Prognostic factors for clinical outcomes in patients with Vogt-Koyanagi-Harada disease treated with high-dose corticosteroids

Abstract

Purpose: To determine prognostic factors in patients with Vogt–Koyanagi–Harada (VKH) disease who were treated with high-dose corticosteroids. Methods: Retrospective analysis of 87 patients (174 eyes). Results: At presentation, there were 53 patients with initial-onset acute VKH disease and 34 patients with chronic recurrent VKH disease. Chronic recurrent presentation was significantly associated with more severe anterior segment inflammation at presentation as indicated by presence of mutton-fat keratic precipitates, anterior chamber reaction ≥2+, iris nodules and posterior synechiae (p < 0.001 for all comparisons), less exudative retinal detachment at presentation (p < 0.001), more complications during the follow-up period (p < 0.001) and a worse visual outcome (p < 0.001). The use of immunomodulatory therapy (cyclosporine and mycophenolate mofetil) as first-line therapy significantly reduced the development of complications in the whole study group (p = 0.006) and in initial-onset acute group (p = 0.024) and improved visual outcome in the whole study group (p = 0.004) and in chronic recurrent group (p = 0.024). In the whole study group, final visual acuity of 20/20 was significantly associated with good initial visual acuity of >20/200 [odds ratio = 4.25; 95% Confidence interval (CI) = 1.53–11.89] and age older than 16 years was significantly associated with the development of complications (odds ratio = 3.15; 95% CI = 1.04–9.48). Conclusions: Chronic recurrent VKH disease is significantly associated with more severe anterior segment inflammation and less exudative retinal detachment at presentation, more ocular complications and a worse visual outcome than initial-onset acute VKH disease. Use of immunomodulatory therapy significantly improved the clinical outcomes. Vogt–Koyanagi–Harada (VKH) disease is an autoimmune disease directed against one or more antigens found on or associated with uveal melanocytes, with subsequent chorioretinal atrophy giving rise to the characteristic ‘sunset glow fundus’ appearance (Yamaki et al. 2000a,b; Gocho et al. 2001). VKH is characterized by bilateral granulomatous panuveitis and exudative retinal detachment. During the acute phase, systemic manifestations include central nervous system and auditory signs. In the chronic or recurrent phase, poliosis, vitiligo and alopecia may develop. The prevalence of the disease varies among different populations of the world, and it commonly affects pigmented races and people of certain genetic predispositions (Moorthy et al. 1995; Fang & Yang 2008). VKH disease is one of the most common uveitis entities in Saudi Arabia (Al-Mezaine et al. 2010), and the visual prognosis is generally good with prompt diagnosis and appropriate immunosuppressive treatment (Al-Kharashi et al. 2007; Abu El-Asrar et al. 2008, 2012). The natural history of VKH disease can be divided into acute, chronic and chronic recurrent stages (Moorthy et al. 1995; Fang & Yang 2008). The disease typically begins with choroiditis with exudative retinal detachment and optic disc hyperaemia and swelling, subsequently involving the anterior segment and finally developing into a recurrent granulomatous anterior uveitis if not properly treated with typical ‘sunset glow fundus’ and chorioretinal atrophy (Yang et al. 2007; Fang & Yang 2008). Bacsal et al. (2008) demonstrated, with the use of indocyanine green angiography, the presence of concomitant subclinical choroidal inflammation in patients with clinically isolated anterior segment recurrence. Vision-threatening complications have clearly been recognized to occur in the chronic recurrent phase of VKH disease, namely cataract, glaucoma, subretinal neovascular membranes and subretinal fibrosis. The occurrence of these complications is known to be associated with a worse visual outcome (Ohno et al. 1988; Moorthy et al. 1995; Read et al. 2001b; Al-Kharashi et al. 2007; Yang et al. 2007; Abu El-Asrar et al. 2008; Fang & Yang 2008). The principles of therapy of VKH disease are to suppress the initial intraocular inflammation in the acute posterior uveitis stage with early and high-dose systemic corticosteroids followed by slow tapering (Rubsamen & Gass 1991; Moorthy et al. 1995). Despite proper treatment with corticosteroids, several studies reported the development of chronic recurrent granulomatous inflammation and ‘sunset glow fundus’ with peripapillary atrophy and depigmented small atrophic lesions at the level of retinal pigment epithelium (Al-Kharashi et al. 2007; Chee et al. 2007; Abu El-Asrar et al. 2008; Cuchacovich et al. 2010; Errera et al. 2011). Recently, several studies suggested that the use of nonsteroid immunomodulatory therapy with cyclosporine, azathioprine, methotrexate and mycophenolate mofetil as first-line therapy in addition to corticosteroids is associated with good clinical results (Paredes et al. 2006; Kim & Yu 2007; Cuchacovich et al. 2010; Abu El-Asrar et al. 2012). Although the prognosis for VKH disease has greatly improved, further studies are needed to determine the optimal treatment regime for this disease. The purpose of this study was therefore to determine the prognostic factors that could assist us to optimize the management of this disease in a group of patients with VKH disease who were treated with high-dose corticosteroids. Several previous studies reported the predictive factors of final visual acuity and development of complications in patients with VKH disease Al-Kharashi et al. 2007; Abu El-Asrar et al. 2008; Ohno et al. 1988; Read et al. 2001a,b; Rubsamen & Gass 1991; Chee et al. 2009). Unlike most other studies, we divided patients into two groups based on the clinical pattern of presentation: patients with initial-onset acute VKH disease and those with chronic recurrent VKH disease. Furthermore, initial-onset acute group was divided into two groups: those who received nonsteroid immunomodulatory therapy as first-line therapy combined with systemic corticosteroids and those who were treated with corticosteroid monotherapy. The outcome measures were final visual acuity and the occurrence of ocular complications. The records of all VKH patients seen in the Uveitis Clinic of King Abdulaziz University Hospital, Riyadh, Saudi Arabia, from January 2004 to April 2011 were retrospectively reviewed. Diagnosis of VKH disease was based on the Revised International Diagnostic Criteria (Read et al. 2001a). Charts were reviewed for demographic data (age and gender), initial and final best-corrected Snellen visual acuities, results of slit-lamp examination of the anterior segment, results of dilated fundus examination, results of fluorescein angiograms, duration from onset of symptoms to presentation to our institute, details of therapy, ocular complications and duration of follow-up. The main outcome measures were visual acuity and the development of complications. The Mann–Whitney test was used to compare means from two independent groups, and the Chi-square test or Fisher’s exact test was used, as appropriate, to compare percentages relating to two categorical variables. A p-value less than 0.05 indicated statistical significance. Stepwise logistic regression analysis was conducted to identify the variables that influenced the attainment of visual acuity of 20 of 20 at last follow-up and the development of any complication of cataract or glaucoma or subretinal neovascular membranes during follow-up. Only variables that had complete data for the 174 eyes included in the study were used as the independent (predictor) variables during the conduction of the logistic regression analyses. Program LR (Logistic Regression) from the BMDP 2007 Statistical Software was used for conducting the logistic regression analyses, and spss Version 15 for Windows software was used for the rest of the analyses. A total of 87 patients (174 eyes) were identified. Patients were 50 (57%) females, and 37 (43%) were males. The age at presentation ranged from 4 to 52 years with a mean of 26.3 ± 10.7 years and a median of 26 years. There were 18 (20.7%) patients who were 16 years of age or younger. For the purpose of analysis, we classified these patients into two groups according to the pattern of clinical characteristics at presentation: patients with initial-onset acute VKH disease and those with chronic recurrent intraocular inflammation. In initial-onset acute VKH disease, the interval between the onset of symptoms and presentation ranged from 2 to 30 days, with a mean of 17 ± 9 days and a median of 21 days. In chronic recurrent VKH disease, this interval ranged from 2 to 44 months, with a mean of 11.6 ± 13.7 months and a median of 6 months. Patients with initial-onset acute VKH disease typically showed exudative retinal detachment and optic disc hyperaemia and swelling with or without anterior uveitis. Patients with chronic recurrent VKH disease typically showed active granulomatous anterior uveitis manifesting as mutton-fat keratic precipitates, posterior synechiae and iris nodules (Koeppe and Busacca). There were 53 patients with initial-onset acute VKH disease and 34 patients with chronic recurrent VKH disease in this series. All patients were managed and followed up by one of the authors (AMA). All patients were treated with systemic corticosteroids. Corticosteroid therapy began with intravenous methylprednisolone 1 g/day (15–30 mg/kg of body weight for children) for 3 days followed by oral prednisone (1 mg/kg of body weight/day) that was maintained for at least 3 weeks. The prednisone dose was gradually tapered following improvement of the intraocular inflammation to a maintenance dose of 5–10 mg/day. Prednisone was tapered at 10 mg every 2 weeks until a daily dose of 40 mg was reached. Afterwards, prednisone was tapered at 5 mg every 2 weeks until 5–10 mg/day was reached. The rate of taper was adjusted according to the patient’s clinical evidence of disease activity. Anterior segment inflammation was treated with topical corticosteroids and cycloplegic agents. The interval between starting treatment and tapering prednisone to 10 mg or less per day ranged from 4 to 10 months with a mean of 5.8 ± 2.3 months and a median of 5 months. The duration of systemic corticosteroid therapy ranged from 11 to 22 months with a mean of 17.3 ± 2.4 months and a median of 17.5 months. Immunomodulatory therapy as first-line therapy combined with systemic corticosteroids was prescribed for 36 patients. Immunomodulatory drugs were used in 25 patients with initial-onset acute VKH disease and in 11 patients with chronic recurrent VKH disease. The immunomodulatory drugs used in this group included mycophenolate mofetil in 22 patients and cyclosporine in 14 patients. The initial dosage of mycophenolate mofetil was 2 g daily and cyclosporine 5 mg/kg daily. The duration of immunomodulatory therapy ranged from 5 to 30 months with a mean of 16.8 ± 5.0 months and a median of 17.8 months. The follow-up period ranged from 9 to 87 months with a mean of 34.2 ± 15.9 months and a median of 32 months. Table 1 displays the baseline and final visual acuity of all eyes in the study. The frequencies along the left-to-right diagonal line represent a total of 70 of 174 (40.2%) eyes that had no change in visual acuity. Improvement in vision occurred in a total of 87 of 174 (50%) eyes (frequencies above the diagonal line), and there was deterioration in vision in 17 of 174 (9.8%) eyes (frequencies below the diagonal line). The prevalence of worst vision of 20 of 200 or worse significantly reduced from 28.2% of the eyes at presentation to 12.1% of the eyes at last follow-up (p = 0.001; Student’s t-test for comparing two percentages from the same sample). The prevalence of best vision of 20 of 20 significantly increased from 29.3% of the eyes at presentation to 63.2% of the eyes at last follow-up (p < 0.001). Univariate analysis demonstrated a significant positive association between final visual acuity of 20 of 20 and clinical findings at presentation including initial visual acuity of better than 20 of 200 (p = 0.009), anterior chamber reaction <2+ (p = 0.043), absence of mutton-fat keratic precipitates (p < 0.001), absence of posterior synechiae (p = 0.003), absence of iris nodules (p = 0.049), presence of exudative retinal detachment (p < 0.001) and initial-onset acute presentation (p < 0.001), use of immunomodulatory therapy (p = 0.004) and findings during the follow-up period including absence of any complication of cataract or glaucoma or subretinal neovascular membranes (p < 0.001), absence of cataract (p < 0.001), absence of subretinal neovascular membranes (p = 0.01), absence of ‘sunset glow fundus’ (p = 0.001), absence of multiple areas of chorioretinal atrophy (p = 0.011), and absence of peripapillary atrophy (p = 0.023) (Table 2). When logistic regression analysis was performed, final visual acuity of 20 of 20 was positively associated with age older than 16 years [odds ratio = 5.50; 95% Confidence interval (CI) = 1.63–18.5], initial visual acuity of better than 20 of 200 (odds ratio = 4.25; 95% CI = 1.53–11.8) and presence of exudative retinal detachment at presentation (odds ratio = 9.05; 95% CI = 1.67–49.1). During the follow-up period, the development of ‘sunset glow fundus’ (odds ratio = 0.092; 95% CI = 0.021–0.402) and cataract (odds ratio = 0.047; 95% CI = 0.015–0.142) was negatively associated with final visual acuity of 20 of 20 (Table 3). The ocular complications encountered during the follow-up period were cataract in 62 (35.6%) eyes, glaucoma that necessitated either medical or surgical intervention in 36 (20.5%) eyes and subretinal neovascular membranes in 12 (6.9%) eyes. Overall, 68 (39%) eyes developed at least one of these complications. Cataract extraction was performed on 18 eyes, and 14 eyes required surgical procedures for control of glaucoma. Other complications encountered during the follow-up period were ‘sunset glow fundus’ in 84 (48.3%) eyes, multiple well-circumscribed areas of chorioretinal atrophy in midperiphery in 80 (46%) eyes and peripapillary atrophy in 30 (17.2%) eyes. On univariate analysis, the development of any complication of cataract or glaucoma or subretinal neovascular membranes was significantly associated with findings at presentation including anterior chamber reaction ≥2+ (p < 0.001), presence of mutton-fat keratic precipitates (p < 0.001), presence of posterior synechiae (p = 0.001), presence of iris nodules (p = 0.005), absence of exudative retinal detachment (p < 0.001) and chronic recurrent presentation (p < 0.001), nonuse of immunomodulatory therapy (p = 0.006), and findings during the follow-up period including development of ‘sunset glow fundus’ (p < 0.001), multiple areas of chorioretinal atrophy (p < 0.001), and peripapillary atrophy (p < 0.001) (Table 4). On logistic regression analysis, the development of any complication of cataract or glaucoma or subretinal neovascular membranes was significantly associated with age older than 16 years (odds ratio = 3.15; 95% CI = 1.04–9.48), presence of mutton-fat keratic precipitates at presentation (odds ratio = 4.60; 95% CI = 2.06–10.3) and chronic recurrent presentation (odds ratio = 11.70; 95% CI = 4.90–27.7). The development of any of these complications was negatively associated with the use of immunomodulatory therapy (odds ratio = 0.36; 95% CI = 0.155–0.832) (Table 3). At presentation, there were significantly more patients aged 16 years or younger in the chronic recurrent group (p = 0.037). Patients with chronic recurrent VKH disease had significantly more severe anterior segment inflammation than those with initial-onset acute VKH disease at presentation as indicated by the presence of anterior chamber reaction ≥2+ (p < 0.001), mutton-fat keratic precipitates (p < 0.001), posterior synechiae (p < 0.001) and iris nodules (p < 0.001). On the other hand, patients with initial-onset acute VKH disease had significantly more exudative retinal detachment compared with those with chronic recurrent VKH disease at presentation (p < 0.001) (Table 5). During the follow-up period, the rates of ocular complications including cataract (p < 0.001), glaucoma (p = 0.004), subretinal neovascular membranes (p < 0.001), ‘sunset glow fundus’ (p < 0.001), multiple areas of chorioretinal atrophy (p < 0.001) and peripapillary atrophy (p < 0.001) were significantly more in patients with chronic recurrent VKH disease compared with those with initial-onset acute VKH disease (Table 5). At the final follow-up period, significantly more eyes in initial-onset acute VKH disease group achieved visual acuity of 20 of 20 compared with those in chronic recurrent VKH disease group (p < 0.001). On the other hand, significantly more eyes in the chronic recurrent VKH disease group had visual acuity of 20 of 50 or worse compared with those in the initial-onset acute VKH disease (p < 0.001) (Table 5). There were no significant differences in baseline characteristics between the corticosteroid group and the corticosteroid + immunomodulatory therapy group. During the follow-up period, the use of immunomodulatory agents as first-line therapy significantly reduced the rates of development of any complication of cataract or glaucoma or subretinal neovascular membranes (p = 0.024), ‘sunset glow fundus’ (p = 0.038) and peripapillary atrophy (p = 0.028), (Table 6). There were no significant differences in baseline characteristics between the corticosteroid group and the corticosteroid + immunomodulatory therapy group. At the final follow-up period, significantly more eyes in the corticosteroid + immunomodulatory therapy group achieved visual acuity of 20 of 20 compared with those in the corticosteroid group (p = 0.024). On the other hand, significantly more eyes in the corticosteroid group had visual acuity of 20 of 50 or worse compared with those in the corticosteroid + immunomodulatory therapy group (p = 0.036) (Table 7). In the current series, there were 18 (20.7%) patients who were 16 years of age or younger. Our multivariate analysis identified a significant association between age older than 16 years at presentation and the development of any complication of cataract or glaucoma or subretinal neovascular membranes. Similarly, previous studies found that an older age at the onset of the disease was significantly associated with the development of cataract (Read et al. 2001a,b; Al-Kharashi et al. 2007; Chee et al. 2009). On the other hand, our analysis showed a significant association between age older than 16 years at presentation and final visual acuity of 20 of 20. In the current study, the initial presenting visual acuity was found to correlate well with final visual acuity. Eyes that had good visual acuity at presentation were more likely to have good final visual acuity. This association retained statistical significance in the multivariate analysis. Initial visual acuity was also shown to be significantly associated with final visual acuity in previous studies (Ohno et al. 1988; Read et al. 2001a,b; Al-Kharashi et al. 2007). In addition, Chee et al. (2009) found that good visual acuity at one month after starting treatment was an important good prognostic factor. The current study has identified a significant association between severity of anterior segment inflammation at presentation and the final visual acuity and the development of complications. There was a significant association between severe anterior uveitis at presentation as indicated by the presence of mutton-fat keratic precipitates, anterior chamber reaction of 2+ or iris nodules and posterior synechiae and a worse final visual acuity and the development of complications. logistic regression analysis, the presence of mutton-fat keratic precipitates at presentation was an independent significant for the development of any complication of cataract or glaucoma or subretinal neovascular membranes. In this study, the presence of exudative retinal a of initial-onset acute VKH disease, at presentation was significantly associated with final visual acuity of 20 of 20 and a of developing any complication of cataract or glaucoma or subretinal neovascular membranes. The association between exudative retinal detachment at presentation and final visual acuity at 20 of 20 retained statistical significance in the multivariate analysis. The development of ocular complications was significantly associated with a worse final visual acuity. Our are with previous a significant association between final visual acuity and of complications (Ohno et al. 1988; Read et al. Patients with VKH disease different clinical manifestations on the duration of the disease The disease typically begins with choroiditis associated with exudative retinal detachment and optic disc hyperaemia and The disease to chronic recurrent granulomatous anterior uveitis if not properly treated with typical ‘sunset glow fundus’ and chorioretinal atrophy (Yang et al. 2007; Fang & Yang 2008). In the current study, patients with chronic recurrent VKH disease had significantly more severe anterior segment inflammation than those with initial-onset acute VKH disease at patients with initial-onset acute VKH disease had significantly more exudative retinal detachment compared with those with chronic recurrent VKH disease at results are with those of Fang et al. (2008) who demonstrated, that and were significantly in patients with chronic recurrent VKH disease than in those with initial-onset acute VKH disease. In addition, recurrent inflammation in the anterior segment was more in patients with recurrent VKH disease. During the follow-up period, the rates of ocular complications including cataract, glaucoma, subretinal neovascular ‘sunset glow chorioretinal atrophy and peripapillary atrophy were significantly more in patients with chronic recurrent VKH disease compared with those with initial-onset acute VKH disease. In multivariate analysis, chronic recurrent presentation was significantly associated with the development of any complication of cataract or glaucoma or subretinal neovascular membranes. Our are with previous that complications were more common in patients with chronic recurrent VKH disease who with recurrent granulomatous anterior uveitis with ‘sunset glow fundus’ (Yang et al. 2007). In addition, et al. demonstrated a significant association between the of chronic ocular inflammation and the appearance of ‘sunset glow Furthermore, indocyanine green studies of patients during of isolated anterior segment showed choroidal of inflammation the absence of clinical of posterior segment and this could to the development of ‘sunset glow fundus’ appearance and chorioretinal atrophy et al. 2008). the final visual outcome was worse in patients with chronic recurrent VKH disease compared with those with initial-onset acute VKH disease. Similarly, previous demonstrated a significant association between chronic recurrent anterior segment inflammation and final visual acuity and the development of complications (Rubsamen & Gass 1991; Read et al. 2001a,b; et al. Al-Kharashi et al. 2007). In the study, we high-dose systemic corticosteroids with tapering in all immunomodulatory therapy with cyclosporine and mycophenolate mofetil as first-line therapy combined with systemic corticosteroids was used in 36 patients. On univariate analysis, the use of immunomodulatory therapy significantly improved visual outcome and reduced the development of complications in the whole study group. logistic regression analysis, the use of immunomodulatory therapy was significantly associated with a of developing any complication of cataract or glaucoma or subretinal neovascular membranes. Our are with previous that the use of immunomodulatory therapy as first-line therapy in addition with systemic corticosteroids is associated with good clinical results (Paredes et al. 2006; Kim & Yu 2007; Cuchacovich et al. 2010; Abu El-Asrar et al. 2012). In patients with initial-onset acute VKH disease, the use of immunomodulatory therapy as first-line therapy combined with systemic corticosteroids significantly reduced the development of any complication of cataract or glaucoma or subretinal neovascular peripapillary atrophy and ‘sunset glow There is evidence to that proper treatment with corticosteroids, patients of the fundus in ‘sunset glow fundus’ appearance and chorioretinal after the clinical disease to be control et al. Al-Kharashi et al. 2007; et al. 2009). The current study has a significant association between the development of ‘sunset glow fundus’ and a worse final visual acuity and the development of any complication of cataract or glaucoma or subretinal neovascular membranes. The association between ‘sunset glow fundus’ and final visual acuity of 20 of 20 retained statistical significance in the multivariate analysis. The development of ‘sunset glow fundus’ and chorioretinal atrophy control of ocular inflammation subclinical choroidal inflammation et al. 2008; et al. et al. In a analysis of eyes with ‘sunset glow fundus’ in patients with VKH disease, of in the with of choroidal was findings the presence of subclinical inflammation in the stage of VKH disease & In the study, the use of immunomodulatory therapy as first-line therapy in patients with initial-onset acute VKH disease significantly reduced the development of ‘sunset glow fundus’ that immunomodulatory therapy was in subclinical choroidal inflammation. uveal and chorioretinal atrophy not be as signs. we that be of subclinical inflammation to immunosuppressive clinical with are to our In visual acuity and severe anterior segment inflammation at presentation are significantly associated with a worse Chronic recurrent VKH disease is significantly associated with more severe anterior segment inflammation and less exudative retinal detachment at presentation, more ocular complications and a worse visual outcome compared with initial-onset acute VKH disease. The use of immunomodulatory therapy as first-line therapy combined with systemic corticosteroids significantly improved the clinical outcomes. The authors for This was by in El-Asrar