymal nodules were observed on chest x-rays of ve patients. Twenty-four patients had the chest x-ray before chest CT and patient 7 had the chest x-ray following a positive chest CT. Chest CT showed parenchymal lesions, mediastinal, 500 AMERICAN JOURNAL
OF

and/or hilar adenopathy in 17 of 30 (57%) patients (Figure 1). Fifteen patients had biopsies taken from various sites to obtain conclusive tissue conrmation of noncaseating granulomas to support the diagnosis of sarcoidosis (Figure 2). Ten patients underwent mediastinoscopy/bronchoscopy with biopsy of mediastinal nodes. One patient had a nasal biopsy prompted by recurrent nasal bleeding. Five patients had nondirected conjunctival biopsies, although none of our patients had conjunctival lesions suspected to be granulomas. Two of the ve conjunctival biopsies revealed noncaseating granulomas in patients who had a positive chest CT scan (patients 12, 15). Two additional patients who had positive biopsies had negative chest CT scans: patient 19 had a conjunctival biopsy and patient 22 had a vitreous biopsy. Two of the patients who had negative conjunctival biopsies were subsequently shown to have sarcoidosis after biopsy specimens were obtained by mediastinocospy (patients 11, 17). For all mediastinal tissue, Gridly stains revealed no evidence of fungus. Fite, acid fast, and uorochrome stains for mycobacteria were also negative. Finally, all fungal and mycobacterial cultures were negative as well. Clinical characteristics of patients with positive and negative chest CT are listed in Table 2.

DISCUSSION
ALTHOUGH UVEITIS PRESENTING FOR THE FIRST TIME IN

elderly patients is well recognized, the nal diagnosis is usually idiopathic, making treatment problematic.2 Despite undergoing a standard battery of diagnostic tests, the underlying cause is often not discovered. To complicate matters further, several of the possible diagnoses can have overlapping test results. All patients in our study underwent a battery of tests; however, in 27 patients, a diagnosis was not evident before the chest CT scans were performed. Thirteen of our patients had elevated serum ACE levels. Elevated serum ACE levels are seen in 60%90% of cases of sarcoidosis, secondary to enzyme release from epithelioid cells and macrophages within the sarcoid granulomas; however, elevated serum ACE levels are nonspecic.5 For example, serum ACE levels may be elevated in non-Hodgkin lymphoma, another diagnostic consideration in some of our patients.6 Furthermore, six of our patients with normal levels of serum ACE had positive chest CT scans. Thirteen of our patients had an elevated serum lysozyme level. Elevated serum lysozyme levels have been reported in 40% of patients with sarcoidosis; however, elevated serum lysozyme levels are also nonspecic and may be elevated in any granulomatous disease including tuberculosis.5 We would not consider a positive or negative serum ACE or serum lysozyme level specic enough to conrm or refute the diagnosis of sarcoidosis in elderly patients with uveitis. Gallium citrate scans demonstrate increased uptake in OPHTHALMOLOGY APRIL 2002

FIGURE 1. Axial chest computed tomography with contrast obtained at the level of the aortic arch demonstrates prominent retrocaval adenopathy (arrow).

FIGURE 2. Light photomicrograph of mediastinal lymph node showing noncaseating granulomas (hematoxylin and eosin, 300).

areas with sarcoid granulomas, especially the lungs and lacrimal glands. However, gallium scans are nonspecic since gallium uptake has been reported in a wide variety of inammatory and neoplastic thoracic disorders.7 In fact, gallium scans have been used to detect and stage lymphomas and monitor response to lymphoma treatment and lymphoma was a diagnostic consideration in our patients.7 It should be noted that although elevated serum ACE levels and positive gallium scans are nonspecic when taken alone, they may have increased specicity approaching 90% for the diagnosis of sarcoidosis when taken together in one report.8 However, even in that report, one patient was eventually diagnosed with lymphoma.8 Thus, it is difcult to distinguish between lymphoma, tuberculosis, and sarcoidosis with any one diagnostic modality. By combining several diagnostic tests, the differential diagnosis can be narrowed, but a diagnosis is not always made. It is very important to obtain a denitive diagnosis since the prognosis and treatment for these disorders are so different. VOL. 133, NO. 4

In our patients, the CT scans demonstrated parenchymal lesions, mediastinal, and/or hilar adenopathy in 17 out of 30 (57%) patients. One patient (patient 22) with a negative chest CT scan had a positive biopsy. Patient 22 had a diagnostic vitrectomy that revealed granulomas 2 years before her evaluation at the Uveitis Department of the Cole Eye Institute, by which time her intraocular inammation had become inactive and the mediastinal inammatory process may have also resolved leading to a negative chest CT scan. All of the patients who had a positive chest CT scan in this study had active intraocular inammation at the time the chest CT scan was performed. There was no difference in the clinical ndings between the group of patients who had a positive chest CT scan when compared with the group who had a negative chest CT scan. The ability of chest CT scans to demonstrate mediastinal abnormalities that were obscured by normal anatomy and not detected by standard chest x-ray examination has been well recognized by pulmonologists, thoracic surgeons, and radiologists.9 11 The trachea and large vessels may entirely mask the paratracheal, subcarinal, and periaortic lymph node chains that are often involved in disease processes, making denitive diagnosis with chest x-rays alone difcult. Eleven of 17 (65%) patients who had a negative chest x-ray had a positive chest CT scan. Therefore, we procure chest CT on all elderly patients when chest x-rays are unrevealing and there is suspicion for mediastinal disease that may be associated with ocular ndings. Chest CT is a safe procedure as demonstrated in our study. While chest CT is costly (a chest CT with contrast material costs approximately six times that of a routine chest x-ray), its ability to document pathology in 14 of 17 (82%) of our cases that were missed by a standard chest x-ray appears to justify its expense. Despite the ndings on chest CT, the diagnosis of sarcoidosis in our patients was still not conclusive. Mediastinal lymphadenopathy in elderly patients may be due to other diagnoses including lymphoma, metastatic disease (especially small-cell lung cancer), fungal infection, and tuberculosis.2,11 The usual thoracic presentation of lymphoma is single or multiple, paratracheal or hilar nodes.11 While lymphoma and tuberculosis are uncommon, missing the diagnosis may lead to unnecessary morbidity and, perhaps, mortality. Thus, tissue conrmation is paramount when the diagnosis is still in doubt. In 14 of our patients, chest CT detected hidden mediastinal pathology. This nding directed the procurement of tissue for conrmatory diagnosis in ten patients. Since mediastinoscopy is now performed as a routine ambulatory outpatient procedure, the ability to target the nodes for biopsy using CT adds to the diagnostic yield.10 Of the 15 patients who underwent biopsy for tissue diagnosis, 14 (93%) showed histopathologic evidence of sarcoidosis. Thirteen (93%) of these had positive ndings on chest CT, suggesting a high sensitivity of this diagnostic
OF

CT EVALUATION

UVEITIS

501

502
Positive chest CT 1 69 C F 2 3 4 70 65 76 68 72 61 67 77 67 70 80 70 78 83 70 73 I C C C AA I C C C C C C C C C C F F F F F F F F F F F F F F F F 5 6 7 8 9 10 11 12 13 14 15 16 17 OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS OD OS

TABLE 1. Patient Characteristics

ACE Lysozyme Normal Range PPD Malignancy Skin Test

ON Multifocal Chest Chest Biopsy Normal Patient Age Race Sex Active Bilateral Eye Iritis Vitritis CME Edema Vasculitis Retinitis Choroiditis Scleritis CT x-ray Biopsy Site RPR FTA-ABS Patient Range Patient

H/M H H/M H/M H/P H/M H/M H/M H/M M H/M H/M M H/M H/M M H/M na PN OG OG

na

na MLN

na na na na na na

69 56 35 110 16 56

(348) (348)

8.8 (2.88.0) na na (413) (4.714.5) (413) (413) (413) (413) (017.5) (413) (413) (413) (017.5) (413) (413) na (017.5) A na na na A A na na A A na

na

na MLN

(348) 14 (967) 10 (348) 14 (348) 15 (348) 14 (348) 13 (348) 8.1

AMERICAN JOURNAL
OF

na

na TBB TBB MLN MLN MLN MLN CNJ MLN na na

88 42 30 na 28 79 77 49

OPHTHALMOLOGY APRIL 2002

(348) 22 (348) 30 (348) 12 (348) 16 (348) 17 (348) 16 (348) na

na

na CNJ NM MLN

na

77 29 84 55

(348) 14

Negative chest CT 18 70 C F

OG

PN

na

na

na

na

53

(967)

na

na A (continued on next page)

VOL. 133, NO. 4

Negative chest CT (continued) 19 72 C F 20 21 22 73 70 76 69 72 82 81 79 75 74 72 C AA C AA C C C C C C C F F F F F F F F F F F 23 24 25 26 27 28 29 30 OD OS OD OS OD OS OD OS OD OS OD OS OD (phthisis) OS OD OS OD OS OD OS OD OS OD OS

TABLE 1. Patient Characteristics (Continued)

ACE Lysozyme Normal Range PPD Malignancy Skin Test

ON Multifocal Chest Chest Biopsy Normal Patient Age Race Sex Active Bilateral Eye Iritis Vitritis CME Edema Vasculitis Retinitis Chorolditis Scleritis CT x-ray Biopsy Site RPR FTA-ABS Patient Range Patient

OG P P

PN na na na na

CNJ na na VIT na na na na na na na na na na na na na na

na na na na na na na na na na na na na

77 35 7 43 43 3 18 44 27 37

(348) (348) (348) (967) (348) (348) (348) (348) (348) (348) (348) (348)

42 18 13 na na 15 na 16 11 7 11 na

(017.5) (413) (413) na na (413) na (413) (413) (2.88.0) (413) na

na na na na na na na na

CT EVALUATION
OF

UVEITIS 503

A na na

na na

na na

na

30 75

Patients 16 and 17 are from Kosmorsky et al.4 A anergic; AA African American; ACE serum angiotensin converting enzyme level (U/L); Biopsy for noncaseating granulomas; C Caucasian; CME cystoid macular edema; CNJ conjunctiva; CT computerized tomography; F female; FTA-ABS uorescent treponemal antibody-absorption test; H hilar adenopathy; I Indian; lysozyme serum lysozyme level (mg/L); M mediastinal adenopathy; MLN mediastinal lymph node; na not available; NM nasal mucosa; OG old granuloma; ON optic nerve; P parenchymal adenopathy; PN parenchymal nodule; PPD puried protein derivative skin test; RPR rapid plasma reagin level; TBB transbronchial biopsy; Underline ACE and lysozyme values that are elevated; VIT vitreous.

TABLE 2. Clinical Characteristics

Chest Computerized Tomography Scan Positive N 17 % N 13 Negative % N Biopsy Positive 14 % N All Patients 30 %

Active uveitis Bilateral Iritis Vitritis Cystoid macular edema Optic disc edema Retinal vasculitis Retinitis Multifocal choroiditis Scleritis

17 16 14 16 8 4 2 1 11 1

100% 94% 82% 94% 47% 24% 12% 6% 65% 6%

11 12 12 11 8 4 5 1 5 0

85% 92% 92% 85% 62% 31% 38% 8% 38% 0%

13 13 11 13 6 4 3 2 9 1

93% 93% 79% 93% 43% 29% 21% 14% 64% 7%

28 28 26 27 16 8 7 2 16 1

93% 93% 87% 90% 53% 27% 23% 7% 53% 3%

test in detecting sarcoidosis. Of these same 14 patients, nine (64%) had an elevated ACE, and only three (21%) had a positive result on chest x-ray, indicating that each of these tests was much less sensitive than chest CT in detecting disease. The numbers are too small for meaningful analyses. Therefore, based on our study, we cannot ascertain that CT and mediastinal biopsy is more or less sensitive than other diagnostic techniques, such as nondirected conjunctival biopsy, for the diagnosis of sarcoidosis in the elderly population. Nondirected conjunctival biopsies in our hands and others have had varying yields.1216 Directed biopsy of suspicious conjunctival lesions offers a greater yield, but none of our patients had conjunctival lesions.16 A comparison of conjunctival biopsy to chest CT and mediastinal biopsy may be worthy of future study. As a paradigm to diagnosis, it may be worthwhile to rst perform a chest x-ray and then nondirected conjunctival biopsy when the diagnosis of sarcoidosis is being entertained because of the low morbidity and expense of both evaluations.12 Consider proceeding with chest CT and mediastinoscopy if both studies are negative. Thirteen of 17 patients with a positive chest CT scan underwent biopsy and were found to have histologic evidence of sarcoidosis. One patient was found to have granulomas in a vitreous biopsy 2 years before a negative chest CT scan. Establishing the diagnosis of sarcoidosis led to appropriate therapy that would have been different had other mediastinal disease, such as tuberculosis or lymphoma, been made. The diagnosis had no impact on the treatment of sarcoid eye disease itself, which was treated locally. All of our patients were elderly females and 83% were Caucasian. In the United States, sarcoid eye disease is more commonly seen in young, African Americans.17 Sarcoidosis has been estimated to occur three to eight times more often in African Americans than Caucasians.18,19 In addition, most cases of sarcoid eye disease present between the ages of 20 40 years. Only 10% of 504 AMERICAN JOURNAL
OF

cases present after the age of 60 years.19 However, sarcoidosis may affect the elderly to a disproportionately larger degree than other causes of uveitis when compared with a younger population.1,2 Hershey found noncaseating granulomas in conjunctival biopsies obtained from six of seven patients, 58 years or older, with multifocal choroiditis and panuveitis.20 Of our 16 patients with multifocal choroiditis, 11 had positive chest CT scans and nine (56%) of these patients had a positive biopsy. Thus, our total experience, which includes the previously reported cases, reconrms the fact that in elderly women with uveitis and multifocal choroiditis, the diagnosis of sarcoidosis should always be considered. In the absence of a specic causal agent, sarcoidosis is still considered a diagnosis of exclusion. Even the nonnecrotizing granulomas are nonspecic and can be found associated with a wide spectrum of other disorders including fungal, mycobacteria, parasitic and spirochetal infection, primary biliary cirrhosis, Wegener granulomatosis, and mineral deposition such as silicosis and berylliosis.5 None of our patients had clinical or histopathologic changes consistent with any of these disorders, except for patient 11 who had a positive PPD skin test. This patient was exposed to tuberculosis as a tuberculosis sanitarium worker, but then subsequently underwent mediastinoscopy, for which stains and cultures for tuberculosis on tissue obtained were negative. Elucidating the underlying cause of uveitis in the elderly is often problematic. In addition, a nal diagnosis of sarcoidosis is often presumptive and usually based on nonspecic radiographic, serologic, and nuclear scanning tests. Our report indicates the usefulness of chest CT in the diagnosis of sarcoidosis in elderly women with chronic iridocyclitis, vitritis, and multifocal choroiditis. It is also useful to rule out other diagnoses and can direct bronchoscopy/mediastinoscopy to obtain representative tissue for histopathologic study and denitive diagnosis. OPHTHALMOLOGY APRIL 2002

REFERENCES
1. Barton K, Pavesio CE, Towler HMA, Lightman S. Uveitis presenting de novo in the elderly. Eye 1994;8:288 291. 2. Chatzistefanou K, Markomichelakis NN, Christen W, Soheilian M, Foster CS. Characteristics of uveitis presenting for the rst time in the elderly. Ophthalmology 1998;105:347 352. 3. Favre C, Tran VT, Herbort CP. Uveite de la personne agee. Klin Monatsbl Augenheilkd 1994;204:319 322. 4. Kosmorsky GS, Meisler DM, Rice TW, Meziane MA, Lowder CY. Chest computerized tomography and mediastinoscopy in the diagnosis of sarcoidosis-associated uveitis. Am J Ophthalmol 1998;126:132134. 5. Weinreb RN, Tessler H. Laboratory diagnosis of ophthalmic sarcoidosis. Surv Ophthalmol 1984;28:653 664. 6. DeRemee RA, Banks PM. Non-Hodgkins lymphoma associated with hypercalcemia and increased activity of serum angiotensin-converting enzyme. Mayo Clinic Proc 1986;61: 714 718. 7. Ben-Haim S, Bar-Shalom R, Israel O. Utility of gallium-67 scintigraphy in low-grade non-Hodgkins lymphoma. J Clin Oncol 1996;14:1936 1942. 8. Power WJ, Neves RA, Rodriguez A, Pedroza-Seres M, Foster CS. The value of combined serum angiotensin-converting enzyme and gallium scan in diagnosing ocular sarcoidosis. Ophthalmology 1995;102:20072011. 9. Kirtland SH, Winterbauer RH. Pulmonary sarcoidosis. Seminars in Resp Med 1993;14:344 352.

10. Cybulsky IJ, Bennet WF. Mediastinoscopy as a routine outpatient procedure. Ann Thorac Surg 1994;58:176 178. 11. Scully RE, Mark EJ, McNeely WF, Ebeling SH. Case Records of the Massachusetts General Hospital: Case 35-1998. N Engl J Med 1998;339:1534 1541. 12. Leavitt JA, Campbell RJ. Cost-effectiveness in the diagnosis of sarcoidosis: the conjunctival biopsy. Eye 1998;12:959 962. 13. Merritt JC, Lipper SL, Peiffer RL, Hale LM. Conjunctival biopsy in sarcoidosis. J Natl Med Assoc 1980;72:347349. 14. Nichols CW, Eagle RC, Yanoff M, Menocal NG. Conjunctival biopsy as an aid in the evaluation of the patient with suspected sarcoidosis. Ophthalmology 1980;87:287291. 15. Spaide RF, Ward DL. Conjunctival biopsy in the diagnosis of sarcoidosis. Br J Ophthalmol 1990;74:469 471. 16. Solomon DA, Horn BR, Byrd RB, Lorfel RS, Griggs GA. The diagnosis of sarcoidosis by conjunctival biopsy. Chest 1978;74:271273. 17. Jabs DA, Johns CJ. Ocular involvement in chronic sarcoidosis. Am J Ophthalmol 1986;102:297301. 18. Newman LS, Rose CS, Maier LA. Sarcoidosis. N Engl J Med 1997;336:1224 1234. 19. Lardenoye CWTA, Van der Lelij A, de Loos WS, Treffers WF, Rothova A. Peripheral multifocal chorioretinitis. Ophthalmology 1997;104:1820 1826. 20. Hershey JM, Pulido JS, Folberg R, Folk JC, Massicotte SJ. Noncaseating conjunctival granulomas in patients with multifocal choroiditis and panuveitis. Ophthalmology 1994:101.

VOL. 133, NO. 4

CT EVALUATION

OF

UVEITIS

505