Chronic Severe Uveitis

Abstract

Introduction There is a significant under-registration of blind and visually impaired patients attending ophthalmic departments. Blindness is defined by the World Health Organization (WHO) as the best possible corrected visual acuity in the better eye of less than 3/60 (20/400 or 0.05) or corresponding visual field ≤10°. Severe visual impairment is defined as the best-corrected visual acuity in the better eye of 3/60 (20/400 or 0.05) or more, but less than 6/60 (20/200 or 0.1) (40,49). In 1995, the WHO estimated that nearly 38 million people in the world were blind and 110 million had severe visual impairment. In 1997, the total number of estimated blind people was 44,800,000 (47). Causes of blindness can be classified according to different age-groups and geographic distribution (49). The major causes of blindness in the middle age-group in the Western world are tapetoretinal degeneration (20%), congenital abnormalities (20%), diabetes (20%), uveitis (10%), and accidents (5%) (17,21,25,44,48). Approximately 5%–20% of legal blindness in developed countries is due to uveitis (7,17,21,46). Uveitis is a potentially sight-threatening intraocular inflammation. Because of a variety of diagnostic criteria used by different centers, the epidemiology and etiologic distribution of uveitis have not been clearly defined. As a result, its incidence is reported to range from 17 per 100,000 to 24 per 100,000 residents, and its prevalence from 38 per 100,000 to 204 per 100,000 residents (7,23,24,38,51). Diagnostic workup determines an infectious etiology or an associated condition in about 60%-70% of cases of uveitis (13,15,16,22,30,41,42,50,52). Primary inflammatory disorders potentially associated with systemic manifestations or inflammatory conditions associated with infectious agents are most frequently reported. Infectious conditions are often associated with uveitis and blindness, especially in developing countries (6,31,32,49,53). Usual cases of uveitis, such as B27-associated disease or herpetic uveitis, are easily handled and are not usually referred to a tertiary eye care center. Most of these cases have a favorable evolution. In a number of complicated cases, the diagnosis is not easy to determine. The use of systemic antiinflammatory and immunosuppressive drugs is often justified and has considerably lowered the incidence of complications in these patients. These therapeutic applications are generally performed in specialized centers under the supervision of internists and ophthalmologists (35). We analyzed a series of 927 consecutive patients referred to our center between 1991 and 1996 for diagnostic and/or therapeutic management of severe uveitis. Therapeutic strategy was evaluated according to the level of intraocular inflammation, and the final visual outcome was determined. Patients and Methods The clinical records of all patients with severe or sight-threatening uveitis referred to the Ophthalmology Department of the Pitié-Salpêtrière Hospital, Paris, France, between January 1991 and January 1996 were retrospectively analyzed. Severe uveitis was defined based on the following criteria at the time of presentation: 1) visual acuity less than 20/100, 2) chronic uveitis (duration >3 months), 3) undiagnosed or unclassified cases after more than 2 different workups, and/or 4) uveitis requiring systemic antiinflammatory or immunosuppressive drugs. In all other cases, uveitis was not considered severe, and those patients were excluded. Furthermore, patients presenting with masquerade syndromes (intraocular lymphoma and postoperative or endogenous endophthalmitis) or infection in an immunosuppressed setting (cytomegalovirus retinitis) were excluded. In the latter, infectious or malignant conditions are highly predominant, and ocular inflammation is a secondary problem. On the other hand, cases of uveitis associated with other infectious diseases, such as toxoplasmic retinochoroiditis, Lyme disease, and syphilis or viral infections in immunocompetent patients, were included in this study. Immune disorders seem to play a major role in these conditions and should be treated in association with antiinfectious drugs. A minimal follow-up of 2 years was required for inclusion in the study to estimate final visual outcome. Classification of uveitis The International Uveitis Study Group (IUSG) classification (2), based on the ocular site of inflammation, was used throughout the study. Anterior uveitis was defined as the presence of anterior chamber inflammatory flare or cells associated with keratic precipitates and/or iris lesions, such as synechiae or partial atrophy. Since June 1991, anterior chamber inflammation was objectively quantified with a Laser Flare Cell Meter (KOWA 1000). Intermediate uveitis was defined as the presence of cells in the vitreous with or without the presence of inflammatory cell condensations in the anterior vitreous (snowballs) or the peripheral retina. The term “pars planitis” was used to describe a subgroup of patients with intermediate uveitis presenting with white exudates (snowbanks) over the pars plana. Posterior uveitis was defined as the presence of inflammatory signs involving the retina, choroid, retinal vessels, and/or posterior vitreous humor. Finally, panuveitis was defined as a combination of inflammation involving the 3 previously described sites. Further classifications were established according to other morphologic features. The presence of keratic precipitates visible on slit lamp examination and accumulating on the endothelial surface of the cornea was included. According to their size and their appearance, keratic precipitates were classified as granulomatous and nongranulomatous. Presence of hypopyon, elevated intraocular pressure (>22 mmHg), cystoid macular edema (confirmed by fluorescein angiography), band keratopathy, keratouveitis, cataract, and papillitis was also noted. Diagnostic criteria Based on a detailed clinical history, complete laboratory workup and additional ophthalmologic or specialized medical consultations were performed when the origin of uveitis was uncertain, and always before the institution of intravenous steroids or immunosuppressive drugs. Evaluation of patients with uveitis potentially associated with spondylarthropathies was based on the European Spondylarthropathy Study Group criteria (10). HLA-B27 typing was performed in all nongranulomatous cases of uveitis and was not limited to patients with typical HLA-B27-associated anterior uveitis. It is well known that severe, atypical, corticodependent, or corticoresistant presentations of uveitis may be associated with seronegative or HLA-B27+ spondylarthropathies; uveitis associated with psoriatic arthritis and inflammatory bowel disease is usually the severe form. Behçet disease was diagnosed according to criteria of the International Study Group for Behçet Disease (5). For granulomatous uveitis associated with sarcoidosis, results of serum angiotensin-converting enzyme and serum lysozyme tests, chest X-ray, and gallium scintigraphy were considered. Evidence of a noncaseating granuloma was required for the diagnosis of sarcoidosis, except when uveitis presented with typical bilateral hilar or mediastinal adenopathy. Vogt-Koyanagi-Harada syndrome was diagnosed according to the classification established by the American Uveitis Society (43). Ocular toxoplasmosis was diagnosed based on the presence of a typical fundus change of focal retinochoroiditis and positive IgG anti-Toxoplasma on serology. Syphilis was diagnosed when treponemal hemagglutination (TPHA) and nontreponemal (VDRL) tests were positive. For tuberculous uveitis, 2 different diagnostic groups were determined. A positive culture for mycobacteria in any fluid or tissue sample was necessary for diagnosis of an active infection by the tubercle bacillus. However, a positive purified protein derivative (PPD) skin test associated with a personal or familial history of tuberculosis, abnormalities on chest X-ray, and granulomatous uveitis without positive culture for mycobacteria was associated with putative hypersensitivity to the bacillus. Herpetic keratouveitis or uveitis was diagnosed based on the presence of corneal scars compatible with a previous herpetic keratitis and/or sectorial iris atrophy and/or ocular hypertension and positive IgG serologic findings. For all potentially infectious uveitis presenting with atypical clinical presentations, anterior chamber or vitreous taps were performed with subsequent local antibody titer determination and, when necessary, specific polymerase chain reactions (PCR) to detect Toxoplasma gondii, different herpes viruses, or bacterial agents such as Tropheryma whippelii(3,9). Cases of uveitis that were not associated with a detectable infectious agent or the above systemic disorders were classified as idiopathic. Statistical analysis and evaluation of visual outcome Data were entered into a FileMaker Pro (FileMaker, Inc.) database for descriptive statistical analysis. State of initial and final visual acuity was reviewed in each anatomic group of uveitis. For each group, visual acuity was classified in 4 different subgroups: 1) less than 20/400, 2) between 20/400 and 20/200, 3) between 20/200 and 20/40, and 4) more than 20/40. Initial and final visual acuity were compared in each anatomic group using the chi-square test. Statistical significance was defined as an alpha of less than 0.05, showing an improvement after a minimal follow-up of 2 years. Results Patient characteristics Among 1,080 patients presenting with 1 of the 4 criteria required for severe uveitis as described in the Patients and Methods section, 927 were finally included. We analyzed 1,310 eyes. One hundred fifty-three patients were excluded due to short follow-up (<24 months). The mean age was 37.1 years at the time of presentation, but 34.5 years at the onset of disease. The female to male sex ratio was 1.05. Most patients (92%) were Caucasian; the cohort comprised 80% Europeans and 12% North Africans. Blacks represented 6% and Asians 2% of patients. Uveitis was unilateral in 58.7% and bilateral in 41.3% of cases. At the time of referral, 29% of the patients were undergoing medical treatment (Table 1). Examples of different aspects of uveitis are provided in Figures 1–6.TABLE 1: Clinical characteristics of 927 patients with severe or sight-threatening uveitisFig. 1: Mutton-fat keratic precipitates associated with systemic sarcoidosis.Fig. 2: Posterior synechiae with totally occluded pupil in a patient with tuberculous uveitis.Fig. 3: Posterior subcapsular cataract in a patient with juvenile chronic idiopathic uveitis.Fig. 4: Viral retinal necrosis in an immunocompetent patient.Fig. 5: Retinal hemorrhages, optic nerve infiltration, and vasculitis in a patient with a severe case of Behçet disease.Fig. 6: Toxoplasmic papillitis associated with a central vein occlusion.Anatomic and etiologic distribution Panuveitis was the most frequent form (35%) followed by anterior uveitis (28.5%), posterior uveitis (21.6%), and intermediate uveitis (15%). At the end of the study, a noninfectious condition was established in 34.2% of cases and an infectious etiology in 31.8% of cases; uveitis remained idiopathic in 34% of cases. The distribution of cases according to the anatomic site of inflammation and associated conditions of uveitis is shown in Table 2. Excluding the idiopathic cases (37.7%), Behçet disease was the most frequent diagnosis in the panuveitis group (13.9%). Herpes virus infections (31%) and HLA-B27-associated uveitis (17.4%) were the most frequent diagnoses in the anterior uveitis group. Toxoplasmic retinochoroiditis was the most frequent in patients presenting with posterior uveitis (39%), but was also noted in a large proportion of patients with panuveitis (9.9%). In the intermediate uveitis group, most cases were idiopathic (75.5%). More than 24.9% of noninfectious cases of uveitis were associated with a systemic disease (Table 3).TABLE 2: Etiologic distribution by anatomic classification of uveitisTABLE 3: Systemic diseases associated with noninfectious uveitisTherapeutic strategies Twenty-nine percent of patients were on systemic steroid or immunosuppressive therapy at the time of presentation. The general aspects of our therapeutic management are given in Table 4. The majority of patients received systemic steroids (68%). Intravenous methylprednisolone pulses were administered in 53.8% of cases. Systemic steroids were used in 399 of 632 (63.1%) patients with noninfectious and 231 of 295 (78.3%) patients with infectious uveitis. The best results were obtained in cases of toxoplasmosis involving the zone 1 of Holland, sparing the fovea and associated with a major intraocular inflammation; in complicated cases of sarcoidosis; and in Behçet disease. In severe cases, such as birdshot retinochoroiditis and other corticoresistant or corticodependent types of posterior uveitis or panuveitis, an immunosuppressive regimen was necessary (9.6% of cases). Azathioprine and cyclosporine were most frequently used. In all confirmed or presumed infectious conditions, antimicrobial, antiviral, or antiparasitic treatment was combined with the antiinflammatory regimen in order to avoid local or systemic complications. Intravenous polyclonal immunoglobulin (IVIg) treatment was used successfully in 8 cases of birdshot retinochoroidopathy. IVIg was given at 1.6 g/kg every 4 weeks for 6 months, followed by injections of 1.2–1.6 g/kg at 6-to 8-week intervals (18).TABLE 4: Medical management of patients with severe uveitisEvolution and visual outcome The mean follow-up was 3.5 years (range, 2–7 yr). Clinical evaluation of intraocular inflammation at the end of the follow-up period confirmed improvement in 59%, stabilization in 22%, and impairment in 19% of cases. Mean values of initial and final anterior chamber inflammation evaluated by laser flare photometry were 27.5 ± 24 ph/ms and 14 ± 8.5 ph/ms, respectively. Anterior chamber inflammation was significantly reduced at the end of the observation period (p < 0.01). In terms of visual acuity, values less than 20/200 in unilateral ocular involvement were observed in 19.5% of patients before treatment, but in only 9.7% of patients at the final stage. In bilateral cases, 6.5% of patients presented to our center with initial visual acuity less than 20/200, and only 3.1% had a final visual acuity below 20/200. The etiologic distribution of these uveitis cases associated with severe outcome is summarized in Table 5. We compared the initial and final states of visual acuity in different anatomic subgroups of uveitis in order to evaluate the efficacy of therapeutic strategies. There was a statistically significant improvement in all except the intermediate uveitis group. Treatment did not seem to change the spontaneous evolution of visual acuity in these patients. It is noteworthy that more than 75% of these cases were idiopathic.TABLE 5: Conditions associated with severe uveitis with poor visual outcome (final visual acuity < 20/200)Ocular complications Cystoid macular edema was the main factor influencing visual outcome in this study. Although we noted a significant improvement of cystoid macular edema after treatment, 18.3% of patients still presented this complication at the end of the observation period. Cataract was present in 14.3% of cases. Cataract extraction was performed in 47% of the latter patients during the period of observation. The other 2 principal complications were secondary glaucoma (6.8% of cases with a mean intraocular pressure of 31 ± 1.6 mmHg) and epiretinal membrane (4.2% of cases). Discussion Panuveitis was the most common presentation in this study, followed by anterior, posterior, and intermediate uveitis. Because of the reference bias in this patient population, these data cannot be considered for the general population of uveitis patients. These results are different from other recently published surveys, which is partly explained by the analysis of a subgroup of patients suffering from severe uveitis in our series (15,22,30). Patients included in this study were those who had the highest risk of ocular complications leading to a poor visual outcome. Under such circumstances, collaboration between ophthalmologists and internists is quite important, requiring extensive workup followed by systemic therapy. The tertiary character of our department could explain the relatively low level of anterior uveitis, which is seen frequently in the general practice of ophthalmology, in this study. This result has been discussed in a previous study reported by McCannel et al (22). Because of their relatively favorable visual prognoses, we have not included patients with uncomplicated HLA-B27 or herpetic uveitis, Fuchs heterochromic cyclitis, and Posner Schlossman disease. However, a subgroup of HLA-B27-associated uveitis has a poor outcome mainly due to posterior segment involvement (29,39,45). The systemic disease is frequently diagnosed after the onset of ocular involvement (19,26,27,33,34,45). Associated conditions were defined in 67.5% of cases. The diagnostic rate is higher than that obtained in 1966 in a relatively comparable French population (14). Our study correlates closely with other studies regarding the percentage of cases in which the etiology was determined (57.8%–67%) (16,22,30,42,50,52). Patterns of uveitis similar to those reported by Banares et al (1) were present in most of our patients. However, anterior chamber tap, diagnostic vitrectomy, and other organ biopsies allowed us to confirm atypical presentations such as Whipple disease or unusual conditions with longer clinical evolution (3,28). These cases were considered as uveitis associated with a systemic disease, and patients were on immunosuppressive regimens. Internists play a major role by ruling out a systemic infection such as tuberculosis, Lyme disease, or Whipple disease associated with an ocular inflammation. Moreover, the etiologic diagnosis was often established during the follow-up period. clinical and laboratory evaluation of uveitis should be performed during the This is the best to the diagnostic efficacy in patients with chronic uveitis. As previously reported the highest proportion of idiopathic uveitis was patients with the intermediate The proportion of noninfectious and infectious of uveitis is nearly the in this study. Uveitis associated with or herpetic infections is still an of complication with poor visual outcome. Ocular manifestations associated with are by an active infection or an in the of any infectious agent such as or are to ocular to detect the presence of the 2 of from the eye was of the is usually associated with and the is in This of involvement quite in our the leading of posterior uveitis in the French For noninfectious uveitis, of cases were associated with systemic and associated conditions were These results the of a collaboration between internists and ophthalmologists in patients with severe uveitis. Systemic steroids and/or immunosuppressive were used in of cases. time an infectious etiology was an antiviral, or treatment was usually 2 before the use of antiinflammatory drugs. and agents were required in patients with Behçet disease, birdshot Vogt-Koyanagi-Harada and severe idiopathic uveitis. In anterior and panuveitis, anterior chamber inflammation evaluated by laser flare photometry during treatment was significantly Initial and final visual acuity were compared It is to that was a significant after treatment in all but the group of patients with intermediate uveitis. This be to the level of idiopathic cases and a evolution in this group. These results are similar to those reported by and The principal complications reported in our patients were cystoid macular edema and cataract However, the prevalence of cystoid macular edema is than in other This could be explained by the and antiinflammatory treatment as as infectious conditions were Cystoid macular edema to steroids and required immunosuppressive drugs especially when associated with Behçet disease and birdshot retinochoroidopathy. Cataract required extraction in most patients. The has the leading to a minimal and of the However, the should be performed in a after a period of 3 and with a antiinflammatory In the use of laser flare photometry could the to steroid and to detect a at a in order to the antiinflammatory regimen without The incidence of blindness patients with uveitis is defined. The incidence of total blindness due to uveitis in the was estimated to be published in based on the epidemiology of uveitis and in the of estimated the prevalence of uveitis as a of blindness to be in the range of per 100,000 This to nearly of the prevalence of uveitis determined in per 100,000 More similar data from different European have been published In the present study we retrospectively evaluated the associated conditions and the prevalence of visual impairment in patients with severe uveitis referred to a tertiary eye care center with a follow-up of 2 years 3.5 yr). collaboration between ophthalmologists and internists is especially in sight-threatening cases of uveitis. to ocular could the diagnostic by infectious conditions to be treated In other cases, systemic steroids or immunosuppressive should be more to visual or general complications. studies with follow-up are necessary to different therapeutic and to the prevalence of blindness in patients with severe uveitis. the infectious associated conditions, and visual of patients with uveitis, we retrospectively analyzed the records of 927 consecutive patients with severe ocular inflammation defined as chronic or sight-threatening who were referred for diagnostic and/or therapeutic management between 1991 and A minimal follow-up of 2 years was The mean age at onset of uveitis was 37.1 and the female to male ratio was nearly Most patients were Uveitis was chronic in of cases, associated with an reduced visual acuity in of cases. The disease was unilateral in 58.7% and bilateral in 41.3% of cases. Panuveitis was the most common presentation followed by anterior (28.5%), posterior (21.6%), and intermediate uveitis (15%). associated condition was determined in 67.5% of cases. uveitis and systemic diseases were associated in 24.9% of patients. Most patients required systemic steroids and/or immunosuppressive Cystoid macular edema and cataract were the most common complications. inflammation was in of the patients. visual acuity was less than 20/200 in 3.1% of bilateral and 9.7% of unilateral cases. The management of patients with severe and sight-threatening of uveitis is performed in collaboration with and on a complete diagnostic and As a result, the rate of blindness due to uveitis can be